From owner-linux-xfs@oss.sgi.com Thu Mar  1 10:29:44 2001
Received:  by oss.sgi.com id <S553679AbRCAS3e>;
	Thu, 1 Mar 2001 10:29:34 -0800
Received: from sgi.SGI.COM ([192.48.153.1]:63826 "EHLO sgi.com")
	by oss.sgi.com with ESMTP id <S553650AbRCAS3J>;
	Thu, 1 Mar 2001 10:29:09 -0800
Received: from zeus-fddi.americas.sgi.com (128-162-8-103.americas.sgi.com [128.162.8.103]) 
	by sgi.com (980327.SGI.8.8.8-aspam/980304.SGI-aspam:
       SGI does not authorize the use of its proprietary
       systems or networks for unsolicited or bulk email
       from the Internet.) 
	via ESMTP id KAA01524
	for <linux-xfs@oss.sgi.com>; Thu, 1 Mar 2001 10:29:08 -0800 (PST)
	mail_from (lord@sgi.com)
Received: from daisy-e185.americas.sgi.com (daisy-e194.americas.sgi.com [128.162.194.214]) by zeus-fddi.americas.sgi.com (8.9.3/americas-smart-nospam1.1) with ESMTP id MAA986237 for <linux-xfs@oss.sgi.com>; Thu, 1 Mar 2001 12:27:52 -0600 (CST)
Received: from jen.americas.sgi.com (IDENT:root@jen.americas.sgi.com [128.162.184.86]) by daisy-e185.americas.sgi.com (SGI-8.9.3/SGI-server-1.7) with ESMTP id MAA87524 for <linux-xfs@oss.sgi.com>; Thu, 1 Mar 2001 12:27:51 -0600 (CST)
From:   Steve Lord <lord@sgi.com>
Received: by jen.americas.sgi.com (8.11.0/SGI-client-1.7) id f21IS2404463; Thu, 1 Mar 2001 12:28:02 -0600
Message-Id: <200103011828.f21IS2404463@jen.americas.sgi.com>
Date:   Thu, 1 Mar 2001 12:28:02 -0600
Subject: TAKE - speed up shutdown with xfs root
To:     unlisted-recipients:; (no To-header on input)
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing

Found a common case where remounting the root filesystem readonly would take
a long time (15 to 30 seconds). This gets it down to fractions of a second
instead.

Date:  Thu Mar  1 10:26:09 PST 2001
Workarea:  jen.americas.sgi.com:/src/lord/xfs-linux.2.4

The following file(s) were checked into:
  bonnie.engr.sgi.com:/isms/slinx/2.4.x-xfs


Modid:  2.4.x-xfs:slinx:88676a
linux/fs/xfs/xfsidbg.c - 1.156
	- print ilock_ra symbolically

linux/fs/xfs/linux/xfs_lrw.c - 1.77
	- Clean up the remount readonly code to do less passes and not to sleep
	  as long.

linux/fs/xfs/linux/xfs_super.c - 1.110
	- Flush out delayed write buffers on a remount readonly before flushing
	  inodes, there are cases where inode buffers are in the delayed write
	  queue and will not be flushed until the pagebuf daemon finds them unless
	  we push here. Also remove the messages about remounting readonly or
	  readwrite.



From owner-linux-xfs@oss.sgi.com Thu Mar  1 10:58:14 2001
Received:  by oss.sgi.com id <S553679AbRCAS5y>;
	Thu, 1 Mar 2001 10:57:54 -0800
Received: from deliverator.sgi.com ([204.94.214.10]:4456 "EHLO
        deliverator.sgi.com") by oss.sgi.com with ESMTP id <S553650AbRCAS5q>;
	Thu, 1 Mar 2001 10:57:46 -0800
Received: from eagdhcp-195-23.americas.sgi.com (root@eagdhcp-195-23.americas.sgi.com [128.162.195.179]) by deliverator.sgi.com (980309.SGI.8.8.8-aspam-6.2/980310.SGI-aspam) via ESMTP id KAA27782
	for <linux-xfs@oss.sgi.com>; Thu, 1 Mar 2001 10:56:41 -0800 (PST)
	mail_from (sandeen@eagdhcp-195-23.americas.sgi.com)
Received: (from sandeen@localhost)
	by eagdhcp-195-23.americas.sgi.com (8.11.0/8.11.0) id f21J4bG02904
	for linux-xfs@oss.sgi.com; Thu, 1 Mar 2001 13:04:37 -0600
Date:   Thu, 1 Mar 2001 13:04:37 -0600
From:   Eric Sandeen <sandeen@eagdhcp-195-23.americas.sgi.com>
Message-Id: <200103011904.f21J4bG02904@eagdhcp-195-23.americas.sgi.com>
Subject: TAKE - O_SYNC handling
To:     unlisted-recipients:; (no To-header on input)
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing

These changes implement O_SYNC write handling as well as the osyncisdsync mount option.

Date:  Thu Mar  1 10:56:45 PST 2001
Workarea:  eagdhcp-195-23.americas.sgi.com:/home/sandeen/workarea

The following file(s) were checked into:
  bonnie.engr.sgi.com:/isms/slinx/2.4.x-xfs


Modid:  2.4.x-xfs:slinx:88679a
linux/fs/xfs/linux/xfs_lrw.c - 1.78
	- Real O_SYNC handling, flushes buffers & log on O_SYNC, also handles osyncisdsync mount option

linux/fs/xfs/linux/xfs_file.c - 1.38
	- Test for O_SYNC and pass PBF_SYNC down via VOP_WRITE

linux/fs/pagebuf/page_buf_io.c - 1.58
	- kicks off IO in pagebuf_write_full_page for last (partial) page



From owner-linux-xfs@oss.sgi.com Thu Mar  1 11:04:14 2001
Received:  by oss.sgi.com id <S553679AbRCATDy>;
	Thu, 1 Mar 2001 11:03:54 -0800
Received: from ns5.novsvcs.net ([192.208.44.111]:24590 "EHLO
        novalfsmtp2.novsvcs.net") by oss.sgi.com with ESMTP
	id <S553650AbRCATDt>; Thu, 1 Mar 2001 11:03:49 -0800
Received: from pp-rafiki-chbs.cp.chbs ([168.246.161.139])
	by novalfsmtp2.novsvcs.net (8.10.2/8.9.3) with ESMTP id f21J4tY11702
	for <linux-xfs@oss.sgi.com>; Thu, 1 Mar 2001 14:04:55 -0500
X-Authentication-Warning: novalfsmtp2.novsvcs.net: Host [168.246.161.139] claimed to be pp-rafiki-chbs.cp.chbs
Received: from pp-banzai-chbs.cp.chbs (pp-banzai-chbs.cp.chbs [168.246.161.82])
	by pp-rafiki-chbs.cp.chbs (Build 101 8.9.3/NT-8.9.3) with ESMTP id UAA03203
	for <linux-xfs@oss.sgi.com>; Thu, 01 Mar 2001 20:03:36 +0100
From:   kenneth.leung@syngenta.com
Received: by pp-banzai-chbs with Internet Mail Service (5.5.2653.19)
	id <GDB4WY64>; Thu, 1 Mar 2001 20:03:37 +0100
Message-ID: <AF84AD19B4A8D411B18500508BAF0E0E03ABF2@SE-EXCUR01-USLJ>
To:     linux-xfs@oss.sgi.com
Subject: Usable memory limit?
Date:   Thu, 1 Mar 2001 20:03:36 +0100 
MIME-Version: 1.0
X-Mailer: Internet Mail Service (5.5.2653.19)
Content-Type: text/plain;
	charset="iso-8859-1"
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing

Hello everyone,

On our Linux + XFS machines that have 1GB of memory, the systems only report
seeing 898,424KB of memory. We are viewing this through the /usr/bin/top
utility. On another machine running Linux without XFS (kernel 2.2.16-22)
which has 2GB of memory, the system sees all the memory.

Is there a barrier on how much memory is usable on the current Linux + XFS
pre-release 0.9 version?

Thanks!
Ken 

*******************************************
Kenneth Leung
IT Administrator
Torrey Mesa Research Institute
kenneth.leung@syngenta.com
(858) 812-1223
*******************************************


From owner-linux-xfs@oss.sgi.com Thu Mar  1 11:17:55 2001
Received:  by oss.sgi.com id <S553679AbRCATRp>;
	Thu, 1 Mar 2001 11:17:45 -0800
Received: from [213.51.129.165] ([213.51.129.165]:35495 "EHLO mail4.home.nl")
	by oss.sgi.com with ESMTP id <S553650AbRCATRY>;
	Thu, 1 Mar 2001 11:17:24 -0800
Received: from p2400 ([212.204.188.15]) by mail4.home.nl
          (InterMail vM.4.01.03.00 201-229-121) with SMTP
          id <20010301191706.MTH2685.mail4.home.nl@p2400>;
          Thu, 1 Mar 2001 19:17:06 +0000
Message-ID: <001f01c0a284$3cecab60$020c080a@thuis>
From:   "Aldert Zomer" <zomeral@biol.rug.nl>
To:     <kenneth.leung@syngenta.com>, <linux-xfs@oss.sgi.com>
References: <AF84AD19B4A8D411B18500508BAF0E0E03ABF2@SE-EXCUR01-USLJ>
Subject: Re: Usable memory limit?
Date:   Thu, 1 Mar 2001 20:17:16 +0100
MIME-Version: 1.0
Content-Type: text/plain;
	charset="iso-8859-1"
Content-Transfer-Encoding: 7bit
X-Priority: 3
X-MSMail-Priority: Normal
X-Mailer: Microsoft Outlook Express 5.50.4133.2400
X-MIMEOLE: Produced By Microsoft MimeOLE V5.50.4133.2400
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing


----- Original Message -----
From: <kenneth.leung@syngenta.com>
To: <linux-xfs@oss.sgi.com>
Sent: Thursday 1 March 2001 20:03 PM
Subject: Usable memory limit?


> Hello everyone,
>
> On our Linux + XFS machines that have 1GB of memory, the systems only
report
> seeing 898,424KB of memory. We are viewing this through the /usr/bin/top
> utility. On another machine running Linux without XFS (kernel 2.2.16-22)
> which has 2GB of memory, the system sees all the memory.
>
> Is there a barrier on how much memory is usable on the current Linux + XFS
> pre-release 0.9 version?

We're running rh7 with XFS from cvs on a box with 2 gb. Top reports 2 gb.
Perhaps the kernel in the pre-release version isn't compiled with highmem
support.


From owner-linux-xfs@oss.sgi.com Thu Mar  1 11:28:34 2001
Received:  by oss.sgi.com id <S553679AbRCAT2Y>;
	Thu, 1 Mar 2001 11:28:24 -0800
Received: from [64.169.83.2] ([64.169.83.2]:9044 "EHLO mail.get2chip.com")
	by oss.sgi.com with ESMTP id <S553650AbRCAT1z>;
	Thu, 1 Mar 2001 11:27:55 -0800
Message-ID: <3A9EA26E.4294A01C@get2chip.com>
Date:   Thu, 01 Mar 2001 11:26:38 -0800
From:   Chris Croswhite <ccroswhite@get2chip.com>
MIME-Version: 1.0
To:     Aldert Zomer <zomeral@biol.rug.nl>
CC:     kenneth.leung@syngenta.com, linux-xfs@oss.sgi.com
Subject: Re: Usable memory limit?
References: <AF84AD19B4A8D411B18500508BAF0E0E03ABF2@SE-EXCUR01-USLJ> <001f01c0a284$3cecab60$020c080a@thuis>
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing

I say this exact behavior on a test machine of RedHat Linux 7.x.

If it is a Red Hat system 7.0, RH has 3 boot modes:

Linux
Linux-smp
Linux-up

The linux kernel has high mem support as well as SMP(in RH terms, the
2.2.16-20-enterprise kernel).  The SMP kernel has no high mem support and linux
up has no builtin modules (well some, but less than the other 2 kernel
versions.).

Aldert Zomer wrote:

> ----- Original Message -----
> From: <kenneth.leung@syngenta.com>
> To: <linux-xfs@oss.sgi.com>
> Sent: Thursday 1 March 2001 20:03 PM
> Subject: Usable memory limit?
>
> > Hello everyone,
> >
> > On our Linux + XFS machines that have 1GB of memory, the systems only
> report
> > seeing 898,424KB of memory. We are viewing this through the /usr/bin/top
> > utility. On another machine running Linux without XFS (kernel 2.2.16-22)
> > which has 2GB of memory, the system sees all the memory.
> >
> > Is there a barrier on how much memory is usable on the current Linux + XFS
> > pre-release 0.9 version?
>
> We're running rh7 with XFS from cvs on a box with 2 gb. Top reports 2 gb.
> Perhaps the kernel in the pre-release version isn't compiled with highmem
> support.


From owner-linux-xfs@oss.sgi.com Thu Mar  1 11:45:34 2001
Received:  by oss.sgi.com id <S553679AbRCATpY>;
	Thu, 1 Mar 2001 11:45:24 -0800
Received: from sgi.SGI.COM ([192.48.153.1]:45938 "EHLO sgi.com")
	by oss.sgi.com with ESMTP id <S553650AbRCATpC>;
	Thu, 1 Mar 2001 11:45:02 -0800
Received: from zeus-fddi.americas.sgi.com (128-162-8-103.americas.sgi.com [128.162.8.103]) 
	by sgi.com (980327.SGI.8.8.8-aspam/980304.SGI-aspam:
       SGI does not authorize the use of its proprietary
       systems or networks for unsolicited or bulk email
       from the Internet.) 
	via ESMTP id LAA06915
	for <linux-xfs@oss.sgi.com>; Thu, 1 Mar 2001 11:44:59 -0800 (PST)
	mail_from (sandeen@sgi.com)
Received: from poppy-e185.americas.sgi.com (poppy.americas.sgi.com [128.162.185.207]) by zeus-fddi.americas.sgi.com (8.9.3/americas-smart-nospam1.1) with ESMTP id NAA987323; Thu, 1 Mar 2001 13:43:44 -0600 (CST)
Received: from sgi.com (sandeen@eagdhcp-195-23.americas.sgi.com [128.162.195.179]) by poppy-e185.americas.sgi.com (980427.SGI.8.8.8/SGI-server-1.7) with ESMTP id NAA81209; Thu, 1 Mar 2001 13:43:44 -0600 (CST)
Message-ID: <3A9EA80C.36E9521C@sgi.com>
Date:   Thu, 01 Mar 2001 13:50:36 -0600
From:   Eric Sandeen <sandeen@sgi.com>
X-Mailer: Mozilla 4.76 [en] (X11; U; Linux 2.4.2-XFS i686)
X-Accept-Language: en
MIME-Version: 1.0
To:     Chris Croswhite <ccroswhite@get2chip.com>
CC:     Aldert Zomer <zomeral@biol.rug.nl>, kenneth.leung@syngenta.com,
        linux-xfs@oss.sgi.com
Subject: Re: Usable memory limit?
References: <AF84AD19B4A8D411B18500508BAF0E0E03ABF2@SE-EXCUR01-USLJ> <001f01c0a284$3cecab60$020c080a@thuis> <3A9EA26E.4294A01C@get2chip.com>
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing

Yep, the only x86 kernel on the XFS-PR0.9 CD with HIGHMEM configured is
the i686-enterprise kernel:

kernel-2.4.0-i686-enterprise.config:CONFIG_HIGHMEM64G=y
kernel-2.4.0-i686-enterprise.config:CONFIG_HIGHMEM=y

There are other differences in the enterprise kernel as well, of course
- although I'm not sure what all, offhand...  the XFS kernel RPMs follow
Red Hat's configs very closely.

-Eric

Chris Croswhite wrote:
> 
> I say this exact behavior on a test machine of RedHat Linux 7.x.
> 
> If it is a Red Hat system 7.0, RH has 3 boot modes:
> 
> Linux
> Linux-smp
> Linux-up
> 
> The linux kernel has high mem support as well as SMP(in RH terms, the
> 2.2.16-20-enterprise kernel).  The SMP kernel has no high mem support and linux
> up has no builtin modules (well some, but less than the other 2 kernel
> versions.).

From owner-linux-xfs@oss.sgi.com Thu Mar  1 11:52:34 2001
Received:  by oss.sgi.com id <S553690AbRCATwZ>;
	Thu, 1 Mar 2001 11:52:25 -0800
Received: from tan7.ncr.com ([192.127.94.7]:34080 "EHLO
        opus.sandiegoca.ncr.com") by oss.sgi.com with ESMTP
	id <S553663AbRCATwR>; Thu, 1 Mar 2001 11:52:17 -0800
Received: from localhost (chuckr@localhost)
	by opus.sandiegoca.ncr.com (8.11.1/8.11.1) with ESMTP id f21JqZW40099
	for <linux-xfs@oss.sgi.com>; Thu, 1 Mar 2001 11:52:35 -0800 (PST)
	(envelope-from chuckr@opus.sandiegoca.ncr.com)
Date:   Thu, 1 Mar 2001 11:52:35 -0800 (PST)
From:   Chuck Rouillard <chuckr@opus.sandiegoca.ncr.com>
To:     Linux - XFS <linux-xfs@oss.sgi.com>
Subject: 2GB filesize limit.
Message-ID: <Pine.BSF.4.31.0103011113230.40049-100000@opus.sandiegoca.ncr.com>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing



1. Does Linux/XFS break the 2G filesize limit (on the
   assumption that certain support libraries exist and
   correctly configured)?

   The linux-xfs.sg.com/projects/xfs/ page mentions
   "(no more 2GB limits)" but it isn't clear what exactly
   that refers to, nor if that claim is valid with Linux.

2. If the above is true, do userland tools such as `ls',
   `cp', `mv', et cetera, need mending?

3. Has XFS support on Linux/IA-64 reached the mumbled
   stage?

I have taken a look at the SGI XFS FAQ and sifted through
the mail headers in the 'Mail Thread Index' to no avail.

If the answer lies among the "XFS Talks and Papers", my
apologies.  I have not read through all of them yet.

thanks,

.cer

--------------------------------------------------------------
|  _    . |  _____ |  ____  | Chuck Rouillard                |
| | \   | | |      | |    | | NCR -- RB MPP Systems          |
| |  \  | | |      | |  --  | chuckr@opus.sandiegoca.ncr.com |
| |   \_| | |_____ | |   \  | voice mail 485-2720            |
--------------------------------------------------------------


From owner-linux-xfs@oss.sgi.com Thu Mar  1 12:08:35 2001
Received:  by oss.sgi.com id <S553690AbRCAUI0>;
	Thu, 1 Mar 2001 12:08:26 -0800
Received: from deliverator.sgi.com ([204.94.214.10]:46340 "EHLO
        deliverator.sgi.com") by oss.sgi.com with ESMTP id <S553663AbRCAUIK>;
	Thu, 1 Mar 2001 12:08:10 -0800
Received: from zeus-fddi.americas.sgi.com (zeus-fddi.americas.sgi.com [128.162.8.103]) by deliverator.sgi.com (980309.SGI.8.8.8-aspam-6.2/980310.SGI-aspam) via ESMTP id MAA09367
	for <linux-xfs@oss.sgi.com>; Thu, 1 Mar 2001 12:07:04 -0800 (PST)
	mail_from (lord@sgi.com)
Received: from daisy-e185.americas.sgi.com (daisy-e194.americas.sgi.com [128.162.194.214]) by zeus-fddi.americas.sgi.com (8.9.3/americas-smart-nospam1.1) with ESMTP id OAA987698; Thu, 1 Mar 2001 14:06:52 -0600 (CST)
Received: from jen.americas.sgi.com (IDENT:root@jen.americas.sgi.com [128.162.184.86]) by daisy-e185.americas.sgi.com (SGI-8.9.3/SGI-server-1.7) with ESMTP id OAA01803; Thu, 1 Mar 2001 14:06:52 -0600 (CST)
Received: from jen.americas.sgi.com by jen.americas.sgi.com (8.11.0/SGI-client-1.7) via ESMTP id f21K71Z11156; Thu, 1 Mar 2001 14:07:01 -0600
Message-Id: <200103012007.f21K71Z11156@jen.americas.sgi.com>
To:     Chuck Rouillard <chuckr@opus.sandiegoca.ncr.com>
cc:     Linux - XFS <linux-xfs@oss.sgi.com>
Subject: Re: 2GB filesize limit. 
References: <Pine.BSF.4.31.0103011113230.40049-100000@opus.sandiegoca.ncr.com>
Comments: In-reply-to Chuck Rouillard <chuckr@opus.sandiegoca.ncr.com>
   message dated "Thu, 01 Mar 2001 11:52:35 -0800."
Date:   Thu, 01 Mar 2001 14:07:01 -0600
From:   Steve Lord <lord@sgi.com>
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing

> 
> 
> 1. Does Linux/XFS break the 2G filesize limit (on the
>    assumption that certain support libraries exist and
>    correctly configured)?
> 
>    The linux-xfs.sg.com/projects/xfs/ page mentions
>    "(no more 2GB limits)" but it isn't clear what exactly
>    that refers to, nor if that claim is valid with Linux.

User space needs to be 64 bit clean - which translates to an installation
which has the 'LFS' patches installed. This applies to most modern
distributions now I think. An d to be honest, ext2 will give you 2G+ files
in 2.4 as well.

> 
> 2. If the above is true, do userland tools such as `ls',
>    `cp', `mv', et cetera, need mending?

Done by the above.

> 
> 3. Has XFS support on Linux/IA-64 reached the mumbled
>    stage?

I know we have it up on an ia64 box somewhere and it is functioning,
exactly what hoops had to be jumped through to get it built I do not
know, but I do not believe xfs itself needed changing. You may need
other kernel changes for your flavor if ia64 of course.

Steve

> 
> I have taken a look at the SGI XFS FAQ and sifted through
> the mail headers in the 'Mail Thread Index' to no avail.
> 
> If the answer lies among the "XFS Talks and Papers", my
> apologies.  I have not read through all of them yet.
> 
> thanks,
> 
> .cer
> 
> --------------------------------------------------------------
> |  _    . |  _____ |  ____  | Chuck Rouillard                |
> | | \   | | |      | |    | | NCR -- RB MPP Systems          |
> | |  \  | | |      | |  --  | chuckr@opus.sandiegoca.ncr.com |
> | |   \_| | |_____ | |   \  | voice mail 485-2720            |
> --------------------------------------------------------------

From owner-linux-xfs@oss.sgi.com Thu Mar  1 12:50:45 2001
Received:  by oss.sgi.com id <S553719AbRCAUug>;
	Thu, 1 Mar 2001 12:50:36 -0800
Received: from sgi.SGI.COM ([192.48.153.1]:50712 "EHLO sgi.com")
	by oss.sgi.com with ESMTP id <S553708AbRCAUu3>;
	Thu, 1 Mar 2001 12:50:29 -0800
Received: from madurai.engr.sgi.com ([163.154.5.75]) 
	by sgi.com (980327.SGI.8.8.8-aspam/980304.SGI-aspam:
       SGI does not authorize the use of its proprietary
       systems or networks for unsolicited or bulk email
       from the Internet.) 
	via ESMTP id MAA02320
	for <linux-xfs@oss.sgi.com>; Thu, 1 Mar 2001 12:50:28 -0800 (PST)
	mail_from (ananth@sgi.com)
Received: from sgi.com (mango.engr.sgi.com [163.154.5.76]) by madurai.engr.sgi.com (980427.SGI.8.8.8/970903.SGI.AUTOCF) via ESMTP id MAA01259; Thu, 1 Mar 2001 12:44:33 -0800 (PST)
Message-ID: <3A9EB582.AD175962@sgi.com>
Date:   Thu, 01 Mar 2001 12:48:02 -0800
From:   Rajagopal Ananthanarayanan <ananth@sgi.com>
X-Mailer: Mozilla 4.72 [en] (X11; U; Linux 2.2.16-4SGI_20smp i686)
X-Accept-Language: en
MIME-Version: 1.0
To:     Chuck Rouillard <chuckr@opus.sandiegoca.ncr.com>
CC:     Linux - XFS <linux-xfs@oss.sgi.com>
Subject: Re: 2GB filesize limit.
References: <Pine.BSF.4.31.0103011113230.40049-100000@opus.sandiegoca.ncr.com> <200103012007.f21K71Z11156@jen.americas.sgi.com>
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing

Steve Lord wrote:

> >
> > 3. Has XFS support on Linux/IA-64 reached the mumbled
> >    stage?
> 
> I know we have it up on an ia64 box somewhere and it is functioning,
> exactly what hoops had to be jumped through to get it built I do not
> know, but I do not believe xfs itself needed changing. You may need
> other kernel changes for your flavor if ia64 of course.


I tried XFS on an IA64 at 2.4.0, little over a month ago.
Since then XFS has moved fairly aggressively to keep up
with 2.4.{1,2} ... only in the last couple of days
there are patches available for 2.4.2 for IA64, so I haven't
tried it yet.

On 2.4.0 XFS didn't have any problems on IA64. To generate
the kernel, I applied the IA64 patch from kernel.org to
the XFS tree. There were only trivial merge issues, and
compilation went through fine. Similarly compiling xfs user-space
tools was mostly painless.

ananth.

--------------------------------------------------------------------------
Rajagopal Ananthanarayanan ("ananth")
Member Technical Staff, SGI.
--------------------------------------------------------------------------

From owner-linux-xfs@oss.sgi.com Thu Mar  1 13:16:56 2001
Received:  by oss.sgi.com id <S553714AbRCAVQq>;
	Thu, 1 Mar 2001 13:16:46 -0800
Received: from pneumatic-tube.sgi.com ([204.94.214.22]:51826 "EHLO
        pneumatic-tube.sgi.com") by oss.sgi.com with ESMTP
	id <S553692AbRCAVQX>; Thu, 1 Mar 2001 13:16:23 -0800
Received: from zeus-fddi.americas.sgi.com (zeus-fddi.americas.sgi.com [128.162.8.103]) by pneumatic-tube.sgi.com (980327.SGI.8.8.8-aspam/980310.SGI-aspam) via ESMTP id NAA06812
	for <linux-xfs@oss.sgi.com>; Thu, 1 Mar 2001 13:26:00 -0800 (PST)
	mail_from (lord@sgi.com)
Received: from daisy-e185.americas.sgi.com (daisy-e194.americas.sgi.com [128.162.194.214]) by zeus-fddi.americas.sgi.com (8.9.3/americas-smart-nospam1.1) with ESMTP id PAA987207 for <linux-xfs@oss.sgi.com>; Thu, 1 Mar 2001 15:15:06 -0600 (CST)
Received: from jen.americas.sgi.com (IDENT:root@jen.americas.sgi.com [128.162.184.86]) by daisy-e185.americas.sgi.com (SGI-8.9.3/SGI-server-1.7) with ESMTP id PAA56637 for <linux-xfs@oss.sgi.com>; Thu, 1 Mar 2001 15:15:06 -0600 (CST)
From:   Steve Lord <lord@sgi.com>
Received: by jen.americas.sgi.com (8.11.0/SGI-client-1.7) id f21LFFA21077; Thu, 1 Mar 2001 15:15:15 -0600
Message-Id: <200103012115.f21LFFA21077@jen.americas.sgi.com>
Date:   Thu, 1 Mar 2001 15:15:15 -0600
Subject: TAKE - simplify pagebuf_flush
To:     unlisted-recipients:; (no To-header on input)
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing

This takes advantage of the buffer heads being on the inode dirty list
to do the flush operation. About 30 lines of code collapses down to 3
which are faster.

Date:  Thu Mar  1 13:13:39 PST 2001
Workarea:  jen.americas.sgi.com:/src/lord/xfs-linux.2.4

The following file(s) were checked into:
  bonnie.engr.sgi.com:/isms/slinx/2.4.x-xfs


Modid:  2.4.x-xfs:slinx:88692a
linux/kernel/ksyms.c - 1.80
	- change exports to reflect new pagebuf code

linux/include/linux/page_buf.h - 1.77
	- remove unused definitions

linux/fs/pagebuf/page_buf.c - 1.60
	- remove an unused pagebuf flag and cleanup the buffer head I/O path a little,
	  more to come here.

linux/fs/pagebuf/page_buf_io.c - 1.59
	- Simplify pagebuf_flush to use base linux functions. remove unused
	  pagebuf flag.



From owner-linux-xfs@oss.sgi.com Thu Mar  1 15:40:06 2001
Received:  by oss.sgi.com id <S553766AbRCAXjz>;
	Thu, 1 Mar 2001 15:39:55 -0800
Received: from pneumatic-tube.sgi.com ([204.94.214.22]:27916 "EHLO
        pneumatic-tube.sgi.com") by oss.sgi.com with ESMTP
	id <S553761AbRCAXjm>; Thu, 1 Mar 2001 15:39:42 -0800
Received: from larry.melbourne.sgi.com (larry.melbourne.sgi.com [134.14.52.130]) by pneumatic-tube.sgi.com (980327.SGI.8.8.8-aspam/980310.SGI-aspam) via SMTP id PAA02527
	for <linux-xfs@oss.sgi.com>; Thu, 1 Mar 2001 15:49:19 -0800 (PST)
	mail_from (nathans@wobbly.melbourne.sgi.com)
Received: from wobbly.melbourne.sgi.com (wobbly.melbourne.sgi.com [134.14.55.135]) by larry.melbourne.sgi.com (950413.SGI.8.6.12/950213.SGI.AUTOCF) via ESMTP id KAA15464; Fri, 2 Mar 2001 10:38:19 +1100
Received: (from nathans@localhost) by wobbly.melbourne.sgi.com (980427.SGI.8.8.8/980728.SGI.AUTOCF) id KAA81798; Fri, 2 Mar 2001 10:37:59 +1100 (EDT)
From:   "Nathan Scott" <nathans@wobbly.melbourne.sgi.com>
Message-Id: <10103021037.ZM181828@wobbly.melbourne.sgi.com>
Date:   Fri, 2 Mar 2001 10:37:58 -0400
X-Mailer: Z-Mail (3.2.3 08feb96 MediaMail)
To:     linux-xfs@oss.sgi.com
Subject: Debian packages
Cc:     tbm@debian.org, troup@debian.org, masklin@debian.org,
        broberts@mbhs.edu
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing

hi,

For those folk using Debian, the base XFS userspace tools
(xfsprogs) can now be downloaded as part of the "unstable"
distribution.  Many thanks to the people who've helped out
over the past few months in getting to this point.


# apt-get update
...[snip]...
# 
# apt-get source xfsprogs
Reading Package Lists... Done
Building Dependency Tree... Done
Need to get 722kB of source archives.
Get:1 ftp://ftp.debian.org unstable/main xfsprogs 1.1.3 (dsc) [557B]
Get:2 ftp://ftp.debian.org unstable/main xfsprogs 1.1.3 (tar) [721kB]
Fetched 722kB in 30s (23.3kB/s)                                                
dpkg-source: extracting xfsprogs in xfsprogs-1.1.3
# 
# apt-get install xfsprogs xfslibs-dev
Reading Package Lists... Done
Building Dependency Tree... Done
The following NEW packages will be installed:
  xfslibs-dev xfsprogs 
0 packages upgraded, 2 newly installed, 0 to remove and 6  not upgraded.
Need to get 890kB of archives. After unpacking 3404kB will be used.
Get:1 ftp://ftp.debian.org unstable/main xfsprogs 1.1.3 [672kB]
Get:2 ftp://ftp.debian.org unstable/main xfslibs-dev 1.1.3 [218kB]             
Fetched 890kB in 23s (37.1kB/s)                                                
Selecting previously deselected package xfsprogs.
(Reading database ... 12848 files and directories currently installed.)
Unpacking xfsprogs (from .../xfsprogs_1.1.3_i386.deb) ...
Selecting previously deselected package xfslibs-dev.
Unpacking xfslibs-dev (from .../xfslibs-dev_1.1.3_i386.deb) ...
Setting up xfsprogs (1.1.3) ...

Setting up xfslibs-dev (1.1.3) ...

# 


cheers.

-- 
Nathan

From owner-linux-xfs@oss.sgi.com Thu Mar  1 16:00:06 2001
Received:  by oss.sgi.com id <S553744AbRCAX7z>;
	Thu, 1 Mar 2001 15:59:55 -0800
Received: from pipt.oz.cc.utah.edu ([155.99.2.7]:10993 "EHLO
        pipt.oz.cc.utah.edu") by oss.sgi.com with ESMTP id <S553656AbRCAX7e>;
	Thu, 1 Mar 2001 15:59:34 -0800
Received: from localhost (jdr1529@localhost)
	by pipt.oz.cc.utah.edu (8.9.2/8.9.2) with ESMTP id QAA20442
	for <linux-xfs@oss.sgi.com>; Thu, 1 Mar 2001 16:59:20 -0700 (MST)
Date:   Thu, 1 Mar 2001 16:59:20 -0700 (MST)
From:   james rich <james.rich@m.cc.utah.edu>
To:     linux-xfs@oss.sgi.com
Subject: Re: building 2.4.2 (with XFS) fails
In-Reply-To: <10102251318.ZM160565@wobbly.melbourne.sgi.com>
Message-ID: <Pine.GSO.4.05.10103011655001.18915-100000@pipt.oz.cc.utah.edu>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing

On Sun, 25 Feb 2001, Nathan Scott wrote:

> On Feb 24,  8:53pm, Mike Bernson wrote:
> > Subject: Re: building 2.4.2 (with XFS) fails
> > > ...
> > > If you have a drivers/char/defkeymap.c file, you could try
> > > deleting it & allow a fresh build to regenerate it.  since
> > > your non-XFS 2.4.2 build works, this should too (there are
> > > no XFS changes around this code that I'm aware of).
> > > 
> > > hope this helps.
> > > 
> > 
> > I have been working for the cvs source. The defkeymap.c is
> > empty and loadkey command is now where to be found.
> >-- End of excerpt from Mike Bernson
> 
> yup, an empty defkeymap.c would cause this same problem I think.
> you'll need the console-tools package installed which contains
> loadkeys and /bin will probably need to be in your $PATH.
> 
> The empty file is probably created by the shell redirect in..
> [drivers/char/Makefile, line 202]
> defkeymap.c: defkeymap.map
>         loadkeys --mktable defkeymap.map | sed -e 's/^static *//' > defkeymap.c

Can someone explain why defkeymap.c from XFS CVS needs loadkeys and the
same file from stock 2.4 doesn't?  Does XFS 2.4 need to regenerate this
file or something?  defkeymap.c from stock 2.4 is fine, what is different
for XFS?

James Rich
james.rich@m.cc.utah.edu


From owner-linux-xfs@oss.sgi.com Thu Mar  1 16:01:26 2001
Received:  by oss.sgi.com id <S553744AbRCBABG>;
	Thu, 1 Mar 2001 16:01:06 -0800
Received: from h217n3fls20o974.telia.com ([212.181.166.217]:12292 "HELO
        neoblade.net") by oss.sgi.com with SMTP id <S553656AbRCBABD>;
	Thu, 1 Mar 2001 16:01:03 -0800
Received: from student.liu.se (localhost.localdomain [127.0.0.1])
	by neoblade.net (Postfix) with ESMTP id 370D9188D63
	for <linux-xfs@oss.sgi.com>; Fri,  2 Mar 2001 01:02:42 +0100 (CET)
Message-ID: <3A9EE314.3020302@student.liu.se>
Date:   Fri, 02 Mar 2001 01:02:28 +0100
From:   Joakim Bodin <joabo552@student.liu.se>
User-Agent: Mozilla/5.0 (X11; U; Linux 2.4.2-XFS i686; en-US; 0.8) Gecko/20010219
X-Accept-Language: sv, en
MIME-Version: 1.0
To:     Linux-XFS Mailing List <linux-xfs@oss.sgi.com>
Subject: Latest XFS kupdate oops
Content-Type: multipart/mixed;
 boundary="------------020001040106000104080607"
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing

This is a multi-part message in MIME format.
--------------020001040106000104080607
Content-Type: text/plain; charset=us-ascii; format=flowed
Content-Transfer-Encoding: 7bit

I updated xfs cvs just now (1 March 23:30 GMT) and just after rebooting 
kupdate segfaulted on me and resultet this attached Call Trace. I'm no 
kernel expert and I don't know if this sayes much. I could probably get 
more info about this if given the prober pointers. This is a up to date 
Linux-Mandrake cooker system and the kernel was compiled with RH7.0 
kgcc. I'm attaching my dmesg log as well.

Joakim Bodin

--------------020001040106000104080607
Content-Type: text/plain;
 name="dmesg"
Content-Transfer-Encoding: 7bit
Content-Disposition: inline;
 filename="dmesg"

Linux version 2.4.2-XFS (root@neoblade.net) (gcc version egcs-2.91.66 19990314/Linux (egcs-1.1.2 release)) #15 fre mar 2 00:24:32 CET 2001
BIOS-provided physical RAM map:
 BIOS-e820: 000000000009fc00 @ 0000000000000000 (usable)
 BIOS-e820: 0000000000000400 @ 000000000009fc00 (reserved)
 BIOS-e820: 0000000000010000 @ 00000000000f0000 (reserved)
 BIOS-e820: 0000000007efd000 @ 0000000000100000 (usable)
 BIOS-e820: 0000000000002000 @ 0000000007ffd000 (ACPI data)
 BIOS-e820: 0000000000001000 @ 0000000007fff000 (ACPI NVS)
 BIOS-e820: 0000000000010000 @ 00000000ffff0000 (reserved)
On node 0 totalpages: 32765
zone(0): 4096 pages.
zone(1): 28669 pages.
zone(2): 0 pages.
Kernel command line: BOOT_IMAGE=linux ro root=1606 hdb=scsi ide1=autotune ide0=autotune devfs=mount
ide_setup: hdb=scsi
ide_setup: ide1=autotune
ide_setup: ide0=autotune
Initializing CPU#0
Detected 300.683 MHz processor.
Console: colour VGA+ 80x25
Calibrating delay loop... 599.65 BogoMIPS
Memory: 126464k/131060k available (1283k kernel code, 4208k reserved, 348k data, 184k init, 0k highmem)
Dentry-cache hash table entries: 16384 (order: 5, 131072 bytes)
Buffer-cache hash table entries: 4096 (order: 2, 16384 bytes)
Page-cache hash table entries: 32768 (order: 5, 131072 bytes)
Inode-cache hash table entries: 8192 (order: 4, 65536 bytes)
CPU: Before vendor init, caps: 0080f9ff 00000000 00000000, vendor = 0
CPU: L1 I cache: 16K, L1 D cache: 16K
CPU: L2 cache: 512K
Intel machine check architecture supported.
Intel machine check reporting enabled on CPU#0.
CPU: After vendor init, caps: 0080f9ff 00000000 00000000 00000000
CPU: After generic, caps: 0080f9ff 00000000 00000000 00000000
CPU: Common caps: 0080f9ff 00000000 00000000 00000000
CPU: Intel Pentium II (Klamath) stepping 04
Checking 'hlt' instruction... OK.
POSIX conformance testing by UNIFIX
mtrr: v1.37 (20001109) Richard Gooch (rgooch@atnf.csiro.au)
mtrr: detected mtrr type: Intel
PCI: PCI BIOS revision 2.10 entry at 0xf06d0, last bus=1
PCI: Using configuration type 1
PCI: Probing PCI hardware
Unknown bridge resource 0: assuming transparent
PCI: Using IRQ router PIIX [8086/7110] at 00:04.0
Limiting direct PCI/PCI transfers.
isapnp: Scanning for Pnp cards...
isapnp: Calling quirk for 01:00
isapnp: SB audio device quirk - increasing port range
isapnp: Card 'Creative ViBRA16C PnP'
isapnp: 1 Plug & Play card detected total
Linux NET4.0 for Linux 2.4
Based upon Swansea University Computer Society NET3.039
Starting kswapd v1.8
pty: 256 Unix98 ptys configured
block: queued sectors max/low 83960kB/27986kB, 256 slots per queue
Uniform Multi-Platform E-IDE driver Revision: 6.31
ide: Assuming 33MHz system bus speed for PIO modes; override with idebus=xx
PIIX4: IDE controller on PCI bus 00 dev 21
PIIX4: chipset revision 1
PIIX4: not 100% native mode: will probe irqs later
    ide0: BM-DMA at 0xd800-0xd807, BIOS settings: hda:DMA, hdb:DMA
    ide1: BM-DMA at 0xd808-0xd80f, BIOS settings: hdc:DMA, hdd:DMA
hda: WDC AC36400L, ATA DISK drive
hdb: CD-RW CRX100E, ATAPI CD/DVD-ROM drive
hdc: IBM-DJNA-371800, ATA DISK drive
hdd: HITACHI CDR-8335, ATAPI CD/DVD-ROM drive
ide0 at 0x1f0-0x1f7,0x3f6 on irq 14
ide1 at 0x170-0x177,0x376 on irq 15
hda: 12594960 sectors (6449 MB) w/256KiB Cache, CHS=784/255/63, UDMA(33)
hdc: 35239680 sectors (18043 MB) w/1966KiB Cache, CHS=34960/16/63, UDMA(33)
ide-cd: passing drive hdb to ide-scsi emulation.
hdd: ATAPI 24X CD-ROM drive, 128kB Cache
Uniform CD-ROM driver Revision: 3.12
Partition check:
 /dev/ide/host0/bus0/target0/lun0: p1 p2 p4
 /dev/ide/host0/bus1/target0/lun0: [PTBL] [2193/255/63] p1 p2 < p5 p6 p7 p8 p9 >
Floppy drive(s): fd0 is 1.44M
FDC 0 is a post-1991 82077
Serial driver version 5.02 (2000-08-09) with MANY_PORTS SHARE_IRQ SERIAL_PCI ISAPNP enabled
ttyS00 at 0x03f8 (irq = 4) is a 16550A
ttyS01 at 0x02f8 (irq = 3) is a 16550A
SCSI subsystem driver Revision: 1.00
scsi0 : SCSI host adapter emulation for IDE ATAPI devices
  Vendor: SONY      Model: CD-RW  CRX100E    Rev: 1.0m
  Type:   CD-ROM                             ANSI SCSI revision: 02
Detected scsi CD-ROM sr0 at scsi0, channel 0, id 0, lun 0
sr0: scsi3-mmc drive: 24x/24x writer cd/rw xa/form2 cdda tray
NET4: Linux TCP/IP 1.0 for NET4.0
IP Protocols: ICMP, UDP, TCP
IP: routing cache hash table of 512 buckets, 4Kbytes
TCP: Hash tables configured (established 8192 bind 8192)
NET4: Unix domain sockets 1.0/SMP for Linux NET4.0.
devfs: v0.102 (20000622) Richard Gooch (rgooch@atnf.csiro.au)
devfs: boot_options: 0x0
Start mounting filesystem: ide1(22,6)
Ending clean XFS mount for filesystem: ide1(22,6)
VFS: Mounted root (xfs filesystem) readonly.
Mounted devfs on /dev
Freeing unused kernel memory: 184k freed
Adding Swap: 88320k swap-space (priority -1)
XFS (dev: 22/7) mounting with kiobuf I/O
Start mounting filesystem: ide1(22,7)
Ending clean XFS mount for filesystem: ide1(22,7)
XFS (dev: 22/8) mounting with kiobuf I/O
Start mounting filesystem: ide1(22,8)
Ending clean XFS mount for filesystem: ide1(22,8)
Linux agpgart interface v0.99 (c) Jeff Hartmann
agpgart: Maximum main memory to use for agp memory: 94M
agpgart: Detected Intel 440LX chipset
agpgart: AGP aperture is 256M @ 0xd0000000

--------------020001040106000104080607
Content-Type: text/plain;
 name="warnings"
Content-Transfer-Encoding: 7bit
Content-Disposition: inline;
 filename="warnings"

Mar  2 00:49:55 neoblade kernel:  printing eip:
Mar  2 00:49:55 neoblade kernel: c0159039
Mar  2 00:49:55 neoblade kernel: Oops: 0002
Mar  2 00:49:55 neoblade kernel: CPU:    0
Mar  2 00:49:55 neoblade kernel: EIP:    0010:[pagebuf_write_full_page+221/304]
Mar  2 00:49:55 neoblade kernel: EIP:    0010:[<c0159039>]
Mar  2 00:49:55 neoblade kernel: EFLAGS: 00010286
Mar  2 00:49:55 neoblade kernel: eax: 00000000   ebx: 00000000   ecx: 000000df   edx: 00000339
Mar  2 00:49:55 neoblade kernel: esi: c7011800   edi: 00000000   ebp: c12168a8   esp: c1233f60
Mar  2 00:49:55 neoblade kernel: ds: 0018   es: 0018   ss: 0018
Mar  2 00:49:55 neoblade kernel: Process kupdate (pid: 6, stackpage=c1233000)
Mar  2 00:49:55 neoblade kernel: Stack: c12168a8 c015908c c70118a4 c70118ac 00000000 c7011800 00000000 c0159099 
Mar  2 00:49:55 neoblade kernel:        c12168a8 c12168a8 c015908c c70118a4 c01226b9 c12168a8 00000007 c7011800 
Mar  2 00:49:55 neoblade kernel:        c7f77638 c7f77600 00000003 c015908c c0141cab c70118a4 c1232000 00000000 
Mar  2 00:49:55 neoblade kernel: Call Trace: [pagebuf_write_full_page_unlock+0/92] [pagebuf_write_full_page_unlock+13/92] [pagebuf_write_full_page_unlock+0/92] [filemap_fdatasync+113/216] [pagebuf_write_full_page_unlock+0/92] [sync_inodes+255/384] [sync_old_buffers+14/60] 
Mar  2 00:49:55 neoblade kernel: Call Trace: [<c015908c>] [<c0159099>] [<c015908c>] [<c01226b9>] [<c015908c>] [<c0141cab>] [<c0132fd6>] 
Mar  2 00:49:55 neoblade kernel:        [kupdate+222/232] [kernel_thread+35/48] 
Mar  2 00:49:55 neoblade kernel:        [<c0133286>] [<c010744f>] 
Mar  2 00:49:55 neoblade kernel: 
Mar  2 00:49:55 neoblade kernel: Code: ff 40 10 8d 44 24 18 50 6a 01 6a 01 e8 c6 a1 06 00 8b 44 24 

--------------020001040106000104080607--


From owner-linux-xfs@oss.sgi.com Thu Mar  1 16:08:36 2001
Received:  by oss.sgi.com id <S553761AbRCBAI0>;
	Thu, 1 Mar 2001 16:08:26 -0800
Received: from sgi.SGI.COM ([192.48.153.1]:53613 "EHLO sgi.com")
	by oss.sgi.com with ESMTP id <S553711AbRCBAIH>;
	Thu, 1 Mar 2001 16:08:07 -0800
Received: from larry.melbourne.sgi.com ([134.14.52.130]) 
	by sgi.com (980327.SGI.8.8.8-aspam/980304.SGI-aspam:
       SGI does not authorize the use of its proprietary
       systems or networks for unsolicited or bulk email
       from the Internet.) 
	via SMTP id QAA07127
	for <linux-xfs@oss.sgi.com>; Thu, 1 Mar 2001 16:08:05 -0800 (PST)
	mail_from (nathans@wobbly.melbourne.sgi.com)
Received: from wobbly.melbourne.sgi.com (wobbly.melbourne.sgi.com [134.14.55.135]) by larry.melbourne.sgi.com (950413.SGI.8.6.12/950213.SGI.AUTOCF) via ESMTP id LAA15672; Fri, 2 Mar 2001 11:06:48 +1100
Received: (from nathans@localhost) by wobbly.melbourne.sgi.com (980427.SGI.8.8.8/980728.SGI.AUTOCF) id LAA77794; Fri, 2 Mar 2001 11:06:47 +1100 (EDT)
From:   "Nathan Scott" <nathans@wobbly.melbourne.sgi.com>
Message-Id: <10103021106.ZM182028@wobbly.melbourne.sgi.com>
Date:   Fri, 2 Mar 2001 11:06:46 -0400
In-Reply-To: james rich <james.rich@m.cc.utah.edu>
        "Re: building 2.4.2 (with XFS) fails" (Mar  1,  4:59pm)
References: <Pine.GSO.4.05.10103011655001.18915-100000@pipt.oz.cc.utah.edu>
X-Mailer: Z-Mail (3.2.3 08feb96 MediaMail)
To:     james rich <james.rich@m.cc.utah.edu>, linux-xfs@oss.sgi.com
Subject: Re: building 2.4.2 (with XFS) fails
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing

hi,

On Mar 1,  4:59pm, james rich wrote:
> Subject: Re: building 2.4.2 (with XFS) fails
> > ...
> > The empty file is probably created by the shell redirect in..
> > [drivers/char/Makefile, line 202]
> > defkeymap.c: defkeymap.map
> >         loadkeys --mktable defkeymap.map | sed -e 's/^static *//' > defkeymap.c
> 
> Can someone explain why defkeymap.c from XFS CVS needs loadkeys and the
> same file from stock 2.4 doesn't?  Does XFS 2.4 need to regenerate this
> file or something?  defkeymap.c from stock 2.4 is fine, what is different
> for XFS?
> 

>From a discussion with Keith the other day, I believe
this is because the XFS tree has kdb patched in and kdb
needs access to some of the (otherwise static) symbols
in defkeymap.c - Keith, correct me if I'm wrong.

cheers.

-- 
Nathan

From owner-linux-xfs@oss.sgi.com Thu Mar  1 16:31:16 2001
Received:  by oss.sgi.com id <S553766AbRCBAbG>;
	Thu, 1 Mar 2001 16:31:06 -0800
Received: from pneumatic-tube.sgi.com ([204.94.214.22]:24341 "EHLO
        pneumatic-tube.sgi.com") by oss.sgi.com with ESMTP
	id <S553762AbRCBAax>; Thu, 1 Mar 2001 16:30:53 -0800
Received: from madurai.engr.sgi.com (madurai.engr.sgi.com [163.154.5.75]) by pneumatic-tube.sgi.com (980327.SGI.8.8.8-aspam/980310.SGI-aspam) via ESMTP id QAA04291
	for <linux-xfs@oss.sgi.com>; Thu, 1 Mar 2001 16:40:31 -0800 (PST)
	mail_from (ananth@sgi.com)
Received: from sgi.com (mango.engr.sgi.com [163.154.5.76]) by madurai.engr.sgi.com (980427.SGI.8.8.8/970903.SGI.AUTOCF) via ESMTP id QAA01587; Thu, 1 Mar 2001 16:25:11 -0800 (PST)
Message-ID: <3A9EE937.EA684C76@sgi.com>
Date:   Thu, 01 Mar 2001 16:28:39 -0800
From:   Rajagopal Ananthanarayanan <ananth@sgi.com>
X-Mailer: Mozilla 4.72 [en] (X11; U; Linux 2.2.16-4SGI_20smp i686)
X-Accept-Language: en
MIME-Version: 1.0
To:     Joakim Bodin <joabo552@student.liu.se>
CC:     Linux-XFS Mailing List <linux-xfs@oss.sgi.com>
Subject: Re: Latest XFS kupdate oops
References: <3A9EE314.3020302@student.liu.se>
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing

Joakim Bodin wrote:
> 
> I updated xfs cvs just now (1 March 23:30 GMT) and just after rebooting
> kupdate segfaulted on me and resultet this attached Call Trace. I'm no
> kernel expert and I don't know if this sayes much. I could probably get
> more info about this if given the prober pointers. This is a up to date
> Linux-Mandrake cooker system and the kernel was compiled with RH7.0
> kgcc. I'm attaching my dmesg log as well.
> 


I strongly suspect you are getting a bh value of NULL here,
in fs/pagebuf/page_buf_io.c:

   1066         } else {
   1067                 __pb_block_commit_write_async(inode, page, NULL, 0);
   1068 
   1069                 /*
   1070                  * Kick-start that last write...
   1071                  */
   1072                 atomic_inc(&bh->b_count);
   1073                 ll_rw_block(WRITE, 1, &bh);
   1074                 atomic_dec(&bh->b_count);
   1075         }

Can you please put a debug line in here, just before the "Kick-start ..." comment?
	
		if (!bh || page->buffers != bh) {
			printk("Inconsistent page buffers 0x%p bh 0x%p\n"
				bh, page->buffers);
		}

thanks,

ananth.

--------------------------------------------------------------------------
Rajagopal Ananthanarayanan ("ananth")
Member Technical Staff, SGI.
--------------------------------------------------------------------------

From owner-linux-xfs@oss.sgi.com Thu Mar  1 17:06:46 2001
Received:  by oss.sgi.com id <S553747AbRCBBG0>;
	Thu, 1 Mar 2001 17:06:26 -0800
Received: from deliverator.sgi.com ([204.94.214.10]:16747 "EHLO
        deliverator.sgi.com") by oss.sgi.com with ESMTP id <S553711AbRCBBGK>;
	Thu, 1 Mar 2001 17:06:10 -0800
Received: from madurai.engr.sgi.com (madurai.engr.sgi.com [163.154.5.75]) by deliverator.sgi.com (980309.SGI.8.8.8-aspam-6.2/980310.SGI-aspam) via ESMTP id RAA26555
	for <linux-xfs@oss.sgi.com>; Thu, 1 Mar 2001 17:05:05 -0800 (PST)
	mail_from (ananth@sgi.com)
Received: from sgi.com (mango.engr.sgi.com [163.154.5.76]) by madurai.engr.sgi.com (980427.SGI.8.8.8/970903.SGI.AUTOCF) via ESMTP id RAA01564; Thu, 1 Mar 2001 17:00:21 -0800 (PST)
Message-ID: <3A9EF174.9318AB12@sgi.com>
Date:   Thu, 01 Mar 2001 17:03:48 -0800
From:   Rajagopal Ananthanarayanan <ananth@sgi.com>
X-Mailer: Mozilla 4.72 [en] (X11; U; Linux 2.2.16-4SGI_20smp i686)
X-Accept-Language: en
MIME-Version: 1.0
To:     Joakim Bodin <joabo552@student.liu.se>,
        Linux-XFS Mailing List <linux-xfs@oss.sgi.com>
Subject: Re: Latest XFS kupdate oops
References: <3A9EE314.3020302@student.liu.se> <3A9EE937.EA684C76@sgi.com>
Content-Type: multipart/mixed;
 boundary="------------82B603EB61DE4B14697795BA"
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing

This is a multi-part message in MIME format.
--------------82B603EB61DE4B14697795BA
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit


Here's a possible patch that fixes this oops.
There's one case if a page was read over a "hole"
then no buffers are attached to it. In this case,
buffers are only allocated as part of prepare_write.

ananth.

--------------------------------------------------------------------------
Rajagopal Ananthanarayanan ("ananth")
Member Technical Staff, SGI.
--------------------------------------------------------------------------
--------------82B603EB61DE4B14697795BA
Content-Type: text/plain; charset=us-ascii;
 name="sync.patch"
Content-Transfer-Encoding: 7bit
Content-Disposition: inline;
 filename="sync.patch"

--- ../../xfs-orig/linux/fs/pagebuf/page_buf_io.c	Thu Mar  1 16:54:40 2001
+++ fs/pagebuf/page_buf_io.c	Thu Mar  1 16:56:24 2001
@@ -1036,7 +1036,6 @@
 STATIC int pagebuf_write_full_page(struct page *page)
 {
 	struct inode *inode = (struct inode*)page->mapping->host;
-	struct buffer_head *bh = page->buffers;
 	unsigned long end_index = inode->i_size >> PAGE_CACHE_SHIFT;
 	loff_t offset;
 	int err, pb_flags;
@@ -1064,8 +1063,16 @@
 	if (err) {
 		ClearPageUptodate(page);
 	} else {
+		struct buffer_head *bh;
+
 		__pb_block_commit_write_async(inode, page, NULL, 0);
 		
+		/*
+		 * Pages over holes may have just been allocated
+		 * buffers, so bh needs to be looked up now.
+		 */
+		bh = page->buffers;
+
 		/*
 		 * Kick-start that last write...
 		 */

--------------82B603EB61DE4B14697795BA--


From owner-linux-xfs@oss.sgi.com Thu Mar  1 19:42:38 2001
Received:  by oss.sgi.com id <S553711AbRCBDmT>;
	Thu, 1 Mar 2001 19:42:19 -0800
Received: from pneumatic-tube.sgi.com ([204.94.214.22]:51491 "EHLO
        pneumatic-tube.sgi.com") by oss.sgi.com with ESMTP
	id <S553651AbRCBDmB>; Thu, 1 Mar 2001 19:42:01 -0800
Received: from cthulhu.engr.sgi.com (gate3-relay.engr.sgi.com [130.62.1.234]) by pneumatic-tube.sgi.com (980327.SGI.8.8.8-aspam/980310.SGI-aspam) via ESMTP id TAA07139
	for <linux-xfs@oss.sgi.com>; Thu, 1 Mar 2001 19:51:39 -0800 (PST)
	mail_from (tduffy@engr.sgi.com)
Received: from dbear.engr.sgi.com (dbear.engr.sgi.com [163.154.18.85])
	by cthulhu.engr.sgi.com (SGI-8.9.3/8.9.3) with ESMTP id TAA57601;
	Thu, 1 Mar 2001 19:40:41 -0800 (PST)
Date:   Thu, 1 Mar 2001 19:37:32 -0800 (PST)
From:   Tom Duffy <tduffy@engr.sgi.com>
To:     Eric Sandeen <sandeen@sgi.com>
cc:     Chris Croswhite <ccroswhite@get2chip.com>,
        Aldert Zomer <zomeral@biol.rug.nl>,
        <kenneth.leung@syngenta.com>, <linux-xfs@oss.sgi.com>
Subject: Re: Usable memory limit?
In-Reply-To: <3A9EA80C.36E9521C@sgi.com>
Message-ID: <Pine.LNX.4.30.0103011936190.2555-100000@dbear.engr.sgi.com>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing

redhat recently turned CONFIG_HIGHMEM4G on in all their i686 kernels.  at
least in kernel-2.4.1-0.1.14.  still on on in i386, i586, or athlon.

-tduffy

On Thu, 1 Mar 2001, Eric Sandeen wrote:

> Yep, the only x86 kernel on the XFS-PR0.9 CD with HIGHMEM configured is
> the i686-enterprise kernel:
>
> kernel-2.4.0-i686-enterprise.config:CONFIG_HIGHMEM64G=y
> kernel-2.4.0-i686-enterprise.config:CONFIG_HIGHMEM=y
>
> There are other differences in the enterprise kernel as well, of course
> - although I'm not sure what all, offhand...  the XFS kernel RPMs follow
> Red Hat's configs very closely.
>
> -Eric
>
> Chris Croswhite wrote:
> >
> > I say this exact behavior on a test machine of RedHat Linux 7.x.
> >
> > If it is a Red Hat system 7.0, RH has 3 boot modes:
> >
> > Linux
> > Linux-smp
> > Linux-up
> >
> > The linux kernel has high mem support as well as SMP(in RH terms, the
> > 2.2.16-20-enterprise kernel).  The SMP kernel has no high mem support and linux
> > up has no builtin modules (well some, but less than the other 2 kernel
> > versions.).
>


From owner-linux-xfs@oss.sgi.com Thu Mar  1 21:02:58 2001
Received:  by oss.sgi.com id <S553744AbRCBFCi>;
	Thu, 1 Mar 2001 21:02:38 -0800
Received: from yoop.oz.cc.utah.edu ([155.99.2.5]:10900 "EHLO
        yoop.oz.cc.utah.edu") by oss.sgi.com with ESMTP id <S553651AbRCBFCY>;
	Thu, 1 Mar 2001 21:02:24 -0800
Received: from localhost (jdr1529@localhost)
	by yoop.oz.cc.utah.edu (8.9.2/8.9.2) with ESMTP id WAA00932
	for <linux-xfs@oss.sgi.com>; Thu, 1 Mar 2001 22:02:12 -0700 (MST)
Date:   Thu, 1 Mar 2001 22:02:11 -0700 (MST)
From:   james rich <james.rich@m.cc.utah.edu>
To:     linux-xfs@oss.sgi.com
Subject: Re: building 2.4.2 (with XFS) fails
In-Reply-To: <10103021106.ZM182028@wobbly.melbourne.sgi.com>
Message-ID: <Pine.GSO.4.05.10103012200070.768-100000@yoop.oz.cc.utah.edu>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing

Anyone know where I can find loadkeys then?  I search on freshmeat turns
up nothing.  Preferably source code over RPM or DEB (though a slackware
.tgz wouldn't be too bad).

James Rich
james.rich@m.cc.utah.edu


From owner-linux-xfs@oss.sgi.com Thu Mar  1 21:11:07 2001
Received:  by oss.sgi.com id <S553711AbRCBFK6>;
	Thu, 1 Mar 2001 21:10:58 -0800
Received: from deliverator.sgi.com ([204.94.214.10]:54055 "EHLO
        deliverator.sgi.com") by oss.sgi.com with ESMTP id <S553651AbRCBFKs>;
	Thu, 1 Mar 2001 21:10:48 -0800
Received: from larry.melbourne.sgi.com (larry.melbourne.sgi.com [134.14.52.130]) by deliverator.sgi.com (980309.SGI.8.8.8-aspam-6.2/980310.SGI-aspam) via SMTP id VAA24414
	for <linux-xfs@oss.sgi.com>; Thu, 1 Mar 2001 21:09:38 -0800 (PST)
	mail_from (nathans@wobbly.melbourne.sgi.com)
Received: from wobbly.melbourne.sgi.com (wobbly.melbourne.sgi.com [134.14.55.135]) by larry.melbourne.sgi.com (950413.SGI.8.6.12/950213.SGI.AUTOCF) via ESMTP id QAA17972; Fri, 2 Mar 2001 16:09:25 +1100
Received: (from nathans@localhost) by wobbly.melbourne.sgi.com (980427.SGI.8.8.8/980728.SGI.AUTOCF) id QAA77865; Fri, 2 Mar 2001 16:09:23 +1100 (EDT)
From:   "Nathan Scott" <nathans@wobbly.melbourne.sgi.com>
Message-Id: <10103021609.ZM167815@wobbly.melbourne.sgi.com>
Date:   Fri, 2 Mar 2001 16:09:22 -0400
In-Reply-To: james rich <james.rich@m.cc.utah.edu>
        "Re: building 2.4.2 (with XFS) fails" (Mar  1, 10:02pm)
References: <Pine.GSO.4.05.10103012200070.768-100000@yoop.oz.cc.utah.edu>
X-Mailer: Z-Mail (3.2.3 08feb96 MediaMail)
To:     james rich <james.rich@m.cc.utah.edu>, linux-xfs@oss.sgi.com
Subject: Re: building 2.4.2 (with XFS) fails
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing

hi James,

On Mar 1, 10:02pm, james rich wrote:
> Subject: Re: building 2.4.2 (with XFS) fails
> Anyone know where I can find loadkeys then?  I search on freshmeat turns
> up nothing.  Preferably source code over RPM or DEB (though a slackware
> .tgz wouldn't be too bad).
> 

Try the console-tools package:  http://lct.sourceforge.net/
and download at:  ftp://lct.sourceforge.net/pub/lct/

cheers.

-- 
Nathan

From owner-linux-xfs@oss.sgi.com Thu Mar  1 21:12:18 2001
Received:  by oss.sgi.com id <S553711AbRCBFMJ>;
	Thu, 1 Mar 2001 21:12:09 -0800
Received: from sgi.SGI.COM ([192.48.153.1]:51546 "EHLO sgi.com")
	by oss.sgi.com with ESMTP id <S553651AbRCBFMG>;
	Thu, 1 Mar 2001 21:12:06 -0800
Received: from sydney.sydney.sgi.com ([134.14.48.2]) 
	by sgi.com (980327.SGI.8.8.8-aspam/980304.SGI-aspam:
       SGI does not authorize the use of its proprietary
       systems or networks for unsolicited or bulk email
       from the Internet.) 
	via SMTP id VAA07277
	for <linux-xfs@oss.sgi.com>; Thu, 1 Mar 2001 21:12:04 -0800 (PST)
	mail_from (kaos@melbourne.sgi.com)
Received: from kao2.melbourne.sgi.com by sydney.sydney.sgi.com via ESMTP (950413.SGI.8.6.12/930416.SGI)
	 id QAA26855; Fri, 2 Mar 2001 16:10:43 +1100
X-Mailer: exmh version 2.1.1 10/15/1999
From:   Keith Owens <kaos@melbourne.sgi.com>
To:     james rich <james.rich@m.cc.utah.edu>
cc:     linux-xfs@oss.sgi.com
Subject: Re: building 2.4.2 (with XFS) fails 
In-reply-to: Your message of "Thu, 01 Mar 2001 22:02:11 PDT."
             <Pine.GSO.4.05.10103012200070.768-100000@yoop.oz.cc.utah.edu> 
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Date:   Fri, 02 Mar 2001 16:10:42 +1100
Message-ID: <3046.983509842@kao2.melbourne.sgi.com>
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing

On Thu, 1 Mar 2001 22:02:11 -0700 (MST), 
james rich <james.rich@m.cc.utah.edu> wrote:
>Anyone know where I can find loadkeys then?  I search on freshmeat turns
>up nothing.  Preferably source code over RPM or DEB (though a slackware
>.tgz wouldn't be too bad).

# rpm -qf $(which loadkeys)
console-tools-19990302-17


From owner-linux-xfs@oss.sgi.com Thu Mar  1 23:49:48 2001
Received:  by oss.sgi.com id <S553747AbRCBHt2>;
	Thu, 1 Mar 2001 23:49:28 -0800
Received: from hermes.mixx.net ([212.84.196.2]:26129 "HELO hermes.mixx.net")
	by oss.sgi.com with SMTP id <S553733AbRCBHtL>;
	Thu, 1 Mar 2001 23:49:11 -0800
Received: from mate.bln.innominate.de (cerberus.berlin.innominate.de [212.84.234.251])
	by hermes.mixx.net (Postfix) with ESMTP id 3D800F80B
	for <linux-xfs@oss.sgi.com>; Fri,  2 Mar 2001 08:49:09 +0100 (CET)
Received: by mate.bln.innominate.de (Postfix, from userid 9)
	id 65DA02CA6F; Fri,  2 Mar 2001 08:49:08 +0100 (CET)
From:   Thomas Graichen <news-innominate.list.sgi.xfs@innominate.de>
Reply-To: Thomas Graichen <thomas.graichen@innominate.com>
X-Newsgroups: innominate.list.sgi.xfs
Subject: Re: 2GB filesize limit.
Date:   2 Mar 2001 07:49:08 GMT
Organization: innominate AG, Berlin, Germany
Lines:  16
Distribution: local
Message-ID: <news2mail-97nj9k$e2f$1@mate.bln.innominate.de>
References: <Pine.BSF.4.31.0103011113230.40049-100000@opus.sandiegoca.ncr.com>
X-Trace: mate.bln.innominate.de 983519348 14415 10.0.0.31 (2 Mar 2001 07:49:08 GMT)
X-Complaints-To: news@innominate.de
User-Agent: tin/1.4.4-20000803 ("Vet for the Insane") (UNIX) (Linux/2.4.2-XFS (i686))
To:     linux-xfs@oss.sgi.com
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing

Chuck Rouillard <chuckr@opus.sandiegoca.ncr.com> wrote:

> I have taken a look at the SGI XFS FAQ and sifted through
> the mail headers in the 'Mail Thread Index' to no avail.

just fyi i plan to update the FAQ next week (then i should have some
time for this) ... i hope it will answer more of the now common
questions then

t

-- 
thomas.graichen@innominate.com
                                                         innominate AG
                                                  the linux architects
tel: +49-30-308806-13   fax: -77             http://www.innominate.com

From owner-linux-xfs@oss.sgi.com Thu Mar  1 23:57:27 2001
Received:  by oss.sgi.com id <S553747AbRCBH5S>;
	Thu, 1 Mar 2001 23:57:18 -0800
Received: from hermes.mixx.net ([212.84.196.2]:46353 "HELO hermes.mixx.net")
	by oss.sgi.com with SMTP id <S553733AbRCBH5F>;
	Thu, 1 Mar 2001 23:57:05 -0800
Received: from mate.bln.innominate.de (cerberus.berlin.innominate.de [212.84.234.251])
	by hermes.mixx.net (Postfix) with ESMTP id 2FEB6F80B
	for <linux-xfs@oss.sgi.com>; Fri,  2 Mar 2001 08:57:03 +0100 (CET)
Received: by mate.bln.innominate.de (Postfix, from userid 9)
	id 0D9922CA6F; Fri,  2 Mar 2001 08:57:03 +0100 (CET)
From:   Thomas Graichen <news-innominate.list.sgi.xfs@innominate.de>
Reply-To: Thomas Graichen <thomas.graichen@innominate.com>
X-Newsgroups: innominate.list.sgi.xfs
Subject: kio on non-x86
Date:   2 Mar 2001 07:57:02 GMT
Organization: innominate AG, Berlin, Germany
Lines:  26
Distribution: local
Message-ID: <news2mail-97njoe$e2f$2@mate.bln.innominate.de>
X-Trace: mate.bln.innominate.de 983519822 14415 10.0.0.31 (2 Mar 2001 07:57:02 GMT)
X-Complaints-To: news@innominate.de
User-Agent: tin/1.4.4-20000803 ("Vet for the Insane") (UNIX) (Linux/2.4.2-XFS (i686))
To:     linux-xfs@oss.sgi.com
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing

just updated the kernel on my ppc machine yesterday - worked all fine
so far ... that i tried the kio and kiocluster options on the ppc too
and ran into problems: as soon as i write to the fs it detects an io
error and shuts down ... thus a few questions:

* am i right in assuming that the kio stuff does not contain any
  arch specific parts so it should - in principle - work also on
  the pcc which would leave endian or gcc issues as a possible
  reason for the problem?

* what is the state of XFS on sparc (also big endian)? - if it
  works there: do the kio and kiocluster options work on sparc?

* does anyone have an idea off-mind what might be the reson for
  this? ... i'll turn on all the debugging options and see if i
  get any more information out of it this evening i think

a lot of thanks in advance

t

-- 
thomas.graichen@innominate.com
                                                         innominate AG
                                                  the linux architects
tel: +49-30-308806-13   fax: -77             http://www.innominate.com

From owner-linux-xfs@oss.sgi.com Fri Mar  2 01:04:08 2001
Received:  by oss.sgi.com id <S553766AbRCBJD6>;
	Fri, 2 Mar 2001 01:03:58 -0800
Received: from hermes.mixx.net ([212.84.196.2]:13828 "HELO hermes.mixx.net")
	by oss.sgi.com with SMTP id <S553756AbRCBJDn>;
	Fri, 2 Mar 2001 01:03:43 -0800
Received: from mate.bln.innominate.de (cerberus.berlin.innominate.de [212.84.234.251])
	by hermes.mixx.net (Postfix) with ESMTP id 3D82AF803
	for <linux-xfs@oss.sgi.com>; Fri,  2 Mar 2001 10:03:41 +0100 (CET)
Received: by mate.bln.innominate.de (Postfix, from userid 9)
	id 16F212CA6F; Fri,  2 Mar 2001 10:03:41 +0100 (CET)
From:   Thomas Graichen <news-innominate.list.sgi.xfs@innominate.de>
Reply-To: Thomas Graichen <thomas.graichen@innominate.com>
X-Newsgroups: innominate.list.sgi.xfs
Subject: Re: TAKE - speed up shutdown with xfs root
Date:   2 Mar 2001 09:03:40 GMT
Organization: innominate AG, Berlin, Germany
Lines:  14
Distribution: local
Message-ID: <news2mail-97nnlc$ses$1@mate.bln.innominate.de>
References: <200103011828.f21IS2404463@jen.americas.sgi.com>
X-Trace: mate.bln.innominate.de 983523820 29148 10.0.0.31 (2 Mar 2001 09:03:40 GMT)
X-Complaints-To: news@innominate.de
User-Agent: tin/1.4.4-20000803 ("Vet for the Insane") (UNIX) (Linux/2.4.2-XFS (i686))
To:     linux-xfs@oss.sgi.com
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing

that sounds great ... no more "endless" shutdowns

t

Steve Lord <lord@sgi.com> wrote:
> Found a common case where remounting the root filesystem readonly would take
> a long time (15 to 30 seconds). This gets it down to fractions of a second
> instead.

-- 
thomas.graichen@innominate.com
                                                         innominate AG
                                                  the linux architects
tel: +49-30-308806-13   fax: -77             http://www.innominate.com

From owner-linux-xfs@oss.sgi.com Fri Mar  2 03:22:39 2001
Received:  by oss.sgi.com id <S553783AbRCBLWa>;
	Fri, 2 Mar 2001 03:22:30 -0800
Received: from h217n3fls20o974.telia.com ([212.181.166.217]:36614 "HELO
        neoblade.net") by oss.sgi.com with SMTP id <S553773AbRCBLWK>;
	Fri, 2 Mar 2001 03:22:10 -0800
Received: from student.liu.se (localhost.localdomain [127.0.0.1])
	by neoblade.net (Postfix) with ESMTP
	id 2C268188D63; Fri,  2 Mar 2001 12:24:01 +0100 (CET)
Message-ID: <3A9F82C6.8050003@student.liu.se>
Date:   Fri, 02 Mar 2001 12:23:50 +0100
From:   Joakim Bodin <joabo552@student.liu.se>
User-Agent: Mozilla/5.0 (X11; U; Linux 2.4.2-XFS i686; en-US; 0.8) Gecko/20010219
X-Accept-Language: sv, en
MIME-Version: 1.0
To:     Rajagopal Ananthanarayanan <ananth@sgi.com>
Cc:     Linux-XFS Mailing List <linux-xfs@oss.sgi.com>
Subject: Re: Latest XFS kupdate oops
References: <3A9EE314.3020302@student.liu.se> <3A9EE937.EA684C76@sgi.com> <3A9EF174.9318AB12@sgi.com>
Content-Type: text/plain; charset=us-ascii; format=flowed
Content-Transfer-Encoding: 7bit
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing

Rajagopal Ananthanarayanan wrote:

> Here's a possible patch that fixes this oops.
> There's one case if a page was read over a "hole"
> then no buffers are attached to it. In this case,
> buffers are only allocated as part of prepare_write.
> 
> ananth.
> 
> --------------------------------------------------------------------------
> Rajagopal Ananthanarayanan ("ananth")
> Member Technical Staff, SGI.
> --------------------------------------------------------------------------

> 
I applied this patch and the machine seems to have survived the night 
just fine with a mozilla compile and slocates etc.

Joakim Bodin


From owner-linux-xfs@oss.sgi.com Fri Mar  2 06:08:21 2001
Received:  by oss.sgi.com id <S553738AbRCBOIB>;
	Fri, 2 Mar 2001 06:08:01 -0800
Received: from south.orl-pub.theseus.com ([12.108.42.66]:60440 "EHLO
        thor.theseus.com") by oss.sgi.com with ESMTP id <S553705AbRCBOHr>;
	Fri, 2 Mar 2001 06:07:47 -0800
Received: from theseus.com (IDENT:thebs@fugitive.theseus.com [192.168.0.242])
	by thor.theseus.com (8.9.3/8.9.3) with ESMTP id JAA11919;
	Fri, 2 Mar 2001 09:11:09 -0500
Message-ID: <3A9FA94C.D3CAA4AF@theseus.com>
Date:   Fri, 02 Mar 2001 09:08:12 -0500
From:   Bryan-TheBS-Smith <thebs@theseus.com>
Reply-To: thebs@theseus.com, b.j.smith@ieee.org
Organization: (Personal)
X-Mailer: Mozilla 4.76 [en] (X11; U; Linux 2.2.17-0.17-FUGITIVE i686)
X-Accept-Language: en
MIME-Version: 1.0
To:     Chuck Rouillard <chuckr@opus.sandiegoca.ncr.com>
CC:     Linux - XFS <linux-xfs@oss.sgi.com>
Subject: Re: 2GB filesize limit.
References: <Pine.BSF.4.31.0103011113230.40049-100000@opus.sandiegoca.ncr.com>
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing

Chuck Rouillard wrote:
> 1. Does Linux/XFS break the 2G filesize limit (on the
>    assumption that certain support libraries exist and
>    correctly configured)?

XFS itself breaks the 2GB filesize limit.  I will assume, for a
moment, that the VFS in 2.4 has been changed to accomodate this with
LFS?

>    The linux-xfs.sg.com/projects/xfs/ page mentions
>    "(no more 2GB limits)" but it isn't clear what exactly
>    that refers to, nor if that claim is valid with Linux.

Good question.  I guess I need to dig into the source of XFS and
2.4's VFS to find out for sure.

As far as kernel 2.2, I seriously doubt it works without some
serious modification.  I got LFS to work on 2.2 kernels with Ext3
and NFS, but it took additional patches to not only the kernel, but
Ext3 and NFS parts of the kernel.

> 2. If the above is true, do userland tools such as `ls',
>    `cp', `mv', et cetera, need mending?

That's where GLibC comes in.

E.g., my RedHat 6.2 systems have only GLibC 2.1, so not really.  But
I can boot a 2.2 kernel with Ext3 and NFS, with LFS patches for all,
and my 64-bit Solaris/SPARC systems be able to write >2GB files to
it.  I can also back it up because I'm using "dump/restore"
(although tar/cpio/bru/etc... might have isues).

> 3. Has XFS support on Linux/IA-64 reached the mumbled
>    stage?

No idea.  Sorry.

-- TheBS


--
Bryan "TheBS" Smith          chat:thebs413 @AOL/MSN/Yahoo
Engineer      mailto:b.j.smith@ieee.org,thebs@theseus.com
*********************************************************
"Never apply a Star Trek solution to a Babylon 5 problem"
                                    -- Nicholas C. Weaver

From owner-linux-xfs@oss.sgi.com Fri Mar  2 06:10:51 2001
Received:  by oss.sgi.com id <S553773AbRCBOKc>;
	Fri, 2 Mar 2001 06:10:32 -0800
Received: from south.orl-pub.theseus.com ([12.108.42.66]:1561 "EHLO
        thor.theseus.com") by oss.sgi.com with ESMTP id <S553721AbRCBOKU>;
	Fri, 2 Mar 2001 06:10:20 -0800
Received: from theseus.com (IDENT:thebs@fugitive.theseus.com [192.168.0.242])
	by thor.theseus.com (8.9.3/8.9.3) with ESMTP id JAA12002;
	Fri, 2 Mar 2001 09:13:49 -0500
Message-ID: <3A9FA9EC.82E48C46@theseus.com>
Date:   Fri, 02 Mar 2001 09:10:52 -0500
From:   Bryan-TheBS-Smith <thebs@theseus.com>
Reply-To: thebs@theseus.com, b.j.smith@ieee.org
Organization: (Personal)
X-Mailer: Mozilla 4.76 [en] (X11; U; Linux 2.2.17-0.17-FUGITIVE i686)
X-Accept-Language: en
MIME-Version: 1.0
To:     Chuck Rouillard <chuckr@opus.sandiegoca.ncr.com>
CC:     Linux - XFS <linux-xfs@oss.sgi.com>
Subject: Re: 2GB filesize limit.
References: <Pine.BSF.4.31.0103011113230.40049-100000@opus.sandiegoca.ncr.com> <3A9FA94C.D3CAA4AF@theseus.com>
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing

Bryan-TheBS-Smith wrote:
> XFS itself breaks the 2GB filesize limit.  I will assume, for a
> moment, that the VFS in 2.4 has been changed to accomodate this with
> LFS?

Let me refraze that, I'm "assuming" that the VFS in 2.4 has been
changed so you can simply use 64-bit filesystems, like XFS, on
32-bit platforms and the LFS in the VFS (again, in kernel 2.4)
handles it all correctly.

-- TheBS

--
Bryan "TheBS" Smith          chat:thebs413 @AOL/MSN/Yahoo
Engineer      mailto:b.j.smith@ieee.org,thebs@theseus.com
*********************************************************
"Never apply a Star Trek solution to a Babylon 5 problem"
                                    -- Nicholas C. Weaver

From owner-linux-xfs@oss.sgi.com Fri Mar  2 06:32:21 2001
Received:  by oss.sgi.com id <S553773AbRCBOcB>;
	Fri, 2 Mar 2001 06:32:01 -0800
Received: from deliverator.sgi.com ([204.94.214.10]:6716 "EHLO
        deliverator.sgi.com") by oss.sgi.com with ESMTP id <S553721AbRCBObq>;
	Fri, 2 Mar 2001 06:31:46 -0800
Received: from zeus-fddi.americas.sgi.com (zeus-fddi.americas.sgi.com [128.162.8.103]) by deliverator.sgi.com (980309.SGI.8.8.8-aspam-6.2/980310.SGI-aspam) via ESMTP id GAA05224
	for <linux-xfs@oss.sgi.com>; Fri, 2 Mar 2001 06:30:40 -0800 (PST)
	mail_from (lord@sgi.com)
Received: from daisy-e185.americas.sgi.com (daisy-e194.americas.sgi.com [128.162.194.214]) by zeus-fddi.americas.sgi.com (8.9.3/americas-smart-nospam1.1) with ESMTP id IAA989635; Fri, 2 Mar 2001 08:30:28 -0600 (CST)
Received: from jen.americas.sgi.com (IDENT:root@jen.americas.sgi.com [128.162.184.86]) by daisy-e185.americas.sgi.com (SGI-8.9.3/SGI-server-1.7) with ESMTP id IAA43526; Fri, 2 Mar 2001 08:30:28 -0600 (CST)
Received: from jen.americas.sgi.com by jen.americas.sgi.com (8.11.0/SGI-client-1.7) via ESMTP id f22EUL929467; Fri, 2 Mar 2001 08:30:21 -0600
Message-Id: <200103021430.f22EUL929467@jen.americas.sgi.com>
X-Mailer: exmh version 2.2 06/23/2000 with nmh-1.0.4
To:     Thomas Graichen <thomas.graichen@innominate.com>
cc:     linux-xfs@oss.sgi.com
Subject: Re: kio on non-x86 
In-Reply-To: Message from Thomas Graichen <news-innominate.list.sgi.xfs@innominate.de> 
   of "02 Mar 2001 07:57:02 GMT." <news2mail-97njoe$e2f$2@mate.bln.innominate.de> 
Date:   Fri, 02 Mar 2001 08:30:21 -0600
From:   Steve Lord <lord@sgi.com>
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing

> just updated the kernel on my ppc machine yesterday - worked all fine
> so far ... that i tried the kio and kiocluster options on the ppc too
> and ran into problems: as soon as i write to the fs it detects an io
> error and shuts down ... thus a few questions:

I presume the same devices worked OK without kio, did they work OK after
this shutdown?

> 
> * am i right in assuming that the kio stuff does not contain any
>   arch specific parts so it should - in principle - work also on
>   the pcc which would leave endian or gcc issues as a possible
>   reason for the problem?

I think this is true, there are no os specific components, where you using
ide or scsi? The ide kio code is specific to having dma turned on. So which
combination was this?

> 
> * what is the state of XFS on sparc (also big endian)? - if it
>   works there: do the kio and kiocluster options work on sparc?

I would have to leave this to the list to answer - and by the way, there
is no difference between these two options now, kiocluster will be going
away I suspect.

> 
> * does anyone have an idea off-mind what might be the reson for
>   this? ... i'll turn on all the debugging options and see if i
>   get any more information out of it this evening i think

Chait may have some suggestions, but the specific hardware combination is
the most interesting part of the equation right now.

Steve

> 
> a lot of thanks in advance
> 
> t
> 
> -- 
> thomas.graichen@innominate.com
>                                                          innominate AG
>                                                   the linux architects
> tel: +49-30-308806-13   fax: -77             http://www.innominate.com



From owner-linux-xfs@oss.sgi.com Fri Mar  2 07:06:31 2001
Received:  by oss.sgi.com id <S553773AbRCBPGV>;
	Fri, 2 Mar 2001 07:06:21 -0800
Received: from hermes.mixx.net ([212.84.196.2]:6930 "HELO hermes.mixx.net")
	by oss.sgi.com with SMTP id <S553721AbRCBPGN>;
	Fri, 2 Mar 2001 07:06:13 -0800
Received: from mate.bln.innominate.de (cerberus.berlin.innominate.de [212.84.234.251])
	by hermes.mixx.net (Postfix) with ESMTP id 306F8F80B
	for <linux-xfs@oss.sgi.com>; Fri,  2 Mar 2001 16:06:10 +0100 (CET)
Received: by mate.bln.innominate.de (Postfix, from userid 9)
	id EB1962CA71; Fri,  2 Mar 2001 16:06:09 +0100 (CET)
From:   Thomas Graichen <news-innominate.list.sgi.xfs@innominate.de>
Reply-To: Thomas Graichen <thomas.graichen@innominate.com>
X-Newsgroups: innominate.list.sgi.xfs
Subject: Re: kio on non-x86
Date:   2 Mar 2001 15:06:09 GMT
Organization: innominate AG, Berlin, Germany
Lines:  54
Distribution: local
Message-ID: <news2mail-97oct1$d7c$1@mate.bln.innominate.de>
References: <news-innominate.list.sgi.xfs@innominate.de> <200103021430.f22EUL929467@jen.americas.sgi.com>
X-Trace: mate.bln.innominate.de 983545569 13548 10.0.0.31 (2 Mar 2001 15:06:09 GMT)
X-Complaints-To: news@innominate.de
User-Agent: tin/1.4.4-20000803 ("Vet for the Insane") (UNIX) (Linux/2.4.2-XFS (i686))
To:     linux-xfs@oss.sgi.com
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing

Steve Lord <lord@sgi.com> wrote:
>> just updated the kernel on my ppc machine yesterday - worked all fine
>> so far ... that i tried the kio and kiocluster options on the ppc too
>> and ran into problems: as soon as i write to the fs it detects an io
>> error and shuts down ... thus a few questions:

> I presume the same devices worked OK without kio, did they work OK after
> this shutdown?

just did very short testings (will do more this evening) but i was
able to umount/mount them afterwards and xfs_repair was close to
happy with them - but more details later

without kio XFS on ppc seems to be as stable as on i386 due to my
testings (i am runing fine XFS root - never had any problems so
far)

>> * am i right in assuming that the kio stuff does not contain any
>>   arch specific parts so it should - in principle - work also on
>>   the pcc which would leave endian or gcc issues as a possible
>>   reason for the problem?

> I think this is true, there are no os specific components, where you using
> ide or scsi? The ide kio code is specific to having dma turned on. So which
> combination was this?

it's ide and dma turned on

>> * what is the state of XFS on sparc (also big endian)? - if it
>>   works there: do the kio and kiocluster options work on sparc?

> I would have to leave this to the list to answer - and by the way, there
> is no difference between these two options now, kiocluster will be going
> away I suspect.

yes i was remembering this somehow from the list - ok - it's the
same behaviour for both anyway :-)

>> * does anyone have an idea off-mind what might be the reson for
>>   this? ... i'll turn on all the debugging options and see if i
>>   get any more information out of it this evening i think

> Chait may have some suggestions, but the specific hardware combination is
> the most interesting part of the equation right now.

thanks - as said i waill (and can) do more testing this evening

t

-- 
thomas.graichen@innominate.com
                                                         innominate AG
                                                  the linux architects
tel: +49-30-308806-13   fax: -77             http://www.innominate.com

From owner-linux-xfs@oss.sgi.com Fri Mar  2 08:43:52 2001
Received:  by oss.sgi.com id <S553798AbRCBQnm>;
	Fri, 2 Mar 2001 08:43:42 -0800
Received: from sgi.SGI.COM ([192.48.153.1]:38729 "EHLO sgi.com")
	by oss.sgi.com with ESMTP id <S553788AbRCBQng>;
	Fri, 2 Mar 2001 08:43:36 -0800
Received: from cthulhu.engr.sgi.com (cthulhu.engr.sgi.com [192.26.80.2]) 
	by sgi.com (980327.SGI.8.8.8-aspam/980304.SGI-aspam:
       SGI does not authorize the use of its proprietary
       systems or networks for unsolicited or bulk email
       from the Internet.) 
	via ESMTP id IAA09328
	for <linux-xfs@oss.sgi.com>; Fri, 2 Mar 2001 08:43:33 -0800 (PST)
	mail_from (chait@getafix.engr.sgi.com)
Received: from getafix.engr.sgi.com (IDENT:root@getafix.engr.sgi.com [163.154.5.110])
	by cthulhu.engr.sgi.com (SGI-8.9.3/8.9.3) with ESMTP id IAA39427;
	Fri, 2 Mar 2001 08:42:17 -0800 (PST)
Received: from localhost (chait@localhost)
	by getafix.engr.sgi.com (8.9.3/8.9.3) with ESMTP id IAA20622;
	Fri, 2 Mar 2001 08:38:19 -0500
Date:   Fri, 2 Mar 2001 08:38:18 -0500 (EST)
From:   Chaitanya Tumuluri <chait@getafix.engr.sgi.com>
To:     Thomas Graichen <thomas.graichen@innominate.com>
cc:     linux-xfs@oss.sgi.com
Subject: Re: kio on non-x86
In-Reply-To: <news2mail-97oct1$d7c$1@mate.bln.innominate.de>
Message-ID: <Pine.LNX.4.21.0103020830520.20445-100000@getafix.engr.sgi.com>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing

On 2 Mar 2001, Thomas Graichen wrote:
> Steve Lord <lord@sgi.com> wrote:
> >> just updated the kernel on my ppc machine yesterday - worked all fine
> >> so far ... that i tried the kio and kiocluster options on the ppc too
> >> and ran into problems: as soon as i write to the fs it detects an io
> >> error and shuts down ... thus a few questions:
> 
> > I presume the same devices worked OK without kio, did they work OK after
> > this shutdown?

Hi,

Do you have the error messages? 
AFAIK, kiobuf code is _not_ arch specific.

> >> * am i right in assuming that the kio stuff does not contain any
> >>   arch specific parts so it should - in principle - work also on
> >>   the pcc which would leave endian or gcc issues as a possible
> >>   reason for the problem?
> 
> > I think this is true, there are no os specific components, where you using
> > ide or scsi? The ide kio code is specific to having dma turned on. So which
> > combination was this?
> 
> it's ide and dma turned on

There are no arch specific components to the kiobuf code. Though, if you
are using more than 1GB there may be an issue with highmem bouncing code
in kiobuf I/O. How much memory do you have on your system?

> >> * does anyone have an idea off-mind what might be the reson for
> >>   this? ... i'll turn on all the debugging options and see if i
> >>   get any more information out of it this evening i think
> 
> > Chait may have some suggestions, but the specific hardware combination is
> > the most interesting part of the equation right now.
> 
> thanks - as said i waill (and can) do more testing this evening

Error messages, if any, would help....I'll wait for the results from
your tests.

Cheers,
-Chait. 


From owner-linux-xfs@oss.sgi.com Fri Mar  2 08:56:02 2001
Received:  by oss.sgi.com id <S553796AbRCBQzw>;
	Fri, 2 Mar 2001 08:55:52 -0800
Received: from hermes.mixx.net ([212.84.196.2]:50182 "HELO hermes.mixx.net")
	by oss.sgi.com with SMTP id <S553787AbRCBQzp>;
	Fri, 2 Mar 2001 08:55:45 -0800
Received: from mate.bln.innominate.de (cerberus.berlin.innominate.de [212.84.234.251])
	by hermes.mixx.net (Postfix) with ESMTP id D8997F80B
	for <linux-xfs@oss.sgi.com>; Fri,  2 Mar 2001 17:55:43 +0100 (CET)
Received: by mate.bln.innominate.de (Postfix, from userid 9)
	id 7E67E2CA6F; Fri,  2 Mar 2001 17:55:43 +0100 (CET)
From:   Thomas Graichen <news-innominate.list.sgi.xfs@innominate.de>
Reply-To: Thomas Graichen <thomas.graichen@innominate.com>
X-Newsgroups: innominate.list.sgi.xfs
Subject: Re: kio on non-x86
Date:   2 Mar 2001 16:55:43 GMT
Organization: innominate AG, Berlin, Germany
Lines:  48
Distribution: local
Message-ID: <news2mail-97ojaf$k03$2@mate.bln.innominate.de>
References: <news2mail-97oct1$d7c$1@mate.bln.innominate.de> <Pine.LNX.4.21.0103020830520.20445-100000@getafix.engr.sgi.com>
X-Trace: mate.bln.innominate.de 983552143 20483 10.0.0.31 (2 Mar 2001 16:55:43 GMT)
X-Complaints-To: news@innominate.de
User-Agent: tin/1.4.4-20000803 ("Vet for the Insane") (UNIX) (Linux/2.4.2-XFS (i686))
To:     linux-xfs@oss.sgi.com
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing

Chaitanya Tumuluri <chait@getafix.engr.sgi.com> wrote:
>> > I presume the same devices worked OK without kio, did they work OK after
>> > this shutdown?

> Do you have the error messages? 
> AFAIK, kiobuf code is _not_ arch specific.

none so far - but i did not have any of the debugging options turned
on (no idea if they would help - maybe not)

>> >> * am i right in assuming that the kio stuff does not contain any
>> >>   arch specific parts so it should - in principle - work also on
>> >>   the pcc which would leave endian or gcc issues as a possible
>> >>   reason for the problem?
>> 
>> > I think this is true, there are no os specific components, where you using
>> > ide or scsi? The ide kio code is specific to having dma turned on. So which
>> > combination was this?
>> 
>> it's ide and dma turned on

> There are no arch specific components to the kiobuf code. Though, if you
> are using more than 1GB there may be an issue with highmem bouncing code
> in kiobuf I/O. How much memory do you have on your system?

128MB - so no problem here (also keep in mind this is on ppc not i386)

>> >> * does anyone have an idea off-mind what might be the reson for
>> >>   this? ... i'll turn on all the debugging options and see if i
>> >>   get any more information out of it this evening i think
>> 
>> > Chait may have some suggestions, but the specific hardware combination is
>> > the most interesting part of the equation right now.
>> 
>> thanks - as said i waill (and can) do more testing this evening

> Error messages, if any, would help....I'll wait for the results from
> your tests.

more later - then i'm at the machine

t

-- 
thomas.graichen@innominate.com
                                                         innominate AG
                                                  the linux architects
tel: +49-30-308806-13   fax: -77             http://www.innominate.com

From owner-linux-xfs@oss.sgi.com Fri Mar  2 09:24:32 2001
Received:  by oss.sgi.com id <S553802AbRCBRYM>;
	Fri, 2 Mar 2001 09:24:12 -0800
Received: from pneumatic-tube.sgi.com ([204.94.214.22]:47442 "EHLO
        pneumatic-tube.sgi.com") by oss.sgi.com with ESMTP
	id <S553796AbRCBRXx>; Fri, 2 Mar 2001 09:23:53 -0800
Received: from cthulhu.engr.sgi.com (gate3-relay.engr.sgi.com [130.62.1.234]) by pneumatic-tube.sgi.com (980327.SGI.8.8.8-aspam/980310.SGI-aspam) via ESMTP id JAA01529
	for <linux-xfs@oss.sgi.com>; Fri, 2 Mar 2001 09:33:32 -0800 (PST)
	mail_from (chait@getafix.engr.sgi.com)
Received: from getafix.engr.sgi.com (IDENT:root@getafix.engr.sgi.com [163.154.5.110])
	by cthulhu.engr.sgi.com (SGI-8.9.3/8.9.3) with ESMTP id JAA45849;
	Fri, 2 Mar 2001 09:22:36 -0800 (PST)
Received: from localhost (chait@localhost)
	by getafix.engr.sgi.com (8.9.3/8.9.3) with ESMTP id JAA21085;
	Fri, 2 Mar 2001 09:18:37 -0500
Date:   Fri, 2 Mar 2001 09:18:37 -0500 (EST)
From:   Chaitanya Tumuluri <chait@getafix.engr.sgi.com>
To:     Thomas Graichen <thomas.graichen@innominate.com>
cc:     linux-xfs@oss.sgi.com
Subject: Re: kio on non-x86
In-Reply-To: <news2mail-97ojaf$k03$2@mate.bln.innominate.de>
Message-ID: <Pine.LNX.4.21.0103020916170.21050-100000@getafix.engr.sgi.com>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing

On 2 Mar 2001, Thomas Graichen wrote:
> none so far - but i did not have any of the debugging options turned
> on (no idea if they would help - maybe not)

I was angling for dmesg type outputs that'd indicate drive
specific info. if any. 

> 
> 128MB - so no problem here (also keep in mind this is on ppc not i386)
> 

Gah! Not enough coffee...too early...I _did_ have i386 in mind. :^)

Cheers,
-Chait.


From owner-linux-xfs@oss.sgi.com Fri Mar  2 10:08:01 2001
Received:  by oss.sgi.com id <S553802AbRCBSHw>;
	Fri, 2 Mar 2001 10:07:52 -0800
Received: from sgi.SGI.COM ([192.48.153.1]:52591 "EHLO sgi.com")
	by oss.sgi.com with ESMTP id <S553796AbRCBSHg>;
	Fri, 2 Mar 2001 10:07:36 -0800
Received: from waco.engr.sgi.com ([163.154.18.95]) 
	by sgi.com (980327.SGI.8.8.8-aspam/980304.SGI-aspam:
       SGI does not authorize the use of its proprietary
       systems or networks for unsolicited or bulk email
       from the Internet.) 
	via ESMTP id KAA04907
	for <linux-xfs@oss.sgi.com>; Fri, 2 Mar 2001 10:07:33 -0800 (PST)
	mail_from (ananth@waco.engr.sgi.com)
Received: (from ananth@localhost)
	by waco.engr.sgi.com (8.11.0/8.11.0) id f22I8aS12556
	for linux-xfs@oss.sgi.com; Fri, 2 Mar 2001 10:08:36 -0800
Date:   Fri, 2 Mar 2001 10:08:36 -0800
From:   Ananth Ananthanarayanan <ananth@waco.engr.sgi.com>
Message-Id: <200103021808.f22I8aS12556@waco.engr.sgi.com>
Subject: TAKE - fix fsync oops
To:     unlisted-recipients:; (no To-header on input)
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing

Thanks to Joakim Bodin for reporting & verifying the problem.


Date:  Fri Mar  2 10:04:34 PST 2001
Workarea:  waco.engr.sgi.com:/build1/ananth/xfs-tot

The following file(s) were checked into:
  bonnie.engr.sgi.com:/isms/slinx/2.4.x-xfs


Modid:  2.4.x-xfs:slinx:88775a
linux/fs/pagebuf/page_buf_io.c - 1.60
	- Fix a problem were page->buffers was being looked up before
	  being intialized for pages over holes.

--------------------------------------------------------------------------
Rajagopal Ananthanarayanan ("ananth")
Member Technical Staff, SGI.
--------------------------------------------------------------------------

From owner-linux-xfs@oss.sgi.com Fri Mar  2 11:53:21 2001
Received:  by oss.sgi.com id <S553813AbRCBTxC>;
	Fri, 2 Mar 2001 11:53:02 -0800
Received: from cc19815-a.zwoll1.ov.nl.home.com ([212.204.138.247]:60662 "HELO
        wisdom.myplace.net") by oss.sgi.com with SMTP id <S553765AbRCBTwv>;
	Fri, 2 Mar 2001 11:52:51 -0800
Received: from ws1 (unknown [192.168.1.15])
	by wisdom.myplace.net (Postfix) with SMTP id C6464115
	for <linux-xfs@oss.sgi.com>; Fri,  2 Mar 2001 20:52:48 +0100 (CET)
Message-ID: <008901c0a39d$cac56160$0f01a8c0@myplace.net>
From:   "Bas" <list@showme.wox.org>
To:     <linux-xfs@oss.sgi.com>
Subject: Log on a separate device
Date:   Fri, 2 Mar 2001 20:52:46 -0800
MIME-Version: 1.0
Content-Type: text/plain;
	charset="iso-8859-1"
Content-Transfer-Encoding: 7bit
X-Priority: 3
X-MSMail-Priority: Normal
X-Mailer: Microsoft Outlook Express 5.00.2615.200
X-MimeOLE: Produced By Microsoft MimeOLE V5.00.2615.200
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing

Hi,

Finally had 2.4.2 with xfs running on LVM, and after reading the man pages I
was wondering if it possible to make one log device (/dev/rootvg/log) for
more than 1 file system ?

Anybody has any idea....

Thanks,
Bas


From owner-linux-xfs@oss.sgi.com Fri Mar  2 12:36:11 2001
Received:  by oss.sgi.com id <S553813AbRCBUgC>;
	Fri, 2 Mar 2001 12:36:02 -0800
Received: from hermes.mixx.net ([212.84.196.2]:15373 "HELO hermes.mixx.net")
	by oss.sgi.com with SMTP id <S553765AbRCBUfm>;
	Fri, 2 Mar 2001 12:35:42 -0800
Received: from mate.bln.innominate.de (cerberus.berlin.innominate.de [212.84.234.251])
	by hermes.mixx.net (Postfix) with ESMTP id 1B7C4F826
	for <linux-xfs@oss.sgi.com>; Fri,  2 Mar 2001 21:35:39 +0100 (CET)
Received: by mate.bln.innominate.de (Postfix, from userid 9)
	id B535E2CA6F; Fri,  2 Mar 2001 21:35:38 +0100 (CET)
From:   Thomas Graichen <news-innominate.list.sgi.xfs@innominate.de>
Reply-To: Thomas Graichen <thomas.graichen@innominate.com>
X-Newsgroups: innominate.list.sgi.xfs
Subject: Re: kio on non-x86
Date:   2 Mar 2001 20:35:38 GMT
Organization: innominate AG, Berlin, Germany
Lines:  21
Distribution: local
Message-ID: <news2mail-97p06q$79t$1@mate.bln.innominate.de>
References: <97njoe$e2f$2@mate.bln.innominate.de>
X-Trace: mate.bln.innominate.de 983565338 7485 10.0.0.31 (2 Mar 2001 20:35:38 GMT)
X-Complaints-To: news@innominate.de
User-Agent: tin/1.4.4-20000803 ("Vet for the Insane") (UNIX) (Linux/2.4.2-XFS (i686))
To:     linux-xfs@oss.sgi.com
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing

Thomas Graichen <thomas.graichen@innominate.com> wrote:
> just updated the kernel on my ppc machine yesterday - worked all fine
> so far ... that i tried the kio and kiocluster options on the ppc too
> and ran into problems: as soon as i write to the fs it detects an io
> error and shuts down ... thus a few questions:

ok - i nailed it down further - looks like it's not a problem with
XFS but more with the ide driver (ide-pmac.c) ... tried it with an
usb floppy drive (which uses usb-storage and thus an scsi layer)
and with this one it worked perfectly even with kio ... so it was
a false alarm in the direction of XFS (but maybe jens has an idea
what might be the reason for the kio stuff not working with
ide-pmac.c on ppc?)

t

-- 
thomas.graichen@innominate.com
                                                         innominate AG
                                                  the linux architects
tel: +49-30-308806-13   fax: -77             http://www.innominate.com

From owner-linux-xfs@oss.sgi.com Fri Mar  2 12:49:42 2001
Received:  by oss.sgi.com id <S553813AbRCBUtW>;
	Fri, 2 Mar 2001 12:49:22 -0800
Received: from sgi.SGI.COM ([192.48.153.1]:31797 "EHLO sgi.com")
	by oss.sgi.com with ESMTP id <S553765AbRCBUst>;
	Fri, 2 Mar 2001 12:48:49 -0800
Received: from zeus-fddi.americas.sgi.com (128-162-8-103.americas.sgi.com [128.162.8.103]) 
	by sgi.com (980327.SGI.8.8.8-aspam/980304.SGI-aspam:
       SGI does not authorize the use of its proprietary
       systems or networks for unsolicited or bulk email
       from the Internet.) 
	via ESMTP id MAA08814
	for <linux-xfs@oss.sgi.com>; Fri, 2 Mar 2001 12:48:49 -0800 (PST)
	mail_from (lord@sgi.com)
Received: from daisy-e185.americas.sgi.com (daisy-e194.americas.sgi.com [128.162.194.214]) by zeus-fddi.americas.sgi.com (8.9.3/americas-smart-nospam1.1) with ESMTP id OAA998997; Fri, 2 Mar 2001 14:47:28 -0600 (CST)
Received: from jen.americas.sgi.com (IDENT:root@jen.americas.sgi.com [128.162.184.86]) by daisy-e185.americas.sgi.com (SGI-8.9.3/SGI-server-1.7) with ESMTP id OAA18870; Fri, 2 Mar 2001 14:47:28 -0600 (CST)
Received: from jen.americas.sgi.com by jen.americas.sgi.com (8.11.0/SGI-client-1.7) via ESMTP id f22KlIU02228; Fri, 2 Mar 2001 14:47:18 -0600
Message-Id: <200103022047.f22KlIU02228@jen.americas.sgi.com>
X-Mailer: exmh version 2.2 06/23/2000 with nmh-1.0.4
To:     "Bas" <list@showme.wox.org>
cc:     linux-xfs@oss.sgi.com
Subject: Re: Log on a separate device 
In-Reply-To: Message from "Bas" <list@showme.wox.org> 
   of "Fri, 02 Mar 2001 20:52:46 PST." <008901c0a39d$cac56160$0f01a8c0@myplace.net> 
Date:   Fri, 02 Mar 2001 14:47:18 -0600
From:   Steve Lord <lord@sgi.com>
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing

> Hi,
> 
> Finally had 2.4.2 with xfs running on LVM, and after reading the man pages I
> was wondering if it possible to make one log device (/dev/rootvg/log) for
> more than 1 file system ?
> 
> Anybody has any idea....
> 
> Thanks,
> Bas

The xfs log is a per filesystem structure, you cannot share logs between
different filesystems. The external log feature is there because on irix
there are volume managers which let you create volumes with subvolumes
within them, xfs data, log and realtime being the typical uses. There is
no equivalent on linux at the moment. 

Steve


Steve




From owner-linux-xfs@oss.sgi.com Fri Mar  2 15:09:02 2001
Received:  by oss.sgi.com id <S553849AbRCBXIm>;
	Fri, 2 Mar 2001 15:08:42 -0800
Received: from deliverator.sgi.com ([204.94.214.10]:56165 "EHLO
        deliverator.sgi.com") by oss.sgi.com with ESMTP id <S553838AbRCBXIU>;
	Fri, 2 Mar 2001 15:08:20 -0800
Received: from zeus-fddi.americas.sgi.com (zeus-fddi.americas.sgi.com [128.162.8.103]) by deliverator.sgi.com (980309.SGI.8.8.8-aspam-6.2/980310.SGI-aspam) via ESMTP id PAA25044
	for <linux-xfs@oss.sgi.com>; Fri, 2 Mar 2001 15:07:13 -0800 (PST)
	mail_from (lord@sgi.com)
Received: from daisy-e185.americas.sgi.com (daisy-e194.americas.sgi.com [128.162.194.214]) by zeus-fddi.americas.sgi.com (8.9.3/americas-smart-nospam1.1) with ESMTP id RAA967674 for <linux-xfs@oss.sgi.com>; Fri, 2 Mar 2001 17:07:02 -0600 (CST)
Received: from jen.americas.sgi.com (IDENT:root@jen.americas.sgi.com [128.162.184.86]) by daisy-e185.americas.sgi.com (SGI-8.9.3/SGI-server-1.7) with ESMTP id RAA47758 for <linux-xfs@oss.sgi.com>; Fri, 2 Mar 2001 17:07:01 -0600 (CST)
From:   Steve Lord <lord@sgi.com>
Received: by jen.americas.sgi.com (8.11.0/SGI-client-1.7) id f22N6pE03365; Fri, 2 Mar 2001 17:06:51 -0600
Message-Id: <200103022306.f22N6pE03365@jen.americas.sgi.com>
Date:   Fri, 2 Mar 2001 17:06:51 -0600
Subject: TAKE - return ENOSPC to caller
To:     unlisted-recipients:; (no To-header on input)
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing

There was a case were we did not return ENOSPC to the caller

Date:  Fri Mar  2 15:06:31 PST 2001
Workarea:  jen.americas.sgi.com:/src/lord/xfs-linux.2.4

The following file(s) were checked into:
  bonnie.engr.sgi.com:/isms/slinx/2.4.x-xfs


Modid:  2.4.x-xfs:slinx:88816a
linux/fs/pagebuf/page_buf_io.c - 1.61
	- Fix error return from prepare write - pass the error back to the caller
	  as is without turning it into EIO.



From owner-linux-xfs@oss.sgi.com Fri Mar  2 16:01:02 2001
Received:  by oss.sgi.com id <S553856AbRCCAAx>;
	Fri, 2 Mar 2001 16:00:53 -0800
Received: from hermes.mixx.net ([212.84.196.2]:46609 "HELO hermes.mixx.net")
	by oss.sgi.com with SMTP id <S553850AbRCCAA3>;
	Fri, 2 Mar 2001 16:00:29 -0800
Received: from mate.bln.innominate.de (cerberus.berlin.innominate.de [212.84.234.251])
	by hermes.mixx.net (Postfix) with ESMTP id 71073F847
	for <linux-xfs@oss.sgi.com>; Sat,  3 Mar 2001 01:00:26 +0100 (CET)
Received: by mate.bln.innominate.de (Postfix, from userid 9)
	id BCE382CA71; Sat,  3 Mar 2001 01:00:24 +0100 (CET)
From:   Thomas Graichen <news-innominate.list.sgi.xfs@innominate.de>
Reply-To: Thomas Graichen <thomas.graichen@innominate.com>
X-Newsgroups: innominate.list.sgi.xfs
Subject: attr on ppc
Date:   3 Mar 2001 00:00:24 GMT
Organization: innominate AG, Berlin, Germany
Lines:  24
Distribution: local
Message-ID: <news2mail-97pc6o$kgu$1@mate.bln.innominate.de>
X-Trace: mate.bln.innominate.de 983577624 21022 10.0.0.31 (3 Mar 2001 00:00:24 GMT)
X-Complaints-To: news@innominate.de
User-Agent: tin/1.4.4-20000803 ("Vet for the Insane") (UNIX) (Linux/2.4.2-XFS (i686))
To:     linux-xfs@oss.sgi.com
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing

i'm now again trying to get attr's working on the ppc ... so i added
the sysctl stuff for it etc. ... but if i do an

  attr -s test -V yes /some_file_on_xfs

i only get a "Bad address" ... so far i can follow it to the point
that it gets into sys_attrctl in ext_attr.c and there into the
ATTR_TYPE_LPATH case and the user_path_walk_link gives an error -14
- independent of the existence of the /some_file_on_xfs ... so somehow
my assumption is that somehow maybe the attr_obj_t obj of sys_attrctl
gets trashed on the ppc (this would make clear why there is no
differnece between existence and nonexistence of the file) and thus
everything else later fails ... anyone any more ideas how to get
further from here?

a lot of thanks in advance

t

-- 
thomas.graichen@innominate.com
                                                         innominate AG
                                                  the linux architects
tel: +49-30-308806-13   fax: -77             http://www.innominate.com

From owner-linux-xfs@oss.sgi.com Fri Mar  2 16:03:43 2001
Received:  by oss.sgi.com id <S553856AbRCCADd>;
	Fri, 2 Mar 2001 16:03:33 -0800
Received: from hermes.mixx.net ([212.84.196.2]:49169 "HELO hermes.mixx.net")
	by oss.sgi.com with SMTP id <S553850AbRCCADP>;
	Fri, 2 Mar 2001 16:03:15 -0800
Received: from mate.bln.innominate.de (cerberus.berlin.innominate.de [212.84.234.251])
	by hermes.mixx.net (Postfix) with ESMTP id 3EE2AF847
	for <linux-xfs@oss.sgi.com>; Sat,  3 Mar 2001 01:03:13 +0100 (CET)
Received: by mate.bln.innominate.de (Postfix, from userid 9)
	id E39BD2CA6F; Sat,  3 Mar 2001 01:03:12 +0100 (CET)
From:   Thomas Graichen <news-innominate.list.sgi.xfs@innominate.de>
Reply-To: Thomas Graichen <thomas.graichen@innominate.com>
X-Newsgroups: innominate.list.sgi.xfs
Subject: small fix
Date:   3 Mar 2001 00:03:12 GMT
Organization: innominate AG, Berlin, Germany
Lines:  21
Distribution: local
Message-ID: <news2mail-97pcc0$kgu$2@mate.bln.innominate.de>
X-Trace: mate.bln.innominate.de 983577792 21022 10.0.0.31 (3 Mar 2001 00:03:12 GMT)
X-Complaints-To: news@innominate.de
User-Agent: tin/1.4.4-20000803 ("Vet for the Insane") (UNIX) (Linux/2.4.2-XFS (i686))
To:     linux-xfs@oss.sgi.com
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing

i had to add this to get the attr package compiled on the ppc (and it
sounds ok to me)

--- attr/libattr/attr.c	2001/02/09 07:45:14	1.3
+++ attr/libattr/attr.c	2001/03/03 00:05:35
@@ -32,6 +32,7 @@
 
 #include <errno.h>
 #include <stdlib.h>
+#include <stdio.h>
 #include <sys/types.h>
 #include <asm/types.h>
 #include <asm/unistd.h>

t

-- 
thomas.graichen@innominate.com
                                                         innominate AG
                                                  the linux architects
tel: +49-30-308806-13   fax: -77             http://www.innominate.com

From owner-linux-xfs@oss.sgi.com Sat Mar  3 00:34:28 2001
Received:  by oss.sgi.com id <S553793AbRCCIeI>;
	Sat, 3 Mar 2001 00:34:08 -0800
Received: from sgi.SGI.COM ([192.48.153.1]:51791 "EHLO sgi.com")
	by oss.sgi.com with ESMTP id <S553724AbRCCIdo>;
	Sat, 3 Mar 2001 00:33:44 -0800
Received: from snort.melbourne.sgi.com ([134.14.55.149]) 
	by sgi.com (980327.SGI.8.8.8-aspam/980304.SGI-aspam:
       SGI does not authorize the use of its proprietary
       systems or networks for unsolicited or bulk email
       from the Internet.) 
	via ESMTP id AAA06404
	for <linux-xfs@oss.sgi.com>; Sat, 3 Mar 2001 00:33:38 -0800 (PST)
	mail_from (nathans@snort.melbourne.sgi.com)
Received: (from nathans@localhost)
	by snort.melbourne.sgi.com (SGI-8.9.3/8.9.3) id TAA17763
	for linux-xfs@oss.sgi.com; Sat, 3 Mar 2001 19:32:09 +1100 (EST)
Date:   Sat, 3 Mar 2001 19:32:09 +1100 (EST)
From:   Nathan Scott <nathans@snort.melbourne.sgi.com>
Message-Id: <200103030832.TAA17763@snort.melbourne.sgi.com>
To:     linux-xfs@oss.sgi.com
Subject: TAKE - attr.c
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing

Date:  Sat Mar  3 00:31:28 PST 2001
Workarea:  snort.melbourne.sgi.com:/diskb/build4/nathans/linux-xfs

The following file(s) were checked into:
  bonnie.engr.sgi.com:/isms/slinx/2.4.x-xfs


Modid:  2.4.x-xfs:slinx:88834a
cmd/attr/libattr/attr.c - 1.4
	- include stdio.h for Thomas' PPC builds.



From owner-linux-xfs@oss.sgi.com Sat Mar  3 08:51:20 2001
Received:  by oss.sgi.com id <S553854AbRCCQvK>;
	Sat, 3 Mar 2001 08:51:10 -0800
Received: from relais.videotron.ca ([24.201.245.36]:50054 "EHLO
        VL-MS-MR002.sc1.videotron.ca") by oss.sgi.com with ESMTP
	id <S553735AbRCCQu5>; Sat, 3 Mar 2001 08:50:57 -0800
Received: from giref.ulaval.ca ([24.200.12.70]) by
          VL-MS-MR002.sc1.videotron.ca (Netscape Messaging Server 4.15)
          with ESMTP id G9MS1O02.PK7 for <linux-xfs@oss.sgi.com>; Sat, 3
          Mar 2001 11:49:00 -0500 
Message-ID: <3AA0DB19.A8B69228@giref.ulaval.ca>
Date:   Sat, 03 Mar 2001 11:52:57 +0000
From:   Luc Lalonde <llalonde@giref.ulaval.ca>
X-Mailer: Mozilla 4.76 [en] (X11; U; Linux 2.4.2-XFS i686)
X-Accept-Language: en
MIME-Version: 1.0
To:     linux-xfs@oss.sgi.com
Subject: Convert base ext2 partition to xfs
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing

Hello Folks,

The section on how to convert the root partition from ext2 to xfs is
missing from the FAQ...so here we go:-)

I've converted all my partitions to xfs except the root partition.  I've
done the others by moving stuff around with a spare partition that I
have.  I'd like to do the same with the root partition however, I'm not
able to copy the contents of /proc.  If I just create that directory
will the proc filesystem recreate itself automatically?

Cheers, Luc.
P.S. Keep up the good work.

From owner-linux-xfs@oss.sgi.com Sat Mar  3 10:46:01 2001
Received:  by oss.sgi.com id <S553854AbRCCSpv>;
	Sat, 3 Mar 2001 10:45:51 -0800
Received: from TK146188.telekabel.at ([195.34.146.188]:53120 "HELO
        TK146188.telekabel.at") by oss.sgi.com with SMTP id <S553735AbRCCSp3>;
	Sat, 3 Mar 2001 10:45:29 -0800
Received: from think.at (localhost [127.0.0.1])
	by TK146188.telekabel.at (Postfix) with ESMTP id 9CAB26B9809
	for <linux-xfs@oss.sgi.com>; Sat,  3 Mar 2001 19:45:12 +0100 (CET)
Message-ID: <3AA13BB8.D6BB224C@think.at>
Date:   Sat, 03 Mar 2001 19:45:12 +0100
From:   Georg Kovalcik <gk@think.at>
Organization: think.at GmbH
X-Mailer: Mozilla 4.76 [en] (X11; U; Linux 2.4.2-XFS i686)
X-Accept-Language: en
MIME-Version: 1.0
Cc:     linux-xfs@oss.sgi.com
Subject: Re: Convert base ext2 partition to xfs
References: <3AA0DB19.A8B69228@giref.ulaval.ca>
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
To:     unlisted-recipients:; (no To-header on input)
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing

Luc Lalonde wrote:
> 
> Hello Folks,
> 
> The section on how to convert the root partition from ext2 to xfs is
> missing from the FAQ...so here we go:-)
> 
> I've converted all my partitions to xfs except the root partition.  I've
> done the others by moving stuff around with a spare partition that I
> have.  I'd like to do the same with the root partition however, I'm not
> able to copy the contents of /proc.  If I just create that directory
> will the proc filesystem recreate itself automatically?
> 
> Cheers, Luc.
> P.S. Keep up the good work.

Yes, the proc-filesystem is created at boot-time, all you will have to
do ist to create a mountpoint (the directory /proc) for it.

Cheers,

Georg

-- 
Georg Kovalcik
think.at GmbH
+43/664/5320999

From owner-linux-xfs@oss.sgi.com Sat Mar  3 10:51:21 2001
Received:  by oss.sgi.com id <S553854AbRCCSvM>;
	Sat, 3 Mar 2001 10:51:12 -0800
Received: from main.braxis.co.uk ([212.160.232.26]:63496 "EHLO
        main.braxis.co.uk") by oss.sgi.com with ESMTP id <S553735AbRCCSvG>;
	Sat, 3 Mar 2001 10:51:06 -0800
Received: (from kszysiu@localhost)
	by main.braxis.co.uk (8.9.3/8.9.3) id TAA15654;
	Sat, 3 Mar 2001 19:47:32 +0100
Date:   Sat, 3 Mar 2001 19:47:32 +0100
From:   Krzysztof Rusocki <kszysiu@braxis.co.uk>
To:     Luc Lalonde <llalonde@giref.ulaval.ca>
Cc:     linux-xfs@oss.sgi.com
Subject: Re: Convert base ext2 partition to xfs
Message-ID: <20010303194732.A11322@main.braxis.co.uk>
References: <3AA0DB19.A8B69228@giref.ulaval.ca>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.2.5i
In-Reply-To: <3AA0DB19.A8B69228@giref.ulaval.ca>; from llalonde@giref.ulaval.ca on Sat, Mar 03, 2001 at 11:52:57AM +0000
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing


Hi,

On destination partition create only /proc directory and nothing
more. The /proc is just kernel interface and you get access to it
as soon as you mount -t proc /proc /proc ;-)
/proc is 'virtual' fs, so it does not exist physically on any drive

Also, remember to install boot manager afterwards.

Cheers
Krzysztof

On Sat, Mar 03, 2001 at 11:52:57AM +0000, Luc Lalonde wrote:
> Hello Folks,
> 
> The section on how to convert the root partition from ext2 to xfs is
> missing from the FAQ...so here we go:-)
> 
> I've converted all my partitions to xfs except the root partition.  I've
> done the others by moving stuff around with a spare partition that I
> have.  I'd like to do the same with the root partition however, I'm not
> able to copy the contents of /proc.  If I just create that directory
> will the proc filesystem recreate itself automatically?
> 
> Cheers, Luc.
> P.S. Keep up the good work.

From owner-linux-xfs@oss.sgi.com Sat Mar  3 13:20:41 2001
Received:  by oss.sgi.com id <S553859AbRCCVUc>;
	Sat, 3 Mar 2001 13:20:32 -0800
Received: from mail0.mia.bellsouth.net ([205.152.144.12]:54212 "EHLO
        mail0.mia.bellsouth.net") by oss.sgi.com with ESMTP
	id <S553810AbRCCVUJ>; Sat, 3 Mar 2001 13:20:09 -0800
Received: from spruce (adsl-61-5-192.mia.bellsouth.net [208.61.5.192])
	by mail0.mia.bellsouth.net (3.3.5alt/0.75.2) with SMTP id QAA00829;
	Sat, 3 Mar 2001 16:20:08 -0500 (EST)
From:   "Juan Casero" <casero@bellsouth.net>
To:     <linux-xfs@oss.sgi.com>
Cc:     <casero@bellsouth.net>
Subject: XFS Linux and /dev entries....
Date:   Sat, 3 Mar 2001 16:11:51 -0500
Message-ID: <ELEJLDODGANIPOJJBJMEGEHLCAAA.casero@bellsouth.net>
MIME-Version: 1.0
Content-Type: text/plain;
	charset="iso-8859-1"
Content-Transfer-Encoding: 7bit
X-Priority: 3 (Normal)
X-MSMail-Priority: Normal
X-Mailer: Microsoft Outlook IMO, Build 9.0.2416 (9.0.2911.0)
X-MimeOLE: Produced By Microsoft MimeOLE V5.50.4133.2400
Importance: Normal
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing

I have a question about devfs and XFS on a Linux box.  I run RedHat 7.0
with XFS exclusively on my primary Linux workstation.  This box uses an
NVidia TNT2 Pro video card and XFree86 4.0.1 so naturally I want to install
and run the specilized drivers provided by nvidia for XFree86 4.0.1.  There
is a problem though; when I install the nvidia drivers (after compiling them
myself they create an entry in /dev called nvidiactl or something like that.
Since the entries in /dev are creating dynamically the entries in /dev
created
by the nvidia software get wiped out when the system shuts down.  Therefore
when it comes back up the XFree86 provided nvidia drivers die and I can't
start X.  A clumsy solution would be to boot into runlevel 3 each time and
recreate the entries myself for each session but that is not a good
approach.
Is there a way to force the system to create the entries in /dev for the
nvidia drivers each time it reboots?  If you don't feel like going into
details
about how to do this but you know where documentation is online please point
me in the right direction.


Thanks......

Juan Casero
casero@bellsouth.net


From owner-linux-xfs@oss.sgi.com Sat Mar  3 13:49:51 2001
Received:  by oss.sgi.com id <S553861AbRCCVtm>;
	Sat, 3 Mar 2001 13:49:42 -0800
Received: from garnet.sover.net ([209.198.87.53]:34496 "EHLO garnet.sover.net")
	by oss.sgi.com with ESMTP id <S553856AbRCCVtk>;
	Sat, 3 Mar 2001 13:49:40 -0800
Received: from surreal.localdomain (pm1a21.stj.sover.net [209.198.94.21])
	by garnet.sover.net (8.9.3/8.9.3) with ESMTP id QAA00318;
	Sat, 3 Mar 2001 16:48:36 -0500 (EST)
Comments: SoVerNet Verification (on garnet.sover.net)
	surreal.localdomain from pm1a21.stj.sover.net [209.198.94.21] 209.198.94.21
	Sat, 3 Mar 2001 16:48:36 -0500 (EST)
Date:   Sat, 3 Mar 2001 16:48:53 -0500 (EST)
From:   Jason Walker <unseen@sover.net>
X-Sender: unseen@surreal.localdomain
To:     Juan Casero <casero@bellsouth.net>
cc:     linux-xfs@oss.sgi.com
Subject: Re: XFS Linux and /dev entries....
In-Reply-To: <ELEJLDODGANIPOJJBJMEGEHLCAAA.casero@bellsouth.net>
Message-ID: <Pine.LNX.4.10.10103031644330.3060-100000@surreal.localdomain>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing


	I had the same problem with my nvidia drivers.  You need a patch for
the nvidia drivers to work with devfs.  I got mine from #nvidia on
openprojects IRC server(s).  Go there and ask around. They are quite
helpful

RegEx


On Sat, 3 Mar 2001, Juan Casero wrote:

> I have a question about devfs and XFS on a Linux box.  I run RedHat 7.0
> with XFS exclusively on my primary Linux workstation.  This box uses an
> NVidia TNT2 Pro video card and XFree86 4.0.1 so naturally I want to install
> and run the specilized drivers provided by nvidia for XFree86 4.0.1.  There
> is a problem though; when I install the nvidia drivers (after compiling them
> myself they create an entry in /dev called nvidiactl or something like that.
> Since the entries in /dev are creating dynamically the entries in /dev
> created
> by the nvidia software get wiped out when the system shuts down.  Therefore
> when it comes back up the XFree86 provided nvidia drivers die and I can't
> start X.  A clumsy solution would be to boot into runlevel 3 each time and
> recreate the entries myself for each session but that is not a good
> approach.
> Is there a way to force the system to create the entries in /dev for the
> nvidia drivers each time it reboots?  If you don't feel like going into
> details
> about how to do this but you know where documentation is online please point
> me in the right direction.
> 
> 
> Thanks......
> 
> Juan Casero
> casero@bellsouth.net
> 

-- 
Jason Walker -- unseen@sover.net

/"\
\ /     ASCII RIBBON CAMPAIGN
 X        AGAINST HTML MAIL
/ \
			 


From owner-linux-xfs@oss.sgi.com Sat Mar  3 16:46:22 2001
Received:  by oss.sgi.com id <S553876AbRCDAqM>;
	Sat, 3 Mar 2001 16:46:12 -0800
Received: from [65.100.85.34] ([65.100.85.34]:31368 "EHLO
        roujin.gargoylecc.com") by oss.sgi.com with ESMTP
	id <S553685AbRCDAqB>; Sat, 3 Mar 2001 16:46:01 -0800
Received: from [65.100.85.35] (IDENT:ringram@gargoyle.gargoylecc.com [65.100.85.35])
	by roujin.gargoylecc.com (8.9.3/8.9.3) with ESMTP id IAA15046;
	Sun, 4 Mar 2001 08:57:07 -0700
Date:   Sat, 3 Mar 2001 17:47:02 -0700 (MST)
From:   Russel Ingram <ringram@gargoylecc.com>
X-Sender:  <ringram@localhost.localdomain>
To:     Jason Walker <unseen@sover.net>
cc:     Juan Casero <casero@bellsouth.net>, <linux-xfs@oss.sgi.com>
Subject: Re: XFS Linux and /dev entries....
In-Reply-To: <Pine.LNX.4.10.10103031644330.3060-100000@surreal.localdomain>
Message-ID: <Pine.LNX.4.30.0103031736390.16970-100000@localhost.localdomain>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing

How are the drivers loaded?  If they are kernel modules then they should
be able to be loaded when X calls for them from the kernel.  If you can
see in your logs where the kernel tries to load the module (it will most
likely be looking for a name other than the actual name of the module)
then you can put an alias in your /etc/modules.conf file to load the
module when it is called for (if you haven't already done so).  Then you
can put a line in your /etc/devfsd.conf file that tells it to make a link
to the /dev file that X actually looks for when the module is loaded.  The
line for devfsd.conf would look something like this:

REGISTER  v4l/video0 CFUNCTION /lib/libc.so.6 symlink $devname video

The example I have here is for my webcam and it creates a symbolic link
from /dev/v41/video0 to /dev/video when the module for my webcam is
inserted into the kernel.

On Sat, 3 Mar 2001, Jason Walker wrote:

>
> 	I had the same problem with my nvidia drivers.  You need a patch for
> the nvidia drivers to work with devfs.  I got mine from #nvidia on
> openprojects IRC server(s).  Go there and ask around. They are quite
> helpful
>
> RegEx
>
>
> On Sat, 3 Mar 2001, Juan Casero wrote:
>
> > I have a question about devfs and XFS on a Linux box.  I run RedHat 7.0
> > with XFS exclusively on my primary Linux workstation.  This box uses an
> > NVidia TNT2 Pro video card and XFree86 4.0.1 so naturally I want to install
> > and run the specilized drivers provided by nvidia for XFree86 4.0.1.  There
> > is a problem though; when I install the nvidia drivers (after compiling them
> > myself they create an entry in /dev called nvidiactl or something like that.
> > Since the entries in /dev are creating dynamically the entries in /dev
> > created
> > by the nvidia software get wiped out when the system shuts down.  Therefore
> > when it comes back up the XFree86 provided nvidia drivers die and I can't
> > start X.  A clumsy solution would be to boot into runlevel 3 each time and
> > recreate the entries myself for each session but that is not a good
> > approach.
> > Is there a way to force the system to create the entries in /dev for the
> > nvidia drivers each time it reboots?  If you don't feel like going into
> > details
> > about how to do this but you know where documentation is online please point
> > me in the right direction.
> >
> >
> > Thanks......
> >
> > Juan Casero
> > casero@bellsouth.net
> >
>
>
Russ
-- 
---------------------------------------------------------------
"Bill Gates and Microsoft have ruined the computer industry for
 a long time to come by creating a class of ignorant and lazy
 computer users."  --Russel Ingram

"Mommy ... can I go out and ... KILL TONIGHT!?"
                   --Glen  Danzig, The Misfits
---------------------------------------------------------------
Russ Ingram
Gargoyle Computer Consulting
(307)742-1361
www.gargoylecc.com


From owner-linux-xfs@oss.sgi.com Sat Mar  3 19:27:03 2001
Received:  by oss.sgi.com id <S553648AbRCDD0p>;
	Sat, 3 Mar 2001 19:26:45 -0800
Received: from site3.talontech.com ([208.179.68.88]:51791 "EHLO
        site3.talontech.com") by oss.sgi.com with ESMTP id <S553645AbRCDD0R>;
	Sat, 3 Mar 2001 19:26:17 -0800
Received: from kalifornia.com (linux4all@sc-24-130-146-204.socal.rr.com [24.130.146.204])
	by site3.talontech.com (8.11.0/8.11.0) with ESMTP id f243SQg07421;
	Sat, 3 Mar 2001 19:28:27 -0800
Message-ID: <3AA1B6D7.2080601@kalifornia.com>
Date:   Sat, 03 Mar 2001 19:30:31 -0800
From:   Ben Ford <ben@kalifornia.com>
User-Agent: Mozilla/5.0 (X11; U; Linux 2.4.1 i586; en-US; 0.9) Gecko/20010226
X-Accept-Language: en
MIME-Version: 1.0
To:     Luc Lalonde <llalonde@giref.ulaval.ca>
CC:     linux-xfs@oss.sgi.com
Subject: Re: Convert base ext2 partition to xfs
References: <3AA0DB19.A8B69228@giref.ulaval.ca>
Content-Type: text/plain; charset=us-ascii; format=flowed
Content-Transfer-Encoding: 7bit
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing

Luc Lalonde wrote:

> Hello Folks,
> 
> The section on how to convert the root partition from ext2 to xfs is
> missing from the FAQ...so here we go:-)
> 
> I've converted all my partitions to xfs except the root partition.  I've
> done the others by moving stuff around with a spare partition that I
> have.  I'd like to do the same with the root partition however, I'm not
> able to copy the contents of /proc.  If I just create that directory
> will the proc filesystem recreate itself automatically?
> 
> Cheers, Luc.
> P.S. Keep up the good work.


/proc is a virtual filesystem.  It doesn't really exist, the /proc 
interface is really just a window into the kernel internals.  In short, 
yes, just make the dir and use fstab to mount it.

-b


From owner-linux-xfs@oss.sgi.com Sat Mar  3 23:26:36 2001
Received:  by oss.sgi.com id <S553679AbRCDH01>;
	Sat, 3 Mar 2001 23:26:27 -0800
Received: from groucho.maths.monash.edu.au ([130.194.160.211]:3622 "EHLO
        groucho.maths.monash.edu.au") by oss.sgi.com with ESMTP
	id <S553650AbRCDH0W>; Sat, 3 Mar 2001 23:26:22 -0800
Received: (from rjh@localhost)
	by groucho.maths.monash.edu.au (8.8.8/8.8.8) id HAA23413
	for linux-xfs@oss.sgi.com; Sun, 4 Mar 2001 07:26:04 GMT
From:   Robin Humble <rjh@groucho.maths.monash.edu.au>
Message-Id: <200103040726.HAA23413@groucho.maths.monash.edu.au>
Subject: NFS performance - more info
To:     linux-xfs@oss.sgi.com
Date:   Sun, 4 Mar 2001 18:26:04 +1100 (EDT)
X-Mailer: ELM [version 2.5 PL1]
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing


Hi XFS gurus,

Firstly you're all doing a great job and I am way impressed by the
solidity, and both speed of porting, and speed of the filesystem...

However, (and I guess you knew this was coming :-) I tried writing to
an XFS partition over NFS again - this time using work machines
instead of home, and the performance was a still a bit of a worry
to say the least.

The results are below. Is there anything more that I can do to try
and diagnose the problem for y'all? Or anything that I can use to
boost NFS performance? I heard someone talk about a userland nfsd -
might that help, and where would I find it?

machine/config this time:
dual celeron 500, ASUS bp6, 512M ram, 4x maxtor 60G/7200rpm drives,
2x promise ultra100 controller cards (each drive on its own controller),
RH7.1beta, 2.4.2-XFSsmp-3mar2001 kernel. The network is 100Mbit
ethernet (3c905 I think). Reiserfs was not timed as it was unable to
make >2G files(!!)

time dd if=/dev/zero of=bigFile bs=1024k count=2500

            writes to local disk   writes over nfs   notes
xfs                16 MB/s            0.7 MB/s    load ~90-100% per cpu + (*)
xfs over RAID0     37 MB/s             1  MB/s       ""
xfs over RAID5            didn't try yet
ext2               17 MB/s             8  MB/s    load ~30-40% per cpu
ext2 over RAID0    38 MB/s             8  MB/s       ""
ext2 over RAID5    14 MB/s             6  MB/s       "" + (**)

(*) I tried just one nfsd process as well as the default 8 and the
bandwidth to disk was the same. I thought maybe the knfsd's were
fighting each other, but that seems not to be the case.
(**) this hung the 2nd time I tried it. sigh

So yeah, the disk lights barely flashed at all with XFS writes over
NFS, and speed was about a factor of 10 less than it should be. Using
an NFSv2 or v3 client didn't matter. Nor did writing from IRIX, Tru64
or Linux.  Clearly there's some major struggle going on between knfsd
and xfs :-/ Sad to say, but the speed makes it pretty much unusable.
I didn't try server-side NFSv2 as I need big files.

Please let me know if I can try anything out for you.

cheers,
robin

From owner-linux-xfs@oss.sgi.com Sun Mar  4 01:13:17 2001
Received:  by oss.sgi.com id <S553679AbRCDJNH>;
	Sun, 4 Mar 2001 01:13:07 -0800
Received: from hermes.mixx.net ([212.84.196.2]:63492 "HELO hermes.mixx.net")
	by oss.sgi.com with SMTP id <S553650AbRCDJMw>;
	Sun, 4 Mar 2001 01:12:52 -0800
Received: from mate.bln.innominate.de (cerberus.berlin.innominate.de [212.84.234.251])
	by hermes.mixx.net (Postfix) with ESMTP id 8600EF805
	for <linux-xfs@oss.sgi.com>; Sun,  4 Mar 2001 10:12:49 +0100 (CET)
Received: by mate.bln.innominate.de (Postfix, from userid 9)
	id 33BD62CA6F; Sun,  4 Mar 2001 10:12:49 +0100 (CET)
From:   Thomas Graichen <news-innominate.list.sgi.xfs@innominate.de>
Reply-To: Thomas Graichen <thomas.graichen@innominate.com>
X-Newsgroups: innominate.list.sgi.xfs
Subject: Re: NFS performance - more info
Date:   4 Mar 2001 09:12:49 GMT
Organization: innominate AG, Berlin, Germany
Lines:  49
Distribution: local
Message-ID: <news2mail-97t0uh$5qj$1@mate.bln.innominate.de>
References: <200103040726.HAA23413@groucho.maths.monash.edu.au>
X-Trace: mate.bln.innominate.de 983697169 5971 10.0.0.31 (4 Mar 2001 09:12:49 GMT)
X-Complaints-To: news@innominate.de
User-Agent: tin/1.4.4-20000803 ("Vet for the Insane") (UNIX) (Linux/2.4.2-XFS (i686))
To:     linux-xfs@oss.sgi.com
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing

Robin Humble <rjh@groucho.maths.monash.edu.au> wrote:

> However, (and I guess you knew this was coming :-) I tried writing to
> an XFS partition over NFS again - this time using work machines
> instead of home, and the performance was a still a bit of a worry
> to say the least.
> ...
> machine/config this time:
> dual celeron 500, ASUS bp6, 512M ram, 4x maxtor 60G/7200rpm drives,
> 2x promise ultra100 controller cards (each drive on its own controller),
> RH7.1beta, 2.4.2-XFSsmp-3mar2001 kernel. The network is 100Mbit
> ethernet (3c905 I think). Reiserfs was not timed as it was unable to
> make >2G files(!!)

> time dd if=/dev/zero of=bigFile bs=1024k count=2500

>             writes to local disk   writes over nfs   notes
> xfs                16 MB/s            0.7 MB/s    load ~90-100% per cpu + (*)
> xfs over RAID0     37 MB/s             1  MB/s       ""
> xfs over RAID5            didn't try yet
> ext2               17 MB/s             8  MB/s    load ~30-40% per cpu
> ext2 over RAID0    38 MB/s             8  MB/s       ""
> ext2 over RAID5    14 MB/s             6  MB/s       "" + (**)

i just did a very simple test on the setup i can us right now from
remote and end up at about 2mb/sec (server: up 333mhz pII, 128mb,
6gb udma2 disk - so - no big hardware, ah - i forgot: k6-2/400, 64mb
solaris8 ix86 client) - looks like something is definitely not ok
with your setup (otherwise i would not get better results with my
much slower hardware) - so - some things to try:

* did you try kiocluster's (only work with ide or scsi not with md
  or lvm) - no idea if it might help -> mount -o kiocluster ...
* the other good lines posted often here (but i don't think they
  will help here) -> mount -o logbufs=4,logbsize=32768 ...
* creating the filesystem with a bigger journal is also a good idea
  for real workload (ok - i think not in this case too - because
  here is not much metadata to journal :-)
  -> mkfs -t xfs -l size=32768b ...
* did you try the same with an up kernel? - maybe some smp locking
  thing somethere

t

-- 
thomas.graichen@innominate.com
                                                         innominate AG
                                                  the linux architects
tel: +49-30-308806-13   fax: -77             http://www.innominate.com

From owner-linux-xfs@oss.sgi.com Sun Mar  4 01:19:26 2001
Received:  by oss.sgi.com id <S553679AbRCDJTR>;
	Sun, 4 Mar 2001 01:19:17 -0800
Received: from hermes.mixx.net ([212.84.196.2]:8197 "HELO hermes.mixx.net")
	by oss.sgi.com with SMTP id <S553650AbRCDJTI>;
	Sun, 4 Mar 2001 01:19:08 -0800
Received: from mate.bln.innominate.de (cerberus.berlin.innominate.de [212.84.234.251])
	by hermes.mixx.net (Postfix) with ESMTP id A115BF805
	for <linux-xfs@oss.sgi.com>; Sun,  4 Mar 2001 10:19:06 +0100 (CET)
Received: by mate.bln.innominate.de (Postfix, from userid 9)
	id 4F11E2CA6F; Sun,  4 Mar 2001 10:19:06 +0100 (CET)
From:   Thomas Graichen <news-innominate.list.sgi.xfs@innominate.de>
Reply-To: Thomas Graichen <thomas.graichen@innominate.com>
X-Newsgroups: innominate.list.sgi.xfs
Subject: Re: NFS performance - more info
Date:   4 Mar 2001 09:19:06 GMT
Organization: innominate AG, Berlin, Germany
Lines:  50
Distribution: local
Message-ID: <news2mail-97t1aa$gji$1@mate.bln.innominate.de>
References: <200103040726.HAA23413@groucho.maths.monash.edu.au> <97t0uh$5qj$1@mate.bln.innominate.de>
X-Trace: mate.bln.innominate.de 983697546 17010 10.0.0.31 (4 Mar 2001 09:19:06 GMT)
X-Complaints-To: news@innominate.de
User-Agent: tin/1.4.4-20000803 ("Vet for the Insane") (UNIX) (Linux/2.4.2-XFS (i686))
To:     linux-xfs@oss.sgi.com
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing

Thomas Graichen <thomas.graichen@innominate.com> wrote:
> Robin Humble <rjh@groucho.maths.monash.edu.au> wrote:

>> However, (and I guess you knew this was coming :-) I tried writing to
>> an XFS partition over NFS again - this time using work machines
>> instead of home, and the performance was a still a bit of a worry
>> to say the least.
>> ...
>> machine/config this time:
>> dual celeron 500, ASUS bp6, 512M ram, 4x maxtor 60G/7200rpm drives,
>> 2x promise ultra100 controller cards (each drive on its own controller),
>> RH7.1beta, 2.4.2-XFSsmp-3mar2001 kernel. The network is 100Mbit
>> ethernet (3c905 I think). Reiserfs was not timed as it was unable to
>> make >2G files(!!)

>> time dd if=/dev/zero of=bigFile bs=1024k count=2500

>>             writes to local disk   writes over nfs   notes
>> xfs                16 MB/s            0.7 MB/s    load ~90-100% per cpu + (*)
>> xfs over RAID0     37 MB/s             1  MB/s       ""
>> xfs over RAID5            didn't try yet
>> ext2               17 MB/s             8  MB/s    load ~30-40% per cpu
>> ext2 over RAID0    38 MB/s             8  MB/s       ""
>> ext2 over RAID5    14 MB/s             6  MB/s       "" + (**)

> i just did a very simple test on the setup i can us right now from
> remote and end up at about 2mb/sec (server: up 333mhz pII, 128mb,
> 6gb udma2 disk - so - no big hardware, ah - i forgot: k6-2/400, 64mb
> solaris8 ix86 client) - looks like something is definitely not ok
> with your setup (otherwise i would not get better results with my
> much slower hardware) - so - some things to try:

another datapoint (same setup as above):

              -------Sequential Output-------- ---Sequential Input-- --Random--
              -Per Char- --Block--- -Rewrite-- -Per Char- --Block--- --Seeks---
Machine    MB K/sec %CPU K/sec %CPU K/sec %CPU K/sec %CPU K/sec %CPU  /sec %CPU
          128  4864 33.6  4501 13.0  3484 17.2  4701 28.8  5554 14.4 109.4  7.2

this is bonnie over nfs - looks quite ok to me for that hardware - also
the nfsd on the server only eats up about 5% cpu while doing this - so
it really looks like something on your setup is not ok

t

-- 
thomas.graichen@innominate.com
                                                         innominate AG
                                                  the linux architects
tel: +49-30-308806-13   fax: -77             http://www.innominate.com

From owner-linux-xfs@oss.sgi.com Sun Mar  4 03:04:07 2001
Received:  by oss.sgi.com id <S553683AbRCDLD5>;
	Sun, 4 Mar 2001 03:03:57 -0800
Received: from groucho.maths.monash.edu.au ([130.194.160.211]:21030 "EHLO
        groucho.maths.monash.edu.au") by oss.sgi.com with ESMTP
	id <S553664AbRCDLDc>; Sun, 4 Mar 2001 03:03:32 -0800
Received: (from rjh@localhost)
	by groucho.maths.monash.edu.au (8.8.8/8.8.8) id LAA23941
	for linux-xfs@oss.sgi.com; Sun, 4 Mar 2001 11:03:30 GMT
From:   Robin Humble <rjh@groucho.maths.monash.edu.au>
Message-Id: <200103041103.LAA23941@groucho.maths.monash.edu.au>
Subject: Re: NFS performance - more info
To:     linux-xfs@oss.sgi.com
Date:   Sun, 4 Mar 2001 22:03:30 +1100 (EDT)
In-Reply-To: <news2mail-97t0uh$5qj$1@mate.bln.innominate.de> from "Thomas Graichen" at Mar 04, 2001 09:12:49 AM
X-Mailer: ELM [version 2.5 PL1]
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing


Thomas Graichen writes:
>Robin Humble <rjh@groucho.maths.monash.edu.au> wrote:
>* did you try kiocluster's (only work with ide or scsi not with md
>  or lvm) - no idea if it might help -> mount -o kiocluster ...

No, I didn't try this. I need to use RAID in the end so it's not
much use to me... :-/

>* the other good lines posted often here (but i don't think they
>  will help here) -> mount -o logbufs=4,logbsize=32768 ...
>* creating the filesystem with a bigger journal is also a good idea
>  for real workload (ok - i think not in this case too - because
>  here is not much metadata to journal :-)
>  -> mkfs -t xfs -l size=32768b ...

yeah - good idea - I could try that, but like you say - I don't know
why they would help. It's just one file being written.

>* did you try the same with an up kernel? - maybe some smp locking
>  thing somethere

I just tried the 'dd' between my home machines (server is
Athlon700/ATA66 single processor, single disk, RH6.2, older 2.4.2 cvs
kernel, and the client is p120 laptop RH6.2) and found much the same
behaviour. However because I only have a 10Mbit network at home, the
nfsd loads didn't get high enough to choke the transfers like they do
with the 100Mbit connections at work. At home, writing to either XFS
or ext2 with a dd achieved about ~1MB/s which is what you'd expect.
  time dd if=/dev/zero of=/mnt/xfs/bigFile bs=1M count=256
takes ~270s or so.

For the XFS writes, the nfsd load was about 25% total (4 daemons,
~5.5% each), and it was only about 3% total for ext2 (<1% per nfsd).
So I guess this problem only really shows up on 100Mbit or greater
networks where the traffic to the daemons is 10x higher.

So yeah, because I see roughly the same nfsd load behaviour on a
single procesor box, think that SMP isn't much of an issue, but yes I
should test it. Is there a way to boot an SMP kernel in uni-processor
mode or should I recompile?

Your bonnie setup (in the other email) is possibly a bit flawed in
that it writes only 128M files and the server has 128M of memory, so
in theory the server doesn't have to write to disk at all. Could you
maybe try it again with a file size that's more than twice the server
memory size? That's what bonnie++ tells you to do anyway, I don't
know about bonnie.
Are you using NFSv3 on the server/client? nfsstat will let you know.

cheers,
robin

From owner-linux-xfs@oss.sgi.com Sun Mar  4 03:32:57 2001
Received:  by oss.sgi.com id <S553686AbRCDLcs>;
	Sun, 4 Mar 2001 03:32:48 -0800
Received: from ns.caldera.de ([212.34.180.1]:18955 "EHLO ns.caldera.de")
	by oss.sgi.com with ESMTP id <S553679AbRCDLcY>;
	Sun, 4 Mar 2001 03:32:24 -0800
Received: (from hch@localhost)
	by ns.caldera.de (8.9.3/8.9.3) id MAA15481
	for linux-xfs@oss.sgi.com; Sun, 4 Mar 2001 12:32:02 +0100
Date:   Sun, 4 Mar 2001 12:32:02 +0100
From:   Christoph Hellwig <hch@caldera.de>
To:     linux-xfs@oss.sgi.com
Subject: [PATCH] kill lookup_vfsmnt()
Message-ID: <20010304123202.A14717@caldera.de>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
X-Mailer: Mutt 1.0i
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing

Hi,

the lookup_vfsmnt function in the xfs tree is broken by design
as it tries to get a struct vfsmnt for a given kdev_t but a
filesystem can be mounted at multiple mount points in Linux 2.4
and so the mapping is not unique.

The actual usage of the function is not broken as it is used only
to get the struct super_block from the vfsmnt which is again
unique for a kdev_t (but doesn't have to - I know of at least one
fs design where there might be multiple sb's for a dev).

Suggestion: kill lookup_vfsmnt and use get_super instead.
Patch attached.

	Christoph

-- 
Of course it doesn't work. We've performed a software upgrade.


Index: fs/super.c
===================================================================
RCS file: /cvs/linux-2.4-xfs/linux/fs/super.c,v
retrieving revision 1.42
diff -u -b -B -r1.42 super.c
--- fs/super.c	2001/02/22 21:09:04	1.42
+++ fs/super.c	2001/03/03 18:29:44
@@ -456,24 +456,6 @@
         return sp - buf;	/* no trailing NUL */
 }
 
-struct vfsmount *lookup_vfsmnt(kdev_t dev)
-{
-	struct list_head *p;
-
-	if (list_empty(&vfsmntlist))
-		return NULL;
-
-	for (p = vfsmntlist.next; p != &vfsmntlist; p = p->next) {
-		struct vfsmount *tmp = list_entry(p, struct vfsmount, mnt_list);
-		if (tmp->mnt_sb->s_dev == dev) {
-			return (tmp);
-		}
-	}
-
-	return NULL;
-}
-
-
 static struct proc_fs_info {
 	int flag;
 	char *str;
Index: fs/xfs/linux/xfs_vfs.c
===================================================================
RCS file: /cvs/linux-2.4-xfs/linux/fs/xfs/linux/xfs_vfs.c,v
retrieving revision 1.21
diff -u -b -B -r1.21 xfs_vfs.c
--- fs/xfs/linux/xfs_vfs.c	2000/11/22 06:53:25	1.21
+++ fs/xfs/linux/xfs_vfs.c	2001/03/03 18:30:20
@@ -211,14 +211,16 @@
 	long s;
 	struct vfs *vfsp;
 	kdev_t	kdev = MKDEV(MAJOR(dev), MINOR(dev));
-	struct vfsmount *entry;
+	struct super_block *sb;
 
 	lock_kernel();
-	entry = lookup_vfsmnt(kdev);
-
+	sb = get_super(kdev);
 	unlock_kernel();
-	if (entry) {
-		vfsp = LINVFS_GET_VFS(entry->mnt_sb);
+
+	if (!sb)
+		return NULL;
+	
+	vfsp = LINVFS_GET_VFS(sb);
 again:
 		spin_lock_irqsave(&vfslock, s);
 		if (vfsp->vfs_dev == dev &&
@@ -241,9 +243,6 @@
 		}
 		spin_unlock_irqrestore(&vfslock, s);
 		return vfsp;
-	} else {
-		return NULL;
-	}
 }
 
 void
@@ -312,14 +311,13 @@
 {
         register struct vfs *vfsp;
 	kdev_t	kdev = MKDEV(MAJOR(dev), MINOR(dev));
-	struct vfsmount *entry;
+	struct super_block *sb;
 
 	lock_kernel();
-	entry = lookup_vfsmnt(kdev);
-
+	sb = get_super(kdev);
 
-        if (entry) {
-		vfsp = LINVFS_GET_VFS(entry->mnt_sb);
+	if (sb) {
+		vfsp = LINVFS_GET_VFS(sb);
                 if ((vfsp->vfs_dev == dev) &&
                     (fstype == VFS_FSTYPE_ANY || fstype == vfsp->vfs_fstype)) {
 			unlock_kernel();
Index: include/linux/mount.h
===================================================================
RCS file: /cvs/linux-2.4-xfs/linux/include/linux/mount.h,v
retrieving revision 1.12
diff -u -b -B -r1.12 mount.h
--- include/linux/mount.h	2000/10/23 18:56:35	1.12
+++ include/linux/mount.h	2001/03/03 18:30:35
@@ -38,7 +38,6 @@
 		atomic_inc(&mnt->mnt_count);
 	return mnt;
 }
-struct vfsmount *lookup_vfsmnt(kdev_t dev);
 
 static inline void mntput(struct vfsmount *mnt)
 {
Index: kernel/ksyms.c
===================================================================
RCS file: /cvs/linux-2.4-xfs/linux/kernel/ksyms.c,v
retrieving revision 1.80
diff -u -b -B -r1.80 ksyms.c
--- kernel/ksyms.c	2001/03/01 21:13:39	1.80
+++ kernel/ksyms.c	2001/03/03 18:30:40
@@ -143,7 +143,6 @@
 EXPORT_SYMBOL(get_fs_type);
 EXPORT_SYMBOL(get_super);
 EXPORT_SYMBOL(get_empty_super);
-EXPORT_SYMBOL(lookup_vfsmnt);
 EXPORT_SYMBOL(getname);
 EXPORT_SYMBOL(names_cachep);
 EXPORT_SYMBOL(fput);

From owner-linux-xfs@oss.sgi.com Sun Mar  4 04:37:58 2001
Received:  by oss.sgi.com id <S553686AbRCDMhj>;
	Sun, 4 Mar 2001 04:37:39 -0800
Received: from hermes.mixx.net ([212.84.196.2]:19464 "HELO hermes.mixx.net")
	by oss.sgi.com with SMTP id <S553679AbRCDMhT>;
	Sun, 4 Mar 2001 04:37:19 -0800
Received: from mate.bln.innominate.de (cerberus.berlin.innominate.de [212.84.234.251])
	by hermes.mixx.net (Postfix) with ESMTP id 0252DF817
	for <linux-xfs@oss.sgi.com>; Sun,  4 Mar 2001 13:37:14 +0100 (CET)
Received: by mate.bln.innominate.de (Postfix, from userid 9)
	id CA07C2CA6F; Sun,  4 Mar 2001 13:37:13 +0100 (CET)
From:   Thomas Graichen <news-innominate.list.sgi.xfs@innominate.de>
Reply-To: Thomas Graichen <thomas.graichen@innominate.com>
X-Newsgroups: innominate.list.sgi.xfs
Subject: Re: NFS performance - more info
Date:   4 Mar 2001 12:37:13 GMT
Organization: innominate AG, Berlin, Germany
Lines:  69
Distribution: local
Message-ID: <news2mail-97tctp$pm3$1@mate.bln.innominate.de>
References: <news2mail-97t0uh$5qj$1@mate.bln.innominate.de> <200103041103.LAA23941@groucho.maths.monash.edu.au>
X-Trace: mate.bln.innominate.de 983709433 26307 10.0.0.31 (4 Mar 2001 12:37:13 GMT)
X-Complaints-To: news@innominate.de
User-Agent: tin/1.4.4-20000803 ("Vet for the Insane") (UNIX) (Linux/2.4.2-XFS (i686))
To:     linux-xfs@oss.sgi.com
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing

Robin Humble <rjh@groucho.maths.monash.edu.au> wrote:

> Thomas Graichen writes:
>>Robin Humble <rjh@groucho.maths.monash.edu.au> wrote:
>>* did you try kiocluster's (only work with ide or scsi not with md
>>  or lvm) - no idea if it might help -> mount -o kiocluster ...

> No, I didn't try this. I need to use RAID in the end so it's not
> much use to me... :-/

but it might be a good idea to try it anyway to circle the problem
down a bit

>>* the other good lines posted often here (but i don't think they
>>  will help here) -> mount -o logbufs=4,logbsize=32768 ...
>>* creating the filesystem with a bigger journal is also a good idea
>>  for real workload (ok - i think not in this case too - because
>>  here is not much metadata to journal :-)
>>  -> mkfs -t xfs -l size=32768b ...

> yeah - good idea - I could try that, but like you say - I don't know
> why they would help. It's just one file being written.

>>* did you try the same with an up kernel? - maybe some smp locking
>>  thing somethere

> I just tried the 'dd' between my home machines (server is
> Athlon700/ATA66 single processor, single disk, RH6.2, older 2.4.2 cvs
> kernel, and the client is p120 laptop RH6.2) and found much the same
> behaviour. However because I only have a 10Mbit network at home, the
> nfsd loads didn't get high enough to choke the transfers like they do
> with the 100Mbit connections at work. At home, writing to either XFS
> or ext2 with a dd achieved about ~1MB/s which is what you'd expect.
>   time dd if=/dev/zero of=/mnt/xfs/bigFile bs=1M count=256
> takes ~270s or so.

> For the XFS writes, the nfsd load was about 25% total (4 daemons,
> ~5.5% each), and it was only about 3% total for ext2 (<1% per nfsd).
> So I guess this problem only really shows up on 100Mbit or greater
> networks where the traffic to the daemons is 10x higher.

> So yeah, because I see roughly the same nfsd load behaviour on a
> single procesor box, think that SMP isn't much of an issue, but yes I
> should test it. Is there a way to boot an SMP kernel in uni-processor
> mode or should I recompile?

> Your bonnie setup (in the other email) is possibly a bit flawed in
> that it writes only 128M files and the server has 128M of memory, so
> in theory the server doesn't have to write to disk at all. Could you
> maybe try it again with a file size that's more than twice the server
> memory size? That's what bonnie++ tells you to do anyway, I don't
> know about bonnie.
> Are you using NFSv3 on the server/client? nfsstat will let you know.

oh yes - you are right - i took 128 because the client had 64 - but
you are right the server has 128 and that counts here ... it'S v3
btw.

one other idea: does the size of the files matter - so do you get
different results for say 25, 250 and 2500mb (always first run to
ignore caching)?

t

-- 
thomas.graichen@innominate.com
                                                         innominate AG
                                                  the linux architects
tel: +49-30-308806-13   fax: -77             http://www.innominate.com

From owner-linux-xfs@oss.sgi.com Sun Mar  4 04:51:28 2001
Received:  by oss.sgi.com id <S553686AbRCDMvI>;
	Sun, 4 Mar 2001 04:51:08 -0800
Received: from hermes.mixx.net ([212.84.196.2]:32008 "HELO hermes.mixx.net")
	by oss.sgi.com with SMTP id <S553679AbRCDMui>;
	Sun, 4 Mar 2001 04:50:38 -0800
Received: from mate.bln.innominate.de (cerberus.berlin.innominate.de [212.84.234.251])
	by hermes.mixx.net (Postfix) with ESMTP id 79B4EF817
	for <linux-xfs@oss.sgi.com>; Sun,  4 Mar 2001 13:50:36 +0100 (CET)
Received: by mate.bln.innominate.de (Postfix, from userid 9)
	id 299342CA6F; Sun,  4 Mar 2001 13:50:36 +0100 (CET)
From:   Thomas Graichen <news-innominate.list.sgi.xfs@innominate.de>
Reply-To: Thomas Graichen <thomas.graichen@innominate.com>
X-Newsgroups: innominate.list.sgi.xfs
Subject: Re: NFS performance - more info
Date:   4 Mar 2001 12:50:36 GMT
Organization: innominate AG, Berlin, Germany
Lines:  36
Distribution: local
Message-ID: <news2mail-97tdms$pm3$2@mate.bln.innominate.de>
References: <news2mail-97t0uh$5qj$1@mate.bln.innominate.de> <200103041103.LAA23941@groucho.maths.monash.edu.au> <97tctp$pm3$1@mate.bln.innominate.de>
X-Trace: mate.bln.innominate.de 983710236 26307 10.0.0.31 (4 Mar 2001 12:50:36 GMT)
X-Complaints-To: news@innominate.de
User-Agent: tin/1.4.4-20000803 ("Vet for the Insane") (UNIX) (Linux/2.4.2-XFS (i686))
To:     linux-xfs@oss.sgi.com
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing

Thomas Graichen <thomas.graichen@innominate.com> wrote:

>> Your bonnie setup (in the other email) is possibly a bit flawed in
>> that it writes only 128M files and the server has 128M of memory, so
>> in theory the server doesn't have to write to disk at all. Could you
>> maybe try it again with a file size that's more than twice the server
>> memory size? That's what bonnie++ tells you to do anyway, I don't
>> know about bonnie.
>> Are you using NFSv3 on the server/client? nfsstat will let you know.

> oh yes - you are right - i took 128 because the client had 64 - but
> you are right the server has 128 and that counts here ... it'S v3
> btw.

> one other idea: does the size of the files matter - so do you get
> different results for say 25, 250 and 2500mb (always first run to
> ignore caching)?

ok - again with 256mb

              -------Sequential Output-------- ---Sequential Input-- --Random--
              -Per Char- --Block--- -Rewrite-- -Per Char- --Block--- --Seeks---
Machine    MB K/sec %CPU K/sec %CPU K/sec %CPU K/sec %CPU K/sec %CPU  /sec %CPU
          256  5179 35.9  5276 16.5  3501 17.8  4724 30.8  5672 15.5  77.4  5.1

but i just noticed it was nfsv2 (the typical redhat no3 in the nfs script)

will try 3 tomorrow - have to go now

t

-- 
thomas.graichen@innominate.com
                                                         innominate AG
                                                  the linux architects
tel: +49-30-308806-13   fax: -77             http://www.innominate.com

From owner-linux-xfs@oss.sgi.com Sun Mar  4 10:34:00 2001
Received:  by oss.sgi.com id <S553738AbRCDSdk>;
	Sun, 4 Mar 2001 10:33:40 -0800
Received: from sgi.SGI.COM ([192.48.153.1]:46207 "EHLO sgi.com")
	by oss.sgi.com with ESMTP id <S553733AbRCDSdZ>;
	Sun, 4 Mar 2001 10:33:25 -0800
Received: from cthulhu.engr.sgi.com (cthulhu.engr.sgi.com [192.26.80.2]) 
	by sgi.com (980327.SGI.8.8.8-aspam/980304.SGI-aspam:
       SGI does not authorize the use of its proprietary
       systems or networks for unsolicited or bulk email
       from the Internet.) 
	via ESMTP id KAA03010
	for <linux-xfs@oss.sgi.com>; Sun, 4 Mar 2001 10:33:25 -0800 (PST)
	mail_from (chait@getafix.engr.sgi.com)
Received: from getafix.engr.sgi.com (IDENT:root@getafix.engr.sgi.com [163.154.5.110])
	by cthulhu.engr.sgi.com (SGI-8.9.3/8.9.3) with ESMTP id KAA99494;
	Sun, 4 Mar 2001 10:32:08 -0800 (PST)
Received: from localhost (chait@localhost)
	by getafix.engr.sgi.com (8.9.3/8.9.3) with ESMTP id KAA28916;
	Sun, 4 Mar 2001 10:27:56 -0500
Date:   Sun, 4 Mar 2001 10:27:56 -0500 (EST)
From:   Chaitanya Tumuluri <chait@getafix.engr.sgi.com>
To:     Thomas Graichen <thomas.graichen@innominate.com>
cc:     linux-xfs@oss.sgi.com
Subject: Re: NFS performance - more info
In-Reply-To: <news2mail-97tctp$pm3$1@mate.bln.innominate.de>
Message-ID: <Pine.LNX.4.21.0103041026120.27128-100000@getafix.engr.sgi.com>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing

On 4 Mar 2001, Thomas Graichen wrote:
> Robin Humble <rjh@groucho.maths.monash.edu.au> wrote:
> 
> > Thomas Graichen writes:
> >>Robin Humble <rjh@groucho.maths.monash.edu.au> wrote:
> >>* did you try kiocluster's (only work with ide or scsi not with md
> >>  or lvm) - no idea if it might help -> mount -o kiocluster ...
> 
> > No, I didn't try this. I need to use RAID in the end so it's not
> > much use to me... :-/
> 
> but it might be a good idea to try it anyway to circle the problem
> down a bit

I'll second that. Also, if it does turn out to give better performance,
then it could provide us motivation to make md aware of kiobufs. We're
still in the early stages of doing this, BTW.

Cheers,
-Chait.


From owner-linux-xfs@oss.sgi.com Sun Mar  4 13:48:43 2001
Received:  by oss.sgi.com id <S553748AbRCDVsd>;
	Sun, 4 Mar 2001 13:48:33 -0800
Received: from 40bcc504.dsl.flashcom.net ([64.188.197.4]:48496 "EHLO flux")
	by oss.sgi.com with ESMTP id <S553744AbRCDVsF>;
	Sun, 4 Mar 2001 13:48:05 -0800
Received: from itp by flux with local (Exim 3.12 #1 (Debian))
	id 14ZgM9-0008RR-00
	for <linux-xfs@oss.sgi.com>; Sun, 04 Mar 2001 16:47:57 -0500
Date:   Sun, 4 Mar 2001 16:47:57 -0500
To:     linux-xfs@oss.sgi.com
Subject: xfs memory usage (reporting?)
Message-ID: <20010304164757.A32419@flux>
Mail-Followup-To: linux-xfs@oss.sgi.com
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.3.15i
From:   Ian Peters <itp@ximian.com>
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing

Before I start the question, let me say I've been lurking on the list
for a while and just took the plunge to install xfs on my laptop.
Wow.  You guys have done an excellent job.

I was playing around with the system, creating directories with lots
of dummy files (~100000), deleting them, selectively deleting files,
etc., and I noticed that my memory usage was way high and not
dropping.  I puzzled over it for a while, trying to get a handle on
what situations cause memory usage to climb, and noticed the following
anomoly:

itp@flux:~$ free
             total       used       free     shared    buffers     cached
Mem:        255764     252080       3684          0          0      69020
-/+ buffers/cache:     183060      72704
Swap:        68032          0      68032

I was somewhat surprised to see shared and buffers both 0.  Playing
around with it some more, it seems like much of the memory in use is
acting like it should be in buffers, or shared, but is getting
reported as used instead.

If I start something memory intensive and then quit, I find that the
used memory drops back down significantly, indicating to me that some
of the memory previously in use before starting the program has been
flushed, which sounds more like memory used as buffers.

Is this behaviour to be expected?  It appears to be a cosmetic problem
only, as memory usage tools make things look dire, when in fact my
system will continue to run normally.

-- 
Ian Peters      "...it is 5 am and the sun has charred the other
itp@gnu.org      side of the earth and come back to us and painted
itp@ximian.com   the smoke over our heads an imperial violet..."

From owner-linux-xfs@oss.sgi.com Sun Mar  4 18:26:25 2001
Received:  by oss.sgi.com id <S553764AbRCEC0P>;
	Sun, 4 Mar 2001 18:26:15 -0800
Received: from 40bcc504.dsl.flashcom.net ([64.188.197.4]:51314 "EHLO flux")
	by oss.sgi.com with ESMTP id <S553652AbRCECZx>;
	Sun, 4 Mar 2001 18:25:53 -0800
Received: from itp by flux with local (Exim 3.12 #1 (Debian))
	id 14Zkgx-00061A-00
	for <linux-xfs@oss.sgi.com>; Sun, 04 Mar 2001 21:25:43 -0500
Date:   Sun, 4 Mar 2001 21:25:43 -0500
To:     linux-xfs@oss.sgi.com
Subject: Re: xfs memory usage (reporting?)
Message-ID: <20010304212543.A23121@flux>
Mail-Followup-To: linux-xfs@oss.sgi.com
References: <20010304164757.A32419@flux>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.3.15i
In-Reply-To: <20010304164757.A32419@flux>; from itp@ximian.com on Sun, Mar 04, 2001 at 04:47:57PM -0500
From:   Ian Peters <itp@ximian.com>
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing

On Sun, Mar 04, 2001 at 04:47:57PM -0500, Ian Peters wrote:
> Before I start the question, let me say I've been lurking on the list
> for a while and just took the plunge to install xfs on my laptop.
> Wow.  You guys have done an excellent job.

I apologize for failing to mention this, but my kernel is current to
HEAD from cvs.

-- 
Ian Peters      "...it is 5 am and the sun has charred the other
itp@gnu.org      side of the earth and come back to us and painted
itp@ximian.com   the smoke over our heads an imperial violet..."

From owner-linux-xfs@oss.sgi.com Sun Mar  4 19:28:55 2001
Received:  by oss.sgi.com id <S553765AbRCED2f>;
	Sun, 4 Mar 2001 19:28:35 -0800
Received: from 40bcc504.dsl.flashcom.net ([64.188.197.4]:16240 "EHLO
        rain.ximian.com") by oss.sgi.com with ESMTP id <S553652AbRCED2M>;
	Sun, 4 Mar 2001 19:28:12 -0800
Received: from vladimir by rain.ximian.com with local (Exim 3.12 #1 (Debian))
	id 14ZlfN-00051A-00
	for <linux-xfs@oss.sgi.com>; Sun, 04 Mar 2001 22:28:09 -0500
From:   Vladimir Vukicevic <vladimir@ximian.com>
To:     linux-xfs@oss.sgi.com
Subject: questions about some benchmarks
Message-Id: <E14ZlfN-00051A-00@rain.ximian.com>
Date:   Sun, 04 Mar 2001 22:28:09 -0500
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing


Howdy.. I just converted my laptop's root partition to XFS; while
doing some hacking I ended up doing a quick-and-dirty benchmark
between XFS and ext2.

The freedb database (www.freedb.org) is a 59MB .tar.bz2 that expands
out into 1.1GB worth of little files, all 1-2K in size, mostly towards
the 1K size. (Why they use the filesystem as the database and not use
libdb or something is beyond me. Anyway.) So, I timed the untar and
rm -rf times of this tree.

The two machines involved are an IBM X20 laptop, P3-600, 392MB ram,
20GB IBM Travelstar 4200 RPM drive, UDMA mode3, xfs filesystem. On my
desktop, which is an Athlon 700, VIA VT82Cxxx Apollo chipset, 512MB
ram, tar file on an UATA/100 drive on the primary on-board interface,
destination drive on a promise UATA/100 IDE controller with an IBM
Deskstar 7200 RPM 45GB drive on it, with an ext2 filesystem:

On my laptop (xfs),

tar xjf ../freedb-complete-20010219.tar.bz2  138.88s user
  207.17s system 27% cpu 21:17.84 total 
rm -rf db  1.45s user 130.99s system 9% cpu 24:24.66 total

On my desktop (ext2),

tar xjf ~vladimir/freedb-complete-20010219.tar.bz2  153.76s user
  2772.26s system 98% cpu 49:28.47 total
rm -rf db  0.16s user 73.39s system 94% cpu 1:18.24 total

Now, I converted the filesystem on my desktop to XFS.  Here's the same
numbers again, this time for xfs.

tar xjf ~vladimir/freedb-complete-20010219.tar.bz2  156.77s user
  199.28s system 61% cpu 9:41.88 total 
rm -rf db  1.73s user 131.59s system 19% cpu 11:07.25 total


So:

System                      tar time  cpu      rm time    cpu
P3-600/4200rpm, xfs         21:17.84  27%      24:24.66    9%
K7-700/7200rpm, xfs          9:41.88  61%      11:07.25   19%
K7-700/7200rpm, ext2        49:28.47  98%       1:18.24   94%

So, while xfs obviously kicks ext2's butt, I'm curious as to the cpu
usage numbers of both the untar and the remove, and also of xfs's
significantly slower remove times (even though its cpu usage was
also significantly lower).. is there anything about xfs that
causes removes to be significantly slower?  (I just realized that
I didn't time a 'sync' after the relevant removes though, so
a good chunk of this could have even been buffered, maybe...)

I realize that this was an ad-hoc benchmark, but the numbers seemed
interesting nonetheless. (If anyone's interested in testing the same,
the database is available at ftp://ftp.freedb.org/pub/freedb/ ).

    - Vlad

From owner-linux-xfs@oss.sgi.com Sun Mar  4 19:54:35 2001
Received:  by oss.sgi.com id <S553767AbRCEDyZ>;
	Sun, 4 Mar 2001 19:54:25 -0800
Received: from taz.cs.utah.edu ([155.99.203.51]:20698 "HELO taz.cs.utah.edu")
	by oss.sgi.com with SMTP id <S553681AbRCEDyD>;
	Sun, 4 Mar 2001 19:54:03 -0800
Received: by taz.cs.utah.edu (Mailer, from userid 1363)
	id D6B10B27748; Sun,  4 Mar 2001 20:54:01 -0700 (MST)
Message-Id: <10103042054.ZM378885@taz.cs.utah.edu>
In-Reply-To: Vladimir Vukicevic <vladimir@ximian.com>
        "questions about some benchmarks" (Mar  4, 10:28pm)
References: <E14ZlfN-00051A-00@rain.ximian.com>
X-Mailer: Z-Mail-SGI (3.2S.3 08feb96 MediaMail)
To:     Vladimir Vukicevic <vladimir@ximian.com>, linux-xfs@oss.sgi.com
Subject: Re: questions about some benchmarks
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Date:   Sun,  4 Mar 2001 20:54:01 -0700 (MST)
From:   sparker@cs.utah.edu (Steven G. Parker)
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing

Hopefully someone will get back to you with a technical explanation,
but the rm times for XFS have always been slow - even under IRIX.
For example, the untar time that you presented is *less than* the
rm time.  That is particularly annoying when running an untar at
the same time as an rm of another directory with the hope that disk
won't fill up.  Unfortunately, it can fill the disk faster than
it can clean it off, which is a little suprising the first time
you run out of disk space because of this.

XFS does have to log the removes, but it seems like that should be
more like a factor of 2 hit than the factor of 10 that you got
over ext2.

Steve
(Another one interested in the real answer)


On Mar 4, 10:28pm, Vladimir Vukicevic wrote:
> Subject: questions about some benchmarks
>
> Howdy.. I just converted my laptop's root partition to XFS; while
> doing some hacking I ended up doing a quick-and-dirty benchmark
> between XFS and ext2.
>
> The freedb database (www.freedb.org) is a 59MB .tar.bz2 that expands
> out into 1.1GB worth of little files, all 1-2K in size, mostly towards
> the 1K size. (Why they use the filesystem as the database and not use
> libdb or something is beyond me. Anyway.) So, I timed the untar and
> rm -rf times of this tree.
>
> The two machines involved are an IBM X20 laptop, P3-600, 392MB ram,
> 20GB IBM Travelstar 4200 RPM drive, UDMA mode3, xfs filesystem. On my
> desktop, which is an Athlon 700, VIA VT82Cxxx Apollo chipset, 512MB
> ram, tar file on an UATA/100 drive on the primary on-board interface,
> destination drive on a promise UATA/100 IDE controller with an IBM
> Deskstar 7200 RPM 45GB drive on it, with an ext2 filesystem:
>
> On my laptop (xfs),
>
> tar xjf ../freedb-complete-20010219.tar.bz2  138.88s user
>   207.17s system 27% cpu 21:17.84 total
> rm -rf db  1.45s user 130.99s system 9% cpu 24:24.66 total
>
> On my desktop (ext2),
>
> tar xjf ~vladimir/freedb-complete-20010219.tar.bz2  153.76s user
>   2772.26s system 98% cpu 49:28.47 total
> rm -rf db  0.16s user 73.39s system 94% cpu 1:18.24 total
>
> Now, I converted the filesystem on my desktop to XFS.  Here's the same
> numbers again, this time for xfs.
>
> tar xjf ~vladimir/freedb-complete-20010219.tar.bz2  156.77s user
>   199.28s system 61% cpu 9:41.88 total
> rm -rf db  1.73s user 131.59s system 19% cpu 11:07.25 total
>
>
> So:
>
> System                      tar time  cpu      rm time    cpu
> P3-600/4200rpm, xfs         21:17.84  27%      24:24.66    9%
> K7-700/7200rpm, xfs          9:41.88  61%      11:07.25   19%
> K7-700/7200rpm, ext2        49:28.47  98%       1:18.24   94%
>
> So, while xfs obviously kicks ext2's butt, I'm curious as to the cpu
> usage numbers of both the untar and the remove, and also of xfs's
> significantly slower remove times (even though its cpu usage was
> also significantly lower).. is there anything about xfs that
> causes removes to be significantly slower?  (I just realized that
> I didn't time a 'sync' after the relevant removes though, so
> a good chunk of this could have even been buffered, maybe...)
>
> I realize that this was an ad-hoc benchmark, but the numbers seemed
> interesting nonetheless. (If anyone's interested in testing the same,
> the database is available at ftp://ftp.freedb.org/pub/freedb/ ).
>
>     - Vlad
>-- End of excerpt from Vladimir Vukicevic



From owner-linux-xfs@oss.sgi.com Sun Mar  4 22:02:35 2001
Received:  by oss.sgi.com id <S553775AbRCEGC0>;
	Sun, 4 Mar 2001 22:02:26 -0800
Received: from pneumatic-tube.sgi.com ([204.94.214.22]:29540 "EHLO
        pneumatic-tube.sgi.com") by oss.sgi.com with ESMTP
	id <S553769AbRCEGCR>; Sun, 4 Mar 2001 22:02:17 -0800
Received: from snort.melbourne.sgi.com (snort.melbourne.sgi.com [134.14.55.149]) by pneumatic-tube.sgi.com (980327.SGI.8.8.8-aspam/980310.SGI-aspam) via ESMTP id WAA04673
	for <linux-xfs@oss.sgi.com>; Sun, 4 Mar 2001 22:11:57 -0800 (PST)
	mail_from (nathans@snort.melbourne.sgi.com)
Received: (from nathans@localhost)
	by snort.melbourne.sgi.com (SGI-8.9.3/8.9.3) id RAA84669
	for linux-xfs@oss.sgi.com; Mon, 5 Mar 2001 17:00:57 +1100 (EST)
Date:   Mon, 5 Mar 2001 17:00:57 +1100 (EST)
From:   Nathan Scott <nathans@snort.melbourne.sgi.com>
Message-Id: <200103050600.RAA84669@snort.melbourne.sgi.com>
To:     linux-xfs@oss.sgi.com
Subject: TAKE - quota
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing

Date:  Sun Mar  4 22:00:33 PST 2001
Workarea:  snort.melbourne.sgi.com:/diskb/build4/nathans/linux-xfs

The following file(s) were checked into:
  bonnie.engr.sgi.com:/isms/slinx/2.4.x-xfs


Modid:  2.4.x-xfs:slinx:88862a
cmd/quota/README.xfs - 1.8
	- revert edquota -p behaviour - Jan has convinced me.

cmd/quota/edquota.c - 1.6
	- revert edquota -p behaviour - Jan has convinced me, plus some minor
	  changes to aid in merging.

cmd/quota/quota.c - 1.6
cmd/quota/quotacheck.c - 1.5
cmd/quota/quotaon.c - 1.11
	- small changes to aid in merging of new vfs quota format later.

cmd/quota/repquota.c - 1.14
	- remove an unused struct field.

cmd/quota/setquota.c - 1.7
	- small changes to aid in merging of new vfs quota format later.



From owner-linux-xfs@oss.sgi.com Mon Mar  5 08:57:01 2001
Received:  by oss.sgi.com id <S553791AbRCEQ4v>;
	Mon, 5 Mar 2001 08:56:51 -0800
Received: from deliverator.sgi.com ([204.94.214.10]:39985 "EHLO
        deliverator.sgi.com") by oss.sgi.com with ESMTP id <S553673AbRCEQ4g>;
	Mon, 5 Mar 2001 08:56:36 -0800
Received: from zeus-fddi.americas.sgi.com (zeus-fddi.americas.sgi.com [128.162.8.103]) by deliverator.sgi.com (980309.SGI.8.8.8-aspam-6.2/980310.SGI-aspam) via ESMTP id IAA19705
	for <linux-xfs@oss.sgi.com>; Mon, 5 Mar 2001 08:55:30 -0800 (PST)
	mail_from (lord@sgi.com)
Received: from daisy-e185.americas.sgi.com (daisy-e194.americas.sgi.com [128.162.194.214]) by zeus-fddi.americas.sgi.com (8.9.3/americas-smart-nospam1.1) with ESMTP id KAA1023879 for <linux-xfs@oss.sgi.com>; Mon, 5 Mar 2001 10:55:19 -0600 (CST)
Received: from jen.americas.sgi.com (IDENT:root@jen.americas.sgi.com [128.162.184.86]) by daisy-e185.americas.sgi.com (SGI-8.9.3/SGI-server-1.7) with ESMTP id KAA10531 for <linux-xfs@oss.sgi.com>; Mon, 5 Mar 2001 10:55:19 -0600 (CST)
From:   Steve Lord <lord@sgi.com>
Received: by jen.americas.sgi.com (8.11.0/SGI-client-1.7) id f25GsbI16800; Mon, 5 Mar 2001 10:54:37 -0600
Message-Id: <200103051654.f25GsbI16800@jen.americas.sgi.com>
Date:   Mon, 5 Mar 2001 10:54:37 -0600
Subject: TAKE - implement direct read and write in XFS
To:     unlisted-recipients:; (no To-header on input)
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing

This implements direct I/O for xfs on linux, read and write should function.
Testing has been limited so far. There is cache coherency between buffered
and direct I/O to the extent that the direct calls will flush or flush and
invalidate the cache before starting I/O. A systune has been added at
/proc/sys/vm/pagebuf/max_dio_pages which defines how many pages will be
locked down at once during an individual read or write call, this does not
limit the size of the I/O, just the size of the requests which will be
issued to the block layer.

Steve

Date:  Mon Mar  5 08:47:52 PST 2001
Workarea:  jen.americas.sgi.com:/src/lord/xfs-linux.2.4

The following file(s) were checked into:
  bonnie.engr.sgi.com:/isms/slinx/2.4.x-xfs


Modid:  2.4.x-xfs:slinx:88886a
linux/fs/xfs/xfs_rw.c - 1.330
	- fix xfs_zero_eof in direct I/O case

linux/fs/xfs/xfs_buf.h - 1.67
	- minor cleanup

linux/fs/xfs/linux/xfs_lrw.c - 1.79
	- Add bounds checking for direct I/O, do the cache invalidation for
	  data coherency on direct I/O.

linux/fs/xfs/linux/xfs_file.c - 1.39
	- Minor tweaking

linux/fs/xfs/linux/xfs_vnode.c - 1.48
	- remove unneeded functions

linux/fs/xfs/linux/xfs_fs_subr.c - 1.27
	- limit flush/toss and flushinval to when there is data to work on.

linux/fs/xfs/linux/xfs_iops.c - 1.95
	- fix dioinfo ioctl return values

linux/include/linux/page_buf.h - 1.78
	- Add systune for direct I/O maximum pages

linux/fs/xfs/linux/xfs_ioctl.c - 1.31
	- fix dioinfo ioctl return values

linux/fs/pagebuf/page_buf.c - 1.61
	- Add systune for the maximum amount of pages to lock down for one
	  direct I/O request

linux/fs/pagebuf/page_buf_io.c - 1.62
	- Implement direct read and write support

linux/fs/xfs/linux/xfs_vnode.h - 1.16
	- implement VN_CACHED, VN_DIRTY and VN_MAPPED as direct tests rather than
	  function calls.



From owner-linux-xfs@oss.sgi.com Mon Mar  5 08:57:21 2001
Received:  by oss.sgi.com id <S553792AbRCEQ5B>;
	Mon, 5 Mar 2001 08:57:01 -0800
Received: from chaos.egr.duke.edu ([152.3.195.82]:24849 "EHLO
        chaos.egr.duke.edu") by oss.sgi.com with ESMTP id <S553788AbRCEQ4p>;
	Mon, 5 Mar 2001 08:56:45 -0800
Received: from localhost (jlb@localhost)
	by chaos.egr.duke.edu (8.9.3/8.9.3) with ESMTP id LAA28644
	for <linux-xfs@oss.sgi.com>; Mon, 5 Mar 2001 11:56:39 -0500
X-Authentication-Warning: chaos.egr.duke.edu: jlb owned process doing -bs
Date:   Mon, 5 Mar 2001 11:56:39 -0500 (EST)
From:   Joshua Baker-LePain <jlb17@duke.edu>
X-Sender:  <jlb@chaos.egr.duke.edu>
To:     Linux xfs mailing list <linux-xfs@oss.sgi.com>
Subject: Getting started...
Message-ID: <Pine.LNX.4.30.0103051143120.7060-100000@chaos.egr.duke.edu>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing

I (well, OK, the lab) will soon be the proud papa of a big IDE-SCSI RAID
array.  Basically, it's a tower full of IDE disks, a RAID controller, and
a protocol adapter that makes it look like U2W SCSI to the host.  Fill it
with 8 80GB disks in a RAID5, and I'm looking at a 560GB partition (for
$6K, btw).  That, in my opinion excludes ext2 -- I'd rather not fsck
560GB, thank you very much.  I'll be exporting this via NFS over a full
duplex 100Mbit network to the users, and that seems to preclude ReiserFS.
Which brings me to xfs (which I also like, having cut my teeth on some SGI
boxen).

Which, finally, brings me to my question.  What is the current
conventional wisdom on the best way to get xfs up and running?  I'll
obviously be testing before I put this into production.  Should I be
grabbing the latest CVS snapshots, or using the 0.9 pre-release?  I'll
probably be running on RH7.0 (as it seems a bit harder to wedge 2.4.x onto
6.2) with the plan of moving to 7.1 (along with the rest of my network) a
short while after it comes out.  Any and all input is much appreciated.

Thank you, and thanks for bringing xfs to Linux.

-- 
Joshua Baker-LePain
Department of Biomedical Engineering
Duke University



From owner-linux-xfs@oss.sgi.com Mon Mar  5 09:09:51 2001
Received:  by oss.sgi.com id <S553788AbRCERJb>;
	Mon, 5 Mar 2001 09:09:31 -0800
Received: from ns.suse.de ([213.95.15.193]:55814 "HELO Cantor.suse.de")
	by oss.sgi.com with SMTP id <S553673AbRCERJC>;
	Mon, 5 Mar 2001 09:09:02 -0800
Received: from Hermes.suse.de (Hermes.suse.de [213.95.15.136])
	by Cantor.suse.de (Postfix) with ESMTP
	id 9D3361E297; Mon,  5 Mar 2001 18:08:50 +0100 (MET)
Date:   Mon, 5 Mar 2001 18:08:44 +0100
From:   Andi Kleen <ak@suse.de>
To:     Steve Lord <lord@sgi.com>
Cc:     linux-xfs@oss.sgi.com
Subject: Re: TAKE - implement direct read and write in XFS
Message-ID: <20010305180844.A19952@gruyere.muc.suse.de>
References: <200103051654.f25GsbI16800@jen.americas.sgi.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.2.5i
In-Reply-To: <200103051654.f25GsbI16800@jen.americas.sgi.com>; from lord@sgi.com on Mon, Mar 05, 2001 at 10:54:37AM -0600
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing

On Mon, Mar 05, 2001 at 10:54:37AM -0600, Steve Lord wrote:
> This implements direct I/O for xfs on linux, read and write should function.
> Testing has been limited so far. There is cache coherency between buffered
> and direct I/O to the extent that the direct calls will flush or flush and
> invalidate the cache before starting I/O. A systune has been added at
> /proc/sys/vm/pagebuf/max_dio_pages which defines how many pages will be
> locked down at once during an individual read or write call, this does not
> limit the size of the I/O, just the size of the requests which will be
> issued to the block layer.

Very cool :-)

-Andi

From owner-linux-xfs@oss.sgi.com Mon Mar  5 09:23:21 2001
Received:  by oss.sgi.com id <S553788AbRCERXM>;
	Mon, 5 Mar 2001 09:23:12 -0800
Received: from pneumatic-tube.sgi.com ([204.94.214.22]:10256 "EHLO
        pneumatic-tube.sgi.com") by oss.sgi.com with ESMTP
	id <S553673AbRCERWs>; Mon, 5 Mar 2001 09:22:48 -0800
Received: from zeus-fddi.americas.sgi.com (zeus-fddi.americas.sgi.com [128.162.8.103]) by pneumatic-tube.sgi.com (980327.SGI.8.8.8-aspam/980310.SGI-aspam) via ESMTP id JAA09086
	for <linux-xfs@oss.sgi.com>; Mon, 5 Mar 2001 09:32:29 -0800 (PST)
	mail_from (lord@sgi.com)
Received: from daisy-e185.americas.sgi.com (daisy-e194.americas.sgi.com [128.162.194.214]) by zeus-fddi.americas.sgi.com (8.9.3/americas-smart-nospam1.1) with ESMTP id LAA1019096; Mon, 5 Mar 2001 11:21:15 -0600 (CST)
Received: from jen.americas.sgi.com (IDENT:root@jen.americas.sgi.com [128.162.184.86]) by daisy-e185.americas.sgi.com (SGI-8.9.3/SGI-server-1.7) with ESMTP id LAA88616; Mon, 5 Mar 2001 11:21:15 -0600 (CST)
Received: from jen.americas.sgi.com by jen.americas.sgi.com (8.11.0/SGI-client-1.7) via ESMTP id f25HKWo20876; Mon, 5 Mar 2001 11:20:32 -0600
Message-Id: <200103051720.f25HKWo20876@jen.americas.sgi.com>
X-Mailer: exmh version 2.2 06/23/2000 with nmh-1.0.4
To:     Vladimir Vukicevic <vladimir@ximian.com>
cc:     linux-xfs@oss.sgi.com
Subject: Re: questions about some benchmarks 
In-Reply-To: Message from Vladimir Vukicevic <vladimir@ximian.com> 
   of "Sun, 04 Mar 2001 22:28:09 EST." <E14ZlfN-00051A-00@rain.ximian.com> 
Date:   Mon, 05 Mar 2001 11:20:32 -0600
From:   Steve Lord <lord@sgi.com>
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing


Removing files is not XFS's strong point, there is a known way to implement
a fix for this, but it involves major surgery, and it has never been attempted.
Basically you are looking at the most disk bound operation in XFS, the
remove operation can contain synchronous disk I/O - this is why your
cpu is not 100% busy during it.

I will dig out the description of what needs doing (which involves
intimate knowledge of xfs workings), in the meantime, you can make
things better with two configuration changes.

1. use a larger than default log size

	mkfs -t xfs -f -l size=32768b /dev/xxx

   this cannot be applied retro actively to existing filesystems.

2. use the following mount option

	mount -o logbufs=4

or

	mount -o logbufs=8

The should both help the situation, but will not totally fix it.

The reason ext2 is so much slower is that you keep modifying the same
buffers in memory without writing them to disk. With XFS you keep writing
new transactions into the log - and in the process of reconsuming log space
have to flush metadata that goes with it. The mkfs option makes the log
bigger, the mount option gives you more in memory buffers so you can have
more log writes going in parallel.

I have also found that multiple parallel removes of different trees will
go faster than a single one of the whole tree. 

Steve

> 
> Howdy.. I just converted my laptop's root partition to XFS; while
> doing some hacking I ended up doing a quick-and-dirty benchmark
> between XFS and ext2.
> 
> The freedb database (www.freedb.org) is a 59MB .tar.bz2 that expands
> out into 1.1GB worth of little files, all 1-2K in size, mostly towards
> the 1K size. (Why they use the filesystem as the database and not use
> libdb or something is beyond me. Anyway.) So, I timed the untar and
> rm -rf times of this tree.
> 
> The two machines involved are an IBM X20 laptop, P3-600, 392MB ram,
> 20GB IBM Travelstar 4200 RPM drive, UDMA mode3, xfs filesystem. On my
> desktop, which is an Athlon 700, VIA VT82Cxxx Apollo chipset, 512MB
> ram, tar file on an UATA/100 drive on the primary on-board interface,
> destination drive on a promise UATA/100 IDE controller with an IBM
> Deskstar 7200 RPM 45GB drive on it, with an ext2 filesystem:
> 
> On my laptop (xfs),
> 
> tar xjf ../freedb-complete-20010219.tar.bz2  138.88s user
>   207.17s system 27% cpu 21:17.84 total 
> rm -rf db  1.45s user 130.99s system 9% cpu 24:24.66 total
> 
> On my desktop (ext2),
> 
> tar xjf ~vladimir/freedb-complete-20010219.tar.bz2  153.76s user
>   2772.26s system 98% cpu 49:28.47 total
> rm -rf db  0.16s user 73.39s system 94% cpu 1:18.24 total
> 
> Now, I converted the filesystem on my desktop to XFS.  Here's the same
> numbers again, this time for xfs.
> 
> tar xjf ~vladimir/freedb-complete-20010219.tar.bz2  156.77s user
>   199.28s system 61% cpu 9:41.88 total 
> rm -rf db  1.73s user 131.59s system 19% cpu 11:07.25 total
> 
> 
> So:
> 
> System                      tar time  cpu      rm time    cpu
> P3-600/4200rpm, xfs         21:17.84  27%      24:24.66    9%
> K7-700/7200rpm, xfs          9:41.88  61%      11:07.25   19%
> K7-700/7200rpm, ext2        49:28.47  98%       1:18.24   94%
> 
> So, while xfs obviously kicks ext2's butt, I'm curious as to the cpu
> usage numbers of both the untar and the remove, and also of xfs's
> significantly slower remove times (even though its cpu usage was
> also significantly lower).. is there anything about xfs that
> causes removes to be significantly slower?  (I just realized that
> I didn't time a 'sync' after the relevant removes though, so
> a good chunk of this could have even been buffered, maybe...)
> 
> I realize that this was an ad-hoc benchmark, but the numbers seemed
> interesting nonetheless. (If anyone's interested in testing the same,
> the database is available at ftp://ftp.freedb.org/pub/freedb/ ).
> 
>     - Vlad



From owner-linux-xfs@oss.sgi.com Mon Mar  5 09:25:01 2001
Received:  by oss.sgi.com id <S553788AbRCERYw>;
	Mon, 5 Mar 2001 09:24:52 -0800
Received: from pneumatic-tube.sgi.com ([204.94.214.22]:18448 "EHLO
        pneumatic-tube.sgi.com") by oss.sgi.com with ESMTP
	id <S553673AbRCERYl>; Mon, 5 Mar 2001 09:24:41 -0800
Received: from zeus-fddi.americas.sgi.com (zeus-fddi.americas.sgi.com [128.162.8.103]) by pneumatic-tube.sgi.com (980327.SGI.8.8.8-aspam/980310.SGI-aspam) via ESMTP id JAA09882
	for <linux-xfs@oss.sgi.com>; Mon, 5 Mar 2001 09:34:22 -0800 (PST)
	mail_from (lord@sgi.com)
Received: from daisy-e185.americas.sgi.com (daisy-e194.americas.sgi.com [128.162.194.214]) by zeus-fddi.americas.sgi.com (8.9.3/americas-smart-nospam1.1) with ESMTP id LAA1023857; Mon, 5 Mar 2001 11:23:23 -0600 (CST)
Received: from jen.americas.sgi.com (IDENT:root@jen.americas.sgi.com [128.162.184.86]) by daisy-e185.americas.sgi.com (SGI-8.9.3/SGI-server-1.7) with ESMTP id LAA02122; Mon, 5 Mar 2001 11:23:23 -0600 (CST)
Received: from jen.americas.sgi.com by jen.americas.sgi.com (8.11.0/SGI-client-1.7) via ESMTP id f25HMe420887; Mon, 5 Mar 2001 11:22:40 -0600
Message-Id: <200103051722.f25HMe420887@jen.americas.sgi.com>
X-Mailer: exmh version 2.2 06/23/2000 with nmh-1.0.4
To:     Christoph Hellwig <hch@caldera.de>
cc:     linux-xfs@oss.sgi.com
Subject: Re: [PATCH] kill lookup_vfsmnt() 
In-Reply-To: Message from Christoph Hellwig <hch@caldera.de> 
   of "Sun, 04 Mar 2001 12:32:02 +0100." <20010304123202.A14717@caldera.de> 
Date:   Mon, 05 Mar 2001 11:22:40 -0600
From:   Steve Lord <lord@sgi.com>
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing


Thanks,

Did you try this?

I think it has a locking problem when mounting the root filesystem, it
hangs very hard during root mount, probably too many lock kernels in
some path.

I like the idea of getting rid of the call, but it looks like it needs
a little work yet (I will dig into it).

Steve


> Hi,
> 
> the lookup_vfsmnt function in the xfs tree is broken by design
> as it tries to get a struct vfsmnt for a given kdev_t but a
> filesystem can be mounted at multiple mount points in Linux 2.4
> and so the mapping is not unique.
> 
> The actual usage of the function is not broken as it is used only
> to get the struct super_block from the vfsmnt which is again
> unique for a kdev_t (but doesn't have to - I know of at least one
> fs design where there might be multiple sb's for a dev).
> 
> Suggestion: kill lookup_vfsmnt and use get_super instead.
> Patch attached.
> 
> 	Christoph
> 
> -- 
> Of course it doesn't work. We've performed a software upgrade.
> 
> 
> Index: fs/super.c
> ===================================================================
> RCS file: /cvs/linux-2.4-xfs/linux/fs/super.c,v
> retrieving revision 1.42
> diff -u -b -B -r1.42 super.c
> --- fs/super.c	2001/02/22 21:09:04	1.42
> +++ fs/super.c	2001/03/03 18:29:44
> @@ -456,24 +456,6 @@
>          return sp - buf;	/* no trailing NUL */
>  }
>  
> -struct vfsmount *lookup_vfsmnt(kdev_t dev)
> -{
> -	struct list_head *p;
> -
> -	if (list_empty(&vfsmntlist))
> -		return NULL;
> -
> -	for (p = vfsmntlist.next; p != &vfsmntlist; p = p->next) {
> -		struct vfsmount *tmp = list_entry(p, struct vfsmount, mnt_list)
> ;
> -		if (tmp->mnt_sb->s_dev == dev) {
> -			return (tmp);
> -		}
> -	}
> -
> -	return NULL;
> -}
> -
> -
>  static struct proc_fs_info {
>  	int flag;
>  	char *str;
> Index: fs/xfs/linux/xfs_vfs.c
> ===================================================================
> RCS file: /cvs/linux-2.4-xfs/linux/fs/xfs/linux/xfs_vfs.c,v
> retrieving revision 1.21
> diff -u -b -B -r1.21 xfs_vfs.c
> --- fs/xfs/linux/xfs_vfs.c	2000/11/22 06:53:25	1.21
> +++ fs/xfs/linux/xfs_vfs.c	2001/03/03 18:30:20
> @@ -211,14 +211,16 @@
>  	long s;
>  	struct vfs *vfsp;
>  	kdev_t	kdev = MKDEV(MAJOR(dev), MINOR(dev));
> -	struct vfsmount *entry;
> +	struct super_block *sb;
>  
>  	lock_kernel();
> -	entry = lookup_vfsmnt(kdev);
> -
> +	sb = get_super(kdev);
>  	unlock_kernel();
> -	if (entry) {
> -		vfsp = LINVFS_GET_VFS(entry->mnt_sb);
> +
> +	if (!sb)
> +		return NULL;
> +	
> +	vfsp = LINVFS_GET_VFS(sb);
>  again:
>  		spin_lock_irqsave(&vfslock, s);
>  		if (vfsp->vfs_dev == dev &&
> @@ -241,9 +243,6 @@
>  		}
>  		spin_unlock_irqrestore(&vfslock, s);
>  		return vfsp;
> -	} else {
> -		return NULL;
> -	}
>  }
>  
>  void
> @@ -312,14 +311,13 @@
>  {
>          register struct vfs *vfsp;
>  	kdev_t	kdev = MKDEV(MAJOR(dev), MINOR(dev));
> -	struct vfsmount *entry;
> +	struct super_block *sb;
>  
>  	lock_kernel();
> -	entry = lookup_vfsmnt(kdev);
> -
> +	sb = get_super(kdev);
>  
> -        if (entry) {
> -		vfsp = LINVFS_GET_VFS(entry->mnt_sb);
> +	if (sb) {
> +		vfsp = LINVFS_GET_VFS(sb);
>                  if ((vfsp->vfs_dev == dev) &&
>                      (fstype == VFS_FSTYPE_ANY || fstype == vfsp->vfs_fstype)
> ) {
>  			unlock_kernel();
> Index: include/linux/mount.h
> ===================================================================
> RCS file: /cvs/linux-2.4-xfs/linux/include/linux/mount.h,v
> retrieving revision 1.12
> diff -u -b -B -r1.12 mount.h
> --- include/linux/mount.h	2000/10/23 18:56:35	1.12
> +++ include/linux/mount.h	2001/03/03 18:30:35
> @@ -38,7 +38,6 @@
>  		atomic_inc(&mnt->mnt_count);
>  	return mnt;
>  }
> -struct vfsmount *lookup_vfsmnt(kdev_t dev);
>  
>  static inline void mntput(struct vfsmount *mnt)
>  {
> Index: kernel/ksyms.c
> ===================================================================
> RCS file: /cvs/linux-2.4-xfs/linux/kernel/ksyms.c,v
> retrieving revision 1.80
> diff -u -b -B -r1.80 ksyms.c
> --- kernel/ksyms.c	2001/03/01 21:13:39	1.80
> +++ kernel/ksyms.c	2001/03/03 18:30:40
> @@ -143,7 +143,6 @@
>  EXPORT_SYMBOL(get_fs_type);
>  EXPORT_SYMBOL(get_super);
>  EXPORT_SYMBOL(get_empty_super);
> -EXPORT_SYMBOL(lookup_vfsmnt);
>  EXPORT_SYMBOL(getname);
>  EXPORT_SYMBOL(names_cachep);
>  EXPORT_SYMBOL(fput);



From owner-linux-xfs@oss.sgi.com Mon Mar  5 09:34:51 2001
Received:  by oss.sgi.com id <S553792AbRCEReb>;
	Mon, 5 Mar 2001 09:34:31 -0800
Received: from ns.caldera.de ([212.34.180.1]:9741 "EHLO ns.caldera.de")
	by oss.sgi.com with ESMTP id <S553785AbRCERe2>;
	Mon, 5 Mar 2001 09:34:28 -0800
Received: (from hch@localhost)
	by ns.caldera.de (8.9.3/8.9.3) id SAA29691;
	Mon, 5 Mar 2001 18:34:01 +0100
Date:   Mon, 5 Mar 2001 18:34:01 +0100
From:   Christoph Hellwig <hch@caldera.de>
To:     Steve Lord <lord@sgi.com>
Cc:     Christoph Hellwig <hch@caldera.de>, linux-xfs@oss.sgi.com
Subject: Re: [PATCH] kill lookup_vfsmnt()
Message-ID: <20010305183401.A28908@caldera.de>
References: <hch@caldera.de> <200103051722.f25HMe420887@jen.americas.sgi.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
X-Mailer: Mutt 1.0i
In-Reply-To: <200103051722.f25HMe420887@jen.americas.sgi.com>; from lord@sgi.com on Mon, Mar 05, 2001 at 11:22:40AM -0600
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing

On Mon, Mar 05, 2001 at 11:22:40AM -0600, Steve Lord wrote:
> 
> Thanks,
> 
> Did you try this?

Yes - a simple mount/umount succeded.  I did no further testing as
I'm not using xfs on my test machine currently.

> I think it has a locking problem when mounting the root filesystem, it
> hangs very hard during root mount, probably too many lock kernels in
> some path.
> 
> I like the idea of getting rid of the call, but it looks like it needs
> a little work yet (I will dig into it).

Okay.

	Christoph

-- 
Of course it doesn't work. We've performed a software upgrade.

From owner-linux-xfs@oss.sgi.com Mon Mar  5 10:51:51 2001
Received:  by oss.sgi.com id <S553796AbRCESvl>;
	Mon, 5 Mar 2001 10:51:41 -0800
Received: from h217n3fls20o974.telia.com ([212.181.166.217]:2827 "HELO
        neoblade.net") by oss.sgi.com with SMTP id <S553791AbRCESva>;
	Mon, 5 Mar 2001 10:51:30 -0800
Received: from student.liu.se (neoblade.net [192.168.1.10])
	by neoblade.net (Postfix) with ESMTP id 16577189131
	for <linux-xfs@oss.sgi.com>; Mon,  5 Mar 2001 19:53:16 +0100 (CET)
Message-ID: <3AA3E08F.9090208@student.liu.se>
Date:   Mon, 05 Mar 2001 19:53:03 +0100
From:   Joakim Bodin <joabo552@student.liu.se>
User-Agent: Mozilla/5.0 (X11; U; Linux 2.4.2-XFS i686; en-US; 0.9) Gecko/20010305
X-Accept-Language: sv, en
MIME-Version: 1.0
To:     Linux-XFS Mailing List <linux-xfs@oss.sgi.com>
Subject: fsr.xfs locking up computer
Content-Type: text/plain; charset=us-ascii; format=flowed
Content-Transfer-Encoding: 7bit
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing

I've been using fsr.xfs for many months now and it's been working fine. 
The exception to that was when I tried to run it yesterday (2001 march 
4) and it hardlocked my system. This "feature" seem to be a bit 
unsupported from reading the documentation but I still felt that I 
should report it as it has worked before. I didn't find anything in the 
logs after this though.

Joakim Bodin


From owner-linux-xfs@oss.sgi.com Mon Mar  5 10:52:42 2001
Received:  by oss.sgi.com id <S553795AbRCESwW>;
	Mon, 5 Mar 2001 10:52:22 -0800
Received: from pipt.oz.cc.utah.edu ([155.99.2.7]:39361 "EHLO
        pipt.oz.cc.utah.edu") by oss.sgi.com with ESMTP id <S553791AbRCESwP>;
	Mon, 5 Mar 2001 10:52:15 -0800
Received: from localhost (jdr1529@localhost)
	by pipt.oz.cc.utah.edu (8.9.2/8.9.2) with ESMTP id LAA29466;
	Mon, 5 Mar 2001 11:51:51 -0700 (MST)
Date:   Mon, 5 Mar 2001 11:51:51 -0700 (MST)
From:   james rich <james.rich@m.cc.utah.edu>
To:     Joshua Baker-LePain <jlb17@duke.edu>
cc:     Linux xfs mailing list <linux-xfs@oss.sgi.com>
Subject: Re: Getting started...
In-Reply-To: <Pine.LNX.4.30.0103051143120.7060-100000@chaos.egr.duke.edu>
Message-ID: <Pine.GSO.4.05.10103051148090.28604-100000@pipt.oz.cc.utah.edu>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing

On Mon, 5 Mar 2001, Joshua Baker-LePain wrote:

> Which, finally, brings me to my question.  What is the current
> conventional wisdom on the best way to get xfs up and running?  I'll
> obviously be testing before I put this into production.  Should I be
> grabbing the latest CVS snapshots, or using the 0.9 pre-release?  I'll

Probably you should begin with the pre-release.  Do your testing, and then
try current CVS.  Do some more testing.  My experience has been very good
with CVS (once I got it compiled - thanks guys!) but I don't use it in a
big, high demand, mission critical environment (unless you call my
slackware mirror and mp3 collection mission critical :) ).

James Rich
james.rich@m.cc.utah.edu


From owner-linux-xfs@oss.sgi.com Mon Mar  5 10:56:22 2001
Received:  by oss.sgi.com id <S553796AbRCES4M>;
	Mon, 5 Mar 2001 10:56:12 -0800
Received: from sgi.SGI.COM ([192.48.153.1]:42819 "EHLO sgi.com")
	by oss.sgi.com with ESMTP id <S553795AbRCESzv>;
	Mon, 5 Mar 2001 10:55:51 -0800
Received: from zeus-fddi.americas.sgi.com (128-162-8-103.americas.sgi.com [128.162.8.103]) 
	by sgi.com (980327.SGI.8.8.8-aspam/980304.SGI-aspam:
       SGI does not authorize the use of its proprietary
       systems or networks for unsolicited or bulk email
       from the Internet.) 
	via ESMTP id KAA03891
	for <linux-xfs@oss.sgi.com>; Mon, 5 Mar 2001 10:55:47 -0800 (PST)
	mail_from (lord@sgi.com)
Received: from daisy-e185.americas.sgi.com (daisy-e194.americas.sgi.com [128.162.194.214]) by zeus-fddi.americas.sgi.com (8.9.3/americas-smart-nospam1.1) with ESMTP id MAA1013335; Mon, 5 Mar 2001 12:54:30 -0600 (CST)
Received: from jen.americas.sgi.com (IDENT:root@jen.americas.sgi.com [128.162.184.86]) by daisy-e185.americas.sgi.com (SGI-8.9.3/SGI-server-1.7) with ESMTP id MAA76414; Mon, 5 Mar 2001 12:54:30 -0600 (CST)
From:   Steve Lord <lord@sgi.com>
Received: by jen.americas.sgi.com (8.11.0/SGI-client-1.7) id f25Irlx26350; Mon, 5 Mar 2001 12:53:47 -0600
Message-Id: <200103051853.f25Irlx26350@jen.americas.sgi.com>
Date:   Mon, 5 Mar 2001 12:53:47 -0600
Subject: TAKE - patch from Christoph Hellwig to remove lookup_vfsmnt()
To:     unlisted-recipients:; (no To-header on input)
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing

This removes another xfs intrusion into the mainline code. Patch is
from Christoph Hellwig with a fix from me for doing root mounts.


Date:  Mon Mar  5 10:53:08 PST 2001
Workarea:  jen.americas.sgi.com:/src/lord/xfs-linux.2.4

The following file(s) were checked into:
  bonnie.engr.sgi.com:/isms/slinx/2.4.x-xfs


Modid:  2.4.x-xfs:slinx:88899a
linux/kernel/ksyms.c - 1.81
	- remove lookup_vfsmnt export

linux/include/linux/mount.h - 1.13
	- remove lookup_vfsmnt() prototype

linux/fs/super.c - 1.43
	- remove lookup_vfsmnt()

linux/fs/xfs/xfs_vfsops.c - 1.308
	- No need to verify a device is not already mounted during the mount code,
	  linux has already done this.

linux/fs/xfs/linux/xfs_vfs.c - 1.22
	- use get_super() instead of lookup_vfsmnt()



From owner-linux-xfs@oss.sgi.com Mon Mar  5 11:07:41 2001
Received:  by oss.sgi.com id <S553796AbRCETHW>;
	Mon, 5 Mar 2001 11:07:22 -0800
Received: from deliverator.sgi.com ([204.94.214.10]:63855 "EHLO
        deliverator.sgi.com") by oss.sgi.com with ESMTP id <S553791AbRCETHK>;
	Mon, 5 Mar 2001 11:07:10 -0800
Received: from zeus-fddi.americas.sgi.com (zeus-fddi.americas.sgi.com [128.162.8.103]) by deliverator.sgi.com (980309.SGI.8.8.8-aspam-6.2/980310.SGI-aspam) via ESMTP id LAA18079
	for <linux-xfs@oss.sgi.com>; Mon, 5 Mar 2001 11:06:03 -0800 (PST)
	mail_from (lord@sgi.com)
Received: from daisy-e185.americas.sgi.com (daisy-e194.americas.sgi.com [128.162.194.214]) by zeus-fddi.americas.sgi.com (8.9.3/americas-smart-nospam1.1) with ESMTP id NAA1023009 for <linux-xfs@oss.sgi.com>; Mon, 5 Mar 2001 13:05:53 -0600 (CST)
Received: from jen.americas.sgi.com (IDENT:root@jen.americas.sgi.com [128.162.184.86]) by daisy-e185.americas.sgi.com (SGI-8.9.3/SGI-server-1.7) with ESMTP id NAA38514 for <linux-xfs@oss.sgi.com>; Mon, 5 Mar 2001 13:05:53 -0600 (CST)
From:   Steve Lord <lord@sgi.com>
Received: by jen.americas.sgi.com (8.11.0/SGI-client-1.7) id f25J5Aj26430; Mon, 5 Mar 2001 13:05:10 -0600
Message-Id: <200103051905.f25J5Aj26430@jen.americas.sgi.com>
Date:   Mon, 5 Mar 2001 13:05:10 -0600
Subject: TAKE - small fix for ram disk from Jens Axboe
To:     unlisted-recipients:; (no To-header on input)
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing

Date:  Mon Mar  5 11:05:13 PST 2001
Workarea:  jen.americas.sgi.com:/src/lord/xfs-linux.2.4

The following file(s) were checked into:
  bonnie.engr.sgi.com:/isms/slinx/2.4.x-xfs


Modid:  2.4.x-xfs:slinx:88900a
linux/drivers/block/rd.c - 1.27
	- Fix function args for kiobuf enabled kernel



From owner-linux-xfs@oss.sgi.com Mon Mar  5 12:10:52 2001
Received:  by oss.sgi.com id <S553797AbRCEUKn>;
	Mon, 5 Mar 2001 12:10:43 -0800
Received: from sgi.SGI.COM ([192.48.153.1]:34917 "EHLO sgi.com")
	by oss.sgi.com with ESMTP id <S553794AbRCEUK2>;
	Mon, 5 Mar 2001 12:10:28 -0800
Received: from zeus-fddi.americas.sgi.com (128-162-8-103.americas.sgi.com [128.162.8.103]) 
	by sgi.com (980327.SGI.8.8.8-aspam/980304.SGI-aspam:
       SGI does not authorize the use of its proprietary
       systems or networks for unsolicited or bulk email
       from the Internet.) 
	via ESMTP id MAA05110
	for <linux-xfs@oss.sgi.com>; Mon, 5 Mar 2001 12:10:01 -0800 (PST)
	mail_from (lord@sgi.com)
Received: from daisy-e185.americas.sgi.com (daisy-e194.americas.sgi.com [128.162.194.214]) by zeus-fddi.americas.sgi.com (8.9.3/americas-smart-nospam1.1) with ESMTP id OAA1028944; Mon, 5 Mar 2001 14:08:41 -0600 (CST)
Received: from jen.americas.sgi.com (IDENT:root@jen.americas.sgi.com [128.162.184.86]) by daisy-e185.americas.sgi.com (SGI-8.9.3/SGI-server-1.7) with ESMTP id OAA31251; Mon, 5 Mar 2001 14:08:41 -0600 (CST)
Received: from jen.americas.sgi.com by jen.americas.sgi.com (8.11.0/SGI-client-1.7) via ESMTP id f25K7wl03508; Mon, 5 Mar 2001 14:07:58 -0600
Message-Id: <200103052007.f25K7wl03508@jen.americas.sgi.com>
X-Mailer: exmh version 2.2 06/23/2000 with nmh-1.0.4
To:     Joakim Bodin <joabo552@student.liu.se>
cc:     Linux-XFS Mailing List <linux-xfs@oss.sgi.com>
Subject: Re: fsr.xfs locking up computer 
In-Reply-To: Message from Joakim Bodin <joabo552@student.liu.se> 
   of "Mon, 05 Mar 2001 19:53:03 +0100." <3AA3E08F.9090208@student.liu.se> 
Date:   Mon, 05 Mar 2001 14:07:57 -0600
From:   Steve Lord <lord@sgi.com>
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing

> I've been using fsr.xfs for many months now and it's been working fine. 
> The exception to that was when I tried to run it yesterday (2001 march 
> 4) and it hardlocked my system. This "feature" seem to be a bit 
> unsupported from reading the documentation but I still felt that I 
> should report it as it has worked before. I didn't find anything in the 
> logs after this though.
> 
> Joakim Bodin


How exacty are you running fsr? Have you updated the fsr binary recently,
or are you running an old one (I doubt this matters). I tried a few basic
tests and fsr appeared to be functioning.

Steve





From owner-linux-xfs@oss.sgi.com Mon Mar  5 12:16:52 2001
Received:  by oss.sgi.com id <S553796AbRCEUQm>;
	Mon, 5 Mar 2001 12:16:42 -0800
Received: from sgi.SGI.COM ([192.48.153.1]:57447 "EHLO sgi.com")
	by oss.sgi.com with ESMTP id <S553791AbRCEUQ1>;
	Mon, 5 Mar 2001 12:16:27 -0800
Received: from waco.engr.sgi.com ([163.154.18.95]) 
	by sgi.com (980327.SGI.8.8.8-aspam/980304.SGI-aspam:
       SGI does not authorize the use of its proprietary
       systems or networks for unsolicited or bulk email
       from the Internet.) 
	via ESMTP id MAA01949
	for <linux-xfs@oss.sgi.com>; Mon, 5 Mar 2001 12:16:15 -0800 (PST)
	mail_from (ananth@waco.engr.sgi.com)
Received: (from ananth@localhost)
	by waco.engr.sgi.com (8.11.0/8.11.0) id f25KHcu18470
	for linux-xfs@oss.sgi.com; Mon, 5 Mar 2001 12:17:38 -0800
Date:   Mon, 5 Mar 2001 12:17:38 -0800
From:   Ananth Ananthanarayanan <ananth@waco.engr.sgi.com>
Message-Id: <200103052017.f25KHcu18470@waco.engr.sgi.com>
Subject: TAKE - properly account clustered IO
To:     unlisted-recipients:; (no To-header on input)
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing

This change improves dbench performance for smaller number of clients by 5-10%.

Date:  Mon Mar  5 12:12:26 PST 2001
Workarea:  waco.engr.sgi.com:/build1/ananth/xfs-tot

The following file(s) were checked into:
  bonnie.engr.sgi.com:/isms/slinx/2.4.x-xfs


Modid:  2.4.x-xfs:slinx:88916a
linux/fs/buffer.c - 1.58
	- Account for the fact that write_buffer for clustered IO can push out
	  more than 1 block at a time. This makes back ground scanning only as
	  aggressive as necessary in bdflush.

linux/fs/pagebuf/page_buf_io.c - 1.63
	- Return number of pages clustered for use by write_buffer so
	  watermark accounting is done properly. Other misc. cleanups.



From owner-linux-xfs@oss.sgi.com Mon Mar  5 12:18:32 2001
Received:  by oss.sgi.com id <S553795AbRCEUSW>;
	Mon, 5 Mar 2001 12:18:22 -0800
Received: from deliverator.sgi.com ([204.94.214.10]:37898 "EHLO
        deliverator.sgi.com") by oss.sgi.com with ESMTP id <S553791AbRCEUSQ>;
	Mon, 5 Mar 2001 12:18:16 -0800
Received: from zeus-fddi.americas.sgi.com (zeus-fddi.americas.sgi.com [128.162.8.103]) by deliverator.sgi.com (980309.SGI.8.8.8-aspam-6.2/980310.SGI-aspam) via ESMTP id MAA29380
	for <linux-xfs@oss.sgi.com>; Mon, 5 Mar 2001 12:17:09 -0800 (PST)
	mail_from (lord@sgi.com)
Received: from daisy-e185.americas.sgi.com (daisy-e194.americas.sgi.com [128.162.194.214]) by zeus-fddi.americas.sgi.com (8.9.3/americas-smart-nospam1.1) with ESMTP id OAA1029601; Mon, 5 Mar 2001 14:16:58 -0600 (CST)
Received: from jen.americas.sgi.com (IDENT:root@jen.americas.sgi.com [128.162.184.86]) by daisy-e185.americas.sgi.com (SGI-8.9.3/SGI-server-1.7) with ESMTP id OAA37123; Mon, 5 Mar 2001 14:16:57 -0600 (CST)
Received: from jen.americas.sgi.com by jen.americas.sgi.com (8.11.0/SGI-client-1.7) via ESMTP id f25KGDK04905; Mon, 5 Mar 2001 14:16:13 -0600
Message-Id: <200103052016.f25KGDK04905@jen.americas.sgi.com>
X-Mailer: exmh version 2.2 06/23/2000 with nmh-1.0.4
To:     james rich <james.rich@m.cc.utah.edu>
cc:     Joshua Baker-LePain <jlb17@duke.edu>,
        Linux xfs mailing list <linux-xfs@oss.sgi.com>
Subject: Re: Getting started... 
In-Reply-To: Message from james rich <james.rich@m.cc.utah.edu> 
   of "Mon, 05 Mar 2001 11:51:51 MST." <Pine.GSO.4.05.10103051148090.28604-100000@pipt.oz.cc.utah.edu> 
Date:   Mon, 05 Mar 2001 14:16:13 -0600
From:   Steve Lord <lord@sgi.com>
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing

> On Mon, 5 Mar 2001, Joshua Baker-LePain wrote:
> 
> > Which, finally, brings me to my question.  What is the current
> > conventional wisdom on the best way to get xfs up and running?  I'll
> > obviously be testing before I put this into production.  Should I be
> > grabbing the latest CVS snapshots, or using the 0.9 pre-release?  I'll
> 
> Probably you should begin with the pre-release.  Do your testing, and then
> try current CVS.  Do some more testing.  My experience has been very good
> with CVS (once I got it compiled - thanks guys!) but I don't use it in a
> big, high demand, mission critical environment (unless you call my
> slackware mirror and mp3 collection mission critical :) ).
> 
> James Rich
> james.rich@m.cc.utah.edu


I would say that the beta images (with some provisos about mouse configuration
due to devfs) is the easiest way to get a system up from scratch. I do NFS
based installs in about 20 minutes - grab all of redhats updated iso images 
as well though. The first thing I do after this is build by own kernel with
the drivers I use built into the kernel and devfs turned off, this gets rid
of the need for an initial ram disk and usually speeds the boot process.

The CVS tree should be more stable for you, and offers more features
[2.4.2 base, better VM integration of delayed write, O_SYNC support,
O_DIRECT support, faster shutdown], and probably better performance than
the beta images do. Upgrading the kernel with the cvs tree version after
installing the beta should be a good thing to do. In general we are trying
not to break the cvs tree now, but it is always good to keep one kernel
you know works (the beta images for example) in your lilo configuration.

Steve




From owner-linux-xfs@oss.sgi.com Mon Mar  5 12:20:52 2001
Received:  by oss.sgi.com id <S553795AbRCEUUm>;
	Mon, 5 Mar 2001 12:20:42 -0800
Received: from deliverator.sgi.com ([204.94.214.10]:22283 "EHLO
        deliverator.sgi.com") by oss.sgi.com with ESMTP id <S553791AbRCEUUd>;
	Mon, 5 Mar 2001 12:20:33 -0800
Received: from zeus-fddi.americas.sgi.com (zeus-fddi.americas.sgi.com [128.162.8.103]) by deliverator.sgi.com (980309.SGI.8.8.8-aspam-6.2/980310.SGI-aspam) via ESMTP id MAA29723
	for <linux-xfs@oss.sgi.com>; Mon, 5 Mar 2001 12:19:26 -0800 (PST)
	mail_from (lord@sgi.com)
Received: from daisy-e185.americas.sgi.com (daisy-e194.americas.sgi.com [128.162.194.214]) by zeus-fddi.americas.sgi.com (8.9.3/americas-smart-nospam1.1) with ESMTP id OAA1026735 for <linux-xfs@oss.sgi.com>; Mon, 5 Mar 2001 14:19:16 -0600 (CST)
Received: from jen.americas.sgi.com (IDENT:root@jen.americas.sgi.com [128.162.184.86]) by daisy-e185.americas.sgi.com (SGI-8.9.3/SGI-server-1.7) with ESMTP id OAA31330 for <linux-xfs@oss.sgi.com>; Mon, 5 Mar 2001 14:19:16 -0600 (CST)
From:   Steve Lord <lord@sgi.com>
Received: by jen.americas.sgi.com (8.11.0/SGI-client-1.7) id f25KIXp05000; Mon, 5 Mar 2001 14:18:33 -0600
Message-Id: <200103052018.f25KIXp05000@jen.americas.sgi.com>
Date:   Mon, 5 Mar 2001 14:18:33 -0600
Subject: TAKE - turn on direct I/O in fsstress test
To:     unlisted-recipients:; (no To-header on input)
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing

Date:  Mon Mar  5 12:18:45 PST 2001
Workarea:  jen.americas.sgi.com:/src/lord/xfs-linux.2.4

The following file(s) were checked into:
  bonnie.engr.sgi.com:/isms/slinx/2.4.x-xfs


Modid:  2.4.x-xfs:slinx:88918a
cmd/xfstests/src/fsstress.c - 1.3
	- Turn back on direct I/O in the testing mix



From owner-linux-xfs@oss.sgi.com Mon Mar  5 12:22:52 2001
Received:  by oss.sgi.com id <S553795AbRCEUWm>;
	Mon, 5 Mar 2001 12:22:42 -0800
Received: from deliverator.sgi.com ([204.94.214.10]:53003 "EHLO
        deliverator.sgi.com") by oss.sgi.com with ESMTP id <S553791AbRCEUWb>;
	Mon, 5 Mar 2001 12:22:31 -0800
Received: from zeus-fddi.americas.sgi.com (zeus-fddi.americas.sgi.com [128.162.8.103]) by deliverator.sgi.com (980309.SGI.8.8.8-aspam-6.2/980310.SGI-aspam) via ESMTP id MAA29946
	for <linux-xfs@oss.sgi.com>; Mon, 5 Mar 2001 12:21:24 -0800 (PST)
	mail_from (lord@sgi.com)
Received: from daisy-e185.americas.sgi.com (daisy-e194.americas.sgi.com [128.162.194.214]) by zeus-fddi.americas.sgi.com (8.9.3/americas-smart-nospam1.1) with ESMTP id OAA1027505 for <linux-xfs@oss.sgi.com>; Mon, 5 Mar 2001 14:21:14 -0600 (CST)
Received: from jen.americas.sgi.com (IDENT:root@jen.americas.sgi.com [128.162.184.86]) by daisy-e185.americas.sgi.com (SGI-8.9.3/SGI-server-1.7) with ESMTP id OAA83485 for <linux-xfs@oss.sgi.com>; Mon, 5 Mar 2001 14:21:14 -0600 (CST)
From:   Steve Lord <lord@sgi.com>
Received: by jen.americas.sgi.com (8.11.0/SGI-client-1.7) id f25KKUe05065; Mon, 5 Mar 2001 14:20:30 -0600
Message-Id: <200103052020.f25KKUe05065@jen.americas.sgi.com>
Date:   Mon, 5 Mar 2001 14:20:30 -0600
Subject: TAKE - replace some deleted tests for dirty files
To:     unlisted-recipients:; (no To-header on input)
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing

Prior to the delayed buffer implementation it was pretty much impossible
to tell if an inode contained dirty data cheaply. Now we have this ability,
restore a couple of deleted uses of this test.

Date:  Mon Mar  5 12:19:59 PST 2001
Workarea:  jen.americas.sgi.com:/src/lord/xfs-linux.2.4

The following file(s) were checked into:
  bonnie.engr.sgi.com:/isms/slinx/2.4.x-xfs


Modid:  2.4.x-xfs:slinx:88919a
linux/fs/xfs/xfs_vnodeops.c - 1.489
linux/fs/xfs/xfs_vfsops.c - 1.309
	- Add back VN_DIRTY since it is easy to work this out now



From owner-linux-xfs@oss.sgi.com Mon Mar  5 13:12:02 2001
Received:  by oss.sgi.com id <S553795AbRCEVLm>;
	Mon, 5 Mar 2001 13:11:42 -0800
Received: from sgi.SGI.COM ([192.48.153.1]:6148 "EHLO sgi.com")
	by oss.sgi.com with ESMTP id <S553791AbRCEVLS>;
	Mon, 5 Mar 2001 13:11:18 -0800
Received: from zeus-fddi.americas.sgi.com (128-162-8-103.americas.sgi.com [128.162.8.103]) 
	by sgi.com (980327.SGI.8.8.8-aspam/980304.SGI-aspam:
       SGI does not authorize the use of its proprietary
       systems or networks for unsolicited or bulk email
       from the Internet.) 
	via ESMTP id NAA08604
	for <linux-xfs@oss.sgi.com>; Mon, 5 Mar 2001 13:11:04 -0800 (PST)
	mail_from (lord@sgi.com)
Received: from daisy-e185.americas.sgi.com (daisy-e194.americas.sgi.com [128.162.194.214]) by zeus-fddi.americas.sgi.com (8.9.3/americas-smart-nospam1.1) with ESMTP id PAA1026014 for <linux-xfs@oss.sgi.com>; Mon, 5 Mar 2001 15:09:47 -0600 (CST)
Received: from jen.americas.sgi.com (IDENT:root@jen.americas.sgi.com [128.162.184.86]) by daisy-e185.americas.sgi.com (SGI-8.9.3/SGI-server-1.7) with ESMTP id PAA43792 for <linux-xfs@oss.sgi.com>; Mon, 5 Mar 2001 15:09:47 -0600 (CST)
From:   Steve Lord <lord@sgi.com>
Received: by jen.americas.sgi.com (8.11.0/SGI-client-1.7) id f25L93n06014; Mon, 5 Mar 2001 15:09:03 -0600
Message-Id: <200103052109.f25L93n06014@jen.americas.sgi.com>
Date:   Mon, 5 Mar 2001 15:09:03 -0600
Subject: TAKE - more pagebuf code cleanup, minor tweak
To:     unlisted-recipients:; (no To-header on input)
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing

Date:  Mon Mar  5 12:58:46 PST 2001
Workarea:  jen.americas.sgi.com:/src/lord/xfs-linux.2.4

The following file(s) were checked into:
  bonnie.engr.sgi.com:/isms/slinx/2.4.x-xfs


Modid:  2.4.x-xfs:slinx:88924a
linux/fs/pagebuf/page_buf.c - 1.62
linux/fs/pagebuf/page_buf_io.c - 1.64
	- more code cleanup



From owner-linux-xfs@oss.sgi.com Mon Mar  5 14:47:33 2001
Received:  by oss.sgi.com id <S553797AbRCEWrX>;
	Mon, 5 Mar 2001 14:47:23 -0800
Received: from mh.instantis.com ([216.216.10.58]:47260 "EHLO
        halftrack.instantis.com") by oss.sgi.com with ESMTP
	id <S553660AbRCEWq7>; Mon, 5 Mar 2001 14:46:59 -0800
Received: from instantis.com (buggy.instantis.com [10.0.0.146])
	by halftrack.instantis.com (8.9.1b+Sun/8.9.1) with ESMTP id OAA03924
	for <linux-xfs@oss.sgi.com>; Mon, 5 Mar 2001 14:52:16 -0800 (PST)
Message-ID: <3AA3FC2C.732878E7@instantis.com>
Date:   Mon, 05 Mar 2001 14:50:52 -0600
From:   Sid Wilroy <sid@instantis.com>
Reply-To: sid@instantis.com
Organization: Instantis.com
X-Mailer: Mozilla 4.7 [en] (WinNT; I)
X-Accept-Language: en
MIME-Version: 1.0
To:     linux-xfs@oss.sgi.com
Subject: web site problems
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing

Link to "contact project owner(s)" doesn't work..
http://linux-xfs.sgi.com/projects/xfs/survey.html doesn't work...


From owner-linux-xfs@oss.sgi.com Mon Mar  5 14:55:32 2001
Received:  by oss.sgi.com id <S553797AbRCEWzW>;
	Mon, 5 Mar 2001 14:55:22 -0800
Received: from pneumatic-tube.sgi.com ([204.94.214.22]:46905 "EHLO
        pneumatic-tube.sgi.com") by oss.sgi.com with ESMTP
	id <S553660AbRCEWzG>; Mon, 5 Mar 2001 14:55:06 -0800
Received: from zeus-fddi.americas.sgi.com (zeus-fddi.americas.sgi.com [128.162.8.103]) by pneumatic-tube.sgi.com (980327.SGI.8.8.8-aspam/980310.SGI-aspam) via ESMTP id PAA05876
	for <linux-xfs@oss.sgi.com>; Mon, 5 Mar 2001 15:04:47 -0800 (PST)
	mail_from (lord@sgi.com)
Received: from daisy-e185.americas.sgi.com (daisy-e194.americas.sgi.com [128.162.194.214]) by zeus-fddi.americas.sgi.com (8.9.3/americas-smart-nospam1.1) with ESMTP id QAA1030646; Mon, 5 Mar 2001 16:53:48 -0600 (CST)
Received: from jen.americas.sgi.com (IDENT:root@jen.americas.sgi.com [128.162.184.86]) by daisy-e185.americas.sgi.com (SGI-8.9.3/SGI-server-1.7) with ESMTP id QAA75188; Mon, 5 Mar 2001 16:53:47 -0600 (CST)
Received: from jen.americas.sgi.com by jen.americas.sgi.com (8.11.0/SGI-client-1.7) via ESMTP id f25Mr3611164; Mon, 5 Mar 2001 16:53:03 -0600
Message-Id: <200103052253.f25Mr3611164@jen.americas.sgi.com>
X-Mailer: exmh version 2.2 06/23/2000 with nmh-1.0.4
To:     sid@instantis.com
cc:     linux-xfs@oss.sgi.com
Subject: Re: web site problems 
In-Reply-To: Message from Sid Wilroy <sid@instantis.com> 
   of "Mon, 05 Mar 2001 14:50:52 CST." <3AA3FC2C.732878E7@instantis.com> 
Date:   Mon, 05 Mar 2001 16:53:03 -0600
From:   Steve Lord <lord@sgi.com>
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing

> Link to "contact project owner(s)" doesn't work..
> http://linux-xfs.sgi.com/projects/xfs/survey.html doesn't work...


Thanks,

The website was moved to a separate machine from the original oss (an all
XFS machine by the way) looks like we are still missing bits.

Steve



From owner-linux-xfs@oss.sgi.com Mon Mar  5 15:30:12 2001
Received:  by oss.sgi.com id <S553799AbRCEXaD>;
	Mon, 5 Mar 2001 15:30:03 -0800
Received: from mail11.jump.net ([206.196.91.11]:1162 "EHLO mail11.jump.net")
	by oss.sgi.com with ESMTP id <S553796AbRCEX3v>;
	Mon, 5 Mar 2001 15:29:51 -0800
Received: from Porter (aus-dsl-dhcp1-26.customer.jump.net [63.163.168.26])
	by mail11.jump.net (8.10.2/) with ESMTP id f25NTmX18783;
	Mon, 5 Mar 2001 17:29:48 -0600 (CST)
Message-Id: <200103052329.f25NTmX18783@mail11.jump.net>
Subject: Re: web site problems
From:   Eric Sandeen <sandeen@jump.net>
To:     Steve Lord <lord@sgi.com>
Cc:     linux-xfs@oss.sgi.com
In-Reply-To: <200103052253.f25Mr3611164@jen.americas.sgi.com>
Content-Type: text/plain
X-Mailer: Evolution (0.8/+cvs.2001.02.15.08.59 - Preview Release)
Date:   05 Mar 2001 17:30:24 -0600
Mime-Version: 1.0
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing

I checked in a fix (maybe) for this - it was a relative link which only
exists on oss.sgi.com - I made it absolute.  It points to a cgi script
which (I guess) looks at the referring URL to find out who to mail to -
if it parses ok from linux-xfs.sgi.com URL it should work.  At least the
link won't be broken.  :)

-Eric


On 05 Mar 2001 16:53:03 -0600, Steve Lord wrote:
> > Link to "contact project owner(s)" doesn't work..
> > http://linux-xfs.sgi.com/projects/xfs/survey.html doesn't work...
> 
> 
> Thanks,
> 
> The website was moved to a separate machine from the original oss (an all
> XFS machine by the way) looks like we are still missing bits.
> 
> Steve
> 


From owner-linux-xfs@oss.sgi.com Mon Mar  5 16:15:14 2001
Received:  by oss.sgi.com id <S553813AbRCFAPE>;
	Mon, 5 Mar 2001 16:15:04 -0800
Received: from mh.instantis.com ([216.216.10.58]:38046 "EHLO
        halftrack.instantis.com") by oss.sgi.com with ESMTP
	id <S553795AbRCFAOy>; Mon, 5 Mar 2001 16:14:54 -0800
Received: from instantis.com (buggy.instantis.com [10.0.0.146])
	by halftrack.instantis.com (8.9.1b+Sun/8.9.1) with ESMTP id QAA07862;
	Mon, 5 Mar 2001 16:20:10 -0800 (PST)
Message-ID: <3AA410C6.ECEC5A86@instantis.com>
Date:   Mon, 05 Mar 2001 16:18:46 -0600
From:   Sid Wilroy <sid@instantis.com>
Reply-To: sid@instantis.com
Organization: Instantis.com
X-Mailer: Mozilla 4.7 [en] (WinNT; I)
X-Accept-Language: en
MIME-Version: 1.0
To:     Steve Lord <lord@sgi.com>
CC:     linux-xfs@oss.sgi.com
Subject: qmail xfs
References: <200103052253.f25Mr3611164@jen.americas.sgi.com>
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing

You guys rock!! I installed xfs to test qmail qeueing. Got the cpu load up to
40 using uptime. No server carsh!!! I couldn't enter a command but it didn't
crash and kept processing mail. I'm very new to xfs so maybe someone could
point me into the right direction about optimizing xfs. Currently I only have
one file system running xfs /var/qmail/queue. Has anyone else setup xfs on
linux dealing with mail queues?

Steve Lord wrote:

> > Link to "contact project owner(s)" doesn't work..
> > http://linux-xfs.sgi.com/projects/xfs/survey.html doesn't work...
>
> Thanks,
>
> The website was moved to a separate machine from the original oss (an all
> XFS machine by the way) looks like we are still missing bits.
>
> Steve


From owner-linux-xfs@oss.sgi.com Mon Mar  5 16:17:23 2001
Received:  by oss.sgi.com id <S553813AbRCFARO>;
	Mon, 5 Mar 2001 16:17:14 -0800
Received: from h217n3fls20o974.telia.com ([212.181.166.217]:10244 "HELO
        neoblade.net") by oss.sgi.com with SMTP id <S553795AbRCFARA>;
	Mon, 5 Mar 2001 16:17:00 -0800
Received: from student.liu.se (neoblade.net [192.168.1.10])
	by neoblade.net (Postfix) with ESMTP
	id 1A9C7189131; Tue,  6 Mar 2001 01:18:57 +0100 (CET)
Message-ID: <3AA42CE6.3070103@student.liu.se>
Date:   Tue, 06 Mar 2001 01:18:46 +0100
From:   Joakim Bodin <joabo552@student.liu.se>
User-Agent: Mozilla/5.0 (X11; U; Linux 2.4.2-XFS i686; en-US; 0.9) Gecko/20010305
X-Accept-Language: sv, en
MIME-Version: 1.0
To:     Steve Lord <lord@sgi.com>
Cc:     Linux-XFS Mailing List <linux-xfs@oss.sgi.com>
Subject: Re: fsr.xfs locking up computer
References: <200103052007.f25K7wl03508@jen.americas.sgi.com>
Content-Type: text/plain; charset=us-ascii; format=flowed
Content-Transfer-Encoding: 7bit
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing

Steve Lord wrote:

>> I've been using fsr.xfs for many months now and it's been working fine. 
>> The exception to that was when I tried to run it yesterday (2001 march 
>> 4) and it hardlocked my system. This "feature" seem to be a bit 
>> unsupported from reading the documentation but I still felt that I 
>> should report it as it has worked before. I didn't find anything in the 
>> logs after this though.
>> 
>> Joakim Bodin
> 
> 
> 
> How exacty are you running fsr? Have you updated the fsr binary recently,
> or are you running an old one (I doubt this matters). I tried a few basic
> tests and fsr appeared to be functioning.
> 
> Steve
> 
> 
I updated my xfsprogs rpm to version 1.1.3 and updated my kernel to run 
the latest cvs (as of march 6 0:10 GMT) and tried running the command 
"fsr.xfs" as root. The kernel oops:ed shortly after this and not even 
sysrq command worked, after doing a hard reset the computer came up fine 
repairing the xfs partitions. The oops wasn't recorded in any logs. I 
compiled the xfsprogs with gcc-2.96 and the kernel with kgcc, I'm 
running ide udma33 with kio enabled.

Joakim Bodin


From owner-linux-xfs@oss.sgi.com Mon Mar  5 18:13:43 2001
Received:  by oss.sgi.com id <S553833AbRCFCNd>;
	Mon, 5 Mar 2001 18:13:33 -0800
Received: from pneumatic-tube.sgi.com ([204.94.214.22]:13644 "EHLO
        pneumatic-tube.sgi.com") by oss.sgi.com with ESMTP
	id <S553829AbRCFCNN>; Mon, 5 Mar 2001 18:13:13 -0800
Received: from larry.melbourne.sgi.com (larry.melbourne.sgi.com [134.14.52.130]) by pneumatic-tube.sgi.com (980327.SGI.8.8.8-aspam/980310.SGI-aspam) via SMTP id SAA00089
	for <linux-xfs@oss.sgi.com>; Mon, 5 Mar 2001 18:22:54 -0800 (PST)
	mail_from (nathans@wobbly.melbourne.sgi.com)
Received: from wobbly.melbourne.sgi.com (wobbly.melbourne.sgi.com [134.14.55.135]) by larry.melbourne.sgi.com (950413.SGI.8.6.12/950213.SGI.AUTOCF) via ESMTP id NAA14408; Tue, 6 Mar 2001 13:11:48 +1100
Received: (from nathans@localhost) by wobbly.melbourne.sgi.com (980427.SGI.8.8.8/980728.SGI.AUTOCF) id NAA82462; Tue, 6 Mar 2001 13:11:45 +1100 (EDT)
From:   "Nathan Scott" <nathans@wobbly.melbourne.sgi.com>
Message-Id: <10103061311.ZM191464@wobbly.melbourne.sgi.com>
Date:   Tue, 6 Mar 2001 13:11:43 -0400
In-Reply-To: Joakim Bodin <joabo552@student.liu.se>
        "Re: fsr.xfs locking up computer" (Mar  6,  1:18am)
References: <200103052007.f25K7wl03508@jen.americas.sgi.com> 
	<3AA42CE6.3070103@student.liu.se>
X-Mailer: Z-Mail (3.2.3 08feb96 MediaMail)
To:     Joakim Bodin <joabo552@student.liu.se>, Steve Lord <lord@sgi.com>
Subject: Re: fsr.xfs locking up computer
Cc:     Linux-XFS Mailing List <linux-xfs@oss.sgi.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing

On Mar 6,  1:18am, Joakim Bodin wrote:
> Subject: Re: fsr.xfs locking up computer
> > 
> I updated my xfsprogs rpm to version 1.1.3 and updated my kernel to run 

xfs_fsr is in the xfsdump package (it relies on the extended
attributes library too) - did you update that too?  Currently
it's at version 1.0.2.  Not sure this well help though - this
is most likely to be an issue in the kernel code.

cheers.

-- 
Nathan

From owner-linux-xfs@oss.sgi.com Mon Mar  5 19:26:24 2001
Received:  by oss.sgi.com id <S553859AbRCFD0P>;
	Mon, 5 Mar 2001 19:26:15 -0800
Received: from sgi.SGI.COM ([192.48.153.1]:65037 "EHLO sgi.com")
	by oss.sgi.com with ESMTP id <S553854AbRCFD0B>;
	Mon, 5 Mar 2001 19:26:01 -0800
Received: from madurai.engr.sgi.com ([163.154.5.75]) 
	by sgi.com (980327.SGI.8.8.8-aspam/980304.SGI-aspam:
       SGI does not authorize the use of its proprietary
       systems or networks for unsolicited or bulk email
       from the Internet.) 
	via ESMTP id TAA04916
	for <linux-xfs@oss.sgi.com>; Mon, 5 Mar 2001 19:26:01 -0800 (PST)
	mail_from (ananth@sgi.com)
Received: from sgi.com (mango.engr.sgi.com [163.154.5.76]) by madurai.engr.sgi.com (980427.SGI.8.8.8/970903.SGI.AUTOCF) via ESMTP id TAA06244 for <linux-xfs@oss.sgi.com>; Mon, 5 Mar 2001 19:20:40 -0800 (PST)
Message-ID: <3AA45852.B0499C02@sgi.com>
Date:   Mon, 05 Mar 2001 19:24:02 -0800
From:   Rajagopal Ananthanarayanan <ananth@sgi.com>
X-Mailer: Mozilla 4.72 [en] (X11; U; Linux 2.2.16-4SGI_20smp i686)
X-Accept-Language: en
MIME-Version: 1.0
To:     linux-xfs@oss.sgi.com
Subject: XFS and NFS
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing


Some in this list had asked about interactions between NFS & XFS,
so I decided to do a small test: compile a kernel over NFS.
In short, except for administrative issues (I have never configured
NFS on linux before), the actual compilation job went through fine.
The only odd message from NFS was this on the client side:

	nfs_stat_to_errno: bad nfs status return value: 11

but it didn't seem to affect the compilation: the resultant kernel booted fine.
The server: 
	o 1 CPU 1GB exporting XFS SCSI Disk
	o installed with XFS 0.9 CD
	o running latest CVS kernel
The client:
	o 2 CPU 64MB SCSI Disk
	o installed with Redhat 6.2
	o running latest CVS kernel

For compilation I used make -j6.

Hope this helps,

--------------------------------------------------------------------------
Rajagopal Ananthanarayanan ("ananth")
Member Technical Staff, SGI.
--------------------------------------------------------------------------

From owner-linux-xfs@oss.sgi.com Tue Mar  6 00:19:18 2001
Received:  by oss.sgi.com id <S553895AbRCFITI>;
	Tue, 6 Mar 2001 00:19:08 -0800
Received: from groucho.maths.monash.edu.au ([130.194.160.211]:27183 "EHLO
        groucho.maths.monash.edu.au") by oss.sgi.com with ESMTP
	id <S553735AbRCFISm>; Tue, 6 Mar 2001 00:18:42 -0800
Received: (from rjh@localhost)
	by groucho.maths.monash.edu.au (8.8.8/8.8.8) id IAA02777
	for linux-xfs@oss.sgi.com; Tue, 6 Mar 2001 08:18:37 GMT
From:   Robin Humble <rjh@groucho.maths.monash.edu.au>
Message-Id: <200103060818.IAA02777@groucho.maths.monash.edu.au>
Subject: more on NFS performance
To:     linux-xfs@oss.sgi.com
Date:   Tue, 6 Mar 2001 19:18:36 +1100 (EDT)
X-Mailer: ELM [version 2.5 PL1]
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing


More about slow NFS writes... sorry it's a bit long.
It gets more interesting towards the end :)

I'm running cvs from today on a dual celeron 500, 512M memory. RedHat
7.1beta.  I'm using a uni-processor kernel and no RAID of any sort.
The disk is a 7200RPM Maxtor ATA100 UDMA mode 5 and is mounted with
logbufs=8,kiocluster. The fs was made with -l size=32768b. I test it
with
  time dd if=/dev/zero of=bigFile bs=1024k count=2500
as this is close to the real application it'll be used for (multi-
gigabyte writes of simulation data).

the summary is:
writes to local disk run at ~30MByte/s. writes over NFSv3 or v2 run
at ~1 MByte/s after the memory cache on the server fils up, and the
cpu load is 90%+ with all of it being system time. The light on the
drive is on maybe 5% of the time. The expected write speed is ~10MByte/s
as this is a 100Mbit network. Throughputs for various count= sizes
are here:

 write size (MByte)   throughput    time
  same as count=      (MBytes/s)
        50               7         7.2s
       100               6         17.7s
       200               4         53s
       400               2.3       2mins 48s
       800               1.6       8mins 14s

using ext2 instead of XFS (same disk, just umount, mkfs, remount, re-export):
       800               9.6       1min 23s

When using XFS the machine appears to be basically using all its CPU
running nfsd's. A machine with a faster CPU could well write faster???
'top' says that there is consistant 90%+ system time usage.  When
using ext2 there's about 50% cpu system time being used by the nfsd's.

Like I've reported before, the behaviour is empirically the same on my
machine at home with completely different hardware. It's also the
same asymptotic ~1MB/s bandwidth whether kio or kiocluster or neither
are used. The logbufs option doesn't matter, nor does NFS v2 or v3 on
the server, or whether I'm writing from a Linux box, SGI or Tru64 machine.

When using kiocluster I did once get messages like these logged about
one every few seconds:
  ll_rw_kio: request size [10752] not a multiple of device [21:01] block-size [4096]
  ll_rw_kio: request size [17920] not a multiple of device [21:01] block-size [4096]
  ll_rw_kio: request size [25088] not a multiple of device [21:01] block-size [4096]
is this a clue???



Ok, I did the smart thing to enable bonnie++ runs in non-infinite time
and pulled all except 128M of memory out of the machine :-) for all
of the above tests it was 512MB. Now I get:

 write size (MByte)   throughput   time
  same as count=      (MBytes/s)
        50               7         7.2s
       100               5         18.6s
       200               4         45.5s
       400               4         1min 39s
       800              3.8        3mins 31s

using ext2 instead of XFS (same blurb as before):
       800               9.6       1min 23s

Ok, so this is MASSIVELY weird. Why should LESS memory in the server
boost speed of NFS writes to XFS??!?!?!?!?! The nfsd load on the
server is still 90%+ in system time. Are the buffers in memory
thrashing themselves and thrashing nfsd?

Anyway, this is lots better but still a factor of 2.5 away from ext2 :-(

I had another thought - I turned off the nfs locking daemons with:
 /etc/rc.d/init.d/nfslock stop
but this makes no difference.

Ok, bonnie++ results on NFS mounted dirs. This is from an IRIX64 O2k with
4G of memory to the celeron500 with 128M memory:

Version 1.00h     ------Sequential Output------ --Sequential Input- --Random-
                  -Per Chr- --Block-- -Rewrite- -Per Chr- --Block-- --Seeks--
Machine      Size K/sec %CP K/sec %CP K/sec %CP K/sec %CP K/sec %CP  /sec %CP
to XFS       256M  1217  30  1300   7  2380  18  4107  98 107168 98 1222.4 31
to ext2      256M  2790  73  4807  27  8107  35  4110  99 104429 99 1228.3 31

This shows some of the same behaviour as all my dd tests I guess - XFS
writes being down by a factor of 2.5 over ext2.
Load on the server is the usual 80%+ for XFS and maybe 30% for ext2.
The O2k has 4G of memory so that's probably why some read results are
way too high - but I don't care about them really - only interested in
writes for now. I'll try out reads later!! :)

Anyway - enough - 3am and time to go home after 10 hours on this stuff.

Please let me know if anyone else sees these same effects and if there
are any explanations, suggestions, solutions, or if there's anything
else you'd like me to try out :-)

cheers,
robin

From owner-linux-xfs@oss.sgi.com Tue Mar  6 04:27:11 2001
Received:  by oss.sgi.com id <S553682AbRCFM1B>;
	Tue, 6 Mar 2001 04:27:01 -0800
Received: from esparrall.udg.es ([130.206.124.16]:49820 "EHLO esparrall.udg.es")
	by oss.sgi.com with ESMTP id <S553650AbRCFM0h>;
	Tue, 6 Mar 2001 04:26:37 -0800
Received: from gcs by esparrall.udg.es with local (Exim 3.22 #1 (Debian))
	id 14aGXA-0003de-00
	for <linux-xfs@oss.sgi.com>; Tue, 06 Mar 2001 13:25:44 +0100
Date:   Tue, 6 Mar 2001 13:25:44 +0100
From:   GCS <gcs@agentsinside.com>
To:     linux-xfs@oss.sgi.com
Subject: Stable Debian + bonnie -> unmount problems
Message-ID: <20010306132544.A13949@esparrall.udg.es>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.3.15i
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing

Hello all!

I have a stable Debian (2.2 rev2), only comes from people.debian.org. This is
the way to install 4.0.2 on Potato. Kernel 2.4.2-XFS from CVS, a week ago.
Created a 400 megs file, and xfs on it. Mounted on loopback, and I have
tried 'bonnie -s 380', which caused a segmentation fault. It should not be
a problem, but the 'filesystem' looks full, without any file.
# umount /mnt
umount: /mnt: device is busy

# fuser -v /mnt
           USER        PID ACCESS COMMAND
/mnt       root     kernel mount  /mnt
What can I do?
Thanks, Laszlo

From owner-linux-xfs@oss.sgi.com Tue Mar  6 07:01:45 2001
Received:  by oss.sgi.com id <S553902AbRCFPBf>;
	Tue, 6 Mar 2001 07:01:35 -0800
Received: from pneumatic-tube.sgi.com ([204.94.214.22]:15468 "EHLO
        pneumatic-tube.sgi.com") by oss.sgi.com with ESMTP
	id <S553899AbRCFPBQ>; Tue, 6 Mar 2001 07:01:16 -0800
Received: from zeus-fddi.americas.sgi.com (zeus-fddi.americas.sgi.com [128.162.8.103]) by pneumatic-tube.sgi.com (980327.SGI.8.8.8-aspam/980310.SGI-aspam) via ESMTP id HAA02549
	for <linux-xfs@oss.sgi.com>; Tue, 6 Mar 2001 07:10:58 -0800 (PST)
	mail_from (lord@sgi.com)
Received: from daisy-e185.americas.sgi.com (daisy-e194.americas.sgi.com [128.162.194.214]) by zeus-fddi.americas.sgi.com (8.9.3/americas-smart-nospam1.1) with ESMTP id IAA1033339; Tue, 6 Mar 2001 08:59:54 -0600 (CST)
Received: from jen.americas.sgi.com (IDENT:root@jen.americas.sgi.com [128.162.184.86]) by daisy-e185.americas.sgi.com (SGI-8.9.3/SGI-server-1.7) with ESMTP id IAA40040; Tue, 6 Mar 2001 08:59:54 -0600 (CST)
Received: from jen.americas.sgi.com by jen.americas.sgi.com (8.11.0/SGI-client-1.7) via ESMTP id f26Ewo218045; Tue, 6 Mar 2001 08:58:51 -0600
Message-Id: <200103061458.f26Ewo218045@jen.americas.sgi.com>
X-Mailer: exmh version 2.2 06/23/2000 with nmh-1.0.4
To:     GCS <gcs@agentsinside.com>
cc:     linux-xfs@oss.sgi.com
Subject: Re: Stable Debian + bonnie -> unmount problems 
In-Reply-To: Message from GCS <gcs@agentsinside.com> 
   of "Tue, 06 Mar 2001 13:25:44 +0100." <20010306132544.A13949@esparrall.udg.es> 
Date:   Tue, 06 Mar 2001 08:58:50 -0600
From:   Steve Lord <lord@sgi.com>
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing

> Hello all!
> 
> I have a stable Debian (2.2 rev2), only comes from people.debian.org. This is
> the way to install 4.0.2 on Potato. Kernel 2.4.2-XFS from CVS, a week ago.
> Created a 400 megs file, and xfs on it. Mounted on loopback, and I have
> tried 'bonnie -s 380', which caused a segmentation fault. It should not be
> a problem, but the 'filesystem' looks full, without any file.
> # umount /mnt
> umount: /mnt: device is busy
> 
> # fuser -v /mnt
>            USER        PID ACCESS COMMAND
> /mnt       root     kernel mount  /mnt
> What can I do?
> Thanks, Laszlo

loop may be the problem here - it is broken in 2.4.2 which our tree is 
currently based on. Any chance of finding a small disk partition to use
instead, this would verify the problem as being in the loop device? 
Jens Axboe's loop patch does not apply cleanly to our tree or I would
tell you to apply it.

Any volunteers to make a version of loop6 which applies to 2.4.2-XFS?

Steve

p.s. looks like 2.4.3 will fix loop fs problems though.



From owner-linux-xfs@oss.sgi.com Tue Mar  6 07:33:25 2001
Received:  by oss.sgi.com id <S553906AbRCFPdP>;
	Tue, 6 Mar 2001 07:33:15 -0800
Received: from sgi.SGI.COM ([192.48.153.1]:13140 "EHLO sgi.com")
	by oss.sgi.com with ESMTP id <S553901AbRCFPdI>;
	Tue, 6 Mar 2001 07:33:08 -0800
Received: from zeus-fddi.americas.sgi.com (128-162-8-103.americas.sgi.com [128.162.8.103]) 
	by sgi.com (980327.SGI.8.8.8-aspam/980304.SGI-aspam:
       SGI does not authorize the use of its proprietary
       systems or networks for unsolicited or bulk email
       from the Internet.) 
	via ESMTP id HAA06984
	for <linux-xfs@oss.sgi.com>; Tue, 6 Mar 2001 07:33:07 -0800 (PST)
	mail_from (lord@sgi.com)
Received: from daisy-e185.americas.sgi.com (daisy-e194.americas.sgi.com [128.162.194.214]) by zeus-fddi.americas.sgi.com (8.9.3/americas-smart-nospam1.1) with ESMTP id JAA1036956 for <linux-xfs@oss.sgi.com>; Tue, 6 Mar 2001 09:31:51 -0600 (CST)
Received: from jen.americas.sgi.com (IDENT:root@jen.americas.sgi.com [128.162.184.86]) by daisy-e185.americas.sgi.com (SGI-8.9.3/SGI-server-1.7) with ESMTP id JAA16152 for <linux-xfs@oss.sgi.com>; Tue, 6 Mar 2001 09:31:50 -0600 (CST)
From:   Steve Lord <lord@sgi.com>
Received: by jen.americas.sgi.com (8.11.0/SGI-client-1.7) id f26FUlH22945; Tue, 6 Mar 2001 09:30:47 -0600
Message-Id: <200103061530.f26FUlH22945@jen.americas.sgi.com>
Date:   Tue, 6 Mar 2001 09:30:47 -0600
Subject: TAKE - fix the DIOINFO ioctl call
To:     unlisted-recipients:; (no To-header on input)
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing

XFS supports an ioctl call to report valid sizes for direct I/O operations,
one of the fields (maximum size) was returning zero which confused user space
code.

Date:  Tue Mar  6 07:30:57 PST 2001
Workarea:  jen.americas.sgi.com:/src/lord/xfs-linux.2.4

The following file(s) were checked into:
  bonnie.engr.sgi.com:/isms/slinx/2.4.x-xfs


Modid:  2.4.x-xfs:slinx:89039a
linux/fs/xfs/xfs_dmapi.c - 1.31
	- Fix d_maxiosz calculation.

linux/include/linux/page_buf.h - 1.79
	- prototype for max direct io size call.

linux/fs/xfs/linux/xfs_ioctl.c - 1.32
	- In the DIOINFO ioctl call, report the correct d_maxiosz to user space.

linux/fs/pagebuf/page_buf_io.c - 1.65
	- Add call to report the maximum direct I/O size which will be performed as
	  a single operation.



From owner-linux-xfs@oss.sgi.com Tue Mar  6 08:20:45 2001
Received:  by oss.sgi.com id <S553910AbRCFQUg>;
	Tue, 6 Mar 2001 08:20:36 -0800
Received: from sgi.SGI.COM ([192.48.153.1]:39026 "EHLO sgi.com")
	by oss.sgi.com with ESMTP id <S553907AbRCFQUS>;
	Tue, 6 Mar 2001 08:20:18 -0800
Received: from zeus-fddi.americas.sgi.com (128-162-8-103.americas.sgi.com [128.162.8.103]) 
	by sgi.com (980327.SGI.8.8.8-aspam/980304.SGI-aspam:
       SGI does not authorize the use of its proprietary
       systems or networks for unsolicited or bulk email
       from the Internet.) 
	via ESMTP id IAA02091
	for <linux-xfs@oss.sgi.com>; Tue, 6 Mar 2001 08:20:18 -0800 (PST)
	mail_from (lord@sgi.com)
Received: from daisy-e185.americas.sgi.com (daisy-e194.americas.sgi.com [128.162.194.214]) by zeus-fddi.americas.sgi.com (8.9.3/americas-smart-nospam1.1) with ESMTP id KAA1036986; Tue, 6 Mar 2001 10:18:59 -0600 (CST)
Received: from jen.americas.sgi.com (IDENT:root@jen.americas.sgi.com [128.162.184.86]) by daisy-e185.americas.sgi.com (SGI-8.9.3/SGI-server-1.7) with ESMTP id KAA87585; Tue, 6 Mar 2001 10:18:59 -0600 (CST)
Received: from jen.americas.sgi.com by jen.americas.sgi.com (8.11.0/SGI-client-1.7) via ESMTP id f26GHsh24687; Tue, 6 Mar 2001 10:17:55 -0600
Message-Id: <200103061617.f26GHsh24687@jen.americas.sgi.com>
X-Mailer: exmh version 2.2 06/23/2000 with nmh-1.0.4
To:     Robin Humble <rjh@groucho.maths.monash.edu.au>
cc:     linux-xfs@oss.sgi.com
Subject: Re: more on NFS performance 
In-Reply-To: Message from Robin Humble <rjh@groucho.maths.monash.edu.au> 
   of "Tue, 06 Mar 2001 19:18:36 +1100." <200103060818.IAA02777@groucho.maths.monash.edu.au> 
Date:   Tue, 06 Mar 2001 10:17:54 -0600
From:   Steve Lord <lord@sgi.com>
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing


OK, I think I know what the problem with NFS - to a certain extent anyway,
for my next trick, I need to work out how we fix it!

Basically it looks like the stateless nature of NFS is the killer here,
at the end of each RPC call we drop the reference count on the inode,
this causes us to truncate out extra space beyond the end of file. On
the next write we go do it again...... There is code in the irix version
to deal with this, but I am not sure if it needs NFS changes or not.

The nfsd threads end up in places like this most of the time:

[1]kdb> btp 1270
    EBP       EIP         Function(args)
0xc74f3d84 0xc0127c15 truncate_list_pages+0x21
                               kernel .text 0xc0100000 0xc0127bf4 0xc0127dec
           0xc0127e48 truncate_inode_pages+0x5c (0xc33b6c64, 0x1006e000, 0x0)
                               kernel .text 0xc0100000 0xc0127dec 0xc0127e80
           0xc018f6ce pagebuf_inval+0x1a (0xc33b6bc0, 0x1006e000, 0x0, 0x0)
                               kernel .text 0xc0100000 0xc018f6b4 0xc018f6d4
           0xc01ece41 fs_tosspages+0x29 (0xc17a3d30, 0x1006e000, 0x0, 0xffffffff, 0xffffffff)
                               kernel .text 0xc0100000 0xc01ece18 0xc01ece48
           0xc01cd21f xfs_itruncate_start+0x8f (0xc17a3d18, 0x1, 0x1006e000, 0x0, 0xc17a3d18)
                               kernel .text 0xc0100000 0xc01cd190 0xc01cd228
           0xc01e55f1 xfs_inactive_free_eofblocks+0x1e5 (0xc75bac00, 0xc17a3d18)
                               kernel .text 0xc0100000 0xc01e540c 0xc01e56dc
           0xc01e5d54 xfs_release+0x74 (0xc17a3d30)
                               kernel .text 0xc0100000 0xc01e5ce0 0xc01e5db8
           0xc01ecbb8 linvfs_release+0x24 (0xc33b6bc0, 0xc74f3ec4)
                               kernel .text 0xc0100000 0xc01ecb94 0xc01ecbc0
           0xc015f35e nfsd_close+0x1e (0xc74f3ec4)
                               kernel .text 0xc0100000 0xc015f340 0xc015f390
           0xc015f97d nfsd_write+0x295 (0xc5ff1600, 0xc64416e0, 0x1006c000, 0x0, 0xc64500ec)
                               kernel .text 0xc0100000 0xc015f6e8 0xc015f990
           0xc015cba4 nfsd_proc_write+0xb4 (0xc5ff1600, 0xc64415e0, 0xc64416e0)
[1]more> 
                               kernel .text 0xc0100000 0xc015caf0 0xc015cbac
           0xc015c213 nfsd_dispatch+0xcb (0xc5ff1600, 0xc1d10014)
                               kernel .text 0xc0100000 0xc015c148 0xc015c2b0
           0xc0298948 svc_process+0x2ac (0xc3a11b60, 0xc5ff1600)
                               kernel .text 0xc0100000 0xc029869c 0xc0298be0
           0xc015bfba nfsd+0x1ca
                               kernel .text 0xc0100000 0xc015bdf0 0xc015c148
           0xc010750f kernel_thread+0x23
                               kernel .text 0xc0100000 0xc01074ec 0xc010751c

The slower with more memory thing I am not sure about though.

Steve


> 
> More about slow NFS writes... sorry it's a bit long.
> It gets more interesting towards the end :)
> 
> I'm running cvs from today on a dual celeron 500, 512M memory. RedHat
> 7.1beta.  I'm using a uni-processor kernel and no RAID of any sort.
> The disk is a 7200RPM Maxtor ATA100 UDMA mode 5 and is mounted with
> logbufs=8,kiocluster. The fs was made with -l size=32768b. I test it
> with
>   time dd if=/dev/zero of=bigFile bs=1024k count=2500
> as this is close to the real application it'll be used for (multi-
> gigabyte writes of simulation data).
> 
> the summary is:
> writes to local disk run at ~30MByte/s. writes over NFSv3 or v2 run
> at ~1 MByte/s after the memory cache on the server fils up, and the
> cpu load is 90%+ with all of it being system time. The light on the
> drive is on maybe 5% of the time. The expected write speed is ~10MByte/s
> as this is a 100Mbit network. Throughputs for various count= sizes
> are here:
> 
>  write size (MByte)   throughput    time
>   same as count=      (MBytes/s)
>         50               7         7.2s
>        100               6         17.7s
>        200               4         53s
>        400               2.3       2mins 48s
>        800               1.6       8mins 14s
> 
> using ext2 instead of XFS (same disk, just umount, mkfs, remount, re-export):
>        800               9.6       1min 23s
> 
> When using XFS the machine appears to be basically using all its CPU
> running nfsd's. A machine with a faster CPU could well write faster???
> 'top' says that there is consistant 90%+ system time usage.  When
> using ext2 there's about 50% cpu system time being used by the nfsd's.
> 
> Like I've reported before, the behaviour is empirically the same on my
> machine at home with completely different hardware. It's also the
> same asymptotic ~1MB/s bandwidth whether kio or kiocluster or neither
> are used. The logbufs option doesn't matter, nor does NFS v2 or v3 on
> the server, or whether I'm writing from a Linux box, SGI or Tru64 machine.
> 
> When using kiocluster I did once get messages like these logged about
> one every few seconds:
>   ll_rw_kio: request size [10752] not a multiple of device [21:01] block-size
>  [4096]
>   ll_rw_kio: request size [17920] not a multiple of device [21:01] block-size
>  [4096]
>   ll_rw_kio: request size [25088] not a multiple of device [21:01] block-size
>  [4096]
> is this a clue???
> 
> 
> 
> Ok, I did the smart thing to enable bonnie++ runs in non-infinite time
> and pulled all except 128M of memory out of the machine :-) for all
> of the above tests it was 512MB. Now I get:
> 
>  write size (MByte)   throughput   time
>   same as count=      (MBytes/s)
>         50               7         7.2s
>        100               5         18.6s
>        200               4         45.5s
>        400               4         1min 39s
>        800              3.8        3mins 31s
> 
> using ext2 instead of XFS (same blurb as before):
>        800               9.6       1min 23s
> 
> Ok, so this is MASSIVELY weird. Why should LESS memory in the server
> boost speed of NFS writes to XFS??!?!?!?!?! The nfsd load on the
> server is still 90%+ in system time. Are the buffers in memory
> thrashing themselves and thrashing nfsd?
> 
> Anyway, this is lots better but still a factor of 2.5 away from ext2 :-(
> 
> I had another thought - I turned off the nfs locking daemons with:
>  /etc/rc.d/init.d/nfslock stop
> but this makes no difference.
> 
> Ok, bonnie++ results on NFS mounted dirs. This is from an IRIX64 O2k with
> 4G of memory to the celeron500 with 128M memory:
> 
> Version 1.00h     ------Sequential Output------ --Sequential Input- --Random-
>                   -Per Chr- --Block-- -Rewrite- -Per Chr- --Block-- --Seeks--
> Machine      Size K/sec %CP K/sec %CP K/sec %CP K/sec %CP K/sec %CP  /sec %CP
> to XFS       256M  1217  30  1300   7  2380  18  4107  98 107168 98 1222.4 31
> to ext2      256M  2790  73  4807  27  8107  35  4110  99 104429 99 1228.3 31
> 
> This shows some of the same behaviour as all my dd tests I guess - XFS
> writes being down by a factor of 2.5 over ext2.
> Load on the server is the usual 80%+ for XFS and maybe 30% for ext2.
> The O2k has 4G of memory so that's probably why some read results are
> way too high - but I don't care about them really - only interested in
> writes for now. I'll try out reads later!! :)
> 
> Anyway - enough - 3am and time to go home after 10 hours on this stuff.
> 
> Please let me know if anyone else sees these same effects and if there
> are any explanations, suggestions, solutions, or if there's anything
> else you'd like me to try out :-)
> 
> cheers,
> robin



From owner-linux-xfs@oss.sgi.com Tue Mar  6 09:03:25 2001
Received:  by oss.sgi.com id <S553922AbRCFRDP>;
	Tue, 6 Mar 2001 09:03:15 -0800
Received: from pneumatic-tube.sgi.com ([204.94.214.22]:17273 "EHLO
        pneumatic-tube.sgi.com") by oss.sgi.com with ESMTP
	id <S553916AbRCFRDD>; Tue, 6 Mar 2001 09:03:03 -0800
Received: from zeus-fddi.americas.sgi.com (zeus-fddi.americas.sgi.com [128.162.8.103]) by pneumatic-tube.sgi.com (980327.SGI.8.8.8-aspam/980310.SGI-aspam) via ESMTP id JAA06955
	for <linux-xfs@oss.sgi.com>; Tue, 6 Mar 2001 09:12:46 -0800 (PST)
	mail_from (lord@sgi.com)
Received: from daisy-e185.americas.sgi.com (daisy-e194.americas.sgi.com [128.162.194.214]) by zeus-fddi.americas.sgi.com (8.9.3/americas-smart-nospam1.1) with ESMTP id LAA1037671; Tue, 6 Mar 2001 11:01:46 -0600 (CST)
Received: from jen.americas.sgi.com (IDENT:root@jen.americas.sgi.com [128.162.184.86]) by daisy-e185.americas.sgi.com (SGI-8.9.3/SGI-server-1.7) with ESMTP id LAA61064; Tue, 6 Mar 2001 11:01:45 -0600 (CST)
Received: from jen.americas.sgi.com by jen.americas.sgi.com (8.11.0/SGI-client-1.7) via ESMTP id f26H0fA30616; Tue, 6 Mar 2001 11:00:41 -0600
Message-Id: <200103061700.f26H0fA30616@jen.americas.sgi.com>
X-Mailer: exmh version 2.2 06/23/2000 with nmh-1.0.4
To:     Robin Humble <rjh@groucho.maths.monash.edu.au>,
        linux-xfs@oss.sgi.com
Subject: Re: more on NFS performance 
In-Reply-To: Message from Steve Lord <lord@sgi.com> 
   of "Tue, 06 Mar 2001 10:17:54 CST." <200103061617.f26GHsh24687@jen.americas.sgi.com> 
Date:   Tue, 06 Mar 2001 11:00:41 -0600
From:   Steve Lord <lord@sgi.com>
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing


Hi,

Is anyone on the list NFS savvy? I need a way from the filp or dentry of
telling that a write came from the NFS server, and I cannot really look
at the ops field in the dentry, since NFS built as a module will mess this up.

I was hoping for some flag or something which I can use to test for this
from within xfs, currently I am doing a strcmp on the task name - which is
not something I want to do really.

Steve


From owner-linux-xfs@oss.sgi.com Tue Mar  6 09:29:26 2001
Received:  by oss.sgi.com id <S553916AbRCFR3G>;
	Tue, 6 Mar 2001 09:29:06 -0800
Received: from ns.suse.de ([213.95.15.193]:6149 "HELO Cantor.suse.de")
	by oss.sgi.com with SMTP id <S553919AbRCFR2z>;
	Tue, 6 Mar 2001 09:28:55 -0800
Received: from Hermes.suse.de (Hermes.suse.de [213.95.15.136])
	by Cantor.suse.de (Postfix) with ESMTP
	id 8B1931E193; Tue,  6 Mar 2001 18:28:49 +0100 (MET)
Date:   Tue, 6 Mar 2001 18:28:42 +0100
From:   Andi Kleen <ak@suse.de>
To:     Steve Lord <lord@sgi.com>
Cc:     Robin Humble <rjh@groucho.maths.monash.edu.au>,
        linux-xfs@oss.sgi.com
Subject: Re: more on NFS performance
Message-ID: <20010306182842.A10901@gruyere.muc.suse.de>
References: <lord@sgi.com> <200103061700.f26H0fA30616@jen.americas.sgi.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.2.5i
In-Reply-To: <200103061700.f26H0fA30616@jen.americas.sgi.com>; from lord@sgi.com on Tue, Mar 06, 2001 at 11:00:41AM -0600
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing

On Tue, Mar 06, 2001 at 11:00:41AM -0600, Steve Lord wrote:
> 
> Hi,
> 
> Is anyone on the list NFS savvy? I need a way from the filp or dentry of
> telling that a write came from the NFS server, and I cannot really look
> at the ops field in the dentry, since NFS built as a module will mess this up.
> 
> I was hoping for some flag or something which I can use to test for this
> from within xfs, currently I am doing a strcmp on the task name - which is
> not something I want to do really.

There isn't a clean way currently AFAIK, but I guess a flag (O_MORE?) could 
be set in nfsd_open() in the filp.   I guess it would be something similar 
to MSG_MORE to turn off the early truncation? If you do that, don't you need a 
timer to do the truncation later?


-Andi

From owner-linux-xfs@oss.sgi.com Tue Mar  6 09:35:36 2001
Received:  by oss.sgi.com id <S553922AbRCFRf0>;
	Tue, 6 Mar 2001 09:35:26 -0800
Received: from usintout02.bms.com ([165.89.129.232]:22739 "EHLO
        usintout02.net.bms.com") by oss.sgi.com with ESMTP
	id <S553919AbRCFRfN>; Tue, 6 Mar 2001 09:35:13 -0800
Received: from mendel.lvl.pri.bms.com ([127.0.0.1])
 by mochilla.bms.com (PMDF V5.2-33 #38470)
 with ESMTP id <0G9S00J3DE3V1D@mochilla.bms.com> for linux-xfs@oss.sgi.com;
 Tue,  6 Mar 2001 12:33:32 -0500 (EST)
Received: from bms.com (localhost [127.0.0.1])	by mendel.lvl.pri.bms.com
 (8.9.3/8.9.3) with ESMTP id MAA93431	for <linux-xfs@oss.sgi.com>; Tue,
 06 Mar 2001 12:33:12 -0500 (EST)
Date:   Tue, 06 Mar 2001 12:33:12 -0500
From:   Stewart Samuels <ssamuels@bms.com>
Subject: XFS problems
To:     linux-xfs@oss.sgi.com
Message-id: <3AA51F58.A97A7053@bms.com>
MIME-version: 1.0
X-Mailer: Mozilla 4.7C-SGI [en] (X11; I; IRIX64 6.5 IP30)
Content-type: multipart/alternative;
 boundary="------------97C2B323B57D260763C2E1A8"
X-Accept-Language: en
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing


--------------97C2B323B57D260763C2E1A8
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit

To whom it may concern,

I am experiencing several problems with the XFS for Linux Pre-Release.
I used the SGI
XFS for Red Hat 7.0 installer on a CD that I got from the SGI booth at
the LinuxWorld Conference and
Expo in New York.  I have logged into your web site and read all the
caveats but am still
experiencing major difficulties from several perspectives.  They are as
follows:

1)    After completing the XFS installation for some reason, trying to
boot from the Red Hat
        7.0 CDs is now extremely slow.  The process to bring up Anaconda
or for that matter
        just load vmlinuz from the CD takes about 20-30 minutes.
Normally, it only takes
        my system (1 gHz Athlon) less than a minute to boot this far.

2)    I cannot reload the "up2date", "up2date-gnome", "LPRng", and
"usermode" packages
        once the system boots because the system does not setup and of
the cdrom (of which I
        have 2) devices.

3)    I get error messages that ppp requires other packages (possible
due to not being able to
        reload those packages in paragraph 2 above) when trying to use
my modem to establish
        a ppp connection.  I can dial out manually, but not with ppp.

4)    XFree86 and gpm are not configured.  This seems to be a problem
with /devfs in that everytime
        I reboot the system I must relink the modem and the mouse to the
appropriate device.


Please help by sending me procedures which would allow me to get around
these problems if possible.
Thanks.


            Stewart

--

                                            Stewart

-------------------------------------------------------------------
                           .-----.
                          / \-_-/ \
                         /_-_\ /_-_\
                         ----- -----
                         \-_-/ \-_-/
                          \ /_-_\ /
                           `-----'
    Bristol-Myers Squibb Pharmaceutical Research Institute
    ------------------------------------------------------

Stewart Samuels           Internet Address: ssamuels@bms.com
Senior Systems Engineer   Telephone       : 609-252-5575
Macromolecular Modeling   FAX             : 609-252-6030
===================================================================



--------------97C2B323B57D260763C2E1A8
Content-Type: text/html; charset=us-ascii
Content-Transfer-Encoding: 7bit

<!doctype html public "-//w3c//dtd html 4.0 transitional//en">
<html>
To whom it may concern,
<p>I am experiencing several problems with the XFS for Linux Pre-Release.&nbsp;
I used the SGI
<br>XFS for Red Hat 7.0 installer on a CD that I got from the SGI booth
at the LinuxWorld Conference and
<br>Expo in New York.&nbsp; I have logged into your web site and read all
the caveats but am still
<br>experiencing major difficulties from several perspectives.&nbsp; They
are as follows:
<p>1)&nbsp;&nbsp;&nbsp; After completing the XFS installation for some
reason, trying to boot from the Red Hat
<br>&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; 7.0 CDs is now extremely
slow.&nbsp; The process to bring up Anaconda or for that matter
<br>&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; just load vmlinuz from the
CD takes about 20-30 minutes.&nbsp; Normally, it only takes
<br>&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; my system (1 gHz Athlon)
less than a minute to boot this far.
<p>2)&nbsp;&nbsp;&nbsp; I cannot reload the "up2date", "up2date-gnome",
"LPRng", and "usermode" packages
<br>&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; once the system boots because
the system does not setup and of the cdrom (of which I
<br>&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; have 2) devices.
<p>3)&nbsp;&nbsp;&nbsp; I get error messages that ppp requires other packages
(possible due to not being able to
<br>&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; reload those packages in
paragraph 2 above) when trying to use my modem to establish
<br>&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; a ppp connection.&nbsp;
I can dial out manually, but not with ppp.
<p>4)&nbsp;&nbsp;&nbsp; XFree86 and gpm are not configured.&nbsp; This
seems to be a problem with /devfs in that everytime
<br>&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; I reboot the system I must
relink the modem and the mouse to the appropriate device.
<br>&nbsp;
<p>Please help by sending me procedures which would allow me to get around
these problems if possible.
<br>Thanks.
<br>&nbsp;
<p>&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; Stewart
<pre>--&nbsp;

&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; Stewart

-------------------------------------------------------------------
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; .-----.
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; / \-_-/ \&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; /_-_\ /_-_\&nbsp;&nbsp;
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; ----- -----&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; \-_-/ \-_-/&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; \ /_-_\ /
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; `-----'
&nbsp;&nbsp;&nbsp; Bristol-Myers Squibb Pharmaceutical Research Institute
&nbsp;&nbsp;&nbsp; ------------------------------------------------------

Stewart Samuels&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; Internet Address: ssamuels@bms.com
Senior Systems Engineer&nbsp;&nbsp; Telephone&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; : 609-252-5575
Macromolecular Modeling&nbsp;&nbsp; FAX&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; : 609-252-6030
===================================================================</pre>
&nbsp;</html>

--------------97C2B323B57D260763C2E1A8--


From owner-linux-xfs@oss.sgi.com Tue Mar  6 09:49:06 2001
Received:  by oss.sgi.com id <S553922AbRCFRs4>;
	Tue, 6 Mar 2001 09:48:56 -0800
Received: from pneumatic-tube.sgi.com ([204.94.214.22]:2431 "EHLO
        pneumatic-tube.sgi.com") by oss.sgi.com with ESMTP
	id <S553919AbRCFRsq>; Tue, 6 Mar 2001 09:48:46 -0800
Received: from zeus-fddi.americas.sgi.com (zeus-fddi.americas.sgi.com [128.162.8.103]) by pneumatic-tube.sgi.com (980327.SGI.8.8.8-aspam/980310.SGI-aspam) via ESMTP id JAA06687
	for <linux-xfs@oss.sgi.com>; Tue, 6 Mar 2001 09:58:28 -0800 (PST)
	mail_from (lord@sgi.com)
Received: from daisy-e185.americas.sgi.com (daisy-e194.americas.sgi.com [128.162.194.214]) by zeus-fddi.americas.sgi.com (8.9.3/americas-smart-nospam1.1) with ESMTP id LAA1035781; Tue, 6 Mar 2001 11:47:27 -0600 (CST)
Received: from jen.americas.sgi.com (IDENT:root@jen.americas.sgi.com [128.162.184.86]) by daisy-e185.americas.sgi.com (SGI-8.9.3/SGI-server-1.7) with ESMTP id LAA28897; Tue, 6 Mar 2001 11:47:27 -0600 (CST)
Received: from jen.americas.sgi.com by jen.americas.sgi.com (8.11.0/SGI-client-1.7) via ESMTP id f26HkMj31473; Tue, 6 Mar 2001 11:46:22 -0600
Message-Id: <200103061746.f26HkMj31473@jen.americas.sgi.com>
X-Mailer: exmh version 2.2 06/23/2000 with nmh-1.0.4
To:     Andi Kleen <ak@suse.de>
cc:     Robin Humble <rjh@groucho.maths.monash.edu.au>,
        linux-xfs@oss.sgi.com
Subject: Re: more on NFS performance 
In-Reply-To: Message from Andi Kleen <ak@suse.de> 
   of "Tue, 06 Mar 2001 18:28:42 +0100." <20010306182842.A10901@gruyere.muc.suse.de> 
Date:   Tue, 06 Mar 2001 11:46:22 -0600
From:   Steve Lord <lord@sgi.com>
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing

> On Tue, Mar 06, 2001 at 11:00:41AM -0600, Steve Lord wrote:
> > 
> > Hi,
> > 
> > Is anyone on the list NFS savvy? I need a way from the filp or dentry of
> > telling that a write came from the NFS server, and I cannot really look
> > at the ops field in the dentry, since NFS built as a module will mess this 
> up.
> > 
> > I was hoping for some flag or something which I can use to test for this
> > from within xfs, currently I am doing a strcmp on the task name - which is
> > not something I want to do really.
> 
> There isn't a clean way currently AFAIK, but I guess a flag (O_MORE?) could 
> be set in nfsd_open() in the filp.   I guess it would be something similar 
> to MSG_MORE to turn off the early truncation? If you do that, don't you need 
> a 
> timer to do the truncation later?

The reference cache does its own cleanup internally, it is effectively
done out of write_super. Under Irix an internal open flag is the way
it is done, I just didn't want to dink with NFS to fix this, but maybe
I should.

Steve


> 
> 
> -Andi



From owner-linux-xfs@oss.sgi.com Tue Mar  6 10:05:36 2001
Received:  by oss.sgi.com id <S553922AbRCFSF0>;
	Tue, 6 Mar 2001 10:05:26 -0800
Received: from mail11.jump.net ([206.196.91.11]:13515 "EHLO mail11.jump.net")
	by oss.sgi.com with ESMTP id <S553919AbRCFSFO>;
	Tue, 6 Mar 2001 10:05:14 -0800
Received: from Porter (aus-dsl-dhcp1-26.customer.jump.net [63.163.168.26])
	by mail11.jump.net (8.10.2/) with ESMTP id f26I5Cl12414;
	Tue, 6 Mar 2001 12:05:12 -0600 (CST)
Message-Id: <200103061805.f26I5Cl12414@mail11.jump.net>
Subject: Re: XFS problems
From:   Eric Sandeen <sandeen@jump.net>
To:     Stewart Samuels <ssamuels@bms.com>
Cc:     linux-xfs@oss.sgi.com
In-Reply-To: <3AA51F58.A97A7053@bms.com>
Content-Type: text/plain
X-Mailer: Evolution (0.8/+cvs.2001.02.15.08.59 - Preview Release)
Date:   06 Mar 2001 12:05:55 -0600
Mime-Version: 1.0
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing

On 06 Mar 2001 12:33:12 -0500, Stewart Samuels wrote:

> To whom it may concern,

> 1)    After completing the XFS installation for some reason, trying to
> boot from the Red Hat 7.0 CDs is now extremely slow.  The process to 
> bring up Anaconda or for that matter just load vmlinuz from the CD 
> takes about 20-30 minutes.  Normally, it only takes my system 
> (1 gHz Athlon) less than a minute to boot this far.

The install CD itself boots slowly?  I've never seen this - the system
will boot somewhat slowly after installation due to the large ramdisk,
and you'll want to recompile the kernel to avoid this.  Not sure what
to tell you if the CD is slow...

> 2)    I cannot reload the "up2date", "up2date-gnome", "LPRng", and
> "usermode" packages once the system boots because the system does not 
> setup and of the cdrom (of which I have 2) devices.

This is because of the (now infamous) decision to enable devfs on the
system - read up on the devfs faq to see how to load mount your CD-ROMS
(you'll need to load the appropriate modules to get the devices in /dev
to appear)

> 3)    I get error messages that ppp requires other packages (possible
> due to not being able to reload those packages in paragraph 2 above) 
> when trying to use my modem to establish a ppp connection.  I can dial
> out manually, but not with ppp.

What packages does it complain about?  Once you get your cdrom mounted
and install those packages hopefully it'll work.
 
> 4)    XFree86 and gpm are not configured.  This seems to be a problem
> with /devfs in that everytime I reboot the system I must relink the 
> modem and the mouse to the appropriate device.

This is devfs again.  /dev/modem and /dev/mouse aren't there anymore,
you can tell devfsd to create them, or point your config files to the
new devfs devices.

-Eric


From owner-linux-xfs@oss.sgi.com Tue Mar  6 11:43:45 2001
Received:  by oss.sgi.com id <S553930AbRCFTn0>;
	Tue, 6 Mar 2001 11:43:26 -0800
Received: from esparrall.udg.es ([130.206.124.16]:9600 "EHLO esparrall.udg.es")
	by oss.sgi.com with ESMTP id <S553926AbRCFTnY>;
	Tue, 6 Mar 2001 11:43:24 -0800
Received: from gcs by esparrall.udg.es with local (Exim 3.22 #1 (Debian))
	id 14aNM8-0000FZ-00
	for <linux-xfs@oss.sgi.com>; Tue, 06 Mar 2001 20:42:48 +0100
Date:   Tue, 6 Mar 2001 20:42:48 +0100
From:   GCS <gcs@agentsinside.com>
To:     linux-xfs@oss.sgi.com
Subject: Still module problems
Message-ID: <20010306204248.A845@esparrall.udg.es>
Mime-Version: 1.0
Content-Type: text/plain; charset=iso-8859-1
Content-Disposition: inline
Content-Transfer-Encoding: 8bit
User-Agent: Mutt/1.3.15i
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing

Hello all,

If someone has Debian, please contact me personally. I do not know why, but
I still have unresolved symbols in all modules. I have deleted the kernel
source + System.[map|old] + vmlinuz[.old], cvs update, cp -Rd linux /usr/src/,
mv linux linux-2.4.2-XFS etc.
After make install, 'ls -l /boot':
lrwxrwxrwx    1 root     root           20 MÁR  6 20:09 System.map -> System.map-2.4.2-XFS
-rw-r--r--    1 root     root       269967 DEC 25 15:54 System.map-2.2.18pre21
-rw-r--r--    1 root     root       405781 MÁR  6 20:09 System.map-2.4.2-XFS
lrwxrwxrwx    1 root     root           17 MÁR  6 20:09 vmlinuz -> vmlinuz-2.4.2-XFS
-rwxr-xr-x    1 root     root       985303 DEC 25 15:54 vmlinuz-2.2.18pre21
-rw-r--r--    1 root     root       703718 MÁR  6 20:09 vmlinuz-2.4.2-XFS
(Well, I just realized it does not have +x, but ...)

The directory /lib/modules/2.4.2-XFS exists and contains the modules.
Result -> unresolved symbols, even printk, which is such basic like the air
for us (imho).
Then, I unpacked the original kernel source, copied .config into it, and
built the kernel and booted without problems. :-0 What can it be?
I could 'tight it down' to this: /var/log/kern.log
Mar  6 20:06:50 esparrall kernel: Inspecting /boot/System.map
Mar  6 20:06:51 esparrall kernel: Loaded 14344 symbols from /boot/System.map.
Mar  6 20:06:51 esparrall kernel: Symbols match kernel version 2.4.2.
				  ^-- It should be 2.4.2-XFS, should not be?
Mar  6 20:06:51 esparrall kernel: No module symbols loaded.
				  ^-- Any reason for that?
Mar  6 20:06:51 esparrall kernel: Linux version 2.4.2-XFS (root@esparrall)

For the original kernel it is different:
Mar  6 20:14:42 esparrall kernel: Inspecting /boot/System.map
Mar  6 20:14:43 esparrall kernel: Loaded 14029 symbols from /boot/System.map.
Mar  6 20:14:43 esparrall kernel: Symbols match kernel version 2.4.2.
Mar  6 20:14:43 esparrall kernel: Loaded 201 symbols from 7 modules.
Mar  6 20:14:43 esparrall kernel: Linux version 2.4.2 (root@esparrall)

Modutils is version 2.4.2.

Any help is greatly appreciated. Thanks, Laszlo
Ps:Previously I had the same problem, but somewhen it began working, and I
did not care. But now it is the same, and I am very curious...

From owner-linux-xfs@oss.sgi.com Tue Mar  6 11:44:45 2001
Received:  by oss.sgi.com id <S553930AbRCFTof>;
	Tue, 6 Mar 2001 11:44:35 -0800
Received: from e21.nc.us.ibm.com ([32.97.136.227]:16368 "EHLO
        e21.nc.us.ibm.com") by oss.sgi.com with ESMTP id <S553926AbRCFToV>;
	Tue, 6 Mar 2001 11:44:21 -0800
Received: from southrelay02.raleigh.ibm.com (southrelay02.raleigh.ibm.com [9.37.3.209])
	by e21.nc.us.ibm.com (8.9.3/8.9.3) with ESMTP id OAA220622
	for <linux-xfs@oss.sgi.com>; Tue, 6 Mar 2001 14:39:29 -0600
Received: from d04nm201.raleigh.ibm.com (d04nm201.raleigh.ibm.com [9.67.228.38])
	by southrelay02.raleigh.ibm.com (8.11.2/NCO v4.95) with ESMTP id f26JhtF70444
	for <linux-xfs@oss.sgi.com>; Tue, 6 Mar 2001 14:44:00 -0500
Importance: Normal
Subject: [PATCH] for comments about kiobufs support for IDE
To:     linux-xfs@oss.sgi.com
From:   "Steve Best" <sbest@us.ibm.com>
Date:   Tue, 6 Mar 2001 14:43:53 -0500
Message-ID: <OFB21EB6B1.5B791E27-ON85256A07.006B80B6@raleigh.ibm.com>
X-MIMETrack: Serialize by Router on D04NM201/04/M/IBM(Release 5.0.6 |December 14, 2000) at
 03/06/2001 02:44:02 PM
MIME-Version: 1.0
Content-type: text/plain; charset=us-ascii
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing

Small comment update for ll_rw_kio. These comments probably
should have been updated when the ide support for kiobufs came in.


--- ll_rw_blk.org   Tue Mar  6 13:16:51 2001
+++ ll_rw_blk.c     Tue Mar  6 13:19:08 2001
@@ -1363,11 +1363,11 @@
     int correct_size, i;

     /*
-     * Only support SCSI disk for now.
+     * Only support SCSI and IDE disks for now.
      *
      * ENOSYS to indicate caller
      * should try ll_rw_block()
-     * for non-SCSI (e.g. IDE) disks.
+     * for other than SCSI or IDE disks.
      */
     if (!SCSI_DISK_MAJOR(MAJOR(dev)) && !IDE_DISK_MAJOR(MAJOR(dev))) {
          *error = -ENOSYS;

Steve


From owner-linux-xfs@oss.sgi.com Tue Mar  6 11:50:35 2001
Received:  by oss.sgi.com id <S553930AbRCFTuP>;
	Tue, 6 Mar 2001 11:50:15 -0800
Received: from sgi.SGI.COM ([192.48.153.1]:5390 "EHLO sgi.com")
	by oss.sgi.com with ESMTP id <S553926AbRCFTt6>;
	Tue, 6 Mar 2001 11:49:58 -0800
Received: from zeus-fddi.americas.sgi.com (128-162-8-103.americas.sgi.com [128.162.8.103]) 
	by sgi.com (980327.SGI.8.8.8-aspam/980304.SGI-aspam:
       SGI does not authorize the use of its proprietary
       systems or networks for unsolicited or bulk email
       from the Internet.) 
	via ESMTP id LAA07322
	for <linux-xfs@oss.sgi.com>; Tue, 6 Mar 2001 11:49:57 -0800 (PST)
	mail_from (lord@sgi.com)
Received: from daisy-e185.americas.sgi.com (daisy-e194.americas.sgi.com [128.162.194.214]) by zeus-fddi.americas.sgi.com (8.9.3/americas-smart-nospam1.1) with ESMTP id NAA1039027 for <linux-xfs@oss.sgi.com>; Tue, 6 Mar 2001 13:48:41 -0600 (CST)
Received: from jen.americas.sgi.com (IDENT:root@jen.americas.sgi.com [128.162.184.86]) by daisy-e185.americas.sgi.com (SGI-8.9.3/SGI-server-1.7) with ESMTP id NAA24113 for <linux-xfs@oss.sgi.com>; Tue, 6 Mar 2001 13:48:37 -0600 (CST)
From:   Steve Lord <lord@sgi.com>
Received: by jen.americas.sgi.com (8.11.0/SGI-client-1.7) id f26Joch03775; Tue, 6 Mar 2001 13:50:38 -0600
Message-Id: <200103061950.f26Joch03775@jen.americas.sgi.com>
Date:   Tue, 6 Mar 2001 13:50:38 -0600
Subject: TAKE - reintroduce the nfs inode reference cache
To:     unlisted-recipients:; (no To-header on input)
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing

This should help nfs server performance over xfs for writes. It puts
back some code which was removed last year and makes it work. This is
a reference cache for inodes written to via nfs so that we basically
keep the inodes in the system for a while and avoid their count dropping
down to zero and going through a lot of cleanup code.

the network here is a bit primitive (10 Mbit/sec) and I can easily
saturate it. Can someone operating over 100 Mbit ethernet let me
know if this helps, just write a file bigger than memory into nfs
over xfs.

Steve

Date:  Tue Mar  6 11:44:15 PST 2001
Workarea:  jen.americas.sgi.com:/src/lord/xfs-linux.2.4

The following file(s) were checked into:
  bonnie.engr.sgi.com:/isms/slinx/2.4.x-xfs


Modid:  2.4.x-xfs:slinx:89063a
linux/fs/xfs/xfs_rw.h - 1.58
	- prototypes for nfs reference cache

linux/fs/xfs/xfs_rw.c - 1.331
	- Reintroduce nfs reference cache

linux/fs/xfs/xfs_vnodeops.c - 1.490
	- remove inode being deleted  from nfs reference cache

linux/fs/xfs/xfs_vfsops.c - 1.310
	- purge nfs reference cache on unmount, and prune some entries during
	  sync activity.

linux/fs/xfs/xfs_inode.h - 1.144
	- Add reference cache field to inode

linux/fs/xfs/xfs_rename.c - 1.30
	- remove inode being renamed from nfs reference cache

linux/fs/xfs/linux/xfs_lrw.c - 1.80
	- Insert inodes from nfs writes into the inode reference cache



From owner-linux-xfs@oss.sgi.com Tue Mar  6 12:02:05 2001
Received:  by oss.sgi.com id <S553930AbRCFUB4>;
	Tue, 6 Mar 2001 12:01:56 -0800
Received: from deliverator.sgi.com ([204.94.214.10]:31584 "EHLO
        deliverator.sgi.com") by oss.sgi.com with ESMTP id <S553926AbRCFUBa>;
	Tue, 6 Mar 2001 12:01:30 -0800
Received: from cthulhu.engr.sgi.com (gate3-relay.engr.sgi.com [130.62.1.234]) by deliverator.sgi.com (980309.SGI.8.8.8-aspam-6.2/980310.SGI-aspam) via ESMTP id MAA21062
	for <linux-xfs@oss.sgi.com>; Tue, 6 Mar 2001 12:00:24 -0800 (PST)
	mail_from (chait@getafix.engr.sgi.com)
Received: from getafix.engr.sgi.com (IDENT:root@getafix.engr.sgi.com [163.154.5.110])
	by cthulhu.engr.sgi.com (SGI-8.9.3/8.9.3) with ESMTP id MAA61325;
	Tue, 6 Mar 2001 12:00:13 -0800 (PST)
Received: from localhost (chait@localhost)
	by getafix.engr.sgi.com (8.9.3/8.9.3) with ESMTP id LAA30549;
	Tue, 6 Mar 2001 11:56:05 -0500
Date:   Tue, 6 Mar 2001 11:56:05 -0500 (EST)
From:   Chaitanya Tumuluri <chait@getafix.engr.sgi.com>
To:     Steve Best <sbest@us.ibm.com>
cc:     linux-xfs@oss.sgi.com
Subject: Re: [PATCH] for comments about kiobufs support for IDE
In-Reply-To: <OFB21EB6B1.5B791E27-ON85256A07.006B80B6@raleigh.ibm.com>
Message-ID: <Pine.LNX.4.21.0103061154220.30317-100000@getafix.engr.sgi.com>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing

Hmmmm...I guess the changes to the comments got dropped in one of
our tree merges...I remember that Jens Axboe did have changes to 
these comments to reflect his IDE changes. Oh well...

Thanks for the patch, Steve.

Cheers,
-Chait.

On Tue, 6 Mar 2001, Steve Best wrote:
> Small comment update for ll_rw_kio. These comments probably
> should have been updated when the ide support for kiobufs came in.
> 
> 
> --- ll_rw_blk.org   Tue Mar  6 13:16:51 2001
> +++ ll_rw_blk.c     Tue Mar  6 13:19:08 2001
> @@ -1363,11 +1363,11 @@
>      int correct_size, i;
> 
>      /*
> -     * Only support SCSI disk for now.
> +     * Only support SCSI and IDE disks for now.
>       *
>       * ENOSYS to indicate caller
>       * should try ll_rw_block()
> -     * for non-SCSI (e.g. IDE) disks.
> +     * for other than SCSI or IDE disks.
>       */
>      if (!SCSI_DISK_MAJOR(MAJOR(dev)) && !IDE_DISK_MAJOR(MAJOR(dev))) {
>           *error = -ENOSYS;
> 
> Steve
> 


From owner-linux-xfs@oss.sgi.com Tue Mar  6 12:24:36 2001
Received:  by oss.sgi.com id <S553931AbRCFUYQ>;
	Tue, 6 Mar 2001 12:24:16 -0800
Received: from pneumatic-tube.sgi.com ([204.94.214.22]:55830 "EHLO
        pneumatic-tube.sgi.com") by oss.sgi.com with ESMTP
	id <S553926AbRCFUYA>; Tue, 6 Mar 2001 12:24:00 -0800
Received: from zeus-fddi.americas.sgi.com (zeus-fddi.americas.sgi.com [128.162.8.103]) by pneumatic-tube.sgi.com (980327.SGI.8.8.8-aspam/980310.SGI-aspam) via ESMTP id MAA03701
	for <linux-xfs@oss.sgi.com>; Tue, 6 Mar 2001 12:33:42 -0800 (PST)
	mail_from (lord@sgi.com)
Received: from daisy-e185.americas.sgi.com (daisy-e194.americas.sgi.com [128.162.194.214]) by zeus-fddi.americas.sgi.com (8.9.3/americas-smart-nospam1.1) with ESMTP id OAA1040716 for <linux-xfs@oss.sgi.com>; Tue, 6 Mar 2001 14:22:43 -0600 (CST)
Received: from jen.americas.sgi.com (IDENT:root@jen.americas.sgi.com [128.162.184.86]) by daisy-e185.americas.sgi.com (SGI-8.9.3/SGI-server-1.7) with ESMTP id OAA09686 for <linux-xfs@oss.sgi.com>; Tue, 6 Mar 2001 14:22:42 -0600 (CST)
Received: from jen.americas.sgi.com by jen.americas.sgi.com (8.11.0/SGI-client-1.7) via ESMTP id f26KOhb04680; Tue, 6 Mar 2001 14:24:43 -0600
Message-Id: <200103062024.f26KOhb04680@jen.americas.sgi.com>
X-Mailer: exmh version 2.2 06/23/2000 with nmh-1.0.4
To:     linux-xfs@oss.sgi.com
Subject: Re: TAKE - reintroduce the nfs inode reference cache 
In-Reply-To: Message from Steve Lord <lord@sgi.com> 
   of "Tue, 06 Mar 2001 13:50:38 CST." <200103061950.f26Joch03775@jen.americas.sgi.com> 
Date:   Tue, 06 Mar 2001 14:24:43 -0600
From:   Steve Lord <lord@sgi.com>
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing


OK, I am probably talking rubbish with this code, if someone feels like trying
it out please do and let me know if it helps. I thought I was seeing 
improvements, but thinking about the code, and trying the tests again I
am not seeing improvements.

Steve

> This should help nfs server performance over xfs for writes. It puts
> back some code which was removed last year and makes it work. This is
> a reference cache for inodes written to via nfs so that we basically
> keep the inodes in the system for a while and avoid their count dropping
> down to zero and going through a lot of cleanup code.
> 
> the network here is a bit primitive (10 Mbit/sec) and I can easily
> saturate it. Can someone operating over 100 Mbit ethernet let me
> know if this helps, just write a file bigger than memory into nfs
> over xfs.
> 
> Steve
> 
> Date:  Tue Mar  6 11:44:15 PST 2001
> Workarea:  jen.americas.sgi.com:/src/lord/xfs-linux.2.4
> 
> The following file(s) were checked into:
>   bonnie.engr.sgi.com:/isms/slinx/2.4.x-xfs
> 
> 
> Modid:  2.4.x-xfs:slinx:89063a
> linux/fs/xfs/xfs_rw.h - 1.58
> 	- prototypes for nfs reference cache
> 
> linux/fs/xfs/xfs_rw.c - 1.331
> 	- Reintroduce nfs reference cache
> 
> linux/fs/xfs/xfs_vnodeops.c - 1.490
> 	- remove inode being deleted  from nfs reference cache
> 
> linux/fs/xfs/xfs_vfsops.c - 1.310
> 	- purge nfs reference cache on unmount, and prune some entries during
> 	  sync activity.
> 
> linux/fs/xfs/xfs_inode.h - 1.144
> 	- Add reference cache field to inode
> 
> linux/fs/xfs/xfs_rename.c - 1.30
> 	- remove inode being renamed from nfs reference cache
> 
> linux/fs/xfs/linux/xfs_lrw.c - 1.80
> 	- Insert inodes from nfs writes into the inode reference cache
> 



From owner-linux-xfs@oss.sgi.com Tue Mar  6 12:30:46 2001
Received:  by oss.sgi.com id <S553931AbRCFUag>;
	Tue, 6 Mar 2001 12:30:36 -0800
Received: from pneumatic-tube.sgi.com ([204.94.214.22]:21783 "EHLO
        pneumatic-tube.sgi.com") by oss.sgi.com with ESMTP
	id <S553926AbRCFUaR>; Tue, 6 Mar 2001 12:30:17 -0800
Received: from madurai.engr.sgi.com (madurai.engr.sgi.com [163.154.5.75]) by pneumatic-tube.sgi.com (980327.SGI.8.8.8-aspam/980310.SGI-aspam) via ESMTP id MAA09772
	for <linux-xfs@oss.sgi.com>; Tue, 6 Mar 2001 12:39:59 -0800 (PST)
	mail_from (ananth@sgi.com)
Received: from sgi.com (mango.engr.sgi.com [163.154.5.76]) by madurai.engr.sgi.com (980427.SGI.8.8.8/970903.SGI.AUTOCF) via ESMTP id MAA07201; Tue, 6 Mar 2001 12:24:49 -0800 (PST)
Message-ID: <3AA54859.7CDCFA@sgi.com>
Date:   Tue, 06 Mar 2001 12:28:09 -0800
From:   Rajagopal Ananthanarayanan <ananth@sgi.com>
X-Mailer: Mozilla 4.72 [en] (X11; U; Linux 2.2.16-4SGI_20smp i686)
X-Accept-Language: en
MIME-Version: 1.0
To:     Steve Lord <lord@sgi.com>
CC:     linux-xfs@oss.sgi.com
Subject: Re: TAKE - reintroduce the nfs inode reference cache
References: <200103062024.f26KOhb04680@jen.americas.sgi.com>
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing

Steve Lord wrote:
> 
> OK, I am probably talking rubbish with this code, if someone feels like trying
> it out please do and let me know if it helps. I thought I was seeing
> improvements, but thinking about the code, and trying the tests again I
> am not seeing improvements.
> 

Ok, that's good ;-) I didn't see any improvements either,
so it isn't a NFS setup thing as I was starting to suspect ...

cheers,

ananth.

From owner-linux-xfs@oss.sgi.com Tue Mar  6 12:37:06 2001
Received:  by oss.sgi.com id <S553931AbRCFUgq>;
	Tue, 6 Mar 2001 12:36:46 -0800
Received: from sgi.SGI.COM ([192.48.153.1]:14883 "EHLO sgi.com")
	by oss.sgi.com with ESMTP id <S553926AbRCFUgk>;
	Tue, 6 Mar 2001 12:36:40 -0800
Received: from zeus-fddi.americas.sgi.com (128-162-8-103.americas.sgi.com [128.162.8.103]) 
	by sgi.com (980327.SGI.8.8.8-aspam/980304.SGI-aspam:
       SGI does not authorize the use of its proprietary
       systems or networks for unsolicited or bulk email
       from the Internet.) 
	via ESMTP id MAA00033
	for <linux-xfs@oss.sgi.com>; Tue, 6 Mar 2001 12:36:22 -0800 (PST)
	mail_from (lord@sgi.com)
Received: from daisy-e185.americas.sgi.com (daisy-e194.americas.sgi.com [128.162.194.214]) by zeus-fddi.americas.sgi.com (8.9.3/americas-smart-nospam1.1) with ESMTP id OAA1041209; Tue, 6 Mar 2001 14:35:04 -0600 (CST)
Received: from jen.americas.sgi.com (IDENT:root@jen.americas.sgi.com [128.162.184.86]) by daisy-e185.americas.sgi.com (SGI-8.9.3/SGI-server-1.7) with ESMTP id OAA56342; Tue, 6 Mar 2001 14:35:03 -0600 (CST)
Received: from jen.americas.sgi.com by jen.americas.sgi.com (8.11.0/SGI-client-1.7) via ESMTP id f26Kb4f06258; Tue, 6 Mar 2001 14:37:04 -0600
Message-Id: <200103062037.f26Kb4f06258@jen.americas.sgi.com>
X-Mailer: exmh version 2.2 06/23/2000 with nmh-1.0.4
To:     Rajagopal Ananthanarayanan <ananth@sgi.com>
cc:     Steve Lord <lord@sgi.com>, linux-xfs@oss.sgi.com
Subject: Re: TAKE - reintroduce the nfs inode reference cache 
In-Reply-To: Message from Rajagopal Ananthanarayanan <ananth@sgi.com> 
   of "Tue, 06 Mar 2001 12:28:09 PST." <3AA54859.7CDCFA@sgi.com> 
Date:   Tue, 06 Mar 2001 14:37:03 -0600
From:   Steve Lord <lord@sgi.com>
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing

> Steve Lord wrote:
> > 
> > OK, I am probably talking rubbish with this code, if someone feels like try
> ing
> > it out please do and let me know if it helps. I thought I was seeing
> > improvements, but thinking about the code, and trying the tests again I
> > am not seeing improvements.
> > 
> 
> Ok, that's good ;-) I didn't see any improvements either,
> so it isn't a NFS setup thing as I was starting to suspect ...
> 
> cheers,
> 
> ananth.

But I do see a way to reuse this cache for a different purpose, a problem
is we end up in linvfs_release at the end of each write via NFS. This is used
in the close case to free extra space allocated beyond the end of file. I want
to use the cache to stop this from happening to nfs on each write.

Steve




From owner-linux-xfs@oss.sgi.com Tue Mar  6 12:43:45 2001
Received:  by oss.sgi.com id <S553931AbRCFUng>;
	Tue, 6 Mar 2001 12:43:36 -0800
Received: from ns.suse.de ([213.95.15.193]:12300 "HELO Cantor.suse.de")
	by oss.sgi.com with SMTP id <S553926AbRCFUn1>;
	Tue, 6 Mar 2001 12:43:27 -0800
Received: from Hermes.suse.de (Hermes.suse.de [213.95.15.136])
	by Cantor.suse.de (Postfix) with ESMTP
	id CD8511E19D; Tue,  6 Mar 2001 21:43:25 +0100 (MET)
Date:   Tue, 6 Mar 2001 21:43:22 +0100
From:   Andi Kleen <ak@suse.de>
To:     Steve Lord <lord@sgi.com>
Cc:     linux-xfs@oss.sgi.com
Subject: Re: TAKE - reintroduce the nfs inode reference cache
Message-ID: <20010306214322.A14781@gruyere.muc.suse.de>
References: <lord@sgi.com> <200103062024.f26KOhb04680@jen.americas.sgi.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.2.5i
In-Reply-To: <200103062024.f26KOhb04680@jen.americas.sgi.com>; from lord@sgi.com on Tue, Mar 06, 2001 at 02:24:43PM -0600
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing

On Tue, Mar 06, 2001 at 02:24:43PM -0600, Steve Lord wrote:
> 
> OK, I am probably talking rubbish with this code, if someone feels like trying
> it out please do and let me know if it helps. I thought I was seeing 
> improvements, but thinking about the code, and trying the tests again I
> am not seeing improvements.

If XFS is doing the truncate on linux inode count zeroing (in i_ops->release)
then it should only happen when the file is expunged from the dcache, which
can be never and should not happen after every NFS RPC unless the machine
is trashing heavily. 


-Andi

From owner-linux-xfs@oss.sgi.com Tue Mar  6 12:56:35 2001
Received:  by oss.sgi.com id <S553933AbRCFU40>;
	Tue, 6 Mar 2001 12:56:26 -0800
Received: from pneumatic-tube.sgi.com ([204.94.214.22]:2330 "EHLO
        pneumatic-tube.sgi.com") by oss.sgi.com with ESMTP
	id <S553926AbRCFU4I>; Tue, 6 Mar 2001 12:56:08 -0800
Received: from zeus-fddi.americas.sgi.com (zeus-fddi.americas.sgi.com [128.162.8.103]) by pneumatic-tube.sgi.com (980327.SGI.8.8.8-aspam/980310.SGI-aspam) via ESMTP id NAA01886
	for <linux-xfs@oss.sgi.com>; Tue, 6 Mar 2001 13:05:50 -0800 (PST)
	mail_from (lord@sgi.com)
Received: from daisy-e185.americas.sgi.com (daisy-e194.americas.sgi.com [128.162.194.214]) by zeus-fddi.americas.sgi.com (8.9.3/americas-smart-nospam1.1) with ESMTP id OAA1041382; Tue, 6 Mar 2001 14:54:50 -0600 (CST)
Received: from jen.americas.sgi.com (IDENT:root@jen.americas.sgi.com [128.162.184.86]) by daisy-e185.americas.sgi.com (SGI-8.9.3/SGI-server-1.7) with ESMTP id OAA39690; Tue, 6 Mar 2001 14:54:50 -0600 (CST)
Received: from jen.americas.sgi.com by jen.americas.sgi.com (8.11.0/SGI-client-1.7) via ESMTP id f26Kuo106999; Tue, 6 Mar 2001 14:56:50 -0600
Message-Id: <200103062056.f26Kuo106999@jen.americas.sgi.com>
X-Mailer: exmh version 2.2 06/23/2000 with nmh-1.0.4
To:     Andi Kleen <ak@suse.de>
cc:     Steve Lord <lord@sgi.com>, linux-xfs@oss.sgi.com
Subject: Re: TAKE - reintroduce the nfs inode reference cache 
In-Reply-To: Message from Andi Kleen <ak@suse.de> 
   of "Tue, 06 Mar 2001 21:43:22 +0100." <20010306214322.A14781@gruyere.muc.suse.de> 
Date:   Tue, 06 Mar 2001 14:56:50 -0600
From:   Steve Lord <lord@sgi.com>
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing

> On Tue, Mar 06, 2001 at 02:24:43PM -0600, Steve Lord wrote:
> > 
> > OK, I am probably talking rubbish with this code, if someone feels like try
> ing
> > it out please do and let me know if it helps. I thought I was seeing 
> > improvements, but thinking about the code, and trying the tests again I
> > am not seeing improvements.
> 
> If XFS is doing the truncate on linux inode count zeroing (in i_ops->release)
> then it should only happen when the file is expunged from the dcache, which
> can be never and should not happen after every NFS RPC unless the machine
> is trashing heavily. 
> 
> 
> -Andi


This is f_ops release, not iops, the normal case is from close. A quick
mod to the code to not do the release in this case but delay it until later
appears to help. Ananth is going to test it on his 100Mbit network and we
will see if this helps. However, should the f_ops release code be the culprit
we may be able to find a simpler way to fix this.

Steve



From owner-linux-xfs@oss.sgi.com Tue Mar  6 13:05:06 2001
Received:  by oss.sgi.com id <S553933AbRCFVEr>;
	Tue, 6 Mar 2001 13:04:47 -0800
Received: from sgi.SGI.COM ([192.48.153.1]:45870 "EHLO sgi.com")
	by oss.sgi.com with ESMTP id <S553926AbRCFVEj>;
	Tue, 6 Mar 2001 13:04:39 -0800
Received: from madurai.engr.sgi.com ([163.154.5.75]) 
	by sgi.com (980327.SGI.8.8.8-aspam/980304.SGI-aspam:
       SGI does not authorize the use of its proprietary
       systems or networks for unsolicited or bulk email
       from the Internet.) 
	via ESMTP id NAA02567
	for <linux-xfs@oss.sgi.com>; Tue, 6 Mar 2001 13:04:37 -0800 (PST)
	mail_from (ananth@sgi.com)
Received: from sgi.com (mango.engr.sgi.com [163.154.5.76]) by madurai.engr.sgi.com (980427.SGI.8.8.8/970903.SGI.AUTOCF) via ESMTP id MAA07243; Tue, 6 Mar 2001 12:58:17 -0800 (PST)
Message-ID: <3AA55031.A4DE4994@sgi.com>
Date:   Tue, 06 Mar 2001 13:01:37 -0800
From:   Rajagopal Ananthanarayanan <ananth@sgi.com>
X-Mailer: Mozilla 4.72 [en] (X11; U; Linux 2.2.16-4SGI_20smp i686)
X-Accept-Language: en
MIME-Version: 1.0
To:     Steve Lord <lord@sgi.com>
CC:     Andi Kleen <ak@suse.de>, linux-xfs@oss.sgi.com
Subject: Re: TAKE - reintroduce the nfs inode reference cache
References: <200103062056.f26Kuo106999@jen.americas.sgi.com>
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing

Steve Lord wrote:

> 
> This is f_ops release, not iops, the normal case is from close. A quick
> mod to the code to not do the release in this case but delay it until later
> appears to help. Ananth is going to test it on his 100Mbit network and we
> will see if this helps. However, should the f_ops release code be the culprit
> we may be able to find a simpler way to fix this.


A quick check reveals that the lmddtest's that had problems before
now runs pretty good ... as fast as writing to ext2 over NFS.
In both cases, lmdd (sequential) write is giving about 11.2 Mbyte/sec.

Also, the server runs much "cooler": only about 30-40% cpu utilization,
again comparable to ext2 exported case. Ship it ;-)

ananth.

From owner-linux-xfs@oss.sgi.com Tue Mar  6 13:08:16 2001
Received:  by oss.sgi.com id <S553933AbRCFVIG>;
	Tue, 6 Mar 2001 13:08:06 -0800
Received: from sgi.SGI.COM ([192.48.153.1]:10544 "EHLO sgi.com")
	by oss.sgi.com with ESMTP id <S553926AbRCFVH7>;
	Tue, 6 Mar 2001 13:07:59 -0800
Received: from zeus-fddi.americas.sgi.com (128-162-8-103.americas.sgi.com [128.162.8.103]) 
	by sgi.com (980327.SGI.8.8.8-aspam/980304.SGI-aspam:
       SGI does not authorize the use of its proprietary
       systems or networks for unsolicited or bulk email
       from the Internet.) 
	via ESMTP id NAA06159
	for <linux-xfs@oss.sgi.com>; Tue, 6 Mar 2001 13:08:00 -0800 (PST)
	mail_from (lord@sgi.com)
Received: from daisy-e185.americas.sgi.com (daisy-e194.americas.sgi.com [128.162.194.214]) by zeus-fddi.americas.sgi.com (8.9.3/americas-smart-nospam1.1) with ESMTP id PAA1041428 for <linux-xfs@oss.sgi.com>; Tue, 6 Mar 2001 15:06:41 -0600 (CST)
Received: from jen.americas.sgi.com (IDENT:root@jen.americas.sgi.com [128.162.184.86]) by daisy-e185.americas.sgi.com (SGI-8.9.3/SGI-server-1.7) with ESMTP id PAA29932 for <linux-xfs@oss.sgi.com>; Tue, 6 Mar 2001 15:06:41 -0600 (CST)
From:   Steve Lord <lord@sgi.com>
Received: by jen.americas.sgi.com (8.11.0/SGI-client-1.7) id f26L8f907105; Tue, 6 Mar 2001 15:08:41 -0600
Message-Id: <200103062108.f26L8f907105@jen.americas.sgi.com>
Date:   Tue, 6 Mar 2001 15:08:41 -0600
Subject: TAKE - nfs reference cache version 2
To:     unlisted-recipients:; (no To-header on input)
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing

OK, this actually should fix the performance of writes, I will think
some more about ways to do this without all the code this adds, there
is probably a simpler way.

Date:  Tue Mar  6 13:05:35 PST 2001
Workarea:  jen.americas.sgi.com:/src/lord/xfs-linux.2.4

The following file(s) were checked into:
  bonnie.engr.sgi.com:/isms/slinx/2.4.x-xfs


Modid:  2.4.x-xfs:slinx:89070a
linux/fs/xfs/xfs_rw.c - 1.332
	- Change reference cache to do a VOP_RELEASE on removal from cache,
	  this will truncate out space allocated beyond end of file.

linux/fs/xfs/xfs_vnodeops.c - 1.491
	- Change release code to do nothing if an inode is in the reference cache



From owner-linux-xfs@oss.sgi.com Tue Mar  6 13:24:46 2001
Received:  by oss.sgi.com id <S553933AbRCFVYg>;
	Tue, 6 Mar 2001 13:24:36 -0800
Received: from ns.suse.de ([213.95.15.193]:47375 "HELO Cantor.suse.de")
	by oss.sgi.com with SMTP id <S553817AbRCFVY2>;
	Tue, 6 Mar 2001 13:24:28 -0800
Received: from Hermes.suse.de (Hermes.suse.de [213.95.15.136])
	by Cantor.suse.de (Postfix) with ESMTP
	id 881651E1B0; Tue,  6 Mar 2001 22:24:27 +0100 (MET)
Date:   Tue, 6 Mar 2001 22:23:56 +0100
From:   Andi Kleen <ak@suse.de>
To:     Steve Lord <lord@sgi.com>
Cc:     Andi Kleen <ak@suse.de>, linux-xfs@oss.sgi.com,
        neilb@cse.unsw.edu.au
Subject: Re: TAKE - reintroduce the nfs inode reference cache
Message-ID: <20010306222356.A15312@gruyere.muc.suse.de>
References: <ak@suse.de> <200103062056.f26Kuo106999@jen.americas.sgi.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.2.5i
In-Reply-To: <200103062056.f26Kuo106999@jen.americas.sgi.com>; from lord@sgi.com on Tue, Mar 06, 2001 at 02:56:50PM -0600
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing

On Tue, Mar 06, 2001 at 02:56:50PM -0600, Steve Lord wrote:
> > On Tue, Mar 06, 2001 at 02:24:43PM -0600, Steve Lord wrote:
> > > 
> > > OK, I am probably talking rubbish with this code, if someone feels like try
> > ing
> > > it out please do and let me know if it helps. I thought I was seeing 
> > > improvements, but thinking about the code, and trying the tests again I
> > > am not seeing improvements.
> > 
> > If XFS is doing the truncate on linux inode count zeroing (in i_ops->release)
> > then it should only happen when the file is expunged from the dcache, which
> > can be never and should not happen after every NFS RPC unless the machine
> > is trashing heavily. 
> > 
> > 
> > -Andi
> 
> 
> This is f_ops release, not iops, the normal case is from close. A quick
> mod to the code to not do the release in this case but delay it until later
> appears to help. Ananth is going to test it on his 100Mbit network and we
> will see if this helps. However, should the f_ops release code be the culprit
> we may be able to find a simpler way to fix this.

I think it would be best to fix nfsd to cache file handles instead of dentries
and only do release when the nfsfh leaves the nfsfh cache.
This way could also help ext2, which caches readahead context in struct file.

Put Neil Brown into cc. Neil, do you see any obstacles with that?


-Andi


From owner-linux-xfs@oss.sgi.com Tue Mar  6 13:48:56 2001
Received:  by oss.sgi.com id <S553926AbRCFVsg>;
	Tue, 6 Mar 2001 13:48:36 -0800
Received: from pneumatic-tube.sgi.com ([204.94.214.22]:27679 "EHLO
        pneumatic-tube.sgi.com") by oss.sgi.com with ESMTP
	id <S553817AbRCFVsK>; Tue, 6 Mar 2001 13:48:10 -0800
Received: from snort.melbourne.sgi.com (snort.melbourne.sgi.com [134.14.55.149]) by pneumatic-tube.sgi.com (980327.SGI.8.8.8-aspam/980310.SGI-aspam) via ESMTP id NAA05870
	for <linux-xfs@oss.sgi.com>; Tue, 6 Mar 2001 13:57:48 -0800 (PST)
	mail_from (nathans@snort.melbourne.sgi.com)
Received: (from nathans@localhost)
	by snort.melbourne.sgi.com (SGI-8.9.3/8.9.3) id IAA31477
	for linux-xfs@oss.sgi.com; Wed, 7 Mar 2001 08:46:47 +1100 (EST)
Date:   Wed, 7 Mar 2001 08:46:47 +1100 (EST)
From:   Nathan Scott <nathans@snort.melbourne.sgi.com>
Message-Id: <200103062146.IAA31477@snort.melbourne.sgi.com>
To:     linux-xfs@oss.sgi.com
Subject: TAKE - mkfile
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing

You'll need Steve's recent ioctl fix for this to work,
otherwise it gets itself into an infinite write loop.


Date:  Tue Mar  6 13:43:21 PST 2001
Workarea:  snort.melbourne.sgi.com:/diskb/build4/nathans/linux-xfs

The following file(s) were checked into:
  bonnie.engr.sgi.com:/isms/slinx/2.4.x-xfs


Modid:  2.4.x-xfs:slinx:89081a
cmd/xfsprogs/mkfile/xfs_mkfile.c - 1.2
	- reenable use of direct IO now that DIOINFO and O_DIRECT
	  are working.



From owner-linux-xfs@oss.sgi.com Tue Mar  6 14:09:56 2001
Received:  by oss.sgi.com id <S553933AbRCFWJh>;
	Tue, 6 Mar 2001 14:09:37 -0800
Received: from sgi.SGI.COM ([192.48.153.1]:12875 "EHLO sgi.com")
	by oss.sgi.com with ESMTP id <S553895AbRCFWJL>;
	Tue, 6 Mar 2001 14:09:11 -0800
Received: from snort.melbourne.sgi.com ([134.14.55.149]) 
	by sgi.com (980327.SGI.8.8.8-aspam/980304.SGI-aspam:
       SGI does not authorize the use of its proprietary
       systems or networks for unsolicited or bulk email
       from the Internet.) 
	via ESMTP id OAA06414
	for <linux-xfs@oss.sgi.com>; Tue, 6 Mar 2001 14:09:09 -0800 (PST)
	mail_from (nathans@snort.melbourne.sgi.com)
Received: (from nathans@localhost)
	by snort.melbourne.sgi.com (SGI-8.9.3/8.9.3) id JAA91907
	for linux-xfs@oss.sgi.com; Wed, 7 Mar 2001 09:07:52 +1100 (EST)
Date:   Wed, 7 Mar 2001 09:07:52 +1100 (EST)
From:   Nathan Scott <nathans@snort.melbourne.sgi.com>
Message-Id: <200103062207.JAA91907@snort.melbourne.sgi.com>
To:     linux-xfs@oss.sgi.com
Subject: TAKE - debug build
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing

Date:  Tue Mar  6 14:07:31 PST 2001
Workarea:  snort.melbourne.sgi.com:/diskb/build4/nathans/linux-xfs

The following file(s) were checked into:
  bonnie.engr.sgi.com:/isms/slinx/2.4.x-xfs


Modid:  2.4.x-xfs:slinx:89087a
linux/fs/xfs/xfs_rw.c - 1.333
	- fix a debug build - v_count field of vnode_t no longer exists.



From owner-linux-xfs@oss.sgi.com Tue Mar  6 15:12:26 2001
Received:  by oss.sgi.com id <S553936AbRCFXMQ>;
	Tue, 6 Mar 2001 15:12:16 -0800
Received: from ppp0.ocs.com.au ([203.34.97.3]:63503 "HELO mail.ocs.com.au")
	by oss.sgi.com with SMTP id <S553930AbRCFXMC>;
	Tue, 6 Mar 2001 15:12:02 -0800
Received: (qmail 16861 invoked from network); 6 Mar 2001 23:11:57 -0000
Received: from ocs3.ocs-net (192.168.255.3)
  by mail.ocs.com.au with SMTP; 6 Mar 2001 23:11:57 -0000
X-Mailer: exmh version 2.1.1 10/15/1999
From:   Keith Owens <kaos@melbourne.sgi.com>
To:     GCS <gcs@agentsinside.com>
cc:     linux-xfs@oss.sgi.com
Subject: Re: Still module problems 
In-reply-to: Your message of "Tue, 06 Mar 2001 20:42:48 BST."
             <20010306204248.A845@esparrall.udg.es> 
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Date:   Wed, 07 Mar 2001 10:11:56 +1100
Message-ID: <4511.983920316@ocs3.ocs-net>
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing

On Tue, 6 Mar 2001 20:42:48 +0100, 
GCS <gcs@agentsinside.com> wrote:
>If someone has Debian, please contact me personally. I do not know why, but
>I still have unresolved symbols in all modules.

http://www.tux.org/lkml/#s8-8

>Mar  6 20:06:50 esparrall kernel: Inspecting /boot/System.map
>Mar  6 20:06:51 esparrall kernel: Loaded 14344 symbols from /boot/System.map.
>Mar  6 20:06:51 esparrall kernel: Symbols match kernel version 2.4.2.
>				  ^-- It should be 2.4.2-XFS, should not be?

klogd handling of symbols and system.maps is broken.  Run klogd as
klogd -x.


From owner-linux-xfs@oss.sgi.com Tue Mar  6 15:12:37 2001
Received:  by oss.sgi.com id <S553934AbRCFXM1>;
	Tue, 6 Mar 2001 15:12:27 -0800
Received: from note.orchestra.cse.unsw.EDU.AU ([129.94.242.29]:24324 "HELO
        note.orchestra.cse.unsw.EDU.AU") by oss.sgi.com with SMTP
	id <S553935AbRCFXMO>; Tue, 6 Mar 2001 15:12:14 -0800
Received: From
	notabene.cse.unsw.edu.au ([129.94.211.194] == dulcimer.orchestra.cse.unsw.EDU.AU)
	(for <ak@suse.de>) (for <lord@sgi.com>) (for <linux-xfs@oss.sgi.com>)
	By note With Smtp ; Wed, 7 Mar 2001 10:11:50 +1100 
Received: from neilb by notabene.cse.unsw.edu.au with local (Exim 3.12 #1 (Debian))
	id 14aQcX-0002va-00; Wed, 07 Mar 2001 10:11:57 +1100
From:   Neil Brown <neilb@cse.unsw.edu.au>
To:     Andi Kleen <ak@suse.de>
Date:   Wed, 7 Mar 2001 10:11:56 +1100 (EST)
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Message-ID: <15013.28348.768980.329307@notabene.cse.unsw.edu.au>
Cc:     Steve Lord <lord@sgi.com>, linux-xfs@oss.sgi.com
Subject: Re: TAKE - reintroduce the nfs inode reference cache
In-Reply-To: message from Andi Kleen on Tuesday March 6
References: <ak@suse.de>
	<200103062056.f26Kuo106999@jen.americas.sgi.com>
	<20010306222356.A15312@gruyere.muc.suse.de>
X-Mailer: VM 6.72 under Emacs 20.7.2
X-face:	[Gw_3E*Gng}4rRrKRYotwlE?.2|**#s9D<ml'fY1Vw+@XfR[fRCsUoP?K6bt3YD\ui5Fh?f
	LONpR';(ql)VM_TQ/<l_^D3~B:z$\YC7gUCuC=sYm/80G=$tt"98mr8(l))QzVKCk$6~gldn~*FK9x
	8`;pM{3S8679sP+MbP,72<3_PIH-$I&iaiIb|hV1d%cYg))BmI)AZ
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing

On Tuesday March 6, ak@suse.de wrote:
> 
> I think it would be best to fix nfsd to cache file handles instead of dentries
> and only do release when the nfsfh leaves the nfsfh cache.
> This way could also help ext2, which caches readahead context in struct file.
> 
> Put Neil Brown into cc. Neil, do you see any obstacles with that?
> 
> 
> -Andi

Firstly, lets check that I understand the situation.

 You find that under high memory pressure dentries for files that are
 currently being used by an NFS client get pruned from the dcache (and
 so the associated inodes leave the icache) resulting in slower
 filehandle-to-dentry lookup times when the next access is made.

 You are suggesting that some sort of filehandle cache go into nfsd to
 hold onto these denties a bit longer so that on the next lookup they
 are more likely to still be there.

 Is that right?

I had hoped that the normal dcache mechanisms would have held onto
them long enough, but I can see that under high memory pressure this
may not be long enough.
It might be enough just to add:

   if (!IS_ERR(result) result->d_flags |= DCACHE_REFERENCED;

after the first call to nfsd_iget in find_fh_dentry.
This would give the dentry an extra turn around the lru list.
If that isn't enough, I think that the best fix would be to modify
prune_dcache to be a bit more generous.
Something like:

add

#define DCACHE_DELAY 0x0010  /* recently used, don't discard for 20ms */

to dache.h, then change prune_dcache to add

	    if ((dentry->d_flags & DCACHE_DELAY) && 
	           time_before(jiffies, dentry->d_reftime+(HZ-1)/50+1)) {
		    list_add(&dentry->d_lru, &dentry_unused);
		    goto next;
            }

Then whenever nfsd lets go of a dentry after an operation that would
likely be followed by some other operations on the same dentry, the
DCACHE_DELAY bit would get set.

Ofcourse you would have to try to get that change passed Linux and
AlViro, but they might go for it, especially if you can point to
performance numbers.  The "20ms" is a bit icky but I don't know what
else to do about that.

You could ofcourse do all this inside knfsd, or even inside the
individual filesystem, but it makes the code a lot more complex.  You
need to make sure you flush the cache in a timely fashion, and let go
of denties that have been unlinked and all that stuff.
If we get the dcache to do the delay, then it is easy as it already
copes with those other things.

NeilBrown

From owner-linux-xfs@oss.sgi.com Tue Mar  6 15:26:06 2001
Received:  by oss.sgi.com id <S553935AbRCFXZ5>;
	Tue, 6 Mar 2001 15:25:57 -0800
Received: from sgi.SGI.COM ([192.48.153.1]:29716 "EHLO sgi.com")
	by oss.sgi.com with ESMTP id <S553930AbRCFXZn>;
	Tue, 6 Mar 2001 15:25:43 -0800
Received: from zeus-fddi.americas.sgi.com (128-162-8-103.americas.sgi.com [128.162.8.103]) 
	by sgi.com (980327.SGI.8.8.8-aspam/980304.SGI-aspam:
       SGI does not authorize the use of its proprietary
       systems or networks for unsolicited or bulk email
       from the Internet.) 
	via ESMTP id PAA07160
	for <linux-xfs@oss.sgi.com>; Tue, 6 Mar 2001 15:25:39 -0800 (PST)
	mail_from (lord@sgi.com)
Received: from daisy-e185.americas.sgi.com (daisy-e194.americas.sgi.com [128.162.194.214]) by zeus-fddi.americas.sgi.com (8.9.3/americas-smart-nospam1.1) with ESMTP id RAA1041802; Tue, 6 Mar 2001 17:24:21 -0600 (CST)
Received: from jen.americas.sgi.com (IDENT:root@jen.americas.sgi.com [128.162.184.86]) by daisy-e185.americas.sgi.com (SGI-8.9.3/SGI-server-1.7) with ESMTP id RAA06973; Tue, 6 Mar 2001 17:24:20 -0600 (CST)
Received: from jen.americas.sgi.com by jen.americas.sgi.com (8.11.0/SGI-client-1.7) via ESMTP id f26NQKc07294; Tue, 6 Mar 2001 17:26:20 -0600
Message-Id: <200103062326.f26NQKc07294@jen.americas.sgi.com>
X-Mailer: exmh version 2.2 06/23/2000 with nmh-1.0.4
To:     Neil Brown <neilb@cse.unsw.edu.au>
cc:     Andi Kleen <ak@suse.de>, Steve Lord <lord@sgi.com>,
        linux-xfs@oss.sgi.com
Subject: Re: TAKE - reintroduce the nfs inode reference cache 
In-Reply-To: Message from Neil Brown <neilb@cse.unsw.edu.au> 
   of "Wed, 07 Mar 2001 10:11:56 +1100." <15013.28348.768980.329307@notabene.cse.unsw.edu.au> 
Date:   Tue, 06 Mar 2001 17:26:20 -0600
From:   Steve Lord <lord@sgi.com>
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing


Hi,

What we were actually getting bitten by was much worse than this, but 
purely an xfs issue. We have an f_ops->release method which is getting
called out of nfsd_close, it does some operations more in line with a
close system call than every write to the file. I made changes within the
xfs code to avoid this issue and our performance went way up. So I cannot
say we have a scenario where the dcache entry is getting pruned out from 
under us all the time.

Steve

> On Tuesday March 6, ak@suse.de wrote:
> > 
> > I think it would be best to fix nfsd to cache file handles instead of dentr
> ies
> > and only do release when the nfsfh leaves the nfsfh cache.
> > This way could also help ext2, which caches readahead context in struct fil
> e.
> > 
> > Put Neil Brown into cc. Neil, do you see any obstacles with that?
> > 
> > 
> > -Andi
> 
> Firstly, lets check that I understand the situation.
> 
>  You find that under high memory pressure dentries for files that are
>  currently being used by an NFS client get pruned from the dcache (and
>  so the associated inodes leave the icache) resulting in slower
>  filehandle-to-dentry lookup times when the next access is made.
> 
>  You are suggesting that some sort of filehandle cache go into nfsd to
>  hold onto these denties a bit longer so that on the next lookup they
>  are more likely to still be there.
> 
>  Is that right?
> 
> I had hoped that the normal dcache mechanisms would have held onto
> them long enough, but I can see that under high memory pressure this
> may not be long enough.
> It might be enough just to add:
> 
>    if (!IS_ERR(result) result->d_flags |= DCACHE_REFERENCED;
> 
> after the first call to nfsd_iget in find_fh_dentry.
> This would give the dentry an extra turn around the lru list.
> If that isn't enough, I think that the best fix would be to modify
> prune_dcache to be a bit more generous.
> Something like:
> 
> add
> 
> #define DCACHE_DELAY 0x0010  /* recently used, don't discard for 20ms */
> 
> to dache.h, then change prune_dcache to add
> 
> 	    if ((dentry->d_flags & DCACHE_DELAY) && 
> 	           time_before(jiffies, dentry->d_reftime+(HZ-1)/50+1)) {
> 		    list_add(&dentry->d_lru, &dentry_unused);
> 		    goto next;
>             }
> 
> Then whenever nfsd lets go of a dentry after an operation that would
> likely be followed by some other operations on the same dentry, the
> DCACHE_DELAY bit would get set.
> 
> Ofcourse you would have to try to get that change passed Linux and
> AlViro, but they might go for it, especially if you can point to
> performance numbers.  The "20ms" is a bit icky but I don't know what
> else to do about that.
> 
> You could ofcourse do all this inside knfsd, or even inside the
> individual filesystem, but it makes the code a lot more complex.  You
> need to make sure you flush the cache in a timely fashion, and let go
> of denties that have been unlinked and all that stuff.
> If we get the dcache to do the delay, then it is easy as it already
> copes with those other things.
> 
> NeilBrown



From owner-linux-xfs@oss.sgi.com Tue Mar  6 15:47:47 2001
Received:  by oss.sgi.com id <S553935AbRCFXr1>;
	Tue, 6 Mar 2001 15:47:27 -0800
Received: from oboe.it.uc3m.es ([163.117.139.101]:27140 "EHLO oboe.it.uc3m.es")
	by oss.sgi.com with ESMTP id <S553930AbRCFXrK>;
	Tue, 6 Mar 2001 15:47:10 -0800
Received: (from ptb@localhost)
	by oboe.it.uc3m.es (8.11.0/8.11.0) id f26Nl6T05798
	for linux-xfs@oss.sgi.com; Wed, 7 Mar 2001 00:47:06 +0100
From:   "Peter T. Breuer" <ptb@it.uc3m.es>
Message-Id: <200103062347.f26Nl6T05798@oboe.it.uc3m.es>
Subject: Re: Still module problems
To:     linux-xfs@oss.sgi.com
Date:   Wed, 7 Mar 2001 00:47:06 +0100 (MET)
X-Anonymously-To:                                 
Reply-To: ptb@it.uc3m.es
X-Mailer: ELM [version 2.4ME+ PL66 (25)]
MIME-Version: 1.0
Content-Type: text/plain; charset=US-ASCII
Content-Transfer-Encoding: 7bit
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing

"GCS wrote:"
> If someone has Debian, please contact me personally. I do not know why, but
> I still have unresolved symbols in all modules. I have deleted the kernel

I'm running on debian, but with the january patch against 2.4.0.

I would appreciate it if the patch were less invasive. Has this been
done?  Surely changing make_request's type is intrinsically a bad idea?
I agree that the implementation is clever - setting the bh argument null
signals the new make_request to use the remaining args, kiobuff and
friends, instead, so only kiobuff-aware code will make use of it.  But
in that case why not write a separate function?  The kernel has
provisions for overwriting make_request _cleanly_.

And surely kiobuff + sector + count should be passed as a single
struct?

What are the issues with respect to buffer locking that I see vaguely
referred to in the faq? I am running xfs atop raid1 atop a block
driver that talks to the net, and a local partition or logical volume. I
was seeing lockups when the net driver ran against localhost (i.e. like
a loopback device).

Peter


From owner-linux-xfs@oss.sgi.com Tue Mar  6 15:55:07 2001
Received:  by oss.sgi.com id <S553935AbRCFXy5>;
	Tue, 6 Mar 2001 15:54:57 -0800
Received: from perninha.conectiva.com.br ([200.250.58.156]:35088 "HELO
        postfix.conectiva.com.br") by oss.sgi.com with SMTP
	id <S553930AbRCFXyt>; Tue, 6 Mar 2001 15:54:49 -0800
Received: from freak.distro.conectiva (freak.distro.conectiva [10.0.17.22])
	by postfix.conectiva.com.br (Postfix) with ESMTP
	id AE43216B0E; Tue,  6 Mar 2001 20:54:39 -0300 (EST)
Date:   Tue, 6 Mar 2001 19:09:29 -0300 (BRT)
From:   Marcelo Tosatti <marcelo@conectiva.com.br>
X-Sender: marcelo@freak.distro.conectiva
To:     linux-xfs@oss.sgi.com
Cc:     Steve Lord <lord@sgi.com>,
        Rajagopal Ananthanarayanan <ananth@sgi.com>
Subject: New allocation flag for non critical allocations
Message-ID: <Pine.LNX.4.21.0103061900220.852-100000@freak.distro.conectiva>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing


Hi, 

I'm attaching a patch to add a new allocation flag which can be used by
non critical allocations.

The kernel memory allocator will fail relatively easy if this flag is set
for the current allocation. 

This is very useful for allocators which are trying to optimize things by
allocating memory, but want to fail in case memory is short. (e.g. IO
clustering)

The patch also changes pagebuf's writepage() code to use this new flag
when trying to allocate page pointers to do IO clustering. 

Comments are welcome. 



Index: mm/page_alloc.c
===================================================================
RCS file: /cvs/linux-2.4-xfs/linux/mm/page_alloc.c,v
retrieving revision 1.38
diff -u -r1.38 page_alloc.c
--- mm/page_alloc.c	2001/02/27 00:36:49	1.38
+++ mm/page_alloc.c	2001/03/06 23:29:05
@@ -398,6 +398,10 @@
 	if (page)
 		return page;
 
+	 /* For non critical allocations we fail previously. */
+	if (gfp_mask & __GFP_FAIL)
+		return NULL;
+
 	/*
 	 * Damn, we didn't succeed.
 	 *
Index: include/linux/mm.h
===================================================================
RCS file: /cvs/linux-2.4-xfs/linux/include/linux/mm.h,v
retrieving revision 1.50
diff -u -r1.50 mm.h
--- include/linux/mm.h	2001/02/27 00:36:49	1.50
+++ include/linux/mm.h	2001/03/06 23:29:33
@@ -468,7 +468,9 @@
 #endif
 #define __GFP_VM	0x20
 #define __GFP_PAGE_IO	0x40
+#define __GFP_FAIL	0x80
 
+#define GFP_FAIL	(__GFP_HIGH | __GFP_WAIT | __GFP_FAIL)
 #define GFP_PAGE_IO	(__GFP_HIGH | __GFP_WAIT | __GFP_PAGE_IO)
 #define GFP_BUFFER	(__GFP_HIGH | __GFP_WAIT)
 #define GFP_ATOMIC	(__GFP_HIGH)
Index: fs/pagebuf/page_buf_io.c
===================================================================
RCS file: /cvs/linux-2.4-xfs/linux/fs/pagebuf/page_buf_io.c,v
retrieving revision 1.64
diff -u -r1.64 page_buf_io.c
--- fs/pagebuf/page_buf_io.c	2001/03/05 20:58:46	1.64
+++ fs/pagebuf/page_buf_io.c	2001/03/06 23:29:51
@@ -989,7 +989,7 @@
 	current->flags |= PF_MEMALLOC;
 #if defined(KIOCLUSTER)
 	cpages = kmalloc(CLUSTER_PAGE_LIST_SIZE * sizeof(struct page *),
-			GFP_PAGE_IO);
+			GFP_FAIL);
 #endif
 	do_write_full_page++;
 
Index: include/linux/slab.h
===================================================================
RCS file: /cvs/linux-2.4-xfs/linux/include/linux/slab.h,v
retrieving revision 1.13
diff -u -r1.13 slab.h
--- include/linux/slab.h	2001/02/22 21:29:07	1.13
+++ include/linux/slab.h	2001/03/06 23:30:34
@@ -22,8 +22,10 @@
 #define	SLAB_NFS		GFP_NFS
 #define	SLAB_DMA		GFP_DMA
 #define SLAB_PAGE_IO		GFP_PAGE_IO
+#define SLAB_FAIL		GFP_FAIL
 
-#define SLAB_LEVEL_MASK		(__GFP_WAIT|__GFP_HIGH|__GFP_IO|__GFP_PAGE_IO)
+#define SLAB_LEVEL_MASK		(__GFP_WAIT|__GFP_HIGH|__GFP_IO|__GFP_PAGE_IO|__GFP_FAIL)
+
 #define	SLAB_NO_GROW		0x00001000UL	/* don't grow a cache */
 
 /* flags to pass to kmem_cache_create().



From owner-linux-xfs@oss.sgi.com Tue Mar  6 16:16:47 2001
Received:  by oss.sgi.com id <S553935AbRCGAQh>;
	Tue, 6 Mar 2001 16:16:37 -0800
Received: from deliverator.sgi.com ([204.94.214.10]:19753 "EHLO
        deliverator.sgi.com") by oss.sgi.com with ESMTP id <S553930AbRCGAQW>;
	Tue, 6 Mar 2001 16:16:22 -0800
Received: from cthulhu.engr.sgi.com (gate3-relay.engr.sgi.com [130.62.1.234]) by deliverator.sgi.com (980309.SGI.8.8.8-aspam-6.2/980310.SGI-aspam) via ESMTP id QAA27858
	for <linux-xfs@oss.sgi.com>; Tue, 6 Mar 2001 16:15:15 -0800 (PST)
	mail_from (chait@getafix.engr.sgi.com)
Received: from getafix.engr.sgi.com (IDENT:root@getafix.engr.sgi.com [163.154.5.110])
	by cthulhu.engr.sgi.com (SGI-8.9.3/8.9.3) with ESMTP id QAA27802;
	Tue, 6 Mar 2001 16:15:06 -0800 (PST)
Received: from localhost (chait@localhost)
	by getafix.engr.sgi.com (8.9.3/8.9.3) with ESMTP id QAA01288;
	Tue, 6 Mar 2001 16:10:57 -0500
Date:   Tue, 6 Mar 2001 16:10:57 -0500 (EST)
From:   Chaitanya Tumuluri <chait@getafix.engr.sgi.com>
To:     "Peter T. Breuer" <ptb@it.uc3m.es>
cc:     linux-xfs@oss.sgi.com
Subject: Re: Still module problems
In-Reply-To: <200103062347.f26Nl6T05798@oboe.it.uc3m.es>
Message-ID: <Pine.LNX.4.21.0103061552590.649-100000@getafix.engr.sgi.com>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing

On Wed, 7 Mar 2001, Peter T. Breuer wrote:
> "GCS wrote:"
> I would appreciate it if the patch were less invasive. Has this been
> done?  Surely changing make_request's type is intrinsically a bad idea?
>
> I agree that the implementation is clever - setting the bh argument null
> signals the new make_request to use the remaining args, kiobuff and
> friends, instead, so only kiobuff-aware code will make use of it.  But
> in that case why not write a separate function?  The kernel has
> provisions for overwriting make_request _cleanly_.

Agreed. The current state of the code is a historical artifact of trying
to keep pace with the evolutions in ll_rw_block/__make_request() and
friends. Originally, we did have a separate make_kio_request() function.
The provision you mention came in later, at which point in time it was going 
to be a _lot_ more effort to provide for lvm/md type of remapping in the
make_kio_request() version. A quicker/easier fix was to modify the 
make_request() signature to handle kiobufs.

> And surely kiobuff + sector + count should be passed as a single
> struct?

Indeed. And that is in the works. Jens Axboe and Christoph Hellwig are fine 
tuning their new I/O container which can handle both sector-based (a la 
buffer-heads) and page-based (a la kiobufs) dma area info. So, as far
as the block-queueing layer is concerned it'll only have to deal with this
new I/O container. This _should_, in theory, obviate the need for a separate
buffer-head/kiobuf make_request() function altogether.

Perhaps they can provide more info....

> What are the issues with respect to buffer locking that I see vaguely
> referred to in the faq? I am running xfs atop raid1 atop a block

Could you be more specific (wrt. "vaguely referred to in the faq")?

> driver that talks to the net, and a local partition or logical volume. I
> was seeing lockups when the net driver ran against localhost (i.e. like
> a loopback device).

Hmmm....this may be a loopback device codepath issue. I'm not sure. Perhaps
someone else can comment on this..

Cheers,
-Chait.



From owner-linux-xfs@oss.sgi.com Tue Mar  6 16:34:47 2001
Received:  by oss.sgi.com id <S553935AbRCGAeh>;
	Tue, 6 Mar 2001 16:34:37 -0800
Received: from deliverator.sgi.com ([204.94.214.10]:19245 "EHLO
        deliverator.sgi.com") by oss.sgi.com with ESMTP id <S553930AbRCGAeS>;
	Tue, 6 Mar 2001 16:34:18 -0800
Received: from snort.melbourne.sgi.com (snort.melbourne.sgi.com [134.14.55.149]) by deliverator.sgi.com (980309.SGI.8.8.8-aspam-6.2/980310.SGI-aspam) via ESMTP id QAA29902
	for <linux-xfs@oss.sgi.com>; Tue, 6 Mar 2001 16:33:10 -0800 (PST)
	mail_from (nathans@snort.melbourne.sgi.com)
Received: (from nathans@localhost)
	by snort.melbourne.sgi.com (SGI-8.9.3/8.9.3) id LAA28740
	for linux-xfs@oss.sgi.com; Wed, 7 Mar 2001 11:32:59 +1100 (EST)
Date:   Wed, 7 Mar 2001 11:32:59 +1100 (EST)
From:   Nathan Scott <nathans@snort.melbourne.sgi.com>
Message-Id: <200103070032.LAA28740@snort.melbourne.sgi.com>
To:     linux-xfs@oss.sgi.com
Subject: TAKE - quota
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing

Date:  Tue Mar  6 16:32:11 PST 2001
Workarea:  snort.melbourne.sgi.com:/diskb/build4/nathans/linux-xfs

The following file(s) were checked into:
  bonnie.engr.sgi.com:/isms/slinx/2.4.x-xfs


Modid:  2.4.x-xfs:slinx:89118a
cmd/quota/hasquota.c - 1.6
	- ensure fstab entries for xfs filesystems with UUID=/LABEL=
	  device fields are correctly translated to device names before
	  attempting a quotactl.



From owner-linux-xfs@oss.sgi.com Tue Mar  6 18:17:39 2001
Received:  by oss.sgi.com id <S553935AbRCGCR3>;
	Tue, 6 Mar 2001 18:17:29 -0800
Received: from h00059aa0e40d.ne.mediaone.net ([66.31.89.164]:38653 "EHLO
        flowers.house.larsshack.org") by oss.sgi.com with ESMTP
	id <S553930AbRCGCRA>; Tue, 6 Mar 2001 18:17:00 -0800
Received: from localhost (lars@localhost)
	by flowers.house.larsshack.org (8.11.0/8.11.0) with ESMTP id f272GbT01326
	for <linux-xfs@oss.sgi.com>; Tue, 6 Mar 2001 21:16:37 -0500
X-Authentication-Warning: flowers.house.larsshack.org: lars owned process doing -bs
Date:   Tue, 6 Mar 2001 21:16:37 -0500 (EST)
From:   Lars Kellogg-Stedman <lars@larsshack.org>
X-Sender:  <lars@flowers.house.larsshack.org>
To:     <linux-xfs@oss.sgi.com>
Subject: More raid5 issues: Kernel panic running xfs_db
In-Reply-To: <200103070032.LAA28740@snort.melbourne.sgi.com>
Message-ID: <Pine.LNX.4.30.0103062020020.1277-100000@flowers.house.larsshack.org>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing

I got a kernel panic trying to run xfs_admin earlier today to label some
filesytems.  Further experiments revealed:

  (a) The panic is apparently caused when xfs_admin calls xfs_db, and
  (b) The panic only occurs on the soft raid5 volume.  Other filesystems
      presented no problems.

The kernel panic looked like:

  Mar  6 20:08:40 flowers kernel: Kernel panic: Invalid blocksize passed to set_blocksize

Which looks like it's coming from fs/buffer.c.

The xfs device that triggered this was /dev/md/0, which is a soft raid5
volume over 4 scsi disks.

-- Lars

-- 
Lars Kellogg-Stedman <lars@larsshack.org> --> http://www.larsshack.org/



From owner-linux-xfs@oss.sgi.com Tue Mar  6 18:45:09 2001
Received:  by oss.sgi.com id <S553936AbRCGCou>;
	Tue, 6 Mar 2001 18:44:50 -0800
Received: from pneumatic-tube.sgi.com ([204.94.214.22]:7994 "EHLO
        pneumatic-tube.sgi.com") by oss.sgi.com with ESMTP
	id <S553930AbRCGCoY>; Tue, 6 Mar 2001 18:44:24 -0800
Received: from larry.melbourne.sgi.com (larry.melbourne.sgi.com [134.14.52.130]) by pneumatic-tube.sgi.com (980327.SGI.8.8.8-aspam/980310.SGI-aspam) via SMTP id SAA01902
	for <linux-xfs@oss.sgi.com>; Tue, 6 Mar 2001 18:54:05 -0800 (PST)
	mail_from (nathans@wobbly.melbourne.sgi.com)
Received: from wobbly.melbourne.sgi.com (wobbly.melbourne.sgi.com [134.14.55.135]) by larry.melbourne.sgi.com (950413.SGI.8.6.12/950213.SGI.AUTOCF) via ESMTP id NAA24596; Wed, 7 Mar 2001 13:43:04 +1100
Received: (from nathans@localhost) by wobbly.melbourne.sgi.com (980427.SGI.8.8.8/980728.SGI.AUTOCF) id NAA77486; Wed, 7 Mar 2001 13:43:04 +1100 (EDT)
From:   "Nathan Scott" <nathans@wobbly.melbourne.sgi.com>
Message-Id: <10103071343.ZM191493@wobbly.melbourne.sgi.com>
Date:   Wed, 7 Mar 2001 13:43:03 -0400
In-Reply-To: Lars Kellogg-Stedman <lars@larsshack.org>
        "More raid5 issues: Kernel panic running xfs_db" (Mar  6,  9:16pm)
References: <Pine.LNX.4.30.0103062020020.1277-100000@flowers.house.larsshack.org>
X-Mailer: Z-Mail (3.2.3 08feb96 MediaMail)
To:     Lars Kellogg-Stedman <lars@larsshack.org>, <linux-xfs@oss.sgi.com>
Subject: Re: More raid5 issues: Kernel panic running xfs_db
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing

hi Lars,

On Mar 6,  9:16pm, Lars Kellogg-Stedman wrote:
> Subject: More raid5 issues: Kernel panic running xfs_db
> I got a kernel panic trying to run xfs_admin earlier today to label some
> filesytems.  Further experiments revealed:
> 
>   (a) The panic is apparently caused when xfs_admin calls xfs_db, and
>   (b) The panic only occurs on the soft raid5 volume.  Other filesystems
>       presented no problems.
> 
> The kernel panic looked like:
> 
>   Mar  6 20:08:40 flowers kernel: Kernel panic: Invalid blocksize passed to set_blocksize
> 
> Which looks like it's coming from fs/buffer.c.
> 
> The xfs device that triggered this was /dev/md/0, which is a soft raid5
> volume over 4 scsi disks.
> 

Were you were running xfs_db (via xfs_admin script) on a
mounted filesystem?  That shouldn't work at all - would be a
bug in xfs_db if it tries to write to a filesystem while its
mounted.

Hmm - just tried it - seems to do the right thing...

# sh -x xfs_admin -L test /dev/hdb10
+ OPTS=
+ USAGE=Usage: xfs_admin [-flu] [-L label] [-U uuid] special
+ getopts fluL:U: c
+ OPTS= -c 'label test'
+ getopts fluL:U: c
+ set -- extra -L test /dev/hdb10
+ shift 3
+ eval xfs_db -x -p xfs_admin -c ''\''label' 'test'\''' /dev/hdb10
++ xfs_db -x -p xfs_admin -c 'label test' /dev/hdb10
xfs_admin: /dev/hdb10 contains a mounted filesystem

fatal error -- couldn't initialize XFS library
+ status=1
+ exit 1

So I guess you're not using a mounted filesystem - in which
case, xfs_db is doing writes to the raw device, not via xfs.
Perhaps the use of xfs_db/xfs_admin is a red herring - do
you see the panic while the filesystem is mounted?  (after
you've run xfs_db & mount).  Are you using Russell/Martin's
md patch?

cheers.

-- 
Nathan

From owner-linux-xfs@oss.sgi.com Tue Mar  6 18:56:09 2001
Received:  by oss.sgi.com id <S553935AbRCGCzt>;
	Tue, 6 Mar 2001 18:55:49 -0800
Received: from h00059aa0e40d.ne.mediaone.net ([66.31.89.164]:56317 "EHLO
        flowers.house.larsshack.org") by oss.sgi.com with ESMTP
	id <S553930AbRCGCzX>; Tue, 6 Mar 2001 18:55:23 -0800
Received: from localhost (lars@localhost)
	by flowers.house.larsshack.org (8.11.0/8.11.0) with ESMTP id f272tAq01335
	for <linux-xfs@oss.sgi.com>; Tue, 6 Mar 2001 21:55:10 -0500
X-Authentication-Warning: flowers.house.larsshack.org: lars owned process doing -bs
Date:   Tue, 6 Mar 2001 21:55:10 -0500 (EST)
From:   Lars Kellogg-Stedman <lars@larsshack.org>
X-Sender:  <lars@flowers.house.larsshack.org>
To:     <linux-xfs@oss.sgi.com>
Subject: Re: More raid5 issues: Kernel panic running xfs_db (fwd)
Message-ID: <Pine.LNX.4.30.0103062154450.1276-100000@flowers.house.larsshack.org>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing

(I meant to cc this to the list.)

On Wed, 7 Mar 2001, Nathan Scott wrote:

> Were you were running xfs_db (via xfs_admin script) on a
> mounted filesystem?

As one would expect -- and as you surmised -- that doesn't work.

> Perhaps the use of xfs_db/xfs_admin is a red herring - do
> you see the panic while the filesystem is mounted?

No, the panic only occurs when running xfs_db (or, in fact, xfs_logprint,
which makes me suspect other xfs command may exhibit the same problem).
The panic only occurs when the raid5 volume is involved.

> Are you using Russell/Martin's md patch?

Possibly.  I'm using a patch Scott (Smyth) sent me, which I presume
incorporates this work -- the patch fixed the problem with the raid5 resync
process wedging when the volume was mounted.

-- Lars

-- 
Lars Kellogg-Stedman <lars@larsshack.org> --> http://www.larsshack.org/



From owner-linux-xfs@oss.sgi.com Tue Mar  6 19:09:09 2001
Received:  by oss.sgi.com id <S553930AbRCGDI7>;
	Tue, 6 Mar 2001 19:08:59 -0800
Received: from sgi.SGI.COM ([192.48.153.1]:34156 "EHLO sgi.com")
	by oss.sgi.com with ESMTP id <S553934AbRCGDIh>;
	Tue, 6 Mar 2001 19:08:37 -0800
Received: from larry.melbourne.sgi.com ([134.14.52.130]) 
	by sgi.com (980327.SGI.8.8.8-aspam/980304.SGI-aspam:
       SGI does not authorize the use of its proprietary
       systems or networks for unsolicited or bulk email
       from the Internet.) 
	via SMTP id TAA08489
	for <linux-xfs@oss.sgi.com>; Tue, 6 Mar 2001 19:08:37 -0800 (PST)
	mail_from (nathans@wobbly.melbourne.sgi.com)
Received: from wobbly.melbourne.sgi.com (wobbly.melbourne.sgi.com [134.14.55.135]) by larry.melbourne.sgi.com (950413.SGI.8.6.12/950213.SGI.AUTOCF) via ESMTP id OAA24753; Wed, 7 Mar 2001 14:07:18 +1100
Received: (from nathans@localhost) by wobbly.melbourne.sgi.com (980427.SGI.8.8.8/980728.SGI.AUTOCF) id OAA94700; Wed, 7 Mar 2001 14:07:17 +1100 (EDT)
From:   "Nathan Scott" <nathans@wobbly.melbourne.sgi.com>
Message-Id: <10103071407.ZM195249@wobbly.melbourne.sgi.com>
Date:   Wed, 7 Mar 2001 14:07:16 -0400
In-Reply-To: Lars Kellogg-Stedman <lars@larsshack.org>
        "Re: More raid5 issues: Kernel panic running xfs_db (fwd)" (Mar  6,  9:55pm)
References: <Pine.LNX.4.30.0103062154450.1276-100000@flowers.house.larsshack.org>
X-Mailer: Z-Mail (3.2.3 08feb96 MediaMail)
To:     Lars Kellogg-Stedman <lars@larsshack.org>,
        "Martin K. Petersen" <mkp@linuxcare.com>
Subject: Re: More raid5 issues: Kernel panic running xfs_db (fwd)
Cc:     linux-xfs@oss.sgi.com
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing

On Mar 6,  9:55pm, Lars Kellogg-Stedman wrote:
> Subject: Re: More raid5 issues: Kernel panic running xfs_db (fwd)
> (I meant to cc this to the list.)
> 
> On Wed, 7 Mar 2001, Nathan Scott wrote:
> ...
> > Perhaps the use of xfs_db/xfs_admin is a red herring - do
> > you see the panic while the filesystem is mounted?
> 
> No,

OK, cool - thats an important data point, I think.

> the panic only occurs when running xfs_db (or, in fact, xfs_logprint,
> which makes me suspect other xfs command may exhibit the same problem).
> The panic only occurs when the raid5 volume is involved.
> 

The only other thing I can think of that may be implicated
here is the BLKSETSIZE ioctl, which all of the xfs tools
will use to try and set the device blksize to 512 before
they get started.  Other than that, its all reads/writes.

Martin, does this sound fishy to you?  Any idea if the use
of this ioctl is the cause here?

> > Are you using Russell/Martin's md patch?
> 
> Possibly.  I'm using a patch Scott (Smyth) sent me, which I presume
> incorporates this work -- the patch fixed the problem with the raid5 resync
> process wedging when the volume was mounted.
> 


cheers.

-- 
Nathan

From owner-linux-xfs@oss.sgi.com Tue Mar  6 19:27:09 2001
Received:  by oss.sgi.com id <S553937AbRCGD07>;
	Tue, 6 Mar 2001 19:26:59 -0800
Received: from deliverator.sgi.com ([204.94.214.10]:6226 "EHLO
        deliverator.sgi.com") by oss.sgi.com with ESMTP id <S553934AbRCGD0h>;
	Tue, 6 Mar 2001 19:26:37 -0800
Received: from larry.melbourne.sgi.com (larry.melbourne.sgi.com [134.14.52.130]) by deliverator.sgi.com (980309.SGI.8.8.8-aspam-6.2/980310.SGI-aspam) via SMTP id TAA18824
	for <linux-xfs@oss.sgi.com>; Tue, 6 Mar 2001 19:25:09 -0800 (PST)
	mail_from (nathans@wobbly.melbourne.sgi.com)
Received: from wobbly.melbourne.sgi.com (wobbly.melbourne.sgi.com [134.14.55.135]) by larry.melbourne.sgi.com (950413.SGI.8.6.12/950213.SGI.AUTOCF) via ESMTP id OAA24866; Wed, 7 Mar 2001 14:24:52 +1100
Received: (from nathans@localhost) by wobbly.melbourne.sgi.com (980427.SGI.8.8.8/980728.SGI.AUTOCF) id OAA77991; Wed, 7 Mar 2001 14:24:51 +1100 (EDT)
From:   "Nathan Scott" <nathans@wobbly.melbourne.sgi.com>
Message-Id: <10103071424.ZM185399@wobbly.melbourne.sgi.com>
Date:   Wed, 7 Mar 2001 14:24:50 -0400
In-Reply-To: Lars Kellogg-Stedman <lars@larsshack.org>
        "Re: More raid5 issues: Kernel panic running xfs_db (fwd)" (Mar  6,  9:55pm)
References: <Pine.LNX.4.30.0103062154450.1276-100000@flowers.house.larsshack.org>
X-Mailer: Z-Mail (3.2.3 08feb96 MediaMail)
To:     Lars Kellogg-Stedman <lars@larsshack.org>
Subject: Re: More raid5 issues: Kernel panic running xfs_db (fwd)
Cc:     linux-xfs@oss.sgi.com
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing

hi Lars,

On Mar 6,  9:55pm, Lars Kellogg-Stedman wrote:
> Subject: Re: More raid5 issues: Kernel panic running xfs_db (fwd)
> ... the panic only occurs when running xfs_db (or, in fact, xfs_logprint,
> which makes me suspect other xfs command may exhibit the same problem).
> The panic only occurs when the raid5 volume is involved.
> 

Just realized - you can easily test whether the BLKSETSIZE
ioctl is whats causing the problem by commenting out the
call at line ~206 in cmd/xfsprogs/libxfs/init.c, then
recompiling the library and all the tools, then use the
new tools to try and reproduce the problem.

cheers.

-- 
Nathan

From owner-linux-xfs@oss.sgi.com Wed Mar  7 00:45:51 2001
Received:  by oss.sgi.com id <S553844AbRCGIpm>;
	Wed, 7 Mar 2001 00:45:42 -0800
Received: from ns.suse.de ([213.95.15.193]:25607 "HELO Cantor.suse.de")
	by oss.sgi.com with SMTP id <S553840AbRCGIpY>;
	Wed, 7 Mar 2001 00:45:24 -0800
Received: from Hermes.suse.de (Hermes.suse.de [213.95.15.136])
	by Cantor.suse.de (Postfix) with ESMTP
	id 0A80D1E240; Wed,  7 Mar 2001 09:45:23 +0100 (MET)
Date:   Wed, 7 Mar 2001 09:45:17 +0100
From:   Andi Kleen <ak@suse.de>
To:     Neil Brown <neilb@cse.unsw.edu.au>
Cc:     Andi Kleen <ak@suse.de>, Steve Lord <lord@sgi.com>,
        linux-xfs@oss.sgi.com
Subject: Re: TAKE - reintroduce the nfs inode reference cache
Message-ID: <20010307094517.A23323@gruyere.muc.suse.de>
References: <ak@suse.de> <200103062056.f26Kuo106999@jen.americas.sgi.com> <20010306222356.A15312@gruyere.muc.suse.de> <15013.28348.768980.329307@notabene.cse.unsw.edu.au>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.2.5i
In-Reply-To: <15013.28348.768980.329307@notabene.cse.unsw.edu.au>; from neilb@cse.unsw.edu.au on Wed, Mar 07, 2001 at 10:11:56AM +1100
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing

On Wed, Mar 07, 2001 at 10:11:56AM +1100, Neil Brown wrote:
> On Tuesday March 6, ak@suse.de wrote:
> > 
> > I think it would be best to fix nfsd to cache file handles instead of dentries
> > and only do release when the nfsfh leaves the nfsfh cache.
> > This way could also help ext2, which caches readahead context in struct file.
> > 
> > Put Neil Brown into cc. Neil, do you see any obstacles with that?
> > 
> > 
> > -Andi
> 
> Firstly, lets check that I understand the situation.

[...]

Main problem is that struct file is not preserved over RPCs. Various
file systems put state into struct file (ext2 readahead context, XFS 
did discard preallocation in f_op->release, which is rather costly).
So I was looking for a way to preserve struct file across multiple RPCs.

[This is somewhat similar to another long standing NFS bug BTW -- 
it shares UDP sockets and there is excessive ARP traffic because the neighbour
state machine times out all the time and does ARP reprobes]


> You could ofcourse do all this inside knfsd, or even inside the
> individual filesystem, but it makes the code a lot more complex.  You
> need to make sure you flush the cache in a timely fashion, and let go
> of denties that have been unlinked and all that stuff.
> If we get the dcache to do the delay, then it is easy as it already
> copes with those other things.

Storing a struct file in the dcache could unfortunately get a bit tricky -
it's not a 1:1 mapping and it would require new fields and new destructors.
I guess it would be easier to put a new cache inbetween similar to what 2.2
nfsd did. 


-Andi

From owner-linux-xfs@oss.sgi.com Wed Mar  7 00:57:52 2001
Received:  by oss.sgi.com id <S553844AbRCGI5m>;
	Wed, 7 Mar 2001 00:57:42 -0800
Received: from groucho.maths.monash.edu.au ([130.194.160.211]:35635 "EHLO
        groucho.maths.monash.edu.au") by oss.sgi.com with ESMTP
	id <S553840AbRCGI5V>; Wed, 7 Mar 2001 00:57:21 -0800
Received: (from rjh@localhost)
	by groucho.maths.monash.edu.au (8.8.8/8.8.8) id IAA06201
	for linux-xfs@oss.sgi.com; Wed, 7 Mar 2001 08:57:18 GMT
From:   Robin Humble <rjh@groucho.maths.monash.edu.au>
Message-Id: <200103070857.IAA06201@groucho.maths.monash.edu.au>
Subject: Re: TAKE - nfs reference cache version 2
To:     linux-xfs@oss.sgi.com
Date:   Wed, 7 Mar 2001 19:57:17 +1100 (EDT)
In-Reply-To: <200103062108.f26L8f907105@jen.americas.sgi.com> from "Steve Lord" at Mar 06, 2001 03:08:41 PMMar 06, 2001 03:08:41 PM
X-Mailer: ELM [version 2.5 PL1]
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing


>OK, this actually should fix the performance of writes, I will think

Ta for the quick response! Quite a flurry of traffic today...

Yup - Initial tests indicate that you've sorted it. It's up to 7MB/s
on writes to a SMP celeron XFS+RAID0 box, and nfsd cpu loads are down
to roughly ext2 levels. cool

Later this week I'll try to re-run the exact same set of controlled
uni-processor non-RAID quiet network tests that I did before, but I
suspect you've eliminated the major bottleneck fer sure... :-)

Well done and thanks heaps!!

cheers,
robin

From owner-linux-xfs@oss.sgi.com Wed Mar  7 02:29:42 2001
Received:  by oss.sgi.com id <S553939AbRCGK3d>;
	Wed, 7 Mar 2001 02:29:33 -0800
Received: from oboe.it.uc3m.es ([163.117.139.101]:13578 "EHLO oboe.it.uc3m.es")
	by oss.sgi.com with ESMTP id <S553847AbRCGK3O>;
	Wed, 7 Mar 2001 02:29:14 -0800
Received: (from ptb@localhost)
	by oboe.it.uc3m.es (8.11.0/8.11.0) id f27ASuU16888;
	Wed, 7 Mar 2001 11:28:56 +0100
From:   "Peter T. Breuer" <ptb@it.uc3m.es>
Message-Id: <200103071028.f27ASuU16888@oboe.it.uc3m.es>
Subject: Re: More raid5 issues: Kernel panic running xfs_db (fwd)
In-Reply-To: <Pine.LNX.4.30.0103062154450.1276-100000@flowers.house.larsshack.org>
 from "Lars Kellogg-Stedman" at "Mar 6, 2001 09:55:10 pm"
To:     "Lars Kellogg-Stedman" <lars@larsshack.org>
Date:   Wed, 7 Mar 2001 11:28:56 +0100 (MET)
CC:     linux-xfs@oss.sgi.com
X-Anonymously-To:                                 
Reply-To: ptb@it.uc3m.es
X-Mailer: ELM [version 2.4ME+ PL66 (25)]
MIME-Version: 1.0
Content-Type: text/plain; charset=US-ASCII
Content-Transfer-Encoding: 7bit
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing

"A month of sundays ago Lars Kellogg-Stedman wrote:"
> > Are you using Russell/Martin's md patch?
> 
> Possibly.  I'm using a patch Scott (Smyth) sent me, which I presume
> incorporates this work -- the patch fixed the problem with the raid5 resync
> process wedging when the volume was mounted.

I think scott also gave me this patch.  I applied it over the Jan xfs
patch for 2.4.0.  It's a nice patch but the introduction of the
BLKSETSIZE ioctl in md.c causes instant lockups for me when mounting or
making an xfs over raid1.  Commenting those three lines fixes things for
me.  I surmise that parts of the code somewhere are not ready to deal
with a nonstandard blksize, so its better if nobody succeeds in setting
one.

The rest of the patch deals with fixing a purported bug in remove_from_lru
in buffer.c (there wasn't a bug, and the patch doesn't fix it, but it
shouldn't do any real harm either - the correct approach is to remove
the guards looking for null pointers since the list is circular and
doesn't ever have null pointers), and in raid5 it converts 
raid5_sync_request to using sector counts instead of block counts.
That's fine, but in one place it makes an additional change to sector
counts in a call to md_done_sync that looks unwarranted to me (in my xfs
snapshot) since md_done_sync uses blocks, not sectors:

-               md_done_sync(conf->mddev, (sh->size>>10) - sh->sync_redone,1);
+               md_done_sync(conf->mddev, (sh->size>>9) - sh->sync_redone, 1);

I'm not using raid5 so I can't comment, but that bit seemed offbase
to me when I read the patch.

  void md_done_sync(mddev_t *mddev, int blocks, int ok)
         // stuff that subtracts the blocks count from the 
         // to-be-recovered list

The patch does some other nice things. I didn't know its intention
was to unblock background resync ... I'll have a look to see where that
effect comes from.


Peter

From owner-linux-xfs@oss.sgi.com Wed Mar  7 04:21:02 2001
Received:  by oss.sgi.com id <S553665AbRCGMUn>;
	Wed, 7 Mar 2001 04:20:43 -0800
Received: from gate2-b.volkswagen.de ([193.23.96.7]:38283 "EHLO
        gate2-b.volkswagen.de") by oss.sgi.com with ESMTP
	id <S553655AbRCGMUk>; Wed, 7 Mar 2001 04:20:40 -0800
Received: (from smtpd@localhost)
	by gate2-b.volkswagen.de (8.9.1a/8.9.1) id NAA17214;
	Wed, 7 Mar 2001 13:20:37 +0100 (MET)
Received: from Volkswagen by relay2.volkswagen.de, id smtpdAAA4ft5R_; Wed Mar  7 13:20:20 2001
Received: by devwagwodx0008.wob.vw.de with Internet Mail Service (5.5.2653.19)
	id <F54WFN2H>; Wed, 7 Mar 2001 13:20:18 +0100
Message-ID: <96C14D00FA99D311A8D60008C791F3640982677C@DEVWAGWODX0009>
From:   "Pasch, Thomas (ACTGRO)" <extern.thomas.pasch@volkswagen.de>
Reply-To: pasch@netzGeneration.com
To:     "'linux-xfs@oss.sgi.com'" <linux-xfs@oss.sgi.com>
Cc:     "'netzGeneration'" <pasch@netzGeneration.com>
Subject: FAQ - Update and (new) question
Date:   Wed, 7 Mar 2001 13:16:26 +0100 
MIME-Version: 1.0
X-Mailer: Internet Mail Service (5.5.2653.19)
Content-Type: text/plain
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing

Hi,

Q: Does XFS run with the linux software RAID md driver? 
Nobody has tested it so far but expect it not to work like for all the other
journaling filesystems due to internal buffer locking problems I think. If
anyone tries this - please let me know. Also keep in mind - if you want to
try it - also the md driver will definitely not work with the kiobuf option
enabled. 

This is wrong. ReiserFS is doing this perfectly on my system for some months
now.
But you must use the raid implementation 0.9 (raidtools-0.9) to do. So the 
question comes up again: Does XFS run with the linux software RAID?

What about nfs? 

There are some problems with new filesystems and nfs in linux
(see discussion on http://lwn.net/2001/0222/kernel.php3
for details). So: is XFS supporting NFS (or better vice versa).

Best regards,

Thomas


From owner-linux-xfs@oss.sgi.com Wed Mar  7 04:21:13 2001
Received:  by oss.sgi.com id <S553669AbRCGMVC>;
	Wed, 7 Mar 2001 04:21:02 -0800
Received: from 66-2-81-26.customer.algx.net ([66.2.81.26]:1360 "EHLO
        wiley.ceo.com") by oss.sgi.com with ESMTP id <S553655AbRCGMUv>;
	Wed, 7 Mar 2001 04:20:51 -0800
Received: from mindspring.com (IDENT:danny@localhost [127.0.0.1])
	by wiley.ceo.com (8.9.3/8.9.3) with ESMTP id HAA00949;
	Wed, 7 Mar 2001 07:32:20 -0500
Message-ID: <3AA62A54.D5BB42F1@mindspring.com>
Date:   Wed, 07 Mar 2001 07:32:20 -0500
From:   Danny <danscox@mindspring.com>
Reply-To: dcox@connex.com
Organization: Connex Inc
X-Mailer: Mozilla 4.76 [en] (X11; U; Linux 2.4.2-XFS i686)
X-Accept-Language: en
MIME-Version: 1.0
To:     ptb@it.uc3m.es
CC:     linux-xfs@oss.sgi.com
Subject: Re: More raid5 issues: Kernel panic running xfs_db (fwd)
References: <200103071028.f27ASuU16888@oboe.it.uc3m.es>
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing

Peter,

"Peter T. Breuer" wrote:
> 
> "A month of sundays ago Lars Kellogg-Stedman wrote:"
> > > Are you using Russell/Martin's md patch?
> >
> > Possibly.  I'm using a patch Scott (Smyth) sent me, which I presume
> > incorporates this work -- the patch fixed the problem with the raid5 resync
> > process wedging when the volume was mounted.
> 
> doesn't ever have null pointers), and in raid5 it converts
> raid5_sync_request to using sector counts instead of block counts.
> That's fine, but in one place it makes an additional change to sector
> counts in a call to md_done_sync that looks unwarranted to me (in my xfs
> snapshot) since md_done_sync uses blocks, not sectors:
> 
> -               md_done_sync(conf->mddev, (sh->size>>10) - sh->sync_redone,1);
> +               md_done_sync(conf->mddev, (sh->size>>9) - sh->sync_redone, 1);
> 
> I'm not using raid5 so I can't comment, but that bit seemed offbase
> to me when I read the patch.
> 
>   void md_done_sync(mddev_t *mddev, int blocks, int ok)
>          // stuff that subtracts the blocks count from the
>          // to-be-recovered list

	Then you don't have the full patch.  The patch from Russell/Martin
changed md.c also to use sectors instead of block in all cases.  What
their patch didn't include was changes to raid1.c, which I stumbled
across when I first attempted to use it.  I've sent a patch to them for
raid1.c, and also a patch to make the BLKSETSIZE ioctl work, too, but
I've not heard anything from either of them on it.

-- 
"Men occasionally stumble over the truth, but most of them pick 
themselves up and hurry off as if nothing had happened." 
   -- Winston Churchill 

Danny

From owner-linux-xfs@oss.sgi.com Wed Mar  7 07:32:06 2001
Received:  by oss.sgi.com id <S553672AbRCGPb5>;
	Wed, 7 Mar 2001 07:31:57 -0800
Received: from lips.borg.umn.edu ([160.94.232.50]:29452 "EHLO
        lips.borg.umn.edu") by oss.sgi.com with ESMTP id <S553669AbRCGPbm>;
	Wed, 7 Mar 2001 07:31:42 -0800
Received: from thebarn.com (nic-31-c12-219.mn.mediaone.net [24.31.12.219])
	by lips.borg.umn.edu (8.11.2/8.10.1) with ESMTP id f27FVSb69281;
	Wed, 7 Mar 2001 09:31:29 -0600 (CST)
Message-ID: <3AA6544A.FB746BF0@thebarn.com>
Date:   Wed, 07 Mar 2001 09:31:23 -0600
From:   Russell Cattelan <cattelan@thebarn.com>
X-Mailer: Mozilla 4.74 [en] (X11; U; Linux 2.2.12 i386)
X-Accept-Language: en
MIME-Version: 1.0
To:     dcox@connex.com
CC:     ptb@it.uc3m.es, linux-xfs@oss.sgi.com
Subject: Re: More raid5 issues: Kernel panic running xfs_db (fwd)
References: <200103071028.f27ASuU16888@oboe.it.uc3m.es> <3AA62A54.D5BB42F1@mindspring.com>
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing

Danny wrote:

> Peter,
>
> "Peter T. Breuer" wrote:
> >
> > "A month of sundays ago Lars Kellogg-Stedman wrote:"
> > > > Are you using Russell/Martin's md patch?
> > >
> > > Possibly.  I'm using a patch Scott (Smyth) sent me, which I presume
> > > incorporates this work -- the patch fixed the problem with the raid5 resync
> > > process wedging when the volume was mounted.
> >
> > doesn't ever have null pointers), and in raid5 it converts
> > raid5_sync_request to using sector counts instead of block counts.
> > That's fine, but in one place it makes an additional change to sector
> > counts in a call to md_done_sync that looks unwarranted to me (in my xfs
> > snapshot) since md_done_sync uses blocks, not sectors:
> >
> > -               md_done_sync(conf->mddev, (sh->size>>10) - sh->sync_redone,1);
> > +               md_done_sync(conf->mddev, (sh->size>>9) - sh->sync_redone, 1);
> >
> > I'm not using raid5 so I can't comment, but that bit seemed offbase
> > to me when I read the patch.
> >
> >   void md_done_sync(mddev_t *mddev, int blocks, int ok)
> >          // stuff that subtracts the blocks count from the
> >          // to-be-recovered list
>
>         Then you don't have the full patch.  The patch from Russell/Martin
> changed md.c also to use sectors instead of block in all cases.  What
> their patch didn't include was changes to raid1.c, which I stumbled
> across when I first attempted to use it.  I've sent a patch to them for
> raid1.c, and also a patch to make the BLKSETSIZE ioctl work, too, but
> I've not heard anything from either of them on it.

Martin has promissed to clean up the patch and get it checked into the main tree by

the end of the week.

If not I'll make sure it gets there.

>
>
> --
> "Men occasionally stumble over the truth, but most of them pick
> themselves up and hurry off as if nothing had happened."
>    -- Winston Churchill
>
> Danny

--
Russell Cattelan
cattelan@thebarn.com




From owner-linux-xfs@oss.sgi.com Wed Mar  7 07:45:15 2001
Received:  by oss.sgi.com id <S553672AbRCGPpG>;
	Wed, 7 Mar 2001 07:45:06 -0800
Received: from tux.mkp.net ([130.225.60.11]:50440 "EHLO tux.mkp.net")
	by oss.sgi.com with ESMTP id <S553669AbRCGPov>;
	Wed, 7 Mar 2001 07:44:51 -0800
Received: from tux.mkp.net ([130.225.60.11] helo=jaguar.mkp.net)
	by tux.mkp.net with esmtp (Exim 3.16 #1)
	id 14ag7I-0005ER-00; Wed, 07 Mar 2001 16:44:46 +0100
Received: (from mkp@localhost)
	by jaguar.mkp.net (8.9.3/8.9.3) id KAA04304;
	Wed, 7 Mar 2001 10:43:42 -0500
X-Authentication-Warning: jaguar.mkp.net: mkp set sender to mkp@mkp.net using -f
To:     "Nathan Scott" <nathans@wobbly.melbourne.sgi.com>
Cc:     Lars Kellogg-Stedman <lars@larsshack.org>, linux-xfs@oss.sgi.com
Subject: Re: More raid5 issues: Kernel panic running xfs_db (fwd)
References: <Pine.LNX.4.30.0103062154450.1276-100000@flowers.house.larsshack.org>
	<10103071407.ZM195249@wobbly.melbourne.sgi.com>
From:   "Martin K. Petersen" <mkp@linuxcare.com>
Organization: Linuxcare, Inc.
Date:   07 Mar 2001 10:43:41 -0500
In-Reply-To: <10103071407.ZM195249@wobbly.melbourne.sgi.com>
Message-ID: <yq1snkp4maa.fsf@jaguar.mkp.net>
Lines:  27
User-Agent: Gnus/5.0808 (Gnus v5.8.8) XEmacs/21.1 (Canyonlands)
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing

>>>>> "Nathan" == Nathan Scott <nathans@wobbly.melbourne.sgi.com> writes:

>> the panic only occurs when running xfs_db (or, in fact,
>> xfs_logprint, which makes me suspect other xfs command may exhibit
>> the same problem).  The panic only occurs when the raid5 volume is
>> involved.

Nathan> The only other thing I can think of that may be implicated
Nathan> here is the BLKSETSIZE ioctl, which all of the xfs tools will
Nathan> use to try and set the device blksize to 512 before they get
Nathan> started.  Other than that, its all reads/writes.

Nathan> Martin, does this sound fishy to you?  Any idea if the use of
Nathan> this ioctl is the cause here?

This should be fixed if my RAID5 patch has been applied.


In any case we decided yesterday to include the temporary RAID
workarounds in the XFS tree since proper fixes will be a while.

I'll commit my patches later today.

-- 
Martin K. Petersen, Principal Linux Consultant, Linuxcare, Inc.
mkp@linuxcare.com, http://www.linuxcare.com/
SGI XFS for Linux Developer, http://oss.sgi.com/projects/xfs/

From owner-linux-xfs@oss.sgi.com Wed Mar  7 07:53:25 2001
Received:  by oss.sgi.com id <S553672AbRCGPxQ>;
	Wed, 7 Mar 2001 07:53:16 -0800
Received: from south.orl-pub.theseus.com ([12.108.42.66]:60444 "EHLO
        thor.theseus.com") by oss.sgi.com with ESMTP id <S553669AbRCGPxI>;
	Wed, 7 Mar 2001 07:53:08 -0800
Received: from theseus.com (IDENT:thebs@fugitive.theseus.com [192.168.0.242])
	by thor.theseus.com (8.9.3/8.9.3) with ESMTP id KAA14061;
	Wed, 7 Mar 2001 10:56:51 -0500
Message-ID: <3AA6598C.A1DA2FE9@theseus.com>
Date:   Wed, 07 Mar 2001 10:53:48 -0500
From:   Bryan-TheBS-Smith <thebs@theseus.com>
Reply-To: thebs@theseus.com, b.j.smith@ieee.org
Organization: (Personal)
X-Mailer: Mozilla 4.76 [en] (X11; U; Linux 2.2.17-0.17-FUGITIVE i686)
X-Accept-Language: en
MIME-Version: 1.0
To:     "Martin K. Petersen" <mkp@linuxcare.com>
CC:     Nathan Scott <nathans@wobbly.melbourne.sgi.com>,
        Lars Kellogg-Stedman <lars@larsshack.org>,
        linux-xfs@oss.sgi.com
Subject: Re: More raid5 issues: Kernel panic running xfs_db (fwd)
References: <Pine.LNX.4.30.0103062154450.1276-100000@flowers.house.larsshack.org>
		<10103071407.ZM195249@wobbly.melbourne.sgi.com> <yq1snkp4maa.fsf@jaguar.mkp.net>
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing

"Martin K. Petersen" wrote:
> In any case we decided yesterday to include the temporary RAID
> workarounds in the XFS tree since proper fixes will be a while.
> I'll commit my patches later today.

I guess I need to get "more involved" in the addition of patches. 
I'm currently trying to figure out what RedHat patches to merge with
the XFS CVS tree when creating a new batch of RPMs.

I take it I should get familiar with the "TAKE" stuff that comes
down the list?

-- TheBS

--
Bryan "TheBS" Smith          chat:thebs413 @AOL/MSN/Yahoo
Engineer      mailto:b.j.smith@ieee.org,thebs@theseus.com
*********************************************************
"Never apply a Star Trek solution to a Babylon 5 problem"
                                    -- Nicholas C. Weaver

From owner-linux-xfs@oss.sgi.com Wed Mar  7 07:54:36 2001
Received:  by oss.sgi.com id <S553672AbRCGPy0>;
	Wed, 7 Mar 2001 07:54:26 -0800
Received: from tux.mkp.net ([130.225.60.11]:57352 "EHLO tux.mkp.net")
	by oss.sgi.com with ESMTP id <S553669AbRCGPyQ>;
	Wed, 7 Mar 2001 07:54:16 -0800
Received: from tux.mkp.net ([130.225.60.11] helo=jaguar.mkp.net)
	by tux.mkp.net with esmtp (Exim 3.16 #1)
	id 14agGI-0005Er-00; Wed, 07 Mar 2001 16:54:03 +0100
Received: (from mkp@localhost)
	by jaguar.mkp.net (8.9.3/8.9.3) id KAA04310;
	Wed, 7 Mar 2001 10:52:54 -0500
X-Authentication-Warning: jaguar.mkp.net: mkp set sender to mkp@mkp.net using -f
To:     dcox@connex.com
Cc:     ptb@it.uc3m.es, linux-xfs@oss.sgi.com
Subject: Re: More raid5 issues: Kernel panic running xfs_db (fwd)
References: <200103071028.f27ASuU16888@oboe.it.uc3m.es>
	<3AA62A54.D5BB42F1@mindspring.com>
From:   "Martin K. Petersen" <mkp@linuxcare.com>
Organization: Linuxcare, Inc.
Date:   07 Mar 2001 10:52:53 -0500
In-Reply-To: <3AA62A54.D5BB42F1@mindspring.com>
Message-ID: <yq1k8614luy.fsf@jaguar.mkp.net>
Lines:  16
User-Agent: Gnus/5.0808 (Gnus v5.8.8) XEmacs/21.1 (Canyonlands)
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing

>>>>> "Danny" == Danny  <danscox@mindspring.com> writes:

Danny> What their patch didn't include was changes to raid1.c, which I
Danny> stumbled across when I first attempted to use it.  I've sent a
Danny> patch to them for raid1.c, and also a patch to make the
Danny> BLKSETSIZE ioctl work, too, but I've not heard anything from
Danny> either of them on it.

I have both of them in my to-be-committed RAID tree.

Thanks.

-- 
Martin K. Petersen, Principal Linux Consultant, Linuxcare, Inc.
mkp@linuxcare.com, http://www.linuxcare.com/
SGI XFS for Linux Developer, http://oss.sgi.com/projects/xfs/

From owner-linux-xfs@oss.sgi.com Wed Mar  7 09:35:36 2001
Received:  by oss.sgi.com id <S553698AbRCGRf1>;
	Wed, 7 Mar 2001 09:35:27 -0800
Received: from [141.154.95.3] ([141.154.95.3]:11001 "EHLO rain.ximian.com")
	by oss.sgi.com with ESMTP id <S553687AbRCGRfO>;
	Wed, 7 Mar 2001 09:35:14 -0800
Received: from localhost ([127.0.0.1])
	by rain.ximian.com with esmtp (Exim 3.12 #1 (Debian))
	id 14ahpz-0007hC-00
	for <linux-xfs@oss.sgi.com>; Wed, 07 Mar 2001 12:34:59 -0500
Subject: kernel panic with nfsd and xfs filesystem
From:   Vladimir Vukicevic <vladimir@ximian.com>
To:     linux-xfs@oss.sgi.com
Content-Type: text/plain
X-Mailer: Evolution (0.9 - Preview Release)
Date:   07 Mar 2001 12:34:49 -0500
Mime-Version: 1.0
Message-Id: <E14ahpz-0007hC-00@rain.ximian.com>
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing


I just upgraded the kernel to CVS as of Mar 7 12:10 EST on our nfs
server with a nice big xfs partition, and got this Oops from nfsd
whenever a client tries to mount (filtered through ksymoops):


ksymoops 2.3.4 on i686 2.4.2-XFS.  Options used
     -v /usr/src/linux/vmlinux (specified)
     -K (specified)
     -l /proc/modules (default)
     -o /lib/modules/2.4.2-XFS (specified)
     -m /usr/src/linux/System.map (specified)

No modules in ksyms, skipping objects
No ksyms, skipping lsmod
Reading Oops report from the terminal
Unable to handle kernel NULL pointer dereference at virtual address 00000000
 printing eip:
c0169d64
*pde = 00000000
Oops: 0000
CPU:    1
EIP:    0010:[<c0169d64>]
EFLAGS: 00010246
eax: 00000000   ebx: c6efb980   ecx: 00000000   edx: c7c27f20
esi: 00000000   edi: c02b901e   ebp: c751b000   esp: c6e75efc
ds: 0018   es: 0018   ss: 0018
Process nfsd (pid: 584, stackpage=c6e75000)
Stack: c6efb980 c6efb780 00000000 c751b000 c7484160 c7c27f20 c6e69f9c c6efb980 
       c7554800 c6efb9a4 c751b000 c016dc2d c531bb04 c0167d49 c751b000 c6efb980 
       00000000 00000000 c6efb780 c751b000 c0301de0 c6e70014 c0167633 c751b000 
Call Trace: [<c016dc2d>] [<c0167d49>] [<c0167633>] [<c0290b28>] [<c01673da>] [<c
0107503>] 

Code: ac ae 75 08 84 c0 75 f8 31 c0 eb 04 19 c0 0c 01 85 c0 75 3c 
Unable to handle kernel NULL pointer dereference at virtual address 00000000
c0169d64
*pde = 00000000
Oops: 0000
CPU:    1
EIP:    0010:[<c0169d64>]
Using defaults from ksymoops -t elf32-i386 -a i386
EFLAGS: 00010246
eax: 00000000   ebx: c6efb980   ecx: 00000000   edx: c7c27f20
esi: 00000000   edi: c02b901e   ebp: c751b000   esp: c6e75efc
ds: 0018   es: 0018   ss: 0018
Process nfsd (pid: 584, stackpage=c6e75000)
Stack: c6efb980 c6efb780 00000000 c751b000 c7484160 c7c27f20 c6e69f9c c6efb980 
       c7554800 c6efb9a4 c751b000 c016dc2d c531bb04 c0167d49 c751b000 c6efb980 
       00000000 00000000 c6efb780 c751b000 c0301de0 c6e70014 c0167633 c751b000 
Call Trace: [<c016dc2d>] [<c0167d49>] [<c0167633>] [<c0290b28>] [<c01673da>] [<c
0107503>] 
Code: ac ae 75 08 84 c0 75 f8 31 c0 eb 04 19 c0 0c 01 85 c0 75 3c 

>>EIP; c0169d64 <nfsd_lookup+94/52c>   <=====
Trace; c016dc2d <nfssvc_decode_diropargs+5d/d0>
Trace; c0167d49 <nfsd_proc_lookup+7d/90>
Trace; c0167633 <nfsd_dispatch+cb/168>
Trace; c0290b28 <svc_process+2ac/544>
Trace; c01673da <nfsd+1ca/358>
Trace; c0107503 <kernel_thread+23/30>
Code;  c0169d64 <nfsd_lookup+94/52c>
00000000 <_EIP>:
Code;  c0169d64 <nfsd_lookup+94/52c>   <=====
   0:   ac                        lods   %ds:(%esi),%al   <=====
Code;  c0169d65 <nfsd_lookup+95/52c>
   1:   ae                        scas   %es:(%edi),%al
Code;  c0169d66 <nfsd_lookup+96/52c>
   2:   75 08                     jne    c <_EIP+0xc> c0169d70 <nfsd_lookup+a0/5
2c>
Code;  c0169d68 <nfsd_lookup+98/52c>
   4:   84 c0                     test   %al,%al
Code;  c0169d6a <nfsd_lookup+9a/52c>
   6:   75 f8                     jne    0 <_EIP>
Code;  c0169d6c <nfsd_lookup+9c/52c>
   8:   31 c0                     xor    %eax,%eax
Code;  c0169d6e <nfsd_lookup+9e/52c>
   a:   eb 04                     jmp    10 <_EIP+0x10> c0169d74 <nfsd_lookup+a4
/52c>
Code;  c0169d70 <nfsd_lookup+a0/52c>
   c:   19 c0                     sbb    %eax,%eax
Code;  c0169d72 <nfsd_lookup+a2/52c>
   e:   0c 01                     or     $0x1,%al
Code;  c0169d74 <nfsd_lookup+a4/52c>
  10:   85 c0                     test   %eax,%eax
Code;  c0169d76 <nfsd_lookup+a6/52c>
  12:   75 3c                     jne    50 <_EIP+0x50> c0169db4 <nfsd_lookup+e4
/52c>

(And I really hope Evolution doesn't munge up that previous bit ;-)

I just backed down to the previous working kernel, which was CVS from
about last Friday, and everything is back to normal.

Has anyone seen anything similar?  I had upgraded in the hopes of fixing
some (infrequent) I/O errors when accessing the nfs exported partition and
benefitting from the nfs speed improvements...

Thanks,

    - Vlad




From owner-linux-xfs@oss.sgi.com Wed Mar  7 12:22:08 2001
Received:  by oss.sgi.com id <S553779AbRCGUV6>;
	Wed, 7 Mar 2001 12:21:58 -0800
Received: from deliverator.sgi.com ([204.94.214.10]:41510 "EHLO
        deliverator.sgi.com") by oss.sgi.com with ESMTP id <S553763AbRCGUVi>;
	Wed, 7 Mar 2001 12:21:38 -0800
Received: from zeus-fddi.americas.sgi.com (zeus-fddi.americas.sgi.com [128.162.8.103]) by deliverator.sgi.com (980309.SGI.8.8.8-aspam-6.2/980310.SGI-aspam) via ESMTP id MAA25833
	for <linux-xfs@oss.sgi.com>; Wed, 7 Mar 2001 12:20:31 -0800 (PST)
	mail_from (lord@sgi.com)
Received: from daisy-e185.americas.sgi.com (daisy-e194.americas.sgi.com [128.162.194.214]) by zeus-fddi.americas.sgi.com (8.9.3/americas-smart-nospam1.1) with ESMTP id OAA1051924 for <linux-xfs@oss.sgi.com>; Wed, 7 Mar 2001 14:20:22 -0600 (CST)
Received: from jen.americas.sgi.com (IDENT:root@jen.americas.sgi.com [128.162.184.86]) by daisy-e185.americas.sgi.com (SGI-8.9.3/SGI-server-1.7) with ESMTP id OAA14852 for <linux-xfs@oss.sgi.com>; Wed, 7 Mar 2001 14:20:22 -0600 (CST)
From:   Steve Lord <lord@sgi.com>
Received: by jen.americas.sgi.com (8.11.0/SGI-client-1.7) id f27KMN007761; Wed, 7 Mar 2001 14:22:23 -0600
Message-Id: <200103072022.f27KMN007761@jen.americas.sgi.com>
Date:   Wed, 7 Mar 2001 14:22:23 -0600
Subject: TAKE - fix compilation of realtime code
To:     unlisted-recipients:; (no To-header on input)
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing

This is just pushing code into the tree and out of my workarea, this
is not useful yet, as there is still no way to direct I/O requests to
the realtime device.

Date:  Wed Mar  7 12:19:24 PST 2001
Workarea:  jen.americas.sgi.com:/src/lord/xfs-linux.2.4

The following file(s) were checked into:
  bonnie.engr.sgi.com:/isms/slinx/2.4.x-xfs


Modid:  2.4.x-xfs:slinx:89231a
linux/fs/xfs/xfs_rtalloc.c - 1.63
	- make realtime code compile - fix division issues and incorrectly turned
	  off code.

linux/fs/xfs/Makefile - 1.117
	- Break realtime and grio apart as build options

linux/fs/xfs/xfsrtstubs.c - 1.11
	- move growfs_rt stub into the stub file where it belongs

linux/fs/xfs/xfs_fsops.c - 1.64
	- move growfs_rt code out of here



From owner-linux-xfs@oss.sgi.com Wed Mar  7 12:46:59 2001
Received:  by oss.sgi.com id <S553781AbRCGUqt>;
	Wed, 7 Mar 2001 12:46:49 -0800
Received: from pneumatic-tube.sgi.com ([204.94.214.22]:51215 "EHLO
        pneumatic-tube.sgi.com") by oss.sgi.com with ESMTP
	id <S553774AbRCGUqh>; Wed, 7 Mar 2001 12:46:37 -0800
Received: from zeus-fddi.americas.sgi.com (zeus-fddi.americas.sgi.com [128.162.8.103]) by pneumatic-tube.sgi.com (980327.SGI.8.8.8-aspam/980310.SGI-aspam) via ESMTP id MAA06340
	for <linux-xfs@oss.sgi.com>; Wed, 7 Mar 2001 12:56:21 -0800 (PST)
	mail_from (lord@sgi.com)
Received: from daisy-e185.americas.sgi.com (daisy-e194.americas.sgi.com [128.162.194.214]) by zeus-fddi.americas.sgi.com (8.9.3/americas-smart-nospam1.1) with ESMTP id OAA1055020; Wed, 7 Mar 2001 14:45:19 -0600 (CST)
Received: from jen.americas.sgi.com (IDENT:root@jen.americas.sgi.com [128.162.184.86]) by daisy-e185.americas.sgi.com (SGI-8.9.3/SGI-server-1.7) with ESMTP id OAA49740; Wed, 7 Mar 2001 14:45:19 -0600 (CST)
Received: from jen.americas.sgi.com by jen.americas.sgi.com (8.11.0/SGI-client-1.7) via ESMTP id f27KlKB09885; Wed, 7 Mar 2001 14:47:21 -0600
Message-Id: <200103072047.f27KlKB09885@jen.americas.sgi.com>
X-Mailer: exmh version 2.2 06/23/2000 with nmh-1.0.4
To:     Vladimir Vukicevic <vladimir@ximian.com>
cc:     linux-xfs@oss.sgi.com
Subject: Re: kernel panic with nfsd and xfs filesystem 
In-Reply-To: Message from Vladimir Vukicevic <vladimir@ximian.com> 
   of "07 Mar 2001 12:34:49 EST." <E14ahpz-0007hC-00@rain.ximian.com> 
Date:   Wed, 07 Mar 2001 14:47:20 -0600
From:   Steve Lord <lord@sgi.com>
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing


Hmm, there is nothing in the changed code which should affect anything
but the write path in nfs. Have you tried a make clean and a complete
rebuild - I have been having problems myself in that area. However, there
have been a lot of changes since last friday.

This is pretty much a shot in the dark, but might make it go away.

Steve

> 
> I just upgraded the kernel to CVS as of Mar 7 12:10 EST on our nfs
> server with a nice big xfs partition, and got this Oops from nfsd
> whenever a client tries to mount (filtered through ksymoops):
> 
> 
> ksymoops 2.3.4 on i686 2.4.2-XFS.  Options used
>      -v /usr/src/linux/vmlinux (specified)
>      -K (specified)
>      -l /proc/modules (default)
>      -o /lib/modules/2.4.2-XFS (specified)
>      -m /usr/src/linux/System.map (specified)
> 
> No modules in ksyms, skipping objects
> No ksyms, skipping lsmod
> Reading Oops report from the terminal
> Unable to handle kernel NULL pointer dereference at virtual address 00000000
>  printing eip:
> c0169d64
> *pde = 00000000
> Oops: 0000
> CPU:    1
> EIP:    0010:[<c0169d64>]
> EFLAGS: 00010246
> eax: 00000000   ebx: c6efb980   ecx: 00000000   edx: c7c27f20
> esi: 00000000   edi: c02b901e   ebp: c751b000   esp: c6e75efc
> ds: 0018   es: 0018   ss: 0018
> Process nfsd (pid: 584, stackpage=c6e75000)
> Stack: c6efb980 c6efb780 00000000 c751b000 c7484160 c7c27f20 c6e69f9c c6efb98
> 0 
>        c7554800 c6efb9a4 c751b000 c016dc2d c531bb04 c0167d49 c751b000 c6efb98
> 0 
>        00000000 00000000 c6efb780 c751b000 c0301de0 c6e70014 c0167633 c751b00
> 0 
> Call Trace: [<c016dc2d>] [<c0167d49>] [<c0167633>] [<c0290b28>] [<c01673da>] 
> [<c
> 0107503>] 
> 
> Code: ac ae 75 08 84 c0 75 f8 31 c0 eb 04 19 c0 0c 01 85 c0 75 3c 
> Unable to handle kernel NULL pointer dereference at virtual address 00000000
> c0169d64
> *pde = 00000000
> Oops: 0000
> CPU:    1
> EIP:    0010:[<c0169d64>]
> Using defaults from ksymoops -t elf32-i386 -a i386
> EFLAGS: 00010246
> eax: 00000000   ebx: c6efb980   ecx: 00000000   edx: c7c27f20
> esi: 00000000   edi: c02b901e   ebp: c751b000   esp: c6e75efc
> ds: 0018   es: 0018   ss: 0018
> Process nfsd (pid: 584, stackpage=c6e75000)
> Stack: c6efb980 c6efb780 00000000 c751b000 c7484160 c7c27f20 c6e69f9c c6efb98
> 0 
>        c7554800 c6efb9a4 c751b000 c016dc2d c531bb04 c0167d49 c751b000 c6efb98
> 0 
>        00000000 00000000 c6efb780 c751b000 c0301de0 c6e70014 c0167633 c751b00
> 0 
> Call Trace: [<c016dc2d>] [<c0167d49>] [<c0167633>] [<c0290b28>] [<c01673da>] 
> [<c
> 0107503>] 
> Code: ac ae 75 08 84 c0 75 f8 31 c0 eb 04 19 c0 0c 01 85 c0 75 3c 
> 
> >>EIP; c0169d64 <nfsd_lookup+94/52c>   <=====
> Trace; c016dc2d <nfssvc_decode_diropargs+5d/d0>
> Trace; c0167d49 <nfsd_proc_lookup+7d/90>
> Trace; c0167633 <nfsd_dispatch+cb/168>
> Trace; c0290b28 <svc_process+2ac/544>
> Trace; c01673da <nfsd+1ca/358>
> Trace; c0107503 <kernel_thread+23/30>
> Code;  c0169d64 <nfsd_lookup+94/52c>
> 00000000 <_EIP>:
> Code;  c0169d64 <nfsd_lookup+94/52c>   <=====
>    0:   ac                        lods   %ds:(%esi),%al   <=====
> Code;  c0169d65 <nfsd_lookup+95/52c>
>    1:   ae                        scas   %es:(%edi),%al
> Code;  c0169d66 <nfsd_lookup+96/52c>
>    2:   75 08                     jne    c <_EIP+0xc> c0169d70 <nfsd_lookup+a
> 0/5
> 2c>
> Code;  c0169d68 <nfsd_lookup+98/52c>
>    4:   84 c0                     test   %al,%al
> Code;  c0169d6a <nfsd_lookup+9a/52c>
>    6:   75 f8                     jne    0 <_EIP>
> Code;  c0169d6c <nfsd_lookup+9c/52c>
>    8:   31 c0                     xor    %eax,%eax
> Code;  c0169d6e <nfsd_lookup+9e/52c>
>    a:   eb 04                     jmp    10 <_EIP+0x10> c0169d74 <nfsd_lookup
> +a4
> /52c>
> Code;  c0169d70 <nfsd_lookup+a0/52c>
>    c:   19 c0                     sbb    %eax,%eax
> Code;  c0169d72 <nfsd_lookup+a2/52c>
>    e:   0c 01                     or     $0x1,%al
> Code;  c0169d74 <nfsd_lookup+a4/52c>
>   10:   85 c0                     test   %eax,%eax
> Code;  c0169d76 <nfsd_lookup+a6/52c>
>   12:   75 3c                     jne    50 <_EIP+0x50> c0169db4 <nfsd_lookup
> +e4
> /52c>
> 
> (And I really hope Evolution doesn't munge up that previous bit ;-)
> 
> I just backed down to the previous working kernel, which was CVS from
> about last Friday, and everything is back to normal.
> 
> Has anyone seen anything similar?  I had upgraded in the hopes of fixing
> some (infrequent) I/O errors when accessing the nfs exported partition and
> benefitting from the nfs speed improvements...
> 
> Thanks,
> 
>     - Vlad
> 
> 



From owner-linux-xfs@oss.sgi.com Wed Mar  7 14:46:40 2001
Received:  by oss.sgi.com id <S553806AbRCGWqa>;
	Wed, 7 Mar 2001 14:46:30 -0800
Received: from e21.nc.us.ibm.com ([32.97.136.227]:28100 "EHLO
        e21.nc.us.ibm.com") by oss.sgi.com with ESMTP id <S553789AbRCGWqX>;
	Wed, 7 Mar 2001 14:46:23 -0800
Received: from southrelay02.raleigh.ibm.com (southrelay02.raleigh.ibm.com [9.37.3.209])
	by e21.nc.us.ibm.com (8.9.3/8.9.3) with ESMTP id RAA35774;
	Wed, 7 Mar 2001 17:41:33 -0600
Received: from d04nm201.raleigh.ibm.com (d04nm201.raleigh.ibm.com [9.67.228.38])
	by southrelay02.raleigh.ibm.com (8.11.2/NCO v4.95) with ESMTP id f27MkGE95272;
	Wed, 7 Mar 2001 17:46:16 -0500
Importance: Normal
Subject: Re: Still module problems
To:     Chaitanya Tumuluri <chait@getafix.engr.sgi.com>
Cc:     linux-xfs@oss.sgi.com, "Peter T. Breuer" <ptb@it.uc3m.es>
From:   "Steve Best" <sbest@us.ibm.com>
Date:   Wed, 7 Mar 2001 17:46:15 -0500
Message-ID: <OF0F26BA5D.468FC2B2-ON85256A08.007ABDCC@raleigh.ibm.com>
X-MIMETrack: Serialize by Router on D04NM201/04/M/IBM(Release 5.0.6 |December 14, 2000) at
 03/07/2001 05:46:16 PM
MIME-Version: 1.0
Content-type: text/plain; charset=us-ascii
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing

On Wed, 7 Mar 2001, Chaitanya Tumuluri wrote:
>> On Wed, 7 Mar 2001, Peter T. Breuer wrote:
>> "GCS wrote:"
>> I would appreciate it if the patch were less invasive. Has this been
>> done?  Surely changing make_request's type is intrinsically a bad idea?
>>
>> I agree that the implementation is clever - setting the bh argument null
>> signals the new make_request to use the remaining args, kiobuff and
>> friends, instead, so only kiobuff-aware code will make use of it.  But
>> in that case why not write a separate function?  The kernel has
>> provisions for overwriting make_request _cleanly_.

> Agreed. The current state of the code is a historical artifact of trying
> to keep pace with the evolutions in ll_rw_block/__make_request() and
> friends. Originally, we did have a separate make_kio_request() function.
> The provision you mention came in later, at which point in time it was
going
> to be a _lot_ more effort to provide for lvm/md type of remapping in the
> make_kio_request() version. A quicker/easier fix was to modify the
> make_request() signature to handle kiobufs.

Is there any way to determine if the kiobufs support is in the kernel at
compile time? Community members would like to build JFS on the SGI kernel
and in our code we use
generic_make_request(READ, &bp->l_bh); /* 2.4.x kernel */

This call would change to
generic_make_request(READ, &bp->l_bh, NULL, 0, 0, 0); /* kiobufs support */

I would like to find a clean way to ifdef this. Anyone have any ideas
with this one.

Thanks,
Steve


From owner-linux-xfs@oss.sgi.com Wed Mar  7 14:53:20 2001
Received:  by oss.sgi.com id <S553806AbRCGWxL>;
	Wed, 7 Mar 2001 14:53:11 -0800
Received: from deliverator.sgi.com ([204.94.214.10]:14167 "EHLO
        deliverator.sgi.com") by oss.sgi.com with ESMTP id <S553789AbRCGWw5>;
	Wed, 7 Mar 2001 14:52:57 -0800
Received: from zeus-fddi.americas.sgi.com (zeus-fddi.americas.sgi.com [128.162.8.103]) by deliverator.sgi.com (980309.SGI.8.8.8-aspam-6.2/980310.SGI-aspam) via ESMTP id OAA19943
	for <linux-xfs@oss.sgi.com>; Wed, 7 Mar 2001 14:51:50 -0800 (PST)
	mail_from (lord@sgi.com)
Received: from daisy-e185.americas.sgi.com (daisy-e194.americas.sgi.com [128.162.194.214]) by zeus-fddi.americas.sgi.com (8.9.3/americas-smart-nospam1.1) with ESMTP id QAA1056678; Wed, 7 Mar 2001 16:51:39 -0600 (CST)
Received: from jen.americas.sgi.com (IDENT:root@jen.americas.sgi.com [128.162.184.86]) by daisy-e185.americas.sgi.com (SGI-8.9.3/SGI-server-1.7) with ESMTP id QAA30055; Wed, 7 Mar 2001 16:51:35 -0600 (CST)
Received: from jen.americas.sgi.com by jen.americas.sgi.com (8.11.0/SGI-client-1.7) via ESMTP id f27Mral17300; Wed, 7 Mar 2001 16:53:36 -0600
Message-Id: <200103072253.f27Mral17300@jen.americas.sgi.com>
X-Mailer: exmh version 2.2 06/23/2000 with nmh-1.0.4
To:     "Steve Best" <sbest@us.ibm.com>
cc:     Chaitanya Tumuluri <chait@getafix.engr.sgi.com>,
        linux-xfs@oss.sgi.com, "Peter T. Breuer" <ptb@it.uc3m.es>
Subject: Re: Still module problems 
In-Reply-To: Message from "Steve Best" <sbest@us.ibm.com> 
   of "Wed, 07 Mar 2001 17:46:15 EST." <OF0F26BA5D.468FC2B2-ON85256A08.007ABDCC@raleigh.ibm.com> 
Date:   Wed, 07 Mar 2001 16:53:36 -0600
From:   Steve Lord <lord@sgi.com>
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing


Hi Steve,

We should still talk about kiobufs I guess....

There are some defines you could use as a trigger, for instance in

linux/fs.h there is

#define IS_KIOBUFIO(inode)      __IS_FLG(inode, MS_KIOBUFIO)

so an 

#ifdef IS_KIOBUFIO

would do it.

Not pretty, but it would work. 

Steve


> On Wed, 7 Mar 2001, Chaitanya Tumuluri wrote:
> >> On Wed, 7 Mar 2001, Peter T. Breuer wrote:
> >> "GCS wrote:"
> >> I would appreciate it if the patch were less invasive. Has this been
> >> done?  Surely changing make_request's type is intrinsically a bad idea?
> >>
> >> I agree that the implementation is clever - setting the bh argument null
> >> signals the new make_request to use the remaining args, kiobuff and
> >> friends, instead, so only kiobuff-aware code will make use of it.  But
> >> in that case why not write a separate function?  The kernel has
> >> provisions for overwriting make_request _cleanly_.
> 
> > Agreed. The current state of the code is a historical artifact of trying
> > to keep pace with the evolutions in ll_rw_block/__make_request() and
> > friends. Originally, we did have a separate make_kio_request() function.
> > The provision you mention came in later, at which point in time it was
> going
> > to be a _lot_ more effort to provide for lvm/md type of remapping in the
> > make_kio_request() version. A quicker/easier fix was to modify the
> > make_request() signature to handle kiobufs.
> 
> Is there any way to determine if the kiobufs support is in the kernel at
> compile time? Community members would like to build JFS on the SGI kernel
> and in our code we use
> generic_make_request(READ, &bp->l_bh); /* 2.4.x kernel */
> 
> This call would change to
> generic_make_request(READ, &bp->l_bh, NULL, 0, 0, 0); /* kiobufs support */
> 
> I would like to find a clean way to ifdef this. Anyone have any ideas
> with this one.
> 
> Thanks,
> Steve



From owner-linux-xfs@oss.sgi.com Wed Mar  7 15:26:00 2001
Received:  by oss.sgi.com id <S553836AbRCGXZu>;
	Wed, 7 Mar 2001 15:25:50 -0800
Received: from hermes.mixx.net ([212.84.196.2]:48647 "HELO hermes.mixx.net")
	by oss.sgi.com with SMTP id <S553830AbRCGXZs>;
	Wed, 7 Mar 2001 15:25:48 -0800
Received: from mate.bln.innominate.de (cerberus.berlin.innominate.de [212.84.234.251])
	by hermes.mixx.net (Postfix) with ESMTP id 331D2F825
	for <linux-xfs@oss.sgi.com>; Thu,  8 Mar 2001 00:25:42 +0100 (CET)
Received: by mate.bln.innominate.de (Postfix, from userid 9)
	id 0A98D2CA6F; Thu,  8 Mar 2001 00:25:42 +0100 (CET)
From:   Thomas Graichen <news-innominate.list.sgi.xfs@innominate.de>
Reply-To: Thomas Graichen <thomas.graichen@innominate.com>
X-Newsgroups: innominate.list.sgi.xfs
Subject: Re: FAQ - Update and (new) question
Date:   7 Mar 2001 23:25:41 GMT
Organization: innominate AG, Berlin, Germany
Lines:  35
Distribution: local
Message-ID: <news2mail-986g1l$mf3$1@mate.bln.innominate.de>
References: <96C14D00FA99D311A8D60008C791F3640982677C@DEVWAGWODX0009>
X-Trace: mate.bln.innominate.de 984007541 23011 10.0.0.31 (7 Mar 2001 23:25:41 GMT)
X-Complaints-To: news@innominate.de
User-Agent: tin/1.4.4-20000803 ("Vet for the Insane") (UNIX) (Linux/2.4.2-XFS (i686))
To:     linux-xfs@oss.sgi.com
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing

i plan to update the XFS FAQ till weekend - will add all this too

t

"Pasch, Thomas (ACTGRO)" <extern.thomas.pasch@volkswagen.de> wrote:
> Hi,

> Q: Does XFS run with the linux software RAID md driver? 
> Nobody has tested it so far but expect it not to work like for all the other
> journaling filesystems due to internal buffer locking problems I think. If
> anyone tries this - please let me know. Also keep in mind - if you want to
> try it - also the md driver will definitely not work with the kiobuf option
> enabled. 

> This is wrong. ReiserFS is doing this perfectly on my system for some months
> now.
> But you must use the raid implementation 0.9 (raidtools-0.9) to do. So the 
> question comes up again: Does XFS run with the linux software RAID?

> What about nfs? 

> There are some problems with new filesystems and nfs in linux
> (see discussion on http://lwn.net/2001/0222/kernel.php3
> for details). So: is XFS supporting NFS (or better vice versa).

> Best regards,

> Thomas


-- 
thomas.graichen@innominate.com
                                                         innominate AG
                                                  the linux architects
tel: +49-30-308806-13   fax: -77             http://www.innominate.com

From owner-linux-xfs@oss.sgi.com Wed Mar  7 16:18:41 2001
Received:  by oss.sgi.com id <S553843AbRCHASb>;
	Wed, 7 Mar 2001 16:18:31 -0800
Received: from hermes.mixx.net ([212.84.196.2]:61448 "HELO hermes.mixx.net")
	by oss.sgi.com with SMTP id <S553835AbRCHASK>;
	Wed, 7 Mar 2001 16:18:10 -0800
Received: from mate.bln.innominate.de (cerberus.berlin.innominate.de [212.84.234.251])
	by hermes.mixx.net (Postfix) with ESMTP id 98C36F838
	for <linux-xfs@oss.sgi.com>; Thu,  8 Mar 2001 01:18:08 +0100 (CET)
Received: by mate.bln.innominate.de (Postfix, from userid 9)
	id 461BC2CA6F; Thu,  8 Mar 2001 01:18:08 +0100 (CET)
From:   Thomas Graichen <news-innominate.list.sgi.xfs@innominate.de>
Reply-To: Thomas Graichen <thomas.graichen@innominate.com>
X-Newsgroups: innominate.list.sgi.xfs
Subject: Re: Stable Debian + bonnie -> unmount problems
Date:   8 Mar 2001 00:18:08 GMT
Organization: innominate AG, Berlin, Germany
Lines:  28
Distribution: local
Message-ID: <news2mail-986j40$11c$2@mate.bln.innominate.de>
References: <gcs@agentsinside.com> <200103061458.f26Ewo218045@jen.americas.sgi.com>
X-Trace: mate.bln.innominate.de 984010688 1068 10.0.0.31 (8 Mar 2001 00:18:08 GMT)
X-Complaints-To: news@innominate.de
User-Agent: tin/1.4.4-20000803 ("Vet for the Insane") (UNIX) (Linux/2.4.2-XFS (i686))
To:     linux-xfs@oss.sgi.com
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing

Steve Lord <lord@sgi.com> wrote:

> loop may be the problem here - it is broken in 2.4.2 which our tree is 
> currently based on. Any chance of finding a small disk partition to use
> instead, this would verify the problem as being in the loop device? 
> Jens Axboe's loop patch does not apply cleanly to our tree or I would
> tell you to apply it.

> Any volunteers to make a version of loop6 which applies to 2.4.2-XFS?

> p.s. looks like 2.4.3 will fix loop fs problems though.

just put the following together:

  http://innominate.org/~graichen/projects/xfs/misc/loop-6-xfs.diff

which is jens' loop-6 patch with all his suggested changes for
the XFS tree - haven't tried it yet - but it should at least
apply cleanly and it builds fine (and i think it also should
work) ... so if anyone needs it - get it

t

-- 
thomas.graichen@innominate.com
                                                         innominate AG
                                                  the linux architects
tel: +49-30-308806-13   fax: -77             http://www.innominate.com

From owner-linux-xfs@oss.sgi.com Wed Mar  7 16:28:12 2001
Received:  by oss.sgi.com id <S553843AbRCHA2D>;
	Wed, 7 Mar 2001 16:28:03 -0800
Received: from ns.virtualhost.dk ([195.184.98.160]:22799 "EHLO virtualhost.dk")
	by oss.sgi.com with ESMTP id <S553835AbRCHA1z>;
	Wed, 7 Mar 2001 16:27:55 -0800
Received: from burns.home.kernel.dk ([192.168.0.2] ident=root)
	by virtualhost.dk with esmtp (Exim 3.22 #4)
	id 14aoGd-0006On-00; Thu, 08 Mar 2001 01:26:55 +0100
Received: from axboe by burns.home.kernel.dk with local (Exim 3.13 #1 (Debian))
	id 14aoGb-0003Aj-00; Thu, 08 Mar 2001 01:26:53 +0100
Date:   Thu, 8 Mar 2001 01:26:53 +0100
From:   Jens Axboe <axboe@suse.de>
To:     Thomas Graichen <thomas.graichen@innominate.com>
Cc:     linux-xfs@oss.sgi.com
Subject: Re: Stable Debian + bonnie -> unmount problems
Message-ID: <20010308012653.E4653@suse.de>
References: <gcs@agentsinside.com> <200103061458.f26Ewo218045@jen.americas.sgi.com> <news2mail-986j40$11c$2@mate.bln.innominate.de>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <news2mail-986j40$11c$2@mate.bln.innominate.de>; from news-innominate.list.sgi.xfs@innominate.de on Thu, Mar 08, 2001 at 12:18:08AM +0000
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing

On Thu, Mar 08 2001, Thomas Graichen wrote:
> > loop may be the problem here - it is broken in 2.4.2 which our tree is 
> > currently based on. Any chance of finding a small disk partition to use
> > instead, this would verify the problem as being in the loop device? 
> > Jens Axboe's loop patch does not apply cleanly to our tree or I would
> > tell you to apply it.
> 
> > Any volunteers to make a version of loop6 which applies to 2.4.2-XFS?
> 
> > p.s. looks like 2.4.3 will fix loop fs problems though.
> 
> just put the following together:
> 
>   http://innominate.org/~graichen/projects/xfs/misc/loop-6-xfs.diff
> 
> which is jens' loop-6 patch with all his suggested changes for
> the XFS tree - haven't tried it yet - but it should at least
> apply cleanly and it builds fine (and i think it also should
> work) ... so if anyone needs it - get it

Majordomo (or something) unfortunately ate my message, but I sent a
loop-xfs-7C an hour or so after Steve's original message... So sorry
for any extra work you have done. I've put the loop-xfs-7C patch
here now:

*.kernel.org/pub/linux/kernel/people/axboe/xfs

and now the rest of the list also knows about it's existance :-)

-- 
Jens Axboe


From owner-linux-xfs@oss.sgi.com Wed Mar  7 16:55:42 2001
Received:  by oss.sgi.com id <S553843AbRCHAzd>;
	Wed, 7 Mar 2001 16:55:33 -0800
Received: from pneumatic-tube.sgi.com ([204.94.214.22]:33834 "EHLO
        pneumatic-tube.sgi.com") by oss.sgi.com with ESMTP
	id <S553835AbRCHAzK>; Wed, 7 Mar 2001 16:55:10 -0800
Received: from ledzep.americas.sgi.com (ledzep.americas.sgi.com [137.38.226.97]) by pneumatic-tube.sgi.com (980327.SGI.8.8.8-aspam/980310.SGI-aspam) via ESMTP id RAA07338
	for <linux-xfs@oss.sgi.com>; Wed, 7 Mar 2001 17:04:53 -0800 (PST)
	mail_from (cattelan@thebarn.com)
Received: from gibble.americas.sgi.com (gibble.americas.sgi.com [128.162.195.80]) by ledzep.americas.sgi.com (SGI-SGI-8.9.3/americas-smart-nospam1.1) with ESMTP id SAA91086; Wed, 7 Mar 2001 18:53:53 -0600 (CST)
Received: from thebarn.com (localhost [127.0.0.1])
	by gibble.americas.sgi.com (8.11.2/8.11.2) with ESMTP id f280pUn19537;
	Wed, 7 Mar 2001 19:51:30 -0500
Message-ID: <3AA6D790.BB1A6126@thebarn.com>
Date:   Wed, 07 Mar 2001 19:51:28 -0500
From:   Russell Cattelan <cattelan@thebarn.com>
X-Mailer: Mozilla 4.76 [en] (X11; U; Linux 2.4.2-XFS i686)
X-Accept-Language: en
MIME-Version: 1.0
To:     Jens Axboe <axboe@suse.de>
CC:     Thomas Graichen <thomas.graichen@innominate.com>,
        linux-xfs@oss.sgi.com
Subject: Re: Stable Debian + bonnie -> unmount problems
References: <gcs@agentsinside.com> <200103061458.f26Ewo218045@jen.americas.sgi.com> <news2mail-986j40$11c$2@mate.bln.innominate.de> <20010308012653.E4653@suse.de>
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing

Jens Axboe wrote:

Just as a side note... oss is very close to having zmailer expunged as the
mail server. Hopefully the switch to sendmail will clear
up this dropped message problem.



> On Thu, Mar 08 2001, Thomas Graichen wrote:
> > > loop may be the problem here - it is broken in 2.4.2 which our tree is
> > > currently based on. Any chance of finding a small disk partition to use
> > > instead, this would verify the problem as being in the loop device?
> > > Jens Axboe's loop patch does not apply cleanly to our tree or I would
> > > tell you to apply it.
> >
> > > Any volunteers to make a version of loop6 which applies to 2.4.2-XFS?
> >
> > > p.s. looks like 2.4.3 will fix loop fs problems though.
> >
> > just put the following together:
> >
> >   http://innominate.org/~graichen/projects/xfs/misc/loop-6-xfs.diff
> >
> > which is jens' loop-6 patch with all his suggested changes for
> > the XFS tree - haven't tried it yet - but it should at least
> > apply cleanly and it builds fine (and i think it also should
> > work) ... so if anyone needs it - get it
>
> Majordomo (or something) unfortunately ate my message, but I sent a
> loop-xfs-7C an hour or so after Steve's original message... So sorry
> for any extra work you have done. I've put the loop-xfs-7C patch
> here now:
>
> *.kernel.org/pub/linux/kernel/people/axboe/xfs
>
> and now the rest of the list also knows about it's existance :-)
>
> --
> Jens Axboe

--
Russell Cattelan
--
Digital Elves inc. -- Currently on loan to SGI
Linux XFS core developer.




From owner-linux-xfs@oss.sgi.com Wed Mar  7 19:28:11 2001
Received:  by oss.sgi.com id <S553681AbRCHD2C>;
	Wed, 7 Mar 2001 19:28:02 -0800
Received: from tux.mkp.net ([130.225.60.11]:43021 "EHLO tux.mkp.net")
	by oss.sgi.com with ESMTP id <S553648AbRCHD1u>;
	Wed, 7 Mar 2001 19:27:50 -0800
Received: from tux.mkp.net ([130.225.60.11] helo=jaguar.mkp.net)
	by tux.mkp.net with esmtp (Exim 3.16 #1)
	id 14ar5d-0005g0-00; Thu, 08 Mar 2001 04:27:47 +0100
Received: (from mkp@localhost)
	by jaguar.mkp.net (8.9.3/8.9.3) id WAA08370;
	Wed, 7 Mar 2001 22:26:45 -0500
X-Authentication-Warning: jaguar.mkp.net: mkp set sender to mkp@mkp.net using -f
To:     linux-xfs@oss.sgi.com
Subject: TAKE - BLKBSZSET, BLKBSZGET ioctls
From:   "Martin K. Petersen" <mkp@linuxcare.com>
Organization: Linuxcare, Inc.
Date:   07 Mar 2001 22:26:45 -0500
Message-ID: <yq1y9uh2b62.fsf@jaguar.mkp.net>
Lines:  30
User-Agent: Gnus/5.0808 (Gnus v5.8.8) XEmacs/21.1 (Canyonlands)
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing


This is the consensus reached after some discussions with Andries
Brouwer.  Except for the name, BLKBSZSET is identical to BLKGETSIZE
(from userland's perspective anyway) and uses the same ioctl number.
I.e. we maintain binary compatibility, so people won't have to upgrade
their xfs utilities.

Date:  Wed Mar  7 19:20:22 PST 2001
Workarea:  sshgate.corp.sgi.com:/export/d0/mkp/XFS/slinx-blk

The following file(s) were checked into:
  bonnie.engr.sgi.com:/isms/slinx/2.4.x-xfs


Modid:  2.4.x-xfs:slinx:89288a
linux/include/linux/fs.h - 1.83
linux/drivers/scsi/sd.c - 1.33
linux/arch/sparc64/kernel/ioctl32.c - 1.31
linux/drivers/block/blkpg.c - 1.5
linux/drivers/ide/ide.c - 1.18
linux/arch/mips64/kernel/ioctl32.c - 1.5
linux/drivers/md/lvm.c - 1.7
linux/drivers/md/md.c - 1.8
        - Replaced BLKSETSIZE with BLKBSZSET.  BLKBSZGET is new kid on
          the block (boo, hiss, bad pun)...

-- 
Martin K. Petersen, Principal Linux Consultant, Linuxcare, Inc.
mkp@linuxcare.com, http://www.linuxcare.com/
SGI XFS for Linux Developer, http://oss.sgi.com/projects/xfs/

From owner-linux-xfs@oss.sgi.com Wed Mar  7 19:32:12 2001
Received:  by oss.sgi.com id <S553681AbRCHDcC>;
	Wed, 7 Mar 2001 19:32:02 -0800
Received: from tux.mkp.net ([130.225.60.11]:45325 "EHLO tux.mkp.net")
	by oss.sgi.com with ESMTP id <S553648AbRCHDbs>;
	Wed, 7 Mar 2001 19:31:48 -0800
Received: from tux.mkp.net ([130.225.60.11] helo=jaguar.mkp.net)
	by tux.mkp.net with esmtp (Exim 3.16 #1)
	id 14ar9V-0005gA-00; Thu, 08 Mar 2001 04:31:46 +0100
Received: (from mkp@localhost)
	by jaguar.mkp.net (8.9.3/8.9.3) id WAA08465;
	Wed, 7 Mar 2001 22:30:45 -0500
X-Authentication-Warning: jaguar.mkp.net: mkp set sender to mkp@mkp.net using -f
To:     linux-xfs@oss.sgi.com
Subject: TAKE - BLKBSZSET vs. init.c
From:   "Martin K. Petersen" <mkp@linuxcare.com>
Organization: Linuxcare, Inc.
Date:   07 Mar 2001 22:30:45 -0500
Message-ID: <yq1u2552aze.fsf@jaguar.mkp.net>
Lines:  16
User-Agent: Gnus/5.0808 (Gnus v5.8.8) XEmacs/21.1 (Canyonlands)
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing


Date:  Wed Mar  7 19:29:45 PST 2001
Workarea:  sshgate.corp.sgi.com:/export/d0/mkp/XFS/slinx-blk

The following file(s) were checked into:
  bonnie.engr.sgi.com:/isms/slinx/2.4.x-xfs


Modid:  2.4.x-xfs:slinx:89289a
cmd/xfsprogs/libxfs/init.c - 1.2
        - Replaced BLKSETSIZE with BLKBSZSET.

-- 
Martin K. Petersen, Principal Linux Consultant, Linuxcare, Inc.
mkp@linuxcare.com, http://www.linuxcare.com/
SGI XFS for Linux Developer, http://oss.sgi.com/projects/xfs/

From owner-linux-xfs@oss.sgi.com Wed Mar  7 20:53:13 2001
Received:  by oss.sgi.com id <S553681AbRCHExD>;
	Wed, 7 Mar 2001 20:53:03 -0800
Received: from pneumatic-tube.sgi.com ([204.94.214.22]:61244 "EHLO
        pneumatic-tube.sgi.com") by oss.sgi.com with ESMTP
	id <S553648AbRCHEwp>; Wed, 7 Mar 2001 20:52:45 -0800
Received: from info.engr.sgi.com (info.engr.sgi.com [192.26.80.216]) by pneumatic-tube.sgi.com (980327.SGI.8.8.8-aspam/980310.SGI-aspam) via ESMTP id VAA04884; Wed, 7 Mar 2001 21:02:29 -0800 (PST)
	mail_from (nobody@info.engr.sgi.com)
Received: (from nobody@localhost)
	by info.engr.sgi.com (SGI-8.9.3/8.9.3) id UAA43148;
	Wed, 7 Mar 2001 20:52:44 -0800 (PST)
Date:   Wed, 7 Mar 2001 20:52:44 -0800 (PST)
Message-Id: <200103080452.UAA43148@info.engr.sgi.com>
X-Pv-Incident: 817252 
webPV:  proxy2.melbourne.sgi.com 
webExec: webpvsubmit,PvProjectIncident 
Reply-To: sgi.bugs.xfs@fido.engr.sgi.com
From:   pv@relay.sgi.com (dxm@engr.sgi.com)
Subject: BUG 817252 - xfs qa 013 trips assertion
To:     nb@sgi.com
Cc:     linux-xfs@oss.sgi.com
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing

View Incident: http://co-op.engr.sgi.com/BugWorks/code/bwxquery.cgi?search=Search&wlong=1&view_type=Bug&wi=817252

Submitter : dxm                       Submitter Domain : engr               
Assigned Engineer : nb                Assigned Domain : sgi.com             
Assigned Group : xfs-linux            Category : software                   
Customer Reported : F                 Priority : 2                          
Project : xfs-linux                   Status : open                         
Description :
fuzzy.melbourne - sgi 1200, scsi, 2 cpu:

xfs qa 013 now reliably trips this assertion:

XFS assertion failed: ismrlocked(io->io_iolock, MR_UPDATE), file: xfs_lrw.c, line: 414
kernel BUG at debug.c:48!

Entering kdb (current=0xc3e30000, pid 1475) on processor 0 Oops: invalid operand
due to oops @ 0xc8827d7d
eax = 0x0000001a ebx = 0xc5592a18 ecx = 0x00000001 edx = 0x00000001 
esi = 0xc5592918 edi = 0x00000000 esp = 0xc3e31e30 eip = 0xc8827d7d 
ebp = 0xc3e31e3c xss = 0x00000018 xcs = 0x00000010 eflags = 0x00010286 
xds = 0x00000018 xes = 0x00000018 origeax = 0xffffffff &regs = 0xc3e31dfc
[0]kdb> bt
    EBP       EIP         Function(args)
0xc3e31e3c 0xc8827d7d [xfs_support]assfail+0x2d (0xc89ba980, 0xc89ba817, 0x19e)
                               xfs_support .text 0xc8827060 0xc8827d50 0xc8827d84
0xc3e31ebc 0xc899c2a9 [xfs]xfs_zero_eof+0x79 (0xc400cd24, 0xc5592a18, 0x17b000, 0x0, 0xa8cea)
                               xfs .text 0xc8929060 0xc899c230 0xc899c8e0
0xc3e31f34 0xc899cb8e [xfs]xfs_write+0x2ae (0xc5592918, 0xc3e31f74, 0x0, 0x0, 0x0)
                               xfs .text 0xc8929060 0xc899c8e0 0xc899cdfc
0xc3e31f98 0xc899872c [xfs]linvfs_write+0x138 (0xc4aa0440, 0x40116000, 0x5000, 0xc4aa0460)
                               xfs .text 0xc8929060 0xc89985f4 0xc8998754
0xc3e31fbc 0xc01349e5 sys_write+0x95 (0x3, 0x40116000, 0x5000, 0x482967d7, 0x59b7)
                               kernel .text 0xc0100000 0xc0134950 0xc0134a20
           0xc01090d7 system_call+0x33
                               kernel .text 0xc0100000 0xc01090a4 0xc01090dc

From owner-linux-xfs@oss.sgi.com Wed Mar  7 20:58:33 2001
Received:  by oss.sgi.com id <S553681AbRCHE6X>;
	Wed, 7 Mar 2001 20:58:23 -0800
Received: from deliverator.sgi.com ([204.94.214.10]:45340 "EHLO
        deliverator.sgi.com") by oss.sgi.com with ESMTP id <S553648AbRCHE6M>;
	Wed, 7 Mar 2001 20:58:12 -0800
Received: from info.engr.sgi.com (info.engr.sgi.com [192.26.80.216]) by deliverator.sgi.com (980309.SGI.8.8.8-aspam-6.2/980310.SGI-aspam) via ESMTP id UAA24236; Wed, 7 Mar 2001 20:57:05 -0800 (PST)
	mail_from (nobody@info.engr.sgi.com)
Received: (from nobody@localhost)
	by info.engr.sgi.com (SGI-8.9.3/8.9.3) id UAA34762;
	Wed, 7 Mar 2001 20:58:11 -0800 (PST)
Date:   Wed, 7 Mar 2001 20:58:11 -0800 (PST)
Message-Id: <200103080458.UAA34762@info.engr.sgi.com>
X-Pv-Incident: 799616 
webPV:  proxy2.melbourne.sgi.com 
webExec: webpvupdate,pvincident 
Reply-To: sgi.bugs.xfs@fido.engr.sgi.com
From:   pv@relay.sgi.com (dxm@engr.sgi.com)
Subject: REASSIGN 799616 - RESVSP & UNRESVSP breakage
To:     nb@sgi.com
Cc:     linux-xfs@oss.sgi.com
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing

View Incident: http://co-op.engr.sgi.com/BugWorks/code/bwxquery.cgi?search=Search&wlong=1&view_type=Bug&wi=799616

 Status : open                         Priority : 3                         
*Assigned Engineer : nb               *Assigned Domain : sgi.com            
 Submitter : dxm                       Project : xfs-linux                  
 Assigned Group : xfs-linux            Opened Date : 08/21/00               
*Modified User : dxm                  *Modified User Domain : engr          
*Description :
when running the xfscrash corrupt test, an ASSERT can
get tripped:

XFS assertion failed: imap.br_startblock != DELAYSTARTBLOCK, file: xfs_vnodeops.c, line: 5803
kernel BUG at xfs_debug.c:50!

Entering kdb (0xedeb6000) on processor 1 Panic: invalid operand
due to panic @ 0xf89028a9
eax = 0x0000001e ebx = 0x000f8fff ecx = 0xc02b6f2c edx = 0x00000000 
esi = 0x00000000 edi = 0x00000000 esp = 0xedeb79b4 eip = 0xf89028a9 

.....


==========================
ADDITIONAL INFORMATION (REASSIGN)
From: dxm@engr (BugWorks)
Date: Mar 07 2001 08:58:11PM
==========================

Reassiging to nb to indicate I've no idea how to fix this
and it's there for whoever wants it...

From owner-linux-xfs@oss.sgi.com Thu Mar  8 04:24:25 2001
Received:  by oss.sgi.com id <S553846AbRCHMYP>;
	Thu, 8 Mar 2001 04:24:15 -0800
Received: from esparrall.udg.es ([130.206.124.16]:59778 "EHLO esparrall.udg.es")
	by oss.sgi.com with ESMTP id <S553843AbRCHMYJ>;
	Thu, 8 Mar 2001 04:24:09 -0800
Received: from gcs by esparrall.udg.es with local (Exim 3.22 #1 (Debian))
	id 14azS4-0004yJ-00
	for <linux-xfs@oss.sgi.com>; Thu, 08 Mar 2001 13:23:28 +0100
Date:   Thu, 8 Mar 2001 13:23:28 +0100
From:   GCS <gcs@agentsinside.com>
To:     linux-xfs@oss.sgi.com
Subject: XFS root filesystem running on Debian
Message-ID: <20010308132328.C18951@esparrall.udg.es>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.3.15i
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing

Hello,

I changed to XFS for a while. It was strange, I had some errors too.
The most strange things were:
- mount said '/' is ext2, even if fstab contains xfs.
- on reboot it could not unmount ext2 filesystems correctly, boot time it
had to fsck all the ext2 filesystems on the hdd.

Is the white paper ready 'how to change the root filesystem to xfs'?
Thanks.

From owner-linux-xfs@oss.sgi.com Thu Mar  8 07:35:28 2001
Received:  by oss.sgi.com id <S553683AbRCHPfT>;
	Thu, 8 Mar 2001 07:35:19 -0800
Received: from sgi.SGI.COM ([192.48.153.1]:29537 "EHLO sgi.com")
	by oss.sgi.com with ESMTP id <S553661AbRCHPfD>;
	Thu, 8 Mar 2001 07:35:03 -0800
Received: from info.engr.sgi.com ([192.26.80.216]) 
	by sgi.com (980327.SGI.8.8.8-aspam/980304.SGI-aspam:
       SGI does not authorize the use of its proprietary
       systems or networks for unsolicited or bulk email
       from the Internet.) 
	via ESMTP id HAA04646; Thu, 8 Mar 2001 07:34:58 -0800 (PST)
	mail_from (nobody@info.engr.sgi.com)
Received: (from nobody@localhost)
	by info.engr.sgi.com (SGI-8.9.3/8.9.3) id HAA39151;
	Thu, 8 Mar 2001 07:34:57 -0800 (PST)
Date:   Thu, 8 Mar 2001 07:34:57 -0800 (PST)
Message-Id: <200103081534.HAA39151@info.engr.sgi.com>
X-Pv-Incident: 817252 
webPV:  jen.americas.sgi.com 
webExec: webpvupdate,pvincident 
Reply-To: sgi.bugs.xfs@fido.engr.sgi.com
From:   pv@odin.corp.sgi.com (lord@sgi.com)
Subject: REASSIGN 817252 - xfs qa 013 trips assertion
To:     lord@sgi.com
Cc:     linux-xfs@oss.sgi.com
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing

View Incident: http://co-op.engr.sgi.com/BugWorks/code/bwxquery.cgi?search=Search&wlong=1&view_type=Bug&wi=817252

 Status : open                         Priority : 2                         
*Assigned Engineer : lord              Submitter : dxm                      
 Project : xfs-linux                  *Assigned Group : xfs-linux           
 Opened Date : 03/07/01               *Modified User : lord                 
*Modified User Domain : sgi.com       *Description :
fuzzy.melbourne - sgi 1200, scsi, 2 cpu:

xfs qa 013 now reliably trips this assertion:

XFS assertion failed: ismrlocked(io->io_iolock, MR_UPDATE), file: xfs_lrw.c, line: 414
kernel BUG at debug.c:48!

Entering kdb (current=0xc3e30000, pid 1475) on processor 0 Oops: invalid operand
due to oops @ 0xc8827d7d
eax = 0x0000001a ebx = 0xc5592a18 ecx = 0x00000001 edx = 0x00000001 

.....


==========================
ADDITIONAL INFORMATION (REASSIGN)
From: lord@sgi.com (BugWorks)
Date: Mar 08 2001 07:34:57AM
==========================

This is easy

From owner-linux-xfs@oss.sgi.com Thu Mar  8 07:55:28 2001
Received:  by oss.sgi.com id <S553679AbRCHPzS>;
	Thu, 8 Mar 2001 07:55:18 -0800
Received: from lucy.physik.TU-Cottbus.De ([141.43.75.1]:55840 "HELO convert =?ISO-8859-1?Q?rfc822-to-8bitomY=9D?=
        lucy.physik.tu-cottbus.de") by oss.sgi.com with SMTP
	id <S553711AbRCHPzJ>; Thu, 8 Mar 2001 07:55:09 -0800
Received: (qmail 105796 invoked from network); 8 Mar 2001 15:54:54 -0000
Received: from wynton.physik.tu-cottbus.de (george@141.43.75.16)
  by lucy.physik.tu-cottbus.de with SMTP; 8 Mar 2001 15:54:54 -0000
Date:   Thu, 8 Mar 2001 16:54:54 +0100
From:   Ionut Georgescu <george@physik.tu-cottbus.de>
To:     GCS <gcs@agentsinside.com>
cc:     linux-xfs@oss.sgi.com
Subject: Re: Still module problems
In-Reply-To: <20010306204248.A845@esparrall.udg.es>
Message-ID: <Pine.SGI.4.20.0103081646080.37931-100000@wynton.physik.tu-cottbus.de>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=X-UNKNOWN
Content-Transfer-Encoding: 8BIT
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing

Ionut Georgescu sends greetings to GCS

>Hello all,
>
>If someone has Debian, please contact me personally. I do not know why, but
>I still have unresolved symbols in all modules. I have deleted the kernel
>source + System.[map|old] + vmlinuz[.old], cvs update, cp -Rd linux /usr/src/,
>mv linux linux-2.4.2-XFS etc.
>After make install, 'ls -l /boot':
>lrwxrwxrwx    1 root     root           20 MÁR  6 20:09 System.map -> System.map-2.4.2-XFS
>-rw-r--r--    1 root     root       269967 DEC 25 15:54 System.map-2.2.18pre21
>-rw-r--r--    1 root     root       405781 MÁR  6 20:09 System.map-2.4.2-XFS
>lrwxrwxrwx    1 root     root           17 MÁR  6 20:09 vmlinuz -> vmlinuz-2.4.2-XFS
>-rwxr-xr-x    1 root     root       985303 DEC 25 15:54 vmlinuz-2.2.18pre21
>-rw-r--r--    1 root     root       703718 MÁR  6 20:09 vmlinuz-2.4.2-XFS
>(Well, I just realized it does not have +x, but ...)
>
>The directory /lib/modules/2.4.2-XFS exists and contains the modules.
>Result -> unresolved symbols, even printk, which is such basic like the air
>for us (imho).
>Then, I unpacked the original kernel source, copied .config into it, and
>built the kernel and booted without problems. :-0 What can it be?
>I could 'tight it down' to this: /var/log/kern.log
>Mar  6 20:06:50 esparrall kernel: Inspecting /boot/System.map
>Mar  6 20:06:51 esparrall kernel: Loaded 14344 symbols from /boot/System.map.
>Mar  6 20:06:51 esparrall kernel: Symbols match kernel version 2.4.2.
>				  ^-- It should be 2.4.2-XFS, should not be?
>Mar  6 20:06:51 esparrall kernel: No module symbols loaded.
>				  ^-- Any reason for that?
>Mar  6 20:06:51 esparrall kernel: Linux version 2.4.2-XFS (root@esparrall)

This is a typical case for using the --flavour switch with make-kpkg and
an unpatched Makefile. The original Makefile is not aware of flavours, so
everything will look fine (module dirs, etc ... ) up to the kernel
version. Try giving uname. If it shows only 2.4.2 then this is you. Try to
use the patch given in /usr/share/doc/kernel-packages/Flavours.gz . It
will fail at the 2nd hunk, as it was ment for 2.2 kernels. Look in
Makefile.rej and do it by hand.

Beware! For as far as I remember debian refuses big caps in kernel
version!


>Any help is greatly appreciated. Thanks, Laszlo
>Ps:Previously I had the same problem, but somewhen it began working, and I
>did not care. But now it is the same, and I am very curious...

This happened to me too. Until I patched the Makefile. Then it worked
whatsoever.


***************
* Ionut Georgescu      
* http://www.physik.tu-cottbus.de/~george/
* ICQ: 38973105
* "In Windows you can do everything Microsoft wants you to do; in Unix you
*                can do anything the computer is able to do."



From owner-linux-xfs@oss.sgi.com Thu Mar  8 08:03:49 2001
Received:  by oss.sgi.com id <S553711AbRCHQDj>;
	Thu, 8 Mar 2001 08:03:39 -0800
Received: from [65.100.85.34] ([65.100.85.34]:1418 "EHLO roujin.gargoylecc.com")
	by oss.sgi.com with ESMTP id <S553679AbRCHQDY>;
	Thu, 8 Mar 2001 08:03:24 -0800
Received: from [65.100.85.35] (IDENT:ringram@gargoyle.gargoylecc.com [65.100.85.35])
	by roujin.gargoylecc.com (8.9.3/8.9.3) with ESMTP id AAA21661;
	Fri, 9 Mar 2001 00:17:24 -0700
Date:   Thu, 8 Mar 2001 09:06:48 -0700 (MST)
From:   Russel Ingram <ringram@gargoylecc.com>
X-Sender:  <ringram@localhost.localdomain>
To:     GCS <gcs@agentsinside.com>
cc:     <linux-xfs@oss.sgi.com>
Subject: Re: XFS root filesystem running on Debian
In-Reply-To: <20010308132328.C18951@esparrall.udg.es>
Message-ID: <Pine.LNX.4.30.0103080903030.14413-100000@localhost.localdomain>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing

On Thu, 8 Mar 2001, GCS wrote:

> Is the white paper ready 'how to change the root filesystem to xfs'?
> Thanks.
I don't know if the guys at SGI are planning a white paper as such, but I
have a howto on moving to XFS that includes some pointers on changing the
/ filesystem to XFS at:

http://www.gargoylecc.com/Linux+XFS-HOWTO.txt

See section 4.1.  I hope it's helpful to you.

Russ

-- 
---------------------------------------------------------------
"Bill Gates and Microsoft have ruined the computer industry for
 a long time to come by creating a class of ignorant and lazy
 computer users."  --Russel Ingram

"Mommy ... can I go out and ... KILL TONIGHT!?"
                   --Glen  Danzig, The Misfits
---------------------------------------------------------------
Russ Ingram
Gargoyle Computer Consulting
(307)742-1361
www.gargoylecc.com


From owner-linux-xfs@oss.sgi.com Thu Mar  8 17:47:22 2001
Received:  by oss.sgi.com id <S553952AbRCIBrN>;
	Thu, 8 Mar 2001 17:47:13 -0800
Received: from note.orchestra.cse.unsw.EDU.AU ([129.94.242.29]:34564 "HELO
        note.orchestra.cse.unsw.EDU.AU") by oss.sgi.com with SMTP
	id <S553702AbRCIBrC>; Thu, 8 Mar 2001 17:47:02 -0800
Received: From
	notabene.cse.unsw.edu.au ([129.94.211.194] == dulcimer.orchestra.cse.unsw.EDU.AU)
	(for <ak@suse.de>) (for <lord@sgi.com>) (for <linux-xfs@oss.sgi.com>)
	By note With Smtp ; Fri, 9 Mar 2001 12:46:38 +1100 
Received: from neilb by notabene.cse.unsw.edu.au with local (Exim 3.12 #1 (Debian))
	id 14bBzX-0004YE-00; Fri, 09 Mar 2001 12:46:51 +1100
From:   Neil Brown <neilb@cse.unsw.edu.au>
To:     Andi Kleen <ak@suse.de>
Date:   Fri, 9 Mar 2001 12:46:50 +1100 (EST)
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Message-ID: <15016.13834.766880.379970@notabene.cse.unsw.edu.au>
Cc:     Steve Lord <lord@sgi.com>, linux-xfs@oss.sgi.com
Subject: Re: TAKE - reintroduce the nfs inode reference cache
In-Reply-To: message from Andi Kleen on Wednesday March 7
References: <ak@suse.de>
	<200103062056.f26Kuo106999@jen.americas.sgi.com>
	<20010306222356.A15312@gruyere.muc.suse.de>
	<15013.28348.768980.329307@notabene.cse.unsw.edu.au>
	<20010307094517.A23323@gruyere.muc.suse.de>
X-Mailer: VM 6.72 under Emacs 20.7.2
X-face:	[Gw_3E*Gng}4rRrKRYotwlE?.2|**#s9D<ml'fY1Vw+@XfR[fRCsUoP?K6bt3YD\ui5Fh?f
	LONpR';(ql)VM_TQ/<l_^D3~B:z$\YC7gUCuC=sYm/80G=$tt"98mr8(l))QzVKCk$6~gldn~*FK9x
	8`;pM{3S8679sP+MbP,72<3_PIH-$I&iaiIb|hV1d%cYg))BmI)AZ
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing

On Wednesday March 7, ak@suse.de wrote:
> 
> Main problem is that struct file is not preserved over RPCs. Various
> file systems put state into struct file (ext2 readahead context, XFS 
> did discard preallocation in f_op->release, which is rather costly).
> So I was looking for a way to preserve struct file across multiple
> RPCs.

Well.... ext2 doesn't really put any state in the "struct file".  The
readahead context is done by the VFS layer and knfsd makes a point of
copying that out into a cache after a read, and putting it back before
the next read on the same file.
But I can appreciate that it makes sense for filesystems to put
context in the "struct file".  Pre-allocation for writes is an obvious
example (though ext2 seems to put that in the inode).

Certainly caching "struct file"s would be cleaner than the current
code for caching read-ahead context.
A tricky bit is deciding whether a particular IO request on a
particular file should use a cached "struct file" or whether it should
use a new one.

One could imagine a scenario where two clients are sequentially
reading a large file but are up to different parts of the file.  If we
used the same struct file for both clients (as the current read-ahead
cache effectively does) then it will look like random access and no
readahead will happen.

The obvious response to this would be to have the cache keyed not just
on inode, but also on "client" and maybe "offset".

Just adding the "client" to the key wouldn't help if two programs on
the same client were accessing the file.
Adding the "offset" to cope with this would cause problems in the client
sent, for example, three consecutive reads on the same file.
They would be handled by different threads and would run in parallel
and so would probably confuse the read-ahead code (which assumes
sequential reads).

Another problem in using the current offset is that it assumes that
any context that the filesystem attaches to a file handle is only
really useful for sequential accesses.  This is true for readahead and
presumably for preallocation, but is that the only sort of context?

Ofcourse, the mantra of NFS is "file sharing is rare" so we could just
not worry, and that would be no worse than the current situation, and
possibly better.

Another question is "Will holding a struct file for longer cause any
other problems?".

I know that the userspace NFS server caches file descriptors (which
are like struct files) and this has the effect that if you write to an
executable file via NFS, you cannot execute it on the server until the
cache gets flushed, as Linux stops you from executing a file which is
open for write access.

This would not be a problem for the kernel server because
"get_write_access" is separate from holding a struct file, but I would
need to study the code a bit to be sure there are not other similar
problems.


> 
> [This is somewhat similar to another long standing NFS bug BTW -- 
> it shares UDP sockets and there is excessive ARP traffic because the neighbour
> state machine times out all the time and does ARP reprobes]
> 

I don't understand this I'm afraid, but I am curious.  Could you give
a bit more detail?


NeilBrown

From owner-linux-xfs@oss.sgi.com Thu Mar  8 18:13:22 2001
Received:  by oss.sgi.com id <S553768AbRCICNN>;
	Thu, 8 Mar 2001 18:13:13 -0800
Received: from [207.198.250.232] ([207.198.250.232]:45326 "HELO perilith.com")
	by oss.sgi.com with SMTP id <S553762AbRCICMr>;
	Thu, 8 Mar 2001 18:12:47 -0800
Received: by perilith.com (Postfix, from userid 1034)
	id BE1E13E087; Thu,  8 Mar 2001 21:12:45 -0500 (EST)
Date:   Thu, 8 Mar 2001 18:12:45 -0800
From:   Drew Bloechl <drew@cesspool.net>
To:     linux-xfs@oss.sgi.com
Subject: Messages from last few days
Message-ID: <20010308181245.I31627@perilith.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.3.15i
X-PGP-Fingerprint: B8B4 8A05 A58B 5252 FFC5  E455 D831 31A3 3385 5516
X-PGP-Public-Key: http://cesspool.net/drew.pubkey.txt
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing

I had some domain issues a few days ago and as a result fell off this
list.  Would anyone be willing to do me a big favor and make an mbox
file with the messages between March 02 and now in it and make it
available somewhere?  

Contact me privately, please. 

-- 
Drew Bloechl
drew@cesspool.net
PGP key ID: 33855516

From owner-linux-xfs@oss.sgi.com Thu Mar  8 19:07:13 2001
Received:  by oss.sgi.com id <S553952AbRCIDHD>;
	Thu, 8 Mar 2001 19:07:03 -0800
Received: from deliverator.sgi.com ([204.94.214.10]:27712 "EHLO
        deliverator.sgi.com") by oss.sgi.com with ESMTP id <S553652AbRCIDGp>;
	Thu, 8 Mar 2001 19:06:45 -0800
Received: from boing.melbourne.sgi.com (boing.melbourne.sgi.com [134.14.55.141]) by deliverator.sgi.com (980309.SGI.8.8.8-aspam-6.2/980310.SGI-aspam) via ESMTP id TAA16794
	for <linux-xfs@oss.sgi.com>; Thu, 8 Mar 2001 19:05:37 -0800 (PST)
	mail_from (tes@boing.melbourne.sgi.com)
Received: (from tes@localhost)
	by boing.melbourne.sgi.com (SGI-8.9.3/8.9.3) id OAA45962;
	Fri, 9 Mar 2001 14:05:26 +1100 (EDT)
From:   Timothy Shimmin <tes@boing.melbourne.sgi.com>
Message-Id: <200103090305.OAA45962@boing.melbourne.sgi.com>
Subject: Re: Messages from last few days
To:     drew@cesspool.net (Drew Bloechl)
Date:   Fri, 9 Mar 2001 14:05:25 +1100 (EDT)
Cc:     linux-xfs@oss.sgi.com
In-Reply-To: <20010308181245.I31627@perilith.com> from "Drew Bloechl" at Mar 08, 2001 06:12:45 PM
X-Mailer: ELM [version 2.5 PL2]
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing

Hi Drew,
> 
> I had some domain issues a few days ago and as a result fell off this
> list.  Would anyone be willing to do me a big favor and make an mbox
> file with the messages between March 02 and now in it and make it
> available somewhere?  
> 
You could look at the mailing-list archive for March:

http://linux-xfs.sgi.com/projects/xfs/mail_archive/0103/maillist.html

--Tim

From owner-linux-xfs@oss.sgi.com Thu Mar  8 21:41:44 2001
Received:  by oss.sgi.com id <S553952AbRCIFle>;
	Thu, 8 Mar 2001 21:41:34 -0800
Received: from tux.mkp.net ([130.225.60.11]:41733 "EHLO tux.mkp.net")
	by oss.sgi.com with ESMTP id <S553765AbRCIFlM>;
	Thu, 8 Mar 2001 21:41:12 -0800
Received: from tux.mkp.net ([130.225.60.11] helo=jaguar.mkp.net)
	by tux.mkp.net with esmtp (Exim 3.16 #1)
	id 14bFeG-0006Pq-00; Fri, 09 Mar 2001 06:41:09 +0100
Received: (from mkp@localhost)
	by jaguar.mkp.net (8.9.3/8.9.3) id AAA09329;
	Fri, 9 Mar 2001 00:40:03 -0500
X-Authentication-Warning: jaguar.mkp.net: mkp set sender to mkp@mkp.net using -f
To:     linux-xfs@oss.sgi.com
Subject: TAKE - BLKBSZSET fixes
From:   "Martin K. Petersen" <mkp@linuxcare.com>
Organization: Linuxcare, Inc.
Date:   09 Mar 2001 00:40:03 -0500
Message-ID: <yq13dcn1owc.fsf@jaguar.mkp.net>
Lines:  20
User-Agent: Gnus/5.0808 (Gnus v5.8.8) XEmacs/21.1 (Canyonlands)
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing


Date:  Thu Mar  8 21:38:39 PST 2001
Workarea:  sshgate.corp.sgi.com:/export/d0/mkp/XFS/slinx-blk

The following file(s) were checked into:
  bonnie.engr.sgi.com:/isms/slinx/2.4.x-xfs


Modid:  2.4.x-xfs:slinx:89398a
cmd/xfsprogs/include/libxfs.h - 1.2
cmd/xfsprogs/repair/init.c - 1.2
cmd/xfsprogs/mkfs/xfs_mkfs.c - 1.2
cmd/xfsprogs/libxfs/init.c - 1.3
        - Only try to set block size when running mkfs or repair on an
          unmounted device.

-- 
Martin K. Petersen, Principal Linux Consultant, Linuxcare, Inc.
mkp@linuxcare.com, http://www.linuxcare.com/
SGI XFS for Linux Developer, http://oss.sgi.com/projects/xfs/

From owner-linux-xfs@oss.sgi.com Fri Mar  9 02:24:57 2001
Received:  by oss.sgi.com id <S553956AbRCIKYs>;
	Fri, 9 Mar 2001 02:24:48 -0800
Received: from groucho.maths.monash.edu.au ([130.194.160.211]:44604 "EHLO
        groucho.maths.monash.edu.au") by oss.sgi.com with ESMTP
	id <S553784AbRCIKYi>; Fri, 9 Mar 2001 02:24:38 -0800
Received: (from rjh@localhost)
	by groucho.maths.monash.edu.au (8.8.8/8.8.8) id KAA00128
	for linux-xfs@oss.sgi.com; Fri, 9 Mar 2001 10:24:35 GMT
From:   Robin Humble <rjh@groucho.maths.monash.edu.au>
Message-Id: <200103091024.KAA00128@groucho.maths.monash.edu.au>
Subject: Re: TAKE - nfs reference cache version 2
To:     linux-xfs@oss.sgi.com
Date:   Fri, 9 Mar 2001 21:24:34 +1100 (EDT)
In-Reply-To: <200103062108.f26L8f907105@jen.americas.sgi.com> from "Steve Lord" at Mar 06, 2001 03:08:41 PM
X-Mailer: ELM [version 2.5 PL1]
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing


I re-ran the bonnie++ over NFS and dd over NFS tests with the fixed
kernel and the results are good. The system is as before: uni-processor,
128M ram, no RAID:

  time dd if=/dev/zero of=bigFile bs=1024k count=??
 write size (MByte)   throughput    time
  same as count=      (MBytes/s)
  xfs   50               8.6        5.8s
   "   100               6.8        14.8s
   "   200               7.7        26s
   "   400               7.8        51s
   "   800               7.5       1min 47s

  ext2 800               8.6       1min 33s

So that's pretty close to the ext2 speeds. Massively better than before.

and bonnie++
Version 1.00h     ------Sequential Output------ --Sequential Input- --Random-
                  -Per Chr- --Block-- -Rewrite- -Per Chr- --Block-- --Seeks--
Machine      Size K/sec %CP K/sec %CP K/sec %CP K/sec %CP K/sec %CP  /sec %CP
NFS to xfs    64M  3277  72  4518  27  2518  19  5046  99 +++++ +++ 1254.9  28
NFS to ext2   64M  3190  69  4989  28  2471  21  5048  99 +++++ +++ 1255.6  28
NFS to xfs   256M  3272  72  4660  26  8078  35  4983  99 104856  98 1222.9  3
NFS to ext2  256M  2682  59  4952  28  8177  36  4978  99 103544  98 1224.1  31
                  ------Sequential Create------ --------Random Create--------
                  -Create-- --Read--- -Delete-- -Create-- --Read--- -Delete--
            files  /sec %CP  /sec %CP  /sec %CP  /sec %CP  /sec %CP  /sec %CP
NFS to xfs     16    78  76   213   1    95  53    25  91  1472  26    37  92
NFS to ext2    16    61  58   163   3    88  54    22  84    30  97    40  90

Looks fine.
The NFS block read results are obviously skewed 'cos the O2k box I
was testing from had 4G ram.
NFS + xfs over RAID0 with a SMP kernel also works just as well as
ext2 now. 512M of ram in the machine works without a slowdown also.


The one down side is that whilst testing these out I got a few weird
messages logged. Like this one sometimes when shutting down NFS
exports on the xfs box:
...
Mar  7 22:14:01 jersey portmap: portmap shutdown succeeded
Mar  7 22:14:08 jersey kernel: xfs_unmount: xfs_ibusy says error/16
Mar  7 22:14:08 jersey kernel: XFS unmount got error 16
Mar  7 22:14:08 jersey kernel: linvfs_put_super: vfsp/0xc1289320 left dangling!
Mar  7 22:14:08 jersey kernel: VFS: Busy inodes after unmount. Self-destruct in 5 seconds.  Have a nice day...

and seemingly at random-ish times I got this twice:

Mar  7 22:30:00 jersey kernel: kmem_cache_destroy: Can't free all objects c12f0890
Mar  7 22:30:00 jersey kernel: kmem_cache_destroy: Can't free all objects c12f0c14
Mar  7 22:30:00 jersey kernel: kmem_cache_destroy: Can't free all objects c12f0c78

and twice I got a 'kernel BUG' message when trying to mount an xfs
partition that had mounted fine just before:

# mount /scratch
mount: fs type xfs not supported by kernel
# lsmod
Module                  Size  Used by
xfs                   439408   1  (initializing)
xfs_support             9232   0  (autoclean) [xfs]
pagebuf                28352   1  (autoclean) [xfs]
3c59x                  23712   1  (autoclean)
agpgart                22560   0  (unused)

and I had to reboot before I can mount an xfs disk again. Apart from
that the machine apparently still worked ok. The message logged was:

Mar  7 20:43:27 jersey kernel: kernel BUG at slab.c:804!
Mar  7 20:43:27 jersey kernel: invalid operand: 0000
Mar  7 20:43:27 jersey kernel: CPU:    0
Mar  7 20:43:27 jersey kernel: EIP:    0010:[kmem_cache_create+889/960]
Mar  7 20:43:27 jersey kernel: EIP:    0010:[<c01288f1>]
Mar  7 20:43:27 jersey kernel: EFLAGS: 00010296
Mar  7 20:43:27 jersey kernel: eax: 0000001a   ebx: c12f08ec   ecx: c77e2000   edx: 00000001
Mar  7 20:43:27 jersey kernel: esi: c12f08e2   edi: c88a9e3f   ebp: c12f0ae8   esp: c170de48
Mar  7 20:43:27 jersey kernel: ds: 0018   es: 0018   ss: 0018
Mar  7 20:43:27 jersey kernel: Process modprobe (pid: 1043, stackpage=c170d000)
Mar  7 20:43:27 jersey kernel: Stack: c01f3f77 c01f4017 00000324 c8842000 c88a3d78 00000002 c88ad478 00000002
Mar  7 20:43:27 jersey kernel:        00000118 00000000 00000158 c170de80 c12f0afc 00000004 00000118 c880069b
Mar  7 20:43:27 jersey kernel:        c88a9e35 00000218 00000020 00000000 00000000 00000000 c889168b 00000218
Mar  7 20:43:27 jersey kernel: Call Trace: [3c59x:__insmod_3c59x_O/lib/modules/2.4.2-XFS/kernel/drivers/net/3+-1114112/ ...
...

once the Call Trace was related to the 3c59x module and the other time
it was for some serial module that I don't even use. Both BUG listings
were related to 'Process modprobe', and '[kmem_cache_create+889/960]'.
The register and Stack parts differed a bit. Let me know if you'd
like to see the full thing.

cheers,
robin

From owner-linux-xfs@oss.sgi.com Fri Mar  9 02:30:47 2001
Received:  by oss.sgi.com id <S553952AbRCIKah>;
	Fri, 9 Mar 2001 02:30:37 -0800
Received: from ppp0.ocs.com.au ([203.34.97.3]:34055 "HELO mail.ocs.com.au")
	by oss.sgi.com with SMTP id <S553777AbRCIKa0>;
	Fri, 9 Mar 2001 02:30:26 -0800
Received: (qmail 25379 invoked from network); 9 Mar 2001 10:30:20 -0000
Received: from ocs3.ocs-net (192.168.255.3)
  by mail.ocs.com.au with SMTP; 9 Mar 2001 10:30:20 -0000
X-Mailer: exmh version 2.1.1 10/15/1999
From:   Keith Owens <kaos@melbourne.sgi.com>
To:     Robin Humble <rjh@groucho.maths.monash.edu.au>
cc:     linux-xfs@oss.sgi.com
Subject: Re: TAKE - nfs reference cache version 2 
In-reply-to: Your message of "Fri, 09 Mar 2001 21:24:34 +1100."
             <200103091024.KAA00128@groucho.maths.monash.edu.au> 
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Date:   Fri, 09 Mar 2001 21:30:18 +1100
Message-ID: <30493.984133818@ocs3.ocs-net>
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing

On Fri, 9 Mar 2001 21:24:34 +1100 (EDT), 
Robin Humble <rjh@groucho.maths.monash.edu.au> wrote:
>Mar  7 20:43:27 jersey kernel: Call Trace: [3c59x:__insmod_3c59x_O/lib/modules/2.4.2-XFS/kernel/drivers/net/3+-1114112/ ...

Yet another broken trace from klogd.  ALways run klogd as "klogd -x" so
it leaves the trace alone, then run the oops through ksymoops.


From owner-linux-xfs@oss.sgi.com Fri Mar  9 06:43:50 2001
Received:  by oss.sgi.com id <S553958AbRCIOnk>;
	Fri, 9 Mar 2001 06:43:40 -0800
Received: from [141.154.95.3] ([141.154.95.3]:44535 "EHLO rain.ximian.com")
	by oss.sgi.com with ESMTP id <S553955AbRCIOnR>;
	Fri, 9 Mar 2001 06:43:17 -0800
Received: from localhost ([127.0.0.1])
	by rain.ximian.com with esmtp (Exim 3.12 #1 (Debian))
	id 14bO6X-0006xk-00
	for <linux-xfs@oss.sgi.com>; Fri, 09 Mar 2001 09:42:53 -0500
Subject: Re: kernel panic with nfsd and xfs filesystem
From:   Vladimir Vukicevic <vladimir@ximian.com>
To:     linux-xfs@oss.sgi.com
In-Reply-To: <200103072047.f27KlKB09885@jen.americas.sgi.com>
Content-Type: text/plain
X-Mailer: Evolution (0.9/+cvs.2001.03.07.21.20 - Preview Release)
Date:   09 Mar 2001 09:42:52 -0500
Mime-Version: 1.0
Message-Id: <E14bO6X-0006xk-00@rain.ximian.com>
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing

On 07 Mar 2001 14:47:20 -0600, Steve Lord wrote:
> 
> Hmm, there is nothing in the changed code which should affect anything
> but the write path in nfs. Have you tried a make clean and a complete
> rebuild - I have been having problems myself in that area. However, there
> have been a lot of changes since last friday.
> 
> This is pretty much a shot in the dark, but might make it go away.
> 


I just clean'd and recompiled, and things seem to be happy now.  We'll
see how it goes -- we're hitting this machine fairly well over NFS (one
100mbit ethernet interface, about 25 clients reading/writing avg. 60MB
files).  It seems to be noticably faster copying files to/from the
machine, and we aren't getting and I/O errors like we were with the
previous kernel..


    - Vlad



From owner-linux-xfs@oss.sgi.com Fri Mar  9 06:51:20 2001
Received:  by oss.sgi.com id <S553958AbRCIOvK>;
	Fri, 9 Mar 2001 06:51:10 -0800
Received: from sgi.SGI.COM ([192.48.153.1]:38219 "EHLO sgi.com")
	by oss.sgi.com with ESMTP id <S553955AbRCIOu7>;
	Fri, 9 Mar 2001 06:50:59 -0800
Received: from zeus-fddi.americas.sgi.com (128-162-8-103.americas.sgi.com [128.162.8.103]) 
	by sgi.com (980327.SGI.8.8.8-aspam/980304.SGI-aspam:
       SGI does not authorize the use of its proprietary
       systems or networks for unsolicited or bulk email
       from the Internet.) 
	via ESMTP id GAA06178
	for <linux-xfs@oss.sgi.com>; Fri, 9 Mar 2001 06:50:58 -0800 (PST)
	mail_from (lord@sgi.com)
Received: from daisy-e185.americas.sgi.com (daisy-e194.americas.sgi.com [128.162.194.214]) by zeus-fddi.americas.sgi.com (8.9.3/americas-smart-nospam1.1) with ESMTP id IAA1059885; Fri, 9 Mar 2001 08:49:41 -0600 (CST)
Received: from jen.americas.sgi.com (IDENT:root@jen.americas.sgi.com [128.162.184.86]) by daisy-e185.americas.sgi.com (SGI-8.9.3/SGI-server-1.7) with ESMTP id IAA64489; Fri, 9 Mar 2001 08:49:41 -0600 (CST)
Received: from jen.americas.sgi.com by jen.americas.sgi.com (8.11.0/SGI-client-1.7) via ESMTP id f29Ep9Q15552; Fri, 9 Mar 2001 08:51:09 -0600
Message-Id: <200103091451.f29Ep9Q15552@jen.americas.sgi.com>
X-Mailer: exmh version 2.2 06/23/2000 with nmh-1.0.4
To:     Vladimir Vukicevic <vladimir@ximian.com>
cc:     linux-xfs@oss.sgi.com
Subject: Re: kernel panic with nfsd and xfs filesystem 
In-Reply-To: Message from Vladimir Vukicevic <vladimir@ximian.com> 
   of "09 Mar 2001 09:42:52 EST." <E14bO6X-0006xk-00@rain.ximian.com> 
Date:   Fri, 09 Mar 2001 08:51:09 -0600
From:   Steve Lord <lord@sgi.com>
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing


This is good to hear - any chance you can tell the world what Ximian is
doing with xfs (would been cool to be underneath red-carpet)?

Steve

> On 07 Mar 2001 14:47:20 -0600, Steve Lord wrote:
> > 
> > Hmm, there is nothing in the changed code which should affect anything
> > but the write path in nfs. Have you tried a make clean and a complete
> > rebuild - I have been having problems myself in that area. However, there
> > have been a lot of changes since last friday.
> > 
> > This is pretty much a shot in the dark, but might make it go away.
> > 
> 
> 
> I just clean'd and recompiled, and things seem to be happy now.  We'll
> see how it goes -- we're hitting this machine fairly well over NFS (one
> 100mbit ethernet interface, about 25 clients reading/writing avg. 60MB
> files).  It seems to be noticably faster copying files to/from the
> machine, and we aren't getting and I/O errors like we were with the
> previous kernel..
> 
> 
>     - Vlad
> 



From owner-linux-xfs@oss.sgi.com Fri Mar  9 07:07:20 2001
Received:  by oss.sgi.com id <S553958AbRCIPHL>;
	Fri, 9 Mar 2001 07:07:11 -0800
Received: from [141.154.95.3] ([141.154.95.3]:49658 "EHLO rain.ximian.com")
	by oss.sgi.com with ESMTP id <S553955AbRCIPGx>;
	Fri, 9 Mar 2001 07:06:53 -0800
Received: from localhost ([127.0.0.1])
	by rain.ximian.com with esmtp (Exim 3.12 #1 (Debian))
	id 14bOTT-0006yo-00; Fri, 09 Mar 2001 10:06:35 -0500
Subject: Re: kernel panic with nfsd and xfs filesystem
From:   Vladimir Vukicevic <vladimir@ximian.com>
To:     Steve Lord <lord@sgi.com>
Cc:     linux-xfs@oss.sgi.com
In-Reply-To: <200103091451.f29Ep9Q15552@jen.americas.sgi.com>
Content-Type: text/plain
X-Mailer: Evolution (0.9/+cvs.2001.03.07.21.20 - Preview Release)
Date:   09 Mar 2001 10:06:35 -0500
Mime-Version: 1.0
Message-Id: <E14bOTT-0006yo-00@rain.ximian.com>
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing

On 09 Mar 2001 08:51:09 -0600, Steve Lord wrote:
> 
> This is good to hear - any chance you can tell the world what Ximian is
> doing with xfs (would been cool to be underneath red-carpet)?
> 
> Steve


Well, it's currently underneath two of the Red Carpet developers'
machines :-) (Mine and Ian's)  Right now, we're evaluating XFS by using
it on the server for our massively parallel CD ripping and encoding
project (for testing purposes only, of course). We'd like to eventually
move to XFS on both the Red Carpet servers and on our internal build
machines.

With the NFS case, we're still getting ocassional I/O Errors from cp (cp
-r)... when an md5sum was attempted on the file in question, md5sum just
bombed out claiming that it could not read the file in question;
however, no kernel messages/oops's on the server.


    - Vlad



From owner-linux-xfs@oss.sgi.com Fri Mar  9 07:10:11 2001
Received:  by oss.sgi.com id <S553958AbRCIPKB>;
	Fri, 9 Mar 2001 07:10:01 -0800
Received: from esparrall.udg.es ([130.206.124.16]:12160 "EHLO esparrall.udg.es")
	by oss.sgi.com with ESMTP id <S553955AbRCIPJy>;
	Fri, 9 Mar 2001 07:09:54 -0800
Received: from gcs by esparrall.udg.es with local (Exim 3.22 #1 (Debian))
	id 14bOVz-0001nh-00
	for <linux-xfs@oss.sgi.com>; Fri, 09 Mar 2001 16:09:11 +0100
Date:   Fri, 9 Mar 2001 16:09:11 +0100
From:   GCS <gcs@agentsinside.com>
To:     linux-xfs@oss.sgi.com
Subject: Local performance
Message-ID: <20010309160911.A5191@esparrall.udg.es>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.3.15i
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing

Hello,

I have installed a kernel based on the cvs on this morning with gcc 2.95.3
(20010219) and the loop fix patch applied (previously I was standing in
/usr/src/linux instead of ~/cvs/linux-2.4-xfs that's why I could not apply it).
The loop fix fixed my crash 'xfs over loopback'. The performance for a
PIII/666Mhz 320Mb RAM, 20Gb Seagate HDD:
Bonnie
              -------Sequential Output-------- ---Sequential Input-- --Random--
              -Per Char- --Block--- -Rewrite-- -Per Char- --Block--- --Seeks---
Machine    MB K/sec %CPU K/sec %CPU K/sec %CPU K/sec %CPU K/sec %CPU  /sec %CPU
       1* 900  8113 92.5 19879 33.1  4735  3.6  7818 83.2 21999  9.2  95.3  1.0 

Bonnie++
Version  1.00h      ------Sequential Output------ --Sequential Input- --Random-
                    -Per Chr- --Block-- -Rewrite- -Per Chr- --Block-- --Seeks--
Machine        Size K/sec %CP K/sec %CP K/sec %CP K/sec %CP K/sec %CP  /sec %CP
esparrall      900M  7194  89 21480  31  4957   8  5250  58 27062  11  75.3   0
                    ------Sequential Create------ --------Random Create--------
                    -Create-- --Read--- -Delete-- -Create-- --Read--- -Delete--
files:max:min        /sec %CP  /sec %CP  /sec %CP  /sec %CP  /sec %CP  /sec %CP
esparrall        16  2974  76 +++++ +++  3649  82  2755  72 +++++ +++  1759  45

Bye, GCS

From owner-linux-xfs@oss.sgi.com Fri Mar  9 07:16:11 2001
Received:  by oss.sgi.com id <S553958AbRCIPQB>;
	Fri, 9 Mar 2001 07:16:01 -0800
Received: from pneumatic-tube.sgi.com ([204.94.214.22]:48938 "EHLO
        pneumatic-tube.sgi.com") by oss.sgi.com with ESMTP
	id <S553955AbRCIPP5>; Fri, 9 Mar 2001 07:15:57 -0800
Received: from zeus-fddi.americas.sgi.com (zeus-fddi.americas.sgi.com [128.162.8.103]) by pneumatic-tube.sgi.com (980327.SGI.8.8.8-aspam/980310.SGI-aspam) via ESMTP id HAA02219
	for <linux-xfs@oss.sgi.com>; Fri, 9 Mar 2001 07:25:43 -0800 (PST)
	mail_from (lord@sgi.com)
Received: from daisy-e185.americas.sgi.com (daisy-e194.americas.sgi.com [128.162.194.214]) by zeus-fddi.americas.sgi.com (8.9.3/americas-smart-nospam1.1) with ESMTP id JAA1076047; Fri, 9 Mar 2001 09:14:40 -0600 (CST)
Received: from jen.americas.sgi.com (IDENT:root@jen.americas.sgi.com [128.162.184.86]) by daisy-e185.americas.sgi.com (SGI-8.9.3/SGI-server-1.7) with ESMTP id JAA07861; Fri, 9 Mar 2001 09:14:40 -0600 (CST)
Received: from jen.americas.sgi.com by jen.americas.sgi.com (8.11.0/SGI-client-1.7) via ESMTP id f29FG8R16362; Fri, 9 Mar 2001 09:16:08 -0600
Message-Id: <200103091516.f29FG8R16362@jen.americas.sgi.com>
X-Mailer: exmh version 2.2 06/23/2000 with nmh-1.0.4
To:     GCS <gcs@agentsinside.com>
cc:     linux-xfs@oss.sgi.com
Subject: Re: Local performance 
In-Reply-To: Message from GCS <gcs@agentsinside.com> 
   of "Fri, 09 Mar 2001 16:09:11 +0100." <20010309160911.A5191@esparrall.udg.es> 
Date:   Fri, 09 Mar 2001 09:16:08 -0600
From:   Steve Lord <lord@sgi.com>
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing

> Hello,
> 
> I have installed a kernel based on the cvs on this morning with gcc 2.95.3
> (20010219) and the loop fix patch applied (previously I was standing in
> /usr/src/linux instead of ~/cvs/linux-2.4-xfs that's why I could not apply it
> ).
> The loop fix fixed my crash 'xfs over loopback'. The performance for a
> PIII/666Mhz 320Mb RAM, 20Gb Seagate HDD:
> Bonnie
>               -------Sequential Output-------- ---Sequential Input-- --Random
> --
>               -Per Char- --Block--- -Rewrite-- -Per Char- --Block--- --Seeks-
> --
> Machine    MB K/sec %CPU K/sec %CPU K/sec %CPU K/sec %CPU K/sec %CPU  /sec %C
> PU
>        1* 900  8113 92.5 19879 33.1  4735  3.6  7818 83.2 21999  9.2  95.3  1
> .0 
> 
> Bonnie++
> Version  1.00h      ------Sequential Output------ --Sequential Input- --Rando
> m-
>                     -Per Chr- --Block-- -Rewrite- -Per Chr- --Block-- --Seeks
> --
> Machine        Size K/sec %CP K/sec %CP K/sec %CP K/sec %CP K/sec %CP  /sec %
> CP
> esparrall      900M  7194  89 21480  31  4957   8  5250  58 27062  11  75.3  
>  0
>                     ------Sequential Create------ --------Random Create------
> --
>                     -Create-- --Read--- -Delete-- -Create-- --Read--- -Delete
> --
> files:max:min        /sec %CP  /sec %CP  /sec %CP  /sec %CP  /sec %CP  /sec %
> CP
> esparrall        16  2974  76 +++++ +++  3649  82  2755  72 +++++ +++  1759  
> 45
> 
> Bye, GCS

Are these xfs over loop numbers?  They are deceptive, since xfs thinks it is
writing to a device and when I/O completion comes back on a write it has
really gone to the device. In the loop case it has not gone to the device
it has gone to the buffers on the underlying file. For me xfs over loop in
an xfs filesystem is faster than the underlying xfs filesystem because
of this.

Steve

p.s. a fairly major change to the read path should arrive today, it should
not affect bonnie, but it will affect larger reads.




From owner-linux-xfs@oss.sgi.com Fri Mar  9 07:18:20 2001
Received:  by oss.sgi.com id <S553958AbRCIPSL>;
	Fri, 9 Mar 2001 07:18:11 -0800
Received: from pneumatic-tube.sgi.com ([204.94.214.22]:61994 "EHLO
        pneumatic-tube.sgi.com") by oss.sgi.com with ESMTP
	id <S553955AbRCIPSH>; Fri, 9 Mar 2001 07:18:07 -0800
Received: from zeus-fddi.americas.sgi.com (zeus-fddi.americas.sgi.com [128.162.8.103]) by pneumatic-tube.sgi.com (980327.SGI.8.8.8-aspam/980310.SGI-aspam) via ESMTP id HAA02459
	for <linux-xfs@oss.sgi.com>; Fri, 9 Mar 2001 07:27:53 -0800 (PST)
	mail_from (lord@sgi.com)
Received: from daisy-e185.americas.sgi.com (daisy-e194.americas.sgi.com [128.162.194.214]) by zeus-fddi.americas.sgi.com (8.9.3/americas-smart-nospam1.1) with ESMTP id JAA1073447; Fri, 9 Mar 2001 09:16:50 -0600 (CST)
Received: from jen.americas.sgi.com (IDENT:root@jen.americas.sgi.com [128.162.184.86]) by daisy-e185.americas.sgi.com (SGI-8.9.3/SGI-server-1.7) with ESMTP id JAA75357; Fri, 9 Mar 2001 09:16:49 -0600 (CST)
Received: from jen.americas.sgi.com by jen.americas.sgi.com (8.11.0/SGI-client-1.7) via ESMTP id f29FIHw16372; Fri, 9 Mar 2001 09:18:17 -0600
Message-Id: <200103091518.f29FIHw16372@jen.americas.sgi.com>
X-Mailer: exmh version 2.2 06/23/2000 with nmh-1.0.4
To:     Vladimir Vukicevic <vladimir@ximian.com>
cc:     Steve Lord <lord@sgi.com>, linux-xfs@oss.sgi.com
Subject: Re: kernel panic with nfsd and xfs filesystem 
In-Reply-To: Message from Vladimir Vukicevic <vladimir@ximian.com> 
   of "09 Mar 2001 10:06:35 EST." <E14bOTT-0006yo-00@rain.ximian.com> 
Date:   Fri, 09 Mar 2001 09:18:17 -0600
From:   Steve Lord <lord@sgi.com>
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing

> On 09 Mar 2001 08:51:09 -0600, Steve Lord wrote:
> > 
> > This is good to hear - any chance you can tell the world what Ximian is
> > doing with xfs (would been cool to be underneath red-carpet)?
> > 
> > Steve
> 
> 
> Well, it's currently underneath two of the Red Carpet developers'
> machines :-) (Mine and Ian's)  Right now, we're evaluating XFS by using
> it on the server for our massively parallel CD ripping and encoding
> project (for testing purposes only, of course). We'd like to eventually
> move to XFS on both the Red Carpet servers and on our internal build
> machines.

Very cool, please keep us posted on how things are going for you.

> 
> With the NFS case, we're still getting ocassional I/O Errors from cp (cp
> -r)... when an md5sum was attempted on the file in question, md5sum just
> bombed out claiming that it could not read the file in question;
> however, no kernel messages/oops's on the server.
> 

Can you supply some more details on your setup (mount options, roughly
what you are doing in terms of how much data you are pushing when this
happens etc). We can try and replicate this here.

Thanks

   Steve


> 
>     - Vlad
> 



From owner-linux-xfs@oss.sgi.com Fri Mar  9 07:43:11 2001
Received:  by oss.sgi.com id <S553959AbRCIPnB>;
	Fri, 9 Mar 2001 07:43:01 -0800
Received: from tux.mkp.net ([130.225.60.11]:44551 "EHLO tux.mkp.net")
	by oss.sgi.com with ESMTP id <S553957AbRCIPmf>;
	Fri, 9 Mar 2001 07:42:35 -0800
Received: from tux.mkp.net ([130.225.60.11] helo=jaguar.mkp.net)
	by tux.mkp.net with esmtp (Exim 3.16 #1)
	id 14bP2F-0006jJ-00; Fri, 09 Mar 2001 16:42:33 +0100
Received: (from mkp@localhost)
	by jaguar.mkp.net (8.9.3/8.9.3) id KAA09465;
	Fri, 9 Mar 2001 10:41:24 -0500
X-Authentication-Warning: jaguar.mkp.net: mkp set sender to mkp@mkp.net using -f
To:     linux-xfs@oss.sgi.com
Subject: TAKE - Resolve ioctl conflicts
From:   "Martin K. Petersen" <mkp@linuxcare.com>
Organization: Linuxcare, Inc.
Date:   09 Mar 2001 10:41:24 -0500
Message-ID: <yq1d7brymor.fsf@jaguar.mkp.net>
Lines:  18
User-Agent: Gnus/5.0808 (Gnus v5.8.8) XEmacs/21.1 (Canyonlands)
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing


Date:  Fri Mar  9 07:40:51 PST 2001
Workarea:  sshgate.corp.sgi.com:/export/d0/mkp/XFS/slinx-pristine

The following file(s) were checked into:
  bonnie.engr.sgi.com:/isms/slinx/2.4.x-xfs


Modid:  2.4.x-xfs:slinx:89417a
linux/include/linux/fs.h - 1.84
cmd/xfsprogs/libxfs/init.c - 1.4
        - Bumped ioctl numbers for BLKBSZSET and BLKBSZGET to avoid
          conflicts with IA64 EFI ioctls.

-- 
Martin K. Petersen, Principal Linux Consultant, Linuxcare, Inc.
mkp@linuxcare.com, http://www.linuxcare.com/
SGI XFS for Linux Developer, http://oss.sgi.com/projects/xfs/

From owner-linux-xfs@oss.sgi.com Fri Mar  9 07:47:11 2001
Received:  by oss.sgi.com id <S553960AbRCIPrB>;
	Fri, 9 Mar 2001 07:47:01 -0800
Received: from sgi.SGI.COM ([192.48.153.1]:56673 "EHLO sgi.com")
	by oss.sgi.com with ESMTP id <S553957AbRCIPqs>;
	Fri, 9 Mar 2001 07:46:48 -0800
Received: from zeus-fddi.americas.sgi.com (128-162-8-103.americas.sgi.com [128.162.8.103]) 
	by sgi.com (980327.SGI.8.8.8-aspam/980304.SGI-aspam:
       SGI does not authorize the use of its proprietary
       systems or networks for unsolicited or bulk email
       from the Internet.) 
	via ESMTP id HAA05966
	for <linux-xfs@oss.sgi.com>; Fri, 9 Mar 2001 07:46:46 -0800 (PST)
	mail_from (lord@sgi.com)
Received: from daisy-e185.americas.sgi.com (daisy-e194.americas.sgi.com [128.162.194.214]) by zeus-fddi.americas.sgi.com (8.9.3/americas-smart-nospam1.1) with ESMTP id JAA1075027 for <linux-xfs@oss.sgi.com>; Fri, 9 Mar 2001 09:45:31 -0600 (CST)
Received: from jen.americas.sgi.com (IDENT:root@jen.americas.sgi.com [128.162.184.86]) by daisy-e185.americas.sgi.com (SGI-8.9.3/SGI-server-1.7) with ESMTP id JAA29599 for <linux-xfs@oss.sgi.com>; Fri, 9 Mar 2001 09:45:31 -0600 (CST)
Received: from jen.americas.sgi.com by jen.americas.sgi.com (8.11.0/SGI-client-1.7) via ESMTP id f29Fkwh18021; Fri, 9 Mar 2001 09:46:58 -0600
Message-Id: <200103091546.f29Fkwh18021@jen.americas.sgi.com>
X-Mailer: exmh version 2.2 06/23/2000 with nmh-1.0.4
To:     linux-xfs@oss.sgi.com
Subject: ALERT, XFS KERNEL API CHANGE
In-Reply-To: Message from "Martin K. Petersen" <mkp@linuxcare.com> 
   of "09 Mar 2001 10:41:24 EST." <yq1d7brymor.fsf@jaguar.mkp.net> 
Date:   Fri, 09 Mar 2001 09:46:58 -0600
From:   Steve Lord <lord@sgi.com>
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing


Note that this change will require a rebuild of the user space tools
before you can use them with the new kernel.

Steve


> 
> Date:  Fri Mar  9 07:40:51 PST 2001
> Workarea:  sshgate.corp.sgi.com:/export/d0/mkp/XFS/slinx-pristine
> 
> The following file(s) were checked into:
>   bonnie.engr.sgi.com:/isms/slinx/2.4.x-xfs
> 
> 
> Modid:  2.4.x-xfs:slinx:89417a
> linux/include/linux/fs.h - 1.84
> cmd/xfsprogs/libxfs/init.c - 1.4
>         - Bumped ioctl numbers for BLKBSZSET and BLKBSZGET to avoid
>           conflicts with IA64 EFI ioctls.
> 
> -- 
> Martin K. Petersen, Principal Linux Consultant, Linuxcare, Inc.
> mkp@linuxcare.com, http://www.linuxcare.com/
> SGI XFS for Linux Developer, http://oss.sgi.com/projects/xfs/



From owner-linux-xfs@oss.sgi.com Fri Mar  9 07:56:12 2001
Received:  by oss.sgi.com id <S553957AbRCIP4C>;
	Fri, 9 Mar 2001 07:56:02 -0800
Received: from tux.mkp.net ([130.225.60.11]:53255 "EHLO tux.mkp.net")
	by oss.sgi.com with ESMTP id <S553724AbRCIPzu>;
	Fri, 9 Mar 2001 07:55:50 -0800
Received: from tux.mkp.net ([130.225.60.11] helo=jaguar.mkp.net)
	by tux.mkp.net with esmtp (Exim 3.16 #1)
	id 14bPF5-0006k2-00; Fri, 09 Mar 2001 16:55:48 +0100
Received: (from mkp@localhost)
	by jaguar.mkp.net (8.9.3/8.9.3) id KAA09468;
	Fri, 9 Mar 2001 10:54:40 -0500
X-Authentication-Warning: jaguar.mkp.net: mkp set sender to mkp@mkp.net using -f
To:     linux-xfs@oss.sgi.com
Subject: IMPORTANT: I just broke the XFS userland utilities
From:   "Martin K. Petersen" <mkp@linuxcare.com>
Organization: Linuxcare, Inc.
Date:   09 Mar 2001 10:54:40 -0500
Message-ID: <yq18zmfym2n.fsf@jaguar.mkp.net>
Lines:  19
User-Agent: Gnus/5.0808 (Gnus v5.8.8) XEmacs/21.1 (Canyonlands)
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing


Turns out the IA64 folks had picked the same ioctls numbers as we had
for some block layer magic.  A few weeks ago I tried convincing them
to move their numbers to avoid conflicts with ours.

However, since I partially broke binary compatibility anyway yesterday
(by switching some of our own ioctls to the officially approved ones),
I have decided to give in and bump our numbers instead.

This means that if you are using the latest kernel from CVS (as of
now) you will have to upgrade your userland utilities before creating
new file systems.

Sorry about the inconvenience...

-- 
Martin K. Petersen, Principal Linux Consultant, Linuxcare, Inc.
mkp@linuxcare.com, http://www.linuxcare.com/
SGI XFS for Linux Developer, http://oss.sgi.com/projects/xfs/

From owner-linux-xfs@oss.sgi.com Fri Mar  9 07:58:42 2001
Received:  by oss.sgi.com id <S553957AbRCIP6c>;
	Fri, 9 Mar 2001 07:58:32 -0800
Received: from pneumatic-tube.sgi.com ([204.94.214.22]:6447 "EHLO
        pneumatic-tube.sgi.com") by oss.sgi.com with ESMTP
	id <S553724AbRCIP62>; Fri, 9 Mar 2001 07:58:28 -0800
Received: from zeus-fddi.americas.sgi.com (zeus-fddi.americas.sgi.com [128.162.8.103]) by pneumatic-tube.sgi.com (980327.SGI.8.8.8-aspam/980310.SGI-aspam) via ESMTP id IAA04626
	for <linux-xfs@oss.sgi.com>; Fri, 9 Mar 2001 08:08:14 -0800 (PST)
	mail_from (lord@sgi.com)
Received: from daisy-e185.americas.sgi.com (daisy-e194.americas.sgi.com [128.162.194.214]) by zeus-fddi.americas.sgi.com (8.9.3/americas-smart-nospam1.1) with ESMTP id JAA1078585 for <linux-xfs@oss.sgi.com>; Fri, 9 Mar 2001 09:57:11 -0600 (CST)
Received: from jen.americas.sgi.com (IDENT:root@jen.americas.sgi.com [128.162.184.86]) by daisy-e185.americas.sgi.com (SGI-8.9.3/SGI-server-1.7) with ESMTP id JAA96004 for <linux-xfs@oss.sgi.com>; Fri, 9 Mar 2001 09:57:11 -0600 (CST)
From:   Steve Lord <lord@sgi.com>
Received: by jen.americas.sgi.com (8.11.0/SGI-client-1.7) id f29FvA418908; Fri, 9 Mar 2001 09:57:10 -0600
Message-Id: <200103091557.f29FvA418908@jen.americas.sgi.com>
Date:   Fri, 9 Mar 2001 09:57:10 -0600
Subject: TAKE - More I/O path rework
To:     unlisted-recipients:; (no To-header on input)
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing

This bumps read speed for large I/Os (well 16K plus) by using
generic_file_read all the time - pagebuf based read is gone
except in the direct case. Also includes some fixes for direct
write.

Date:  Fri Mar  9 07:55:37 PST 2001
Workarea:  jen.americas.sgi.com:/src/lord/xfs-linux.2.4

The following file(s) were checked into:
  bonnie.engr.sgi.com:/isms/slinx/2.4.x-xfs


Modid:  2.4.x-xfs:slinx:89419a
linux/mm/filemap.c - 1.71
	- remove pagebuf code from do_generic_file_read

linux/include/linux/fs.h - 1.85
	- get pagebuf out of the inode ops business.

linux/fs/xfs/linux/xfs_lrw.c - 1.81
	- Buffered read calls generic_file_read, fixes for write path and data
	  flushing.

linux/fs/xfs/linux/xfs_super.c - 1.111
	- remove pagebuf_ioinitiate it is a) not used and b) not defined any more

linux/fs/xfs/linux/xfs_iops.c - 1.96
	- Add linvfs methods to pass linvfs bmap function into pagebuf.

linux/fs/xfs/linux/xfs_iops.h - 1.8
	- external for linvfs_pb_bmap

linux/include/linux/page_buf.h - 1.80
	- changed prototypes

linux/fs/pagebuf/page_buf.c - 1.63
	- Remove references to the pagebuf_bmap inode method, it is gone.

linux/fs/pagebuf/page_buf_io.c - 1.66
	- Obtain bmap function via function call parameters, do not get it from
	  the inode ops. The pagebuf read path is now only used by direct I/O.



From owner-linux-xfs@oss.sgi.com Fri Mar  9 08:00:12 2001
Received:  by oss.sgi.com id <S553957AbRCIQAD>;
	Fri, 9 Mar 2001 08:00:03 -0800
Received: from [141.154.95.3] ([141.154.95.3]:24311 "EHLO rain.ximian.com")
	by oss.sgi.com with ESMTP id <S553724AbRCIP7v>;
	Fri, 9 Mar 2001 07:59:51 -0800
Received: from localhost ([127.0.0.1])
	by rain.ximian.com with esmtp (Exim 3.12 #1 (Debian))
	id 14bPIk-00077i-00; Fri, 09 Mar 2001 10:59:34 -0500
Subject: Re: kernel panic with nfsd and xfs filesystem
From:   Vladimir Vukicevic <vladimir@ximian.com>
To:     Steve Lord <lord@sgi.com>
Cc:     linux-xfs@oss.sgi.com
In-Reply-To: <200103091518.f29FIHw16372@jen.americas.sgi.com>
Content-Type: text/plain
X-Mailer: Evolution (0.9/+cvs.2001.03.07.21.20 - Preview Release)
Date:   09 Mar 2001 10:59:33 -0500
Mime-Version: 1.0
Message-Id: <E14bPIk-00077i-00@rain.ximian.com>
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing

On 09 Mar 2001 09:18:17 -0600, Steve Lord wrote:

> Can you supply some more details on your setup (mount options, roughly
> what you are doing in terms of how much data you are pushing when this
> happens etc). We can try and replicate this here.


The server is a Dual-300 MHz Pentium II on an intel 440BX-based
motherboard.  There is an Adaptec 2940UW SCSI card, on which / is
(Seagate ST34501W).  Root is currently ext2.

The onboard PIIX4 IDE controller has:

hda: IBM-DTLA-307075, ATA DISK drive
hdc: ATAPI CD ROM DRIVE 50X MAX, ATAPI CD/DVD-ROM drive
ide0 at 0x1f0-0x1f7,0x3f6 on irq 14
ide1 at 0x170-0x177,0x376 on irq 15
hda: 150136560 sectors (76870 MB) w/1916KiB Cache, CHS=9345/255/63,
UDMA(33)
hdc: ATAPI 50X CD-ROM drive, 128kB Cache, UDMA(33)

/dev/hda1 is formatted to be one big XFS partition and is exported via
NFS, v1 or v2 only. (We had previous problems with v3 on the clients, so
decided to just stick to v2.)  NFS is from Red Hat 7.0
nfs-utils-0.1.9.1-7 -- the system is basically stock RH7 with all of the
errata applied (via Red Carpet, of course ;-), and then a kernel built
from linux-2.4-xfs CVS as of this morning.

The directory is exported "*(rw,no_root_squash)" (it's on an internal
subnet), rpc.mountd is running with --no-nfs-version 3, and we have the
kernel nfsd running.  Hmm.  Kernel nfsd is indeed running.  Maybe I'll
try this with userspace nfsd, yes?  Dunno if that would have an effect
though...

The client is running roughly the same version of the XFS kernel and is
just doing a 'cp -r' of a directory with roughly 12 or so files, total
70MB.  The last file is often the one that gets an IO error. This cp is
happening while a few other clients are reading/writing uncompressed
.wav's to/from the nfs drive. Also, if the files are copied over via
tar, it can apparently succeed, but tar has to recover from some errors.
Ian might be able to provide more information when he sees this.

Hope this helps; I'll keep poking at it on our end, and I will give
userspace nfsd a try as well (and turn on debugging perhaps).

    - Vlad



From owner-linux-xfs@oss.sgi.com Fri Mar  9 08:46:42 2001
Received:  by oss.sgi.com id <S553957AbRCIQqd>;
	Fri, 9 Mar 2001 08:46:33 -0800
Received: from [141.154.95.3] ([141.154.95.3]:46831 "EHLO rain.ximian.com")
	by oss.sgi.com with ESMTP id <S553724AbRCIQqT>;
	Fri, 9 Mar 2001 08:46:19 -0800
Received: from localhost ([127.0.0.1])
	by rain.ximian.com with esmtp (Exim 3.12 #1 (Debian))
	id 14bQ1Y-0007cM-00; Fri, 09 Mar 2001 11:45:52 -0500
Subject: Re: kernel panic with nfsd and xfs filesystem
From:   Vladimir Vukicevic <vladimir@ximian.com>
To:     Steve Lord <lord@sgi.com>
Cc:     linux-xfs@oss.sgi.com
In-Reply-To: <200103091518.f29FIHw16372@jen.americas.sgi.com>
Content-Type: text/plain
X-Mailer: Evolution (0.9/+cvs.2001.03.07.21.20 - Preview Release)
Date:   09 Mar 2001 11:45:51 -0500
Mime-Version: 1.0
Message-Id: <E14bQ1Y-0007cM-00@rain.ximian.com>
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing



Hmm.  I now have a repeatable case of this I/O error, but as the other
end is running the kernel nfsd server, I'm not exactly sure how to debug
it or turn on debugging on the other end... A 'cat foo.ogg /dev/null' on
the mounted partition repeatably gives an I/O error.

Any thoughts on how to diagnose this?  I'll keep poking..


    - Vlad



From owner-linux-xfs@oss.sgi.com Fri Mar  9 09:19:22 2001
Received:  by oss.sgi.com id <S553960AbRCIRTN>;
	Fri, 9 Mar 2001 09:19:13 -0800
Received: from [141.154.95.3] ([141.154.95.3]:60148 "EHLO rain.ximian.com")
	by oss.sgi.com with ESMTP id <S553790AbRCIRS4>;
	Fri, 9 Mar 2001 09:18:56 -0800
Received: from localhost ([127.0.0.1])
	by rain.ximian.com with esmtp (Exim 3.12 #1 (Debian))
	id 14bQXC-0007hT-00; Fri, 09 Mar 2001 12:18:34 -0500
Subject: Re: kernel panic with nfsd and xfs filesystem
From:   Vladimir Vukicevic <vladimir@ximian.com>
To:     Vladimir Vukicevic <vladimir@ximian.com>
Cc:     Steve Lord <lord@sgi.com>, linux-xfs@oss.sgi.com
In-Reply-To: <E14bQ1Y-0007cM-00@rain.ximian.com>
Content-Type: text/plain
X-Mailer: Evolution (0.9/+cvs.2001.03.07.21.20 - Preview Release)
Date:   09 Mar 2001 12:18:34 -0500
Mime-Version: 1.0
Message-Id: <E14bQXC-0007hT-00@rain.ximian.com>
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing

On 09 Mar 2001 11:45:51 -0500, Vladimir Vukicevic wrote:
> 
> 
> Hmm.  I now have a repeatable case of this I/O error, but as the other
> end is running the kernel nfsd server, I'm not exactly sure how to debug
> it or turn on debugging on the other end... A 'cat foo.ogg /dev/null' on
> the mounted partition repeatably gives an I/O error.
> 
> Any thoughts on how to diagnose this?  I'll keep poking..


Doh. Forgot about tcpdump. :-)  So, this is what I'm seeing:

11:57:46.159530 rain.ximian.priv.4040915765 > ogg.nfs: 140 lookup fh
Unknown/1 "01-letters_from_the_wasteland.ogg" (DF)
11:57:46.163988 ogg.nfs > rain.ximian.priv.4040915765: reply ok 128
lookup fh Unknown/1 (DF)

So, the lookup goes okay. Then the weirdness starts.

11:57:46.173585 rain.ximian.priv.4057692981 > ogg.nfs: 112 read fh
Unknown/1 4096 bytes @ 2408448 (DF)
11:57:46.217986 ogg > rain.ximian.priv: (frag 20288:1480@1480+)
11:57:46.220115 ogg.nfs > rain.ximian.priv.4057692981: reply ok 1472
read (frag 20288:1480@0+)

Looking at this more in ethereal, this call/reply sequence has XID
0xf0db7b35. It appears to succeed (although I'm confused why it's
reading @ offset 2408448). However, the next 3 calls have the exact same
XID (marked as dup's in ethereal), and it's reading same size/offset.

11:57:46.865276 rain.ximian.priv.4057692981 > ogg.nfs: 112 read fh
Unknown/1 4096 bytes @ 2408448 (DF)
11:57:46.907584 ogg > rain.ximian.priv: (frag 20544:1480@1480+)
11:57:46.909674 ogg.nfs > rain.ximian.priv.4057692981: reply ok 1472
read (frag 20544:1480@0+)

11:57:48.265276 rain.ximian.priv.4057692981 > ogg.nfs: 112 read fh
Unknown/1 4096 bytes @ 2408448 (DF)
11:57:48.319183 ogg > rain.ximian.priv: (frag 20800:1480@1480+)
11:57:48.321456 ogg.nfs > rain.ximian.priv.4057692981: reply ok 1472
read (frag 20800:1480@0+)

11:57:51.065342 rain.ximian.priv.4057692981 > ogg.nfs: 112 read fh
Unknown/1 4096 bytes @ 2408448 (DF)
11:57:51.114002 ogg > rain.ximian.priv: (frag 21056:1480@1480+)
11:57:51.115317 ogg.nfs > rain.ximian.priv.4057692981: reply ok 1472
read (frag 21056:1480@0+)

Then, it switches to a new XID, and the same set repeats itself.

11:57:56.666020 rain.ximian.priv.4074470197 > ogg.nfs: 112 read fh
Unknown/1 4096 bytes @ 2408448 (DF)
11:57:56.720428 ogg > rain.ximian.priv: (frag 21312:1480@1480+)
11:57:56.722696 ogg.nfs > rain.ximian.priv.4074470197: reply ok 1472
read (frag 21312:1480@0+)

11:57:57.365349 rain.ximian.priv.4074470197 > ogg.nfs: 112 read fh
Unknown/1 4096 bytes @ 2408448 (DF)
11:57:57.411418 ogg > rain.ximian.priv: (frag 21568:1480@1480+)
11:57:57.413422 ogg.nfs > rain.ximian.priv.4074470197: reply ok 1472
read (frag 21568:1480@0+)

11:57:58.765279 rain.ximian.priv.4074470197 > ogg.nfs: 112 read fh
Unknown/1 4096 bytes @ 2408448 (DF)
11:57:58.814856 ogg > rain.ximian.priv: (frag 21824:1480@1480+)
11:57:58.817140 ogg.nfs > rain.ximian.priv.4074470197: reply ok 1472
read (frag 21824:1480@0+)

11:58:01.565281 rain.ximian.priv.4074470197 > ogg.nfs: 112 read fh
Unknown/1 4096 bytes @ 2408448 (DF)
11:58:01.617302 ogg > rain.ximian.priv: (frag 22080:1480@1480+)
11:58:01.619220 ogg.nfs > rain.ximian.priv.4074470197: reply ok 1472
read (frag 22080:1480@0+)


... and then cp dies with an I/O error.

So, from looking at this, I'm going to blame the client side NFS stuff
here -- especially since the file is perfectly fine on the server
itself.  Alan Cox said that he wasn't aware of any nfs client-side
patches that have gone in since 2.4.2 came out. 

Note that this is actually a similar error to what I was seeing while
running my iPAQ with a NFS'd root filesystem -- certain files just give
I/O errors.  This is one of them; I recreated by copying the file to an
ext2 filesystem and getting the same I/O error.

So, this isn't XFS related (whew!), but nfs on linux does indeed suck.
:-P


    - Vlad



From owner-linux-xfs@oss.sgi.com Fri Mar  9 09:32:43 2001
Received:  by oss.sgi.com id <S553960AbRCIRcX>;
	Fri, 9 Mar 2001 09:32:23 -0800
Received: from deliverator.sgi.com ([204.94.214.10]:53092 "EHLO
        deliverator.sgi.com") by oss.sgi.com with ESMTP id <S553788AbRCIRcS>;
	Fri, 9 Mar 2001 09:32:18 -0800
Received: from zeus-fddi.americas.sgi.com (zeus-fddi.americas.sgi.com [128.162.8.103]) by deliverator.sgi.com (980309.SGI.8.8.8-aspam-6.2/980310.SGI-aspam) via ESMTP id JAA08868
	for <linux-xfs@oss.sgi.com>; Fri, 9 Mar 2001 09:31:00 -0800 (PST)
	mail_from (lord@sgi.com)
Received: from daisy-e185.americas.sgi.com (daisy-e194.americas.sgi.com [128.162.194.214]) by zeus-fddi.americas.sgi.com (8.9.3/americas-smart-nospam1.1) with ESMTP id LAA1079769; Fri, 9 Mar 2001 11:30:14 -0600 (CST)
Received: from jen.americas.sgi.com (IDENT:root@jen.americas.sgi.com [128.162.184.86]) by daisy-e185.americas.sgi.com (SGI-8.9.3/SGI-server-1.7) with ESMTP id LAA79680; Fri, 9 Mar 2001 11:30:14 -0600 (CST)
Received: from jen.americas.sgi.com by jen.americas.sgi.com (8.11.0/SGI-client-1.7) via ESMTP id f29HUC411306; Fri, 9 Mar 2001 11:30:12 -0600
Message-Id: <200103091730.f29HUC411306@jen.americas.sgi.com>
X-Mailer: exmh version 2.2 06/23/2000 with nmh-1.0.4
To:     Vladimir Vukicevic <vladimir@ximian.com>
cc:     Steve Lord <lord@sgi.com>, linux-xfs@oss.sgi.com
Subject: Re: kernel panic with nfsd and xfs filesystem 
In-Reply-To: Message from Vladimir Vukicevic <vladimir@ximian.com> 
   of "09 Mar 2001 12:18:34 EST." <E14bQXC-0007hT-00@rain.ximian.com> 
Date:   Fri, 09 Mar 2001 11:30:12 -0600
From:   Steve Lord <lord@sgi.com>
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing


Sorry for the silence on this end - got dragged into a conference call on
a totally different topic, sat and listened for over an hour.

Anyway, looks like you are willing to place the blame elsewhere,
I have not seen similar, but my NFS usage between linux boxes is not
huge. You mentioned switching to user space nfsd - the kernel version
should be OK from the XFS point of view now we have the speed problem
worked out - I still plan on going back and cleaning up what I did
there, sounds like there may be unmount problems later.

Not sure I can help with the NFS problem itself though.

Steve

> On 09 Mar 2001 11:45:51 -0500, Vladimir Vukicevic wrote:
> > 
> > 
> > Hmm.  I now have a repeatable case of this I/O error, but as the other
> > end is running the kernel nfsd server, I'm not exactly sure how to debug
> > it or turn on debugging on the other end... A 'cat foo.ogg /dev/null' on
> > the mounted partition repeatably gives an I/O error.
> > 
> > Any thoughts on how to diagnose this?  I'll keep poking..
> 
> 
> Doh. Forgot about tcpdump. :-)  So, this is what I'm seeing:
> 
> 11:57:46.159530 rain.ximian.priv.4040915765 > ogg.nfs: 140 lookup fh
> Unknown/1 "01-letters_from_the_wasteland.ogg" (DF)
> 11:57:46.163988 ogg.nfs > rain.ximian.priv.4040915765: reply ok 128
> lookup fh Unknown/1 (DF)
> 
> So, the lookup goes okay. Then the weirdness starts.
> 
> 11:57:46.173585 rain.ximian.priv.4057692981 > ogg.nfs: 112 read fh
> Unknown/1 4096 bytes @ 2408448 (DF)
> 11:57:46.217986 ogg > rain.ximian.priv: (frag 20288:1480@1480+)
> 11:57:46.220115 ogg.nfs > rain.ximian.priv.4057692981: reply ok 1472
> read (frag 20288:1480@0+)
> 
> Looking at this more in ethereal, this call/reply sequence has XID
> 0xf0db7b35. It appears to succeed (although I'm confused why it's
> reading @ offset 2408448). However, the next 3 calls have the exact same
> XID (marked as dup's in ethereal), and it's reading same size/offset.
> 
> 11:57:46.865276 rain.ximian.priv.4057692981 > ogg.nfs: 112 read fh
> Unknown/1 4096 bytes @ 2408448 (DF)
> 11:57:46.907584 ogg > rain.ximian.priv: (frag 20544:1480@1480+)
> 11:57:46.909674 ogg.nfs > rain.ximian.priv.4057692981: reply ok 1472
> read (frag 20544:1480@0+)
> 
> 11:57:48.265276 rain.ximian.priv.4057692981 > ogg.nfs: 112 read fh
> Unknown/1 4096 bytes @ 2408448 (DF)
> 11:57:48.319183 ogg > rain.ximian.priv: (frag 20800:1480@1480+)
> 11:57:48.321456 ogg.nfs > rain.ximian.priv.4057692981: reply ok 1472
> read (frag 20800:1480@0+)
> 
> 11:57:51.065342 rain.ximian.priv.4057692981 > ogg.nfs: 112 read fh
> Unknown/1 4096 bytes @ 2408448 (DF)
> 11:57:51.114002 ogg > rain.ximian.priv: (frag 21056:1480@1480+)
> 11:57:51.115317 ogg.nfs > rain.ximian.priv.4057692981: reply ok 1472
> read (frag 21056:1480@0+)
> 
> Then, it switches to a new XID, and the same set repeats itself.
> 
> 11:57:56.666020 rain.ximian.priv.4074470197 > ogg.nfs: 112 read fh
> Unknown/1 4096 bytes @ 2408448 (DF)
> 11:57:56.720428 ogg > rain.ximian.priv: (frag 21312:1480@1480+)
> 11:57:56.722696 ogg.nfs > rain.ximian.priv.4074470197: reply ok 1472
> read (frag 21312:1480@0+)
> 
> 11:57:57.365349 rain.ximian.priv.4074470197 > ogg.nfs: 112 read fh
> Unknown/1 4096 bytes @ 2408448 (DF)
> 11:57:57.411418 ogg > rain.ximian.priv: (frag 21568:1480@1480+)
> 11:57:57.413422 ogg.nfs > rain.ximian.priv.4074470197: reply ok 1472
> read (frag 21568:1480@0+)
> 
> 11:57:58.765279 rain.ximian.priv.4074470197 > ogg.nfs: 112 read fh
> Unknown/1 4096 bytes @ 2408448 (DF)
> 11:57:58.814856 ogg > rain.ximian.priv: (frag 21824:1480@1480+)
> 11:57:58.817140 ogg.nfs > rain.ximian.priv.4074470197: reply ok 1472
> read (frag 21824:1480@0+)
> 
> 11:58:01.565281 rain.ximian.priv.4074470197 > ogg.nfs: 112 read fh
> Unknown/1 4096 bytes @ 2408448 (DF)
> 11:58:01.617302 ogg > rain.ximian.priv: (frag 22080:1480@1480+)
> 11:58:01.619220 ogg.nfs > rain.ximian.priv.4074470197: reply ok 1472
> read (frag 22080:1480@0+)
> 
> 
> ... and then cp dies with an I/O error.
> 
> So, from looking at this, I'm going to blame the client side NFS stuff
> here -- especially since the file is perfectly fine on the server
> itself.  Alan Cox said that he wasn't aware of any nfs client-side
> patches that have gone in since 2.4.2 came out. 
> 
> Note that this is actually a similar error to what I was seeing while
> running my iPAQ with a NFS'd root filesystem -- certain files just give
> I/O errors.  This is one of them; I recreated by copying the file to an
> ext2 filesystem and getting the same I/O error.
> 
> So, this isn't XFS related (whew!), but nfs on linux does indeed suck.
> :-P
> 
> 
>     - Vlad
> 



From owner-linux-xfs@oss.sgi.com Fri Mar  9 10:59:03 2001
Received:  by oss.sgi.com id <S553773AbRCIS6n>;
	Fri, 9 Mar 2001 10:58:43 -0800
Received: from [207.198.250.232] ([207.198.250.232]:24595 "HELO perilith.com")
	by oss.sgi.com with SMTP id <S553769AbRCIS6l>;
	Fri, 9 Mar 2001 10:58:41 -0800
Received: by perilith.com (Postfix, from userid 1034)
	id C6DCE3E03D; Fri,  9 Mar 2001 13:58:39 -0500 (EST)
Date:   Fri, 9 Mar 2001 10:58:39 -0800
From:   Drew Bloechl <drew@cesspool.net>
To:     linux-xfs@oss.sgi.com
Subject: Re: Messages from last few days
Message-ID: <20010309105839.J31627@perilith.com>
References: <20010308181245.I31627@perilith.com> <200103090305.OAA45962@boing.melbourne.sgi.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.3.15i
In-Reply-To: <200103090305.OAA45962@boing.melbourne.sgi.com>; from tes@boing.melbourne.sgi.com on Fri, Mar 09, 2001 at 02:05:25PM +1100
X-PGP-Fingerprint: B8B4 8A05 A58B 5252 FFC5  E455 D831 31A3 3385 5516
X-PGP-Public-Key: http://cesspool.net/drew.pubkey.txt
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing

On Fri, Mar 09, 2001 at 02:05:25PM +1100, Timothy Shimmin wrote:
> You could look at the mailing-list archive for March:

I really was hoping for mbox so I can use mutt's searching/threading
functions rather than a web interface.  

-- 
Drew Bloechl
drew@cesspool.net
PGP key ID: 33855516

From owner-linux-xfs@oss.sgi.com Fri Mar  9 12:38:12 2001
Received:  by oss.sgi.com id <S554015AbRCIUhw>;
	Fri, 9 Mar 2001 12:37:52 -0800
Received: from ns.suse.de ([213.95.15.193]:26889 "HELO Cantor.suse.de")
	by oss.sgi.com with SMTP id <S554011AbRCIUhp>;
	Fri, 9 Mar 2001 12:37:45 -0800
Received: from Hermes.suse.de (Hermes.suse.de [213.95.15.136])
	by Cantor.suse.de (Postfix) with ESMTP
	id 788291E2AE; Fri,  9 Mar 2001 21:37:44 +0100 (MET)
Date:   Fri, 9 Mar 2001 21:37:37 +0100
From:   Andi Kleen <ak@suse.de>
To:     Vladimir Vukicevic <vladimir@ximian.com>
Cc:     Steve Lord <lord@sgi.com>, linux-xfs@oss.sgi.com
Subject: Re: kernel panic with nfsd and xfs filesystem
Message-ID: <20010309213737.A18494@gruyere.muc.suse.de>
References: <E14bQ1Y-0007cM-00@rain.ximian.com> <E14bQXC-0007hT-00@rain.ximian.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.2.5i
In-Reply-To: <E14bQXC-0007hT-00@rain.ximian.com>; from vladimir@ximian.com on Fri, Mar 09, 2001 at 12:18:34PM -0500
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing

On Fri, Mar 09, 2001 at 12:18:34PM -0500, Vladimir Vukicevic wrote:
> On 09 Mar 2001 11:45:51 -0500, Vladimir Vukicevic wrote:
> > 
> > 
> > Hmm.  I now have a repeatable case of this I/O error, but as the other
> > end is running the kernel nfsd server, I'm not exactly sure how to debug
> > it or turn on debugging on the other end... A 'cat foo.ogg /dev/null' on
> > the mounted partition repeatably gives an I/O error.
> > 
> > Any thoughts on how to diagnose this?  I'll keep poking..
> 
> 
> Doh. Forgot about tcpdump. :-)  So, this is what I'm seeing:
> 
> 11:57:46.159530 rain.ximian.priv.4040915765 > ogg.nfs: 140 lookup fh
> Unknown/1 "01-letters_from_the_wasteland.ogg" (DF)
> 11:57:46.163988 ogg.nfs > rain.ximian.priv.4040915765: reply ok 128
> lookup fh Unknown/1 (DF)
> 
> So, the lookup goes okay. Then the weirdness starts.

You're probably overflowing some of the too small default buffers and 
therefore get packet loss. You can check using netstat -s. If you see
IpReasmFails increase /proc/sys/net/ipv4/ipfrag_high_thresh
If you see UdpInErrs increase the socket buffer  (net/core/rmem_{default,max})
This will probably need a reload of the nfs modules to take effect.
If you see IpInErrs or device errors then you're out of luck in 2.4 currently.

Note that when you overflow the socket buffer you'll likely also not see
it in ethertap, because it tends to use the same too small socket buffer sizes
on its packet sniffing socket. The single fragments you saw probably point
to this.


-Andi


From owner-linux-xfs@oss.sgi.com Fri Mar  9 13:35:32 2001
Received:  by oss.sgi.com id <S554017AbRCIVfW>;
	Fri, 9 Mar 2001 13:35:22 -0800
Received: from deliverator.sgi.com ([204.94.214.10]:39743 "EHLO
        deliverator.sgi.com") by oss.sgi.com with ESMTP id <S553804AbRCIVfE>;
	Fri, 9 Mar 2001 13:35:04 -0800
Received: from snort.melbourne.sgi.com (snort.melbourne.sgi.com [134.14.55.149]) by deliverator.sgi.com (980309.SGI.8.8.8-aspam-6.2/980310.SGI-aspam) via ESMTP id NAA20362
	for <linux-xfs@oss.sgi.com>; Fri, 9 Mar 2001 13:33:56 -0800 (PST)
	mail_from (nathans@snort.melbourne.sgi.com)
Received: (from nathans@localhost)
	by snort.melbourne.sgi.com (SGI-8.9.3/8.9.3) id IAA66642
	for linux-xfs@oss.sgi.com; Sat, 10 Mar 2001 08:33:45 +1100 (EST)
Date:   Sat, 10 Mar 2001 08:33:45 +1100 (EST)
From:   Nathan Scott <nathans@snort.melbourne.sgi.com>
Message-Id: <200103092133.IAA66642@snort.melbourne.sgi.com>
To:     linux-xfs@oss.sgi.com
Subject: TAKE - version
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing

Date:  Fri Mar  9 13:32:37 PST 2001
Workarea:  snort.melbourne.sgi.com:/diskb/build4/nathans/linux-xfs

The following file(s) were checked into:
  bonnie.engr.sgi.com:/isms/slinx/2.4.x-xfs


Modid:  2.4.x-xfs:slinx:89465a
cmd/xfsprogs/doc/CHANGES - 1.5
	- document recent api changes in this rev.

cmd/xfsprogs/VERSION - 1.5
	- bump minor release number.



From owner-linux-xfs@oss.sgi.com Fri Mar  9 15:43:53 2001
Received:  by oss.sgi.com id <S553809AbRCIXnn>;
	Fri, 9 Mar 2001 15:43:43 -0800
Received: from pneumatic-tube.sgi.com ([204.94.214.22]:16241 "EHLO
        pneumatic-tube.sgi.com") by oss.sgi.com with ESMTP
	id <S553795AbRCIXnX>; Fri, 9 Mar 2001 15:43:23 -0800
Received: from zeus-fddi.americas.sgi.com (zeus-fddi.americas.sgi.com [128.162.8.103]) by pneumatic-tube.sgi.com (980327.SGI.8.8.8-aspam/980310.SGI-aspam) via ESMTP id PAA01928
	for <linux-xfs@oss.sgi.com>; Fri, 9 Mar 2001 15:53:09 -0800 (PST)
	mail_from (lord@sgi.com)
Received: from daisy-e185.americas.sgi.com (daisy-e194.americas.sgi.com [128.162.194.214]) by zeus-fddi.americas.sgi.com (8.9.3/americas-smart-nospam1.1) with ESMTP id RAA1083580 for <linux-xfs@oss.sgi.com>; Fri, 9 Mar 2001 17:42:06 -0600 (CST)
Received: from jen.americas.sgi.com (IDENT:root@jen.americas.sgi.com [128.162.184.86]) by daisy-e185.americas.sgi.com (SGI-8.9.3/SGI-server-1.7) with ESMTP id RAA99090 for <linux-xfs@oss.sgi.com>; Fri, 9 Mar 2001 17:42:05 -0600 (CST)
From:   Steve Lord <lord@sgi.com>
Received: by jen.americas.sgi.com (8.11.0/SGI-client-1.7) id f29Ni8K02927; Fri, 9 Mar 2001 17:44:08 -0600
Message-Id: <200103092344.f29Ni8K02927@jen.americas.sgi.com>
Date:   Fri, 9 Mar 2001 17:44:08 -0600
Subject: TAKE - fix unmount of xfs filesystem after nfs export
To:     unlisted-recipients:; (no To-header on input)
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing

Recent changes this week introduced some optimizations for nfs exporting
xfs, and this bug - unmounts would report a busy inode. This should 
fix it.

Date:  Fri Mar  9 15:41:04 PST 2001
Workarea:  jen.americas.sgi.com:/src/lord/xfs-linux.2.4

The following file(s) were checked into:
  bonnie.engr.sgi.com:/isms/slinx/2.4.x-xfs


Modid:  2.4.x-xfs:slinx:89479a
linux/fs/xfs/linux/xfs_lrw.c - 1.82
	- Fix xfs_write to correctly cleanup after reference cache overflow, fixes
	  unmount of xfs on nfs server.



From owner-linux-xfs@oss.sgi.com Fri Mar  9 15:50:42 2001
Received:  by oss.sgi.com id <S553809AbRCIXud>;
	Fri, 9 Mar 2001 15:50:33 -0800
Received: from cc19815-a.zwoll1.ov.nl.home.com ([212.204.138.247]:13040 "HELO
        wisdom.myplace.net") by oss.sgi.com with SMTP id <S553795AbRCIXuV>;
	Fri, 9 Mar 2001 15:50:21 -0800
Received: from ws1 (ws1.myplace.net [192.168.1.15])
	by wisdom.myplace.net (Postfix) with SMTP id B588040BBE
	for <linux-xfs@oss.sgi.com>; Sat, 10 Mar 2001 00:50:18 +0100 (CET)
Message-ID: <000a01c0a93f$22643440$0f01a8c0@myplace.net>
From:   "Bas" <list@showme.wox.org>
To:     <linux-xfs@oss.sgi.com>
Subject: 2.4.2, LVM 0.9.1 beta 5a and XFS.
Date:   Sat, 10 Mar 2001 00:50:18 -0800
MIME-Version: 1.0
Content-Type: text/plain;
	charset="iso-8859-1"
Content-Transfer-Encoding: 7bit
X-Priority: 3
X-MSMail-Priority: Normal
X-Mailer: Microsoft Outlook Express 5.00.2615.200
X-MimeOLE: Produced By Microsoft MimeOLE V5.00.2615.200
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing

Hi,

Finally got XFS running on my machine. I'm using it next to reiserfs which
is a little faster, but I believe XFS to be more stable.

I got the tree through CVS, and compiled it last week. When I extend the LV
that XFS is on, and I run xfs_growfs, it returns:

xfs_growfs: cannot determine geometry of filesystem mounted at /mnt: Invalid
argument

I didn't see any patches for LVM to use XFS, and on the XFS site it didn't
find anything to patch for XFS. How do I fix this ?

Thanks,
Bas



From owner-linux-xfs@oss.sgi.com Fri Mar  9 16:04:43 2001
Received:  by oss.sgi.com id <S553811AbRCJAEY>;
	Fri, 9 Mar 2001 16:04:24 -0800
Received: from pike.sover.net ([209.198.87.34]:64968 "EHLO pike.sover.net")
	by oss.sgi.com with ESMTP id <S553795AbRCJAET>;
	Fri, 9 Mar 2001 16:04:19 -0800
Received: from surreal.localdomain (pm1a30.stj.sover.net [209.198.94.30])
	by pike.sover.net (8.9.3/8.9.3) with ESMTP id TAA19731;
	Fri, 9 Mar 2001 19:04:01 -0500 (EST)
Comments: SoVerNet Verification (on pike.sover.net)
	surreal.localdomain from pm1a30.stj.sover.net [209.198.94.30] 209.198.94.30
	Fri, 9 Mar 2001 19:04:01 -0500 (EST)
Date:   Fri, 9 Mar 2001 19:04:53 -0500 (EST)
From:   Jason Walker <unseen@sover.net>
X-Sender: unseen@surreal.localdomain
To:     Bas <list@showme.wox.org>
cc:     linux-xfs@oss.sgi.com
Subject: Re: 2.4.2, LVM 0.9.1 beta 5a and XFS.
In-Reply-To: <000a01c0a93f$22643440$0f01a8c0@myplace.net>
Message-ID: <Pine.LNX.4.10.10103091902400.3576-100000@surreal.localdomain>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing


I use LVM as well, and I believ eyou need to aply the patches for LVM to
the kernel source. did you do that? there will be a PATCHES/ directory in
your LVM source. read the README in there, it says how to build the patch
against your source tree. pretty simple proccess. see if that helps you.

I am running LVM 0.9.1beta5, btw

RegEx

On Sat, 10 Mar 2001, Bas wrote:

> Hi,
> 
> Finally got XFS running on my machine. I'm using it next to reiserfs which
> is a little faster, but I believe XFS to be more stable.
> 
> I got the tree through CVS, and compiled it last week. When I extend the LV
> that XFS is on, and I run xfs_growfs, it returns:
> 
> xfs_growfs: cannot determine geometry of filesystem mounted at /mnt: Invalid
> argument
> 
> I didn't see any patches for LVM to use XFS, and on the XFS site it didn't
> find anything to patch for XFS. How do I fix this ?
> 
> Thanks,
> Bas
> 

-- 
Jason Walker -- unseen@sover.net, perlgod@hotmail.com
UIN: 110493687
AIM: Nightface

/"\
\ /     ASCII RIBBON CAMPAIGN
 X        AGAINST HTML MAIL
/ \
			 


From owner-linux-xfs@oss.sgi.com Sat Mar 10 02:08:26 2001
Received:  by oss.sgi.com id <S553802AbRCJKIR>;
	Sat, 10 Mar 2001 02:08:17 -0800
Received: from cc19815-a.zwoll1.ov.nl.home.com ([212.204.138.247]:22512 "HELO
        wisdom.myplace.net") by oss.sgi.com with SMTP id <S553795AbRCJKH6>;
	Sat, 10 Mar 2001 02:07:58 -0800
Received: from ws1 (ws1.myplace.net [192.168.1.15])
	by wisdom.myplace.net (Postfix) with SMTP
	id 9FE5A40BBE; Sat, 10 Mar 2001 11:07:55 +0100 (CET)
Message-ID: <001001c0a995$693e84a0$0f01a8c0@myplace.net>
From:   "Bas" <list@showme.wox.org>
To:     "Jason Walker" <unseen@sover.net>, <linux-xfs@oss.sgi.com>
References: <Pine.LNX.4.10.10103091902400.3576-100000@surreal.localdomain>
Subject: Re: 2.4.2, LVM 0.9.1 beta 5a and XFS.
Date:   Sat, 10 Mar 2001 11:07:54 -0800
MIME-Version: 1.0
Content-Type: text/plain;
	charset="iso-8859-1"
Content-Transfer-Encoding: 7bit
X-Priority: 3
X-MSMail-Priority: Normal
X-Mailer: Microsoft Outlook Express 5.00.2615.200
X-MimeOLE: Produced By Microsoft MimeOLE V5.00.2615.200
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing

Hi,

I used the patches that came with LVM before building the kernel.

It doesn't matter if I use a striped LV or a LV on one disk, it returns the
same error.

Thanks,
Bas

----- Original Message -----
From: Jason Walker <unseen@sover.net>
To: Bas <list@showme.wox.org>
Cc: <linux-xfs@oss.sgi.com>
Sent: vrijdag 9 maart 2001 16:04
Subject: Re: 2.4.2, LVM 0.9.1 beta 5a and XFS.


>
> I use LVM as well, and I believ eyou need to aply the patches for LVM to
> the kernel source. did you do that? there will be a PATCHES/ directory in
> your LVM source. read the README in there, it says how to build the patch
> against your source tree. pretty simple proccess. see if that helps you.
>
> I am running LVM 0.9.1beta5, btw
>
> RegEx
>
> On Sat, 10 Mar 2001, Bas wrote:
>
> > Hi,
> >
> > Finally got XFS running on my machine. I'm using it next to reiserfs
which
> > is a little faster, but I believe XFS to be more stable.
> >
> > I got the tree through CVS, and compiled it last week. When I extend the
LV
> > that XFS is on, and I run xfs_growfs, it returns:
> >
> > xfs_growfs: cannot determine geometry of filesystem mounted at /mnt:
Invalid
> > argument
> >
> > I didn't see any patches for LVM to use XFS, and on the XFS site it
didn't
> > find anything to patch for XFS. How do I fix this ?
> >
> > Thanks,
> > Bas
> >
>
> --
> Jason Walker -- unseen@sover.net, perlgod@hotmail.com
> UIN: 110493687
> AIM: Nightface
>
> /"\
> \ /     ASCII RIBBON CAMPAIGN
>  X        AGAINST HTML MAIL
> / \
>
>


From owner-linux-xfs@oss.sgi.com Sat Mar 10 05:29:29 2001
Received:  by oss.sgi.com id <S553796AbRCJN3T>;
	Sat, 10 Mar 2001 05:29:19 -0800
Received: from bernstein.gruntwerk.net ([209.249.19.41]:28800 "EHLO
        thor.gruntwerk.net") by oss.sgi.com with ESMTP id <S553651AbRCJN3I>;
	Sat, 10 Mar 2001 05:29:08 -0800
Received: from gruntwerk.net (IDENT:doood@localhost.localdomain [127.0.0.1])
	by thor.gruntwerk.net (8.11.0/8.11.0) with ESMTP id f2A5T0f06450
	for <linux-xfs@oss.sgi.com>; Fri, 9 Mar 2001 21:29:00 -0800
Message-ID: <3AA9BB9C.2C3B8E1A@gruntwerk.net>
Date:   Fri, 09 Mar 2001 21:29:00 -0800
From:   dave <doood@gruntwerk.net>
X-Mailer: Mozilla 4.75 [en] (X11; U; Linux 2.4.0 i686)
X-Accept-Language: en
MIME-Version: 1.0
To:     linux-xfs@oss.sgi.com
Subject: nfs issues
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing

Greetings,

    I have been running the xfs filesystem for about 1mo now on a redhat
7.0 system. The box was installed using the SGI RedHat install iso, and
its kernel re-compiled using the CVS tree.  Its /usr directory is
exported to 4 other linux boxes over nfs v2 and v3. We are constanly
having : nfs, cant get request slot errors between our workstations and
the server. At times the kernel will dump a stack trace and print out:
kernel bug 1339! I'm assuming it caused by the md driver, or the nfs
load, or a combination of both. Any suggestions to a work around? Are
there any new drivers I wouldn't get from CVS?
Thanks in advance.

Dave


From owner-linux-xfs@oss.sgi.com Sat Mar 10 06:14:19 2001
Received:  by oss.sgi.com id <S553796AbRCJOOA>;
	Sat, 10 Mar 2001 06:14:00 -0800
Received: from pneumatic-tube.sgi.com ([204.94.214.22]:14 "EHLO
        pneumatic-tube.sgi.com") by oss.sgi.com with ESMTP
	id <S553651AbRCJONn>; Sat, 10 Mar 2001 06:13:43 -0800
Received: from zeus-fddi.americas.sgi.com (zeus-fddi.americas.sgi.com [128.162.8.103]) by pneumatic-tube.sgi.com (980327.SGI.8.8.8-aspam/980310.SGI-aspam) via ESMTP id GAA03504
	for <linux-xfs@oss.sgi.com>; Sat, 10 Mar 2001 06:23:29 -0800 (PST)
	mail_from (lord@sgi.com)
Received: from daisy-e185.americas.sgi.com (daisy-e194.americas.sgi.com [128.162.194.214]) by zeus-fddi.americas.sgi.com (8.9.3/americas-smart-nospam1.1) with ESMTP id IAA1089968; Sat, 10 Mar 2001 08:12:25 -0600 (CST)
Received: from jen.americas.sgi.com (IDENT:root@jen.americas.sgi.com [128.162.184.86]) by daisy-e185.americas.sgi.com (SGI-8.9.3/SGI-server-1.7) with ESMTP id IAA75882; Sat, 10 Mar 2001 08:12:24 -0600 (CST)
Received: from jen.americas.sgi.com by jen.americas.sgi.com (8.11.0/SGI-client-1.7) via ESMTP id f2AEEI903763; Sat, 10 Mar 2001 08:14:19 -0600
Message-Id: <200103101414.f2AEEI903763@jen.americas.sgi.com>
X-Mailer: exmh version 2.2 06/23/2000 with nmh-1.0.4
To:     dave <doood@gruntwerk.net>
cc:     linux-xfs@oss.sgi.com
Subject: Re: nfs issues 
In-Reply-To: Message from dave <doood@gruntwerk.net> 
   of "Fri, 09 Mar 2001 21:29:00 PST." <3AA9BB9C.2C3B8E1A@gruntwerk.net> 
Date:   Sat, 10 Mar 2001 08:14:18 -0600
From:   Steve Lord <lord@sgi.com>
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing

> Greetings,
> 
>     I have been running the xfs filesystem for about 1mo now on a redhat
> 7.0 system. The box was installed using the SGI RedHat install iso, and
> its kernel re-compiled using the CVS tree.  Its /usr directory is
> exported to 4 other linux boxes over nfs v2 and v3. We are constanly
> having : nfs, cant get request slot errors between our workstations and
> the server. At times the kernel will dump a stack trace and print out:
> kernel bug 1339! I'm assuming it caused by the md driver, or the nfs
> load, or a combination of both. Any suggestions to a work around? Are
> there any new drivers I wouldn't get from CVS?
> Thanks in advance.
> 
> Dave


I have seen this error from my workstation accessing NFS servers which are
not running Linux, it may be a generic problem. This message comes from
this code:

        if (clnt->cl_chatty && !(task->tk_flags & RPC_CALL_MAJORSEEN)) {
                task->tk_flags |= RPC_CALL_MAJORSEEN;
                if (req)
                        printk(KERN_NOTICE "%s: server %s not responding, still trying\n",
                                clnt->cl_protname, clnt->cl_server);
#ifdef RPC_DEBUG
                else
                        printk(KERN_NOTICE "%s: task %d can't get a request slot\n",
                                clnt->cl_protname, task->tk_pid);
#endif
        }

in an rpc timeout path, so this is a request timeout.

I am not sure if this could be happening because or XFS or not. How recent is
your kernel from cvs, there were changes this week (not complete until
yesterday) which help nfs performance considerably.

Steve



From owner-linux-xfs@oss.sgi.com Sat Mar 10 06:19:59 2001
Received:  by oss.sgi.com id <S553796AbRCJOTu>;
	Sat, 10 Mar 2001 06:19:50 -0800
Received: from tux.mkp.net ([130.225.60.11]:47883 "EHLO tux.mkp.net")
	by oss.sgi.com with ESMTP id <S553651AbRCJOTc>;
	Sat, 10 Mar 2001 06:19:32 -0800
Received: from tux.mkp.net ([130.225.60.11] helo=jaguar.mkp.net)
	by tux.mkp.net with esmtp (Exim 3.16 #1)
	id 14bkDF-0007V4-00; Sat, 10 Mar 2001 15:19:19 +0100
Received: (from mkp@localhost)
	by jaguar.mkp.net (8.9.3/8.9.3) id JAA09929;
	Sat, 10 Mar 2001 09:18:01 -0500
X-Authentication-Warning: jaguar.mkp.net: mkp set sender to mkp@mkp.net using -f
To:     "Bas" <list@showme.wox.org>
Cc:     <linux-xfs@oss.sgi.com>
Subject: Re: 2.4.2, LVM 0.9.1 beta 5a and XFS.
References: <000a01c0a93f$22643440$0f01a8c0@myplace.net>
From:   "Martin K. Petersen" <mkp@linuxcare.com>
Organization: Linuxcare, Inc.
Date:   10 Mar 2001 09:18:01 -0500
In-Reply-To: <000a01c0a93f$22643440$0f01a8c0@myplace.net>
Message-ID: <yq14rx1zp0m.fsf@jaguar.mkp.net>
Lines:  16
User-Agent: Gnus/5.0808 (Gnus v5.8.8) XEmacs/21.1 (Canyonlands)
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing

>>>>> "Bas" == Bas  <list@showme.wox.org> writes:

Bas> I got the tree through CVS, and compiled it last week. When I
Bas> extend the LV that XFS is on, and I run xfs_growfs, it returns:

Bas> xfs_growfs: cannot determine geometry of filesystem mounted at
Bas> /mnt: Invalid argument

I'll put it on my list to have a look.  Merge to LVM 0.9 is next on my
list of things to do, once I get the RAID1+5 resync speed throttling
right.

-- 
Martin K. Petersen, Principal Linux Consultant, Linuxcare, Inc.
mkp@linuxcare.com, http://www.linuxcare.com/
SGI XFS for Linux Developer, http://oss.sgi.com/projects/xfs/

From owner-linux-xfs@oss.sgi.com Sat Mar 10 12:04:06 2001
Received:  by oss.sgi.com id <S553818AbRCJUDr>;
	Sat, 10 Mar 2001 12:03:47 -0800
Received: from pike.sover.net ([209.198.87.34]:44483 "EHLO pike.sover.net")
	by oss.sgi.com with ESMTP id <S553802AbRCJUDQ>;
	Sat, 10 Mar 2001 12:03:16 -0800
Received: from surreal.localdomain (pm1a13.stj.sover.net [209.198.94.13])
	by pike.sover.net (8.9.3/8.9.3) with ESMTP id PAA25684
	for <linux-xfs@oss.sgi.com>; Sat, 10 Mar 2001 15:03:08 -0500 (EST)
Comments: SoVerNet Verification (on pike.sover.net)
	surreal.localdomain from pm1a13.stj.sover.net [209.198.94.13] 209.198.94.13
	Sat, 10 Mar 2001 15:03:08 -0500 (EST)
Date:   Sat, 10 Mar 2001 15:04:06 -0500 (EST)
From:   Jason Walker <unseen@sover.net>
X-Sender: unseen@surreal.localdomain
To:     SGI XFS Mailing List <linux-xfs@oss.sgi.com>
Subject: mkfs.xfs warning on LVM
Message-ID: <Pine.LNX.4.10.10103101456450.14964-100000@surreal.localdomain>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing


Hello all,

	I made a new lv with LVM, and did this:

[root@surreal:/raid]# mkfs.xfs /dev/raid/tarballs    
mkfs.xfs: warning - cannot set blocksize on block device
/dev/raid/tarballs: Invalid argument
meta-data=/dev/raid/tarballs     isize=256    agcount=8, agsize=65536 blks
data     =                       bsize=4096   blocks=524288, imaxpct=25
         =                       sunit=0      swidth=0 blks, unwritten=0
		 naming   =version 2              bsize=4096  
		 log      =internal log           bsize=4096   blocks=1200
		 realtime =none                   extsz=65536  blocks=0,
		 rtextents=0

	That warning, should I be concerned about it? Everything *seems* to be
in order. it mounts fine, and I just copied a few gb of data to it,
everything seems fine. Warnings make me nervous, that's all :)

Jason

-- 
Jason Walker -- unseen@sover.net, perlgod@hotmail.com
UIN: 110493687
AIM: Nightface

/"\
\ /     ASCII RIBBON CAMPAIGN
 X        AGAINST HTML MAIL
/ \
			 


From owner-linux-xfs@oss.sgi.com Sun Mar 11 23:13:34 2001
Received:  by oss.sgi.com id <S553681AbRCLHNZ>;
	Sun, 11 Mar 2001 23:13:25 -0800
Received: from hermes.mixx.net ([212.84.196.2]:61448 "HELO hermes.mixx.net")
	by oss.sgi.com with SMTP id <S553648AbRCLHNF>;
	Sun, 11 Mar 2001 23:13:05 -0800
Received: from mate.bln.innominate.de (cerberus.berlin.innominate.de [212.84.234.251])
	by hermes.mixx.net (Postfix) with ESMTP id CFF1BF806
	for <linux-xfs@oss.sgi.com>; Mon, 12 Mar 2001 08:12:59 +0100 (CET)
Received: by mate.bln.innominate.de (Postfix, from userid 9)
	id 348442CA6F; Mon, 12 Mar 2001 08:12:59 +0100 (CET)
From:   Thomas Graichen <news-innominate.list.sgi.xfs@innominate.de>
Reply-To: Thomas Graichen <thomas.graichen@innominate.com>
X-Newsgroups: innominate.list.sgi.xfs
Subject: FAQ updated
Date:   12 Mar 2001 07:12:59 GMT
Organization: innominate AG, Berlin, Germany
Lines:  36
Distribution: local
Message-ID: <news2mail-98hstr$rjm$1@mate.bln.innominate.de>
X-Trace: mate.bln.innominate.de 984381179 28278 10.0.0.31 (12 Mar 2001 07:12:59 GMT)
X-Complaints-To: news@innominate.de
User-Agent: tin/1.4.4-20000803 ("Vet for the Insane") (UNIX) (Linux/2.4.2-XFS (i686))
To:     linux-xfs@oss.sgi.com
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing

ok - i've updated the XFS FAQ a bit yesterday and committed it right
now (so it should appear on oss soon i think - it's the new one if
the last questions is "Mounting or umounting an XFS root ...")

please check it out if i have missed something or did something wrong
so that i can update all this in a second round. i am especially
interested in information about the following points:

* current status of XFS on sw-RAID (russel are your patches in now -
  i don't think so - right? - any plans within this week?)
* the current state of NFS on XFS - after the latest fixes - are
  there any issues remaining?
* gcc 2.96 - does the kernel now work with the updated redhat gcc
  2.96 or with 2.95.2 - anyone checked it lately?
* what is the current state of XFS on sparc64? - on the alpha (i may
  try it on the one here this week - but maybe someone else has it
  running right now)? - other arches so far?
* does XFS on LVM work out of the box (i.e. XFS tree plus lvm beta
  patches) now?
* does anyone else haveany useful links for XFS (the "installing XFS
  as root fs" HOWTO, the precompiled rpm's, maybe some hidden docs
  about XFS etc.)? - if yes - please send me the url ...
* is anyone using XFS in an production environment right now? - is
  anyone using XFS in a large scale (big SMP machine or large file-
  system) system on a regular basis?

a lot of thanks in advance - please let me know if you have any
condensed information about the above points

t

-- 
thomas.graichen@innominate.com
                                                         innominate AG
                                                  the linux architects
tel: +49-30-308806-13   fax: -77             http://www.innominate.com

From owner-linux-xfs@oss.sgi.com Mon Mar 12 00:32:26 2001
Received:  by oss.sgi.com id <S553681AbRCLIcQ>;
	Mon, 12 Mar 2001 00:32:16 -0800
Received: from [65.100.85.34] ([65.100.85.34]:18048 "EHLO
        roujin.gargoylecc.com") by oss.sgi.com with ESMTP
	id <S553648AbRCLIbw>; Mon, 12 Mar 2001 00:31:52 -0800
Received: from roujin.gargoylecc.com (IDENT:ringram@roujin.gargoylecc.com [65.100.85.34])
	by roujin.gargoylecc.com (8.9.3/8.9.3) with ESMTP id QAA07253;
	Mon, 12 Mar 2001 16:43:33 -0700
Date:   Mon, 12 Mar 2001 16:43:33 -0700 (MST)
From:   Russel Ingram <ringram@gargoylecc.com>
To:     Thomas Graichen <thomas.graichen@innominate.com>
cc:     linux-xfs@oss.sgi.com
Subject: Re: FAQ updated
In-Reply-To: <news2mail-98hstr$rjm$1@mate.bln.innominate.de>
Message-ID: <Pine.LNX.4.21.0103121639590.7247-100000@roujin.gargoylecc.com>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing

On 12 Mar 2001, Thomas Graichen wrote:

> * does anyone else haveany useful links for XFS (the "installing XFS
>   as root fs" HOWTO, the precompiled rpm's, maybe some hidden docs
>   about XFS etc.)? - if yes - please send me the url ...
The howto I wrote for installing XFS as the root fs (as well as the rest
of the system for that matter) is here:

http://www.gargoylecc.com/Linux+XFS-HOWTO.html

I just (mostly) finished the port to HTML.  It still needs some
navigation links put in but other than that it's complete.  

Russ 

-- 
---------------------------------------------------------------
"Bill Gates and Microsoft have ruined the computer industry for
 a long time to come by creating a class of ignorant and lazy
 computer users."  --Russel Ingram

"Mommy ... can I go out and ... KILL TONIGHT!?"
                   --Glen  Danzig, The Misfits
---------------------------------------------------------------
Russ Ingram
Gargoyle Computer Consulting
(307)742-1361
www.gargoylecc.com


From owner-linux-xfs@oss.sgi.com Mon Mar 12 03:30:05 2001
Received:  by oss.sgi.com id <S553665AbRCLL34>;
	Mon, 12 Mar 2001 03:29:56 -0800
Received: from horizon.barak.net.il ([206.49.94.218]:36563 "EHLO
        horizon.barak-online.net") by oss.sgi.com with ESMTP
	id <S553655AbRCLL3f>; Mon, 12 Mar 2001 03:29:35 -0800
Received: from yahoo.com (pop02-1-ras3-p155.barak.net.il [212.150.104.155])
	by horizon.barak-online.net (8.11.2/8.9.1) with ESMTP id f2CBTfr10489
	for <linux-xfs@oss.sgi.com>; Mon, 12 Mar 2001 13:29:41 +0200 (IST)
Message-ID: <3AACB386.6040607@yahoo.com>
Date:   Mon, 12 Mar 2001 13:31:18 +0200
From:   Yaacov Akiba Slama <slamaya@yahoo.com>
User-Agent: Mozilla/5.0 (Windows; U; Windows NT 5.0; en-US; 0.8) Gecko/20010215
X-Accept-Language: en
MIME-Version: 1.0
To:     linux-xfs@oss.sgi.com
Subject: 2.4.2-acxx and XFS
Content-Type: text/plain; charset=us-ascii; format=flowed
Content-Transfer-Encoding: 7bit
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing

Hello !
I wanted to use XFS with the patches from Alan Cox (I need the 
modifications in 2.4.2-ac18), but it seems to be a conflict between the 
ac patches and the Feb232001devel.patch .

What can I do ?
Thanks,
YAS


From owner-linux-xfs@oss.sgi.com Mon Mar 12 04:16:15 2001
Received:  by oss.sgi.com id <S553770AbRCLMQG>;
	Mon, 12 Mar 2001 04:16:06 -0800
Received: from ubr-35.28.151.oviedo.cfl.rr.com ([65.35.28.151]:25589 "EHLO
        home.smithconcepts.com") by oss.sgi.com with ESMTP
	id <S553708AbRCLMPy>; Mon, 12 Mar 2001 04:15:54 -0800
Received: from ieee.org (IDENT:bjsmith@bitman.oviedo.smithconcepts.com [172.24.24.192])
	by home.smithconcepts.com (8.9.3/8.9.3) with ESMTP id HAA16112;
	Mon, 12 Mar 2001 07:10:25 -0500
Message-ID: <3AACC014.E1D9BC28@ieee.org>
Date:   Mon, 12 Mar 2001 07:24:52 -0500
From:   "Bryan J. Smith" <b.j.smith@ieee.org>
Reply-To: b.j.smith@ieee.org, thebs@theseus.com
Organization: SmithConcepts/Personal
X-Mailer: Mozilla 4.76 [en] (X11; U; Linux 2.2.18-0.13 i686)
X-Accept-Language: en
MIME-Version: 1.0
To:     Thomas Graichen <thomas.graichen@innominate.com>
CC:     linux-xfs@oss.sgi.com
Subject: Re: FAQ updated  2.96, RH7 RPMs w/HOWTO, JFS presentation ...
References: <news2mail-98hstr$rjm$1@mate.bln.innominate.de>
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing

Thomas Graichen wrote:
> * gcc 2.96 - does the kernel now work with the updated redhat gcc
>   2.96 or with 2.95.2 - anyone checked it lately?

I believe my RPMs were built on 2.96.  For the life of me, I cannot
remember how to check (brain fart).  Maybe it was 2.91 -- not sure
(could have swore it was 2.96).

> * does anyone else haveany useful links for XFS (the "installing XFS
>   as root fs" HOWTO, the precompiled rpm's, maybe some hidden docs
>   about XFS etc.)? - if yes - please send me the url ...

My "first batch" of RPMs from the original 2.4.2 release are here:

http://www.smithconcepts.com/files/xfs/kernel-xfscvs_feb24

[ My releases will always be in
  http://www.smithconcepts.com/files/xfs
  and I have setup rsync for you'all ]

I am in the middle of building a "more complete" RPM with numerous
patches applied.  If anyone has additional patches they want to
suggest (other than those in the RedHat RPMs), please do so.

I'm sorry I have not updated them yet.  I also run a LUG and we are
admist moving servers.  That will be completed tonight so I can
direct my full attention to making more XFS RPMs.

I recently did a JFS presentation at my LUG.  It was a "quick'n
dirty" one (about 30 slides, should have been more like 100 -- but I
got lazy).  If anyone is interested in them, the PDF (36KB) is here:

http://www.smithconcepts.com/files/presentations/ELUG_JFS_2001Mar05a.pdf

> * is anyone using XFS in an production environment right now? - is
>   anyone using XFS in a large scale (big SMP machine or large file-
>   system) system on a regular basis?

I am in the middle of my standard "3 month testing" of XFS, just
like I did with Ext3.  XFS is going on a few of my 1GHz Pentium III
workstations at work later this week (or whenever I get the new
batch of RPMs).  I am also working on making a "2.4 firewall
distro/HOWTO" and XFS is at the center of that as well (and will be
the subject of my LUG's April presentation).

> a lot of thanks in advance - please let me know if you have any
> condensed information about the above points

My (somewhat dated) HOWTO is here in HTML:
http://www.smithconcepts.com/files/xfs/kernel-2.4.2-xfscvs_feb24/SCXFS-HOWTO.html
And PDF:
http://www.smithconcepts.com/files/xfs/kernel-2.4.2-xfscvs_feb24/SCXFS-HOWTO.pdf

You'll find I make what I like to call "RedHat microscopic HOWTOs"
which I am regularly bashed for.  I'd love to make more "generic"
HOWTOs, but most newbies like the RedHat-specific ones anyway.

-- Bryan "TheBS" Smith, SmithConcepts
   Engineer/SysAdmin, Theseus Logic

-- 
Bryan "TheBS" Smith         chat:thebs413 @AOL/MSN/Yahoo
Engineer     mailto:b.j.smith@ieee.org,thebs@theseus.com
********************************************************
"Linux will do for applications what the Internet did to
 networks" -- Sam Palmisano, IBM Chief Operating Officer

From owner-linux-xfs@oss.sgi.com Mon Mar 12 04:35:06 2001
Received:  by oss.sgi.com id <S553775AbRCLMeq>;
	Mon, 12 Mar 2001 04:34:46 -0800
Received: from main.braxis.co.uk ([212.160.232.26]:56843 "EHLO
        main.braxis.co.uk") by oss.sgi.com with ESMTP id <S553757AbRCLMef>;
	Mon, 12 Mar 2001 04:34:35 -0800
Received: (from kszysiu@localhost)
	by main.braxis.co.uk (8.9.3/8.9.3) id NAA07989;
	Mon, 12 Mar 2001 13:23:13 +0100
Date:   Mon, 12 Mar 2001 13:23:13 +0100
From:   Krzysztof Rusocki <kszysiu@braxis.co.uk>
To:     Yaacov Akiba Slama <slamaya@yahoo.com>
Cc:     linux-xfs@oss.sgi.com
Subject: Re: 2.4.2-acxx and XFS
Message-ID: <20010312132313.A6858@main.braxis.co.uk>
References: <3AACB386.6040607@yahoo.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.2.5i
In-Reply-To: <3AACB386.6040607@yahoo.com>; from slamaya@yahoo.com on Mon, Mar 12, 2001 at 01:31:18PM +0200
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing


Hi,

Actually guys at XFS do not merge XFS with -ac series kernels.
Probably you'll have to apply the part of -ac patch which concerns you by
yourself... or ask somebody else do it :)

Cheers,
Krzysztof


On Mon, Mar 12, 2001 at 01:31:18PM +0200, Yaacov Akiba Slama wrote:
> Hello !
> I wanted to use XFS with the patches from Alan Cox (I need the 
> modifications in 2.4.2-ac18), but it seems to be a conflict between the 
> ac patches and the Feb232001devel.patch .
> 
> What can I do ?
> Thanks,
> YAS

From owner-linux-xfs@oss.sgi.com Mon Mar 12 05:37:08 2001
Received:  by oss.sgi.com id <S553775AbRCLNg7>;
	Mon, 12 Mar 2001 05:36:59 -0800
Received: from tux.mkp.net ([130.225.60.11]:64785 "EHLO tux.mkp.net")
	by oss.sgi.com with ESMTP id <S553757AbRCLNgf>;
	Mon, 12 Mar 2001 05:36:35 -0800
Received: from tux.mkp.net ([130.225.60.11] helo=jaguar.mkp.net)
	by tux.mkp.net with esmtp (Exim 3.16 #1)
	id 14cSUx-0001F1-00; Mon, 12 Mar 2001 14:36:32 +0100
Received: (from mkp@localhost)
	by jaguar.mkp.net (8.9.3/8.9.3) id IAA10294;
	Mon, 12 Mar 2001 08:35:08 -0500
X-Authentication-Warning: jaguar.mkp.net: mkp set sender to mkp@mkp.net using -f
To:     Jason Walker <unseen@sover.net>
Cc:     SGI XFS Mailing List <linux-xfs@oss.sgi.com>
Subject: Re: mkfs.xfs warning on LVM
References: <Pine.LNX.4.10.10103101456450.14964-100000@surreal.localdomain>
From:   "Martin K. Petersen" <mkp@linuxcare.com>
Organization: Linuxcare, Inc.
Date:   12 Mar 2001 08:35:08 -0500
In-Reply-To: <Pine.LNX.4.10.10103101456450.14964-100000@surreal.localdomain>
Message-ID: <yq1u24zxg8j.fsf@jaguar.mkp.net>
Lines:  13
User-Agent: Gnus/5.0808 (Gnus v5.8.8) XEmacs/21.1 (Canyonlands)
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing

>>>>> "Jason" == Jason Walker <unseen@sover.net> writes:

Jason> 	I made a new lv with LVM, and did this:

Jason> [root@surreal:/raid]# mkfs.xfs /dev/raid/tarballs mkfs.xfs:
Jason> warning - cannot set blocksize on block device

You need to upgrade your xfsprogs to the latest-and-greatest from CVS.

-- 
Martin K. Petersen, Principal Linux Consultant, Linuxcare, Inc.
mkp@linuxcare.com, http://www.linuxcare.com/
SGI XFS for Linux Developer, http://oss.sgi.com/projects/xfs/

From owner-linux-xfs@oss.sgi.com Mon Mar 12 05:46:09 2001
Received:  by oss.sgi.com id <S553780AbRCLNpt>;
	Mon, 12 Mar 2001 05:45:49 -0800
Received: from tux.mkp.net ([130.225.60.11]:1554 "EHLO tux.mkp.net")
	by oss.sgi.com with ESMTP id <S553775AbRCLNpe>;
	Mon, 12 Mar 2001 05:45:34 -0800
Received: from tux.mkp.net ([130.225.60.11] helo=jaguar.mkp.net)
	by tux.mkp.net with esmtp (Exim 3.16 #1)
	id 14cSdf-0001FJ-00; Mon, 12 Mar 2001 14:45:32 +0100
Received: (from mkp@localhost)
	by jaguar.mkp.net (8.9.3/8.9.3) id IAA10297;
	Mon, 12 Mar 2001 08:44:08 -0500
X-Authentication-Warning: jaguar.mkp.net: mkp set sender to mkp@mkp.net using -f
To:     Thomas Graichen <thomas.graichen@innominate.com>
Cc:     linux-xfs@oss.sgi.com
Subject: Re: FAQ updated
References: <news2mail-98hstr$rjm$1@mate.bln.innominate.de>
From:   "Martin K. Petersen" <mkp@linuxcare.com>
Organization: Linuxcare, Inc.
Date:   12 Mar 2001 08:44:08 -0500
In-Reply-To: <news2mail-98hstr$rjm$1@mate.bln.innominate.de>
Message-ID: <yq1pufnxftj.fsf@jaguar.mkp.net>
Lines:  35
User-Agent: Gnus/5.0808 (Gnus v5.8.8) XEmacs/21.1 (Canyonlands)
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing

>>>>> "Thomas" == Thomas Graichen <news-innominate.list.sgi.xfs@innominate.de> writes:

Thomas,

Thomas> * current status of XFS on sw-RAID (russel are your patches in
Thomas>   now - i don't think so - right? - any plans within this
Thomas>   week?)

The patches are not in, no.  The resync throttling code is extremely
brittle and has been bugging me for days now.

I think I nailed the bug Friday, but my testbox ratted out on me (time
to get an X10 relay), so I haven't been able to verify.  I'll try it
as soon as I get to the office.  Barring new bugs, I'll check in the
fixes.

If throttling still doesn't work correctly, I'll probably check in the
MD patches anyway and then fix it later.  It's not really a problem
unless you run your RAID in degraded mode.


Thomas> * does XFS on LVM work out of the box (i.e. XFS tree plus lvm
Thomas>   beta patches) now?

I have prepared an upgrade to 0.9 for xfsprogs this weekend.  I'll
commit today.

Due to the frequent LVM IOP protocol changes, both userland and kernel
need to be at the same version, so I ponder upgrading TOT to 0.9beta6
once it gets out.

-- 
Martin K. Petersen, Principal Linux Consultant, Linuxcare, Inc.
mkp@linuxcare.com, http://www.linuxcare.com/
SGI XFS for Linux Developer, http://oss.sgi.com/projects/xfs/

From owner-linux-xfs@oss.sgi.com Mon Mar 12 07:58:19 2001
Received:  by oss.sgi.com id <S553789AbRCLP6K>;
	Mon, 12 Mar 2001 07:58:10 -0800
Received: from deliverator.sgi.com ([204.94.214.10]:8051 "EHLO
        deliverator.sgi.com") by oss.sgi.com with ESMTP id <S553692AbRCLP5p>;
	Mon, 12 Mar 2001 07:57:45 -0800
Received: from zeus-fddi.americas.sgi.com (zeus-fddi.americas.sgi.com [128.162.8.103]) by deliverator.sgi.com (980309.SGI.8.8.8-aspam-6.2/980310.SGI-aspam) via ESMTP id HAA16647
	for <linux-xfs@oss.sgi.com>; Mon, 12 Mar 2001 07:56:37 -0800 (PST)
	mail_from (lord@sgi.com)
Received: from daisy-e185.americas.sgi.com (daisy-e194.americas.sgi.com [128.162.194.214]) by zeus-fddi.americas.sgi.com (8.9.3/americas-smart-nospam1.1) with ESMTP id JAA1097382; Mon, 12 Mar 2001 09:56:27 -0600 (CST)
Received: from jen.americas.sgi.com (IDENT:root@jen.americas.sgi.com [128.162.184.86]) by daisy-e185.americas.sgi.com (SGI-8.9.3/SGI-server-1.7) with ESMTP id JAA20410; Mon, 12 Mar 2001 09:56:27 -0600 (CST)
Received: from jen.americas.sgi.com by jen.americas.sgi.com (8.11.0/SGI-client-1.7) via ESMTP id f2CFvrk24545; Mon, 12 Mar 2001 09:57:53 -0600
Message-Id: <200103121557.f2CFvrk24545@jen.americas.sgi.com>
X-Mailer: exmh version 2.2 06/23/2000 with nmh-1.0.4
To:     Thomas Graichen <thomas.graichen@innominate.com>
cc:     linux-xfs@oss.sgi.com
Subject: Re: FAQ updated 
In-Reply-To: Message from Thomas Graichen <news-innominate.list.sgi.xfs@innominate.de> 
   of "12 Mar 2001 07:12:59 GMT." <news2mail-98hstr$rjm$1@mate.bln.innominate.de> 
Date:   Mon, 12 Mar 2001 09:57:53 -0600
From:   Steve Lord <lord@sgi.com>
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing

> ok - i've updated the XFS FAQ a bit yesterday and committed it right
> now (so it should appear on oss soon i think - it's the new one if
> the last questions is "Mounting or umounting an XFS root ...")
> 
> please check it out if i have missed something or did something wrong
> so that i can update all this in a second round. i am especially
> interested in information about the following points:
> 
> * current status of XFS on sw-RAID (russel are your patches in now -
>   i don't think so - right? - any plans within this week?)
> * the current state of NFS on XFS - after the latest fixes - are
>   there any issues remaining?

So far as I know, the CVS tree should now be OK on NFS, last week was
all performance problems, with an intermediate couple of days between
the first and last part of the fix.


> * gcc 2.96 - does the kernel now work with the updated redhat gcc
>   2.96 or with 2.95.2 - anyone checked it lately?

The latest redhat gcc does not build xfs - there are still compile errors
associated with the use of the do_div macros. Someone internally will get
to this eventually, but external help would probably get this sorted out
faster than we will right now. So for instance:

xfs_bmap.c:2830:36: warning: pasting would not give a valid preprocessing token
xfs_bmap.c: In function `xfs_bmap_del_extent':
xfs_bmap.c:3130: Unrecognizable insn:
(insn/i 254 252 4392 (parallel[ 
            (set (reg:SI 0 eax)
                (asm_operands ("") ("=a") 0[ 
                        (reg:DI 1 edx)
                    ] 
                    [ 
                        (asm_input:DI ("A"))
                    ]  ("linux/xfs_linux.h") 267))
            (set (reg:SI 1 edx)
                (asm_operands ("") ("=d") 1[ 
                        (reg:DI 1 edx)
                    ] 
                    [ 
                        (asm_input:DI ("A"))
                    ]  ("linux/xfs_linux.h") 267))
            (clobber (reg:QI 19 dirflag))
            (clobber (reg:QI 18 fpsr))
            (clobber (reg:QI 17 flags))
        ] ) -1 (nil)
    (nil))
xfs_bmap.c:3130: confused by earlier errors, bailing out


> * what is the current state of XFS on sparc64? - on the alpha (i may
>   try it on the one here this week - but maybe someone else has it
>   running right now)? - other arches so far?

No Knowledge here.

> * does XFS on LVM work out of the box (i.e. XFS tree plus lvm beta
>   patches) now?
> * does anyone else haveany useful links for XFS (the "installing XFS
>   as root fs" HOWTO, the precompiled rpm's, maybe some hidden docs
>   about XFS etc.)? - if yes - please send me the url ...
> * is anyone using XFS in an production environment right now? - is
>   anyone using XFS in a large scale (big SMP machine or large file-
>   system) system on a regular basis?

I will have to leave it to external people to comment on this, I think
there are some which would classify as production, I am not sure how
big things get in terms of machine sizes etc.

> 
> a lot of thanks in advance - please let me know if you have any
> condensed information about the above points
> 
> t
> 
> -- 
> thomas.graichen@innominate.com
>                                                          innominate AG
>                                                   the linux architects
> tel: +49-30-308806-13   fax: -77             http://www.innominate.com



From owner-linux-xfs@oss.sgi.com Mon Mar 12 08:47:19 2001
Received:  by oss.sgi.com id <S553781AbRCLQrK>;
	Mon, 12 Mar 2001 08:47:10 -0800
Received: from deliverator.sgi.com ([204.94.214.10]:23679 "EHLO
        deliverator.sgi.com") by oss.sgi.com with ESMTP id <S553692AbRCLQqp>;
	Mon, 12 Mar 2001 08:46:45 -0800
Received: from info.engr.sgi.com (info.engr.sgi.com [192.26.80.216]) by deliverator.sgi.com (980309.SGI.8.8.8-aspam-6.2/980310.SGI-aspam) via ESMTP id IAA22881; Mon, 12 Mar 2001 08:45:38 -0800 (PST)
	mail_from (nobody@info.engr.sgi.com)
Received: (from nobody@localhost)
	by info.engr.sgi.com (SGI-8.9.3/8.9.3) id IAA88465;
	Mon, 12 Mar 2001 08:46:43 -0800 (PST)
Date:   Mon, 12 Mar 2001 08:46:43 -0800 (PST)
Message-Id: <200103121646.IAA88465@info.engr.sgi.com>
X-Pv-Incident: 817252 
webPV:  lord-h1.americas.sgi.com 
webExec: webpvupdate,pvincident 
Reply-To: sgi.bugs.xfs@fido.engr.sgi.com
From:   pv@relay.sgi.com (lord@sgi.com)
Subject: CLOSE 817252 - xfs qa 013 trips assertion
To:     lord@sgi.com
Cc:     linux-xfs@oss.sgi.com
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing

View Incident: http://co-op.engr.sgi.com/BugWorks/code/bwxquery.cgi?search=Search&wlong=1&view_type=Bug&wi=817252

*Status : closed                       Priority : 2                         
 Assigned Engineer : lord              Submitter : dxm                      
 Opened Date : 03/07/01               *Closed Date : 03/12/01               
*Fixed By : lord                      *Fixed By Domain : sgi.com            
*Modified Date : 03/12/01             *Modified User : lord                 
*Modified User Domain : sgi.com       *Fix Description :
==========================
ADDITIONAL INFORMATION (CLOSE)
From: lord@sgi.com (BugWorks)
Date: Mar 12 2001 08:46:43AM
==========================

This was fixed in the last direct I/O change check in.

From owner-linux-xfs@oss.sgi.com Mon Mar 12 09:13:50 2001
Received:  by oss.sgi.com id <S553781AbRCLRNa>;
	Mon, 12 Mar 2001 09:13:30 -0800
Received: from deliverator.sgi.com ([204.94.214.10]:25867 "EHLO
        deliverator.sgi.com") by oss.sgi.com with ESMTP id <S553692AbRCLRNB>;
	Mon, 12 Mar 2001 09:13:01 -0800
Received: from zeus-fddi.americas.sgi.com (zeus-fddi.americas.sgi.com [128.162.8.103]) by deliverator.sgi.com (980309.SGI.8.8.8-aspam-6.2/980310.SGI-aspam) via ESMTP id JAA26606
	for <linux-xfs@oss.sgi.com>; Mon, 12 Mar 2001 09:11:52 -0800 (PST)
	mail_from (lord@sgi.com)
Received: from daisy-e185.americas.sgi.com (daisy-e194.americas.sgi.com [128.162.194.214]) by zeus-fddi.americas.sgi.com (8.9.3/americas-smart-nospam1.1) with ESMTP id LAA1101632 for <linux-xfs@oss.sgi.com>; Mon, 12 Mar 2001 11:11:44 -0600 (CST)
Received: from jen.americas.sgi.com (IDENT:root@jen.americas.sgi.com [128.162.184.86]) by daisy-e185.americas.sgi.com (SGI-8.9.3/SGI-server-1.7) with ESMTP id LAA41829 for <linux-xfs@oss.sgi.com>; Mon, 12 Mar 2001 11:11:43 -0600 (CST)
From:   Steve Lord <lord@sgi.com>
Received: by jen.americas.sgi.com (8.11.0/SGI-client-1.7) id f2CHD9p26914; Mon, 12 Mar 2001 11:13:09 -0600
Message-Id: <200103121713.f2CHD9p26914@jen.americas.sgi.com>
Date:   Mon, 12 Mar 2001 11:13:09 -0600
Subject: TAKE - fix problem with randholes test case
To:     unlisted-recipients:; (no To-header on input)
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing

Recent pagebuf changes left a bug in the zeroing of space in a file
between the existing end of file and the new end of file.

Date:  Mon Mar 12 09:10:23 PST 2001
Workarea:  jen.americas.sgi.com:/src/lord/xfs-linux.2.4

The following file(s) were checked into:
  bonnie.engr.sgi.com:/isms/slinx/2.4.x-xfs


Modid:  2.4.x-xfs:slinx:89541a
linux/fs/xfs/linux/xfs_lrw.c - 1.83
	- Explicitly initialize pb_bn to PAGE_BUF_DADDR_NULL in the zero eof code
	  before calling pagebuf_iozero. The default initialization to a null
	  block number is gone now.



From owner-linux-xfs@oss.sgi.com Mon Mar 12 09:43:20 2001
Received:  by oss.sgi.com id <S553781AbRCLRnK>;
	Mon, 12 Mar 2001 09:43:10 -0800
Received: from f116.law11.hotmail.com ([64.4.17.116]:50951 "EHLO hotmail.com")
	by oss.sgi.com with ESMTP id <S553692AbRCLRmv>;
	Mon, 12 Mar 2001 09:42:51 -0800
Received: from mail pickup service by hotmail.com with Microsoft SMTPSVC;
	 Mon, 12 Mar 2001 09:42:46 -0800
Received: from 204.69.198.2 by lw11fd.law11.hotmail.msn.com with HTTP;	Mon, 12 Mar 2001 17:42:45 GMT
X-Originating-IP: [204.69.198.2]
From:   "Ying Chen" <yingchenb@hotmail.com>
To:     linux-xfs@oss.sgi.com
Subject: xfs buffer cache
Date:   Mon, 12 Mar 2001 09:42:45 -0800
Mime-Version: 1.0
Content-Type: text/plain; format=flowed
Message-ID: <F116ddAkOhw2RlC6aVE0000755e@hotmail.com>
X-OriginalArrivalTime: 12 Mar 2001 17:42:46.0066 (UTC) FILETIME=[D911C120:01C0AB1B]
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing

Hi,

On the linux port, does xfs have its own buffer cache, instead of the 
generic Linux VM/buffer cache that ext2 and many other file systems use?

Thanks,

Ying

_________________________________________________________________
Get your FREE download of MSN Explorer at http://explorer.msn.com


From owner-linux-xfs@oss.sgi.com Mon Mar 12 09:51:09 2001
Received:  by oss.sgi.com id <S553796AbRCLRuu>;
	Mon, 12 Mar 2001 09:50:50 -0800
Received: from pneumatic-tube.sgi.com ([204.94.214.22]:19821 "EHLO
        pneumatic-tube.sgi.com") by oss.sgi.com with ESMTP
	id <S553719AbRCLRul>; Mon, 12 Mar 2001 09:50:41 -0800
Received: from zeus-fddi.americas.sgi.com (zeus-fddi.americas.sgi.com [128.162.8.103]) by pneumatic-tube.sgi.com (980327.SGI.8.8.8-aspam/980310.SGI-aspam) via ESMTP id KAA07856
	for <linux-xfs@oss.sgi.com>; Mon, 12 Mar 2001 10:00:30 -0800 (PST)
	mail_from (lord@sgi.com)
Received: from daisy-e185.americas.sgi.com (daisy-e194.americas.sgi.com [128.162.194.214]) by zeus-fddi.americas.sgi.com (8.9.3/americas-smart-nospam1.1) with ESMTP id LAA1106324; Mon, 12 Mar 2001 11:49:23 -0600 (CST)
Received: from jen.americas.sgi.com (IDENT:root@jen.americas.sgi.com [128.162.184.86]) by daisy-e185.americas.sgi.com (SGI-8.9.3/SGI-server-1.7) with ESMTP id LAA69582; Mon, 12 Mar 2001 11:49:23 -0600 (CST)
Received: from jen.americas.sgi.com by jen.americas.sgi.com (8.11.0/SGI-client-1.7) via ESMTP id f2CHomx31206; Mon, 12 Mar 2001 11:50:48 -0600
Message-Id: <200103121750.f2CHomx31206@jen.americas.sgi.com>
X-Mailer: exmh version 2.2 06/23/2000 with nmh-1.0.4
To:     Thomas Graichen <thomas.graichen@innominate.com>
cc:     linux-xfs@oss.sgi.com
Subject: Re: FAQ updated 
In-Reply-To: Message from Thomas Graichen <news-innominate.list.sgi.xfs@innominate.de> 
   of "12 Mar 2001 07:12:59 GMT." <news2mail-98hstr$rjm$1@mate.bln.innominate.de> 
Date:   Mon, 12 Mar 2001 11:50:48 -0600
From:   Steve Lord <lord@sgi.com>
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing


> * gcc 2.96 - does the kernel now work with the updated redhat gcc
>   2.96 or with 2.95.2 - anyone checked it lately?

I can categorically say that if you fix the compilation problems (by making
the xfs_do_div and xfs_do_mod functions real functions rather than inlines,
that xfs DOES NOT WORK with this compiler. I have a machine which cannot
find its own root filesystem right now.

Steve




From owner-linux-xfs@oss.sgi.com Mon Mar 12 09:52:40 2001
Received:  by oss.sgi.com id <S553796AbRCLRwV>;
	Mon, 12 Mar 2001 09:52:21 -0800
Received: from pneumatic-tube.sgi.com ([204.94.214.22]:34669 "EHLO
        pneumatic-tube.sgi.com") by oss.sgi.com with ESMTP
	id <S553719AbRCLRwG>; Mon, 12 Mar 2001 09:52:06 -0800
Received: from zeus-fddi.americas.sgi.com (zeus-fddi.americas.sgi.com [128.162.8.103]) by pneumatic-tube.sgi.com (980327.SGI.8.8.8-aspam/980310.SGI-aspam) via ESMTP id KAA08097
	for <linux-xfs@oss.sgi.com>; Mon, 12 Mar 2001 10:01:55 -0800 (PST)
	mail_from (lord@sgi.com)
Received: from daisy-e185.americas.sgi.com (daisy-e194.americas.sgi.com [128.162.194.214]) by zeus-fddi.americas.sgi.com (8.9.3/americas-smart-nospam1.1) with ESMTP id LAA1109152; Mon, 12 Mar 2001 11:50:48 -0600 (CST)
Received: from jen.americas.sgi.com (IDENT:root@jen.americas.sgi.com [128.162.184.86]) by daisy-e185.americas.sgi.com (SGI-8.9.3/SGI-server-1.7) with ESMTP id LAA56054; Mon, 12 Mar 2001 11:50:48 -0600 (CST)
Received: from jen.americas.sgi.com by jen.americas.sgi.com (8.11.0/SGI-client-1.7) via ESMTP id f2CHqD231214; Mon, 12 Mar 2001 11:52:13 -0600
Message-Id: <200103121752.f2CHqD231214@jen.americas.sgi.com>
X-Mailer: exmh version 2.2 06/23/2000 with nmh-1.0.4
To:     "Ying Chen" <yingchenb@hotmail.com>
cc:     linux-xfs@oss.sgi.com
Subject: Re: xfs buffer cache 
In-Reply-To: Message from "Ying Chen" <yingchenb@hotmail.com> 
   of "Mon, 12 Mar 2001 09:42:45 PST." <F116ddAkOhw2RlC6aVE0000755e@hotmail.com> 
Date:   Mon, 12 Mar 2001 11:52:13 -0600
From:   Steve Lord <lord@sgi.com>
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing

> Hi,
> 
> On the linux port, does xfs have its own buffer cache, instead of the 
> generic Linux VM/buffer cache that ext2 and many other file systems use?
> 
> Thanks,
> 
> Ying
> 
> _________________________________________________________________
> Get your FREE download of MSN Explorer at http://explorer.msn.com

It depends, file data is handled fairly closely to the ext2 case,
but metadata is a totally distinct cache. Metadata is actually cached
in pages just like filedata, we do not use buffer_heads at all for
metadata though.

Steve



From owner-linux-xfs@oss.sgi.com Mon Mar 12 11:18:10 2001
Received:  by oss.sgi.com id <S553660AbRCLTSA>;
	Mon, 12 Mar 2001 11:18:00 -0800
Received: from deliverator.sgi.com ([204.94.214.10]:42547 "EHLO
        deliverator.sgi.com") by oss.sgi.com with ESMTP id <S553645AbRCLTRj>;
	Mon, 12 Mar 2001 11:17:39 -0800
Received: from ledzep.americas.sgi.com (ledzep.americas.sgi.com [137.38.226.97]) by deliverator.sgi.com (980309.SGI.8.8.8-aspam-6.2/980310.SGI-aspam) via ESMTP id LAA16790
	for <linux-xfs@oss.sgi.com>; Mon, 12 Mar 2001 11:16:27 -0800 (PST)
	mail_from (sandeen@gibble.americas.sgi.com)
Received: from gibble.americas.sgi.com (gibble.americas.sgi.com [128.162.195.80]) by ledzep.americas.sgi.com (SGI-SGI-8.9.3/americas-smart-nospam1.1) with ESMTP id NAA68077 for <linux-xfs@oss.sgi.com>; Mon, 12 Mar 2001 13:16:18 -0600 (CST)
Received: (from sandeen@localhost)
	by gibble.americas.sgi.com (8.11.2/8.11.2) id f2CJDqM18927
	for linux-xfs@oss.sgi.com; Mon, 12 Mar 2001 14:13:52 -0500
Date:   Mon, 12 Mar 2001 14:13:52 -0500
From:   Eric Sandeen <sandeen@sgi.com>
Message-Id: <200103121913.f2CJDqM18927@gibble.americas.sgi.com>
Subject: TAKE - irix6.5f:irix:00908a
To:     unlisted-recipients:; (no To-header on input)
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing

Date:  Mon Mar 12 11:10:09 PST 2001
Workarea:  gibble.americas.sgi.com:/home/poppy20/sandeen/xfs/workarea

The following file(s) were checked into:
  bonnie.engr.sgi.com:/isms/slinx/2.4.x-xfs 

  Merge of irix mod irix6.5f:irix:00908a
  PV Incidents affected: 784874

Modid:  2.4.x-xfs:slinx:89549a
linux/fs/xfs/mac_xfs.c - 1.10
	- Allow deletion of files with bad MAC labels. (mod irix6.5f:irix:00908a)



From owner-linux-xfs@oss.sgi.com Mon Mar 12 11:23:00 2001
Received:  by oss.sgi.com id <S553660AbRCLTWu>;
	Mon, 12 Mar 2001 11:22:50 -0800
Received: from deliverator.sgi.com ([204.94.214.10]:20789 "EHLO
        deliverator.sgi.com") by oss.sgi.com with ESMTP id <S553645AbRCLTWh>;
	Mon, 12 Mar 2001 11:22:37 -0800
Received: from ledzep.americas.sgi.com (ledzep.americas.sgi.com [137.38.226.97]) by deliverator.sgi.com (980309.SGI.8.8.8-aspam-6.2/980310.SGI-aspam) via ESMTP id LAA17608
	for <linux-xfs@oss.sgi.com>; Mon, 12 Mar 2001 11:21:29 -0800 (PST)
	mail_from (sandeen@gibble.americas.sgi.com)
Received: from gibble.americas.sgi.com (gibble.americas.sgi.com [128.162.195.80]) by ledzep.americas.sgi.com (SGI-SGI-8.9.3/americas-smart-nospam1.1) with ESMTP id NAA64012 for <linux-xfs@oss.sgi.com>; Mon, 12 Mar 2001 13:21:21 -0600 (CST)
Received: (from sandeen@localhost)
	by gibble.americas.sgi.com (8.11.2/8.11.2) id f2CJIto19018
	for linux-xfs@oss.sgi.com; Mon, 12 Mar 2001 14:18:55 -0500
Date:   Mon, 12 Mar 2001 14:18:55 -0500
From:   Eric Sandeen <sandeen@sgi.com>
Message-Id: <200103121918.f2CJIto19018@gibble.americas.sgi.com>
Subject: TAKE - irix6.5f:irix:03700a
To:     unlisted-recipients:; (no To-header on input)
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing

Date:  Mon Mar 12 11:19:33 PST 2001
Workarea:  gibble.americas.sgi.com:/home/poppy20/sandeen/xfs/workarea

The following file(s) were checked into:
  bonnie.engr.sgi.com:/isms/slinx/2.4.x-xfs

Merge of Irix mod irix6.5f:irix:03700a
PV Incidents affected:	 801763

Modid:  2.4.x-xfs:slinx:89553a
linux/fs/xfs/xfs_log.c - 1.229
	- pv 801763 rv nathans - don't set freed kmem on mount fail (irix mod irix6.5f:irix:03700a)



From owner-linux-xfs@oss.sgi.com Mon Mar 12 11:26:50 2001
Received:  by oss.sgi.com id <S553660AbRCLT0k>;
	Mon, 12 Mar 2001 11:26:40 -0800
Received: from pneumatic-tube.sgi.com ([204.94.214.22]:55930 "EHLO
        pneumatic-tube.sgi.com") by oss.sgi.com with ESMTP
	id <S553645AbRCLT0e>; Mon, 12 Mar 2001 11:26:34 -0800
Received: from zeus-fddi.americas.sgi.com (zeus-fddi.americas.sgi.com [128.162.8.103]) by pneumatic-tube.sgi.com (980327.SGI.8.8.8-aspam/980310.SGI-aspam) via ESMTP id LAA06236
	for <linux-xfs@oss.sgi.com>; Mon, 12 Mar 2001 11:36:23 -0800 (PST)
	mail_from (lord@sgi.com)
Received: from daisy-e185.americas.sgi.com (daisy-e194.americas.sgi.com [128.162.194.214]) by zeus-fddi.americas.sgi.com (8.9.3/americas-smart-nospam1.1) with ESMTP id NAA1111514 for <linux-xfs@oss.sgi.com>; Mon, 12 Mar 2001 13:25:17 -0600 (CST)
Received: from jen.americas.sgi.com (IDENT:root@jen.americas.sgi.com [128.162.184.86]) by daisy-e185.americas.sgi.com (SGI-8.9.3/SGI-server-1.7) with ESMTP id NAA62693 for <linux-xfs@oss.sgi.com>; Mon, 12 Mar 2001 13:25:08 -0600 (CST)
From:   Steve Lord <lord@sgi.com>
Received: by jen.americas.sgi.com (8.11.0/SGI-client-1.7) id f2CJQXt01662; Mon, 12 Mar 2001 13:26:33 -0600
Message-Id: <200103121926.f2CJQXt01662@jen.americas.sgi.com>
Date:   Mon, 12 Mar 2001 13:26:33 -0600
Subject: TAKE - remove the last used case of address space remapping from xfs
To:     unlisted-recipients:; (no To-header on input)
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing

For the default filesystem configuration anyway. There was one remaining
path in the default block sizes allowed on linux where we need to use
address space manipulation. This removes that case and turns off the
remapping code by default. Dealing with filesystems with a block
size bigger than a page will still require something like this
feature, and dealing with directory blocks bigger than a page will
as well (this is turned off in mkfs at the moment).

Date:  Mon Mar 12 11:22:23 PST 2001
Workarea:  jen.americas.sgi.com:/src/lord/xfs-linux.2.4

The following file(s) were checked into:
  bonnie.engr.sgi.com:/isms/slinx/2.4.x-xfs


Modid:  2.4.x-xfs:slinx:89552a
linux/fs/xfs/xfs_rw.c - 1.334
	- allow better flag passing down from xfs to pagebuf

linux/fs/xfs/xfs_buf.h - 1.68
	- define xfs_biomove to call pagebuf_iomove, remove definition of 
	  xfs_bp_mapin

linux/fs/xfs/xfs_attr.c - 1.82
	- Use xfs_biomove to move extended attribute data in and out of buffers.

linux/include/linux/page_buf.h - 1.81
	- define pagebuf_iomove, remove mapin and mapout prototypes

linux/fs/pagebuf/page_buf.c - 1.64
	- Add pagebuf_iomove to copy arbiraty sized chunks of memory in and
	  out of pagebufs without mapping the memory. Make all the code to
	  do with address space remapping not compile by default.



From owner-linux-xfs@oss.sgi.com Mon Mar 12 11:31:51 2001
Received:  by oss.sgi.com id <S553660AbRCLTbl>;
	Mon, 12 Mar 2001 11:31:41 -0800
Received: from deliverator.sgi.com ([204.94.214.10]:17976 "EHLO
        deliverator.sgi.com") by oss.sgi.com with ESMTP id <S553645AbRCLTbd>;
	Mon, 12 Mar 2001 11:31:33 -0800
Received: from ledzep.americas.sgi.com (ledzep.americas.sgi.com [137.38.226.97]) by deliverator.sgi.com (980309.SGI.8.8.8-aspam-6.2/980310.SGI-aspam) via ESMTP id LAA18986
	for <linux-xfs@oss.sgi.com>; Mon, 12 Mar 2001 11:30:25 -0800 (PST)
	mail_from (sandeen@gibble.americas.sgi.com)
Received: from gibble.americas.sgi.com (gibble.americas.sgi.com [128.162.195.80]) by ledzep.americas.sgi.com (SGI-SGI-8.9.3/americas-smart-nospam1.1) with ESMTP id NAA67549 for <linux-xfs@oss.sgi.com>; Mon, 12 Mar 2001 13:30:16 -0600 (CST)
Received: (from sandeen@localhost)
	by gibble.americas.sgi.com (8.11.2/8.11.2) id f2CJRoQ19141
	for linux-xfs@oss.sgi.com; Mon, 12 Mar 2001 14:27:50 -0500
Date:   Mon, 12 Mar 2001 14:27:50 -0500
From:   Eric Sandeen <sandeen@sgi.com>
Message-Id: <200103121927.f2CJRoQ19141@gibble.americas.sgi.com>
Subject: TAKE -  irix6.5f:irix:71608a,72205a
To:     unlisted-recipients:; (no To-header on input)
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing

Date:  Mon Mar 12 11:27:48 PST 2001
Workarea:  gibble.americas.sgi.com:/home/poppy20/sandeen/xfs/workarea

The following file(s) were checked into:
  bonnie.engr.sgi.com:/isms/slinx/2.4.x-xfs

Merge of Irix mods irix6.5f:irix:71608a & irix6.5f:irix:72205a
PVs affected:  793264, 798940

Modid:  2.4.x-xfs:slinx:89555a
linux/fs/xfs/xfs_bmap.c - 1.264
	- Do an unlock if an error condition is detected.
	  In xfs_bmap_add_attrfork() remove extra xfs_iunlock() under label error0.
	  Fix for bug 798940.



From owner-linux-xfs@oss.sgi.com Mon Mar 12 11:36:40 2001
Received:  by oss.sgi.com id <S553660AbRCLTga>;
	Mon, 12 Mar 2001 11:36:30 -0800
Received: from deliverator.sgi.com ([204.94.214.10]:28474 "EHLO
        deliverator.sgi.com") by oss.sgi.com with ESMTP id <S553645AbRCLTgS>;
	Mon, 12 Mar 2001 11:36:18 -0800
Received: from ledzep.americas.sgi.com (ledzep.americas.sgi.com [137.38.226.97]) by deliverator.sgi.com (980309.SGI.8.8.8-aspam-6.2/980310.SGI-aspam) via ESMTP id LAA20046
	for <linux-xfs@oss.sgi.com>; Mon, 12 Mar 2001 11:35:10 -0800 (PST)
	mail_from (sandeen@gibble.americas.sgi.com)
Received: from gibble.americas.sgi.com (gibble.americas.sgi.com [128.162.195.80]) by ledzep.americas.sgi.com (SGI-SGI-8.9.3/americas-smart-nospam1.1) with ESMTP id NAA75691 for <linux-xfs@oss.sgi.com>; Mon, 12 Mar 2001 13:35:01 -0600 (CST)
Received: (from sandeen@localhost)
	by gibble.americas.sgi.com (8.11.2/8.11.2) id f2CJWZW19242
	for linux-xfs@oss.sgi.com; Mon, 12 Mar 2001 14:32:35 -0500
Date:   Mon, 12 Mar 2001 14:32:35 -0500
From:   Eric Sandeen <sandeen@sgi.com>
Message-Id: <200103121932.f2CJWZW19242@gibble.americas.sgi.com>
Subject: TAKE - irix6.5f:irix:80991a
To:     unlisted-recipients:; (no To-header on input)
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing

Date:  Mon Mar 12 11:32:25 PST 2001
Workarea:  gibble.americas.sgi.com:/home/poppy20/sandeen/xfs/workarea

The following file(s) were checked into:
  bonnie.engr.sgi.com:/isms/slinx/2.4.x-xfs

Merge of Irix mod irix6.5f:irix:80991a
PVs affected:  810492
(Linux version wrapped in "#if CONFIG_XFS_DMAPI")

Modid:  2.4.x-xfs:slinx:89558a
linux/fs/xfs/xfs_log_recover.c - 1.203
	- In xlog_recover_process_iunlinks(), always clear the inode DMAPI event
	  mask to prevent any DMAPI events from being sent.  This is not a problem
	  for XFS since the file system is not mounted during recovery, but it is
	  a problem for CXFS which does recovery on mounted (but quiescent) file
	  systems.  Fix for bug 810492.



From owner-linux-xfs@oss.sgi.com Mon Mar 12 11:40:40 2001
Received:  by oss.sgi.com id <S553660AbRCLTkb>;
	Mon, 12 Mar 2001 11:40:31 -0800
Received: from deliverator.sgi.com ([204.94.214.10]:39227 "EHLO
        deliverator.sgi.com") by oss.sgi.com with ESMTP id <S553645AbRCLTkN>;
	Mon, 12 Mar 2001 11:40:13 -0800
Received: from ledzep.americas.sgi.com (ledzep.americas.sgi.com [137.38.226.97]) by deliverator.sgi.com (980309.SGI.8.8.8-aspam-6.2/980310.SGI-aspam) via ESMTP id LAA20646
	for <linux-xfs@oss.sgi.com>; Mon, 12 Mar 2001 11:39:05 -0800 (PST)
	mail_from (sandeen@gibble.americas.sgi.com)
Received: from gibble.americas.sgi.com (gibble.americas.sgi.com [128.162.195.80]) by ledzep.americas.sgi.com (SGI-SGI-8.9.3/americas-smart-nospam1.1) with ESMTP id NAA78339 for <linux-xfs@oss.sgi.com>; Mon, 12 Mar 2001 13:38:56 -0600 (CST)
Received: (from sandeen@localhost)
	by gibble.americas.sgi.com (8.11.2/8.11.2) id f2CJaUR19375
	for linux-xfs@oss.sgi.com; Mon, 12 Mar 2001 14:36:30 -0500
Date:   Mon, 12 Mar 2001 14:36:30 -0500
From:   Eric Sandeen <sandeen@sgi.com>
Message-Id: <200103121936.f2CJaUR19375@gibble.americas.sgi.com>
Subject: TAKE - irix6.5f:irix:82811a
To:     unlisted-recipients:; (no To-header on input)
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing

Date:  Mon Mar 12 11:36:49 PST 2001
Workarea:  gibble.americas.sgi.com:/home/poppy20/sandeen/xfs/workarea

The following file(s) were checked into:
  bonnie.engr.sgi.com:/isms/slinx/2.4.x-xfs

Merge Irix mod irix6.5f:irix:82811a
PVs affected:  750712

Modid:  2.4.x-xfs:slinx:89559a
linux/fs/xfs/xfs_log.c - 1.230
	- add another error injection area - to inject errors on log writes

linux/fs/xfs/xfs_error.h - 1.21
	- Restructure xlog_state_do_callbacks to not use GOTOs to create a loop.
	  When scanning the iclogs, always start over at log->l_iclog every time
	  the log lock is released.  It now continues processing iclogs until
	  all iclogs in DONE_SYNC or DO_CALLBACK state have been processed; this
	  is a potential endless loop so print a console message if it seems to
	  be caught in a loop.  However, if there is an I/O error on any iclog,
	  only go through the iclogs exactly once.



From owner-linux-xfs@oss.sgi.com Mon Mar 12 11:45:00 2001
Received:  by oss.sgi.com id <S553759AbRCLTok>;
	Mon, 12 Mar 2001 11:44:40 -0800
Received: from deliverator.sgi.com ([204.94.214.10]:57404 "EHLO
        deliverator.sgi.com") by oss.sgi.com with ESMTP id <S553652AbRCLTof>;
	Mon, 12 Mar 2001 11:44:35 -0800
Received: from ledzep.americas.sgi.com (ledzep.americas.sgi.com [137.38.226.97]) by deliverator.sgi.com (980309.SGI.8.8.8-aspam-6.2/980310.SGI-aspam) via ESMTP id LAA21283
	for <linux-xfs@oss.sgi.com>; Mon, 12 Mar 2001 11:43:27 -0800 (PST)
	mail_from (sandeen@gibble.americas.sgi.com)
Received: from gibble.americas.sgi.com (gibble.americas.sgi.com [128.162.195.80]) by ledzep.americas.sgi.com (SGI-SGI-8.9.3/americas-smart-nospam1.1) with ESMTP id NAA81050 for <linux-xfs@oss.sgi.com>; Mon, 12 Mar 2001 13:43:19 -0600 (CST)
Received: (from sandeen@localhost)
	by gibble.americas.sgi.com (8.11.2/8.11.2) id f2CJeqI19507
	for linux-xfs@oss.sgi.com; Mon, 12 Mar 2001 14:40:52 -0500
Date:   Mon, 12 Mar 2001 14:40:52 -0500
From:   Eric Sandeen <sandeen@sgi.com>
Message-Id: <200103121940.f2CJeqI19507@gibble.americas.sgi.com>
Subject: TAKE - irix6.5f:irix:87583b
To:     unlisted-recipients:; (no To-header on input)
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing

Date:  Mon Mar 12 11:41:17 PST 2001
Workarea:  gibble.americas.sgi.com:/home/poppy20/sandeen/xfs/workarea

The following file(s) were checked into:
  bonnie.engr.sgi.com:/isms/slinx/2.4.x-xfs

Last one for today.  :-)
Merge Irix mod irix6.5f:irix:87583b
PVs affected:  815195

Modid:  2.4.x-xfs:slinx:89561a
linux/fs/xfs/xfs_vnodeops.c - 1.492
	- If the file has attributes, set the xflags to indicate thus.
	  I can't see any real value in XFS.  In CXFS we can check
	  that flag to determine if an RPC is needed.

linux/include/linux/xfs_fs.h - 1.22
	- The value of XFS_XFLAG_HASATTR needs to be defined so that it
	  doesn't get sign extended on a 64 bit machine.



From owner-linux-xfs@oss.sgi.com Mon Mar 12 12:03:30 2001
Received:  by oss.sgi.com id <S553721AbRCLUDL>;
	Mon, 12 Mar 2001 12:03:11 -0800
Received: from sgi.SGI.COM ([192.48.153.1]:41560 "EHLO sgi.com")
	by oss.sgi.com with ESMTP id <S553652AbRCLUCn>;
	Mon, 12 Mar 2001 12:02:43 -0800
Received: from madurai.engr.sgi.com ([163.154.5.75]) 
	by sgi.com (980327.SGI.8.8.8-aspam/980304.SGI-aspam:
       SGI does not authorize the use of its proprietary
       systems or networks for unsolicited or bulk email
       from the Internet.) 
	via ESMTP id MAA01786
	for <linux-xfs@oss.sgi.com>; Mon, 12 Mar 2001 12:02:41 -0800 (PST)
	mail_from (ananth@sgi.com)
Received: from sgi.com (mango.engr.sgi.com [163.154.5.76]) by madurai.engr.sgi.com (980427.SGI.8.8.8/970903.SGI.AUTOCF) via ESMTP id LAA15199; Mon, 12 Mar 2001 11:56:35 -0800 (PST)
Message-ID: <3AAD2AB1.13EE0E67@sgi.com>
Date:   Mon, 12 Mar 2001 11:59:45 -0800
From:   Rajagopal Ananthanarayanan <ananth@sgi.com>
X-Mailer: Mozilla 4.72 [en] (X11; U; Linux 2.2.16-4SGI_20smp i686)
X-Accept-Language: en
MIME-Version: 1.0
To:     Steve Lord <lord@sgi.com>
CC:     Thomas Graichen <thomas.graichen@innominate.com>,
        linux-xfs@oss.sgi.com
Subject: Re: FAQ updated
References: <200103121557.f2CFvrk24545@jen.americas.sgi.com>
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing

Steve Lord wrote:

> > * is anyone using XFS in an production environment right now? - is
> >   anyone using XFS in a large scale (big SMP machine or large file-
> >   system) system on a regular basis?
> 
> I will have to leave it to external people to comment on this, I think
> there are some which would classify as production, I am not sure how
> big things get in terms of machine sizes etc.
> 

Internally, I've been using a 4 CPU, 1.8GB memory, 1 9G disk system
for all my build needs; this machine is also used as to build "pre-release"
versions of SGI ProPak.

There is also linux-xfs.sgi.com which is obviously a production system,
serving XFS web pages, downloads, etc.

ananth.

From owner-linux-xfs@oss.sgi.com Mon Mar 12 13:54:50 2001
Received:  by oss.sgi.com id <S553721AbRCLVyb>;
	Mon, 12 Mar 2001 13:54:31 -0800
Received: from pneumatic-tube.sgi.com ([204.94.214.22]:16404 "EHLO
        pneumatic-tube.sgi.com") by oss.sgi.com with ESMTP
	id <S553692AbRCLVyR>; Mon, 12 Mar 2001 13:54:17 -0800
Received: from snort.melbourne.sgi.com (snort.melbourne.sgi.com [134.14.55.149]) by pneumatic-tube.sgi.com (980327.SGI.8.8.8-aspam/980310.SGI-aspam) via ESMTP id OAA02576
	for <linux-xfs@oss.sgi.com>; Mon, 12 Mar 2001 14:04:05 -0800 (PST)
	mail_from (nathans@snort.melbourne.sgi.com)
Received: (from nathans@localhost)
	by snort.melbourne.sgi.com (SGI-8.9.3/8.9.3) id IAA13763
	for linux-xfs@oss.sgi.com; Tue, 13 Mar 2001 08:52:57 +1100 (EST)
Date:   Tue, 13 Mar 2001 08:52:57 +1100 (EST)
From:   Nathan Scott <nathans@snort.melbourne.sgi.com>
Message-Id: <200103122152.IAA13763@snort.melbourne.sgi.com>
To:     linux-xfs@oss.sgi.com
Subject: TAKE - srcdiff
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing

Date:  Mon Mar 12 13:51:11 PST 2001
Workarea:  snort.melbourne.sgi.com:/diskb/build4/nathans/linux-xfs

The following file(s) were checked into:
  bonnie.engr.sgi.com:/isms/slinx/2.4.x-xfs


Modid:  2.4.x-xfs:slinx:89576a
cmd/xfsprogs/libxfs/xfs_rtalloc.c - 1.2
	- sync with someones recent change to fix xfs_buf_t tab offseting -
	  doesn't affect userspace at all, but keep routine in-sync.

cmd/xfsprogs/include/xfs_inode.h - 1.3
	- sync with recent change to kernel struct xfs_inode - doesn't
	  affect userspace at all, but keep files in-sync.



From owner-linux-xfs@oss.sgi.com Mon Mar 12 20:17:51 2001
Received:  by oss.sgi.com id <S553655AbRCMERb>;
	Mon, 12 Mar 2001 20:17:31 -0800
Received: from tux.mkp.net ([130.225.60.11]:60934 "EHLO tux.mkp.net")
	by oss.sgi.com with ESMTP id <S553645AbRCMERF>;
	Mon, 12 Mar 2001 20:17:05 -0800
Received: from tux.mkp.net ([130.225.60.11] helo=jaguar.mkp.net)
	by tux.mkp.net with esmtp (Exim 3.16 #1)
	id 14cgF4-0001uq-00; Tue, 13 Mar 2001 05:17:03 +0100
Received: (from mkp@localhost)
	by jaguar.mkp.net (8.9.3/8.9.3) id XAA10588;
	Mon, 12 Mar 2001 23:15:37 -0500
X-Authentication-Warning: jaguar.mkp.net: mkp set sender to mkp@mkp.net using -f
To:     linux-xfs@oss.sgi.com
Subject: TAKE - MD RAID1 and RAID5 fixes
From:   "Martin K. Petersen" <mkp@linuxcare.com>
Organization: Linuxcare, Inc.
Date:   12 Mar 2001 23:15:37 -0500
Message-ID: <yq1pufmwbh2.fsf@jaguar.mkp.net>
Lines:  18
User-Agent: Gnus/5.0808 (Gnus v5.8.8) XEmacs/21.1 (Canyonlands)
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing


Date:  Mon Mar 12 20:15:16 PST 2001
Workarea:  sshgate.corp.sgi.com:/home/mkp/XFS/slinx-md

The following file(s) were checked into:
  bonnie.engr.sgi.com:/isms/slinx/2.4.x-xfs


Modid:  2.4.x-xfs:slinx:89619a
linux/drivers/md/raid1.c - 1.6
linux/drivers/md/raid5.c - 1.10
linux/drivers/md/md.c - 1.9
        - Resync fixes for MD RAID1 and RAID5

-- 
Martin K. Petersen, Principal Linux Consultant, Linuxcare, Inc.
mkp@linuxcare.com, http://www.linuxcare.com/
SGI XFS for Linux Developer, http://oss.sgi.com/projects/xfs/

From owner-linux-xfs@oss.sgi.com Mon Mar 12 21:34:53 2001
Received:  by oss.sgi.com id <S553655AbRCMFen>;
	Mon, 12 Mar 2001 21:34:43 -0800
Received: from tux.mkp.net ([130.225.60.11]:13319 "EHLO tux.mkp.net")
	by oss.sgi.com with ESMTP id <S553645AbRCMFeV>;
	Mon, 12 Mar 2001 21:34:21 -0800
Received: from tux.mkp.net ([130.225.60.11] helo=jaguar.mkp.net)
	by tux.mkp.net with esmtp (Exim 3.16 #1)
	id 14chRp-0001xr-00; Tue, 13 Mar 2001 06:34:19 +0100
Received: (from mkp@localhost)
	by jaguar.mkp.net (8.9.3/8.9.3) id AAA10604;
	Tue, 13 Mar 2001 00:32:53 -0500
X-Authentication-Warning: jaguar.mkp.net: mkp set sender to mkp@mkp.net using -f
To:     linux-xfs@oss.sgi.com
Subject: TAKE - Upgrade xfsprogs to LVM 0.9 / IOP10
From:   "Martin K. Petersen" <mkp@linuxcare.com>
Organization: Linuxcare, Inc.
Date:   13 Mar 2001 00:32:53 -0500
Message-ID: <yq1hf0yw7wa.fsf@jaguar.mkp.net>
Lines:  22
User-Agent: Gnus/5.0808 (Gnus v5.8.8) XEmacs/21.1 (Canyonlands)
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing


Date:  Mon Mar 12 21:32:06 PST 2001
Workarea:  sshgate.corp.sgi.com:/export/d0/mkp/XFS/slinx-lvm

The following file(s) were checked into:
  bonnie.engr.sgi.com:/isms/slinx/2.4.x-xfs


Modid:  2.4.x-xfs:slinx:89622a
cmd/xfsprogs/include/lvm_log.h - 1.1
cmd/xfsprogs/include/lvm_config.h - 1.1
cmd/xfsprogs/include/lvm_user.h - 1.2
cmd/xfsprogs/include/lvm.h - 1.2
cmd/xfsprogs/include/liblvm.h - 1.2
cmd/xfsprogs/doc/CHANGES - 1.6
cmd/xfsprogs/VERSION - 1.6
        - Upgrade LVM support to 0.9/IOP 10.

-- 
Martin K. Petersen, Principal Linux Consultant, Linuxcare, Inc.
mkp@linuxcare.com, http://www.linuxcare.com/
SGI XFS for Linux Developer, http://oss.sgi.com/projects/xfs/

From owner-linux-xfs@oss.sgi.com Mon Mar 12 22:16:13 2001
Received:  by oss.sgi.com id <S553659AbRCMGPx>;
	Mon, 12 Mar 2001 22:15:53 -0800
Received: from tux.mkp.net ([130.225.60.11]:29191 "EHLO tux.mkp.net")
	by oss.sgi.com with ESMTP id <S553655AbRCMGPh>;
	Mon, 12 Mar 2001 22:15:37 -0800
Received: from tux.mkp.net ([130.225.60.11] helo=jaguar.mkp.net)
	by tux.mkp.net with esmtp (Exim 3.16 #1)
	id 14ci5m-0001zc-00; Tue, 13 Mar 2001 07:15:35 +0100
Received: (from mkp@localhost)
	by jaguar.mkp.net (8.9.3/8.9.3) id BAA10621;
	Tue, 13 Mar 2001 01:14:05 -0500
X-Authentication-Warning: jaguar.mkp.net: mkp set sender to mkp@mkp.net using -f
To:     "Bas" <list@showme.wox.org>
Cc:     <linux-xfs@oss.sgi.com>
Subject: Re: 2.4.2, LVM 0.9.1 beta 5a and XFS.
References: <000a01c0a93f$22643440$0f01a8c0@myplace.net>
	<yq14rx1zp0m.fsf@jaguar.mkp.net>
From:   "Martin K. Petersen" <mkp@linuxcare.com>
Organization: Linuxcare, Inc.
Date:   13 Mar 2001 01:14:05 -0500
In-Reply-To: <yq14rx1zp0m.fsf@jaguar.mkp.net>
Message-ID: <yq14rwyw5zm.fsf@jaguar.mkp.net>
Lines:  19
User-Agent: Gnus/5.0808 (Gnus v5.8.8) XEmacs/21.1 (Canyonlands)
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing

>>>>> "Martin" == Martin K Petersen <mkp@linuxcare.com> writes:

>>>>> "Bas" == Bas  <list@showme.wox.org> writes:
Bas> I got the tree through CVS, and compiled it last week. When I
Bas> extend the LV that XFS is on, and I run xfs_growfs, it returns:

Bas> xfs_growfs: cannot determine geometry of filesystem mounted at
Bas> /mnt: Invalid argument

Martin> I'll put it on my list to have a look.

Ok, so I had a look.  It works like a charm here.

Are you still experiencing this behaviour?

-- 
Martin K. Petersen, Principal Linux Consultant, Linuxcare, Inc.
mkp@linuxcare.com, http://www.linuxcare.com/
SGI XFS for Linux Developer, http://oss.sgi.com/projects/xfs/

From owner-linux-xfs@oss.sgi.com Mon Mar 12 22:25:13 2001
Received:  by oss.sgi.com id <S553655AbRCMGYy>;
	Mon, 12 Mar 2001 22:24:54 -0800
Received: from pneumatic-tube.sgi.com ([204.94.214.22]:47437 "EHLO
        pneumatic-tube.sgi.com") by oss.sgi.com with ESMTP
	id <S553645AbRCMGYn>; Mon, 12 Mar 2001 22:24:43 -0800
Received: from snort.melbourne.sgi.com (snort.melbourne.sgi.com [134.14.55.149]) by pneumatic-tube.sgi.com (980327.SGI.8.8.8-aspam/980310.SGI-aspam) via ESMTP id WAA04948
	for <linux-xfs@oss.sgi.com>; Mon, 12 Mar 2001 22:34:32 -0800 (PST)
	mail_from (nathans@snort.melbourne.sgi.com)
Received: (from nathans@localhost)
	by snort.melbourne.sgi.com (SGI-8.9.3/8.9.3) id RAA16808
	for linux-xfs@oss.sgi.com; Tue, 13 Mar 2001 17:23:24 +1100 (EST)
Date:   Tue, 13 Mar 2001 17:23:24 +1100 (EST)
From:   Nathan Scott <nathans@snort.melbourne.sgi.com>
Message-Id: <200103130623.RAA16808@snort.melbourne.sgi.com>
To:     linux-xfs@oss.sgi.com
Subject: TAKE - quota
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing

This is a small change in the quotactl syscall semantics for
XFS - current kernel needs current tools otherwise blk counts
will come out funny.  This is in preparation for the merge of
quota-2.00 + XFS + quota-3.00 userspace tools (which is a work
in progress).

cheers.


Date:  Mon Mar 12 22:15:13 PST 2001
Workarea:  snort.melbourne.sgi.com:/diskb/build4/nathans/linux-xfs

The following file(s) were checked into:
  bonnie.engr.sgi.com:/isms/slinx/2.4.x-xfs


Modid:  2.4.x-xfs:slinx:89621a
linux/fs/xfs/xfs_qm_syscalls.c - 1.50
cmd/quota/quotaops.c - 1.9
cmd/quota/repquota.c - 1.15
cmd/quota/rquota_server.c - 1.6
cmd/quota/warnquota.c - 1.5
	- blksize changes - revert to IRIX mechanism for blksize in quota -
	  makes things a little more obtuse for this version of the quota
	  user tools (2.00), but simplifies things alot for the new version.

SOURCES/quota-2.00-xfs.patch - 1.4
	- recent changes - hasxfsquota() & label/uuid fix, blksize changes,
	  other minor tidyups.


From owner-linux-xfs@oss.sgi.com Tue Mar 13 00:52:33 2001
Received:  by oss.sgi.com id <S553671AbRCMIwY>;
	Tue, 13 Mar 2001 00:52:24 -0800
Received: from [211.192.224.67] ([211.192.224.67]:36384 "EHLO
        mailsvr.kornet.net") by oss.sgi.com with ESMTP id <S553648AbRCMIwC>;
	Tue, 13 Mar 2001 00:52:02 -0800
Received: from realweb21.com [10.10.1.1] by mailsvr.kornet.net
  (SMTPD32-6.04) id AB4A3A054A; Mon, 12 Mar 2001 21:04:26 +0900
Message-ID: <000066a45228$00002320$00006767@usun.com>
To:     <nofees@digitex.com.au>
From:   techsupport243@usun.com
Subject: pay nothing for your conference calls!                         26471
Date:   Mon, 12 Mar 2001 04:09:15 -0800
MIME-Version: 1.0
Content-Type: text/html;
	charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable
X-Priority: 3
X-MSMail-Priority: Normal
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing

<HTML>
<BODY>

<FONT face=3D"MS Sans Serif">
<FONT size=3D2> <!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.0 Transitional//=
EN"><BR>
<HTML><HEAD><TITLE>Take Control Of Your Conference Calls</TITLE><META http=
-equiv=3DContent-Type content=3D"text/html; charset=3Dwindows-1252"><META =
content=3D"MSHTML 5.50.4134.100" name=3DGENERATOR></HEAD><BR>
<BODY vLink=3D#c0c0c0 link=3D#c0c0c0 bgColor=3D#000000 leftMargin=3D0><PRE=
><FONT face=3Darial,helvetica><BR>
<HEAD><META content=3DFrontPage.Editor.Document name=3DProgId><BR>
<DIV align=3Dcenter><CENTER><TABLE height=3D789 cellSpacing=3D0 cellPaddin=
g=3D0 width=3D602 border=3D0><TBODY><TR vAlign=3Dtop><TD width=3D602 heigh=
t=3D452><DIV align=3Dcenter><TABLE width=3D470 border=3D0><TBODY><TR><TD><=
P align=3Dcenter><FONT color=3D#0000ff size=3D7><B><FONT color=3Dffffff si=
ze=3D6>Paying Too Much For Your </FONT></B></FONT><FONT color=3D#ffffff si=
ze=3D6><B>Conference Calls?</B></FONT></P></TD></TR></TBODY></TABLE><TABLE=
 width=3D352 border=3D0><TBODY><TR><TD><P align=3Dcenter><FONT color=3D#ff=
0000 size=3D5>Only <U><B>.18 Cents </B></U>per minute (Including long dist=
ance!)</FONT></P></TD></TR></TBODY></TABLE></DIV><UL><UL><UL><UL><UL><LI><=
DIV align=3Dleft><FONT color=3Dffffff size=3D3><B>No setup fees</B></FONT>=
</DIV><LI><DIV align=3Dleft><FONT color=3Dffffff size=3D3><B>No contracts =
or monthly fees</B></FONT></DIV><LI><DIV align=3Dleft><FONT color=3Dffffff=
 size=3D3><B>Call anytime, from anywhere, to anywhere</B></FONT></DIV><LI>=
<DIV align=3Dleft><FONT color=3Dffffff size=3D3><B>International Dial In 1=
8 cents per minute</B></FONT></DIV><LI><DIV align=3Dleft><FONT color=3Dfff=
fff size=3D3><B>Connects up to 100 participants</B></FONT></DIV><LI><DIV a=
lign=3Dleft><FONT color=3Dffffff size=3D3><B>Operator Help available 24/7<=
/B></FONT></DIV></LI></UL></UL></UL></UL></UL><DIV align=3Dcenter><TABLE w=
idth=3D424 border=3D0><TBODY><TR><TD><DIV align=3Dcenter><FONT color=3D#ff=
0000 size=3D6><B><FONT size=3D5>Get the best quality, the easiest to use,<=
/FONT></B></FONT> <FONT color=3D#ff0000 size=3D5><B>and lowest rate in the=
 industry.</B></FONT></DIV></TD></TR></TBODY></TABLE><TABLE width=3D300 bo=
rder=3D0><TBODY><TR><TD height=3D73><DIV align=3Dcenter><P align=3Dcenter>=
<FONT color=3D#ffffff size=3D4>If you like saving money, fill out the form=
 below and one of our consultants will contact you.</FONT></P></DIV></TD><=
/TR></TBODY></TABLE></DIV></BLOCKQUOTE></BLOCKQUOTE><DIV align=3Dcenter><P=
 align=3Dcenter><FONT color=3D#ff0000 size=3D2>Required Input Field</FONT>=
<FONT color=3D#ff0000 size=3D2>*</FONT></P><TABLE cellSpacing=3D0 borderCo=
lorDark=3D#333300 cellPadding=3D3 width=3D600 borderColorLight=3D#ffffcc><=
TBODY><TR><TD><FORM action=3D"mailto:mailbox217@excite.com?subject=3DConfe=
rence_Inquiry" method=3Dpost encType=3Dtext/plain><DIV align=3Dcenter><TAB=
LE width=3D"100%"><TBODY><TR><TD width=3D"49%"><DIV align=3Dright><FONT fa=
ce=3D"Arial, Helvetica, sans-serif" color=3D#ff0000 size=3D2>Name*</FONT><=
/DIV></TD><TD width=3D"51%"><FONT color=3D#ff0000><INPUT name=3DNAME> </FO=
NT></TD></TR><TR><TD width=3D"49%"><DIV align=3Dright><FONT face=3D"Arial,=
 Helvetica, sans-serif" color=3D#ff0000 size=3D2>Web Address*</FONT></DIV>=
</TD><TD width=3D"51%"><FONT color=3D#ff0000><INPUT value=3Dhttp:// name=3D=
URL> </FONT></TD></TR><TR><TD width=3D"49%"><DIV align=3Dright><FONT face=3D=
"Arial, Helvetica, sans-serif" color=3D#ff0000 size=3D2>Company Name*</FON=
T></DIV></TD><TD width=3D"51%"><FONT color=3D#ff0000><INPUT name=3DCOMPANY=
_NAME> </FONT></TD></TR><TR><TD width=3D"49%"><DIV align=3Dright><FONT fac=
e=3D"Arial, Helvetica, sans-serif" color=3D#ff0000 size=3D2>State</FONT></=
DIV></TD><TD width=3D"51%"><FONT color=3D#ff0000><INPUT size=3D2 name=3DST=
ATE> </FONT></TD></TR><TR><TD width=3D"49%"><DIV align=3Dright><FONT face=3D=
"Arial, Helvetica, sans-serif" color=3D#ff0000 size=3D2>Business Phone*</F=
ONT></DIV></TD><TD width=3D"51%"><FONT color=3D#ff0000><INPUT name=3DBUS_P=
HONE> </FONT></TD></TR><TR><TD width=3D"49%"><DIV align=3Dright><FONT face=
=3D"Arial, Helvetica, sans-serif" color=3D#ff0000 size=3D2>Home Phone</FON=
T></DIV></TD><TD width=3D"51%"><FONT color=3D#ff0000><INPUT name=3DHOME_PH=
ONE> </FONT></TD></TR><TR><TD width=3D"49%"><DIV align=3Dright><FONT face=3D=
"Arial, Helvetica, sans-serif" color=3D#ff0000 size=3D2>E-mail*</FONT></DI=
V></TD><TD width=3D"51%"><FONT color=3D#ff0000><INPUT name=3DEMAIL> </FONT=
></TD></TR><TR><TD width=3D"49%"><DIV align=3Dright><FONT face=3D"Arial, H=
elvetica, sans-serif" color=3D#ff0000 size=3D2>Type of Business</FONT></DI=
V></TD><TD width=3D"51%"><FONT color=3D#ff0000><INPUT name=3DTYPE_OF_BUSIN=
ESS> </FONT></TD></TR></TBODY></TABLE><P><BR>&nbsp;&nbsp;&nbsp;&nbsp;&nbsp=
;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&=
nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nb=
sp;&nbsp;&nbsp;&nbsp;&nbsp; <INPUT type=3Dsubmit value=3D"Submit Informati=
on" name=3Dsubmit> </P><P><FONT face=3D"Arial, Helvetica, sans-serif" colo=
r=3D#999999 size=3D1>If you received this e-mail in error or would like to=
 be removed,</FONT><FONT face=3D"Arial, Helvetica, sans-serif" color=3D#ff=
0000 size=3D1> <A href=3D"mailto:remove991@excite.com?subject=3DRemoveName=
_Conferencing"><FONT color=3D#666666>PLEASE CLICK HERE</FONT></A><FONT col=
or=3D#666666>.</FONT></FONT></P></DIV></FORM></TD></TR></TBODY></TABLE><P>=
 </P></DIV></TD></TR></TBODY></TABLE><BR>
    </CENTER><BR>
</DIV><BR>
<DIV align=3Dleft> <BR>
  <DIV align=3Dcenter><BR>
    <P>&nbsp;</P><BR>
    </DIV><BR>
</DIV><BR>
</PRE></FONT></BODY></HTML><BR>
</FONT></FONT>



From owner-linux-xfs@oss.sgi.com Tue Mar 13 03:32:14 2001
Received:  by oss.sgi.com id <S553671AbRCMLby>;
	Tue, 13 Mar 2001 03:31:54 -0800
Received: from hermes.mixx.net ([212.84.196.2]:30727 "HELO hermes.mixx.net")
	by oss.sgi.com with SMTP id <S553648AbRCMLbn>;
	Tue, 13 Mar 2001 03:31:43 -0800
Received: from mate.bln.innominate.de (cerberus.berlin.innominate.de [212.84.234.251])
	by hermes.mixx.net (Postfix) with ESMTP id 9A0FCF80C
	for <linux-xfs@oss.sgi.com>; Tue, 13 Mar 2001 12:31:40 +0100 (CET)
Received: by mate.bln.innominate.de (Postfix, from userid 9)
	id 5F9A12CA70; Tue, 13 Mar 2001 12:31:37 +0100 (CET)
From:   Thomas Graichen <news-innominate.list.sgi.xfs@innominate.de>
Reply-To: Thomas Graichen <thomas.graichen@innominate.com>
X-Newsgroups: innominate.list.sgi.xfs
Subject: Re: FAQ updated
Date:   13 Mar 2001 11:31:37 GMT
Organization: innominate AG, Berlin, Germany
Lines:  17
Distribution: local
Message-ID: <news2mail-98l0ep$fr4$2@mate.bln.innominate.de>
References: <98hstr$rjm$1@mate.bln.innominate.de>
X-Trace: mate.bln.innominate.de 984483097 16228 10.0.0.31 (13 Mar 2001 11:31:37 GMT)
X-Complaints-To: news@innominate.de
User-Agent: tin/1.4.4-20000803 ("Vet for the Insane") (UNIX) (Linux/2.4.2-XFS (i686))
To:     linux-xfs@oss.sgi.com
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing

Thomas Graichen <thomas.graichen@innominate.com> wrote:
> * what is the current state of XFS on sparc64? - on the alpha (i may
>   try it on the one here this week - but maybe someone else has it
>   running right now)? - other arches so far?

just checked it on the alpha with the current cvs tree and it seems
to work just fine out of the box (with blocksize 8k) ... does any-
one know anything about the sparc64 efforts? - ia64 was running
quite well too if i remember correctly

t

-- 
thomas.graichen@innominate.com
                                                         innominate AG
                                                  the linux architects
tel: +49-30-308806-13   fax: -77             http://www.innominate.com

From owner-linux-xfs@oss.sgi.com Tue Mar 13 07:04:55 2001
Received:  by oss.sgi.com id <S553678AbRCMPEf>;
	Tue, 13 Mar 2001 07:04:35 -0800
Received: from tux.mkp.net ([130.225.60.11]:48648 "EHLO tux.mkp.net")
	by oss.sgi.com with ESMTP id <S553671AbRCMPEb>;
	Tue, 13 Mar 2001 07:04:31 -0800
Received: from tux.mkp.net ([130.225.60.11] helo=jaguar.mkp.net)
	by tux.mkp.net with esmtp (Exim 3.16 #1)
	id 14cqLW-0002GV-00; Tue, 13 Mar 2001 16:04:28 +0100
Received: (from mkp@localhost)
	by jaguar.mkp.net (8.9.3/8.9.3) id KAA10696;
	Tue, 13 Mar 2001 10:02:54 -0500
X-Authentication-Warning: jaguar.mkp.net: mkp set sender to mkp@mkp.net using -f
To:     Thomas Graichen <thomas.graichen@innominate.com>
Cc:     linux-xfs@oss.sgi.com
Subject: Re: FAQ updated
References: <98hstr$rjm$1@mate.bln.innominate.de>
	<news2mail-98l0ep$fr4$2@mate.bln.innominate.de>
From:   "Martin K. Petersen" <mkp@linuxcare.com>
Organization: Linuxcare, Inc.
Date:   13 Mar 2001 10:02:54 -0500
In-Reply-To: <news2mail-98l0ep$fr4$2@mate.bln.innominate.de>
Message-ID: <yq1vgpdvhi9.fsf@jaguar.mkp.net>
Lines:  14
User-Agent: Gnus/5.0808 (Gnus v5.8.8) XEmacs/21.1 (Canyonlands)
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing

>>>>> "Thomas" == Thomas Graichen <news-innominate.list.sgi.xfs@innominate.de> writes:

Thomas> just checked it on the alpha with the current cvs tree and it
Thomas> seems to work just fine out of the box (with blocksize 8k)
Thomas> ... does any- one know anything about the sparc64 efforts? -

I check sparc64 and alpha on a semi-regular basis.  My primary
workstations happen to be a Sun and an Alpha respectively, and they
run whatever was latest and greatest last time I rebooted them.

-- 
Martin K. Petersen, Principal Linux Consultant, Linuxcare, Inc.
mkp@linuxcare.com, http://www.linuxcare.com/
SGI XFS for Linux Developer, http://oss.sgi.com/projects/xfs/

From owner-linux-xfs@oss.sgi.com Tue Mar 13 07:33:55 2001
Received:  by oss.sgi.com id <S553671AbRCMPdg>;
	Tue, 13 Mar 2001 07:33:36 -0800
Received: from pneumatic-tube.sgi.com ([204.94.214.22]:60286 "EHLO
        pneumatic-tube.sgi.com") by oss.sgi.com with ESMTP
	id <S552774AbRCMPdI>; Tue, 13 Mar 2001 07:33:08 -0800
Received: from zeus-fddi.americas.sgi.com (zeus-fddi.americas.sgi.com [128.162.8.103]) by pneumatic-tube.sgi.com (980327.SGI.8.8.8-aspam/980310.SGI-aspam) via ESMTP id HAA03793
	for <linux-xfs@oss.sgi.com>; Tue, 13 Mar 2001 07:42:59 -0800 (PST)
	mail_from (lord@sgi.com)
Received: from daisy-e185.americas.sgi.com (daisy-e194.americas.sgi.com [128.162.194.214]) by zeus-fddi.americas.sgi.com (8.9.3/americas-smart-nospam1.1) with ESMTP id JAA1118923; Tue, 13 Mar 2001 09:31:25 -0600 (CST)
Received: from jen.americas.sgi.com (IDENT:root@jen.americas.sgi.com [128.162.184.86]) by daisy-e185.americas.sgi.com (SGI-8.9.3/SGI-server-1.7) with ESMTP id JAA79510; Tue, 13 Mar 2001 09:31:24 -0600 (CST)
Received: from jen.americas.sgi.com by jen.americas.sgi.com (8.11.0/SGI-client-1.7) via ESMTP id f2DFWfd06445; Tue, 13 Mar 2001 09:32:41 -0600
Message-Id: <200103131532.f2DFWfd06445@jen.americas.sgi.com>
X-Mailer: exmh version 2.2 06/23/2000 with nmh-1.0.4
To:     dave <doood@gruntwerk.net>
cc:     linux-xfs@oss.sgi.com
Subject: Re: nfs issues 
References: <200103101414.f2AEEI903763@jen.americas.sgi.com> <3AAD6F05.F221B0F2@gruntwerk.net>
Comments: In-reply-to dave <doood@gruntwerk.net>
   message dated "Mon, 12 Mar 2001 16:51:17 -0800."
Date:   Tue, 13 Mar 2001 09:32:41 -0600
From:   Steve Lord <lord@sgi.com>
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing


It looks like your code base is quite out of sync with the current CVS tree,
well about 2 weeks anyway. The code in question changed on feb 27th and again
on march 5th. Let me know if this happens with current cvs code, and please
run the oops output through ksymoops - it is impossible to interpret the
data without symbol names attached.

Thanks

   Steve


> Steve Lord wrote:
> 
> > > Greetings,
> > >
> > >     I have been running the xfs filesystem for about 1mo now on a redhat
> > > 7.0 system. The box was installed using the SGI RedHat install iso, and
> > > its kernel re-compiled using the CVS tree.  Its /usr directory is
> > > exported to 4 other linux boxes over nfs v2 and v3. We are constanly
> > > having : nfs, cant get request slot errors between our workstations and
> > > the server. At times the kernel will dump a stack trace and print out:
> > > kernel bug 1339! I'm assuming it caused by the md driver, or the nfs
> > > load, or a combination of both. Any suggestions to a work around? Are
> > > there any new drivers I wouldn't get from CVS?
> > > Thanks in advance.
> > >
> > > Dave
> >
> > I have seen this error from my workstation accessing NFS servers which are
> > not running Linux, it may be a generic problem. This message comes from
> > this code:
> >
> >         if (clnt->cl_chatty && !(task->tk_flags & RPC_CALL_MAJORSEEN)) {
> >                 task->tk_flags |= RPC_CALL_MAJORSEEN;
> >                 if (req)
> >                         printk(KERN_NOTICE "%s: server %s not responding, s
> till trying\n",
> >                                 clnt->cl_protname, clnt->cl_server);
> > #ifdef RPC_DEBUG
> >                 else
> >                         printk(KERN_NOTICE "%s: task %d can't get a request
>  slot\n",
> >                                 clnt->cl_protname, task->tk_pid);
> > #endif
> >         }
> >
> > in an rpc timeout path, so this is a request timeout.
> >
> > I am not sure if this could be happening because or XFS or not. How recent 
> is
> > your kernel from cvs, there were changes this week (not complete until
> > yesterday) which help nfs performance considerably.
> >
> > Steve
> 
> Here's the exact kernel message:
> 
> kernel BUG at buffer.c:1399!
> invalid operand: 0000
> CPU:    0
> EIP:    0010:[<c01315b3>]
> EFLAGS: 00010292
> eax: 0000001d   ebx: c541de40   ecx: 00000000   edx: ffffffff
> esi: 000005a8   edi: c4eb2000   ebp: 00000000   esp: c6dfbd98
> ds: 0018   es: 0018   ss: 0018
> Process nfsd (pid: 553, stackpage=c6dfb000)
> Stack: c0297da5 c029801a 00000577 000005a8 000005a8 c4eb2000 00000a58 c541de4
> 0
>        c0122133 c11621d0 00000a58 c6dfbe00 c50f632c c50f6324 c50f6334 0000000
> 0
>        0002abb8 c11621d0 c6dfbe00 c012227b 00000048 c6e30c14 00000a58 0000000
> 0
> Call Trace: [<c0122133>] [<c012227b>] [<c01205e4>] [<c0142c5a>] [<c01ce2c4>] 
> [<c01d9985>]
> [<c01ce2c4>]
>        [<c0142d75>] [<c016a8fc>] [<c0168212>] [<c0167bf6>] [<c0285ef8>] [<c01
> 679c1>]
> [<c0107537>]
> 
> Code: 0f 0b 83 c4 0c 31 c0 66 8b 43 08 8b 73 28 8d 3c 28 39 6c 24
> 
> We have jsut rebuilt from CVS as of today so we'll see how it goes. Thanks ag
> ain.
> 
> Dave
> 



From owner-linux-xfs@oss.sgi.com Tue Mar 13 18:52:02 2001
Received:  by oss.sgi.com id <S553675AbRCNCvn>;
	Tue, 13 Mar 2001 18:51:43 -0800
Received: from [211.106.66.169] ([211.106.66.169]:13322 "EHLO w10.crezio.com")
	by oss.sgi.com with ESMTP id <S553653AbRCNCvV>;
	Tue, 13 Mar 2001 18:51:21 -0800
Received: by w10.crezio.com id LAA0000018576; Wed, 14 Mar 2001 11:50:55 +0900 (KST)
From:   <biotechstox41@juno.com>
To:     linux-xfs@oss.sgi.com
Subject: FREE Biotech Stock Info!    164
Date:   Tue, 13 Mar 2001 21:48:08
Message-Id: <263.463087.198671@juno.com>
Reply-To: biotechinfo2004@yahoo.com
Mime-Version: 1.0
Content-Type: text/html; charset="us-ascii"
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing


<html>

<head>
<meta http-equiv="Content-Type" content="text/html; charset=windows-1252">
<meta name="GENERATOR" content="Microsoft FrontPage 4.0">
<meta name="ProgId" content="FrontPage.Editor.Document">
<title>Do you want to capitalize on the Biotech Revolution</title>
</head>

<body>

<p class="MsoNormal" style="mso-margin-top-alt:auto;mso-margin-bottom-alt:auto" align="center"><img border="0" src="http://www.geocities.com/mailtestbox2000/Kiloh_logo.gif" width="204" height="170"></p>
<p class="MsoNormal" style="mso-margin-top-alt:auto;mso-margin-bottom-alt:auto"><span style="font-family:Arial">Do
you want to capitalize on the Biotech Revolution? Would you like to add
groundbreaking biotech, pharmaceutical and medical device companies to your
portfolio mix? Does hearing about exciting IPO and private placement offerings
from life sciences companies interest you?</span></p>
<p class="MsoNormal" style="mso-margin-top-alt:auto;mso-margin-bottom-alt:auto"><span style="font-family:Arial">The
exclusive <b>Ruddy-Carlisle Biotech Infoline</b> service keeps you abreast of
investment opportunities in the life sciences space. Just sign up for it once
and get important information instantly delivered to study at your leisure. Our
service is <b><u>100% FREE</u></b>! <b><span style="color:blue"><a href="mailto:biotechsubscribe@yahoo.com">Sign
up!</a></span></b></span></p>
<p class="MsoNormal" style="mso-margin-top-alt:auto;mso-margin-bottom-alt:auto"><b><i><span style="font-size:11.0pt;mso-bidi-font-size:12.0pt;font-family:Arial;color:#003366">Ruddy-Carlisle
Biotech Infoline:</span></i></b></p>
<ul type="disc">
  <li class="MsoNormal" style="color:#003366;mso-margin-top-alt:auto;mso-margin-bottom-alt:
     auto;mso-list:l0 level1 lfo1;tab-stops:list .5in"><b><i><span style="font-size:11.0pt;mso-bidi-font-size:12.0pt;font-family:Arial">Instantly
    delivers key life sciences investment information directly to you! </span></i></b><o:p>
    </o:p>
  </li>
  <li class="MsoNormal" style="color:#003366;mso-margin-top-alt:auto;mso-margin-bottom-alt:
     auto;mso-list:l0 level1 lfo1;tab-stops:list .5in"><b><i><span style="font-size:11.0pt;mso-bidi-font-size:12.0pt;font-family:Arial">Learn
    about biotech, pharmaceutical &amp; medical device investment opportunities
    before others! </span></i></b><o:p>
    </o:p>
  </li>
  <li class="MsoNormal" style="color:#003366;mso-margin-top-alt:auto;mso-margin-bottom-alt:
     auto;mso-list:l0 level1 lfo1;tab-stops:list .5in"><b><i><span style="font-size:11.0pt;mso-bidi-font-size:12.0pt;font-family:Arial">Includes
    IPO &amp; private placement information! </span></i></b><o:p>
    </o:p>
  </li>
  <li class="MsoNormal" style="mso-margin-top-alt:auto;mso-margin-bottom-alt:auto;
     mso-list:l0 level1 lfo1;tab-stops:list .5in"><b><i><span style="font-size:
     11.0pt;mso-bidi-font-size:12.0pt;font-family:Arial;color:#003366">100%
    FREE!</span></i></b></li>
</ul>
<p class="MsoNormal" style="mso-margin-top-alt:auto;mso-margin-bottom-alt:auto"><span style="font-family:Arial">For
the entire last decade there were only three profitable biotech companies. At
the end of this year, ten are projected. At the end of 2003, <u>over forty</u>
are projected! The genomic promise is about to be delivered and investors know
it. The <b>Ruddy-Carlisle Biotech Infoline </b>provides you with critical,
decision-making, information that aids the chance of investment success in this
lucrative space. <b><span style="color:blue"><a href="mailto:biotechsubscribe@yahoo.com">Sign
up!</a></span></b></span></p>
<p class="MsoNormal" style="mso-margin-top-alt:auto;mso-margin-bottom-alt:auto"><b><span style="font-family:Arial">Please
Note-</span></b><span style="font-family:Arial"> Your information will only be
shared with companies that are in the life sciences space <u>and</u> pass our
rigorous inspection. Only the best opportunities will come to you.
Ruddy-Carlisle respects your privacy. <b><span style="color:blue"><a href="mailto:biotechsubscribe@yahoo.com">Sign
up!</a></span></b></span></p>
<p class="MsoNormal" style="mso-margin-top-alt:auto;mso-margin-bottom-alt:auto">&nbsp;</p>
<p class="MsoNormal" style="mso-margin-top-alt:auto;mso-margin-bottom-alt:auto">&nbsp;</p>
<b><span style="font-size:10.0pt;mso-bidi-font-size:12.0pt;font-family:Arial;
mso-fareast-font-family:&quot;Times New Roman&quot;;mso-ansi-language:EN-US;mso-fareast-language:
EN-US;mso-bidi-language:AR-SA">
</p>
</p>List Removal Instructions</span></b><span style="font-size:10.0pt;mso-bidi-font-size:12.0pt;font-family:Arial;mso-fareast-font-family:
&quot;Times New Roman&quot;;mso-ansi-language:EN-US;mso-fareast-language:EN-US;
mso-bidi-language:AR-SA">- Simply click here: <b><span style="color:blue"><a href="mailto:remobiotech@yahoo.com">remove</a></span></b>
to be instantly and permanently removed from our list. Send the blank email to
the address specified. Please do not try to reply to this message.</span>

</body>

</html>

From owner-linux-xfs@oss.sgi.com Tue Mar 13 20:11:03 2001
Received:  by oss.sgi.com id <S553678AbRCNEKx>;
	Tue, 13 Mar 2001 20:10:53 -0800
Received: from h00059aa0e40d.ne.mediaone.net ([66.31.89.164]:55283 "EHLO
        flowers.house.larsshack.org") by oss.sgi.com with ESMTP
	id <S553653AbRCNEK3>; Tue, 13 Mar 2001 20:10:29 -0800
Received: from localhost (lars@localhost)
	by flowers.house.larsshack.org (8.11.0/8.11.0) with ESMTP id f2E4A7w29002
	for <linux-xfs@oss.sgi.com>; Tue, 13 Mar 2001 23:10:07 -0500
X-Authentication-Warning: flowers.house.larsshack.org: lars owned process doing -bs
Date:   Tue, 13 Mar 2001 23:10:06 -0500 (EST)
From:   Lars Kellogg-Stedman <lars@larsshack.org>
X-Sender:  <lars@flowers.house.larsshack.org>
To:     <linux-xfs@oss.sgi.com>
Subject: xfsprogs package building problems
In-Reply-To: <200103131532.f2DFWfd06445@jen.americas.sgi.com>
Message-ID: <Pine.LNX.4.30.0103132302320.2905-100000@flowers.house.larsshack.org>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing

Howdy all,

Now that the md patches are in the tree I've rebuilt my kernel; I'd like
to generate a new xfsprogs rpm -- but the xfsprogs/Makepkgs command is no
longer generating a package.  Running 'make dist' in build/rpm yields
several errors along the lines of:

  Processing files: xfsprogs-1.1.5-0
  File listed twice: /usr/sbin/xfs_admin
  File listed twice: /usr/sbin/xfs_admin
  File listed twice: /usr/sbin/xfs_admin
  File listed twice: /usr/sbin/xfs_admin
  File listed twice: /usr/sbin/xfs_check

...and so forth.  Since the spec file appears to be generating the file
lists automatically, I'm not sure exactly what's up.  I'm going to spend a
little time this evening fiddling with things, unless someone out there
already has a quick solution...

-- Lars

-- 
Lars Kellogg-Stedman <lars@larsshack.org>


From owner-linux-xfs@oss.sgi.com Tue Mar 13 20:17:23 2001
Received:  by oss.sgi.com id <S553678AbRCNERN>;
	Tue, 13 Mar 2001 20:17:13 -0800
Received: from pneumatic-tube.sgi.com ([204.94.214.22]:61786 "EHLO
        pneumatic-tube.sgi.com") by oss.sgi.com with ESMTP
	id <S553653AbRCNERB>; Tue, 13 Mar 2001 20:17:01 -0800
Received: from larry.melbourne.sgi.com (larry.melbourne.sgi.com [134.14.52.130]) by pneumatic-tube.sgi.com (980327.SGI.8.8.8-aspam/980310.SGI-aspam) via SMTP id UAA08637
	for <linux-xfs@oss.sgi.com>; Tue, 13 Mar 2001 20:26:51 -0800 (PST)
	mail_from (nathans@wobbly.melbourne.sgi.com)
Received: from wobbly.melbourne.sgi.com (wobbly.melbourne.sgi.com [134.14.55.135]) by larry.melbourne.sgi.com (950413.SGI.8.6.12/950213.SGI.AUTOCF) via ESMTP id PAA16612; Wed, 14 Mar 2001 15:15:42 +1100
Received: (from nathans@localhost) by wobbly.melbourne.sgi.com (980427.SGI.8.8.8/980728.SGI.AUTOCF) id PAA19249; Wed, 14 Mar 2001 15:15:41 +1100 (EDT)
From:   "Nathan Scott" <nathans@wobbly.melbourne.sgi.com>
Message-Id: <10103141515.ZM216228@wobbly.melbourne.sgi.com>
Date:   Wed, 14 Mar 2001 15:15:39 -0400
In-Reply-To: Lars Kellogg-Stedman <lars@larsshack.org>
        "xfsprogs package building problems" (Mar 13, 11:10pm)
References: <Pine.LNX.4.30.0103132302320.2905-100000@flowers.house.larsshack.org>
X-Mailer: Z-Mail (3.2.3 08feb96 MediaMail)
To:     Lars Kellogg-Stedman <lars@larsshack.org>, <linux-xfs@oss.sgi.com>
Subject: Re: xfsprogs package building problems
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing

hi,

On Mar 13, 11:10pm, Lars Kellogg-Stedman wrote:
> Subject: xfsprogs package building problems
> Howdy all,
> 
> Now that the md patches are in the tree I've rebuilt my kernel; I'd like
> to generate a new xfsprogs rpm -- but the xfsprogs/Makepkgs command is no
> longer generating a package.  Running 'make dist' in build/rpm yields
> several errors along the lines of:
> 
>   Processing files: xfsprogs-1.1.5-0
>   File listed twice: /usr/sbin/xfs_admin
>   File listed twice: /usr/sbin/xfs_admin
>   File listed twice: /usr/sbin/xfs_admin
>   File listed twice: /usr/sbin/xfs_admin
>   File listed twice: /usr/sbin/xfs_check
> 
> ...and so forth.  Since the spec file appears to be generating the file
> lists automatically, I'm not sure exactly what's up.  I'm going to spend a
> little time this evening fiddling with things, unless someone out there
> already has a quick solution...
> 

Try a make clean/realclean before Makepkgs.  This is on my
list of things to tidy up, but its way down the bottom...

cheers.

-- 
Nathan

From owner-linux-xfs@oss.sgi.com Wed Mar 14 06:40:07 2001
Received:  by oss.sgi.com id <S553752AbRCNOj5>;
	Wed, 14 Mar 2001 06:39:57 -0800
Received: from deliverator.sgi.com ([204.94.214.10]:27760 "EHLO
        deliverator.sgi.com") by oss.sgi.com with ESMTP id <S553746AbRCNOjn>;
	Wed, 14 Mar 2001 06:39:43 -0800
Received: from zeus-fddi.americas.sgi.com (zeus-fddi.americas.sgi.com [128.162.8.103]) by deliverator.sgi.com (980309.SGI.8.8.8-aspam-6.2/980310.SGI-aspam) via ESMTP id GAA13625
	for <linux-xfs@oss.sgi.com>; Wed, 14 Mar 2001 06:38:32 -0800 (PST)
	mail_from (lord@sgi.com)
Received: from daisy-e185.americas.sgi.com (daisy-e194.americas.sgi.com [128.162.194.214]) by zeus-fddi.americas.sgi.com (8.9.3/americas-smart-nospam1.1) with ESMTP id IAA1134069 for <linux-xfs@oss.sgi.com>; Wed, 14 Mar 2001 08:38:27 -0600 (CST)
Received: from jen.americas.sgi.com (IDENT:root@jen.americas.sgi.com [128.162.184.86]) by daisy-e185.americas.sgi.com (SGI-8.9.3/SGI-server-1.7) with ESMTP id IAA75085 for <linux-xfs@oss.sgi.com>; Wed, 14 Mar 2001 08:38:26 -0600 (CST)
From:   Steve Lord <lord@sgi.com>
Received: by jen.americas.sgi.com (8.11.0/SGI-client-1.7) id f2EEdGE31552; Wed, 14 Mar 2001 08:39:16 -0600
Message-Id: <200103141439.f2EEdGE31552@jen.americas.sgi.com>
Date:   Wed, 14 Mar 2001 08:39:16 -0600
Subject: TAKE - fix a compiler warning
To:     unlisted-recipients:; (no To-header on input)
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing

Date:  Wed Mar 14 06:37:49 PST 2001
Workarea:  jen.americas.sgi.com:/src/lord/xfs-linux.2.4

The following file(s) were checked into:
  bonnie.engr.sgi.com:/isms/slinx/2.4.x-xfs


Modid:  2.4.x-xfs:slinx:89777a
linux/fs/xfs/linux/xfs_vnode.c - 1.49
	- Fix compiler warning.



From owner-linux-xfs@oss.sgi.com Wed Mar 14 07:11:27 2001
Received:  by oss.sgi.com id <S553752AbRCNPLR>;
	Wed, 14 Mar 2001 07:11:17 -0800
Received: from deliverator.sgi.com ([204.94.214.10]:14967 "EHLO
        deliverator.sgi.com") by oss.sgi.com with ESMTP id <S553748AbRCNPLM>;
	Wed, 14 Mar 2001 07:11:12 -0800
Received: from zeus-fddi.americas.sgi.com (zeus-fddi.americas.sgi.com [128.162.8.103]) by deliverator.sgi.com (980309.SGI.8.8.8-aspam-6.2/980310.SGI-aspam) via ESMTP id HAA17041
	for <linux-xfs@oss.sgi.com>; Wed, 14 Mar 2001 07:10:00 -0800 (PST)
	mail_from (lord@sgi.com)
Received: from daisy-e185.americas.sgi.com (daisy-e194.americas.sgi.com [128.162.194.214]) by zeus-fddi.americas.sgi.com (8.9.3/americas-smart-nospam1.1) with ESMTP id JAA1133753 for <linux-xfs@oss.sgi.com>; Wed, 14 Mar 2001 09:09:55 -0600 (CST)
Received: from jen.americas.sgi.com (IDENT:root@jen.americas.sgi.com [128.162.184.86]) by daisy-e185.americas.sgi.com (SGI-8.9.3/SGI-server-1.7) with ESMTP id JAA48321 for <linux-xfs@oss.sgi.com>; Wed, 14 Mar 2001 09:09:55 -0600 (CST)
From:   Steve Lord <lord@sgi.com>
Received: by jen.americas.sgi.com (8.11.0/SGI-client-1.7) id f2EFAiO01328; Wed, 14 Mar 2001 09:10:44 -0600
Message-Id: <200103141510.f2EFAiO01328@jen.americas.sgi.com>
Date:   Wed, 14 Mar 2001 09:10:44 -0600
Subject: TAKE - fix make dep on xfs directory
To:     unlisted-recipients:; (no To-header on input)
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing

The last change to the make files broke make dep for me, removing one
unused line fixed it, I am not sure if this is really the correct fix,
but it is the expedient one.

Date:  Wed Mar 14 07:08:54 PST 2001
Workarea:  jen.americas.sgi.com:/src/lord/linux-base

The following file(s) were checked into:
  bonnie.engr.sgi.com:/isms/slinx/2.4.x-xfs


Modid:  2.4.x-xfs:slinx:89784a
linux/fs/xfs/Makefile - 1.119
	- Comment out grio directory, make dep dies an ugly death otherwise



From owner-linux-xfs@oss.sgi.com Wed Mar 14 07:24:07 2001
Received:  by oss.sgi.com id <S553752AbRCNPX6>;
	Wed, 14 Mar 2001 07:23:58 -0800
Received: from pneumatic-tube.sgi.com ([204.94.214.22]:44152 "EHLO
        pneumatic-tube.sgi.com") by oss.sgi.com with ESMTP
	id <S553748AbRCNPXp>; Wed, 14 Mar 2001 07:23:45 -0800
Received: from zeus-fddi.americas.sgi.com (zeus-fddi.americas.sgi.com [128.162.8.103]) by pneumatic-tube.sgi.com (980327.SGI.8.8.8-aspam/980310.SGI-aspam) via ESMTP id HAA01612
	for <linux-xfs@oss.sgi.com>; Wed, 14 Mar 2001 07:33:36 -0800 (PST)
	mail_from (lord@sgi.com)
Received: from daisy-e185.americas.sgi.com (daisy-e194.americas.sgi.com [128.162.194.214]) by zeus-fddi.americas.sgi.com (8.9.3/americas-smart-nospam1.1) with ESMTP id JAA1134064 for <linux-xfs@oss.sgi.com>; Wed, 14 Mar 2001 09:22:28 -0600 (CST)
Received: from jen.americas.sgi.com (IDENT:root@jen.americas.sgi.com [128.162.184.86]) by daisy-e185.americas.sgi.com (SGI-8.9.3/SGI-server-1.7) with ESMTP id JAA12366 for <linux-xfs@oss.sgi.com>; Wed, 14 Mar 2001 09:22:27 -0600 (CST)
From:   Steve Lord <lord@sgi.com>
Received: by jen.americas.sgi.com (8.11.0/SGI-client-1.7) id f2EFNGh09944; Wed, 14 Mar 2001 09:23:16 -0600
Message-Id: <200103141523.f2EFNGh09944@jen.americas.sgi.com>
Date:   Wed, 14 Mar 2001 09:23:16 -0600
Subject: TAKE - small code size reduction
To:     unlisted-recipients:; (no To-header on input)
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing

Date:  Wed Mar 14 07:21:39 PST 2001
Workarea:  jen.americas.sgi.com:/src/lord/xfs-linux.2.4

The following file(s) were checked into:
  bonnie.engr.sgi.com:/isms/slinx/2.4.x-xfs


Modid:  2.4.x-xfs:slinx:89786a
linux/fs/xfs/xfs_itable.c - 1.96
	- Ifdef out a function we do not use yet.



From owner-linux-xfs@oss.sgi.com Wed Mar 14 07:28:17 2001
Received:  by oss.sgi.com id <S553756AbRCNP2H>;
	Wed, 14 Mar 2001 07:28:07 -0800
Received: from dollar.ecetra.com ([193.164.224.209]:25184 "EHLO msg.ecetra.com")
	by oss.sgi.com with ESMTP id <S553748AbRCNP16>;
	Wed, 14 Mar 2001 07:27:58 -0800
Received: from vie-ac.office.ecetra.com (vie-ac.office.ecetra.com [10.251.148.147] (may be forged))
	by msg.ecetra.com (8.9.3/8.9.3) with ESMTP id QAA10235
	for <linux-xfs@oss.sgi.com>; Wed, 14 Mar 2001 16:27:36 +0100
Received: from localhost (localhost [127.0.0.1])
	by vie-ac.office.ecetra.com (8.9.3/8.9.3) with ESMTP id QAA07459
	for <linux-xfs@oss.sgi.com>; Wed, 14 Mar 2001 16:27:36 +0100
Date:   Wed, 14 Mar 2001 16:27:36 +0100 (CET)
From:   CioccarelliA <alciocca@ecetra.com>
To:     <linux-xfs@oss.sgi.com>
Subject: gcc 2.95.2
Message-ID: <Pine.LNX.4.33.0103141621400.2349-100000@vie-ac.office.ecetra.com>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing

Dear All,

this question has probably been done to death but is gcc 2.95.2 a reliable
compiler for the current (CVS) xfs kernel? I am running the latest 14/03
CVS kernel compiled with 2.95.2 and haven't had any huge problems. My
window manager crashed a few times but recompiling it with gcc 2.95.2
fixed that, also my vmware win2000 died a horrible death with disk
corruption problems but I can't figure out how that could be related to
xfs (more likey just to a kernel compiled with 2.95.2)?

Regards,
Adam Cioccarelli



From owner-linux-xfs@oss.sgi.com Wed Mar 14 08:08:38 2001
Received:  by oss.sgi.com id <S553752AbRCNQI2>;
	Wed, 14 Mar 2001 08:08:28 -0800
Received: from esparrall.udg.es ([130.206.124.16]:35476 "EHLO esparrall.udg.es")
	by oss.sgi.com with ESMTP id <S553748AbRCNQII>;
	Wed, 14 Mar 2001 08:08:08 -0800
Received: from gcs by esparrall.udg.es with local (Exim 3.22 #1 (Debian))
	id 14dDnj-0008KX-00
	for <linux-xfs@oss.sgi.com>; Wed, 14 Mar 2001 17:07:03 +0100
Date:   Wed, 14 Mar 2001 17:07:03 +0100
From:   GCS <gcs@agentsinside.com>
To:     linux-xfs@oss.sgi.com
Subject: Re: gcc 2.95.2
Message-ID: <20010314170703.A31662@esparrall.udg.es>
References: <Pine.LNX.4.33.0103141621400.2349-100000@vie-ac.office.ecetra.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=iso-8859-1
Content-Disposition: inline
Content-Transfer-Encoding: 8bit
User-Agent: Mutt/1.3.15i
In-Reply-To: CioccarelliA wrote on =?iso-8859-1?Q?SZE?=
	=?iso-8859-1?B?LCBNwVI=?= 14, 2001 at 04:27:36 +0100:
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing



Hello,

On SZE, MÁR 14, 2001 at 04:27:36 +0100, CioccarelliA wrote:

> this question has probably been done to death but is gcc 2.95.2 a reliable
> compiler for the current (CVS) xfs kernel?
 Well, I have Debian 2.2rev2 with gcc 2.95.2-13, and about a week ago I can
not compile the kernel (from the CVS). Anyhow, I have an other box, also
Debian 2.2rev2, but I did some upgrades from testing, such as gcc. Now I use
2.95.3-5, and it was fine all the time. I had some misterious crashes in the
past (switching from XFree to a console left my screen totally black, and
the computer stopped responding; no alt+sysrq trick). But that time I used
VMWare also. Thus I think it is more related to VMWare.

> CVS kernel compiled with 2.95.2 and haven't had any huge problems.
 What distribution do you have?
Regards, GCS

From owner-linux-xfs@oss.sgi.com Wed Mar 14 08:16:48 2001
Received:  by oss.sgi.com id <S553752AbRCNQQi>;
	Wed, 14 Mar 2001 08:16:38 -0800
Received: from pneumatic-tube.sgi.com ([204.94.214.22]:44158 "EHLO
        pneumatic-tube.sgi.com") by oss.sgi.com with ESMTP
	id <S553748AbRCNQQ0>; Wed, 14 Mar 2001 08:16:26 -0800
Received: from zeus-fddi.americas.sgi.com (zeus-fddi.americas.sgi.com [128.162.8.103]) by pneumatic-tube.sgi.com (980327.SGI.8.8.8-aspam/980310.SGI-aspam) via ESMTP id IAA04457
	for <linux-xfs@oss.sgi.com>; Wed, 14 Mar 2001 08:26:17 -0800 (PST)
	mail_from (lord@sgi.com)
Received: from daisy-e185.americas.sgi.com (daisy-e194.americas.sgi.com [128.162.194.214]) by zeus-fddi.americas.sgi.com (8.9.3/americas-smart-nospam1.1) with ESMTP id KAA1093323; Wed, 14 Mar 2001 10:15:05 -0600 (CST)
Received: from jen.americas.sgi.com (IDENT:root@jen.americas.sgi.com [128.162.184.86]) by daisy-e185.americas.sgi.com (SGI-8.9.3/SGI-server-1.7) with ESMTP id KAA15267; Wed, 14 Mar 2001 10:15:05 -0600 (CST)
Received: from jen.americas.sgi.com by jen.americas.sgi.com (8.11.0/SGI-client-1.7) via ESMTP id f2EGFrZ13617; Wed, 14 Mar 2001 10:15:53 -0600
Message-Id: <200103141615.f2EGFrZ13617@jen.americas.sgi.com>
X-Mailer: exmh version 2.2 06/23/2000 with nmh-1.0.4
To:     GCS <gcs@agentsinside.com>
cc:     linux-xfs@oss.sgi.com
Subject: Re: gcc 2.95.2 
In-Reply-To: Message from GCS <gcs@agentsinside.com> 
   of "Wed, 14 Mar 2001 17:07:03 +0100." <20010314170703.A31662@esparrall.udg.es> 
Content-Transfer-Encoding: 8bit
Date:   Wed, 14 Mar 2001 10:15:53 -0600
From:   Steve Lord <lord@sgi.com>
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing


Hi,

I cannot speak for 2.95.2, but I can say that 2.96 (including the latest
redhat patched version) does not work at all. I can work past the compile
errors, but then I end up with failing mounts. Unfortunately at the moment
we do not have the bandwidth to be in the compiler debugging business.

Steve

> 
> 
> Hello,
> 
> On SZE, MAR 14, 2001 at 04:27:36 +0100, CioccarelliA wrote:
> 
> > this question has probably been done to death but is gcc 2.95.2 a reliable
> > compiler for the current (CVS) xfs kernel?
>  Well, I have Debian 2.2rev2 with gcc 2.95.2-13, and about a week ago I can
> not compile the kernel (from the CVS). Anyhow, I have an other box, also
> Debian 2.2rev2, but I did some upgrades from testing, such as gcc. Now I use
> 2.95.3-5, and it was fine all the time. I had some misterious crashes in the
> past (switching from XFree to a console left my screen totally black, and
> the computer stopped responding; no alt+sysrq trick). But that time I used
> VMWare also. Thus I think it is more related to VMWare.
> 
> > CVS kernel compiled with 2.95.2 and haven't had any huge problems.
>  What distribution do you have?
> Regards, GCS




From owner-linux-xfs@oss.sgi.com Wed Mar 14 08:20:37 2001
Received:  by oss.sgi.com id <S553752AbRCNQU1>;
	Wed, 14 Mar 2001 08:20:27 -0800
Received: from [209.101.91.34] ([209.101.91.34]:38921 "EHLO mail.compro.net")
	by oss.sgi.com with ESMTP id <S553748AbRCNQUU>;
	Wed, 14 Mar 2001 08:20:20 -0800
Received: from compro.net (PC120.COMPRO.NET [10.10.10.120])
	by mail.compro.net (8.9.3/8.9.3) with ESMTP id MAA15360;
	Wed, 14 Mar 2001 12:35:00 -0500
Message-ID: <3AAF9C95.C0D37650@compro.net>
Date:   Wed, 14 Mar 2001 11:30:13 -0500
From:   Mark Hounschell <markh@compro.net>
Reply-To: markh@compro.net
Organization: Compro Computer Svcs.
X-Mailer: Mozilla 4.76 [en] (X11; U; Linux 2.4.2-4GB i686)
X-Accept-Language: en
MIME-Version: 1.0
To:     CioccarelliA <alciocca@ecetra.com>
CC:     linux-xfs@oss.sgi.com
Subject: Re: gcc 2.95.2
References: <Pine.LNX.4.33.0103141621400.2349-100000@vie-ac.office.ecetra.com>
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing

CioccarelliA wrote:
> 
> Dear All,
> 
> this question has probably been done to death but is gcc 2.95.2 a reliable
> compiler for the current (CVS) xfs kernel? I am running the latest 14/03
> CVS kernel compiled with 2.95.2 and haven't had any huge problems. My
> window manager crashed a few times but recompiling it with gcc 2.95.2
> fixed that, also my vmware win2000 died a horrible death with disk
> corruption problems but I can't figure out how that could be related to
> xfs (more likey just to a kernel compiled with 2.95.2)?
> 
> Regards,
> Adam Cioccarelli

2.95.2 is the official latest stable compiler. Unfortunatly I beleive
there
are still problems compiling xfs stuff with it. Thats what's holding me
back
from using it. Until these issues are resolved this will continue to be
as you
say "Done to death". I had the same problems as you when I used the "XFS
Recomended"
complier with colors in KDE2. Small price for a usable xfs fs but who
knows
what else it could cause.


regards

From owner-linux-xfs@oss.sgi.com Wed Mar 14 08:27:47 2001
Received:  by oss.sgi.com id <S553752AbRCNQ1i>;
	Wed, 14 Mar 2001 08:27:38 -0800
Received: from esparrall.udg.es ([130.206.124.16]:48788 "EHLO esparrall.udg.es")
	by oss.sgi.com with ESMTP id <S553748AbRCNQ11>;
	Wed, 14 Mar 2001 08:27:27 -0800
Received: from gcs by esparrall.udg.es with local (Exim 3.22 #1 (Debian))
	id 14dE6T-0008Nc-00
	for <linux-xfs@oss.sgi.com>; Wed, 14 Mar 2001 17:26:25 +0100
Date:   Wed, 14 Mar 2001 17:26:24 +0100
From:   GCS <gcs@agentsinside.com>
To:     linux-xfs@oss.sgi.com
Subject: Re: gcc 2.95.2
Message-ID: <20010314172624.A32147@esparrall.udg.es>
References: <gcs@agentsinside.com> <200103141615.f2EGFrZ13617@jen.americas.sgi.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=iso-8859-1
Content-Disposition: inline
Content-Transfer-Encoding: 8bit
User-Agent: Mutt/1.3.15i
In-Reply-To: Steve Lord wrote on SZE, =?iso-8859-1?Q?M?=
	=?iso-8859-1?Q?=C1R?= 14, 2001 at 10:15:53 -0600:
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing



Hi,

On SZE, MÁR 14, 2001 at 10:15:53 -0600, Steve Lord wrote:

> I can work past the compile errors, but then I end up with failing mounts.
 What kind of errors do you get?
 I get internal error as:
make[3]: Cambiando a directorio `/usr/src/linux/fs/xfs'
gcc -D__KERNEL__ -I/usr/src/linux/include  -Wall -Wstrict-prototypes -O2 -fno-strict-aliasing -pipe -mpreferred-stack-boundary=2 -march=i686  -Wno-unused  -Wno-parentheses -Wno-uninitialized -I. -funsigned-char  -Wno-unknown-pragmas -DDEBUG -DXFSDEBUG  -c -o xfs_bmap.o xfs_bmap.c
xfs_bmap.c: In function `xfs_bmap_del_extent':
xfs_bmap.c:3130: internal error--unrecognizable insn:
(insn/i 468 467 2149 (parallel[ 
            (set (reg:SI 0 %eax)
                (asm_operands ("") ("=a") 0[ 
                        (reg:DI 1 %edx)
                    ] 
                    [ 
                        (asm_input:DI ("A"))
                    ]  ("linux/xfs_linux.h") 267))
            (set (reg:SI 1 %edx)
                (asm_operands ("") ("=d") 1[ 
                        (reg:DI 1 %edx)
                    ] 
                    [ 
                        (asm_input:DI ("A"))
                    ]  ("linux/xfs_linux.h") 267))
        ] ) -1 (insn_list 435 (nil))
    (nil))
cpp: output pipe has been closed
make[3]: *** [xfs_bmap.o] Error 1

Thanks, Laszlo

From owner-linux-xfs@oss.sgi.com Wed Mar 14 09:12:48 2001
Received:  by oss.sgi.com id <S553752AbRCNRMi>;
	Wed, 14 Mar 2001 09:12:38 -0800
Received: from deliverator.sgi.com ([204.94.214.10]:8988 "EHLO
        deliverator.sgi.com") by oss.sgi.com with ESMTP id <S553748AbRCNRM1>;
	Wed, 14 Mar 2001 09:12:27 -0800
Received: from zeus-fddi.americas.sgi.com (zeus-fddi.americas.sgi.com [128.162.8.103]) by deliverator.sgi.com (980309.SGI.8.8.8-aspam-6.2/980310.SGI-aspam) via ESMTP id JAA03283
	for <linux-xfs@oss.sgi.com>; Wed, 14 Mar 2001 09:11:16 -0800 (PST)
	mail_from (lord@sgi.com)
Received: from daisy-e185.americas.sgi.com (daisy-e194.americas.sgi.com [128.162.194.214]) by zeus-fddi.americas.sgi.com (8.9.3/americas-smart-nospam1.1) with ESMTP id LAA1134678; Wed, 14 Mar 2001 11:11:08 -0600 (CST)
Received: from jen.americas.sgi.com (IDENT:root@jen.americas.sgi.com [128.162.184.86]) by daisy-e185.americas.sgi.com (SGI-8.9.3/SGI-server-1.7) with ESMTP id LAA89077; Wed, 14 Mar 2001 11:11:08 -0600 (CST)
Received: from jen.americas.sgi.com by jen.americas.sgi.com (8.11.0/SGI-client-1.7) via ESMTP id f2EHBu613775; Wed, 14 Mar 2001 11:11:56 -0600
Message-Id: <200103141711.f2EHBu613775@jen.americas.sgi.com>
X-Mailer: exmh version 2.2 06/23/2000 with nmh-1.0.4
To:     GCS <gcs@agentsinside.com>
cc:     linux-xfs@oss.sgi.com
Subject: Re: gcc 2.95.2 
In-Reply-To: Message from GCS <gcs@agentsinside.com> 
   of "Wed, 14 Mar 2001 17:26:24 +0100." <20010314172624.A32147@esparrall.udg.es> 
Content-Transfer-Encoding: 8bit
Date:   Wed, 14 Mar 2001 11:11:56 -0600
From:   Steve Lord <lord@sgi.com>
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing

> 
> 
> Hi,
> 
> On SZE, MAR 14, 2001 at 10:15:53 -0600, Steve Lord wrote:
> 
> > I can work past the compile errors, but then I end up with failing mounts.
>  What kind of errors do you get?
>  I get internal error as:
> make[3]: Cambiando a directorio `/usr/src/linux/fs/xfs'
> gcc -D__KERNEL__ -I/usr/src/linux/include  -Wall -Wstrict-prototypes -O2 -fno
> -strict-aliasing -pipe -mpreferred-stack-boundary=2 -march=i686  -Wno-unused 
>  -Wno-parentheses -Wno-uninitialized -I. -funsigned-char  -Wno-unknown-pragma
> s -DDEBUG -DXFSDEBUG  -c -o xfs_bmap.o xfs_bmap.c
> xfs_bmap.c: In function `xfs_bmap_del_extent':
> xfs_bmap.c:3130: internal error--unrecognizable insn:
> (insn/i 468 467 2149 (parallel[ 
>             (set (reg:SI 0 %eax)
>                 (asm_operands ("") ("=a") 0[ 
>                         (reg:DI 1 %edx)
>                     ] 
>                     [ 
>                         (asm_input:DI ("A"))
>                     ]  ("linux/xfs_linux.h") 267))
>             (set (reg:SI 1 %edx)
>                 (asm_operands ("") ("=d") 1[ 
>                         (reg:DI 1 %edx)
>                     ] 
>                     [ 
>                         (asm_input:DI ("A"))
>                     ]  ("linux/xfs_linux.h") 267))
>         ] ) -1 (insn_list 435 (nil))
>     (nil))
> cpp: output pipe has been closed
> make[3]: *** [xfs_bmap.o] Error 1
> 
> Thanks, Laszlo


This error is easy to get rid of, the xfs_do_div, xfs_do_mod  and roundup_64
inline functions in fs/xfs/linux/xfs_linux.h need to be moved to a .c file and
made into regular functions. This moves the problem to being a runtime one,
if you build a kernel like this, do not do it on a system where parts of the
OS depend on xfs working.

Steve




From owner-linux-xfs@oss.sgi.com Wed Mar 14 09:43:58 2001
Received:  by oss.sgi.com id <S553762AbRCNRnt>;
	Wed, 14 Mar 2001 09:43:49 -0800
Received: from sgi.SGI.COM ([192.48.153.1]:64541 "EHLO sgi.com")
	by oss.sgi.com with ESMTP id <S553753AbRCNRng>;
	Wed, 14 Mar 2001 09:43:36 -0800
Received: from ledzep.americas.sgi.com (relay.sgi.com [137.38.226.97] (may be forged)) 
	by sgi.com (980327.SGI.8.8.8-aspam/980304.SGI-aspam:
       SGI does not authorize the use of its proprietary
       systems or networks for unsolicited or bulk email
       from the Internet.) 
	via ESMTP id JAA05997
	for <linux-xfs@oss.sgi.com>; Wed, 14 Mar 2001 09:43:34 -0800 (PST)
	mail_from (cattelan@thebarn.com)
Received: from gibble.americas.sgi.com (gibble.americas.sgi.com [128.162.195.80]) by ledzep.americas.sgi.com (SGI-SGI-8.9.3/americas-smart-nospam1.1) with ESMTP id LAA23600; Wed, 14 Mar 2001 11:42:18 -0600 (CST)
Received: from thebarn.com (localhost [127.0.0.1])
	by gibble.americas.sgi.com (8.11.2/8.11.2) with ESMTP id f2EHdpn06542;
	Wed, 14 Mar 2001 12:39:51 -0500
Message-ID: <3AAFACE5.87DC16B6@thebarn.com>
Date:   Wed, 14 Mar 2001 12:39:50 -0500
From:   Russell Cattelan <cattelan@thebarn.com>
X-Mailer: Mozilla 4.76 [en] (X11; U; Linux 2.4.2-XFS i686)
X-Accept-Language: en
MIME-Version: 1.0
To:     markh@compro.net
CC:     CioccarelliA <alciocca@ecetra.com>, linux-xfs@oss.sgi.com
Subject: Re: gcc 2.95.2
References: <Pine.LNX.4.33.0103141621400.2349-100000@vie-ac.office.ecetra.com> <3AAF9C95.C0D37650@compro.net>
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing

Mark Hounschell wrote:

> CioccarelliA wrote:
> >
> > Dear All,
> >
> > this question has probably been done to death but is gcc 2.95.2 a reliable
> > compiler for the current (CVS) xfs kernel? I am running the latest 14/03
> > CVS kernel compiled with 2.95.2 and haven't had any huge problems. My
> > window manager crashed a few times but recompiling it with gcc 2.95.2
> > fixed that, also my vmware win2000 died a horrible death with disk
> > corruption problems but I can't figure out how that could be related to
> > xfs (more likey just to a kernel compiled with 2.95.2)?
> >
> > Regards,
> > Adam Cioccarelli
>
> 2.95.2 is the official latest stable compiler. Unfortunatly I beleive
> there
> are still problems compiling xfs stuff with it. Thats what's holding me
> back
> from using it. Until these issues are resolved this will continue to be
> as you
> say "Done to death". I had the same problems as you when I used the "XFS
> Recomended"
> complier with colors in KDE2. Small price for a usable xfs fs but who
> knows
> what else it could cause.

The last time I tried 2.95.2 it appeared to working.
I haven't run extensive tests so I can't say for sure not problems exist.
If somebody does have a specific example of XFS failing send it our way and
we'll look into fixing it.

As far as 2.96 goes ... I don't think anybody really recommends using that
version
for the kernel or for anything else important.

http://gcc.gnu.org/gcc-2.96.html

>
>
> regards

--
Russell Cattelan
--
Digital Elves inc. -- Currently on loan to SGI
Linux XFS core developer.




From owner-linux-xfs@oss.sgi.com Wed Mar 14 10:06:08 2001
Received:  by oss.sgi.com id <S553764AbRCNSF6>;
	Wed, 14 Mar 2001 10:05:58 -0800
Received: from mail.linpro.no ([213.203.57.2]:36625 "HELO linpro.no")
	by oss.sgi.com with SMTP id <S553762AbRCNSFp>;
	Wed, 14 Mar 2001 10:05:45 -0800
Received: (qmail 14839 invoked from network); 14 Mar 2001 18:05:39 -0000
Received: from false.linpro.no (213.203.57.201)
  by mail.linpro.no with SMTP; 14 Mar 2001 18:05:39 -0000
Received: from perbu by false.linpro.no with local (Exim 3.12 #1 (Debian))
	id 14dFeU-0006Rc-00
	for <linux-xfs@oss.sgi.com>; Wed, 14 Mar 2001 19:05:38 +0100
To:     linux-xfs@oss.sgi.com
Subject: Re: gcc 2.95.2
References: <gcs@agentsinside.com>
	<200103141615.f2EGFrZ13617@jen.americas.sgi.com>
From:   Per Andreas Buer <per@linpro.no>
Organization: LinPro AS
Date:   14 Mar 2001 19:05:38 +0100
In-Reply-To: Steve Lord's message of "Wed, 14 Mar 2001 10:15:53 -0600"
Message-ID: <ul6d7bkusy5.fsf@false.linpro.no>
Lines:  61
User-Agent: Gnus/5.0807 (Gnus v5.8.7) Emacs/20.7
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing

Steve Lord <lord@sgi.com> writes:

> I cannot speak for 2.95.2, 

I can. 

The kernel, compiled from CVS 2001-03-04, compiled nicely. Basic testing
( 42 processes doing "cp /tmp/1 /tmp/2" and "sync") for just a few
seconds resulted in the attached crash. With egcs-2.91.66 this kernel
worked flawlessly.


kernel BUG at buffer.c:188!
invalid operand: 0000
CPU:    0
EIP:    0010:[<c0134d53>]
EFLAGS: 00010282
eax: 0000001c   ebx: c2088790   ecx: 00000001   edx: 00000001
esi: 00000000   edi: 00000000   ebp: 00000000   esp: f74ebf6c
ds: 0018   es: 0018   ss: 0018
Process sync (pid: 7963, stackpage=f74eb000)
Stack: c0242405 c0242669 000000bc f7549e40 f7549e40 c0134e58 f7549e40 f74ea000 
       00000000 00000001 bffffb44 00000000 00000000 f7549e40 c0134f7f 00000000 
       00000000 f74ea000 00000001 c0134ffb 00000000 c0108f4b bffffbac 00000002 
Call Trace: [<c0134e58>] [<c0134f7f>] [<c0134ffb>] [<c0108f4b>] 
Code: 0f 0b 83 c4 0c 5b 5e c3 90 83 ec 0c 55 57 56 53 8b 7c 24 20 

>>EIP; c0134d53 <_write_buffer+df/e8>   <=====
Trace; c0134e58 <sync_buffers+fc/1f0>
Trace; c0134f7f <fsync_dev+f/84>
Trace; c0134ffb <sys_sync+7/10>
Trace; c0108f4b <system_call+33/38>
Code;  c0134d53 <_write_buffer+df/e8>
00000000 <_EIP>:
Code;  c0134d53 <_write_buffer+df/e8>   <=====
   0:   0f 0b                     ud2a      <=====
Code;  c0134d55 <_write_buffer+e1/e8>
   2:   83 c4 0c                  add    $0xc,%esp
Code;  c0134d58 <_write_buffer+e4/e8>
   5:   5b                        pop    %ebx
Code;  c0134d59 <_write_buffer+e5/e8>
   6:   5e                        pop    %esi
Code;  c0134d5a <_write_buffer+e6/e8>
   7:   c3                        ret    
Code;  c0134d5b <_write_buffer+e7/e8>
   8:   90                        nop    
Code;  c0134d5c <sync_buffers+0/1f0>
   9:   83 ec 0c                  sub    $0xc,%esp
Code;  c0134d5f <sync_buffers+3/1f0>
   c:   55                        push   %ebp
Code;  c0134d60 <sync_buffers+4/1f0>
   d:   57                        push   %edi
Code;  c0134d61 <sync_buffers+5/1f0>
   e:   56                        push   %esi
Code;  c0134d62 <sync_buffers+6/1f0>
   f:   53                        push   %ebx
Code;  c0134d63 <sync_buffers+7/1f0>
  10:   8b 7c 24 20               mov    0x20(%esp,1),%edi

-- 
Per Andreas Buer

From owner-linux-xfs@oss.sgi.com Wed Mar 14 10:32:58 2001
Received:  by oss.sgi.com id <S553772AbRCNScs>;
	Wed, 14 Mar 2001 10:32:48 -0800
Received: from mpdr0.milwaukee.wi.ameritech.net ([206.141.239.126]:54268 "EHLO
        mailhost.mil.ameritech.net") by oss.sgi.com with ESMTP
	id <S553762AbRCNScm>; Wed, 14 Mar 2001 10:32:42 -0800
Received: from friz.themagicbus.com ([64.108.133.80])
          by mailhost.mil.ameritech.net
          (InterMail v4.01.01.07 201-229-111-110) with SMTP
          id <20010314181935.RBJ26289.mailhost.mil.ameritech.net@friz.themagicbus.com>
          for <linux-xfs@oss.sgi.com>; Wed, 14 Mar 2001 12:19:35 -0600
Content-Type: text/plain;
  charset="iso-8859-1"
From:   james <jdickens@ameritech.net>
To:     linux-xfs@oss.sgi.com
Date:   Wed, 14 Mar 2001 12:44:36 -0600
X-Mailer: KMail [version 1.2]
MIME-Version: 1.0
Message-Id: <01031412430900.13860@friz.themagicbus.com>
Content-Transfer-Encoding: 8bit
Subject: is this a bug or a feature?
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing

Hi 

I was testing xfs out, so i wrote a script that created 2.3 million files in 
a directory, they were zero lenth files, them i deleted them, but now the 
file system is using 500megs, i figure this is directory information that is 
being held for future use. But since i will not be using 2 million files on 
this file system, what is the best way to get this space back.   


here is some relavent information about my system 


[jamesd@magic jamesd]$ uname -a
Linux magic 2.4.2-XFS #7 Mon Mar 12 23:26:54 CST 2001 i686 unknown


Filesystem            Size  Used Avail Use% Mounted on
/dev/hda9             3.9G  580M  3.4G  14% /home4


[jamesd@magic jamesd]$ df -i /home4
Filesystem            Inodes   IUsed   IFree IUse% Mounted on
/dev/hda9               3.9M       3    3.9M    0% /home4


please email if you need any further information 



Thank you 

James Dickens




From owner-linux-xfs@oss.sgi.com Wed Mar 14 10:52:38 2001
Received:  by oss.sgi.com id <S553779AbRCNSw2>;
	Wed, 14 Mar 2001 10:52:28 -0800
Received: from deliverator.sgi.com ([204.94.214.10]:42046 "EHLO
        deliverator.sgi.com") by oss.sgi.com with ESMTP id <S553774AbRCNSwT>;
	Wed, 14 Mar 2001 10:52:19 -0800
Received: from zeus-fddi.americas.sgi.com (zeus-fddi.americas.sgi.com [128.162.8.103]) by deliverator.sgi.com (980309.SGI.8.8.8-aspam-6.2/980310.SGI-aspam) via ESMTP id KAA19859
	for <linux-xfs@oss.sgi.com>; Wed, 14 Mar 2001 10:51:08 -0800 (PST)
	mail_from (lord@sgi.com)
Received: from daisy-e185.americas.sgi.com (daisy-e194.americas.sgi.com [128.162.194.214]) by zeus-fddi.americas.sgi.com (8.9.3/americas-smart-nospam1.1) with ESMTP id MAA1137796; Wed, 14 Mar 2001 12:51:01 -0600 (CST)
Received: from jen.americas.sgi.com (IDENT:root@jen.americas.sgi.com [128.162.184.86]) by daisy-e185.americas.sgi.com (SGI-8.9.3/SGI-server-1.7) with ESMTP id MAA78084; Wed, 14 Mar 2001 12:51:01 -0600 (CST)
Received: from jen.americas.sgi.com by jen.americas.sgi.com (8.11.0/SGI-client-1.7) via ESMTP id f2EIpmK18568; Wed, 14 Mar 2001 12:51:48 -0600
Message-Id: <200103141851.f2EIpmK18568@jen.americas.sgi.com>
X-Mailer: exmh version 2.2 06/23/2000 with nmh-1.0.4
To:     james <jdickens@ameritech.net>
cc:     linux-xfs@oss.sgi.com
Subject: Re: is this a bug or a feature? 
In-Reply-To: Message from james <jdickens@ameritech.net> 
   of "Wed, 14 Mar 2001 12:44:36 CST." <01031412430900.13860@friz.themagicbus.com> 
Date:   Wed, 14 Mar 2001 12:51:48 -0600
From:   Steve Lord <lord@sgi.com>
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing


It is a 'feature'. XFS does not preallocate space for inodes at mkfs time
(ext2 for example does). Instead they are allocated on demand, but they are
never given back. So once you hit a high watermark with inode allocations,
you stay there and show up in the space used. In a filesystem which allocated
inodes at mkfs time I suspect the inode space is never included in the free
space reports.

Steve

> Hi 
> 
> I was testing xfs out, so i wrote a script that created 2.3 million files in 
> a directory, they were zero lenth files, them i deleted them, but now the 
> file system is using 500megs, i figure this is directory information that is 
> being held for future use. But since i will not be using 2 million files on 
> this file system, what is the best way to get this space back.   
> 
> 
> here is some relavent information about my system 
> 
> 
> [jamesd@magic jamesd]$ uname -a
> Linux magic 2.4.2-XFS #7 Mon Mar 12 23:26:54 CST 2001 i686 unknown
> 
> 
> Filesystem            Size  Used Avail Use% Mounted on
> /dev/hda9             3.9G  580M  3.4G  14% /home4
> 
> 
> [jamesd@magic jamesd]$ df -i /home4
> Filesystem            Inodes   IUsed   IFree IUse% Mounted on
> /dev/hda9               3.9M       3    3.9M    0% /home4
> 
> 
> please email if you need any further information 
> 
> 
> 
> Thank you 
> 
> James Dickens
> 
> 



From owner-linux-xfs@oss.sgi.com Wed Mar 14 11:21:31 2001
Received:  by oss.sgi.com id <S553779AbRCNTVX>;
	Wed, 14 Mar 2001 11:21:23 -0800
Received: from ns1.algx.net ([216.99.225.30]:15261 "EHLO dfwns01.algx.net")
	by oss.sgi.com with ESMTP id <S553772AbRCNTVE>;
	Wed, 14 Mar 2001 11:21:04 -0800
Received: from jtsdell (66-2-81-26.customer.algx.net [66.2.81.26])
	by dfwns01.algx.net (8.9.1/8.9.1) with ESMTP id NAA19144
	for <linux-xfs@oss.sgi.com>; Wed, 14 Mar 2001 13:20:44 -0600 (CST)
Message-ID: <XFMail.20010314141910.jtrostel@connex.com>
X-Mailer: XFMail 1.4.8 on Linux
X-Priority: 3 (Normal)
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 8bit
MIME-Version: 1.0
Date:   Wed, 14 Mar 2001 14:19:10 -0500 (EST)
Reply-To: jtrostel@connex.com
Organization: Connex
From:   John Trostel <jtrostel@connex.com>
To:     linux-xfs@oss.sgi.com
Subject: Module compile error
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing

When trying to compile the kernel portion of the latest CVS tree, I found I
needed to insert the following into the Makefile in the ../linux/fs/xfs
directory.  I get 

ld -m elf_i386  -r -o xfs.o xfsrtstubs.o dmapi/dmapi.o xfs_dmapi.o xfs_dquot.o
xfs_trans_dquot.o xfs_dquot_item.o xfs_qm_syscalls.o xfs_qm.o xfs_alloc.o
xfs_alloc_btree.o xfs_attr.o xfs_attr_fetch.o xfs_attr_leaf.o xfs_bit.o
xfs_bmap.o xfs_bmap_btree.o xfs_btree.o xfs_buf_item.o xfs_da_btree.o xfs_dir.o
xfs_dir2.o xfs_dir2_block.o xfs_dir2_data.o xfs_dir2_leaf.o xfs_dir2_node.o
xfs_dir2_sf.o xfs_dir2_trace.o xfs_dir_leaf.o xfs_error.o xfs_extfree_item.o
xfs_fsops.o xfs_ialloc.o xfs_ialloc_btree.o xfs_iget.o xfs_inode.o
xfs_inode_item.o xfs_iocore.o xfs_itable.o xfs_dfrag.o xfs_log.o
xfs_log_recover.o xfs_macros.o xfs_mount.o xfs_rename.o xfs_trans.o
xfs_trans_ail.o xfs_trans_buf.o xfs_trans_extfree.o xfs_trans_inode.o
xfs_trans_item.o xfs_utils.o xfs_vfsops.o xfs_vnodeops.o xfs_rw.o macstubs.o
linux/linux_xfs.o
ld: cannot open dmapi/dmapi.o: No such file or directory
make[2]: *** [xfs.o] Error 1
make[2]: Leaving directory `/home/jt/Coding/linux-2.4-xfs/linux/fs/xfs'
make[1]: *** [_modsubdir_xfs] Error 2
make[1]: Leaving directory `/home/jt/Coding/linux-2.4-xfs/linux/fs'
make: *** [_mod_fs] Error 2

Sure enough the /dmapi subdirectory was there, but there was no dmapi.o inside
it.

I cut out a portion from an older ../linux/fs/xfs/Makefile and slapped it into
the new ../linux/fs/xfs/Makefile and now the modules compile.

subdir-$(CONFIG_XFS_FS)         += linux
#subdir-$(CONFIG_XFS_GRIO)      += grio
subdir-$(CONFIG_XFS_DMAPI)      += dmapi

# added to make dmapi.o get compiled
ifneq ($(CONFIG_XFS_DMAPI),)
  subdir-$(CONFIG_XFS_FS)       += dmapi
endif

Thinking this is vaguely reminiscent of Steve Lord's quick fix for the grio
problem mentioned earlier today (see the #subdir-$(CONFIG_XFS_GRIO) += grio
line above, I thought I'd see if this is related, a stupid mistake on my part,
or whatever.

Your mileage may vary...

-- 
John M. Trostel
Linux OS Engineer
Connex
jtrostel@connex.com

From owner-linux-xfs@oss.sgi.com Wed Mar 14 11:48:31 2001
Received:  by oss.sgi.com id <S553800AbRCNTsV>;
	Wed, 14 Mar 2001 11:48:21 -0800
Received: from deliverator.sgi.com ([204.94.214.10]:47442 "EHLO
        deliverator.sgi.com") by oss.sgi.com with ESMTP id <S553779AbRCNTsL>;
	Wed, 14 Mar 2001 11:48:11 -0800
Received: from zeus-fddi.americas.sgi.com (zeus-fddi.americas.sgi.com [128.162.8.103]) by deliverator.sgi.com (980309.SGI.8.8.8-aspam-6.2/980310.SGI-aspam) via ESMTP id LAA29681
	for <linux-xfs@oss.sgi.com>; Wed, 14 Mar 2001 11:46:59 -0800 (PST)
	mail_from (lord@sgi.com)
Received: from daisy-e185.americas.sgi.com (daisy-e194.americas.sgi.com [128.162.194.214]) by zeus-fddi.americas.sgi.com (8.9.3/americas-smart-nospam1.1) with ESMTP id NAA1137252 for <linux-xfs@oss.sgi.com>; Wed, 14 Mar 2001 13:46:54 -0600 (CST)
Received: from jen.americas.sgi.com (IDENT:root@jen.americas.sgi.com [128.162.184.86]) by daisy-e185.americas.sgi.com (SGI-8.9.3/SGI-server-1.7) with ESMTP id NAA45537 for <linux-xfs@oss.sgi.com>; Wed, 14 Mar 2001 13:46:54 -0600 (CST)
From:   Steve Lord <lord@sgi.com>
Received: by jen.americas.sgi.com (8.11.0/SGI-client-1.7) id f2EJlfx19882; Wed, 14 Mar 2001 13:47:41 -0600
Message-Id: <200103141947.f2EJlfx19882@jen.americas.sgi.com>
Date:   Wed, 14 Mar 2001 13:47:41 -0600
Subject: TAKE - fix up vmalloc_area_pages arguments
To:     unlisted-recipients:; (no To-header on input)
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing

we still have the page remapping code, it just looks a little less
intrusive now.

Date:  Wed Mar 14 11:46:03 PST 2001
Workarea:  jen.americas.sgi.com:/src/lord/xfs-linux.2.4

The following file(s) were checked into:
  bonnie.engr.sgi.com:/isms/slinx/2.4.x-xfs


Modid:  2.4.x-xfs:slinx:89832a
linux/mm/vmalloc.c - 1.28
	- play some some tricks to avoid changing vmalloc_area_pages prototype.

linux/include/linux/vmalloc.h - 1.13
	- fall back to old prototype for vmalloc_area_pages

linux/arch/sparc64/mm/modutil.c - 1.7
	- we no longer need differences in vmalloc_area_pages arguments



From owner-linux-xfs@oss.sgi.com Wed Mar 14 12:40:41 2001
Received:  by oss.sgi.com id <S553814AbRCNUkb>;
	Wed, 14 Mar 2001 12:40:31 -0800
Received: from sgi.SGI.COM ([192.48.153.1]:51291 "EHLO sgi.com")
	by oss.sgi.com with ESMTP id <S553812AbRCNUkH>;
	Wed, 14 Mar 2001 12:40:07 -0800
Received: from info.engr.sgi.com ([192.26.80.216]) 
	by sgi.com (980327.SGI.8.8.8-aspam/980304.SGI-aspam:
       SGI does not authorize the use of its proprietary
       systems or networks for unsolicited or bulk email
       from the Internet.) 
	via ESMTP id MAA03868; Wed, 14 Mar 2001 12:38:09 -0800 (PST)
	mail_from (nobody@info.engr.sgi.com)
Received: (from nobody@localhost)
	by info.engr.sgi.com (SGI-8.9.3/8.9.3) id MAA65020;
	Wed, 14 Mar 2001 12:38:09 -0800 (PST)
Date:   Wed, 14 Mar 2001 12:38:09 -0800 (PST)
Message-Id: <200103142038.MAA65020@info.engr.sgi.com>
X-Pv-Incident: 802340 
webPV:  oldchicago.americas.sgi.com 
webExec: webpvupdate,pvincident 
Reply-To: sgi.bugs.xfs@fido.engr.sgi.com
From:   pv@odin.corp.sgi.com (nb@sgi.com)
Subject: REASSIGN 802340 - kernel panic
To:     raa@sgi.com
Cc:     linux-xfs@oss.sgi.com
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing

View Incident: http://co-op.engr.sgi.com/BugWorks/code/bwxquery.cgi?search=Search&wlong=1&view_type=Bug&wi=802340

 Status : open                         Priority : 1                         
*Assigned Engineer : raa               Submitter : coreym                   
 Project : xfs-linux                  *Assigned Group : xfs-linux           
 Opened Date : 09/20/00               *Modified User : nb                   
*Modified User Domain : sgi.com       *Description :
I was running rwtest (doio and iogen) to stress xfs on tarpon 
and tarpon paniced. I have no way of knowing if it was related 
to my test or not.

The panic string that I received was:
Message from syslogd@tarpon at Fri Sep 15 11:49:03 2000 ...
tarpon kernel: Kernel panic: kmem_alloc: NULL memory on KM_SLEEP request!

==========================
ADDITIONAL INFORMATION (ADD)

.....


==========================
ADDITIONAL INFORMATION (REASSIGN)
From: nb@sgi.com (BugWorks)
Date: Mar 14 2001 12:38:08PM
==========================

From owner-linux-xfs@oss.sgi.com Wed Mar 14 12:57:41 2001
Received:  by oss.sgi.com id <S553824AbRCNU5V>;
	Wed, 14 Mar 2001 12:57:21 -0800
Received: from deliverator.sgi.com ([204.94.214.10]:17003 "EHLO
        deliverator.sgi.com") by oss.sgi.com with ESMTP id <S553822AbRCNU5N>;
	Wed, 14 Mar 2001 12:57:13 -0800
Received: from nodin.corp.sgi.com (nodin.corp.sgi.com [192.26.51.193]) by deliverator.sgi.com (980309.SGI.8.8.8-aspam-6.2/980310.SGI-aspam) via ESMTP id MAA12727; Wed, 14 Mar 2001 12:56:02 -0800 (PST)
	mail_from (nobody@info.engr.sgi.com)
Received: from info.engr.sgi.com (info.engr.sgi.com [192.26.80.216])
	by nodin.corp.sgi.com (8.11.2/8.11.2/nodin-1.0) with ESMTP id f2EKuCE2342662
	for <linux-xfs@oss.sgi.com>; Wed, 14 Mar 2001 12:56:12 -0800 (PST)
Received: (from nobody@localhost)
	by info.engr.sgi.com (SGI-8.9.3/8.9.3) id MAA82985;
	Wed, 14 Mar 2001 12:56:11 -0800 (PST)
Date:   Wed, 14 Mar 2001 12:56:11 -0800 (PST)
Message-Id: <200103142056.MAA82985@info.engr.sgi.com>
X-Pv-Incident: 802021 
webPV:  oldchicago.americas.sgi.com 
webExec: webpvupdate,pvincident 
Reply-To: sgi.bugs.xfs@fido.engr.sgi.com
From:   pv@relay.sgi.com (nb@sgi.com)
Subject: REASSIGN 802021 - hang in _pagebuf_find_lockable_buffer
To:     raa@sgi.com
Cc:     linux-xfs@oss.sgi.com
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing

View Incident: http://co-op.engr.sgi.com/BugWorks/code/bwxquery.cgi?search=Search&wlong=1&view_type=Bug&wi=802021

 Status : open                         Priority : 1                         
*Assigned Engineer : raa               Submitter : nathans                  
 Project : xfs-linux                   Assigned Group : xfs-linux           
 Opened Date : 09/17/00               *Modified User : nb                   
*Modified User Domain : sgi.com       *Description :
An XFS qa run hung my desktop machine over the weekend.  This
was using the top-of-tree source as of Sunday morning (localtime)
- I don't think anything has been checked in since - in qa test
number 013 (fsstress).

The machine was responsive to ping, but not to telnet/rlogin,
the kernel backtrace was as follows...


Entering kdb (current=0xc4622000, pid 5038) due to Keyboard Entry

.....


==========================
ADDITIONAL INFORMATION (REASSIGN)
From: nb@sgi.com (BugWorks)
Date: Mar 14 2001 12:56:10PM
==========================

From owner-linux-xfs@oss.sgi.com Wed Mar 14 12:57:51 2001
Received:  by oss.sgi.com id <S553825AbRCNU5l>;
	Wed, 14 Mar 2001 12:57:41 -0800
Received: from [141.154.95.60] ([141.154.95.60]:25737 "EHLO rain")
	by oss.sgi.com with ESMTP id <S553815AbRCNU52>;
	Wed, 14 Mar 2001 12:57:28 -0800
Received: from localhost ([127.0.0.1])
	by rain with esmtp (Exim 3.12 #1 (Debian))
	id 14dIJq-0000XQ-00; Wed, 14 Mar 2001 15:56:30 -0500
Subject: Re: gcc 2.95.2
From:   Vladimir Vukicevic <vladimir@ximian.com>
To:     Per Andreas Buer <per@linpro.no>
Cc:     linux-xfs@oss.sgi.com
In-Reply-To: <ul6d7bkusy5.fsf@false.linpro.no>
References: <gcs@agentsinside.com>
	<200103141615.f2EGFrZ13617@jen.americas.sgi.com> 
	<ul6d7bkusy5.fsf@false.linpro.no>
Content-Type: text/plain
X-Mailer: Evolution (0.9/+cvs.2001.03.12.09.00 - Preview Release)
Date:   14 Mar 2001 15:56:30 -0500
Mime-Version: 1.0
Message-Id: <E14dIJq-0000XQ-00@rain>
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing

On 14 Mar 2001 19:05:38 +0100, Per Andreas Buer wrote:
> Steve Lord <lord@sgi.com> writes:
> 
> > I cannot speak for 2.95.2, 
> 
> I can. 
> 
> The kernel, compiled from CVS 2001-03-04, compiled nicely. Basic testing
> ( 42 processes doing "cp /tmp/1 /tmp/2" and "sync") for just a few
> seconds resulted in the attached crash. With egcs-2.91.66 this kernel
> worked flawlessly.


The gcc that is currently in Debian testing/unstable, gcc version 2.95.3
20010125 (prerelease), works fine -- two of us here have been running
XFS kernels built with this gcc for a while now...


    - Vlad



From owner-linux-xfs@oss.sgi.com Wed Mar 14 13:03:53 2001
Received:  by oss.sgi.com id <S553825AbRCNVDn>;
	Wed, 14 Mar 2001 13:03:43 -0800
Received: from sgi.SGI.COM ([192.48.153.1]:35936 "EHLO sgi.com")
	by oss.sgi.com with ESMTP id <S553823AbRCNVDe>;
	Wed, 14 Mar 2001 13:03:34 -0800
Received: from info.engr.sgi.com ([192.26.80.216]) 
	by sgi.com (980327.SGI.8.8.8-aspam/980304.SGI-aspam:
       SGI does not authorize the use of its proprietary
       systems or networks for unsolicited or bulk email
       from the Internet.) 
	via ESMTP id NAA07350; Wed, 14 Mar 2001 13:01:23 -0800 (PST)
	mail_from (nobody@info.engr.sgi.com)
Received: (from nobody@localhost)
	by info.engr.sgi.com (SGI-8.9.3/8.9.3) id NAA36286;
	Wed, 14 Mar 2001 13:00:21 -0800 (PST)
Date:   Wed, 14 Mar 2001 13:00:21 -0800 (PST)
Message-Id: <200103142100.NAA36286@info.engr.sgi.com>
X-Pv-Incident: 802340 
webPV:  jen.americas.sgi.com 
webExec: webpvupdate,pvincident 
Reply-To: sgi.bugs.xfs@fido.engr.sgi.com
From:   pv@odin.corp.sgi.com (lord@sgi.com)
Subject: CLOSE 802340 - kernel panic
To:     raa@sgi.com
Cc:     linux-xfs@oss.sgi.com
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing

View Incident: http://co-op.engr.sgi.com/BugWorks/code/bwxquery.cgi?search=Search&wlong=1&view_type=Bug&wi=802340

*Status : closed                       Priority : 1                         
 Assigned Engineer : raa               Submitter : coreym                   
 Opened Date : 09/20/00               *Closed Date : 03/14/01               
*Fixed By : lord                      *Fixed By Domain : sgi.com            
*Modified Date : 03/14/01             *Modified User : lord                 
*Modified User Domain : sgi.com       *Fix Description :
==========================
ADDITIONAL INFORMATION (CLOSE)
From: lord@sgi.com (BugWorks)
Date: Mar 14 2001 01:00:21PM
==========================

I am going to close this one, things have changed a lot since
September. Lets see if this still happens on the 4G system.

From owner-linux-xfs@oss.sgi.com Wed Mar 14 13:08:53 2001
Received:  by oss.sgi.com id <S553825AbRCNVIn>;
	Wed, 14 Mar 2001 13:08:43 -0800
Received: from deliverator.sgi.com ([204.94.214.10]:39791 "EHLO
        deliverator.sgi.com") by oss.sgi.com with ESMTP id <S553823AbRCNVIe>;
	Wed, 14 Mar 2001 13:08:34 -0800
Received: from info.engr.sgi.com (info.engr.sgi.com [192.26.80.216]) by deliverator.sgi.com (980309.SGI.8.8.8-aspam-6.2/980310.SGI-aspam) via ESMTP id NAA14976; Wed, 14 Mar 2001 13:07:23 -0800 (PST)
	mail_from (nobody@info.engr.sgi.com)
Received: (from nobody@localhost)
	by info.engr.sgi.com (SGI-8.9.3/8.9.3) id NAA04965;
	Wed, 14 Mar 2001 13:08:33 -0800 (PST)
Date:   Wed, 14 Mar 2001 13:08:33 -0800 (PST)
Message-Id: <200103142108.NAA04965@info.engr.sgi.com>
X-Pv-Incident: 802021 
webPV:  jen.americas.sgi.com 
webExec: webpvupdate,pvincident 
Reply-To: sgi.bugs.xfs@fido.engr.sgi.com
From:   pv@relay.sgi.com (lord@sgi.com)
Subject: CLOSE 802021 - hang in _pagebuf_find_lockable_buffer
To:     raa@sgi.com
Cc:     linux-xfs@oss.sgi.com
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing

View Incident: http://co-op.engr.sgi.com/BugWorks/code/bwxquery.cgi?search=Search&wlong=1&view_type=Bug&wi=802021

*Status : closed                       Priority : 1                         
 Assigned Engineer : raa               Submitter : nathans                  
 Opened Date : 09/17/00               *Closed Date : 03/14/01               
*Fixed By : lord                      *Fixed By Domain : sgi.com            
*Modified Date : 03/14/01             *Modified User : lord                 
*Modified User Domain : sgi.com       *Fix Description :
==========================
ADDITIONAL INFORMATION (CLOSE)
From: lord@sgi.com (BugWorks)
Date: Mar 14 2001 01:08:33PM
==========================

Closing as ancient history, these tests get run every night, I
have not heard about it happening again.

From owner-linux-xfs@oss.sgi.com Wed Mar 14 13:40:13 2001
Received:  by oss.sgi.com id <S553828AbRCNVjx>;
	Wed, 14 Mar 2001 13:39:53 -0800
Received: from deliverator.sgi.com ([204.94.214.10]:59261 "EHLO
        deliverator.sgi.com") by oss.sgi.com with ESMTP id <S553825AbRCNVjc>;
	Wed, 14 Mar 2001 13:39:32 -0800
Received: from info.engr.sgi.com (info.engr.sgi.com [192.26.80.216]) by deliverator.sgi.com (980309.SGI.8.8.8-aspam-6.2/980310.SGI-aspam) via ESMTP id NAA22834; Wed, 14 Mar 2001 13:38:21 -0800 (PST)
	mail_from (nobody@info.engr.sgi.com)
Received: (from nobody@localhost)
	by info.engr.sgi.com (SGI-8.9.3/8.9.3) id NAA78149;
	Wed, 14 Mar 2001 13:39:31 -0800 (PST)
Date:   Wed, 14 Mar 2001 13:39:31 -0800 (PST)
Message-Id: <200103142139.NAA78149@info.engr.sgi.com>
X-Pv-Incident: 809716 
webPV:  oldchicago.americas.sgi.com 
webExec: webpvupdate,pvincident 
Reply-To: sgi.bugs.xfs@fido.engr.sgi.com
From:   pv@relay.sgi.com (nb@sgi.com)
Subject: REASSIGN 809716 - XFS Linux and kiobuf io results in data corruption.
To:     raa@sgi.com
Cc:     linux-xfs@oss.sgi.com
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing

View Incident: http://co-op.engr.sgi.com/BugWorks/code/bwxquery.cgi?search=Search&wlong=1&view_type=Bug&wi=809716

 Status : open                         Priority : 3                         
*Assigned Engineer : raa               Submitter : cattelan                 
 Project : xfs-linux                   Assigned Group : xfs-linux           
 Opened Date : 12/07/00               *Modified User : nb                   
*Modified User Domain : sgi.com       *Description :
Dual processor machine running linux XFS 2.4.0-test11
heavy doio tests, results in data corruption.

#!/bin/sh
export PATH=$PATH:/usr/tests/rts/bin


iogen -i 0 -T 128b 468973k:doio_1 | doio -akv -n 4 -m 1000 &
iogen -i 0 -T 128b 468973k:doio_2 | doio -akv -n 4 -m 1000 &
iogen -i 0 -T 128b 468973k:doio_3 | doio -akv -n 4 -m 1000 &

.....


==========================
ADDITIONAL INFORMATION (REASSIGN)
From: nb@sgi.com (BugWorks)
Date: Mar 14 2001 01:39:30PM
==========================

From owner-linux-xfs@oss.sgi.com Wed Mar 14 13:41:13 2001
Received:  by oss.sgi.com id <S553828AbRCNVky>;
	Wed, 14 Mar 2001 13:40:54 -0800
Received: from deliverator.sgi.com ([204.94.214.10]:27006 "EHLO
        deliverator.sgi.com") by oss.sgi.com with ESMTP id <S553826AbRCNVkm>;
	Wed, 14 Mar 2001 13:40:42 -0800
Received: from info.engr.sgi.com (info.engr.sgi.com [192.26.80.216]) by deliverator.sgi.com (980309.SGI.8.8.8-aspam-6.2/980310.SGI-aspam) via ESMTP id NAA23104; Wed, 14 Mar 2001 13:39:31 -0800 (PST)
	mail_from (nobody@info.engr.sgi.com)
Received: (from nobody@localhost)
	by info.engr.sgi.com (SGI-8.9.3/8.9.3) id NAA00157;
	Wed, 14 Mar 2001 13:40:40 -0800 (PST)
Date:   Wed, 14 Mar 2001 13:40:40 -0800 (PST)
Message-Id: <200103142140.NAA00157@info.engr.sgi.com>
X-Pv-Incident: 804570 
webPV:  oldchicago.americas.sgi.com 
webExec: bugs_update,pvincident 
Reply-To: sgi.bugs.xfs@fido.engr.sgi.com
From:   pv@relay.sgi.com (nb@sgi.com)
Subject: REASSIGN 804570 - The elevator bug
To:     raa@sgi.com
Cc:     raa@sgi.com, huovinen@sgi.com, cattelan@sgi.com, mann@sgi.com,
        tbd@sgi.com, rrl@sgi.com, alaffin@sgi.com, cbrady@sgi.com,
        eac@sgi.com, hasib@sgi.com, coreym@sgi.com, steiner@sgi.com,
        linux-xfs@oss.sgi.com
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing

View Incident: http://co-op.engr.sgi.com/BugWorks/code/bwxquery.cgi?search=Search&wlong=1&view_type=Bug&wi=804570

*Assigned Engineer : raa              *Assigned Domain : sgi.com            
*Assigned Group : xfs-linux           *Modified Date : 03/14/01             
*Modified User : nb                   *Modified User Domain : sgi.com       
*Description :
I ran rwtest on a xfs-linux filesystem on the machine permit:

rwtest 100000000:/mnt1/file_1 >/tmp/file_1.out 2>&1 &

This caused df, ls -l, fdisk, and top all to hang.

Permit is running Redhat 6.2 with a 2.4.0-test5 kernel

==========================
ADDITIONAL INFORMATION (ADD)

.....


==========================
ADDITIONAL INFORMATION (REASSIGN)
From: nb@sgi.com (BWX)
Date: Mar 14 2001 01:40:40PM
==========================

RE-assigning to Bob

From owner-linux-xfs@oss.sgi.com Wed Mar 14 13:41:13 2001
Received:  by oss.sgi.com id <S553829AbRCNVkx>;
	Wed, 14 Mar 2001 13:40:53 -0800
Received: from deliverator.sgi.com ([204.94.214.10]:28286 "EHLO
        deliverator.sgi.com") by oss.sgi.com with ESMTP id <S553828AbRCNVko>;
	Wed, 14 Mar 2001 13:40:44 -0800
Received: from info.engr.sgi.com (info.engr.sgi.com [192.26.80.216]) by deliverator.sgi.com (980309.SGI.8.8.8-aspam-6.2/980310.SGI-aspam) via ESMTP id NAA23117; Wed, 14 Mar 2001 13:39:32 -0800 (PST)
	mail_from (nobody@info.engr.sgi.com)
Received: (from nobody@localhost)
	by info.engr.sgi.com (SGI-8.9.3/8.9.3) id NAA98807;
	Wed, 14 Mar 2001 13:40:42 -0800 (PST)
Date:   Wed, 14 Mar 2001 13:40:42 -0800 (PST)
Message-Id: <200103142140.NAA98807@info.engr.sgi.com>
X-Pv-Incident: 809716 
webPV:  oldchicago.americas.sgi.com 
webExec: bugs_update,pvincident 
Reply-To: sgi.bugs.xfs@fido.engr.sgi.com
From:   pv@relay.sgi.com (nb@sgi.com)
Subject: REASSIGN 809716 - XFS Linux and kiobuf io results in data corruption.
To:     raa@sgi.com
Cc:     linux-xfs@oss.sgi.com
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing

View Incident: http://co-op.engr.sgi.com/BugWorks/code/bwxquery.cgi?search=Search&wlong=1&view_type=Bug&wi=809716

*Assigned Engineer : raa              *Assigned Domain : sgi.com            
*Assigned Group : xfs-linux           *Modified Date : 03/14/01             
*Modified User : nb                   *Modified User Domain : sgi.com       
*Description :
Dual processor machine running linux XFS 2.4.0-test11
heavy doio tests, results in data corruption.

#!/bin/sh
export PATH=$PATH:/usr/tests/rts/bin


iogen -i 0 -T 128b 468973k:doio_1 | doio -akv -n 4 -m 1000 &
iogen -i 0 -T 128b 468973k:doio_2 | doio -akv -n 4 -m 1000 &
iogen -i 0 -T 128b 468973k:doio_3 | doio -akv -n 4 -m 1000 &

.....


==========================
ADDITIONAL INFORMATION (REASSIGN)
From: nb@sgi.com (BWX)
Date: Mar 14 2001 01:40:42PM
==========================

RE-assigning to Bob

From owner-linux-xfs@oss.sgi.com Wed Mar 14 13:41:14 2001
Received:  by oss.sgi.com id <S553830AbRCNVkx>;
	Wed, 14 Mar 2001 13:40:53 -0800
Received: from deliverator.sgi.com ([204.94.214.10]:30078 "EHLO
        deliverator.sgi.com") by oss.sgi.com with ESMTP id <S553829AbRCNVkp>;
	Wed, 14 Mar 2001 13:40:45 -0800
Received: from info.engr.sgi.com (info.engr.sgi.com [192.26.80.216]) by deliverator.sgi.com (980309.SGI.8.8.8-aspam-6.2/980310.SGI-aspam) via ESMTP id NAA23129; Wed, 14 Mar 2001 13:39:34 -0800 (PST)
	mail_from (nobody@info.engr.sgi.com)
Received: (from nobody@localhost)
	by info.engr.sgi.com (SGI-8.9.3/8.9.3) id NAA79965;
	Wed, 14 Mar 2001 13:40:44 -0800 (PST)
Date:   Wed, 14 Mar 2001 13:40:44 -0800 (PST)
Message-Id: <200103142140.NAA79965@info.engr.sgi.com>
X-Pv-Incident: 811057 
webPV:  oldchicago.americas.sgi.com 
webExec: bugs_update,pvincident 
Reply-To: sgi.bugs.xfs@fido.engr.sgi.com
From:   pv@relay.sgi.com (nb@sgi.com)
Subject: REASSIGN 811057 - XFS linux may not be correctly working with LVM
To:     raa@sgi.com
Cc:     linux-xfs@oss.sgi.com
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing

View Incident: http://co-op.engr.sgi.com/BugWorks/code/bwxquery.cgi?search=Search&wlong=1&view_type=Bug&wi=811057

*Assigned Engineer : raa              *Assigned Domain : sgi.com            
*Assigned Group : xfs-linux           *Modified Date : 03/14/01             
*Modified User : nb                   *Modified User Domain : sgi.com       
*Description :
It appears that we still have a problem with XFS running 
on a striped LVM volume.

At first glance it appears that log writes may be off
by one or more blocks.

The data dump of the xfs_logprint would suggest that
block 3369 should really be at 3368.

This was on test11.

.....


==========================
ADDITIONAL INFORMATION (REASSIGN)
From: nb@sgi.com (BWX)
Date: Mar 14 2001 01:40:43PM
==========================

RE-assigning to Bob

From owner-linux-xfs@oss.sgi.com Wed Mar 14 13:41:14 2001
Received:  by oss.sgi.com id <S553836AbRCNVkx>;
	Wed, 14 Mar 2001 13:40:53 -0800
Received: from deliverator.sgi.com ([204.94.214.10]:31614 "EHLO
        deliverator.sgi.com") by oss.sgi.com with ESMTP id <S553830AbRCNVkr>;
	Wed, 14 Mar 2001 13:40:47 -0800
Received: from info.engr.sgi.com (info.engr.sgi.com [192.26.80.216]) by deliverator.sgi.com (980309.SGI.8.8.8-aspam-6.2/980310.SGI-aspam) via ESMTP id NAA23140; Wed, 14 Mar 2001 13:39:35 -0800 (PST)
	mail_from (nobody@info.engr.sgi.com)
Received: (from nobody@localhost)
	by info.engr.sgi.com (SGI-8.9.3/8.9.3) id NAA44980;
	Wed, 14 Mar 2001 13:40:45 -0800 (PST)
Date:   Wed, 14 Mar 2001 13:40:45 -0800 (PST)
Message-Id: <200103142140.NAA44980@info.engr.sgi.com>
X-Pv-Incident: 813113 
webPV:  oldchicago.americas.sgi.com 
webExec: bugs_update,pvincident 
Reply-To: sgi.bugs.xfs@fido.engr.sgi.com
From:   pv@relay.sgi.com (nb@sgi.com)
Subject: REASSIGN 813113 - kswapd issues
To:     raa@sgi.com
Cc:     linux-xfs@oss.sgi.com
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing

View Incident: http://co-op.engr.sgi.com/BugWorks/code/bwxquery.cgi?search=Search&wlong=1&view_type=Bug&wi=813113

*Assigned Engineer : raa              *Assigned Domain : sgi.com            
*Assigned Group : xfs-linux           *Modified Date : 03/14/01             
*Modified User : nb                   *Modified User Domain : sgi.com       
*Description :
Below is the state of a system that has kswapd (pid = 3)
and a process performing I/O (pid = 1120) that are deadlocked
waiting for each other. I'm sure that the wakeup_kswapd()
changes in 2.4.1preX change this situation somewhat. However,
in the long run, allocating memory under IOLOCK could
still leave avenues for similar problems.

Steve, we'll use this bug to record problems in this area.
This instance is the fdatasync case, while running a 4-threaded
doio on a 64MB machine (takes about an hour to hit it).

.....


==========================
ADDITIONAL INFORMATION (REASSIGN)
From: nb@sgi.com (BWX)
Date: Mar 14 2001 01:40:45PM
==========================

RE-assigning to Bob

From owner-linux-xfs@oss.sgi.com Wed Mar 14 14:02:23 2001
Received:  by oss.sgi.com id <S553830AbRCNWCN>;
	Wed, 14 Mar 2001 14:02:13 -0800
Received: from pneumatic-tube.sgi.com ([204.94.214.22]:53304 "EHLO
        pneumatic-tube.sgi.com") by oss.sgi.com with ESMTP
	id <S553826AbRCNWCB>; Wed, 14 Mar 2001 14:02:01 -0800
Received: from info.engr.sgi.com (info.engr.sgi.com [192.26.80.216]) by pneumatic-tube.sgi.com (980327.SGI.8.8.8-aspam/980310.SGI-aspam) via ESMTP id OAA06907; Wed, 14 Mar 2001 14:11:41 -0800 (PST)
	mail_from (nobody@info.engr.sgi.com)
Received: (from nobody@localhost)
	by info.engr.sgi.com (SGI-8.9.3/8.9.3) id OAA54603;
	Wed, 14 Mar 2001 14:00:35 -0800 (PST)
Date:   Wed, 14 Mar 2001 14:00:35 -0800 (PST)
Message-Id: <200103142200.OAA54603@info.engr.sgi.com>
X-Pv-Incident: 813113 
webPV:  jen.americas.sgi.com 
webExec: webpvupdate,pvincident 
Reply-To: sgi.bugs.xfs@fido.engr.sgi.com
From:   pv@relay.sgi.com (lord@sgi.com)
Subject: CLOSE 813113 - kswapd issues
To:     raa@sgi.com
Cc:     linux-xfs@oss.sgi.com
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing

View Incident: http://co-op.engr.sgi.com/BugWorks/code/bwxquery.cgi?search=Search&wlong=1&view_type=Bug&wi=813113

*Status : closed                       Priority : 2                         
 Assigned Engineer : raa               Submitter : ananth                   
 Opened Date : 01/22/01               *Closed Date : 03/14/01               
*Fixed By : lord                      *Fixed By Domain : sgi.com            
*Modified Date : 03/14/01             *Modified User : lord                 
*Modified User Domain : sgi.com       *Fix Description :
==========================
ADDITIONAL INFORMATION (CLOSE)
From: lord@sgi.com (BugWorks)
Date: Mar 14 2001 02:00:35PM
==========================

We have made changes since this was opened to work around or fix
these issues.

From owner-linux-xfs@oss.sgi.com Wed Mar 14 14:02:32 2001
Received:  by oss.sgi.com id <S553826AbRCNWCX>;
	Wed, 14 Mar 2001 14:02:23 -0800
Received: from pneumatic-tube.sgi.com ([204.94.214.22]:49208 "EHLO
        pneumatic-tube.sgi.com") by oss.sgi.com with ESMTP
	id <S553825AbRCNWCC>; Wed, 14 Mar 2001 14:02:02 -0800
Received: from info.engr.sgi.com (info.engr.sgi.com [192.26.80.216]) by pneumatic-tube.sgi.com (980327.SGI.8.8.8-aspam/980310.SGI-aspam) via ESMTP id OAA09217; Wed, 14 Mar 2001 14:11:40 -0800 (PST)
	mail_from (nobody@info.engr.sgi.com)
Received: (from nobody@localhost)
	by info.engr.sgi.com (SGI-8.9.3/8.9.3) id OAA65739;
	Wed, 14 Mar 2001 14:01:04 -0800 (PST)
Date:   Wed, 14 Mar 2001 14:01:04 -0800 (PST)
Message-Id: <200103142201.OAA65739@info.engr.sgi.com>
X-Pv-Incident: 809716 
webPV:  jen.americas.sgi.com 
webExec: webpvupdate,pvincident 
Reply-To: sgi.bugs.xfs@fido.engr.sgi.com
From:   pv@relay.sgi.com (lord@sgi.com)
Subject: CLOSE 809716 - XFS Linux and kiobuf io results in data corruption.
To:     raa@sgi.com
Cc:     linux-xfs@oss.sgi.com
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing

View Incident: http://co-op.engr.sgi.com/BugWorks/code/bwxquery.cgi?search=Search&wlong=1&view_type=Bug&wi=809716

*Status : closed                       Priority : 3                         
 Assigned Engineer : raa               Submitter : cattelan                 
 Opened Date : 12/07/00               *Closed Date : 03/14/01               
*Fixed By : lord                      *Fixed By Domain : sgi.com            
*Modified Date : 03/14/01             *Modified User : lord                 
*Modified User Domain : sgi.com       *Fix Description :
==========================
ADDITIONAL INFORMATION (CLOSE)
From: lord@sgi.com (BugWorks)
Date: Mar 14 2001 02:01:03PM
==========================

This testcase no longer fails

From owner-linux-xfs@oss.sgi.com Wed Mar 14 14:06:03 2001
Received:  by oss.sgi.com id <S553828AbRCNWFn>;
	Wed, 14 Mar 2001 14:05:43 -0800
Received: from pneumatic-tube.sgi.com ([204.94.214.22]:37433 "EHLO
        pneumatic-tube.sgi.com") by oss.sgi.com with ESMTP
	id <S553825AbRCNWFf>; Wed, 14 Mar 2001 14:05:35 -0800
Received: from info.engr.sgi.com (info.engr.sgi.com [192.26.80.216]) by pneumatic-tube.sgi.com (980327.SGI.8.8.8-aspam/980310.SGI-aspam) via ESMTP id OAA02876; Wed, 14 Mar 2001 14:15:26 -0800 (PST)
	mail_from (nobody@info.engr.sgi.com)
Received: (from nobody@localhost)
	by info.engr.sgi.com (SGI-8.9.3/8.9.3) id OAA18149;
	Wed, 14 Mar 2001 14:05:33 -0800 (PST)
Date:   Wed, 14 Mar 2001 14:05:33 -0800 (PST)
Message-Id: <200103142205.OAA18149@info.engr.sgi.com>
X-Pv-Incident: 804570 
webPV:  jen.americas.sgi.com 
webExec: webpvupdate,pvincident 
Reply-To: sgi.bugs.xfs@fido.engr.sgi.com
From:   pv@relay.sgi.com (lord@sgi.com)
Subject: CLOSE 804570 - The elevator bug
To:     raa@sgi.com
Cc:     raa@sgi.com, huovinen@sgi.com, cattelan@sgi.com, mann@sgi.com,
        tbd@sgi.com, rrl@sgi.com, alaffin@sgi.com, cbrady@sgi.com,
        eac@sgi.com, hasib@sgi.com, coreym@sgi.com, steiner@sgi.com,
        linux-xfs@oss.sgi.com
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing

View Incident: http://co-op.engr.sgi.com/BugWorks/code/bwxquery.cgi?search=Search&wlong=1&view_type=Bug&wi=804570

*Status : closed                       Priority : 3                         
 Assigned Engineer : raa               Submitter : coreym                   
 Opened Date : 10/11/00               *Closed Date : 03/14/01               
*Fixed By : lord                      *Fixed By Domain : sgi.com            
*Modified Date : 03/14/01             *Modified User : lord                 
*Modified User Domain : sgi.com       *Fix Description :
==========================
ADDITIONAL INFORMATION (CLOSE)
From: steiner@sgi.com (BugWorks)
Date: Jan 16 2001 01:54:23PM
==========================

This code was checked in to the 2.4.0-test12 tree. I thought I 
closed the PV but it appears I didnt......

==========================
ADDITIONAL INFORMATION (CLOSE)
From: lord@sgi.com (BugWorks)
Date: Mar 14 2001 02:05:32PM
==========================

We have merged two more kernel releases since this time, plus
Ananth changed the I/O path to do the right thing. If we still
have a problem this can be reopened, but I think it should be
gone now.

From owner-linux-xfs@oss.sgi.com Wed Mar 14 14:10:12 2001
Received:  by oss.sgi.com id <S553832AbRCNWKD>;
	Wed, 14 Mar 2001 14:10:03 -0800
Received: from sgi.SGI.COM ([192.48.153.1]:39030 "EHLO sgi.com")
	by oss.sgi.com with ESMTP id <S553830AbRCNWJk>;
	Wed, 14 Mar 2001 14:09:40 -0800
Received: from info.engr.sgi.com ([192.26.80.216]) 
	by sgi.com (980327.SGI.8.8.8-aspam/980304.SGI-aspam:
       SGI does not authorize the use of its proprietary
       systems or networks for unsolicited or bulk email
       from the Internet.) 
	via ESMTP id OAA03281; Wed, 14 Mar 2001 14:08:49 -0800 (PST)
	mail_from (nobody@info.engr.sgi.com)
Received: (from nobody@localhost)
	by info.engr.sgi.com (SGI-8.9.3/8.9.3) id OAA31799;
	Wed, 14 Mar 2001 14:08:42 -0800 (PST)
Date:   Wed, 14 Mar 2001 14:08:42 -0800 (PST)
Message-Id: <200103142208.OAA31799@info.engr.sgi.com>
X-Pv-Incident: 768265 
webPV:  jen.americas.sgi.com 
webExec: webpvupdate,pvincident 
Reply-To: sgi.bugs.xfs@fido.engr.sgi.com
From:   pv@odin.corp.sgi.com (lord@sgi.com)
Subject: CLOSE 768265 - Implement/Port Direct I/O for XFS on Linux
To:     btg@sgi.com
Cc:     linux-xfs@oss.sgi.com
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing

View Incident: http://co-op.engr.sgi.com/BugWorks/code/bwxquery.cgi?search=Search&wlong=1&view_type=Bug&wi=768265

*Status : closed                       Priority : 2                         
 Assigned Engineer : btg               Submitter : mostek                   
 Opened Date : 09/23/99               *Closed Date : 03/14/01               
*Fixed By : lord                      *Fixed By Domain : sgi.com            
*Modified Date : 03/14/01             *Modified User : lord                 
*Modified User Domain : sgi.com       *Fix Description :
==========================
ADDITIONAL INFORMATION (CLOSE)
From: lord@sgi.com (BugWorks)
Date: Mar 14 2001 02:08:42PM
==========================

This is implemented

From owner-linux-xfs@oss.sgi.com Wed Mar 14 15:19:04 2001
Received:  by oss.sgi.com id <S553840AbRCNXSo>;
	Wed, 14 Mar 2001 15:18:44 -0800
Received: from sgi.SGI.COM ([192.48.153.1]:60688 "EHLO sgi.com")
	by oss.sgi.com with ESMTP id <S553838AbRCNXSN>;
	Wed, 14 Mar 2001 15:18:13 -0800
Received: from zeus-fddi.americas.sgi.com (128-162-8-103.americas.sgi.com [128.162.8.103]) 
	by sgi.com (980327.SGI.8.8.8-aspam/980304.SGI-aspam:
       SGI does not authorize the use of its proprietary
       systems or networks for unsolicited or bulk email
       from the Internet.) 
	via ESMTP id PAA07765
	for <linux-xfs@oss.sgi.com>; Wed, 14 Mar 2001 15:18:06 -0800 (PST)
	mail_from (lord@sgi.com)
Received: from daisy-e185.americas.sgi.com (daisy-e194.americas.sgi.com [128.162.194.214]) by zeus-fddi.americas.sgi.com (8.9.3/americas-smart-nospam1.1) with ESMTP id RAA1136991 for <linux-xfs@oss.sgi.com>; Wed, 14 Mar 2001 17:16:39 -0600 (CST)
Received: from jen.americas.sgi.com (IDENT:root@jen.americas.sgi.com [128.162.184.86]) by daisy-e185.americas.sgi.com (SGI-8.9.3/SGI-server-1.7) with ESMTP id RAA85257 for <linux-xfs@oss.sgi.com>; Wed, 14 Mar 2001 17:16:39 -0600 (CST)
From:   Steve Lord <lord@sgi.com>
Received: by jen.americas.sgi.com (8.11.0/SGI-client-1.7) id f2ENHOn06725; Wed, 14 Mar 2001 17:17:24 -0600
Message-Id: <200103142317.f2ENHOn06725@jen.americas.sgi.com>
Date:   Wed, 14 Mar 2001 17:17:24 -0600
Subject: TAKE - fix pbtrace xxxx in kdb
To:     unlisted-recipients:; (no To-header on input)
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing

This drove me batty for months looking at the code trying to work out
why this was busted. Finally looked at the right code and fixed it!

Date:  Wed Mar 14 15:15:41 PST 2001
Workarea:  jen.americas.sgi.com:/src/lord/linux-base

The following file(s) were checked into:
  bonnie.engr.sgi.com:/isms/slinx/2.4.x-xfs


Modid:  2.4.x-xfs:slinx:89987a
linux/kdb/modules/kdbm_pg.c - 1.30
	- Fix pbtrace command in kdb - make arguments work again.



From owner-linux-xfs@oss.sgi.com Wed Mar 14 15:39:33 2001
Received:  by oss.sgi.com id <S553750AbRCNXjX>;
	Wed, 14 Mar 2001 15:39:23 -0800
Received: from deliverator.sgi.com ([204.94.214.10]:29752 "EHLO
        deliverator.sgi.com") by oss.sgi.com with ESMTP id <S553746AbRCNXi7>;
	Wed, 14 Mar 2001 15:38:59 -0800
Received: from zeus-fddi.americas.sgi.com (zeus-fddi.americas.sgi.com [128.162.8.103]) by deliverator.sgi.com (980309.SGI.8.8.8-aspam-6.2/980310.SGI-aspam) via ESMTP id PAA21120
	for <linux-xfs@oss.sgi.com>; Wed, 14 Mar 2001 15:37:48 -0800 (PST)
	mail_from (lord@sgi.com)
Received: from daisy-e185.americas.sgi.com (daisy-e194.americas.sgi.com [128.162.194.214]) by zeus-fddi.americas.sgi.com (8.9.3/americas-smart-nospam1.1) with ESMTP id RAA1140733 for <linux-xfs@oss.sgi.com>; Wed, 14 Mar 2001 17:37:43 -0600 (CST)
Received: from jen.americas.sgi.com (IDENT:root@jen.americas.sgi.com [128.162.184.86]) by daisy-e185.americas.sgi.com (SGI-8.9.3/SGI-server-1.7) with ESMTP id RAA92168 for <linux-xfs@oss.sgi.com>; Wed, 14 Mar 2001 17:37:42 -0600 (CST)
From:   Steve Lord <lord@sgi.com>
Received: by jen.americas.sgi.com (8.11.0/SGI-client-1.7) id f2ENcRa09059; Wed, 14 Mar 2001 17:38:27 -0600
Message-Id: <200103142338.f2ENcRa09059@jen.americas.sgi.com>
Date:   Wed, 14 Mar 2001 17:38:27 -0600
Subject: TAKE - pagebuf fixes for address space remapping
To:     unlisted-recipients:; (no To-header on input)
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing

The dreaded remapping is back, but mostly for filesystems created with
oddball mkfs options, the standard filesystem config will not touch it
except possibly in recovery. Also fix the mapping for the cases were we
bring stuff into memory via a readahead call. The code in pagebuf is now
ifdefed for easy removal. 

This also contains some cleanup in the interface between xfs and pagebuf,
and a minor fix for the shutdown on error code.

Date:  Wed Mar 14 15:35:05 PST 2001
Workarea:  jen.americas.sgi.com:/src/lord/linux-base

The following file(s) were checked into:
  bonnie.engr.sgi.com:/isms/slinx/2.4.x-xfs


Modid:  2.4.x-xfs:slinx:90004a
linux/fs/xfs/xfs_buf.h - 1.69
	- directly manipulate some pagebuf fields rather than calling one line
	  functions. Fix xfs_bdwrite and xfs_bawrite so they correctly set the
	  strategy callout.

linux/fs/xfs/linux/xfs_lrw.c - 1.84
	- remove some code we do not need

linux/include/linux/page_buf.h - 1.82
	- remove some prototypes no longer needed, streamline some others.

linux/fs/pagebuf/page_buf.c - 1.65
	- Reintroduce the remapping code - it turns out we still have cases which
	  use it, however, now it is switchable at compile time. Remove a coupld
	  of functions we do not need, and fix one case in the address space
	  remapping code where the pages are already in memory, but were put
	  there by a readahead call which did not do the remapping.



From owner-linux-xfs@oss.sgi.com Wed Mar 14 17:03:54 2001
Received:  by oss.sgi.com id <S553753AbRCOBDe>;
	Wed, 14 Mar 2001 17:03:34 -0800
Received: from deliverator.sgi.com ([204.94.214.10]:21584 "EHLO
        deliverator.sgi.com") by oss.sgi.com with ESMTP id <S553752AbRCOBDJ>;
	Wed, 14 Mar 2001 17:03:09 -0800
Received: from sherman.melbourne.sgi.com (sherman.melbourne.sgi.com [134.14.55.175]) by deliverator.sgi.com (980309.SGI.8.8.8-aspam-6.2/980310.SGI-aspam) via ESMTP id RAA03591
	for <linux-xfs@oss.sgi.com>; Wed, 14 Mar 2001 17:01:56 -0800 (PST)
	mail_from (kaos@sherman.melbourne.sgi.com)
Received: (from kaos@localhost)
	by sherman.melbourne.sgi.com (8.9.3/8.9.3) id MAA07184;
	Thu, 15 Mar 2001 12:03:04 +1100
Date:   Thu, 15 Mar 2001 12:03:04 +1100
From:   Keith Owens <kaos@sherman.melbourne.sgi.com>
Message-Id: <200103150103.MAA07184@sherman.melbourne.sgi.com>
Subject: TAKE - Special cases in xfs makefile
To:     unlisted-recipients:; (no To-header on input)
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing

The xfs support reorganization missed a couple of special cases.
Directory grio does not exist and has been removed from xfs/Makefile
until somebody decides what to do.  xfsidgb needs to be built as a
module even when xfs is built in, which breaks a lot of makefile
assumptions.

Date:  Wed Mar 14 17:00:39 PST 2001
Workarea:  sherman.melbourne.sgi.com:/build/kaos/2.4.x-xfs

The following file(s) were checked into:
  bonnie.engr.sgi.com:/isms/slinx/2.4.x-xfs


Modid:  2.4.x-xfs:slinx:90033a
linux/fs/Makefile - 1.29
linux/fs/xfs/Makefile - 1.120


From owner-linux-xfs@oss.sgi.com Thu Mar 15 05:58:55 2001
Received:  by oss.sgi.com id <S553768AbRCON6q>;
	Thu, 15 Mar 2001 05:58:46 -0800
Received: from arthur.aksess.no ([193.217.72.133]:43909 "EHLO arthur.aksess.no")
	by oss.sgi.com with ESMTP id <S553759AbRCON63>;
	Thu, 15 Mar 2001 05:58:29 -0800
Received: (from haralde@localhost)
	by arthur.aksess.no (8.11.0/8.11.0-mogul) id f2FDue004869
	for linux-xfs@oss.sgi.com; Thu, 15 Mar 2001 14:56:40 +0100
Date:   Thu, 15 Mar 2001 14:56:40 +0100
From:   haralde@arthur.aksess.no
Message-Id: <200103151356.f2FDue004869@arthur.aksess.no>
To:     linux-xfs@oss.sgi.com
Subject: oops in xfs-2.4.2 cvs tree from march 15th
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing

Hi
I have this server running current cvs (updated daily) that
keeps crashing with the same errors approx once every day.
(Here are two of the oops'es from /var/log/messages).
It always happens during the slocate cronjob (but not always).
The last time I was able to force it by first running xfs_fsr
followed by slocate.

The server is running a rh7.0 system with only xfs partitions
compiled with egcs 1.1.2 (kgcc).
It's an amd athlon 800mhz machine with 384mb ram and both
ide/scsi disks.

Let me know if you need more info.

regards, Harald


OOPS1:
Mar 13 04:04:04 arthur kernel: kernel BUG at dcache.c:349!
Mar 13 04:04:04 arthur kernel: invalid operand: 0000
Mar 13 04:04:04 arthur kernel: CPU:    0
Mar 13 04:04:04 arthur kernel: EIP:    0010:[prune_dcache+113/336]
Mar 13 04:04:04 arthur kernel: EFLAGS: 00010286
Mar 13 04:04:04 arthur kernel: eax: 0000001c   ebx: cfdd18e0   ecx: 00000000   e
dx: ffffffff
Mar 13 04:04:04 arthur kernel: esi: cfdd18c0   edi: d0a957c0   ebp: c1679fa8   e
sp: c1679f8c
Mar 13 04:04:04 arthur kernel: ds: 0018   es: 0018   ss: 0018
Mar 13 04:04:04 arthur kernel: Process kswapd (pid: 3, stackpage=c1679000)
Mar 13 04:04:04 arthur kernel: Stack: c01d26ae c01d274e 0000015d 00010f00 000000
01 00000004 d0a957c0 c1679fb4 
Mar 13 04:04:04 arthur kernel:        c01423c2 00007e3f c1679fd0 c012a122 000000
06 00000004 00010f00 c01cefd1 
Mar 13 04:04:04 arthur kernel:        c1678239 c1679fec c012a1be 00000004 000000
00 d7fe3f9c c0105000 0063b7e3 
Mar 13 04:04:04 arthur kernel: Call Trace: [shrink_dcache_memory+34/64] [do_try_
to_free_pages+98/144] [kswapd+110/256] [empty_bad_page+0/4096] [kernel_thread+35
/64] 
Mar 13 04:04:04 arthur kernel: 
Mar 13 04:04:04 arthur kernel: Code: 0f 0b 83 c4 0c 8d 46 18 8b 53 f8 8b 48 04 8
9 4a 04 89 11 89 
Mar 13 04:04:05 arthur kernel: kernel BUG at exit.c:458!
Mar 13 04:04:05 arthur kernel: invalid operand: 0000
Mar 13 04:04:05 arthur kernel: CPU:    0
Mar 13 04:04:05 arthur kernel: EIP:    0010:[do_exit+551/576]
Mar 13 04:04:05 arthur kernel: EFLAGS: 00010286
Mar 13 04:04:05 arthur kernel: eax: 0000001a   ebx: c025a600   ecx: 00000000   e
dx: fffffffe
Mar 13 04:04:05 arthur kernel: esi: c1678000   edi: 0000000b   ebp: c1679e94   e
sp: c1679e7c
Mar 13 04:04:05 arthur kernel: ds: 0018   es: 0018   ss: 0018
Mar 13 04:04:05 arthur kernel: Process kswapd (pid: 3, stackpage=c1679000)
Mar 13 04:04:05 arthur kernel: Stack: c01cbc6e c01cbda5 000001ca 00000246 000000
00 00000000 c1679eac c010981a 
Mar 13 04:04:05 arthur kernel:        0000000b c1679f58 00000000 c01099c0 c1679f
48 c0109a3c c01c67fe c1679f58 
Mar 13 04:04:05 arthur kernel:        00000000 c1678000 00000000 00000004 000000
00 00030002 c0142041 c48531e4 

OOPS2:
Mar 15 14:13:36 arthur kernel: kernel BUG at dcache.c:349!
Mar 15 14:13:36 arthur kernel: invalid operand: 0000
Mar 15 14:13:36 arthur kernel: CPU:    0
Mar 15 14:13:36 arthur kernel: EIP:    0010:[prune_dcache+109/332]
Mar 15 14:13:36 arthur kernel: EFLAGS: 00010286
Mar 15 14:13:36 arthur kernel: eax: 0000001c   ebx: cadee600   ecx: 00000000   edx: 00000000
Mar 15 14:13:36 arthur kernel: esi: cadee5e0   edi: c473e2c0   ebp: c167bfa8   esp: c167bf8c
Mar 15 14:13:36 arthur kernel: ds: 0018   es: 0018   ss: 0018
Mar 15 14:13:36 arthur kernel: Process kswapd (pid: 3, stackpage=c167b000)
Mar 15 14:13:36 arthur kernel: Stack: c01cd9ce c01cda6e 0000015d 00010f00 00000182 00000004 c473e2c0 c167bfb4 
Mar 15 14:13:36 arthur kernel:        c0140d02 0000350c c167bfd0 c0129242 00000006 00000004 00010f00 c01ca2f1 
Mar 15 14:13:36 arthur kernel:        c167a239 c167bfec c01292d2 00000004 00000000 d7fe3f9c c0105000 00192d04 
Mar 15 14:13:36 arthur kernel: Call Trace: [shrink_dcache_memory+34/52] [do_try_to_free_pages+98/132] [kswapd+110/248] [empty_bad_page+0/4096] [kernel_thread+35/56] 
Mar 15 14:13:36 arthur kernel: 
Mar 15 14:13:36 arthur kernel: Code: 0f 0b 83 c4 0c 8d 46 18 8b 53 f8 8b 48 04 89 4a 04 89 11 89 
Mar 15 14:13:36 arthur kernel: kernel BUG at exit.c:458!
Mar 15 14:13:36 arthur kernel: invalid operand: 0000
Mar 15 14:13:36 arthur kernel: CPU:    0
Mar 15 14:13:36 arthur kernel: EIP:    0010:[do_exit+539/552]
Mar 15 14:13:36 arthur kernel: EFLAGS: 00010286
Mar 15 14:13:36 arthur kernel: eax: 0000001a   ebx: c0255640   ecx: 00000001   edx: 00000001
Mar 15 14:13:36 arthur kernel: esi: c167a000   edi: 0000000b   ebp: c167be94   esp: c167be7c
Mar 15 14:13:36 arthur kernel: ds: 0018   es: 0018   ss: 0018
Mar 15 14:13:36 arthur kernel: Process kswapd (pid: 3, stackpage=c167b000)
Mar 15 14:13:36 arthur kernel: Stack: c01c6f8e c01c70c5 000001ca 00000246 00000000 00000000 c167beac c01096cd 
Mar 15 14:13:36 arthur kernel:        0000000b c167bf58 00000000 c010986c c167bf48 c01098e1 c01c1abe c167bf58 
Mar 15 14:13:36 arthur kernel:        00000000 c167a000 00000000 00000004 00000000 00030002 c0140991 d7407000 
Mar 15 14:13:36 arthur kernel: Call Trace: [die+249/252] [do_invalid_op+0/132] [do_invalid_op+117/132] [prune_dcache+109/332] [3c59x:__insmod_3c59x_O/lib/modules/2.4.2-XFS/kernel/drivers/net/3+-278666/96] [hide_cursor+129/140] [vt_console_print+724/744] 
Mar 15 14:13:36 arthur kernel:        [error_code+52/64] [prune_dcache+109/332] [shrink_dcache_memory+34/52] [do_try_to_free_pages+98/132] [kswapd+110/248] [empty_bad_page+0/4096] [kernel_thread+35/56] 
Mar 15 14:13:36 arthur kernel: 
Mar 15 14:13:36 arthur kernel: Code: 0f 0b 83 c4 0c e9 37 fe ff ff 8d 76 00 55 89 e5 8b 4d 08 85 
Mar 15 14:13:36 arthur kernel: kernel BUG at exit.c:458!
Mar 15 14:13:36 arthur kernel: invalid operand: 0000
Mar 15 14:13:36 arthur kernel: CPU:    0
Mar 15 14:13:36 arthur kernel: EIP:    0010:[do_exit+539/552]
Mar 15 14:13:36 arthur kernel: EFLAGS: 00010286
Mar 15 14:13:36 arthur kernel: eax: 0000001a   ebx: 00000000   ecx: 00000001   edx: 00000001
Mar 15 14:13:36 arthur kernel: esi: c167a000   edi: 0000000b   ebp: c167bd84   esp: c167bd6c
Mar 15 14:13:36 arthur kernel: ds: 0018   es: 0018   ss: 0018
Mar 15 14:13:36 arthur kernel: Process kswapd (pid: 3, stackpage=c167b000)
Mar 15 14:13:36 arthur kernel: Stack: c01c6f8e c01c70c5 000001ca 00000246 00000000 00000000 c167bd9c c01096cd 
Mar 15 14:13:36 arthur kernel:        0000000b c167be48 00000000 c010986c c167be38 c01098e1 c01c1abe c167be48 
Mar 15 14:13:36 arthur kernel:        00000000 c167a000 00000000 00000004 00000000 00030002 c0115c37 00000000 
Mar 15 14:13:36 arthur kernel: Call Trace: [die+249/252] [do_invalid_op+0/132] [do_invalid_op+117/132] [do_exit+539/552] [lf+57/104] [vt_console_print+724/744] [error_code+52/64] 
Mar 15 14:13:36 arthur kernel:        [do_exit+539/552] [die+249/252] [do_invalid_op+0/132] [do_invalid_op+117/132] [prune_dcache+109/332] [3c59x:__insmod_3c59x_O/lib/modules/2.4.2-XFS/kernel/drivers/net/3+-278666/96] [hide_cursor+129/140] [vt_console_print+724/744] 
Mar 15 14:13:36 arthur kernel:        [error_code+52/64] [prune_dcache+109/332] [shrink_dcache_memory+34/52] [do_try_to_free_pages+98/132] [kswapd+110/248] [empty_bad_page+0/4096] [kernel_thread+35/56] 
Mar 15 14:13:36 arthur kernel: 
Mar 15 14:13:36 arthur kernel: Code: 0f 0b 83 c4 0c e9 37 fe ff ff 8d 76 00 55 89 e5 8b 4d 08 85 
Mar 15 14:13:36 arthur kernel: kernel BUG at exit.c:458!
Mar 15 14:13:36 arthur kernel: invalid operand: 0000
Mar 15 14:13:36 arthur kernel: CPU:    0
Mar 15 14:13:36 arthur kernel: EIP:    0010:[do_exit+539/552]
Mar 15 14:13:36 arthur kernel: EFLAGS: 00010286
Mar 15 14:13:36 arthur kernel: eax: 0000001a   ebx: 00000000   ecx: 00000000   edx: fffffffb
Mar 15 14:13:36 arthur kernel: esi: c167a000   edi: 0000000b   ebp: c167bc74   esp: c167bc5c
Mar 15 14:13:36 arthur kernel: ds: 0018   es: 0018   ss: 0018
Mar 15 14:13:36 arthur kernel: Process kswapd (pid: 3, stackpage=c167b000)

From owner-linux-xfs@oss.sgi.com Thu Mar 15 07:27:05 2001
Received:  by oss.sgi.com id <S553773AbRCOP0q>;
	Thu, 15 Mar 2001 07:26:46 -0800
Received: from heimdal.iee.lu ([194.154.201.2]:35163 "EHLO heimdal.iee.lu")
	by oss.sgi.com with ESMTP id <S553767AbRCOP0P>;
	Thu, 15 Mar 2001 07:26:15 -0800
Received: from picard.findel.ieeintern (picard.findel.ieeintern [10.1.0.3])
	by heimdal.iee.lu (8.9.3/8.9.3/SuSE Linux 8.9.3-0.1) with ESMTP id QAA12422
	for <linux-xfs@oss.sgi.com>; Thu, 15 Mar 2001 16:25:59 +0100
Received: from iee.lu (rkn.findel.ieeintern [10.1.4.2])
	by picard.findel.ieeintern (8.11.0/8.10.2/SuSE Linux 8.10.0-0.3) with ESMTP id f2FFPxa28966
	for <linux-xfs@oss.sgi.com>; Thu, 15 Mar 2001 16:25:59 +0100
X-Authentication-Warning: picard.findel.ieeintern: Host rkn.findel.ieeintern [10.1.4.2] claimed to be iee.lu
Message-ID: <3AB0EAB0.6E92108D@iee.lu>
Date:   Thu, 15 Mar 2001 17:15:44 +0100
From:   Reinhard Knecht <rkn@iee.lu>
X-Mailer: Mozilla 4.76 [en] (X11; U; Linux 2.4.0 i686)
X-Accept-Language: en
MIME-Version: 1.0
To:     linux-xfs@oss.sgi.com
Subject: partition > 2 gigabyte
Content-Type: text/html; charset=us-ascii
Content-Transfer-Encoding: 7bit
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing

<!doctype html public "-//w3c//dtd html 4.0 transitional//en">
<html>
hello,
<p>i've downloaded the linux-2.4-xfs and the linux-2.4-xfs-beta from the
cvs some day's ago.
<br>i've got them compiled with gcc-2.95.2 but when i try to create a filesystem
on partitions larger than 2 gig, i get the following:
<p>belana:~ # mkfs.xfs /dev/hdb2
<br>/dev/hdb2: Invalid argument
<br>can't get size of data subvolume
<br>Usage: mkfs.xfs
<br>/* blocksize */&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; [-b
log=n|size=num]
<br>/* data subvol */&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; [-d agcount=n,agsize=n,file,name=xxx,size=num,
<br>&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;
sunit=value,swidth=value,unwritten=0|1]
<br>/* inode size */&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; [-i log=n|perblock=n|size=num,maxpct=n]
<br>/* log subvol */&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; [-l agnum=n,internal,size=num,logdev=xxx]
<br>/* naming */&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;
[-n log=n|size=num|version=n]
<br>/* prototype file */&nbsp;&nbsp;&nbsp; [-p fname]
<br>/* quiet */&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;
[-q]
<br>/* version */&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;
[-V]
<br>/* realtime subvol */&nbsp;&nbsp; [-r extsize=num,size=num,rtdev=xxx]
<br>&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;
devicename
<br>devicename is required unless -d name=xxx is given
<br>internal 1000 block log is default unless overridden or using a volumemanager
with log
<br>num is xxx (bytes), or xxxb (blocks), or xxxk (xxx KB), or xxxm (xxx
MB)
<br>value is xxx (512 blocks)
<p>for the entire device (/dev/hdb) it works.
<br>with gcc-2.91.66 (egcs 1.12) i can't compile the kernel at all.
<p>any suggestions?
<pre>--&nbsp;
Reinhard Knecht&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; Reinhard.Knecht@iee.lu
International Electronics &amp; Engineering&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; Luxembourg
Tel: (+352) 42 47 37-332&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; Fax: (+352) 42 47 37-200</pre>
&nbsp;</html>


From owner-linux-xfs@oss.sgi.com Thu Mar 15 07:47:35 2001
Received:  by oss.sgi.com id <S553773AbRCOPr0>;
	Thu, 15 Mar 2001 07:47:26 -0800
Received: from pneumatic-tube.sgi.com ([204.94.214.22]:40985 "EHLO
        pneumatic-tube.sgi.com") by oss.sgi.com with ESMTP
	id <S553767AbRCOPrI>; Thu, 15 Mar 2001 07:47:08 -0800
Received: from zeus-fddi.americas.sgi.com (zeus-fddi.americas.sgi.com [128.162.8.103]) by pneumatic-tube.sgi.com (980327.SGI.8.8.8-aspam/980310.SGI-aspam) via ESMTP id HAA06764
	for <linux-xfs@oss.sgi.com>; Thu, 15 Mar 2001 07:56:59 -0800 (PST)
	mail_from (lord@sgi.com)
Received: from daisy-e185.americas.sgi.com (daisy-e194.americas.sgi.com [128.162.194.214]) by zeus-fddi.americas.sgi.com (8.9.3/americas-smart-nospam1.1) with ESMTP id JAA1147331; Thu, 15 Mar 2001 09:45:45 -0600 (CST)
Received: from jen.americas.sgi.com (IDENT:root@jen.americas.sgi.com [128.162.184.86]) by daisy-e185.americas.sgi.com (SGI-8.9.3/SGI-server-1.7) with ESMTP id JAA75305; Thu, 15 Mar 2001 09:45:45 -0600 (CST)
Received: from jen.americas.sgi.com by jen.americas.sgi.com (8.11.0/SGI-client-1.7) via ESMTP id f2FFkEc19676; Thu, 15 Mar 2001 09:46:14 -0600
Message-Id: <200103151546.f2FFkEc19676@jen.americas.sgi.com>
X-Mailer: exmh version 2.2 06/23/2000 with nmh-1.0.4
To:     Reinhard Knecht <rkn@iee.lu>
cc:     linux-xfs@oss.sgi.com
Subject: Re: partition > 2 gigabyte 
In-Reply-To: Message from Reinhard Knecht <rkn@iee.lu> 
   of "Thu, 15 Mar 2001 17:15:44 +0100." <3AB0EAB0.6E92108D@iee.lu> 
Content-Transfer-Encoding: 8bit
Date:   Thu, 15 Mar 2001 09:46:14 -0600
From:   Steve Lord <lord@sgi.com>
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing


You mention downloading the beta and the cvs tree, are you running commands
from the beta with the kernel from the cvs tree?

There will be problems if you are doing this, although your mkfs output
does not go with that. Can you check the /dev entry for hdb2, it should look
like this:

brw-rw----    1 root     disk       3,  66 Aug 24  2000 /dev/hdb2

Also, if you do need to rebuild user space commands from the cvs tree
it is fairly easy. First you need to install the e2fsprogs-devel code,
this is fairly easy to locate on the net if you do not have it (go
to freshmeat and do a search).

cd into the cmd/xfsprogs directory and type

./Makepkgs

this will build the rpm files for the commands, you can just do an rpm -U
with these packages. Even though the name has changed from the beta it
will still update on top of the beta rpms.

Steve


>> hello, 
>> i've downloaded the linux-2.4-xfs and the linux-2.4-xfs-beta from the cvs some day's ago. 
>> i've got them compiled with gcc-2.95.2 but when i try to create a filesystem on partitions larger than 2 gig, i get the following: 
>> belana:~ # mkfs.xfs /dev/hdb2 
>> /dev/hdb2: Invalid argument 
>> can't get size of data subvolume 
>> Usage: mkfs.xfs 
>> /* blocksize */         [-b log=n|size=num] 
>> /* data subvol */       [-d agcount=n,agsize=n,file,name=xxx,size=num, 
>>                             sunit=value,swidth=value,unwritten=0|1] 
>> /* inode size */        [-i log=n|perblock=n|size=num,maxpct=n] 
>> /* log subvol */        [-l agnum=n,internal,size=num,logdev=xxx] 
>> /* naming */            [-n log=n|size=num|version=n] 
>> /* prototype file */    [-p fname] 
>> /* quiet */             [-q] 
>> /* version */           [-V] 
>> /* realtime subvol */   [-r extsize=num,size=num,rtdev=xxx] 
>>                         devicename 
>> devicename is required unless -d name=xxx is given 
>> internal 1000 block log is default unless overridden or using a volumemanager with log 
>> num is xxx (bytes), or xxxb (blocks), or xxxk (xxx KB), or xxxm (xxx MB) 
>> value is xxx (512 blocks) 
>> for the entire device (/dev/hdb) it works. 
>> with gcc-2.91.66 (egcs 1.12) i can't compile the kernel at all. 
>> any suggestions? 




From owner-linux-xfs@oss.sgi.com Thu Mar 15 07:48:36 2001
Received:  by oss.sgi.com id <S553773AbRCOPsR>;
	Thu, 15 Mar 2001 07:48:17 -0800
Received: from deliverator.sgi.com ([204.94.214.10]:2318 "EHLO
        deliverator.sgi.com") by oss.sgi.com with ESMTP id <S553767AbRCOPsG>;
	Thu, 15 Mar 2001 07:48:06 -0800
Received: from info.engr.sgi.com (info.engr.sgi.com [192.26.80.216]) by deliverator.sgi.com (980309.SGI.8.8.8-aspam-6.2/980310.SGI-aspam) via ESMTP id HAA02679; Thu, 15 Mar 2001 07:46:55 -0800 (PST)
	mail_from (nobody@info.engr.sgi.com)
Received: (from nobody@localhost)
	by info.engr.sgi.com (SGI-8.9.3/8.9.3) id HAA58172;
	Thu, 15 Mar 2001 07:48:04 -0800 (PST)
Date:   Thu, 15 Mar 2001 07:48:04 -0800 (PST)
Message-Id: <200103151548.HAA58172@info.engr.sgi.com>
X-Pv-Incident: 817944 
webPV:  fsgi158.americas.sgi.com 
webExec: webpvsubmit,PvProjectIncident 
Reply-To: sgi.bugs.xfs@fido.engr.sgi.com
From:   pv@relay.sgi.com (tbd@sgi.com)
Subject: BUG 817944 - NFS V3 from IRIX or SUN client hangs XFS file system
To:     nb@sgi.com
Cc:     linux-xfs@oss.sgi.com
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing

View Incident: http://co-op.engr.sgi.com/BugWorks/code/bwxquery.cgi?search=Search&wlong=1&view_type=Bug&wi=817944

Submitter : tbd                       Submitter Domain : sgi.com            
Assigned Engineer : nb                Assigned Domain : sgi.com             
Assigned Group : xfs-linux            Category : software                   
Customer Reported : F                 Priority : 1                          
Project : xfs-linux                   Status : open                         
Description :
Exporting an XFS file system on a linux server and using it
NFS V3 server from an Irix or Sun client hangs the XFS directory.

To recreate:
On IRIX machine:
  mkdir /tmp/permit
  mount permit:/mnt_xfs /tmp/permit
  cd /home/daisy08/nettest/tests/nfs/regression
  cp -r sgi.new /tmp/permit/sgi.new
  cd /tmp/permit/sgi.new/special
  ls  ------ this hangs

ls locally on the linux server hangs as well and nfsd seems to hang.

Snoop shows that readdirplus calls are being retransmitted to the
linux server.

I did NOT see this problem under the following conditions:
  linux client (SNIA)
  ext2 file system on the server.
  version 2 NFS

From owner-linux-xfs@oss.sgi.com Thu Mar 15 08:19:06 2001
Received:  by oss.sgi.com id <S553773AbRCOQS4>;
	Thu, 15 Mar 2001 08:18:56 -0800
Received: from sgi.SGI.COM ([192.48.153.1]:34871 "EHLO sgi.com")
	by oss.sgi.com with ESMTP id <S553767AbRCOQSi>;
	Thu, 15 Mar 2001 08:18:38 -0800
Received: from info.engr.sgi.com ([192.26.80.216]) 
	by sgi.com (980327.SGI.8.8.8-aspam/980304.SGI-aspam:
       SGI does not authorize the use of its proprietary
       systems or networks for unsolicited or bulk email
       from the Internet.) 
	via ESMTP id IAA05323; Thu, 15 Mar 2001 08:17:25 -0800 (PST)
	mail_from (nobody@info.engr.sgi.com)
Received: (from nobody@localhost)
	by info.engr.sgi.com (SGI-8.9.3/8.9.3) id IAA79460;
	Thu, 15 Mar 2001 08:17:21 -0800 (PST)
Date:   Thu, 15 Mar 2001 08:17:21 -0800 (PST)
Message-Id: <200103151617.IAA79460@info.engr.sgi.com>
X-Pv-Incident: 817944 
webPV:  jen.americas.sgi.com 
webExec: webpvupdate,pvincident 
Reply-To: sgi.bugs.xfs@fido.engr.sgi.com
From:   pv@odin.corp.sgi.com (lord@sgi.com)
Subject: ADD 817944 - NFS V3 from IRIX or SUN client hangs XFS file system
To:     nb@sgi.com
Cc:     linux-xfs@oss.sgi.com
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing

View Incident: http://co-op.engr.sgi.com/BugWorks/code/bwxquery.cgi?search=Search&wlong=1&view_type=Bug&wi=817944

 Status : open                         Priority : 1                         
 Assigned Engineer : nb                Submitter : tbd                      
*Modified User : lord                 *Modified User Domain : sgi.com       
*Description :
Exporting an XFS file system on a linux server and using it
NFS V3 server from an Irix or Sun client hangs the XFS directory.

To recreate:
On IRIX machine:
  mkdir /tmp/permit
  mount permit:/mnt_xfs /tmp/permit
  cd /home/daisy08/nettest/tests/nfs/regression
  cp -r sgi.new /tmp/permit/sgi.new
  cd /tmp/permit/sgi.new/special

.....


==========================
ADDITIONAL INFORMATION (ADD)
From: lord@sgi.com (BugWorks)
Date: Mar 15 2001 08:17:21AM
==========================

Hmmm,,mount -o vers=3 lord:/xfs /tmp/xfs


 uname -a
IRIX64 cxfs1 6.5-overby-grove2-SN0CELLARRAY 03141602 IP27

# mkdir /tmp/xfs# cd /home/daisy08/nettest/tests/nfs/regression
# cp -r sgi.new /tmp/xfs/sgi.new
# cd /tmp/xfs/sgi.new
# ls
Makefile     Testitems    general      runtests     tests.h      unixdos.h
README       basic        getopt.c     server       tests.init
READWIN.txt  domount.c    lock         special      tools
#

But It hangs in another directory doing ls -lR on the whole
tree. As Russell reported, it appears to hang on a pagebuf lock:

The problem here appears to be the way NFS uses the filldir function,
we call it under all sorts of locks, and it calls back
into xfs again. Almost certainly we are deadlocked on ourself.

Stack traceback for pid 1097
    EBP       EIP         Function(args)
0xc5f7b6c8 0xc01161fa schedule+0x41e (0xc6246e40)
                               kernel .text 0xc0100000 0xc0115ddc 0xc0116410
0xc5f7b6e0 0xc0107a5b __down+0x73
                               kernel .text 0xc0100000 0xc01079e8 0xc0107ab8
           0xc0107c08 __down_failed+0x8 (0xc0196ae8, 0xc6246e40, 0xc61ae820, 0xc13b2fc0, 0xc5f7b7f8)
                               kernel .text 0xc0100000 0xc0107c00 0xc0107c0c
           0xc02acca6 stext_lock+0x35da
                               kernel .text.lock 0xc02a96cc 0xc02a96cc 0xc02b0200
           0xc01969c9 _pagebuf_grab_lock+0x11 (0xc6246e40)
                               kernel .text 0xc0100000 0xc01969b8 0xc01969cc
           0xc0196ae8 _pagebuf_find_lockable_buffer+0x11c (0xc61ae820, 0xc13b2fc0, 0x3be2f000, 0x0, 0x1000)
                               kernel .text 0xc0100000 0xc01969cc 0xc0196bd0
           0xc0196bff _pagebuf_get_lockable_buffer+0x2f (0xc61ae820, 0xc13b2fc0, 0x3be2f000, 0x0, 0x1000)
                               kernel .text 0xc0100000 0xc0196bd0 0xc0196cc0
           0xc0192b02 pagebuf_get+0x126 (0xc61ae820, 0x3be2f000, 0x0, 0x1000, 0x2205)
                               kernel .text 0xc0100000 0xc01929dc 0xc0192c04
           0xc01e5396 xfs_trans_read_buf+0x5e (0xc70eec00, 0x0, 0xc70eed64, 0x1df178, 0x0)
                               kernel .text 0xc0100000 0xc01e5338 0xc01e5674
           0xc01b9bde xfs_da_do_buf+0x612 (0x0, 0xc6046900, 0x0, 0xc5f7b944)
                               kernel .text 0xc0100000 0xc01b95cc 0xc01b9e64
[0]more> 
           0xc01b9ec5 xfs_da_read_buf+0x2d (0x0, 0xc5f7b9d8, 0x0, 0xc01bd5bd, 0x0)
                               kernel .text 0xc0100000 0xc01b9e98 0xc01b9ecc
           0xc01bd5bd xfs_dir2_block_lookup_int+0x35 (0xc5f7b9dc, 0xc5f7b9b8, 0xc5f7b9bc)
                               kernel .text 0xc0100000 0xc01bd588 0xc01bd74c
           0xc01bd4c7 xfs_dir2_block_lookup+0x1b (0xc5f7b9dc)
                               kernel .text 0xc0100000 0xc01bd4ac 0xc01bd588
           0xc01bbcce xfs_dir2_lookup+0xde (0x0, 0xc6046900, 0xc69c0a00, 0x5, 0xc5f7badc)
                               kernel .text 0xc0100000 0xc01bbbf0 0xc01bbd30
           0xc01e633a xfs_dir_lookup_int+0x112 (0x0, 0xc6046918, 0x5, 0xc69c0a00, 0xc5f7bb30)
                               kernel .text 0xc0100000 0xc01e6228 0xc01e65d0
           0xc01eb2ba xfs_lookup+0xb2 (0xc6046918, 0xc69c0a00, 0xc5f7bb2c, 0xc5f7bb30, 0x0)
                               kernel .text 0xc0100000 0xc01eb208 0xc01eb358
           0xc01f47f4 linvfs_lookup+0x84 (0xc64629e0, 0xc69c09a0)
                               kernel .text 0xc0100000 0xc01f4770 0xc01f485c
           0xc014246d lookup_hash+0x9d (0xc5f7bb68, 0xc7b968e0)
                               kernel .text 0xc0100000 0xc01423d0 0xc01424c0
           0xc014250f lookup_one+0x4f (0xc257c039, 0xc7b968e0)
                               kernel .text 0xc0100000 0xc01424c0 0xc0142520
           0xc016849e encode_entry+0x1e6 (0xc5f7be9c, 0xc257c039, 0x3, 0x6, 0x600100)
                               kernel .text 0xc0100000 0xc01682b8 0xc016859c
           0xc01685f4 nfs3svc_encode_entry_plus+0x28 (0xc5f7be9c, 0xc257c039, 0x3, 0x6, 0x600100)
                               kernel .text 0xc0100000 0xc01685cc 0xc01685fc
           0xc01bc784 xfs_dir2_put_dirent64_uio+0x28 (0xc5f7bcd8)
[0]more> 
                               kernel .text 0xc0100000 0xc01bc75c 0xc01bc7a4
           0xc01bd33d xfs_dir2_block_getdents+0x1cd (0x0, 0xc6046900, 0xc5f7be40, 0xc5f7bd78, 0x0)
                               kernel .text 0xc0100000 0xc01bd170 0xc01bd424
           0xc01bbec3 xfs_dir2_getdents+0x6b (0x0, 0xc6046900, 0xc5f7be40, 0xc5f7bd78)
                               kernel .text 0xc0100000 0xc01bbe58 0xc01bbee0
           0xc01ee30a xfs_readdir+0xaa (0xc6046918, 0xc5f7be40, 0xc5f7bd84, 0xc5f7bd78, 0xc64629e0)
                               kernel .text 0xc0100000 0xc01ee260 0xc01ee340
           0xc01f2889 linvfs_readdir+0xc1 (0xc5f7beb4, 0xc5f7be9c, 0xc01685cc)
                               kernel .text 0xc0100000 0xc01f27c8 0xc01f28b0
           0xc0161053 nfsd_readdir+0xeb (0xc5f7de00, 0xc5f7da04, 0x0, 0x0, 0xc01685cc)
                               kernel .text 0xc0100000 0xc0160f68 0xc016116c
           0xc016596b nfsd3_proc_readdirplus+0x123 (0xc5f7de00, 0xc5f7dc00, 0xc5f7da00)
                               kernel .text 0xc0100000 0xc0165848 0xc01659a8
           0xc015c1a3 nfsd_dispatch+0xcb (0xc5f7de00, 0xc5fe0014)
                               kernel .text 0xc0100000 0xc015c0d8 0xc015c240
           0xc029eec8 svc_process+0x2ac (0xc688ee20, 0xc5f7de00)
                               kernel .text 0xc0100000 0xc029ec1c 0xc029f160
           0xc015bf4a nfsd+0x1ca
                               kernel .text 0xc0100000 0xc015bd80 0xc015c0d8
           0xc010750f kernel_thread+0x23
                               kernel .text 0xc0100000 0xc01074ec 0xc010751c





# mount -o vers=3 lord:/xfs /tmp/xfs

From owner-linux-xfs@oss.sgi.com Thu Mar 15 08:28:55 2001
Received:  by oss.sgi.com id <S553778AbRCOQ2q>;
	Thu, 15 Mar 2001 08:28:46 -0800
Received: from pneumatic-tube.sgi.com ([204.94.214.22]:42529 "EHLO
        pneumatic-tube.sgi.com") by oss.sgi.com with ESMTP
	id <S553773AbRCOQ2d>; Thu, 15 Mar 2001 08:28:33 -0800
Received: from info.engr.sgi.com (info.engr.sgi.com [192.26.80.216]) by pneumatic-tube.sgi.com (980327.SGI.8.8.8-aspam/980310.SGI-aspam) via ESMTP id IAA01119; Thu, 15 Mar 2001 08:38:25 -0800 (PST)
	mail_from (nobody@info.engr.sgi.com)
Received: (from nobody@localhost)
	by info.engr.sgi.com (SGI-8.9.3/8.9.3) id IAA77375;
	Thu, 15 Mar 2001 08:28:32 -0800 (PST)
Date:   Thu, 15 Mar 2001 08:28:32 -0800 (PST)
Message-Id: <200103151628.IAA77375@info.engr.sgi.com>
X-Pv-Incident: 817944 
webPV:  jen.americas.sgi.com 
webExec: webpvupdate,pvincident 
Reply-To: sgi.bugs.xfs@fido.engr.sgi.com
From:   pv@relay.sgi.com (lord@sgi.com)
Subject: ADD 817944 - NFS V3 from IRIX or SUN client hangs XFS file system
To:     nb@sgi.com
Cc:     linux-xfs@oss.sgi.com
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing

View Incident: http://co-op.engr.sgi.com/BugWorks/code/bwxquery.cgi?search=Search&wlong=1&view_type=Bug&wi=817944

 Status : open                         Priority : 1                         
 Assigned Engineer : nb                Submitter : tbd                      
*Modified User : lord                 *Modified User Domain : sgi.com       
*Description :
Exporting an XFS file system on a linux server and using it
NFS V3 server from an Irix or Sun client hangs the XFS directory.

To recreate:
On IRIX machine:
  mkdir /tmp/permit
  mount permit:/mnt_xfs /tmp/permit
  cd /home/daisy08/nettest/tests/nfs/regression
  cp -r sgi.new /tmp/permit/sgi.new
  cd /tmp/permit/sgi.new/special

.....


==========================
ADDITIONAL INFORMATION (ADD)
From: lord@sgi.com (BugWorks)
Date: Mar 15 2001 08:28:32AM
==========================

My only other comment here is this is Mostek's fault!

We need to rework where we call the filldir function in the
readdir path. This may take a day or so to fix.

From owner-linux-xfs@oss.sgi.com Thu Mar 15 08:34:46 2001
Received:  by oss.sgi.com id <S553778AbRCOQeg>;
	Thu, 15 Mar 2001 08:34:36 -0800
Received: from sgi.SGI.COM ([192.48.153.1]:10813 "EHLO sgi.com")
	by oss.sgi.com with ESMTP id <S553769AbRCOQeZ>;
	Thu, 15 Mar 2001 08:34:25 -0800
Received: from zeus-fddi.americas.sgi.com (128-162-8-103.americas.sgi.com [128.162.8.103]) 
	by sgi.com (980327.SGI.8.8.8-aspam/980304.SGI-aspam:
       SGI does not authorize the use of its proprietary
       systems or networks for unsolicited or bulk email
       from the Internet.) 
	via ESMTP id IAA03287
	for <linux-xfs@oss.sgi.com>; Thu, 15 Mar 2001 08:34:24 -0800 (PST)
	mail_from (lord@sgi.com)
Received: from daisy-e185.americas.sgi.com (daisy-e194.americas.sgi.com [128.162.194.214]) by zeus-fddi.americas.sgi.com (8.9.3/americas-smart-nospam1.1) with ESMTP id KAA1146495; Thu, 15 Mar 2001 10:33:07 -0600 (CST)
Received: from jen.americas.sgi.com (IDENT:root@jen.americas.sgi.com [128.162.184.86]) by daisy-e185.americas.sgi.com (SGI-8.9.3/SGI-server-1.7) with ESMTP id KAA72482; Thu, 15 Mar 2001 10:33:06 -0600 (CST)
Received: from jen.americas.sgi.com by jen.americas.sgi.com (8.11.0/SGI-client-1.7) via ESMTP id f2FGZHX01418; Thu, 15 Mar 2001 10:35:17 -0600
Message-Id: <200103151635.f2FGZHX01418@jen.americas.sgi.com>
X-Mailer: exmh version 2.2 06/23/2000 with nmh-1.0.4
To:     haralde@arthur.aksess.no
cc:     linux-xfs@oss.sgi.com
Subject: Re: oops in xfs-2.4.2 cvs tree from march 15th 
In-Reply-To: Message from haralde@arthur.aksess.no 
   of "Thu, 15 Mar 2001 14:56:40 +0100." <200103151356.f2FDue004869@arthur.aksess.no> 
Date:   Thu, 15 Mar 2001 10:35:16 -0600
From:   Steve Lord <lord@sgi.com>
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing



This one is somewhat scary - the bug call you hit is at the end of
exit, it is never supposed to get there!

I do not really see how this problem can really be XFS related, it
looks more like a scheduler thing. I wonder if there have been any
external reports of similar problems in linux kernel.

Steve

> Hi
> I have this server running current cvs (updated daily) that
> keeps crashing with the same errors approx once every day.
> (Here are two of the oops'es from /var/log/messages).
> It always happens during the slocate cronjob (but not always).
> The last time I was able to force it by first running xfs_fsr
> followed by slocate.
> 
> The server is running a rh7.0 system with only xfs partitions
> compiled with egcs 1.1.2 (kgcc).
> It's an amd athlon 800mhz machine with 384mb ram and both
> ide/scsi disks.
> 
> Let me know if you need more info.
> 
> regards, Harald
> 
> 
> OOPS1:
> Mar 13 04:04:04 arthur kernel: kernel BUG at dcache.c:349!
> Mar 13 04:04:04 arthur kernel: invalid operand: 0000
> Mar 13 04:04:04 arthur kernel: CPU:    0
> Mar 13 04:04:04 arthur kernel: EIP:    0010:[prune_dcache+113/336]
> Mar 13 04:04:04 arthur kernel: EFLAGS: 00010286
> Mar 13 04:04:04 arthur kernel: eax: 0000001c   ebx: cfdd18e0   ecx: 00000000 
>   e
> dx: ffffffff
> Mar 13 04:04:04 arthur kernel: esi: cfdd18c0   edi: d0a957c0   ebp: c1679fa8 
>   e
> sp: c1679f8c
> Mar 13 04:04:04 arthur kernel: ds: 0018   es: 0018   ss: 0018
> Mar 13 04:04:04 arthur kernel: Process kswapd (pid: 3, stackpage=c1679000)
> Mar 13 04:04:04 arthur kernel: Stack: c01d26ae c01d274e 0000015d 00010f00 000
> 000
> 01 00000004 d0a957c0 c1679fb4 
> Mar 13 04:04:04 arthur kernel:        c01423c2 00007e3f c1679fd0 c012a122 000
> 000
> 06 00000004 00010f00 c01cefd1 
> Mar 13 04:04:04 arthur kernel:        c1678239 c1679fec c012a1be 00000004 000
> 000
> 00 d7fe3f9c c0105000 0063b7e3 
> Mar 13 04:04:04 arthur kernel: Call Trace: [shrink_dcache_memory+34/64] [do_t
> ry_
> to_free_pages+98/144] [kswapd+110/256] [empty_bad_page+0/4096] [kernel_thread
> +35
> /64] 
> Mar 13 04:04:04 arthur kernel: 
> Mar 13 04:04:04 arthur kernel: Code: 0f 0b 83 c4 0c 8d 46 18 8b 53 f8 8b 48 0
> 4 8
> 9 4a 04 89 11 89 
> Mar 13 04:04:05 arthur kernel: kernel BUG at exit.c:458!
> Mar 13 04:04:05 arthur kernel: invalid operand: 0000
> Mar 13 04:04:05 arthur kernel: CPU:    0
> Mar 13 04:04:05 arthur kernel: EIP:    0010:[do_exit+551/576]
> Mar 13 04:04:05 arthur kernel: EFLAGS: 00010286
> Mar 13 04:04:05 arthur kernel: eax: 0000001a   ebx: c025a600   ecx: 00000000 
>   e
> dx: fffffffe
> Mar 13 04:04:05 arthur kernel: esi: c1678000   edi: 0000000b   ebp: c1679e94 
>   e
> sp: c1679e7c
> Mar 13 04:04:05 arthur kernel: ds: 0018   es: 0018   ss: 0018
> Mar 13 04:04:05 arthur kernel: Process kswapd (pid: 3, stackpage=c1679000)
> Mar 13 04:04:05 arthur kernel: Stack: c01cbc6e c01cbda5 000001ca 00000246 000
> 000
> 00 00000000 c1679eac c010981a 
> Mar 13 04:04:05 arthur kernel:        0000000b c1679f58 00000000 c01099c0 c16
> 79f
> 48 c0109a3c c01c67fe c1679f58 
> Mar 13 04:04:05 arthur kernel:        00000000 c1678000 00000000 00000004 000
> 000
> 00 00030002 c0142041 c48531e4 
> 
> OOPS2:
> Mar 15 14:13:36 arthur kernel: kernel BUG at dcache.c:349!
> Mar 15 14:13:36 arthur kernel: invalid operand: 0000
> Mar 15 14:13:36 arthur kernel: CPU:    0
> Mar 15 14:13:36 arthur kernel: EIP:    0010:[prune_dcache+109/332]
> Mar 15 14:13:36 arthur kernel: EFLAGS: 00010286
> Mar 15 14:13:36 arthur kernel: eax: 0000001c   ebx: cadee600   ecx: 00000000 
>   edx: 00000000
> Mar 15 14:13:36 arthur kernel: esi: cadee5e0   edi: c473e2c0   ebp: c167bfa8 
>   esp: c167bf8c
> Mar 15 14:13:36 arthur kernel: ds: 0018   es: 0018   ss: 0018
> Mar 15 14:13:36 arthur kernel: Process kswapd (pid: 3, stackpage=c167b000)
> Mar 15 14:13:36 arthur kernel: Stack: c01cd9ce c01cda6e 0000015d 00010f00 000
> 00182 00000004 c473e2c0 c167bfb4 
> Mar 15 14:13:36 arthur kernel:        c0140d02 0000350c c167bfd0 c0129242 000
> 00006 00000004 00010f00 c01ca2f1 
> Mar 15 14:13:36 arthur kernel:        c167a239 c167bfec c01292d2 00000004 000
> 00000 d7fe3f9c c0105000 00192d04 
> Mar 15 14:13:36 arthur kernel: Call Trace: [shrink_dcache_memory+34/52] [do_t
> ry_to_free_pages+98/132] [kswapd+110/248] [empty_bad_page+0/4096] [kernel_thr
> ead+35/56] 
> Mar 15 14:13:36 arthur kernel: 
> Mar 15 14:13:36 arthur kernel: Code: 0f 0b 83 c4 0c 8d 46 18 8b 53 f8 8b 48 0
> 4 89 4a 04 89 11 89 
> Mar 15 14:13:36 arthur kernel: kernel BUG at exit.c:458!
> Mar 15 14:13:36 arthur kernel: invalid operand: 0000
> Mar 15 14:13:36 arthur kernel: CPU:    0
> Mar 15 14:13:36 arthur kernel: EIP:    0010:[do_exit+539/552]
> Mar 15 14:13:36 arthur kernel: EFLAGS: 00010286
> Mar 15 14:13:36 arthur kernel: eax: 0000001a   ebx: c0255640   ecx: 00000001 
>   edx: 00000001
> Mar 15 14:13:36 arthur kernel: esi: c167a000   edi: 0000000b   ebp: c167be94 
>   esp: c167be7c
> Mar 15 14:13:36 arthur kernel: ds: 0018   es: 0018   ss: 0018
> Mar 15 14:13:36 arthur kernel: Process kswapd (pid: 3, stackpage=c167b000)
> Mar 15 14:13:36 arthur kernel: Stack: c01c6f8e c01c70c5 000001ca 00000246 000
> 00000 00000000 c167beac c01096cd 
> Mar 15 14:13:36 arthur kernel:        0000000b c167bf58 00000000 c010986c c16
> 7bf48 c01098e1 c01c1abe c167bf58 
> Mar 15 14:13:36 arthur kernel:        00000000 c167a000 00000000 00000004 000
> 00000 00030002 c0140991 d7407000 
> Mar 15 14:13:36 arthur kernel: Call Trace: [die+249/252] [do_invalid_op+0/132
> ] [do_invalid_op+117/132] [prune_dcache+109/332] [3c59x:__insmod_3c59x_O/lib/
> modules/2.4.2-XFS/kernel/drivers/net/3+-278666/96] [hide_cursor+129/140] [vt_
> console_print+724/744] 
> Mar 15 14:13:36 arthur kernel:        [error_code+52/64] [prune_dcache+109/33
> 2] [shrink_dcache_memory+34/52] [do_try_to_free_pages+98/132] [kswapd+110/248
> ] [empty_bad_page+0/4096] [kernel_thread+35/56] 
> Mar 15 14:13:36 arthur kernel: 
> Mar 15 14:13:36 arthur kernel: Code: 0f 0b 83 c4 0c e9 37 fe ff ff 8d 76 00 5
> 5 89 e5 8b 4d 08 85 
> Mar 15 14:13:36 arthur kernel: kernel BUG at exit.c:458!
> Mar 15 14:13:36 arthur kernel: invalid operand: 0000
> Mar 15 14:13:36 arthur kernel: CPU:    0
> Mar 15 14:13:36 arthur kernel: EIP:    0010:[do_exit+539/552]
> Mar 15 14:13:36 arthur kernel: EFLAGS: 00010286
> Mar 15 14:13:36 arthur kernel: eax: 0000001a   ebx: 00000000   ecx: 00000001 
>   edx: 00000001
> Mar 15 14:13:36 arthur kernel: esi: c167a000   edi: 0000000b   ebp: c167bd84 
>   esp: c167bd6c
> Mar 15 14:13:36 arthur kernel: ds: 0018   es: 0018   ss: 0018
> Mar 15 14:13:36 arthur kernel: Process kswapd (pid: 3, stackpage=c167b000)
> Mar 15 14:13:36 arthur kernel: Stack: c01c6f8e c01c70c5 000001ca 00000246 000
> 00000 00000000 c167bd9c c01096cd 
> Mar 15 14:13:36 arthur kernel:        0000000b c167be48 00000000 c010986c c16
> 7be38 c01098e1 c01c1abe c167be48 
> Mar 15 14:13:36 arthur kernel:        00000000 c167a000 00000000 00000004 000
> 00000 00030002 c0115c37 00000000 
> Mar 15 14:13:36 arthur kernel: Call Trace: [die+249/252] [do_invalid_op+0/132
> ] [do_invalid_op+117/132] [do_exit+539/552] [lf+57/104] [vt_console_print+724
> /744] [error_code+52/64] 
> Mar 15 14:13:36 arthur kernel:        [do_exit+539/552] [die+249/252] [do_inv
> alid_op+0/132] [do_invalid_op+117/132] [prune_dcache+109/332] [3c59x:__insmod
> _3c59x_O/lib/modules/2.4.2-XFS/kernel/drivers/net/3+-278666/96] [hide_cursor+
> 129/140] [vt_console_print+724/744] 
> Mar 15 14:13:36 arthur kernel:        [error_code+52/64] [prune_dcache+109/33
> 2] [shrink_dcache_memory+34/52] [do_try_to_free_pages+98/132] [kswapd+110/248
> ] [empty_bad_page+0/4096] [kernel_thread+35/56] 
> Mar 15 14:13:36 arthur kernel: 
> Mar 15 14:13:36 arthur kernel: Code: 0f 0b 83 c4 0c e9 37 fe ff ff 8d 76 00 5
> 5 89 e5 8b 4d 08 85 
> Mar 15 14:13:36 arthur kernel: kernel BUG at exit.c:458!
> Mar 15 14:13:36 arthur kernel: invalid operand: 0000
> Mar 15 14:13:36 arthur kernel: CPU:    0
> Mar 15 14:13:36 arthur kernel: EIP:    0010:[do_exit+539/552]
> Mar 15 14:13:36 arthur kernel: EFLAGS: 00010286
> Mar 15 14:13:36 arthur kernel: eax: 0000001a   ebx: 00000000   ecx: 00000000 
>   edx: fffffffb
> Mar 15 14:13:36 arthur kernel: esi: c167a000   edi: 0000000b   ebp: c167bc74 
>   esp: c167bc5c
> Mar 15 14:13:36 arthur kernel: ds: 0018   es: 0018   ss: 0018
> Mar 15 14:13:36 arthur kernel: Process kswapd (pid: 3, stackpage=c167b000)



From owner-linux-xfs@oss.sgi.com Thu Mar 15 08:46:55 2001
Received:  by oss.sgi.com id <S553778AbRCOQqq>;
	Thu, 15 Mar 2001 08:46:46 -0800
Received: from heimdal.iee.lu ([194.154.201.2]:30836 "EHLO heimdal.iee.lu")
	by oss.sgi.com with ESMTP id <S553769AbRCOQq3>;
	Thu, 15 Mar 2001 08:46:29 -0800
Received: from picard.findel.ieeintern (picard.findel.ieeintern [10.1.0.3])
	by heimdal.iee.lu (8.9.3/8.9.3/SuSE Linux 8.9.3-0.1) with ESMTP id RAA13008;
	Thu, 15 Mar 2001 17:46:01 +0100
Received: from iee.lu (rkn.findel.ieeintern [10.1.4.2])
	by picard.findel.ieeintern (8.11.0/8.10.2/SuSE Linux 8.10.0-0.3) with ESMTP id f2FGjwa03519;
	Thu, 15 Mar 2001 17:45:58 +0100
X-Authentication-Warning: picard.findel.ieeintern: Host rkn.findel.ieeintern [10.1.4.2] claimed to be iee.lu
Message-ID: <3AB0FD70.4C18B5E2@iee.lu>
Date:   Thu, 15 Mar 2001 18:35:44 +0100
From:   Reinhard Knecht <rkn@iee.lu>
X-Mailer: Mozilla 4.76 [en] (X11; U; Linux 2.4.0 i686)
X-Accept-Language: en
MIME-Version: 1.0
To:     Steve Lord <lord@sgi.com>
CC:     linux-xfs@oss.sgi.com
Subject: Re: partition > 2 gigabyte
References: <200103151546.f2FFkEc19676@jen.americas.sgi.com>
Content-Type: text/html; charset=us-ascii
Content-Transfer-Encoding: 7bit
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk
Return-Path: <owner-linux-xfs@oss.sgi.com>
X-Orcpt: rfc822;linux-xfs-outgoing

<!doctype html public "-//w3c//dtd html 4.0 transitional//en">
<html>
i don't mix up the kernel and the tools but i got it working now.
<br>before i compiled the xfs as module, now i compiled it into the kernel
and it works (gcc 2.91.66).
<br>btw, what's with the acl's?
<br>they are not available in the beta?
<p>thank's anyway...
<br>&nbsp;
<p>Steve Lord wrote:
<blockquote TYPE=CITE>You mention downloading the beta and the cvs tree,
are you running commands
<br>from the beta with the kernel from the cvs tree?
<p>There will be problems if you are doing this, although your mkfs output
<br>does not go with that. Can you check the /dev entry for hdb2, it should
look
<br>like this:
<p>brw-rw----&nbsp;&nbsp;&nbsp; 1 root&nbsp;&nbsp;&nbsp;&nbsp; disk&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;
3,&nbsp; 66 Aug 24&nbsp; 2000 /dev/hdb2
<p>Also, if you do need to rebuild user space commands from the cvs tree
<br>it is fairly easy. First you need to install the e2fsprogs-devel code,
<br>this is fairly easy to locate on the net if you do not have it (go
<br>to freshmeat and do a search).
<p>cd into the cmd/xfsprogs directory and type
<p>./Makepkgs
<p>this will build the rpm files for the commands, you can just do an rpm
-U
<br>with these packages. Even though the name has changed from the beta
it
<br>will still update on top of the beta rpms.
<p>Steve
<p>>> hello,
<br>>> i've downloaded the linux-2.4-xfs and the linux-2.4-xfs-beta from
the cvs some day's ago.
<br>>> i've got them compiled with gcc-2.95.2 but when i try to create
a filesystem on partitions larger than 2 gig, i get the following:
<br>>> belana:~ # mkfs.xfs /dev/hdb2
<br>>> /dev/hdb2: Invalid argument
<br>>> can't get size of data subvolume
<br>>> Usage: mkfs.xfs
<br>>> /* blocksize */&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;
[-b log=n|size=num]
<br>>> /* data subvol */&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; [-d agcount=n,agsize=n,file,name=xxx,size=num,
<br>>>&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;
sunit=value,swidth=value,unwritten=0|1]
<br>>> /* inode size */&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; [-i log=n|perblock=n|size=num,maxpct=n]
<br>>> /* log subvol */&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; [-l agnum=n,internal,size=num,logdev=xxx]
<br>>> /* naming */&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;
[-n log=n|size=num|version=n]
<br>>> /* prototype file */&nbsp;&nbsp;&nbsp; [-p fname]
<br>>> /* quiet */&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;
[-q]
<br>>> /* version */&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;
[-V]
<br>>> /* realtime subvol */&nbsp;&nbsp; [-r extsize=num,size=num,rtdev=xxx]
<br>>>&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;
devicename
<br>>> devicename is required unless -d name=xxx is given
<br>>> internal 1000 block log is default unless overridden or using a
volumemanager with log
<br>>> num is xxx (bytes), or xxxb (blocks), or xxxk (xxx KB), or xxxm
(xxx MB)
<br>>> value is xxx (512 blocks)
<br>>> for the entire device (/dev/hdb) it works.
<br>>> with gcc-2.91.66 (egcs 1.12) i can't compile the kernel at all.
<br>>> any suggestions?</blockquote>

<pre>--&nbsp;
Reinhard Knecht&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; Reinhard.Knecht@iee.lu
International Electronics &amp; Engineering&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; Luxembourg
Tel: (+352) 42 47 37-332&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; Fax: (+352) 42 47 37-200</pre>
&nbsp;</html>


From owner-linux-xfs@oss.sgi.com Thu Mar 15 16:02:44 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2G02iX21270
	for linux-xfs-outgoing; Thu, 15 Mar 2001 16:02:44 -0800
Received: from lupo.thebarn.com (nic-31-c12-219.mn.mediaone.net [24.31.12.219])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2G02gM21267
	for <linux-xfs@oss.sgi.com>; Thu, 15 Mar 2001 16:02:42 -0800
Received: (from cattelan@localhost)
	by lupo.thebarn.com (8.11.3/8.11.0) id f2FNxFA05475
	for linux-xfs@oss.sgi.com; Thu, 15 Mar 2001 17:59:15 -0600 (CST)
	(envelope-from cattelan)
Date: Thu, 15 Mar 2001 17:59:15 -0600 (CST)
From: Russell Cattelan <cattelan@lupo.thebarn.com>
Message-Id: <200103152359.f2FNxFA05475@lupo.thebarn.com>
To: undisclosed-recipients:;
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

Hi folks another really really important test message

From owner-linux-xfs@oss.sgi.com Thu Mar 15 16:07:04 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2G074L21381
	for linux-xfs-outgoing; Thu, 15 Mar 2001 16:07:04 -0800
Received: from lips.borg.umn.edu (lips.borg.umn.edu [160.94.232.50])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2G074M21378
	for <linux-xfs@oss.sgi.com>; Thu, 15 Mar 2001 16:07:04 -0800
Received: (from cattelan@localhost)
	by lips.borg.umn.edu (8.12.0.Beta5/8.12.0.Beta5) id f2FNwfPF045056
	for linux-xfs@oss.sgi.com; Thu, 15 Mar 2001 17:58:41 -0600 (CST)
Date: Thu, 15 Mar 2001 17:58:41 -0600 (CST)
From: Russell Cattelan <cattelan@thebarn.com>
Message-Id: <200103152358.f2FNwfPF045056@lips.borg.umn.edu>
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

Hi folks another really really important test message

From owner-linux-xfs@oss.sgi.com Thu Mar 15 16:13:46 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2G0Dk321640
	for linux-xfs-outgoing; Thu, 15 Mar 2001 16:13:46 -0800
Received: from sgi.com (sgi.SGI.COM [192.48.153.1])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2G0DjM21636
	for <linux-xfs@oss.sgi.com>; Thu, 15 Mar 2001 16:13:45 -0800
Received: from zeus-fddi.americas.sgi.com (128-162-8-103.americas.sgi.com [128.162.8.103]) 
	by sgi.com (980327.SGI.8.8.8-aspam/980304.SGI-aspam:
       SGI does not authorize the use of its proprietary
       systems or networks for unsolicited or bulk email
       from the Internet.) 
	via ESMTP id PAA05979
	for <linux-xfs@oss.sgi.com>; Thu, 15 Mar 2001 15:36:46 -0800 (PST)
	mail_from (lord@sgi.com)
Received: from daisy-e185.americas.sgi.com (daisy-e194.americas.sgi.com [128.162.194.214]) by zeus-fddi.americas.sgi.com (8.9.3/americas-smart-nospam1.1) with ESMTP id RAA1152979 for <linux-xfs@oss.sgi.com>; Thu, 15 Mar 2001 17:35:29 -0600 (CST)
Received: from jen.americas.sgi.com (IDENT:root@jen.americas.sgi.com [128.162.184.86]) by daisy-e185.americas.sgi.com (SGI-8.9.3/SGI-server-1.7) with ESMTP id RAA79911 for <linux-xfs@oss.sgi.com>; Thu, 15 Mar 2001 17:35:29 -0600 (CST)
From: Steve Lord <lord@sgi.com>
Received: by jen.americas.sgi.com (8.11.0/SGI-client-1.7) id f2FNbaa19682; Thu, 15 Mar 2001 17:37:36 -0600
Message-Id: <200103152337.f2FNbaa19682@jen.americas.sgi.com>
Date: Thu, 15 Mar 2001 17:37:36 -0600
Subject: TAKE - fix xfs readdir under an NFS V3 server
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

This is a reworking of readdir in xfs to avoid deadlocks which 
happened with nfs v3 clients which exercise the readdirplus code
in the linux nfs v3 server. Having said that, this change is to
the base code and will be executed in all getdents system calls.

Date:  Thu Mar 15 15:33:20 PST 2001
Workarea:  jen.americas.sgi.com:/src/lord/xfs-linux.2.4

The following file(s) were checked into:
  bonnie.engr.sgi.com:/isms/slinx/2.4.x-xfs


Modid:  2.4.x-xfs:slinx:90125a
linux/fs/xfs/xfs_dir2_block.c - 1.19
linux/fs/xfs/xfs_dir.c - 1.130
linux/fs/xfs/xfs_dir2_sf.c - 1.21
linux/fs/xfs/xfs_dir_leaf.c - 1.94
linux/fs/xfs/xfs_dir_leaf.h - 1.31
linux/fs/xfs/xfs_dir2_leaf.c - 1.19
linux/fs/xfs/xfs_dir2.h - 1.8
linux/fs/xfs/xfs_dir2.c - 1.27
	- Go back to basically the irix code for readdir, linux stuff is handled 
	  in the upper layers now.

linux/fs/xfs/linux/xfs_file.c - 1.40
	- Change linvfs_readdir to allocate a buffer, call xfs to fill it, and
	  then call the filldir function on each entry. This is instead of doing the
	  filldir deep in the bowels of xfs which causes locking problems.

linux/fs/xfs_support/move.h - 1.3
	- The uio_copy function is gone, no longer needed.



From owner-linux-xfs@oss.sgi.com Thu Mar 15 16:17:57 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2G0HvS21794
	for linux-xfs-outgoing; Thu, 15 Mar 2001 16:17:57 -0800
Received: from sgi.com (sgi.SGI.COM [192.48.153.1])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2G0HvM21791
	for <linux-xfs@oss.sgi.com>; Thu, 15 Mar 2001 16:17:57 -0800
Received: from info.engr.sgi.com ([192.26.80.216]) 
	by sgi.com (980327.SGI.8.8.8-aspam/980304.SGI-aspam:
       SGI does not authorize the use of its proprietary
       systems or networks for unsolicited or bulk email
       from the Internet.) 
	via ESMTP id QAB09993; Thu, 15 Mar 2001 16:17:52 -0800 (PST)
	mail_from (nobody@info.engr.sgi.com)
Received: (from nobody@localhost)
	by info.engr.sgi.com (SGI-8.9.3/8.9.3) id QAA02428;
	Thu, 15 Mar 2001 16:17:51 -0800 (PST)
Date: Thu, 15 Mar 2001 16:17:51 -0800 (PST)
Message-Id: <200103160017.QAA02428@info.engr.sgi.com>
X-Pv-Incident: 817944 
webPV: jen.americas.sgi.com 
webExec: webpvupdate,pvincident 
Reply-To: sgi.bugs.xfs@fido.engr.sgi.com
From: pv@relay.sgi.com (lord@sgi.com)
Subject: CLOSE 817944 - NFS V3 from IRIX or SUN client hangs XFS file system
To: nb@sgi.com
Cc: linux-xfs@oss.sgi.com
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

View Incident: http://co-op.engr.sgi.com/BugWorks/code/bwxquery.cgi?search=Search&wlong=1&view_type=Bug&wi=817944

*Status : closed                       Priority : 1                         
 Assigned Engineer : nb                Submitter : tbd                      
 Opened Date : 03/15/01               *Closed Date : 03/15/01               
*Fixed By : lord                      *Fixed By Domain : sgi.com            
*Modified Date : 03/15/01             *Modified User : lord                 
*Modified User Domain : sgi.com       *Fix Description :
==========================
ADDITIONAL INFORMATION (CLOSE)
From: lord@sgi.com (BugWorks)
Date: Mar 15 2001 04:17:51PM
==========================

Fixed by this:

This is a reworking of readdir in xfs to avoid deadlocks which 
happened with nfs v3 clients which exercise the readdirplus code
in the linux nfs v3 server. Having said that, this change is to
the base code and will be executed in all getdents system calls.

Date:  Thu Mar 15 15:33:20 PST 2001
Workarea:  jen.americas.sgi.com:/src/lord/xfs-linux.2.4

The following file(s) were checked into:
  bonnie.engr.sgi.com:/isms/slinx/2.4.x-xfs


Modid:  2.4.x-xfs:slinx:90125a
linux/fs/xfs/xfs_dir2_block.c - 1.19
linux/fs/xfs/xfs_dir.c - 1.130
linux/fs/xfs/xfs_dir2_sf.c - 1.21
linux/fs/xfs/xfs_dir_leaf.c - 1.94
linux/fs/xfs/xfs_dir_leaf.h - 1.31
linux/fs/xfs/xfs_dir2_leaf.c - 1.19
linux/fs/xfs/xfs_dir2.h - 1.8
linux/fs/xfs/xfs_dir2.c - 1.27
	- Go back to basically the irix code for readdir, linux stuff is handled 
	  in the upper layers now.

linux/fs/xfs/linux/xfs_file.c - 1.40
	- Change linvfs_readdir to allocate a buffer, call xfs to fill it, and
	  then call the filldir function on each entry. This is instead of doing the
	  filldir deep in the bowels of xfs which causes locking problems.

linux/fs/xfs_support/move.h - 1.3
	- The uio_copy function is gone, no longer needed.

From owner-linux-xfs@oss.sgi.com Thu Mar 15 16:17:58 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2G0HwC21807
	for linux-xfs-outgoing; Thu, 15 Mar 2001 16:17:58 -0800
Received: from pneumatic-tube.sgi.com (pneumatic-tube.sgi.com [204.94.214.22])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2G0HwM21801
	for <linux-xfs@oss.sgi.com>; Thu, 15 Mar 2001 16:17:58 -0800
Received: from zeus-fddi.americas.sgi.com (zeus-fddi.americas.sgi.com [128.162.8.103]) by pneumatic-tube.sgi.com (980327.SGI.8.8.8-aspam/980310.SGI-aspam) via ESMTP id PAA03261
	for <linux-xfs@oss.sgi.com>; Thu, 15 Mar 2001 15:21:44 -0800 (PST)
	mail_from (lord@sgi.com)
Received: from daisy-e185.americas.sgi.com (daisy-e194.americas.sgi.com [128.162.194.214]) by zeus-fddi.americas.sgi.com (8.9.3/americas-smart-nospam1.1) with ESMTP id RAA1151016 for <linux-xfs@oss.sgi.com>; Thu, 15 Mar 2001 17:10:34 -0600 (CST)
Received: from jen.americas.sgi.com (IDENT:root@jen.americas.sgi.com [128.162.184.86]) by daisy-e185.americas.sgi.com (SGI-8.9.3/SGI-server-1.7) with ESMTP id RAA73680 for <linux-xfs@oss.sgi.com>; Thu, 15 Mar 2001 17:10:34 -0600 (CST)
From: Steve Lord <lord@sgi.com>
Received: by jen.americas.sgi.com (8.11.0/SGI-client-1.7) id f2FNCfo29503; Thu, 15 Mar 2001 17:12:41 -0600
Message-Id: <200103152312.f2FNCfo29503@jen.americas.sgi.com>
Date: Thu, 15 Mar 2001 17:12:41 -0600
Subject: TAKE - bugfix to log code
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

Only applies in filesystem shutdown case

Date:  Thu Mar 15 15:10:02 PST 2001
Workarea:  jen.americas.sgi.com:/src/lord/xfs-linux.2.4

The following file(s) were checked into:
  bonnie.engr.sgi.com:/isms/slinx/2.4.x-xfs


Modid:  2.4.x-xfs:slinx:90124a
linux/fs/xfs/xfs_log.c - 1.231
	- Small bug fix to last merge from irix, hmm now we have the fix for
	  linux, but not irix....



From owner-linux-xfs@oss.sgi.com Thu Mar 15 16:21:23 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2G0LNu21892
	for linux-xfs-outgoing; Thu, 15 Mar 2001 16:21:23 -0800
Received: from pneumatic-tube.sgi.com (pneumatic-tube.sgi.com [204.94.214.22])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2G0LNM21889
	for <linux-xfs@oss.sgi.com>; Thu, 15 Mar 2001 16:21:23 -0800
Received: from feature.engr.sgi.com (feature.engr.sgi.com [130.62.40.134]) by pneumatic-tube.sgi.com (980327.SGI.8.8.8-aspam/980310.SGI-aspam) via ESMTP id QAA08402; Thu, 15 Mar 2001 16:31:16 -0800 (PST)
	mail_from (pv@feature.engr.sgi.com)
Received: (from pv@localhost) by feature.engr.sgi.com (980427.SGI.8.8.8/970903.SGI.AUTOCF) id QAA12300; Thu, 15 Mar 2001 16:20:04 -0800 (PST)
Date: Thu, 15 Mar 2001 16:20:04 -0800 (PST)
Message-Id: <200103160020.QAA12300@feature.engr.sgi.com>
X-Pv-Incident: 817944 
Reply-To: sgi.bugs.xfs@fido.engr.sgi.com
From: pv@fddi-odin.corp.sgi.com (nb@sgi.com)
Subject: ADD 817944 - NFS V3 from IRIX or SUN client hangs XFS file system
To: nb@sgi.com
Cc: linux-xfs@oss.sgi.com
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

 Submitter : tbd                       Status : closed                      
 Assigned Engineer : nb                Priority : 1                         
*Modified Date : 03/15/01             *Modified User : nb                   
*Modified User Domain : sgi.com       *Description :
Exporting an XFS file system on a linux server and using it
NFS V3 server from an Irix or Sun client hangs the XFS directory.

To recreate:
On IRIX machine:
  mkdir /tmp/permit
  mount permit:/mnt_xfs /tmp/permit
  cd /home/daisy08/nettest/tests/nfs/regression
  cp -r sgi.new /tmp/permit/sgi.new
  cd /tmp/permit/sgi.new/special

.....


==========================
ADDITIONAL INFORMATION (ADD)
From: neil bannister <nb@sgi.com>
Date: Mar 15 2001 04:20:04PM
[pvnews version: 1.71]
==========================
Hello,

I will be out of the office Friday this week, back in Monday and
travelling Monday night thru Tuesday night and then back in the
office Wednesday 21st.

I will be checking email as best I can. If it urgent pls contact 
my admin, Ila Roske, 651-683-3034, vnet 233-3034 or roskei@sgi.com.

Thanks

Neil Bannister

From owner-linux-xfs@oss.sgi.com Thu Mar 15 16:22:41 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2G0MfV21922
	for linux-xfs-outgoing; Thu, 15 Mar 2001 16:22:41 -0800
Received: from sgi.com (sgi.SGI.COM [192.48.153.1])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2G0MfM21919
	for <linux-xfs@oss.sgi.com>; Thu, 15 Mar 2001 16:22:41 -0800
Received: from waco.engr.sgi.com ([163.154.18.95]) 
	by sgi.com (980327.SGI.8.8.8-aspam/980304.SGI-aspam:
       SGI does not authorize the use of its proprietary
       systems or networks for unsolicited or bulk email
       from the Internet.) 
	via ESMTP id NAA07858
	for <linux-xfs@oss.sgi.com>; Thu, 15 Mar 2001 13:52:43 -0800 (PST)
	mail_from (ananth@waco.engr.sgi.com)
Received: (from ananth@localhost)
	by waco.engr.sgi.com (8.11.0/8.11.0) id f2FLrEc04821
	for linux-xfs@oss.sgi.com; Thu, 15 Mar 2001 13:53:14 -0800
Date: Thu, 15 Mar 2001 13:53:14 -0800
From: Ananth Ananthanarayanan <ananth@waco.engr.sgi.com>
Message-Id: <200103152153.f2FLrEc04821@waco.engr.sgi.com>
Subject: TAKE - turn off debug messages
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

Date:  Thu Mar 15 13:51:54 PST 2001
Workarea:  waco.engr.sgi.com:/build1/ananth/xfs-tot

The following file(s) were checked into:
  bonnie.engr.sgi.com:/isms/slinx/2.4.x-xfs


Modid:  2.4.x-xfs:slinx:90106a
linux/drivers/md/raid5.c - 1.11
	- Turn off a couple of chatty messages.



From owner-linux-xfs@oss.sgi.com Thu Mar 15 16:34:00 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2G0Y0N22203
	for linux-xfs-outgoing; Thu, 15 Mar 2001 16:34:00 -0800
Received: from deliverator.sgi.com (deliverator.sgi.com [204.94.214.10])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2G0Y0M22198
	for <linux-xfs@oss.sgi.com>; Thu, 15 Mar 2001 16:34:00 -0800
Received: from info.engr.sgi.com (info.engr.sgi.com [192.26.80.216]) by deliverator.sgi.com (980309.SGI.8.8.8-aspam-6.2/980310.SGI-aspam) via ESMTP id OAA12187; Thu, 15 Mar 2001 14:40:50 -0800 (PST)
	mail_from (nobody@info.engr.sgi.com)
Received: (from nobody@localhost)
	by info.engr.sgi.com (SGI-8.9.3/8.9.3) id OAA25960;
	Thu, 15 Mar 2001 14:42:01 -0800 (PST)
Date: Thu, 15 Mar 2001 14:42:01 -0800 (PST)
Message-Id: <200103152242.OAA25960@info.engr.sgi.com>
X-Pv-Incident: 800061 
webPV: oldchicago.americas.sgi.com 
webExec: webpvupdate,pvincident 
Reply-To: sgi.bugs.xfs@fido.engr.sgi.com
From: pv@relay.sgi.com (nb@sgi.com)
Subject: REASSIGN 800061 - HIGHMEM support is broken
To: raa@sgi.com
Cc: linux-xfs@oss.sgi.com
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

View Incident: http://co-op.engr.sgi.com/BugWorks/code/bwxquery.cgi?search=Search&wlong=1&view_type=Bug&wi=800061

 Status : open                         Priority : 2                         
*Assigned Engineer : raa               Submitter : ananth                   
 Project : xfs-linux                   Assigned Group : xfs-linux           
 Opened Date : 08/24/00               *Modified User : nb                   
*Modified User Domain : sgi.com       *Description :
There are two problems in handling highmem with XFS:

1. consistent use of kmap/kunmap to get at the address
   of a page.

2. supporting highmem with kiobufs.

I'll check-in a short fix that will check, issue warning, etc.
if highmem is turned on.


.....


==========================
ADDITIONAL INFORMATION (REASSIGN)
From: nb@sgi.com (BugWorks)
Date: Mar 15 2001 02:42:00PM
==========================

re-assigning Chait's PV to Bob

From owner-linux-xfs@oss.sgi.com Thu Mar 15 16:44:36 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2G0iaD22632
	for linux-xfs-outgoing; Thu, 15 Mar 2001 16:44:36 -0800
Received: from deliverator.sgi.com (deliverator.sgi.com [204.94.214.10])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2G0iYM22618
	for <linux-xfs@oss.sgi.com>; Thu, 15 Mar 2001 16:44:34 -0800
Received: from cthulhu.engr.sgi.com (gate3-relay.engr.sgi.com [130.62.1.234]) by deliverator.sgi.com (980309.SGI.8.8.8-aspam-6.2/980310.SGI-aspam) via ESMTP id MAA19186
	for <linux-xfs@oss.sgi.com>; Thu, 15 Mar 2001 12:08:05 -0800 (PST)
	mail_from (tduffy@engr.sgi.com)
Received: from dbear.engr.sgi.com (dbear.engr.sgi.com [163.154.18.85])
	by cthulhu.engr.sgi.com (SGI-8.9.3/8.9.3) with ESMTP id MAA34103
	for <linux-xfs@oss.sgi.com>; Thu, 15 Mar 2001 12:07:59 -0800 (PST)
Date: Thu, 15 Mar 2001 12:04:16 -0800 (PST)
From: Tom Duffy <tduffy@engr.sgi.com>
To: <linux-xfs@oss.sgi.com>
Subject: filesystem locking?
Message-ID: <Pine.LNX.4.30.0103151202020.2555-100001@dbear.engr.sgi.com>
MIME-Version: 1.0
Content-Type: MULTIPART/MIXED; BOUNDARY="-1700571886-1030679191-984686648=:2555"
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

  This message is in MIME format.  The first part should be readable text,
  while the remaining parts are likely unreadable without MIME-aware tools.
  Send mail to mime@docserver.cac.washington.edu for more info.

---1700571886-1030679191-984686648=:2555
Content-Type: TEXT/PLAIN; charset=US-ASCII

I got this error running VMware on XFS (see attached jpg).

if xfs does support the file locking semantics that vmware wants (which I
assume), we should let them know so they can include xfs awareness in
their next release...

-tduffy

---1700571886-1030679191-984686648=:2555
Content-Type: IMAGE/jpeg; name="vmware_error.jpg"
Content-Transfer-Encoding: BASE64
Content-ID: <Pine.LNX.4.30.0103151204080.2555@dbear.engr.sgi.com>
Content-Description: 
Content-Disposition: attachment; filename="vmware_error.jpg"
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---1700571886-1030679191-984686648=:2555--

From owner-linux-xfs@oss.sgi.com Fri Mar 16 02:01:54 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2GA1si32275
	for linux-xfs-outgoing; Fri, 16 Mar 2001 02:01:54 -0800
Received: from inpbox.inp.nsk.su (inpbox.inp.nsk.su [193.124.167.24])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2GA1qM32272
	for <linux-xfs@oss.sgi.com>; Fri, 16 Mar 2001 02:01:52 -0800
Received: from inp.nsk.su (IDENT:byrganov@cfs.inp.nsk.su [193.124.167.1])
	by inpbox.inp.nsk.su (8.9.0/8.9.0) with ESMTP id PAA22952
	for <linux-xfs@oss.sgi.com>; Fri, 16 Mar 2001 15:47:26 +0600 (NSK)
Message-ID: <3AB1E12C.29643DEB@inp.nsk.su>
Date: Fri, 16 Mar 2001 15:47:24 +0600
From: "Eugene B. Byrganov" <SkyMaster+remove+it@inp.nsk.su>
X-Mailer: Mozilla 4.75 [ru] (X11; U; Linux 2.2.16-3bc i586)
X-Accept-Language: ru, en, *
MIME-Version: 1.0
To: linux-xfs@oss.sgi.com
Subject: xfsrestory problem.
Content-Type: text/plain; charset=koi8-r
Content-Transfer-Encoding: 8bit
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

Hi!

Current dir: ext2 and 'xfs-root.dump' - dump on Irix 6.5
xfs: CVS 15 Mar 2001

First: xfsrestore -r -f xfs-root.dump
--------------------------------------------------
xfsrestore: NOTE: attempt to reserve 46720 bytes for
/home/sky/xfsrestorehousekeepingdir/dirattr using F_ALLOCSP64 failed:
Inappropriate ioctl for device (25)
xfsrestore: NOTE: attempt to reserve 46720 bytes for
/home/sky/xfsrestorehousekeepingdir/dirattr using F_ALLOCSP64 failed:
Inappropriate ioctl for device (25)
xfsrestore: NOTE: attempt to reserve 45496 bytes for
/home/sky/xfsrestorehousekeepingdir/namreg using F_ALLOCSP64 failed:
Inappropriate ioctl for device (25)
xfsrestore: NOTE: attempt to reserve 45496 bytes for
/home/sky/xfsrestorehousekeepingdir/namreg using F_ALLOCSP64 failed:
Inappropriate ioctl for device (25)
--------------------------------------------------

Don't 'xfsrestore'  know about ext2?

Second:
--------------------------------------------------
[root@cfs sky]# xfsrestore -t -f xfs-root.dump
xfsrestore: version 3.0 - Running single-threaded
xfsrestore: ERROR: Current directory not XFS: /home/sky
--------------------------------------------------
Abort!!!!

Why? I only want get list of files.

--
                      ó ÎÁÉÌÕÞÛÉÍÉ ÐÏÖÅÌÁÎÉÑÍÉ, å×ÇÅÎÉÊ âÙÒÇÁÎÏ×.
                      Best regards, Eugene Byrganov.
                      Viele Gruesse, Eugen Byrganow.

  mailto:E.B.Byrganov@inp.nsk.su




From owner-linux-xfs@oss.sgi.com Fri Mar 16 06:19:38 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2GEJcF05897
	for linux-xfs-outgoing; Fri, 16 Mar 2001 06:19:38 -0800
Received: from web1607.mail.yahoo.com (web1607.mail.yahoo.com [128.11.23.149])
	by oss.sgi.com (8.11.3/8.11.3) with SMTP id f2GEJbM05894
	for <linux-xfs@oss.sgi.com>; Fri, 16 Mar 2001 06:19:37 -0800
Received: (qmail 175 invoked by uid 60001); 16 Mar 2001 14:19:37 -0000
Message-ID: <20010316141937.174.qmail@web1607.mail.yahoo.com>
Received: from [202.95.126.106] by web1607.mail.yahoo.com; Fri, 16 Mar 2001 06:19:37 PST
Date: Fri, 16 Mar 2001 06:19:37 -0800 (PST)
From: girish chandrashekara <girish_c_99@yahoo.com>
Subject: dmapi support in XFS
To: linux-xfs@oss.sgi.com
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

Hi,
  Iam working on dmapi support in VFS layer of linux
to make it generic for underlying file systems. In the
process I went through your source code
for dmapi support in XFS(2.4.0-test5 kernel), I need
some clarifications (correct me if Iam wrongly
understood)

(1)Data events are triggered by the caller(through
xfs_ioctls, xfs_dm_fcntl) except create(truncate
event). How XFS manages (trigger implicitly-initiated
by kernel)to send data event to DM application incase
of read,write and mmap by user application on the
managed region

(2)Nospace event is also triggered through xfs_ioctl
when filesystem  runs out of space. How XFS triggers
nospace event implicitly, when filesystem runs out of
space

thanks in advance

regards
girish_c









__________________________________________________
Do You Yahoo!?
Get email at your own domain with Yahoo! Mail. 
http://personal.mail.yahoo.com/

From owner-linux-xfs@oss.sgi.com Fri Mar 16 06:44:59 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2GEixJ06663
	for linux-xfs-outgoing; Fri, 16 Mar 2001 06:44:59 -0800
Received: from deliverator.sgi.com (deliverator.sgi.com [204.94.214.10])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2GEiwM06660
	for <linux-xfs@oss.sgi.com>; Fri, 16 Mar 2001 06:44:58 -0800
Received: from zeus-fddi.americas.sgi.com (zeus-fddi.americas.sgi.com [128.162.8.103]) by deliverator.sgi.com (980309.SGI.8.8.8-aspam-6.2/980310.SGI-aspam) via ESMTP id GAA19079
	for <linux-xfs@oss.sgi.com>; Fri, 16 Mar 2001 06:43:47 -0800 (PST)
	mail_from (lord@sgi.com)
Received: from daisy-e185.americas.sgi.com (daisy-e194.americas.sgi.com [128.162.194.214]) by zeus-fddi.americas.sgi.com (8.9.3/americas-smart-nospam1.1) with ESMTP id IAA1160896; Fri, 16 Mar 2001 08:43:40 -0600 (CST)
Received: from jen.americas.sgi.com (IDENT:root@jen.americas.sgi.com [128.162.184.86]) by daisy-e185.americas.sgi.com (SGI-8.9.3/SGI-server-1.7) with ESMTP id IAA93779; Fri, 16 Mar 2001 08:43:40 -0600 (CST)
Received: from jen.americas.sgi.com by jen.americas.sgi.com (8.11.0/SGI-client-1.7) via ESMTP id f2GEjYe19328; Fri, 16 Mar 2001 08:45:34 -0600
Message-Id: <200103161445.f2GEjYe19328@jen.americas.sgi.com>
X-Mailer: exmh version 2.2 06/23/2000 with nmh-1.0.4
To: "Eugene B. Byrganov" <SkyMaster@inp.nsk.su>
cc: linux-xfs@oss.sgi.com
Subject: Re: xfsrestory problem. 
In-Reply-To: Message from "Eugene B. Byrganov" <SkyMaster+remove+it@inp.nsk.su> 
   of "Fri, 16 Mar 2001 15:47:24 +0600." <3AB1E12C.29643DEB@inp.nsk.su> 
Content-Transfer-Encoding: 8bit
Date: Fri, 16 Mar 2001 08:45:34 -0600
From: Steve Lord <lord@sgi.com>
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk


I am not sure how much work it would be to get xfsrestore to function
on a non-xfs filesystem. As the names indicate, these are xfs specific
programs - they know about the extra things in xfs filesystems such as
extended attributes and can store these. It also looks like they take
advantage of xfs special system calls.

The second error seems a little draconian.

Our dump/restore experts are in Australia and it is the middle of the
night there, they may have comments later.

Steve

> Hi!
> 
> Current dir: ext2 and 'xfs-root.dump' - dump on Irix 6.5
> xfs: CVS 15 Mar 2001
> 
> First: xfsrestore -r -f xfs-root.dump
> --------------------------------------------------
> xfsrestore: NOTE: attempt to reserve 46720 bytes for
> /home/sky/xfsrestorehousekeepingdir/dirattr using F_ALLOCSP64 failed:
> Inappropriate ioctl for device (25)
> xfsrestore: NOTE: attempt to reserve 46720 bytes for
> /home/sky/xfsrestorehousekeepingdir/dirattr using F_ALLOCSP64 failed:
> Inappropriate ioctl for device (25)
> xfsrestore: NOTE: attempt to reserve 45496 bytes for
> /home/sky/xfsrestorehousekeepingdir/namreg using F_ALLOCSP64 failed:
> Inappropriate ioctl for device (25)
> xfsrestore: NOTE: attempt to reserve 45496 bytes for
> /home/sky/xfsrestorehousekeepingdir/namreg using F_ALLOCSP64 failed:
> Inappropriate ioctl for device (25)
> --------------------------------------------------
> 
> Don't 'xfsrestore'  know about ext2?
> 
> Second:
> --------------------------------------------------
> [root@cfs sky]# xfsrestore -t -f xfs-root.dump
> xfsrestore: version 3.0 - Running single-threaded
> xfsrestore: ERROR: Current directory not XFS: /home/sky
> --------------------------------------------------
> Abort!!!!
> 
> Why? I only want get list of files.
> 
> --
>                       s NAILU^[IMI POVELANIQMI, eWGENIJ bYRGANOW.
>                       Best regards, Eugene Byrganov.
>                       Viele Gruesse, Eugen Byrganow.
> 
>   mailto:E.B.Byrganov@inp.nsk.su
> 
> 



From owner-linux-xfs@oss.sgi.com Fri Mar 16 07:12:56 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2GFCuD07152
	for linux-xfs-outgoing; Fri, 16 Mar 2001 07:12:56 -0800
Received: from stine.vestdata.no (IDENT:0@stine.vestdata.no [195.204.68.10])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2GFCsM07149
	for <linux-xfs@oss.sgi.com>; Fri, 16 Mar 2001 07:12:55 -0800
Received: (from ragnark@localhost)
	by stine.vestdata.no (8.9.3/8.9.3) id PAA29295;
	Fri, 16 Mar 2001 15:39:49 +0100
Date: Fri, 16 Mar 2001 15:39:48 +0100
From: =?iso-8859-1?Q?Ragnar_Kj=F8rstad?= <xfs@ragnark.vestdata.no>
To: girish chandrashekara <girish_c_99@yahoo.com>
Cc: linux-xfs@oss.sgi.com, openxdsm-devel@lists.sourceforge.net
Subject: Re: dmapi support in XFS
Message-ID: <20010316153948.C27307@vestdata.no>
References: <20010316141937.174.qmail@web1607.mail.yahoo.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=iso-8859-1
Content-Transfer-Encoding: 8bit
X-Mailer: Mutt 0.95.5i
In-Reply-To: <20010316141937.174.qmail@web1607.mail.yahoo.com>; from girish chandrashekara on Fri, Mar 16, 2001 at 06:19:37AM -0800
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

On Fri, Mar 16, 2001 at 06:19:37AM -0800, girish chandrashekara wrote:
> Hi,
>   Iam working on dmapi support in VFS layer of linux
> to make it generic for underlying file systems. 

Are you awere of the openxdsm project?
(http://openxdsm.sourceforge.net/)

It's a project to do _exactly_ that. 

The status so far is that we've worked on a list of potential problems
(thanks to the SGI developers for helping us identify them) and ways to
solve them. You'll find some information in our mail-archive.

Please tell us more about your project. (preferably on the openxdsm
list)


-- 
Ragnar Kjørstad
Big Storage

From owner-linux-xfs@oss.sgi.com Fri Mar 16 08:22:47 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2GGMl408624
	for linux-xfs-outgoing; Fri, 16 Mar 2001 08:22:47 -0800
Received: from sgi.com (sgi.SGI.COM [192.48.153.1])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2GGMlM08621
	for <linux-xfs@oss.sgi.com>; Fri, 16 Mar 2001 08:22:47 -0800
Received: from zeus-fddi.americas.sgi.com (128-162-8-103.americas.sgi.com [128.162.8.103]) 
	by sgi.com (980327.SGI.8.8.8-aspam/980304.SGI-aspam:
       SGI does not authorize the use of its proprietary
       systems or networks for unsolicited or bulk email
       from the Internet.) 
	via ESMTP id IAA09420
	for <linux-xfs@oss.sgi.com>; Fri, 16 Mar 2001 08:22:17 -0800 (PST)
	mail_from (lord@sgi.com)
Received: from daisy-e185.americas.sgi.com (daisy-e194.americas.sgi.com [128.162.194.214]) by zeus-fddi.americas.sgi.com (8.9.3/americas-smart-nospam1.1) with ESMTP id KAA1160719 for <linux-xfs@oss.sgi.com>; Fri, 16 Mar 2001 10:20:57 -0600 (CST)
Received: from jen.americas.sgi.com (IDENT:root@jen.americas.sgi.com [128.162.184.86]) by daisy-e185.americas.sgi.com (SGI-8.9.3/SGI-server-1.7) with ESMTP id KAA59250 for <linux-xfs@oss.sgi.com>; Fri, 16 Mar 2001 10:20:57 -0600 (CST)
From: Steve Lord <lord@sgi.com>
Received: by jen.americas.sgi.com (8.11.0/SGI-client-1.7) id f2GGMo202572; Fri, 16 Mar 2001 10:22:50 -0600
Message-Id: <200103161622.f2GGMo202572@jen.americas.sgi.com>
Date: Fri, 16 Mar 2001 10:22:50 -0600
Subject: TAKE - fixes another directory testcase
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

Small hole in yesterdays changes.

Date:  Fri Mar 16 08:20:24 PST 2001
Workarea:  jen.americas.sgi.com:/src/lord/xfs-linux.2.4

The following file(s) were checked into:
  bonnie.engr.sgi.com:/isms/slinx/2.4.x-xfs


Modid:  2.4.x-xfs:slinx:90192a
linux/fs/xfs/linux/xfs_file.c - 1.41
	- Fix setting of file offset in directory read code, there was a case
	  where an entry was accounted for in the offset but not returned to
	  the user.



From owner-linux-xfs@oss.sgi.com Fri Mar 16 09:03:26 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2GH3QV09511
	for linux-xfs-outgoing; Fri, 16 Mar 2001 09:03:26 -0800
Received: from horus.bticc.net (IDENT:root@horus.bticc.net [216.196.51.155])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2GH3OM09508
	for <linux-xfs@oss.sgi.com>; Fri, 16 Mar 2001 09:03:25 -0800
Received: from horus.bticc.net (IDENT:jkinsley@horus.bticc.net [216.196.51.155])
	by horus.bticc.net (8.9.3/8.9.3) with ESMTP id MAA30346;
	Fri, 16 Mar 2001 12:03:19 -0500
Date: Fri, 16 Mar 2001 12:03:19 -0500 (EST)
From: J Kinsley <jkinsley@bticc.net>
To: <proftpd-users@proftpd.org>, <proftpd-devel@proftpd.org>,
   <linux-xfs@oss.sgi.com>
Subject: ProFTPd + linux-2.4.2-XFS = RETR bug
Message-ID: <Pine.LNX.4.30.0103160413420.27608-100000@horus.bticc.net>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

I am running ProFTPd 1.2.1 on an experimental Redhat 7.0/7.1beta
(glibc-2.2) PII box.  The kernel is the stock 2.4.2 specifically
configured for the box it is running on.  The only patch applied
is the Feb232001devel XFS patch from SGI.

(ftp://linux-xfs.sgi.com/projects/xfs/download/patches/Feb232001devel.patch.gz)

Both the kernel and XFS tools were compiled using egcs-1.1.2 (kgcc on
RH70) while ProFTPd and everything else were compiled with the RH71
gcc.

I have ProFTPd configured with one master server and two virtual
hosts.  The DefaultRoot for each site is on separate partition or
disk. All partitions were created using the mkfs.xfs command from the
XFS patch kit.  In addition, the DefaultRoot of the two virtualhosts
is accessable via the web using Apache-1.3.19 which works properly.

The bug arises when the remote client attempts to RETR a file with
filesize >= 16384.  The file is received by the remote as an empty
file (filesize ==0) with all remaining file attributes (name, mtime,
etc) preserved.  However, if a particular file with filesize >= 16384
has recently been accessed by some other process such as Apache, then
the remote FTP client can receive the complete file irregardless of
its filesize.  Once a period of time passes after the file has been
accessed by another process, the file will again be returned empty to
the remote FTP client.  I have not yet determined what the exact
period of time is, but on averate it is one hour.  Files with
filesize < 16384 seem to be uneffected by this bug.

For those that want to observe this bug in action, goto:

ftp://ftp.freeworldbbs.org:4021/

The subdirectory, xfs-test contains contains five test files.  The
filesizes are 8192, 16383, 16384, 16385, and 32768.  The first two
files will always be retrieved while the last three will be empty
files when retrived.  Since I just copied the files to a parallel
FTP tree, I can now download all of them via the remote client.

The parallel FTP tree is identical to the one here except that it is
stored on an ext2 filesystem.  It can be accessed at:

ftp://ftp.freeworldbbs.org/

The ProFTPD configuration for the two sites is identical with the
exception of /mnt being appended to the paths and a Port 4021
directive being aded in the first one.

I have been using ProFTPd for several years without incident, and now
that I have tried XFS, I certainly do not want to deevolve my box
back to ext2.  Unfortunately, my skills are not at the level required
to go poking around in the kernel, XFS, and ProFTPd to figure out a
solution to this problem.  However, I can test any possible patches
to ProFTPd.


Regards,
Jarrod Kinsley


From owner-linux-xfs@oss.sgi.com Fri Mar 16 09:11:59 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2GHBxC09604
	for linux-xfs-outgoing; Fri, 16 Mar 2001 09:11:59 -0800
Received: from deliverator.sgi.com (deliverator.sgi.com [204.94.214.10])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2GHBwM09601
	for <linux-xfs@oss.sgi.com>; Fri, 16 Mar 2001 09:11:58 -0800
Received: from zeus-fddi.americas.sgi.com (zeus-fddi.americas.sgi.com [128.162.8.103]) by deliverator.sgi.com (980309.SGI.8.8.8-aspam-6.2/980310.SGI-aspam) via ESMTP id JAA10033
	for <linux-xfs@oss.sgi.com>; Fri, 16 Mar 2001 09:10:46 -0800 (PST)
	mail_from (lord@sgi.com)
Received: from daisy-e185.americas.sgi.com (daisy-e194.americas.sgi.com [128.162.194.214]) by zeus-fddi.americas.sgi.com (8.9.3/americas-smart-nospam1.1) with ESMTP id LAA1157622; Fri, 16 Mar 2001 11:10:40 -0600 (CST)
Received: from jen.americas.sgi.com (IDENT:root@jen.americas.sgi.com [128.162.184.86]) by daisy-e185.americas.sgi.com (SGI-8.9.3/SGI-server-1.7) with ESMTP id LAA98098; Fri, 16 Mar 2001 11:10:40 -0600 (CST)
Received: from jen.americas.sgi.com by jen.americas.sgi.com (8.11.0/SGI-client-1.7) via ESMTP id f2GHCXi04411; Fri, 16 Mar 2001 11:12:33 -0600
Message-Id: <200103161712.f2GHCXi04411@jen.americas.sgi.com>
X-Mailer: exmh version 2.2 06/23/2000 with nmh-1.0.4
To: J Kinsley <jkinsley@bticc.net>
cc: proftpd-users@proftpd.org, proftpd-devel@proftpd.org,
   linux-xfs@oss.sgi.com
Subject: Re: ProFTPd + linux-2.4.2-XFS = RETR bug 
In-Reply-To: Message from J Kinsley <jkinsley@bticc.net> 
   of "Fri, 16 Mar 2001 12:03:19 EST." <Pine.LNX.4.30.0103160413420.27608-100000@horus.bticc.net> 
Date: Fri, 16 Mar 2001 11:12:33 -0600
From: Steve Lord <lord@sgi.com>
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk


This is almost certainly not proftpd, but something about xfs, if possible
can you download the cvs version of the kernel, Feb 23 is a long time ago
as far as xfs is concerned now, and I would like to determine if the
problem is still present in xfs. 

Steve

> I am running ProFTPd 1.2.1 on an experimental Redhat 7.0/7.1beta
> (glibc-2.2) PII box.  The kernel is the stock 2.4.2 specifically
> configured for the box it is running on.  The only patch applied
> is the Feb232001devel XFS patch from SGI.
> 
> (ftp://linux-xfs.sgi.com/projects/xfs/download/patches/Feb232001devel.patch.g
> z)
> 
> Both the kernel and XFS tools were compiled using egcs-1.1.2 (kgcc on
> RH70) while ProFTPd and everything else were compiled with the RH71
> gcc.
> 
> I have ProFTPd configured with one master server and two virtual
> hosts.  The DefaultRoot for each site is on separate partition or
> disk. All partitions were created using the mkfs.xfs command from the
> XFS patch kit.  In addition, the DefaultRoot of the two virtualhosts
> is accessable via the web using Apache-1.3.19 which works properly.
> 
> The bug arises when the remote client attempts to RETR a file with
> filesize >= 16384.  The file is received by the remote as an empty
> file (filesize ==0) with all remaining file attributes (name, mtime,
> etc) preserved.  However, if a particular file with filesize >= 16384
> has recently been accessed by some other process such as Apache, then
> the remote FTP client can receive the complete file irregardless of
> its filesize.  Once a period of time passes after the file has been
> accessed by another process, the file will again be returned empty to
> the remote FTP client.  I have not yet determined what the exact
> period of time is, but on averate it is one hour.  Files with
> filesize < 16384 seem to be uneffected by this bug.
> 
> For those that want to observe this bug in action, goto:
> 
> ftp://ftp.freeworldbbs.org:4021/
> 
> The subdirectory, xfs-test contains contains five test files.  The
> filesizes are 8192, 16383, 16384, 16385, and 32768.  The first two
> files will always be retrieved while the last three will be empty
> files when retrived.  Since I just copied the files to a parallel
> FTP tree, I can now download all of them via the remote client.
> 
> The parallel FTP tree is identical to the one here except that it is
> stored on an ext2 filesystem.  It can be accessed at:
> 
> ftp://ftp.freeworldbbs.org/
> 
> The ProFTPD configuration for the two sites is identical with the
> exception of /mnt being appended to the paths and a Port 4021
> directive being aded in the first one.
> 
> I have been using ProFTPd for several years without incident, and now
> that I have tried XFS, I certainly do not want to deevolve my box
> back to ext2.  Unfortunately, my skills are not at the level required
> to go poking around in the kernel, XFS, and ProFTPd to figure out a
> solution to this problem.  However, I can test any possible patches
> to ProFTPd.
> 
> 
> Regards,
> Jarrod Kinsley



From owner-linux-xfs@oss.sgi.com Fri Mar 16 11:23:41 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2GJNfC13182
	for linux-xfs-outgoing; Fri, 16 Mar 2001 11:23:41 -0800
Received: from sgi.com (sgi.SGI.COM [192.48.153.1])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2GJNfM13179
	for <linux-xfs@oss.sgi.com>; Fri, 16 Mar 2001 11:23:41 -0800
Received: from zeus-fddi.americas.sgi.com (128-162-8-103.americas.sgi.com [128.162.8.103]) 
	by sgi.com (980327.SGI.8.8.8-aspam/980304.SGI-aspam:
       SGI does not authorize the use of its proprietary
       systems or networks for unsolicited or bulk email
       from the Internet.) 
	via ESMTP id LAA08828
	for <linux-xfs@oss.sgi.com>; Fri, 16 Mar 2001 11:23:31 -0800 (PST)
	mail_from (lord@sgi.com)
Received: from daisy-e185.americas.sgi.com (daisy-e194.americas.sgi.com [128.162.194.214]) by zeus-fddi.americas.sgi.com (8.9.3/americas-smart-nospam1.1) with ESMTP id NAA1164440; Fri, 16 Mar 2001 13:22:15 -0600 (CST)
Received: from jen.americas.sgi.com (IDENT:root@jen.americas.sgi.com [128.162.184.86]) by daisy-e185.americas.sgi.com (SGI-8.9.3/SGI-server-1.7) with ESMTP id NAA28126; Fri, 16 Mar 2001 13:22:15 -0600 (CST)
Received: from jen.americas.sgi.com by jen.americas.sgi.com (8.11.0/SGI-client-1.7) via ESMTP id f2GJO7l31603; Fri, 16 Mar 2001 13:24:07 -0600
Message-Id: <200103161924.f2GJO7l31603@jen.americas.sgi.com>
X-Mailer: exmh version 2.2 06/23/2000 with nmh-1.0.4
To: Tom Duffy <tduffy@engr.sgi.com>
cc: linux-xfs@oss.sgi.com
Subject: Re: filesystem locking? 
In-Reply-To: Message from Tom Duffy <tduffy@engr.sgi.com> 
   of "Thu, 15 Mar 2001 12:04:16 PST." <Pine.LNX.4.30.0103151202020.2555-100001@dbear.engr.sgi.com> 
Date: Fri, 16 Mar 2001 13:24:07 -0600
From: Steve Lord <lord@sgi.com>
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk


> 
> I got this error running VMware on XFS (see attached jpg).
> 
> if xfs does support the file locking semantics that vmware wants (which I
> assume), we should let them know so they can include xfs awareness in
> their next release...
> 
> -tduffy
> 

Hmm, so you think they probe the filesystem type and then decide they
do not know what it can do if they do not recognize the type? They could
try a couple of tests to see what works, even different revisions of 
the same filesystem could have different feature sets.

Any vmware experts on the list know what is going on here?

Steve



From owner-linux-xfs@oss.sgi.com Fri Mar 16 12:18:52 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2GKIq214192
	for linux-xfs-outgoing; Fri, 16 Mar 2001 12:18:52 -0800
Received: from stine.vestdata.no (IDENT:0@stine.vestdata.no [195.204.68.10])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2GKIoM14189
	for <linux-xfs@oss.sgi.com>; Fri, 16 Mar 2001 12:18:51 -0800
Received: (from ragnark@localhost)
	by stine.vestdata.no (8.9.3/8.9.3) id VAA03040;
	Fri, 16 Mar 2001 21:18:13 +0100
Date: Fri, 16 Mar 2001 21:18:11 +0100
From: =?iso-8859-1?Q?Ragnar_Kj=F8rstad?= <xfs@ragnark.vestdata.no>
To: Steve Lord <lord@sgi.com>
Cc: Tom Duffy <tduffy@engr.sgi.com>, linux-xfs@oss.sgi.com
Subject: Re: filesystem locking?
Message-ID: <20010316211811.A2984@vestdata.no>
References: <tduffy@engr.sgi.com> <200103161924.f2GJO7l31603@jen.americas.sgi.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=iso-8859-1
Content-Transfer-Encoding: 8bit
X-Mailer: Mutt 0.95.5i
In-Reply-To: <200103161924.f2GJO7l31603@jen.americas.sgi.com>; from Steve Lord on Fri, Mar 16, 2001 at 01:24:07PM -0600
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

On Fri, Mar 16, 2001 at 01:24:07PM -0600, Steve Lord wrote:
> > I got this error running VMware on XFS (see attached jpg).
> > 
> > if xfs does support the file locking semantics that vmware wants (which I
> > assume), we should let them know so they can include xfs awareness in
> > their next release...
> 
> Hmm, so you think they probe the filesystem type and then decide they
> do not know what it can do if they do not recognize the type? They could
> try a couple of tests to see what works, even different revisions of 
> the same filesystem could have different feature sets.

Yes, that's exactly what they do.
(see long thread on reiserfs mailinglist on this subject long time ago -
only at that time vmware just crashed instead of telling you what to do)
 

-- 
Ragnar Kjørstad
Big Storage

From owner-linux-xfs@oss.sgi.com Fri Mar 16 12:33:20 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2GKXKu14483
	for linux-xfs-outgoing; Fri, 16 Mar 2001 12:33:20 -0800
Received: from sgi.com (sgi.SGI.COM [192.48.153.1])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2GKXKM14480
	for <linux-xfs@oss.sgi.com>; Fri, 16 Mar 2001 12:33:20 -0800
Received: from zeus-fddi.americas.sgi.com (128-162-8-103.americas.sgi.com [128.162.8.103]) 
	by sgi.com (980327.SGI.8.8.8-aspam/980304.SGI-aspam:
       SGI does not authorize the use of its proprietary
       systems or networks for unsolicited or bulk email
       from the Internet.) 
	via ESMTP id MAA08579
	for <linux-xfs@oss.sgi.com>; Fri, 16 Mar 2001 12:33:16 -0800 (PST)
	mail_from (roehrich@sgi.com)
Received: from daisy-e185.americas.sgi.com (daisy-e194.americas.sgi.com [128.162.194.214]) by zeus-fddi.americas.sgi.com (8.9.3/americas-smart-nospam1.1) with ESMTP id OAA1165540; Fri, 16 Mar 2001 14:31:08 -0600 (CST)
Received: from slobber.americas.sgi.com (slobber.americas.sgi.com [128.162.184.30]) by daisy-e185.americas.sgi.com (SGI-8.9.3/SGI-server-1.7) with ESMTP id OAA84867; Fri, 16 Mar 2001 14:31:08 -0600 (CST)
Received: from slobber.americas.sgi.com by slobber.americas.sgi.com (SGI-8.9.3/SGI-client-1.7) via ESMTP id OAA13270; Fri, 16 Mar 2001 14:31:08 -0600 (CST)
Message-Id: <200103162031.OAA13270@slobber.americas.sgi.com>
To: girish chandrashekara <girish_c_99@yahoo.com>
cc: linux-xfs@oss.sgi.com, openxdsm-devel@lists.sourceforge.net
Subject: Re: dmapi support in XFS 
Date: Fri, 16 Mar 2001 14:31:07 -0600
From: Dean Roehrich <roehrich@sgi.com>
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk


>From:  girish chandrashekara <girish_c_99@yahoo.com>
>Hi,
>  Iam working on dmapi support in VFS layer of linux
>to make it generic for underlying file systems. In the
>process I went through your source code
>for dmapi support in XFS(2.4.0-test5 kernel), I need
>some clarifications (correct me if Iam wrongly
>understood)

I'm absolutely positive that I must be misunderstanding what you
are asking for, but here goes...

>(1)Data events are triggered by the caller(through
>xfs_ioctls, xfs_dm_fcntl) except create(truncate
>event). How XFS manages (trigger implicitly-initiated
>by kernel)to send data event to DM application incase
>of read,write and mmap by user application on the
>managed region

DM_EVENT_READ comes from XFS this way:

xfs_read->xfs_dm_send_data_event->dm_send_data_event

In Irix XFS, and Linux XFS, we support just one managed region per file.

Mmap is not working yet.


>(2)Nospace event is also triggered through xfs_ioctl
>when filesystem  runs out of space. How XFS triggers
>nospace event implicitly, when filesystem runs out of
>space

DM_EVENT_NOSPACE comes in one of two ways:

xfs_alloc_file_space->dm_send_namesp_event
xfs_write->dm_send_namesp_event


Dean

From owner-linux-xfs@oss.sgi.com Fri Mar 16 21:26:19 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2H5QJn22045
	for linux-xfs-outgoing; Fri, 16 Mar 2001 21:26:19 -0800
Received: from saboche.muine.org (cr891907-a.slnt1.on.wave.home.com [24.156.26.33])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2H5QHM22036
	for <linux-xfs@oss.sgi.com>; Fri, 16 Mar 2001 21:26:17 -0800
Received: from avocado.muine.org (avocado.muine.org [192.168.1.12])
	by saboche.muine.org (Postfix) with ESMTP id 2872B99312
	for <linux-xfs@oss.sgi.com>; Sat, 17 Mar 2001 00:26:11 -0500 (EST)
Date: Sat, 17 Mar 2001 00:26:14 -0500 (EST)
From: "Hoang Q. Tran" <hoang@muine.org>
To: linux-xfs@oss.sgi.com
Subject: kernel compile error
Message-ID: <Pine.BSF.4.21.0103170015370.23820-100000@avocado.muine.org>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

Hi,

I'm fairly new and this is my first attempt compiling the
linux-2.4-xfs source checked out on March 16, 2001. The
box is a pre-release 0.9/rh7.

Thanks!

Hoang

Here's the snippet during make bzImage:

cc -D__KERNEL__ -I/usr/src/linux-2.4-xfs/linux/include  -Wall
-Wstrict-prototypes -O2 -fno-strict-aliasing -fomit-frame-pointer -pipe
-mpreferred-stack-boundary=2 -march=i586  -Wno-parentheses -I .. -I
/usr/src/linux-2.4-xfs/linux/fs -funsigned-char  -Wno-unknown-pragmas  -c
-o xfs_lrw.o xfs_lrw.c
xfs_lrw.c:1011:56: warning: pasting would not give a valid preprocessing
token
xfs_lrw.c:1134:62: warning: pasting would not give a valid preprocessing
token
xfs_lrw.c:1152:46: warning: pasting would not give a valid preprocessing
token
xfs_lrw.c: In function `xfs_iomap_write_direct':
xfs_lrw.c:1988: Unrecognizable insn:
...
xfs_lrw.c:1988: confused by earlier errors, bailing out
make[4]: *** [xfs_lrw.o] Error 1
make[4]: Leaving directory `/usr/src/linux-2.4-xfs/linux/fs/xfs/linux'
make[3]: *** [first_rule] Error 2


From owner-linux-xfs@oss.sgi.com Fri Mar 16 21:36:54 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2H5as822140
	for linux-xfs-outgoing; Fri, 16 Mar 2001 21:36:54 -0800
Received: from deliverator.sgi.com (deliverator.sgi.com [204.94.214.10])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2H5asM22137
	for <linux-xfs@oss.sgi.com>; Fri, 16 Mar 2001 21:36:54 -0800
Received: from madurai.engr.sgi.com (madurai.engr.sgi.com [163.154.5.75]) by deliverator.sgi.com (980309.SGI.8.8.8-aspam-6.2/980310.SGI-aspam) via ESMTP id VAA22401
	for <linux-xfs@oss.sgi.com>; Fri, 16 Mar 2001 21:35:42 -0800 (PST)
	mail_from (ananth@sgi.com)
Received: from sgi.com (sgigate.sgi.com [198.29.75.75]) by madurai.engr.sgi.com (980427.SGI.8.8.8/970903.SGI.AUTOCF) via ESMTP id VAA21341; Fri, 16 Mar 2001 21:31:12 -0800 (PST)
Message-ID: <3AB2F770.D5A1EB99@sgi.com>
Date: Fri, 16 Mar 2001 21:34:40 -0800
From: Rajagopal Ananthanarayanan <ananth@sgi.com>
X-Mailer: Mozilla 4.6 [en] (X11; I; Linux 2.2.10-1SGI_17 i686)
X-Accept-Language: en
MIME-Version: 1.0
To: "Hoang Q. Tran" <hoang@muine.org>
CC: linux-xfs@oss.sgi.com
Subject: Re: kernel compile error
References: <Pine.BSF.4.21.0103170015370.23820-100000@avocado.muine.org>
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

"Hoang Q. Tran" wrote:
> 
> Hi,
> 
> I'm fairly new and this is my first attempt compiling the
> linux-2.4-xfs source checked out on March 16, 2001. The
> box is a pre-release 0.9/rh7.

This is likely a kgcc issue. Look at the top level linux Makefile,
and make these changes:

----------
#comment out this line if compiling on RH 7.0
#CC             = $(CROSS_COMPILE)gcc -V egcs-2.91.66
# AND uncomment the following line
CC              = $(CROSS_COMPILE)kgcc
--------------


> 
> Thanks!
> 
> Hoang
> 
> Here's the snippet during make bzImage:
> 
> cc -D__KERNEL__ -I/usr/src/linux-2.4-xfs/linux/include  -Wall
> -Wstrict-prototypes -O2 -fno-strict-aliasing -fomit-frame-pointer -pipe
> -mpreferred-stack-boundary=2 -march=i586  -Wno-parentheses -I .. -I
> /usr/src/linux-2.4-xfs/linux/fs -funsigned-char  -Wno-unknown-pragmas  -c
> -o xfs_lrw.o xfs_lrw.c
> xfs_lrw.c:1011:56: warning: pasting would not give a valid preprocessing
> token
> xfs_lrw.c:1134:62: warning: pasting would not give a valid preprocessing
> token
> xfs_lrw.c:1152:46: warning: pasting would not give a valid preprocessing
> token
> xfs_lrw.c: In function `xfs_iomap_write_direct':
> xfs_lrw.c:1988: Unrecognizable insn:
> ...
> xfs_lrw.c:1988: confused by earlier errors, bailing out
> make[4]: *** [xfs_lrw.o] Error 1
> make[4]: Leaving directory `/usr/src/linux-2.4-xfs/linux/fs/xfs/linux'
> make[3]: *** [first_rule] Error 2

-- 
--------------------------------------------------------------------------
Rajagopal Ananthanarayanan ("ananth")
Member Technical Staff, SGI.
--------------------------------------------------------------------------

From owner-linux-xfs@oss.sgi.com Fri Mar 16 23:55:29 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2H7tTj23398
	for linux-xfs-outgoing; Fri, 16 Mar 2001 23:55:29 -0800
Received: from deliverator.sgi.com (deliverator.sgi.com [204.94.214.10])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2H7tTM23395
	for <linux-xfs@oss.sgi.com>; Fri, 16 Mar 2001 23:55:29 -0800
Received: from snort.melbourne.sgi.com (snort.melbourne.sgi.com [134.14.55.149]) by deliverator.sgi.com (980309.SGI.8.8.8-aspam-6.2/980310.SGI-aspam) via ESMTP id XAA03833
	for <linux-xfs@oss.sgi.com>; Fri, 16 Mar 2001 23:54:16 -0800 (PST)
	mail_from (nathans@snort.melbourne.sgi.com)
Received: (from nathans@localhost)
	by snort.melbourne.sgi.com (SGI-8.9.3/8.9.3) id SAA15472;
	Sat, 17 Mar 2001 18:54:07 +1100 (EST)
Date: Sat, 17 Mar 2001 18:54:07 +1100 (EST)
From: Nathan Scott <nathans@snort.melbourne.sgi.com>
Message-Id: <200103170754.SAA15472@snort.melbourne.sgi.com>
To: linux-xfs@oss.sgi.com
Cc: tduffy@engr.sgi.com, jalal@engr.sgi.com
Subject: TAKE - quota
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

Should fix the ia64 build problem you're seeing, Tom/Jalal.
Let me know if there are further issues here -- thanks.

cheers.


Date:  Fri Mar 16 23:47:10 PST 2001
Workarea:  snort.melbourne.sgi.com:/diskb/build4/nathans/linux-xfs

The following file(s) were checked into:
  bonnie.engr.sgi.com:/isms/slinx/2.4.x-xfs


Modid:  2.4.x-xfs:slinx:90286a
cmd/quota/quotaops.c - 1.10
cmd/quota/repquota.c - 1.16
	- don't extern stuff that should be declared in libc headers.



From owner-linux-xfs@oss.sgi.com Sat Mar 17 07:47:01 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2HFl1H28057
	for linux-xfs-outgoing; Sat, 17 Mar 2001 07:47:01 -0800
Received: from wisdom.myplace.net (cc19815-a.zwoll1.ov.nl.home.com [212.204.138.247])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2HFkYM28051
	for <linux-xfs@oss.sgi.com>; Sat, 17 Mar 2001 07:46:36 -0800
Received: from ws1 (ws1.myplace.net [192.168.1.15])
	by wisdom.myplace.net (Postfix) with SMTP id 2254E14B
	for <linux-xfs@oss.sgi.com>; Sat, 17 Mar 2001 16:45:14 +0100 (CET)
Message-ID: <008501c0af44$b15be340$0f01a8c0@myplace.net>
From: "Bas" <list@showme.wox.org>
To: <linux-xfs@oss.sgi.com>
Subject: Damaged filesystem causes oops ?!
Date: Sat, 17 Mar 2001 16:45:13 -0800
MIME-Version: 1.0
Content-Type: multipart/mixed;
	boundary="----=_NextPart_000_0082_01C0AF01.A327DA60"
X-Priority: 3
X-MSMail-Priority: Normal
X-Mailer: Microsoft Outlook Express 5.00.2615.200
X-MimeOLE: Produced By Microsoft MimeOLE V5.00.2615.200
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

This is a multi-part message in MIME format.

------=_NextPart_000_0082_01C0AF01.A327DA60
Content-Type: text/plain;
	charset="iso-8859-1"
Content-Transfer-Encoding: 7bit

Hi,

I use Linux 2.4.2 with XFS support build from the CVS sources which were
available on march 13th. One of the filesystems seemed to be damaged, and
when I rebooted it caused an oops. This oops was there with my previous
kernel too, and made it impossible to mount anything after that, but I was
able to use the system however, whenever I wanted to xfs_repair the LV it
was on, it stopped with a D state.

I included the dmesg output.

But when I rebooted using a rescue cd, repaired the filesystem everything
was and still is working fine.

Thanks,

Bas.




------=_NextPart_000_0082_01C0AF01.A327DA60
Content-Type: application/octet-stream;
	name="oops.out"
Content-Transfer-Encoding: quoted-printable
Content-Disposition: attachment;
	filename="oops.out"

ksymoops 0.7c on i686 2.4.2-XFS.  Options used=0A=
     -V (default)=0A=
     -k /lib/modules/2.4.2-3/ (specified)=0A=
     -l /proc/modules (default)=0A=
     -o /lib/modules/2.4.2-XFS/ (default)=0A=
     -m /boot/System.map-2.4.2-3 (specified)=0A=
=0A=
Error (regular_file): read_ksyms /lib/modules/2.4.2-3/ is not a regular =
file, ignored=0A=
No modules in ksyms, skipping objects=0A=
No ksyms, skipping lsmod=0A=
invalid operand: 0000=0A=
CPU:    0=0A=
EIP:    0010:[<c015dc40>]=0A=
Using defaults from ksymoops -t elf32-i386 -a i386=0A=
EFLAGS: 00010282=0A=
eax: 0000001e   ebx: 02082225   ecx: 00000001   edx: 00000001=0A=
esi: c140d5cc   edi: 00202205   ebp: 00000000   esp: cf635850=0A=
ds: 0018   es: 0018   ss: 0018=0A=
Process mount (pid: 95, stackpage=3Dcf635000)=0A=
Stack: c026caa5 c026cbe7 0000025a cf491680 00002000 049ae000 00000000 =
c140d5cc =0A=
       cfff6280 cff70d28 c140d5a4 00000043 00000000 00000001 000049b0 =
00000002 =0A=
       cf7d6160 00000000 00000002 c015de2e cf5e03a0 00202205 00000000 =
00000000 =0A=
Call Trace: [<c015de2e>] [<c01ad81d>] [<c01ae4a5>] [<c01ae54b>] =
[<c01ae688>] [<c01aefae>] [<c01af415>] =0A=
       [<c01af45e>] [<c01af5d8>] [<c01a94a2>] [<c01b0cc6>] [<c0113428>] =
[<c018dd50>] [<c0107a79>] [<e04de742>] =0A=
       [<c01afc1b>] [<c01b90e7>] [<c01b92c3>] [<c01b92f3>] [<c01cac5a>] =
[<c01c0002>] [<c01cb8ed>] [<c01a0029>] =0A=
       [<c0142b3f>] [<c0143bc6>] [<c01377a4>] [<c01352fc>] [<c013553a>] =
[<c013608b>] [<c0135eb4>] [<c013626c>] =0A=
       [<c0108e97>] =0A=
Code: 0f 0b 83 c4 0c 83 7c 24 24 00 74 1b 8b 5c 24 44 8b 43 08 24 =0A=
=0A=
>>EIP; c015dc40 <_pagebuf_lookup_pages+3e8/41c>   <=3D=3D=3D=3D=3D=0A=
Trace; c015de2e <pagebuf_get+162/218>=0A=
Trace; c01ad81d <xlog_recover_do_buffer_trans+125/294>=0A=
Trace; c01ae4a5 <xlog_recover_do_trans+f9/10c>=0A=
Trace; c01ae54b <xlog_recover_commit_trans+27/3c>=0A=
Trace; c01ae688 <xlog_recover_process_data+114/19c>=0A=
Trace; c01aefae <xlog_do_recovery_pass+1b6/5a8>=0A=
Trace; c01af415 <xlog_do_log_recovery+75/9c>=0A=
Trace; c01af45e <xlog_do_recover+22/f8>=0A=
Trace; c01af5d8 <xlog_recover+a4/d0>=0A=
Trace; c01a94a2 <xfs_log_mount+76/a4>=0A=
Trace; c01b0cc6 <xfs_mountfs+cf2/11d8>=0A=
Trace; c0113428 <schedule+2d8/430>=0A=
Trace; c018dd50 <xfs_dir2_mount+0/120>=0A=
Trace; c0107a79 <__down+a1/ac>=0A=
Trace; e04de742 <END_OF_CODE+201d46d2/????>=0A=
Trace; c01afc1b <xfs_readsb+af/c8>=0A=
Trace; c01b90e7 <xfs_cmountfs+517/584>=0A=
Trace; c01b92c3 <xfs_mount+a7/b4>=0A=
Trace; c01b92f3 <xfs_vfsmount+23/38>=0A=
Trace; c01cac5a <linvfs_read_super+1ea/2d0>=0A=
Trace; c01c0002 <xfs_symlink+692/d10>=0A=
Trace; c01cb8ed <vn_insert_in_linux_hash+19/20>=0A=
Trace; c01a0029 <xfs_inobt_lookup+2a9/328>=0A=
Trace; c0142b3f <destroy_inode+2f/34>=0A=
Trace; c0143bc6 <iput+146/14c>=0A=
Trace; c01377a4 <blkdev_get+108/128>=0A=
Trace; c01352fc <read_super+100/170>=0A=
Trace; c013553a <get_sb_bdev+16e/1c8>=0A=
Trace; c013608b <do_mount+187/2e4>=0A=
Trace; c0135eb4 <copy_mount_options+50/a0>=0A=
Trace; c013626c <sys_mount+84/c8>=0A=
Trace; c0108e97 <system_call+33/38>=0A=
Code;  c015dc40 <_pagebuf_lookup_pages+3e8/41c>=0A=
00000000 <_EIP>:=0A=
Code;  c015dc40 <_pagebuf_lookup_pages+3e8/41c>   <=3D=3D=3D=3D=3D=0A=
   0:   0f 0b                     ud2a      <=3D=3D=3D=3D=3D=0A=
Code;  c015dc42 <_pagebuf_lookup_pages+3ea/41c>=0A=
   2:   83 c4 0c                  add    $0xc,%esp=0A=
Code;  c015dc45 <_pagebuf_lookup_pages+3ed/41c>=0A=
   5:   83 7c 24 24 00            cmpl   $0x0,0x24(%esp,1)=0A=
Code;  c015dc4a <_pagebuf_lookup_pages+3f2/41c>=0A=
   a:   74 1b                     je     27 <_EIP+0x27> c015dc67 =
<_pagebuf_lookup_pages+40f/41c>=0A=
Code;  c015dc4c <_pagebuf_lookup_pages+3f4/41c>=0A=
   c:   8b 5c 24 44               mov    0x44(%esp,1),%ebx=0A=
Code;  c015dc50 <_pagebuf_lookup_pages+3f8/41c>=0A=
  10:   8b 43 08                  mov    0x8(%ebx),%eax=0A=
Code;  c015dc53 <_pagebuf_lookup_pages+3fb/41c>=0A=
  13:   24 00                     and    $0x0,%al=0A=
=0A=
=0A=
1 error issued.  Results may not be reliable.=0A=

------=_NextPart_000_0082_01C0AF01.A327DA60--


From owner-linux-xfs@oss.sgi.com Sat Mar 17 07:56:19 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2HFuJh28397
	for linux-xfs-outgoing; Sat, 17 Mar 2001 07:56:19 -0800
Received: from saboche.muine.org (cr891907-a.slnt1.on.wave.home.com [24.156.26.33])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2HFuHM28394
	for <linux-xfs@oss.sgi.com>; Sat, 17 Mar 2001 07:56:18 -0800
Received: from avocado.muine.org (avocado.muine.org [192.168.1.12])
	by saboche.muine.org (Postfix) with ESMTP id 4D92E99312
	for <linux-xfs@oss.sgi.com>; Sat, 17 Mar 2001 10:56:11 -0500 (EST)
Date: Sat, 17 Mar 2001 10:56:15 -0500 (EST)
From: "Hoang Q. Tran" <hoang@muine.org>
To: linux-xfs@oss.sgi.com
Subject: Re: kernel compile error
In-Reply-To: <3AB2F770.D5A1EB99@sgi.com>
Message-ID: <Pine.BSF.4.21.0103171044250.27584-100000@avocado.muine.org>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

Hi,

Thanks for the quick reply.

I did comment and uncomment the top level Makefile and for
some reason, make doesn't find kgcc. I've found a link regarding
to 2.96. Not sure if it's related to my problem.

http://www.fsf.org/software/gcc/gcc-2.96.html

CC in top level Makefile:

----------------------------------------------------
#comment out this line if compiling on RH 7.0
#CC             = $(CROSS_COMPILE)gcc -V egcs-2.91.66
# AND uncomment the following line
CC              = $(CROSS_COMPILE)kgcc
-----------------------------------------------------

make bzImage error:

[root@blacksheep linux]# make bzImage
gcc -Wall -Wstrict-prototypes -O2 -fomit-frame-pointer -o
scripts/split-include scripts/split-include.c
scripts/split-include include/linux/autoconf.h include/config
kgcc  -D__KERNEL__ -I/usr/src/linux-2.4-xfs/linux/include  -Wall
-Wstrict-prototypes -O2 -fno-strict-aliasing -fomit-frame-pointer -pipe
-march=i586   -c -o init/main.o init/main.c
make: kgcc: Command not found
make: *** [init/main.o] Error 127


From owner-linux-xfs@oss.sgi.com Sat Mar 17 08:53:14 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2HGrEB29063
	for linux-xfs-outgoing; Sat, 17 Mar 2001 08:53:14 -0800
Received: from sttlpop4.sttl.uswest.net (mail.sttl.uswest.net [206.81.192.4])
	by oss.sgi.com (8.11.3/8.11.3) with SMTP id f2HGrCM29060
	for <linux-xfs@oss.sgi.com>; Sat, 17 Mar 2001 08:53:13 -0800
Received: (qmail 32432 invoked by alias); 17 Mar 2001 16:53:06 -0000
Delivered-To: fixup-linux-xfs@oss.sgi.com@fixme
Received: (qmail 32427 invoked by uid 0); 17 Mar 2001 16:53:06 -0000
Received: from unknown (HELO walt400) (63.226.231.5)
  by mail.sttl.uswest.net with SMTP; 17 Mar 2001 16:53:06 -0000
Message-ID: <000c01c0af03$4cbd2200$0200000a@walt400>
From: "Walt H" <waltabbyh@uswest.net>
To: <linux-xfs@oss.sgi.com>
Subject: FS Corruption and Repair Path
Date: Sat, 17 Mar 2001 08:57:03 -0800
MIME-Version: 1.0
Content-Type: multipart/alternative;
	boundary="----=_NextPart_000_0009_01C0AEC0.3C7183C0"
X-Priority: 3
X-MSMail-Priority: Normal
X-Mailer: Microsoft Outlook Express 5.50.4133.2400
X-MimeOLE: Produced By Microsoft MimeOLE V5.50.4133.2400
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

This is a multi-part message in MIME format.

------=_NextPart_000_0009_01C0AEC0.3C7183C0
Content-Type: text/plain;
	charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable

Hello,

I've been using XFS for a couple months now with very good results. =
However, last night had a couple of weird things happen. I pulled latest =
CVS and built a new kernel. In the process of shutting down the system =
for reboot into new kernel, the shutdown scripts stopped executing. The =
system was still up, however, and I was able to sync and emergency =
remount read-only prior to rebooting. The system successfully booted =
into the new kernel and all appeared well. After some time, another =
process got stuck so I proceeded to shutdown and boot off my repair disk =
to see if anything was wrong. xfs_check showed inconsistencies in the =
root FS, so I attempted to repair using xfs_repair. A number of problems =
existed and it attempted to repair them eventually failing with:

FATAL ERROR - XFS_CHECK    DUPLICATE INODE RANGE

(or something very close - typing up message from a diff machine)

No amount of fiddling on my part has been successful in getting past =
this error. I can still successfully mount this partition, however, =
there are damaged directory entries throughout. All the damaged entries =
are named "/ome" or some such where the first letter, "h" in this case =
is replaced with a leading slash. I've tried using xfs_db to step =
through it as well as xfs_ncheck to determine the files at this inode =
location (127069055) but can't find anything. Worst case, I can reformat =
this partition and reinstall, but would like to know what causes this =
and how to avoid it. This machine has locked up a couple times as of =
late (nvidia card w/ latest drivers, not so good) - Most of the time, =
I've been paranoid due to my newness to XFS and have fsck'd the FS from =
my repair boot disk w/o any errors showing up. Not sure what caused =
this?

Machine info:
P2 w/ 384MB ram, 2 IDE drives - partitioned as follows:

Maxtor 11.5GB - 4 partittions, connected to builtin IDE
hda1 =3D vfat32   approx. 7.5 GB
hda2 =3D ext2 (boot)  300 MB
hda3 =3D swap  250 MB
hda4 =3D xfs (/root)  remainder of drive

IBM Deskstar 15GB - 3 Partitions, connected to Promise UDMA 66
hde1 =3D vfat32  2 GB
hde2 =3D xfs (/home)  5 GB
hde3 =3D xfs (/usr)  remainder of drive

Interestingly, hde2 has no problems - and hde3 reports similar =
inconsistencies to hda4. I haven't attempted to repair it yet.
Any help would be greatly appreciated. Thanks.

Walt

------=_NextPart_000_0009_01C0AEC0.3C7183C0
Content-Type: text/html;
	charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable

<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.0 Transitional//EN">
<HTML><HEAD>
<META http-equiv=3DContent-Type content=3D"text/html; =
charset=3Diso-8859-1">
<META content=3D"MSHTML 5.50.4611.1300" name=3DGENERATOR>
<STYLE></STYLE>
</HEAD>
<BODY bgColor=3D#ffffff>
<DIV><FONT size=3D2>Hello,</FONT></DIV>
<DIV><FONT size=3D2></FONT>&nbsp;</DIV>
<DIV><FONT size=3D2>I've been using XFS for a couple months now with =
very good=20
results. However, last night had a couple of weird things happen. I =
pulled=20
latest CVS and built a new kernel. In the process of shutting down the =
system=20
for reboot into new kernel, the shutdown scripts stopped executing. The =
system=20
was still up, however, and I was able to sync and emergency remount =
read-only=20
prior to rebooting. The system successfully booted into the new kernel =
and all=20
appeared well. After some time, another process got stuck so I proceeded =
to=20
shutdown and boot off my repair disk to see if anything was wrong. =
xfs_check=20
showed inconsistencies in the root FS, so I attempted to repair using=20
xfs_repair. A number of problems existed and it attempted to repair them =

eventually failing with:</FONT></DIV>
<DIV><FONT size=3D2></FONT>&nbsp;</DIV>
<DIV><FONT size=3D2>FATAL ERROR - XFS_CHECK&nbsp;&nbsp;&nbsp; DUPLICATE =
INODE=20
RANGE</FONT></DIV>
<DIV><FONT size=3D2></FONT>&nbsp;</DIV>
<DIV><FONT size=3D2>(or something very close - typing up message from a =
diff=20
machine)</FONT></DIV>
<DIV><FONT size=3D2></FONT>&nbsp;</DIV>
<DIV><FONT size=3D2>No amount of fiddling on my part has been successful =
in=20
getting past this error. I can still successfully mount this partition, =
however,=20
there are damaged directory entries throughout. All the damaged entries =
are=20
named "/ome" or some such where the first letter, "h" in this case is =
replaced=20
with a leading slash. I've tried using xfs_db to step through it as well =
as=20
xfs_ncheck to determine the files at this inode location (127069055) but =
can't=20
find anything. Worst case, I can reformat this partition and reinstall, =
but=20
would like to know what causes this and how to avoid it. This machine =
has locked=20
up a couple times as of late (nvidia card w/ latest drivers, not so =
good) - Most=20
of the time, I've been paranoid due to my newness to XFS and have fsck'd =
the FS=20
from my repair boot disk w/o any errors showing up. Not sure what caused =

this?</FONT></DIV>
<DIV><FONT size=3D2></FONT>&nbsp;</DIV>
<DIV><FONT size=3D2>Machine info:</FONT></DIV>
<DIV><FONT size=3D2>P2 w/ 384MB ram, 2 IDE drives - partitioned as=20
follows:</FONT></DIV>
<DIV><FONT size=3D2></FONT>&nbsp;</DIV>
<DIV><FONT size=3D2>Maxtor 11.5GB - 4 partittions, connected to builtin=20
IDE</FONT></DIV>
<DIV><FONT size=3D2>hda1 =3D vfat32&nbsp;&nbsp; approx. 7.5 =
GB</FONT></DIV>
<DIV><FONT size=3D2>hda2 =3D ext2 (boot)&nbsp; 300 MB</FONT></DIV>
<DIV><FONT size=3D2>hda3 =3D swap&nbsp; 250 MB</FONT></DIV>
<DIV><FONT size=3D2>hda4 =3D xfs (/root)&nbsp; remainder of =
drive</FONT></DIV>
<DIV><FONT size=3D2></FONT>&nbsp;</DIV>
<DIV><FONT size=3D2>IBM Deskstar 15GB - 3 Partitions, connected to =
Promise UDMA=20
66</FONT></DIV>
<DIV><FONT size=3D2>hde1 =3D vfat32&nbsp; 2 GB</FONT></DIV>
<DIV><FONT size=3D2>hde2 =3D xfs (/home)&nbsp; 5 GB</FONT></DIV>
<DIV><FONT size=3D2>hde3 =3D xfs (/usr)&nbsp; remainder of =
drive</FONT></DIV>
<DIV><FONT size=3D2></FONT>&nbsp;</DIV>
<DIV><FONT size=3D2>Interestingly, hde2 has no problems - and hde3 =
reports similar=20
inconsistencies to hda4. I haven't attempted to repair it =
yet.</FONT></DIV>
<DIV><FONT size=3D2>Any help would be greatly appreciated. =
Thanks.</FONT></DIV>
<DIV><FONT size=3D2></FONT>&nbsp;</DIV>
<DIV><FONT size=3D2>Walt</FONT></DIV></BODY></HTML>

------=_NextPart_000_0009_01C0AEC0.3C7183C0--


From owner-linux-xfs@oss.sgi.com Sat Mar 17 09:07:25 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2HH7Pl29227
	for linux-xfs-outgoing; Sat, 17 Mar 2001 09:07:25 -0800
Received: from Cantor.suse.de (ns.suse.de [213.95.15.193])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2HH7OM29224
	for <linux-xfs@oss.sgi.com>; Sat, 17 Mar 2001 09:07:24 -0800
Received: from Hermes.suse.de (Hermes.suse.de [213.95.15.136])
	by Cantor.suse.de (Postfix) with ESMTP
	id 8518B1E226; Sat, 17 Mar 2001 18:07:23 +0100 (MET)
Date: Sat, 17 Mar 2001 18:07:13 +0100
From: Andi Kleen <ak@suse.de>
To: Steve Lord <lord@sgi.com>
Cc: Tom Duffy <tduffy@engr.sgi.com>, linux-xfs@oss.sgi.com
Subject: Re: filesystem locking?
Message-ID: <20010317180713.A24505@gruyere.muc.suse.de>
References: <tduffy@engr.sgi.com> <200103161924.f2GJO7l31603@jen.americas.sgi.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.2.5i
In-Reply-To: <200103161924.f2GJO7l31603@jen.americas.sgi.com>; from lord@sgi.com on Fri, Mar 16, 2001 at 01:24:07PM -0600
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

On Fri, Mar 16, 2001 at 01:24:07PM -0600, Steve Lord wrote:
> 
> > 
> > I got this error running VMware on XFS (see attached jpg).
> > 
> > if xfs does support the file locking semantics that vmware wants (which I
> > assume), we should let them know so they can include xfs awareness in
> > their next release...
> > 
> > -tduffy
> > 
> 
> Hmm, so you think they probe the filesystem type and then decide they
> do not know what it can do if they do not recognize the type? They could
> try a couple of tests to see what works, even different revisions of 
> the same filesystem could have different feature sets.

They indeed do that. Another software that also does this unfortuate thing
is the Arkeia backup client.


-Andi

From owner-linux-xfs@oss.sgi.com Sat Mar 17 09:26:11 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2HHQBe29423
	for linux-xfs-outgoing; Sat, 17 Mar 2001 09:26:11 -0800
Received: from mail15.jump.net (mail15.jump.net [206.196.91.15])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2HHQAM29420
	for <linux-xfs@oss.sgi.com>; Sat, 17 Mar 2001 09:26:10 -0800
Received: from sgi.com (aus-dsl-dhcp1-26.customer.jump.net [63.163.168.26])
	by mail15.jump.net (8.10.2/) with ESMTP id f2HHPp302832;
	Sat, 17 Mar 2001 11:25:51 -0600 (CST)
Message-ID: <3AB39E51.EE8334B4@sgi.com>
Date: Sat, 17 Mar 2001 11:26:41 -0600
From: Eric Sandeen <sandeen@sgi.com>
X-Mailer: Mozilla 4.76 [en] (X11; U; Linux 2.4.2-XFS i686)
X-Accept-Language: en
MIME-Version: 1.0
To: "Hoang Q. Tran" <hoang@muine.org>
CC: linux-xfs@oss.sgi.com
Subject: Re: kernel compile error
References: <Pine.BSF.4.21.0103171044250.27584-100000@avocado.muine.org>
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

"Hoang Q. Tran" wrote:

> I did comment and uncomment the top level Makefile and for
> some reason, make doesn't find kgcc. 

Hm... if this is a RH7 box it should be there, but maybe not if you
didn't select "kernel development" when you installed.

kgcc is just an older (stable) version of gcc, and you can get it from

ftp://ftp.rpmfind.net/linux/redhat/redhat-7.0/i386/en/RedHat/RPMS//kgcc-1.1.2-40.i386.rpm

or any Red Hat mirror...

> I've found a link regarding
> to 2.96. Not sure if it's related to my problem.
> 
> http://www.fsf.org/software/gcc/gcc-2.96.html

Yep, this is the FSF complaining about Red Hat releasing a snapshot of
gcc, which won't build the kernel - as you found out.  :)

-Eric

From owner-linux-xfs@oss.sgi.com Sat Mar 17 11:42:07 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2HJg7C31305
	for linux-xfs-outgoing; Sat, 17 Mar 2001 11:42:07 -0800
Received: from saboche.muine.org (cr891907-a.slnt1.on.wave.home.com [24.156.26.33])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2HJg6M31302
	for <linux-xfs@oss.sgi.com>; Sat, 17 Mar 2001 11:42:06 -0800
Received: from avocado.muine.org (avocado.muine.org [192.168.1.12])
	by saboche.muine.org (Postfix) with ESMTP
	id EA16199312; Sat, 17 Mar 2001 14:41:59 -0500 (EST)
Date: Sat, 17 Mar 2001 14:41:59 -0500 (EST)
From: "Hoang Q. Tran" <hoang@muine.org>
To: Eric Sandeen <sandeen@sgi.com>
Cc: linux-xfs@oss.sgi.com
Subject: Re: kernel compile error
In-Reply-To: <3AB39E51.EE8334B4@sgi.com>
Message-ID: <Pine.BSF.4.21.0103171441040.28036-100000@avocado.muine.org>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

Installed kgcc fixed and got the kernel working.

Thanks!

Hoang

On Sat, 17 Mar 2001, Eric Sandeen wrote:

> "Hoang Q. Tran" wrote:
> 
> > I did comment and uncomment the top level Makefile and for
> > some reason, make doesn't find kgcc. 
> 
> Hm... if this is a RH7 box it should be there, but maybe not if you
> didn't select "kernel development" when you installed.
> 
> kgcc is just an older (stable) version of gcc, and you can get it from
> 
> ftp://ftp.rpmfind.net/linux/redhat/redhat-7.0/i386/en/RedHat/RPMS//kgcc-1.1.2-40.i386.rpm
> 
> or any Red Hat mirror...
> 
> > I've found a link regarding
> > to 2.96. Not sure if it's related to my problem.
> > 
> > http://www.fsf.org/software/gcc/gcc-2.96.html
> 
> Yep, this is the FSF complaining about Red Hat releasing a snapshot of
> gcc, which won't build the kernel - as you found out.  :)
> 
> -Eric
> 
> 


From owner-linux-xfs@oss.sgi.com Sat Mar 17 16:30:51 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2I0Up003643
	for linux-xfs-outgoing; Sat, 17 Mar 2001 16:30:51 -0800
Received: from sgi.com (sgi.SGI.COM [192.48.153.1])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2I0UoM03640
	for <linux-xfs@oss.sgi.com>; Sat, 17 Mar 2001 16:30:50 -0800
Received: from waco.engr.sgi.com ([163.154.18.95]) 
	by sgi.com (980327.SGI.8.8.8-aspam/980304.SGI-aspam:
       SGI does not authorize the use of its proprietary
       systems or networks for unsolicited or bulk email
       from the Internet.) 
	via ESMTP id QAA03700
	for <linux-xfs@oss.sgi.com>; Sat, 17 Mar 2001 16:30:50 -0800 (PST)
	mail_from (ananth@waco.engr.sgi.com)
Received: (from ananth@localhost)
	by waco.engr.sgi.com (8.11.0/8.11.0) id f2I0UBG06280
	for linux-xfs@oss.sgi.com; Sat, 17 Mar 2001 16:30:11 -0800
Date: Sat, 17 Mar 2001 16:30:11 -0800
From: Ananth Ananthanarayanan <ananth@waco.engr.sgi.com>
Message-Id: <200103180030.f2I0UBG06280@waco.engr.sgi.com>
Subject: TAKE - handle vn_alloc failure
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

Fixes a panic found during stress tests (dbench).

Date:  Sat Mar 17 16:26:19 PST 2001
Workarea:  waco.engr.sgi.com:/build1/ananth/xfs-tot

The following file(s) were checked into:
  bonnie.engr.sgi.com:/isms/slinx/2.4.x-xfs


Modid:  2.4.x-xfs:slinx:90296a
linux/fs/xfs/xfs_iget.c - 1.131
linux/fs/xfs/linux/xfs_vnode.c - 1.50
	- Handle out of memory conditions on vn_alloc.



From owner-linux-xfs@oss.sgi.com Sat Mar 17 16:42:50 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2I0goq03778
	for linux-xfs-outgoing; Sat, 17 Mar 2001 16:42:50 -0800
Received: from sgi.com (sgi.SGI.COM [192.48.153.1])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2I0gmM03775
	for <linux-xfs@oss.sgi.com>; Sat, 17 Mar 2001 16:42:49 -0800
Received: from madurai.engr.sgi.com ([163.154.5.75]) 
	by sgi.com (980327.SGI.8.8.8-aspam/980304.SGI-aspam:
       SGI does not authorize the use of its proprietary
       systems or networks for unsolicited or bulk email
       from the Internet.) 
	via ESMTP id QAA05212
	for <linux-xfs@oss.sgi.com>; Sat, 17 Mar 2001 16:42:36 -0800 (PST)
	mail_from (ananth@sgi.com)
Received: from sgi.com (sgigate.sgi.com [198.29.75.75]) by madurai.engr.sgi.com (980427.SGI.8.8.8/970903.SGI.AUTOCF) via ESMTP id QAA22164; Sat, 17 Mar 2001 16:36:44 -0800 (PST)
Message-ID: <3AB403ED.C1E707DF@sgi.com>
Date: Sat, 17 Mar 2001 16:40:13 -0800
From: Rajagopal Ananthanarayanan <ananth@sgi.com>
X-Mailer: Mozilla 4.6 [en] (X11; I; Linux 2.2.10-1SGI_17 i686)
X-Accept-Language: en
MIME-Version: 1.0
To: Bas <list@showme.wox.org>
CC: linux-xfs@oss.sgi.com
Subject: Re: Damaged filesystem causes oops ?!
References: <008501c0af44$b15be340$0f01a8c0@myplace.net>
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

Bas wrote:
> 
> Hi,
> 
> I use Linux 2.4.2 with XFS support build from the CVS sources which were
> available on march 13th. One of the filesystems seemed to be damaged, and
> when I rebooted it caused an oops. This oops was there with my previous
> kernel too, and made it impossible to mount anything after that, but I was
> able to use the system however, whenever I wanted to xfs_repair the LV it
> was on, it stopped with a D state.
> 
> I included the dmesg output.


I see that you hit an "ud2a" which is almost always part of some BUG().
Can you give a line # that was printed in the BUG() message on console?

Also, there were some recent bug fixes to  the recovery code, IIRC; perhaps you
can try with a current CVS kernel?

> 
> But when I rebooted using a rescue cd, repaired the filesystem everything
> was and still is working fine.
> 
> Thanks,
> 
> Bas.
> 
>   ------------------------------------------------------------------------------------------------------------------------------------
>                Name: oops.out
>    oops.out    Type: unspecified type (application/octet-stream)
>            Encoding: quoted-printable

-- 
--------------------------------------------------------------------------
Rajagopal Ananthanarayanan ("ananth")
Member Technical Staff, SGI.
--------------------------------------------------------------------------

From owner-linux-xfs@oss.sgi.com Sat Mar 17 17:37:06 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2I1b6m04388
	for linux-xfs-outgoing; Sat, 17 Mar 2001 17:37:06 -0800
Received: from horus.bticc.net (IDENT:root@horus.bticc.net [216.196.51.155])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2I1b4M04385
	for <linux-xfs@oss.sgi.com>; Sat, 17 Mar 2001 17:37:04 -0800
Received: from horus.bticc.net (IDENT:jkinsley@horus.bticc.net [216.196.51.155])
	by horus.bticc.net (8.9.3/8.9.3) with ESMTP id UAA08216;
	Sat, 17 Mar 2001 20:36:51 -0500
Date: Sat, 17 Mar 2001 20:36:51 -0500 (EST)
From: J Kinsley <jkinsley@bticc.net>
To: <proftpd-users@proftpd.org>, <proftpd-devel@proftpd.org>,
   <linux-xfs@oss.sgi.com>
cc: Steve Lord <lord@sgi.com>
Subject: Re: ProFTPd + linux-2.4.2-XFS = RETR bug
In-Reply-To: <Pine.LNX.4.30.0103160413420.27608-100000@horus.bticc.net>
Message-ID: <Pine.LNX.4.30.0103172024120.8147-100000@horus.bticc.net>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

Although the bug reported only showed up when using ProFTPD, it was
indeed an XFS bug.  While the patch Steve uploaded yesterday did not
apply cleanly, I compiled the latest linux-2.4-xfs from CVS this
morning.  After twelve hours of uptime, I have noticed no zero length
files in the FTP server logs, and all test tansfers have been
successful.

Thanks to those who replied to the initial report...

Regards,
Jarrod Kinsley




From owner-linux-xfs@oss.sgi.com Sun Mar 18 20:03:24 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2J43OH23103
	for linux-xfs-outgoing; Sun, 18 Mar 2001 20:03:24 -0800
Received: from web1609.mail.yahoo.com (web1609.mail.yahoo.com [128.11.23.163])
	by oss.sgi.com (8.11.3/8.11.3) with SMTP id f2J43NM23100
	for <linux-xfs@oss.sgi.com>; Sun, 18 Mar 2001 20:03:23 -0800
Received: (qmail 6828 invoked by uid 60001); 19 Mar 2001 04:03:23 -0000
Message-ID: <20010319040323.6827.qmail@web1609.mail.yahoo.com>
Received: from [202.95.126.106] by web1609.mail.yahoo.com; Sun, 18 Mar 2001 20:03:23 PST
Date: Sun, 18 Mar 2001 20:03:23 -0800 (PST)
From: girish chandrashekara <girish_c_99@yahoo.com>
Subject: Re: dmapi support in XFS 
To: Dean Roehrich <roehrich@sgi.com>
Cc: linux-xfs@oss.sgi.com, openxdsm-devel@lists.sourceforge.net
In-Reply-To: <200103162031.OAA13270@slobber.americas.sgi.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

> DM_EVENT_READ comes from XFS this way:
> 
> xfs_read->xfs_dm_send_data_event->dm_send_data_event

In 2.4.0-test5 kernel I can see read path 
xfs_read->xfs_rdwr->pagebuf_generic_file_read->pagebuf_generic_file_read->linvfs_file_read->pagebuf_file_read
nowhere data event(xfs_dm_send_data_event) is
triggered

For nospace event my problem remains same, 
in xfs_write, I don't see any nospace event called 

May be the XFS code I browsed is incomplete or is
there any other explaination?


girish


__________________________________________________
Do You Yahoo!?
Get email at your own domain with Yahoo! Mail. 
http://personal.mail.yahoo.com/

From owner-linux-xfs@oss.sgi.com Sun Mar 18 22:53:27 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2J6rRv25518
	for linux-xfs-outgoing; Sun, 18 Mar 2001 22:53:27 -0800
Received: from sgi.com (sgi.SGI.COM [192.48.153.1])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2J6rQM25515
	for <linux-xfs@oss.sgi.com>; Sun, 18 Mar 2001 22:53:26 -0800
Received: from larry.melbourne.sgi.com ([134.14.52.130]) 
	by sgi.com (980327.SGI.8.8.8-aspam/980304.SGI-aspam:
       SGI does not authorize the use of its proprietary
       systems or networks for unsolicited or bulk email
       from the Internet.) 
	via SMTP id WAA03189
	for <linux-xfs@oss.sgi.com>; Sun, 18 Mar 2001 22:53:25 -0800 (PST)
	mail_from (ivanr@melbourne.sgi.com)
Received: from omen.melbourne.sgi.com (omen.melbourne.sgi.com [134.14.55.139]) by larry.melbourne.sgi.com (950413.SGI.8.6.12/950213.SGI.AUTOCF) via ESMTP id RAA22924; Mon, 19 Mar 2001 17:52:03 +1100
From: ivanr@melbourne.sgi.com (Ivan Rayner)
Received: from localhost (ivanr@localhost)
	by omen.melbourne.sgi.com (SGI-8.9.3/8.9.3) with ESMTP id RAA77995;
	Mon, 19 Mar 2001 17:52:02 +1100 (EST)
X-Authentication-Warning: omen.melbourne.sgi.com: ivanr owned process doing -bs
Date: Mon, 19 Mar 2001 17:52:02 +1100
To: Steve Lord <lord@sgi.com>
cc: "Eugene B. Byrganov" <SkyMaster@inp.nsk.su>, <linux-xfs@oss.sgi.com>
Subject: Re: xfsrestory problem. 
In-Reply-To: <200103161445.f2GEjYe19328@jen.americas.sgi.com>
Message-ID: <Pine.SGI.4.32.0103191729350.610589-100000@omen.melbourne.sgi.com>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

On Fri, 16 Mar 2001, Steve Lord wrote:

> I am not sure how much work it would be to get xfsrestore to function
> on a non-xfs filesystem. As the names indicate, these are xfs specific
> programs - they know about the extra things in xfs filesystems such as
> extended attributes and can store these. It also looks like they take
> advantage of xfs special system calls.

Despite the warnings, I think you'll find the xfsrestore succeeded.  Not
being able to reserve space isn't fatal for xfsrestore, and the rest of
the operation should have succeeded.  I'd like to know if it didn't.

> The second error seems a little draconian.

The second error is fatal, because for the table-of-contents, xfsrestore
wants to create a tempory directory in the current directory.  Well
anyway, that's the theory, and I'm sure it's correct, but it does seem
odd, and I can't see any obvious place where xfsrestore writes to this
directory and actually needs it to be an xfs filesystem.

This error is also present on IRIX and it is annoying when it occurs, but
it's very rare nowadays of course.  It will probably be more common on
Linux, so it's probably worth investigating.  The reservation warnings
could also be avoided fairly easily.  I'll add them to my list of things
to do when I'm bored...

Ivan


>
> Our dump/restore experts are in Australia and it is the middle of the
> night there, they may have comments later.
>
> Steve
>
> > Hi!
> >
> > Current dir: ext2 and 'xfs-root.dump' - dump on Irix 6.5
> > xfs: CVS 15 Mar 2001
> >
> > First: xfsrestore -r -f xfs-root.dump
> > --------------------------------------------------
> > xfsrestore: NOTE: attempt to reserve 46720 bytes for
> > /home/sky/xfsrestorehousekeepingdir/dirattr using F_ALLOCSP64 failed:
> > Inappropriate ioctl for device (25)
> > xfsrestore: NOTE: attempt to reserve 46720 bytes for
> > /home/sky/xfsrestorehousekeepingdir/dirattr using F_ALLOCSP64 failed:
> > Inappropriate ioctl for device (25)
> > xfsrestore: NOTE: attempt to reserve 45496 bytes for
> > /home/sky/xfsrestorehousekeepingdir/namreg using F_ALLOCSP64 failed:
> > Inappropriate ioctl for device (25)
> > xfsrestore: NOTE: attempt to reserve 45496 bytes for
> > /home/sky/xfsrestorehousekeepingdir/namreg using F_ALLOCSP64 failed:
> > Inappropriate ioctl for device (25)
> > --------------------------------------------------
> >
> > Don't 'xfsrestore'  know about ext2?
> >
> > Second:
> > --------------------------------------------------
> > [root@cfs sky]# xfsrestore -t -f xfs-root.dump
> > xfsrestore: version 3.0 - Running single-threaded
> > xfsrestore: ERROR: Current directory not XFS: /home/sky
> > --------------------------------------------------
> > Abort!!!!
> >
> > Why? I only want get list of files.
> >
> > --
> >                       s NAILU^[IMI POVELANIQMI, eWGENIJ bYRGANOW.
> >                       Best regards, Eugene Byrganov.
> >                       Viele Gruesse, Eugen Byrganow.
> >
> >   mailto:E.B.Byrganov@inp.nsk.su
> >
> >
>
>

-- 
Ivan Rayner
ivanr@melbourne.sgi.com


From owner-linux-xfs@oss.sgi.com Mon Mar 19 09:14:05 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2JHE5j03372
	for linux-xfs-outgoing; Mon, 19 Mar 2001 09:14:05 -0800
Received: from deliverator.sgi.com (deliverator.sgi.com [204.94.214.10])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2JHE4M03369
	for <linux-xfs@oss.sgi.com>; Mon, 19 Mar 2001 09:14:04 -0800
Received: from zeus-fddi.americas.sgi.com (zeus-fddi.americas.sgi.com [128.162.8.103]) by deliverator.sgi.com (980309.SGI.8.8.8-aspam-6.2/980310.SGI-aspam) via ESMTP id JAA00553
	for <linux-xfs@oss.sgi.com>; Mon, 19 Mar 2001 09:12:52 -0800 (PST)
	mail_from (roehrich@sgi.com)
Received: from daisy-e185.americas.sgi.com (daisy-e194.americas.sgi.com [128.162.194.214]) by zeus-fddi.americas.sgi.com (8.9.3/americas-smart-nospam1.1) with ESMTP id LAA1188952; Mon, 19 Mar 2001 11:12:47 -0600 (CST)
Received: from slobber.americas.sgi.com (slobber.americas.sgi.com [128.162.184.30]) by daisy-e185.americas.sgi.com (SGI-8.9.3/SGI-server-1.7) with ESMTP id LAA22973; Mon, 19 Mar 2001 11:12:47 -0600 (CST)
Received: from slobber.americas.sgi.com by slobber.americas.sgi.com (SGI-8.9.3/SGI-client-1.7) via ESMTP id LAA49999; Mon, 19 Mar 2001 11:12:46 -0600 (CST)
Message-Id: <200103191712.LAA49999@slobber.americas.sgi.com>
To: girish chandrashekara <girish_c_99@yahoo.com>
cc: linux-xfs@oss.sgi.com, openxdsm-devel@lists.sourceforge.net
Subject: Re: dmapi support in XFS 
Date: Mon, 19 Mar 2001 11:12:46 -0600
From: Dean Roehrich <roehrich@sgi.com>
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk


>From:  girish chandrashekara <girish_c_99@yahoo.com>
>> DM_EVENT_READ comes from XFS this way:
>> 
>> xfs_read->xfs_dm_send_data_event->dm_send_data_event
>
>In 2.4.0-test5 kernel I can see read path 
>xfs_read->xfs_rdwr->pagebuf_generic_file_read->pagebuf_generic_file_read->linv
>fs_file_read->pagebuf_file_read
>nowhere data event(xfs_dm_send_data_event) is
>triggered

xfs_rdwr disappeared in January.  DMAPI read/write events were added to the
read/write path on October 9th, 2000.

Care to update your tree and then we can talk?


Dean

From owner-linux-xfs@oss.sgi.com Mon Mar 19 10:26:36 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2JIQav05982
	for linux-xfs-outgoing; Mon, 19 Mar 2001 10:26:36 -0800
Received: from deliverator.sgi.com (deliverator.sgi.com [204.94.214.10])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2JIQaM05979
	for <linux-xfs@oss.sgi.com>; Mon, 19 Mar 2001 10:26:36 -0800
Received: from waco.engr.sgi.com (waco.engr.sgi.com [163.154.18.95]) by deliverator.sgi.com (980309.SGI.8.8.8-aspam-6.2/980310.SGI-aspam) via ESMTP id KAA11887
	for <linux-xfs@oss.sgi.com>; Mon, 19 Mar 2001 10:25:24 -0800 (PST)
	mail_from (ananth@waco.engr.sgi.com)
Received: (from ananth@localhost)
	by waco.engr.sgi.com (8.11.0/8.11.0) id f2JIR4l22799
	for linux-xfs@oss.sgi.com; Mon, 19 Mar 2001 10:27:04 -0800
Date: Mon, 19 Mar 2001 10:27:04 -0800
From: Ananth Ananthanarayanan <ananth@waco.engr.sgi.com>
Message-Id: <200103191827.f2JIR4l22799@waco.engr.sgi.com>
Subject: TAKE - remove redundant PF_MEMALLOC
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

Date:  Mon Mar 19 10:25:43 PST 2001
Workarea:  waco.engr.sgi.com:/build1/ananth/xfs-tot

The following file(s) were checked into:
  bonnie.engr.sgi.com:/isms/slinx/2.4.x-xfs


Modid:  2.4.x-xfs:slinx:90330a
linux/fs/pagebuf/page_buf_io.c - 1.68
	- Remove use of PF_MEMALLOC in writepage() ... appropriate daemons
	  will have the right flags, or in the case of memory allocators
	  core linux code will set this flag.



From owner-linux-xfs@oss.sgi.com Mon Mar 19 10:28:47 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2JISlu06035
	for linux-xfs-outgoing; Mon, 19 Mar 2001 10:28:47 -0800
Received: from pneumatic-tube.sgi.com (pneumatic-tube.sgi.com [204.94.214.22])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2JISlM06032
	for <linux-xfs@oss.sgi.com>; Mon, 19 Mar 2001 10:28:47 -0800
Received: from waco.engr.sgi.com (waco.engr.sgi.com [163.154.18.95]) by pneumatic-tube.sgi.com (980327.SGI.8.8.8-aspam/980310.SGI-aspam) via ESMTP id KAA06527
	for <linux-xfs@oss.sgi.com>; Mon, 19 Mar 2001 10:38:44 -0800 (PST)
	mail_from (ananth@waco.engr.sgi.com)
Received: (from ananth@localhost)
	by waco.engr.sgi.com (8.11.0/8.11.0) id f2JISUJ22884
	for linux-xfs@oss.sgi.com; Mon, 19 Mar 2001 10:28:30 -0800
Date: Mon, 19 Mar 2001 10:28:30 -0800
From: Ananth Ananthanarayanan <ananth@waco.engr.sgi.com>
Message-Id: <200103191828.f2JISUJ22884@waco.engr.sgi.com>
Subject: TAKE - remove another PF_MEMALLOC in vn_alloc 
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

Date:  Mon Mar 19 10:27:31 PST 2001
Workarea:  waco.engr.sgi.com:/build1/ananth/xfs-tot

The following file(s) were checked into:
  bonnie.engr.sgi.com:/isms/slinx/2.4.x-xfs


Modid:  2.4.x-xfs:slinx:90331a
linux/fs/xfs/linux/xfs_vnode.c - 1.51
	- Remove PF_MEMALLOC during inode allocation, we now handle
	  out-of-memory situations more gracefully.



From owner-linux-xfs@oss.sgi.com Mon Mar 19 14:52:24 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2JMqOS11677
	for linux-xfs-outgoing; Mon, 19 Mar 2001 14:52:24 -0800
Received: from deliverator.sgi.com (deliverator.sgi.com [204.94.214.10])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2JMqNM11674
	for <linux-xfs@oss.sgi.com>; Mon, 19 Mar 2001 14:52:23 -0800
Received: from zeus-fddi.americas.sgi.com (zeus-fddi.americas.sgi.com [128.162.8.103]) by deliverator.sgi.com (980309.SGI.8.8.8-aspam-6.2/980310.SGI-aspam) via ESMTP id OAA26222
	for <linux-xfs@oss.sgi.com>; Mon, 19 Mar 2001 14:51:10 -0800 (PST)
	mail_from (lord@sgi.com)
Received: from daisy-e185.americas.sgi.com (daisy-e194.americas.sgi.com [128.162.194.214]) by zeus-fddi.americas.sgi.com (8.9.3/americas-smart-nospam1.1) with ESMTP id QAA1188554 for <linux-xfs@oss.sgi.com>; Mon, 19 Mar 2001 16:51:06 -0600 (CST)
Received: from jen.americas.sgi.com (IDENT:root@jen.americas.sgi.com [128.162.184.86]) by daisy-e185.americas.sgi.com (SGI-8.9.3/SGI-server-1.7) with ESMTP id QAA45768 for <linux-xfs@oss.sgi.com>; Mon, 19 Mar 2001 16:51:06 -0600 (CST)
From: Steve Lord <lord@sgi.com>
Received: by jen.americas.sgi.com (8.11.0/SGI-client-1.7) id f2JMqKd01983; Mon, 19 Mar 2001 16:52:20 -0600
Message-Id: <200103192252.f2JMqKd01983@jen.americas.sgi.com>
Date: Mon, 19 Mar 2001 16:52:20 -0600
Subject: TAKE - remove extra code from __make_request
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

Looks like some code got left in __make_request which we should have lost 
a while back. Testing shows we run just fine without it, removing.

Date:  Mon Mar 19 14:49:44 PST 2001
Workarea:  jen.americas.sgi.com:/src/lord/xfs-linux.2.4

The following file(s) were checked into:
  bonnie.engr.sgi.com:/isms/slinx/2.4.x-xfs


Modid:  2.4.x-xfs:slinx:90364a
linux/drivers/block/ll_rw_blk.c - 1.63
	- remove extra code - probably merge remainders



From owner-linux-xfs@oss.sgi.com Mon Mar 19 15:30:52 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2JNUqq12656
	for linux-xfs-outgoing; Mon, 19 Mar 2001 15:30:52 -0800
Received: from groucho.maths.monash.edu.au (groucho.maths.monash.edu.au [130.194.160.211])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2JNUpM12653
	for <linux-xfs@oss.sgi.com>; Mon, 19 Mar 2001 15:30:51 -0800
Received: (from rjh@localhost)
	by groucho.maths.monash.edu.au (8.8.8/8.8.8) id XAA01298
	for linux-xfs@oss.sgi.com; Mon, 19 Mar 2001 23:30:49 GMT
From: Robin Humble <rjh@groucho.maths.monash.edu.au>
Message-Id: <200103192330.XAA01298@groucho.maths.monash.edu.au>
Subject: raid5 rebuild time
To: linux-xfs@oss.sgi.com
Date: Tue, 20 Mar 2001 10:30:49 +1100 (EDT)
X-Mailer: ELM [version 2.5 PL1]
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk


Hi,

This is more a RAID5 question than an XFS one, but perhaps it's
relevant anyway - I'm not sure - I'm a bit of a novice at RAID5.

We have just added to our machine to make it 8 60G ATA/100 disks in a
XFS+SoftwareRAID5 setup. /proc/mdstat tells me that it's going to
take 700+ minutes to build the parity data for the full 480G. It's
running at a speed of ~1300K/sec...

Is this about the length of time I'd expect? Seems a lot... :-/

It's a dual celeron 500 system and both CPUs seem to be flat out running
100% system time doing raid5syncd and raid5d. It's running cvs kernel from
yesterday (sunday).

When we had 4 of these 60G disks doing RAID5+XFS I think it took
about 180 minutes to reconstruct all the parity data which was a more
reasonable time.

Am I just CPU speed limited here?
Is there any rule of thumb that governs these rebuild times?
Something depending upon number of disks in the raid set perhaps.

When it boots it prints:
 Mar 19 17:26:49 jersey kernel: raid5: measuring checksumming speed
 Mar 19 17:26:49 jersey kernel:    8regs     :   674.400 MB/sec
 Mar 19 17:26:49 jersey kernel:    32regs    :   455.600 MB/sec
 Mar 19 17:26:49 jersey kernel:    pII_mmx   :  1113.200 MB/sec
 Mar 19 17:26:49 jersey kernel:    p5_mmx    :  1180.000 MB/sec
 Mar 19 17:26:49 jersey kernel: raid5: using function: p5_mmx (1180.000 MB/sec)

which looks like 1GB/sec which is fast - more like 480 seconds for
480G of disk than 11 or 12 hours :)

Anyway - like I said, I have no idea what is "normal" for RAID5 so any
hints are much appreciated.

cheers,
robin

From owner-linux-xfs@oss.sgi.com Mon Mar 19 16:00:21 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2K00Ln13066
	for linux-xfs-outgoing; Mon, 19 Mar 2001 16:00:21 -0800
Received: from tux.mkp.net (tux.mkp.net [130.225.60.11])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2K00GM13063
	for <linux-xfs@oss.sgi.com>; Mon, 19 Mar 2001 16:00:17 -0800
Received: from tux.mkp.net ([130.225.60.11] helo=jaguar.mkp.net)
	by tux.mkp.net with esmtp (Exim 3.16 #1)
	id 14f9ZJ-0007ty-00; Tue, 20 Mar 2001 01:00:10 +0100
Received: (from mkp@localhost)
	by jaguar.mkp.net (8.9.3/8.9.3) id SAA13259;
	Mon, 19 Mar 2001 18:58:08 -0500
X-Authentication-Warning: jaguar.mkp.net: mkp set sender to mkp@mkp.net using -f
To: Robin Humble <rjh@groucho.maths.monash.edu.au>
Cc: linux-xfs@oss.sgi.com
Subject: Re: raid5 rebuild time
References: <200103192330.XAA01298@groucho.maths.monash.edu.au>
From: "Martin K. Petersen" <mkp@linuxcare.com>
Organization: Linuxcare, Inc.
Date: 19 Mar 2001 18:58:08 -0500
In-Reply-To: <200103192330.XAA01298@groucho.maths.monash.edu.au>
Message-ID: <yq11yrttipb.fsf@jaguar.mkp.net>
Lines: 20
User-Agent: Gnus/5.0808 (Gnus v5.8.8) XEmacs/21.1 (Canyonlands)
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

>>>>> "Robin" == Robin Humble <rjh@groucho.maths.monash.edu.au> writes:

Robin> This is more a RAID5 question than an XFS one, but perhaps it's
Robin> relevant anyway - I'm not sure - I'm a bit of a novice at
Robin> RAID5.

Robin> It's running at a speed of ~1300K/sec...

My recent RAID changes seem to have caused a bigger performance
penalty than expected.  But only on some configurations.  My box at
works blasts away on both SCSI and FC-AL.  And here at home my testbox
crawls during resync.

I've begun fixing the stripe cache.  That should hopefully get things
accelerated a bit.

-- 
Martin K. Petersen, Principal Linux Consultant, Linuxcare, Inc.
mkp@linuxcare.com, http://www.linuxcare.com/
SGI XFS for Linux Developer, http://oss.sgi.com/projects/xfs/

From owner-linux-xfs@oss.sgi.com Mon Mar 19 18:07:08 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2K278t15535
	for linux-xfs-outgoing; Mon, 19 Mar 2001 18:07:08 -0800
Received: from sgi.com (sgi.SGI.COM [192.48.153.1])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2K277M15531
	for <linux-xfs@oss.sgi.com>; Mon, 19 Mar 2001 18:07:07 -0800
Received: from snort.melbourne.sgi.com ([134.14.55.149]) 
	by sgi.com (980327.SGI.8.8.8-aspam/980304.SGI-aspam:
       SGI does not authorize the use of its proprietary
       systems or networks for unsolicited or bulk email
       from the Internet.) 
	via ESMTP id SAA09476
	for <linux-xfs@oss.sgi.com>; Mon, 19 Mar 2001 18:06:56 -0800 (PST)
	mail_from (nathans@snort.melbourne.sgi.com)
Received: (from nathans@localhost)
	by snort.melbourne.sgi.com (SGI-8.9.3/8.9.3) id NAA69280
	for linux-xfs@oss.sgi.com; Tue, 20 Mar 2001 13:05:37 +1100 (EST)
Date: Tue, 20 Mar 2001 13:05:37 +1100 (EST)
From: Nathan Scott <nathans@snort.melbourne.sgi.com>
Message-Id: <200103200205.NAA69280@snort.melbourne.sgi.com>
To: linux-xfs@oss.sgi.com
Subject: TAKE - sparc build, quota doc
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

Fixes sparc build issue picked up by the Debian sparc build daemon
(really a glibc problem, but we can work around it easily enough -
see build log at http://vore.debian.org/buildlogs/xfsprogs for the
full details).

Also, added in some documentation on the use of quota with XFS.

cheers.


Date:  Sun Mar 18 16:38:00 PST 2001
Workarea:  snort.melbourne.sgi.com:/diskb/build4/nathans/linux-xfs

The following file(s) were checked into:
  bonnie.engr.sgi.com:/isms/slinx/2.4.x-xfs


Modid:  2.4.x-xfs:slinx:90310a
cmd/xfstests/tools/srcdiff - 1.4
	- support module has moved in the kernel.

cmd/xfsprogs/include/xfs_dir_leaf.h - 1.2
cmd/xfsprogs/include/xfs_dir2.h - 1.2
	- sync with benign (for userspace) kernel header change.

cmd/xfsprogs/debian/changelog - 1.5
	- push 1.1.5 into unstable.


Date:  Mon Mar 19 17:57:36 PST 2001
Workarea:  snort.melbourne.sgi.com:/diskb/build4/nathans/linux-xfs

The following file(s) were checked into:
  bonnie.engr.sgi.com:/isms/slinx/2.4.x-xfs


Modid:  2.4.x-xfs:slinx:90386a
cmd/xfsprogs/doc/README.quota - 1.1
	- document the use of quota and XFS.

cmd/xfsprogs/include/platform_defs.h.in - 1.3
	- fix Debian bug #90211 - XFS commands fail to build on sparc.

cmd/xfsprogs/doc/Makefile - 1.4
	- added in README.quota doc.

cmd/xfsprogs/doc/CHANGES - 1.7
cmd/xfsprogs/debian/changelog - 1.6
	- update wrt sparc+O_DIRECT, quota doc.  push 1.1.6 into unstable.

cmd/xfsprogs/VERSION - 1.7
	- roll minor version number.



From owner-linux-xfs@oss.sgi.com Mon Mar 19 18:07:36 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2K27aU15546
	for linux-xfs-outgoing; Mon, 19 Mar 2001 18:07:36 -0800
Received: from gwyn.tux.org (ident-user@gwyn.tux.org [207.96.122.8])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2K27SM15543
	for <linux-xfs@oss.sgi.com>; Mon, 19 Mar 2001 18:07:31 -0800
Received: (from timball@localhost)
	by gwyn.tux.org (8.9.3/8.9.1) id VAA03107
	for linux-xfs@oss.sgi.com; Mon, 19 Mar 2001 21:07:06 -0500
Date: Mon, 19 Mar 2001 21:07:06 -0500
From: Timothy Ball <timball@tux.org>
To: linux-xfs@oss.sgi.com
Subject: debian + xfs + gcc
Message-ID: <20010319210706.E24087@gwyn.tux.org>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.3.14i
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

I'm trying to compile xfs on debian system. I'd rather not recompile the
compiler to do this, I notice that gcc wants to use gcc 2.91.66, where
as debian sid uses 2.95.3 would it be an issue to change the lines in
the top level Makefile to this:

--snip--snip--snip--
#comment out this line if compiling on RH 7.0
#CC     = $(CROSS_COMPILE)gcc -V egcs-2.91.66
# AND uncomment the following line
CC      = $(CROSS_COMPILE)gcc
--snip--snip--snip--

--timball

-- 
	Send mail with subject "send pgp key" for public key.
pub  1024R/CFF85605 1999-06-10 Timothy L. Ball <timball@sheergenius.com>
     Key fingerprint = 8A 8E 64 D6 21 C0 90 29  9F D6 1E DC F8 18 CB CD

From owner-linux-xfs@oss.sgi.com Mon Mar 19 18:18:22 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2K2IMY15682
	for linux-xfs-outgoing; Mon, 19 Mar 2001 18:18:22 -0800
Received: from sgi.com (sgi.SGI.COM [192.48.153.1])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2K2I5M15676
	for <linux-xfs@oss.sgi.com>; Mon, 19 Mar 2001 18:18:05 -0800
Received: from ledzep.americas.sgi.com (relay.sgi.com [137.38.226.97] (may be forged)) 
	by sgi.com (980327.SGI.8.8.8-aspam/980304.SGI-aspam:
       SGI does not authorize the use of its proprietary
       systems or networks for unsolicited or bulk email
       from the Internet.) 
	via ESMTP id SAA03558
	for <linux-xfs@oss.sgi.com>; Mon, 19 Mar 2001 18:17:51 -0800 (PST)
	mail_from (cattelan@thebarn.com)
Received: from gibble.americas.sgi.com (gibble.americas.sgi.com [128.162.195.80]) by ledzep.americas.sgi.com (SGI-SGI-8.9.3/americas-smart-nospam1.1) with ESMTP id UAA35762; Mon, 19 Mar 2001 20:16:29 -0600 (CST)
Received: from thebarn.com (localhost [127.0.0.1])
	by gibble.americas.sgi.com (8.11.2/8.11.2) with ESMTP id f2K27ce09029;
	Mon, 19 Mar 2001 21:07:38 -0500
Message-ID: <3AB6BB68.70EF46FE@thebarn.com>
Date: Mon, 19 Mar 2001 21:07:37 -0500
From: Russell Cattelan <cattelan@thebarn.com>
X-Mailer: Mozilla 4.76 [en] (X11; U; Linux 2.4.2-XFS i686)
X-Accept-Language: en
MIME-Version: 1.0
To: Timothy Ball <timball@tux.org>
CC: linux-xfs@oss.sgi.com
Subject: Re: debian + xfs + gcc
References: <20010319210706.E24087@gwyn.tux.org>
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

Timothy Ball wrote:

> I'm trying to compile xfs on debian system. I'd rather not recompile the
> compiler to do this, I notice that gcc wants to use gcc 2.91.66, where
> as debian sid uses 2.95.3 would it be an issue to change the lines in
> the top level Makefile to this:

Yes that would work in fact if you look at the normal linux file
that is how it is shipped.


There have been several reports of 2.95.x working ... but it
it still largely untested by anybody at SGI.


>

>
>
> --snip--snip--snip--
> #comment out this line if compiling on RH 7.0
> #CC     = $(CROSS_COMPILE)gcc -V egcs-2.91.66
> # AND uncomment the following line
> CC      = $(CROSS_COMPILE)gcc
> --snip--snip--snip--
>
> --timball
>
> --
>         Send mail with subject "send pgp key" for public key.
> pub  1024R/CFF85605 1999-06-10 Timothy L. Ball <timball@sheergenius.com>
>      Key fingerprint = 8A 8E 64 D6 21 C0 90 29  9F D6 1E DC F8 18 CB CD

--
Russell Cattelan
--
Digital Elves inc. -- Currently on loan to SGI
Linux XFS core developer.




From owner-linux-xfs@oss.sgi.com Mon Mar 19 19:12:19 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2K3CJs17124
	for linux-xfs-outgoing; Mon, 19 Mar 2001 19:12:19 -0800
Received: from yog-sothoth.sgi.com (eugate.sgi.com [192.48.160.10])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2K3CIM17121
	for <linux-xfs@oss.sgi.com>; Mon, 19 Mar 2001 19:12:18 -0800
Received: from snort.melbourne.sgi.com (snort.melbourne.sgi.com [134.14.55.149]) by yog-sothoth.sgi.com (980305.SGI.8.8.8-aspam-6.2/980304.SGI-aspam-europe) via ESMTP id EAA3955435
	for <linux-xfs@oss.sgi.com>; Tue, 20 Mar 2001 04:12:16 +0100 (CET)
	mail_from (nathans@snort.melbourne.sgi.com)
Received: (from nathans@localhost)
	by snort.melbourne.sgi.com (SGI-8.9.3/8.9.3) id OAA74540
	for linux-xfs@oss.sgi.com; Tue, 20 Mar 2001 14:10:55 +1100 (EST)
Date: Tue, 20 Mar 2001 14:10:55 +1100 (EST)
From: Nathan Scott <nathans@snort.melbourne.sgi.com>
Message-Id: <200103200310.OAA74540@snort.melbourne.sgi.com>
To: linux-xfs@oss.sgi.com
Subject: TAKE - specfile
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

Date:  Mon Mar 19 18:17:03 PST 2001
Workarea:  snort.melbourne.sgi.com:/diskb/build4/nathans/linux-xfs

The following file(s) were checked into:
  bonnie.engr.sgi.com:/isms/slinx/2.4.x-xfs


Modid:  2.4.x-xfs:slinx:90388a
cmd/xfstests/common.quota - 1.2
	- check for kernel xfs quota before attempting to check for
	  userspace support.

cmd/xfstests/050 - 1.4
	- move test for xfs kernel quota into common.quota.


Date:  Mon Mar 19 19:09:12 PST 2001
Workarea:  snort.melbourne.sgi.com:/diskb/build4/nathans/linux-xfs

The following file(s) were checked into:
  bonnie.engr.sgi.com:/isms/slinx/2.4.x-xfs


Modid:  2.4.x-xfs:slinx:90394a
cmd/xfsprogs/build/rpm/xfsprogs.spec.in - 1.6
	- change conflicts to provides at {russell,martin}'s suggestion.



From owner-linux-xfs@oss.sgi.com Mon Mar 19 19:24:05 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2K3O5d17413
	for linux-xfs-outgoing; Mon, 19 Mar 2001 19:24:05 -0800
Received: from gwyn.tux.org (ident-user@gwyn.tux.org [207.96.122.8])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2K3O3M17409
	for <linux-xfs@oss.sgi.com>; Mon, 19 Mar 2001 19:24:04 -0800
Received: (from timball@localhost)
	by gwyn.tux.org (8.9.3/8.9.1) id WAA07081
	for linux-xfs@oss.sgi.com; Mon, 19 Mar 2001 22:24:02 -0500
Date: Mon, 19 Mar 2001 22:24:02 -0500
From: Timothy Ball <timball@finux.org>
To: XFS Mailing List <linux-xfs@oss.sgi.com>
Subject: mounting loop devices
Message-ID: <20010319222402.F24087@gwyn.tux.org>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.3.14i
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

Okay well I did it and installed xfs on my debian box. I now have the
interesting issue that mounting a loop devices doesn't seem to work
here's what I did:

dd if=/dev/zero of=/win98/block.dat bs=512 count=2500000
mkfs -t xfs /win98/block.dat
mount -o loop -t xfs /win98/block.dat test/

The mount then hangs.... forever it seems. I'm using the "beta" cvs tree
stuff. I'm a bit wary of installing the "fast moving" tree but if it
fixes that problem I guess I will have to. I did not enable all the
debuging options, mostly cause I didn't know how to deal w/ them.

TIA,
timball

From owner-linux-xfs@oss.sgi.com Mon Mar 19 19:37:09 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2K3b9S17513
	for linux-xfs-outgoing; Mon, 19 Mar 2001 19:37:09 -0800
Received: from lips.borg.umn.edu (lips.borg.umn.edu [160.94.232.50])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2K3b8M17509
	for <linux-xfs@oss.sgi.com>; Mon, 19 Mar 2001 19:37:08 -0800
Received: from thebarn.com (nic-31-c12-219.mn.mediaone.net [24.31.12.219])
	by lips.borg.umn.edu (8.12.0.Beta5/8.12.0.Beta5) with ESMTP id f2K3b6kn076389;
	Mon, 19 Mar 2001 21:37:07 -0600 (CST)
Message-ID: <3AB6D05C.FC62E518@thebarn.com>
Date: Mon, 19 Mar 2001 21:37:01 -0600
From: Russell Cattelan <cattelan@thebarn.com>
X-Mailer: Mozilla 4.74 [en] (X11; U; Linux 2.2.12 i386)
X-Accept-Language: en
MIME-Version: 1.0
To: Timothy Ball <timball@finux.org>
CC: XFS Mailing List <linux-xfs@oss.sgi.com>
Subject: Re: mounting loop devices
References: <20010319222402.F24087@gwyn.tux.org>
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

Timothy Ball wrote:

loop is know to be broken in linux 2.4.0 (which is what the beta tree is
based on)

If you can wait until later tonight or possibly tomorrow test rpm's of the
upcoming prerelease 0.10 will be available on oss.sgi.com.

These rpm's will have the a  loop  patch applied already.




> Okay well I did it and installed xfs on my debian box. I now have the
> interesting issue that mounting a loop devices doesn't seem to work
> here's what I did:
>
> dd if=/dev/zero of=/win98/block.dat bs=512 count=2500000
> mkfs -t xfs /win98/block.dat
> mount -o loop -t xfs /win98/block.dat test/
>
> The mount then hangs.... forever it seems. I'm using the "beta" cvs tree
> stuff. I'm a bit wary of installing the "fast moving" tree but if it
> fixes that problem I guess I will have to. I did not enable all the
> debuging options, mostly cause I didn't know how to deal w/ them.
>
> TIA,
> timball

--
Russell Cattelan
cattelan@thebarn.com




From owner-linux-xfs@oss.sgi.com Mon Mar 19 19:46:44 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2K3ki417676
	for linux-xfs-outgoing; Mon, 19 Mar 2001 19:46:44 -0800
Received: from gwyn.tux.org (ident-user@gwyn.tux.org [207.96.122.8])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2K3kgM17667
	for <linux-xfs@oss.sgi.com>; Mon, 19 Mar 2001 19:46:42 -0800
Received: (from timball@localhost)
	by gwyn.tux.org (8.9.3/8.9.1) id WAA08201;
	Mon, 19 Mar 2001 22:46:38 -0500
Date: Mon, 19 Mar 2001 22:46:38 -0500
From: Timothy Ball <timball@tux.org>
To: Russell Cattelan <cattelan@thebarn.com>
Cc: XFS Mailing List <linux-xfs@oss.sgi.com>
Subject: Re: mounting loop devices
Message-ID: <20010319224638.G24087@gwyn.tux.org>
References: <20010319222402.F24087@gwyn.tux.org> <3AB6D05C.FC62E518@thebarn.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.3.14i
In-Reply-To: <3AB6D05C.FC62E518@thebarn.com>; from cattelan@thebarn.com on Mon, Mar 19, 2001 at 09:37:01PM -0600
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

On Mon, Mar 19, 2001 at 09:37:01PM -0600, Russell Cattelan wrote:
> Timothy Ball wrote:
> 
> loop is know to be broken in linux 2.4.0 (which is what the beta tree is
> based on)
> 
> If you can wait until later tonight or possibly tomorrow test rpm's of the
> upcoming prerelease 0.10 will be available on oss.sgi.com.
> 
> These rpm's will have the a  loop  patch applied already.

Aaah that's the problem :). Is this fix applied to the cvs already? I
guess I can wait a day but boy oh boy do I want to start testing... :)

--timball

-- 
	Send mail with subject "send pgp key" for public key.
pub  1024R/CFF85605 1999-06-10 Timothy L. Ball <timball@sheergenius.com>
     Key fingerprint = 8A 8E 64 D6 21 C0 90 29  9F D6 1E DC F8 18 CB CD

From owner-linux-xfs@oss.sgi.com Mon Mar 19 19:57:41 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2K3vfC17849
	for linux-xfs-outgoing; Mon, 19 Mar 2001 19:57:41 -0800
Received: from mail15.jump.net (mail15.jump.net [206.196.91.15])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2K3veM17846
	for <linux-xfs@oss.sgi.com>; Mon, 19 Mar 2001 19:57:40 -0800
Received: from jump.net (aus-dsl-dhcp1-26.customer.jump.net [63.163.168.26])
	by mail15.jump.net (8.10.2/) with ESMTP id f2K3vMG21670;
	Mon, 19 Mar 2001 21:57:22 -0600 (CST)
Message-ID: <3AB6D559.BC0273AD@jump.net>
Date: Mon, 19 Mar 2001 21:58:17 -0600
From: Eric Sandeen <sandeen@jump.net>
X-Mailer: Mozilla 4.76 [en] (X11; U; Linux 2.4.2-XFS i686)
X-Accept-Language: en
MIME-Version: 1.0
To: Timothy Ball <timball@tux.org>
CC: linux-xfs@oss.sgi.com
Subject: Re: mounting loop devices
References: <20010319222402.F24087@gwyn.tux.org> <3AB6D05C.FC62E518@thebarn.com> <20010319224638.G24087@gwyn.tux.org>
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

Timothy Ball wrote:

> Aaah that's the problem :). Is this fix applied to the cvs already? I
> guess I can wait a day but boy oh boy do I want to start testing... :)

The fix is not in CVS, but if you get CVS and apply
http://www.kernel.org/pub/linux/kernel/people/axboe/xfs/loop-xfs-7C.bz2
you should be good to go...

-Eric

From owner-linux-xfs@oss.sgi.com Mon Mar 19 20:04:05 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2K445f17973
	for linux-xfs-outgoing; Mon, 19 Mar 2001 20:04:05 -0800
Received: from chaos.egr.duke.edu (IDENT:root@chaos.egr.duke.edu [152.3.195.82])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2K444M17970
	for <linux-xfs@oss.sgi.com>; Mon, 19 Mar 2001 20:04:04 -0800
Received: from localhost (jlb@localhost)
	by chaos.egr.duke.edu (8.9.3/8.9.3) with ESMTP id XAA25366
	for <linux-xfs@oss.sgi.com>; Mon, 19 Mar 2001 23:04:02 -0500
X-Authentication-Warning: chaos.egr.duke.edu: jlb owned process doing -bs
Date: Mon, 19 Mar 2001 23:04:02 -0500 (EST)
From: Joshua Baker-LePain <jlb17@duke.edu>
X-Sender:  <jlb@chaos.egr.duke.edu>
To: Linux xfs mailing list <linux-xfs@oss.sgi.com>
Subject: Pre-production questions
Message-ID: <Pine.LNX.4.30.0103192249060.25325-100000@chaos.egr.duke.edu>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

First, let me say thanks to this list and to SGI for all the hard work on
XFS.  It's a rather impressive bit of work, and almost kind of eerie
knowing that I could pull a disk out of my Indigo2, put it in one of the
Linux boxen, and be good to go.  Weird.

To the point at hand, my hardware RAID system came in today, and I'm ready
(having played with xfs on some smaller partitions) to start my real
pre-production testing.  I'll be exporting the RAID to about 20 users
(all Linux clients) over a 100Mb FD network.  I'm running a CVS kernel
from March 13, although I now hear of a new pre-release coming...

My RAID is an IDE-SCSI system -- IDE disks, looks like SCSI to the host.
It has 8 80GB drives in RAID 5, for about 560GB of usable space.  I'd like
it to be just one partition.  My question is, are there any special
parameters I should pass to mkfs for such a big (well, for me at least)
partition?  What about to tailor it to the hardware RAID?

Any hints are much appreciated, and thanks again.

-- 
Joshua Baker-LePain
Department of Biomedical Engineering
Duke University




From owner-linux-xfs@oss.sgi.com Mon Mar 19 20:28:36 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2K4SaU18485
	for linux-xfs-outgoing; Mon, 19 Mar 2001 20:28:36 -0800
Received: from deliverator.sgi.com (deliverator.sgi.com [204.94.214.10])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2K4SaM18482
	for <linux-xfs@oss.sgi.com>; Mon, 19 Mar 2001 20:28:36 -0800
Received: from snort.melbourne.sgi.com (snort.melbourne.sgi.com [134.14.55.149]) by deliverator.sgi.com (980309.SGI.8.8.8-aspam-6.2/980310.SGI-aspam) via ESMTP id UAA24458
	for <linux-xfs@oss.sgi.com>; Mon, 19 Mar 2001 20:27:19 -0800 (PST)
	mail_from (nathans@snort.melbourne.sgi.com)
Received: (from nathans@localhost)
	by snort.melbourne.sgi.com (SGI-8.9.3/8.9.3) id PAA07949
	for linux-xfs@oss.sgi.com; Tue, 20 Mar 2001 15:27:11 +1100 (EST)
Date: Tue, 20 Mar 2001 15:27:11 +1100 (EST)
From: Nathan Scott <nathans@snort.melbourne.sgi.com>
Message-Id: <200103200427.PAA07949@snort.melbourne.sgi.com>
To: linux-xfs@oss.sgi.com
Subject: TAKE - xfsdump specfile
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

Date:  Mon Mar 19 20:26:48 PST 2001
Workarea:  snort.melbourne.sgi.com:/diskb/build4/nathans/linux-xfs

The following file(s) were checked into:
  bonnie.engr.sgi.com:/isms/slinx/2.4.x-xfs


Modid:  2.4.x-xfs:slinx:90399a
cmd/xfsdump/build/rpm/xfsdump.spec.in - 1.4
	- we can no longer conflict with xfs-cmds as xfsprogs now Provides it,
	  but we should be OK cos we Require xfsprogs anyway.



From owner-linux-xfs@oss.sgi.com Mon Mar 19 21:08:21 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2K58Lx18855
	for linux-xfs-outgoing; Mon, 19 Mar 2001 21:08:21 -0800
Received: from home.smithconcepts.com (ubr-35.28.151.oviedo.cfl.rr.com [65.35.28.151])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2K58KM18852
	for <linux-xfs@oss.sgi.com>; Mon, 19 Mar 2001 21:08:20 -0800
Received: from ieee.org (IDENT:bjsmith@bitman.oviedo.smithconcepts.com [172.24.24.192])
	by home.smithconcepts.com (8.9.3/8.9.3) with ESMTP id AAA31386;
	Tue, 20 Mar 2001 00:02:45 -0500
Message-ID: <3AB6E7EC.424FA6DF@ieee.org>
Date: Tue, 20 Mar 2001 00:17:32 -0500
From: "Bryan J. Smith" <b.j.smith@ieee.org>
Reply-To: b.j.smith@ieee.org, thebs@theseus.com
Organization: SmithConcepts/Personal
X-Mailer: Mozilla 4.76 [en] (X11; U; Linux 2.2.18-0.13 i686)
X-Accept-Language: en
MIME-Version: 1.0
To: Eric Sandeen <sandeen@jump.net>
CC: Timothy Ball <timball@tux.org>, linux-xfs@oss.sgi.com
Subject: Re: mounting loop devices  What other XFS patches are recommended?
References: <20010319222402.F24087@gwyn.tux.org> <3AB6D05C.FC62E518@thebarn.com> <20010319224638.G24087@gwyn.tux.org> <3AB6D559.BC0273AD@jump.net>
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

Re: mounting loop devices  What other XFS patches are recommended?

Eric Sandeen wrote:
> The fix is not in CVS, but if you get CVS and apply
> http://www.kernel.org/pub/linux/kernel/people/axboe/xfs/loop-xfs-7C.bz2
> you should be good to go...

I'm admist building some new RPMs.  I added this patch in the mix. 
It seems the other Axobe patch (idekio) has already been applied to
the CVS repository.

Any other patches for XFS anyone wants to recommend?

-- TheBS

P.S.  Is there an XFS patch for mkinitrd-3.0.5?  All I have is the
2.8 release from your PR/Betas.  Should be be using the newer
release (assuming there is an XFS patch)?

-- 
Bryan "TheBS" Smith         chat:thebs413 @AOL/MSN/Yahoo
Engineer     mailto:b.j.smith@ieee.org,thebs@theseus.com
********************************************************
"Linux will do for applications what the Internet did to
 networks" -- Sam Palmisano, IBM Chief Operating Officer

From owner-linux-xfs@oss.sgi.com Mon Mar 19 23:20:07 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2K7K7V20355
	for linux-xfs-outgoing; Mon, 19 Mar 2001 23:20:07 -0800
Received: from yog-sothoth.sgi.com (eugate.sgi.com [192.48.160.10])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2K7K5M20350
	for <linux-xfs@oss.sgi.com>; Mon, 19 Mar 2001 23:20:06 -0800
Received: from snort.melbourne.sgi.com (snort.melbourne.sgi.com [134.14.55.149]) by yog-sothoth.sgi.com (980305.SGI.8.8.8-aspam-6.2/980304.SGI-aspam-europe) via ESMTP id IAA6008649
	for <linux-xfs@oss.sgi.com>; Tue, 20 Mar 2001 08:20:03 +0100 (CET)
	mail_from (nathans@snort.melbourne.sgi.com)
Received: (from nathans@localhost)
	by snort.melbourne.sgi.com (SGI-8.9.3/8.9.3) id SAA14424
	for linux-xfs@oss.sgi.com; Tue, 20 Mar 2001 18:18:35 +1100 (EST)
Date: Tue, 20 Mar 2001 18:18:35 +1100 (EST)
From: Nathan Scott <nathans@snort.melbourne.sgi.com>
Message-Id: <200103200718.SAA14424@snort.melbourne.sgi.com>
To: linux-xfs@oss.sgi.com
Subject: TAKE - qa
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

Date:  Mon Mar 19 23:17:52 PST 2001
Workarea:  snort.melbourne.sgi.com:/diskb/build4/nathans/linux-xfs

The following file(s) were checked into:
  bonnie.engr.sgi.com:/isms/slinx/2.4.x-xfs


Modid:  2.4.x-xfs:slinx:90408a
cmd/xfstests/052.out - 1.1
cmd/xfstests/052 - 1.1
	- test comparing output from quota(1) with xfs_db ondisk dquots.

cmd/xfstests/src/feature.c - 1.2
	- fix up a comment to describe -U/-G.

cmd/xfstests/group - 1.3
	- add test 052.

cmd/xfstests/common.quota - 1.4
	- move common mount stuff in here, fix test for xfs quota in-kernel.

cmd/xfstests/050 - 1.5
	- move mount stuff into common quota file, tidy a little, make
	  usrquota the default case.



From owner-linux-xfs@oss.sgi.com Tue Mar 20 01:18:25 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2K9IPg22485
	for linux-xfs-outgoing; Tue, 20 Mar 2001 01:18:25 -0800
Received: from sgi.com (sgi.SGI.COM [192.48.153.1])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2K9I0M22468
	for <linux-xfs@oss.sgi.com>; Tue, 20 Mar 2001 01:18:00 -0800
Received: from cthulhu.engr.sgi.com (cthulhu.engr.sgi.com [192.26.80.2]) 
	by sgi.com (980327.SGI.8.8.8-aspam/980304.SGI-aspam:
       SGI does not authorize the use of its proprietary
       systems or networks for unsolicited or bulk email
       from the Internet.) 
	via ESMTP id BAA02294
	for <linux-xfs@oss.sgi.com>; Tue, 20 Mar 2001 01:17:01 -0800 (PST)
	mail_from (tduffy@engr.sgi.com)
Received: from dbear.engr.sgi.com (dbear.engr.sgi.com [163.154.18.85])
	by cthulhu.engr.sgi.com (SGI-8.9.3/8.9.3) with ESMTP id BAA17350;
	Tue, 20 Mar 2001 01:15:36 -0800 (PST)
Date: Tue, 20 Mar 2001 01:11:45 -0800 (PST)
From: Tom Duffy <tduffy@engr.sgi.com>
To: <b.j.smith@ieee.org>, <thebs@theseus.com>
cc: Eric Sandeen <sandeen@jump.net>, Timothy Ball <timball@tux.org>,
   <linux-xfs@oss.sgi.com>
Subject: Re: mounting loop devices  What other XFS patches are recommended?
In-Reply-To: <3AB6E7EC.424FA6DF@ieee.org>
Message-ID: <Pine.LNX.4.30.0103200109490.2403-200000@dbear.engr.sgi.com>
MIME-Version: 1.0
Content-Type: MULTIPART/MIXED; BOUNDARY="-1700571886-1111633326-985079505=:2403"
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

  This message is in MIME format.  The first part should be readable text,
  while the remaining parts are likely unreadable without MIME-aware tools.
  Send mail to mime@docserver.cac.washington.edu for more info.

---1700571886-1111633326-985079505=:2403
Content-Type: TEXT/PLAIN; charset=US-ASCII

yeah, i have what you need.  this should do the trick.

-tduffy

On Tue, 20 Mar 2001, Bryan J. Smith wrote:

> Re: mounting loop devices  What other XFS patches are recommended?
>
> Eric Sandeen wrote:
> > The fix is not in CVS, but if you get CVS and apply
> > http://www.kernel.org/pub/linux/kernel/people/axboe/xfs/loop-xfs-7C.bz2
> > you should be good to go...
>
> I'm admist building some new RPMs.  I added this patch in the mix.
> It seems the other Axobe patch (idekio) has already been applied to
> the CVS repository.
>
> Any other patches for XFS anyone wants to recommend?
>
> -- TheBS
>
> P.S.  Is there an XFS patch for mkinitrd-3.0.5?  All I have is the
> 2.8 release from your PR/Betas.  Should be be using the newer
> release (assuming there is an XFS patch)?
>
>

---1700571886-1111633326-985079505=:2403
Content-Type: TEXT/PLAIN; charset=US-ASCII; name="mkinitrd-3.0.5-xfs.patch"
Content-Transfer-Encoding: BASE64
Content-ID: <Pine.LNX.4.30.0103200111450.2403@dbear.engr.sgi.com>
Content-Description: 
Content-Disposition: attachment; filename="mkinitrd-3.0.5-xfs.patch"
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---1700571886-1111633326-985079505=:2403--

From owner-linux-xfs@oss.sgi.com Tue Mar 20 01:45:23 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2K9jNn23148
	for linux-xfs-outgoing; Tue, 20 Mar 2001 01:45:23 -0800
Received: from esparrall.udg.es (esparrall.udg.es [130.206.124.16])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2K9jLM23145
	for <linux-xfs@oss.sgi.com>; Tue, 20 Mar 2001 01:45:21 -0800
Received: from gcs by esparrall.udg.es with local (Exim 3.22 #1 (Debian))
	id 14fIgG-0003lr-00
	for <linux-xfs@oss.sgi.com>; Tue, 20 Mar 2001 10:43:56 +0100
Date: Tue, 20 Mar 2001 10:43:56 +0100
From: GCS <gcs@agentsinside.com>
To: linux-xfs@oss.sgi.com
Subject: Bugreport - not really important
Message-ID: <20010320104356.A14454@esparrall.udg.es>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.3.15i
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

Hello all,

I made up my mind, and decided to change a real partition to XFS. I made a
backup, then formatted my 16Gb slice to XFS, which was ReiserFS previously.
I mounted it without problems, but(!) I still could mount it as ReiserFS!
Then I got back all of my files, like nothing happened with the partition.
Mounted as XFS showed the partition empty. Weird, isn't it? I had to format
the partition as ext2 (bad feeling seeing lost+found :-) ), then reformat as
XFS. This was the way to properly eliminate ReiserFS.
If some of you have a little time, please look into it.

Anyway, we had two power fault since then, and the XFS works perfectly!
Very good work!
Sincerely,
GCS

From owner-linux-xfs@oss.sgi.com Tue Mar 20 02:20:46 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2KAKki23668
	for linux-xfs-outgoing; Tue, 20 Mar 2001 02:20:46 -0800
Received: from msg.ecetra.com (dollar.ecetra.com [193.164.224.209])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2KAKiM23665
	for <linux-xfs@oss.sgi.com>; Tue, 20 Mar 2001 02:20:44 -0800
Received: from vie-ac.office.ecetra.com (vie-ac.office.ecetra.com [10.251.148.147] (may be forged))
	by msg.ecetra.com (8.9.3/8.9.3) with ESMTP id LAA05686;
	Tue, 20 Mar 2001 11:20:22 +0100
Received: from localhost (localhost [127.0.0.1])
	by vie-ac.office.ecetra.com (8.9.3/8.9.3) with ESMTP id LAA10427;
	Tue, 20 Mar 2001 11:20:22 +0100
Date: Tue, 20 Mar 2001 11:20:22 +0100 (CET)
From: Adam Cioccarelli <alciocca@ecetra.com>
To: Russell Cattelan <cattelan@thebarn.com>
cc: Timothy Ball <timball@tux.org>, <linux-xfs@oss.sgi.com>
Subject: Re: debian + xfs + gcc
In-Reply-To: <3AB6BB68.70EF46FE@thebarn.com>
Message-ID: <Pine.LNX.4.33.0103201114440.8969-100000@vie-ac.office.ecetra.com>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

I don't know if it's interesting but I have a well used XFS system running
for around 7 days now compiled with 2.95.2 so it seems OK on a standard
desktop PC at least...

PS. Does anyone on the list have any idea what in the changes since 9 days
ago might effect vmware? It ran perfectly on my box (except for the file
locking messages) until then. I thought it might be because I recompiled
the kernel with 2.95.2 but I went back to 2.91.66 and it still didn't
work. Anyway I seem to be able to live without exchange!


-------------------------------------------------------------------------------
Adam Cioccarelli (B.E Mechanical)                Adam.Cioccarelli@ecetra.com
Database Administrator			         Phone: +43 1 536 89 7725
						 Fax:   +43 1 536 89 7719
ecetra Central European e-Finance AG             Mobile:+43 664 181 4195
-------------------------------------------------------------------------------

On Mon, 19 Mar 2001, Russell Cattelan wrote:

> Timothy Ball wrote:
>
> > I'm trying to compile xfs on debian system. I'd rather not recompile the
> > compiler to do this, I notice that gcc wants to use gcc 2.91.66, where
> > as debian sid uses 2.95.3 would it be an issue to change the lines in
> > the top level Makefile to this:
>
> Yes that would work in fact if you look at the normal linux file
> that is how it is shipped.
>
>
> There have been several reports of 2.95.x working ... but it
> it still largely untested by anybody at SGI.
>
>
> >
>
> >
> >
> > --snip--snip--snip--
> > #comment out this line if compiling on RH 7.0
> > #CC     = $(CROSS_COMPILE)gcc -V egcs-2.91.66
> > # AND uncomment the following line
> > CC      = $(CROSS_COMPILE)gcc
> > --snip--snip--snip--
> >
> > --timball
> >
> > --
> >         Send mail with subject "send pgp key" for public key.
> > pub  1024R/CFF85605 1999-06-10 Timothy L. Ball <timball@sheergenius.com>
> >      Key fingerprint = 8A 8E 64 D6 21 C0 90 29  9F D6 1E DC F8 18 CB CD
>
> --
> Russell Cattelan
> --
> Digital Elves inc. -- Currently on loan to SGI
> Linux XFS core developer.
>
>


From owner-linux-xfs@oss.sgi.com Tue Mar 20 02:34:40 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2KAYee23773
	for linux-xfs-outgoing; Tue, 20 Mar 2001 02:34:40 -0800
Received: from deliverator.sgi.com (deliverator.sgi.com [204.94.214.10])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2KAYcM23770
	for <linux-xfs@oss.sgi.com>; Tue, 20 Mar 2001 02:34:38 -0800
Received: from cthulhu.engr.sgi.com (gate3-relay.engr.sgi.com [130.62.1.234]) by deliverator.sgi.com (980309.SGI.8.8.8-aspam-6.2/980310.SGI-aspam) via ESMTP id CAA26296
	for <linux-xfs@oss.sgi.com>; Tue, 20 Mar 2001 02:33:26 -0800 (PST)
	mail_from (tduffy@engr.sgi.com)
Received: from dbear.engr.sgi.com (dbear.engr.sgi.com [163.154.18.85])
	by cthulhu.engr.sgi.com (SGI-8.9.3/8.9.3) with ESMTP id CAA78174;
	Tue, 20 Mar 2001 02:33:21 -0800 (PST)
Date: Tue, 20 Mar 2001 02:29:30 -0800 (PST)
From: Tom Duffy <tduffy@engr.sgi.com>
To: GCS <gcs@agentsinside.com>
cc: <linux-xfs@oss.sgi.com>
Subject: Re: Bugreport - not really important
In-Reply-To: <20010320104356.A14454@esparrall.udg.es>
Message-ID: <Pine.LNX.4.30.0103200223470.2403-100000@dbear.engr.sgi.com>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

> Hello all,
>
> I made up my mind, and decided to change a real partition to XFS. I made a
> backup, then formatted my 16Gb slice to XFS, which was ReiserFS previously.
> I mounted it without problems, but(!) I still could mount it as ReiserFS!
> Then I got back all of my files, like nothing happened with the partition.
> Mounted as XFS showed the partition empty. Weird, isn't it? I had to format
> the partition as ext2 (bad feeling seeing lost+found :-) ), then reformat as
> XFS. This was the way to properly eliminate ReiserFS.
> If some of you have a little time, please look into it.

xfs will not trash the disk when it "formats" it.  this is most likely
what is happening.  since xfs is lazy in its allocation and use of disk
space, it probably did not blow away the reiser data structures allowing
you to still mount it as reiser.

when you ext2 formatted it, it went through and allocated ext2 inodes and
such and copies of the superblock throughout the disk killing any chance
of reiser being able to mount it.

what I am surprised about is that xfs did not replace the magic number
that tells the kernel and other tools what type of fs it is.  otherwise,
the reiser tools do not verify this number (or assume it could be
corrupted).

anyways, the best way to clean a partition is

dd if=/dev/zero of=/dev/hd[letter][number]

enough times that even the FBI cannot recover your data.

-tduffy



From owner-linux-xfs@oss.sgi.com Tue Mar 20 03:56:12 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2KBuCg24780
	for linux-xfs-outgoing; Tue, 20 Mar 2001 03:56:12 -0800
Received: from esparrall.udg.es (esparrall.udg.es [130.206.124.16])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2KBuAM24777
	for <linux-xfs@oss.sgi.com>; Tue, 20 Mar 2001 03:56:11 -0800
Received: from gcs by esparrall.udg.es with local (Exim 3.22 #1 (Debian))
	id 14fKiq-0006ob-00
	for <linux-xfs@oss.sgi.com>; Tue, 20 Mar 2001 12:54:44 +0100
Date: Tue, 20 Mar 2001 12:54:44 +0100
From: GCS <gcs@agentsinside.com>
To: linux-xfs@oss.sgi.com
Subject: Re: Bugreport - not really important
Message-ID: <20010320125444.A25268@esparrall.udg.es>
References: <20010320104356.A14454@esparrall.udg.es> <Pine.LNX.4.30.0103200223470.2403-100000@dbear.engr.sgi.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=iso-8859-1
Content-Disposition: inline
Content-Transfer-Encoding: 8bit
User-Agent: Mutt/1.3.15i
In-Reply-To: Tom Duffy wrote on K, =?iso-8859-1?Q?M=C1?=
	=?iso-8859-1?Q?R?= 20, 2001 at 02:29:30 -0800:
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk



Hello,

On K, MÁR 20, 2001 at 02:29:30 -0800, Tom Duffy wrote:

> xfs will not trash the disk when it "formats" it.  this is most likely
> what is happening.  since xfs is lazy in its allocation and use of disk
> space, it probably did not blow away the reiser data structures allowing
> you to still mount it as reiser.
 I think XFS should do it, or things like this will happen. I know that XFS
dinamicaly allocate more inodes as there is a need. I like this idea to being
lazy in this way. But do you know if it happens in the opposite direction?
Let's say I create a lot of file, XFS allocates a lot of inodes which take
a considerable amount of disk space. Then I delete my million numbers of files,
and tries to save a big file. Just to realize I have a lot of inodes for
nothing, but does not have enough space for the file itself. What does XFS
do in this case? Does it delete some inode blocks, or not?

> when you ext2 formatted it, it went through and allocated ext2 inodes and
> such and copies of the superblock throughout the disk killing any chance
> of reiser being able to mount it.
 Sure, mkfs.ext2 took more time.

> what I am surprised about is that xfs did not replace the magic number
> that tells the kernel and other tools what type of fs it is.  otherwise,
> the reiser tools do not verify this number (or assume it could be
> corrupted).
 I think it is more XFS than ReiserFS. At mount time ReiserFS print some
debug information, and it said 'clean mount' -> no problems. Thus I think the
magic number is not changed by XFS.

> anyways, the best way to clean a partition is
> dd if=/dev/zero of=/dev/hd[letter][number]
 Sure, but specify bs=16384 (or depends on your amount of memory; at least
my test show when copying a 4Gb slice, it makes things faster). :-)

> enough times that even the FBI cannot recover your data.
 One is enough isn't it?

Thanks, Laszlo

From owner-linux-xfs@oss.sgi.com Tue Mar 20 05:30:20 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2KDUKC26191
	for linux-xfs-outgoing; Tue, 20 Mar 2001 05:30:20 -0800
Received: from tux.mkp.net (tux.mkp.net [130.225.60.11])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2KDUJM26188
	for <linux-xfs@oss.sgi.com>; Tue, 20 Mar 2001 05:30:19 -0800
Received: from tux.mkp.net ([130.225.60.11] helo=jaguar.mkp.net)
	by tux.mkp.net with esmtp (Exim 3.16 #1)
	id 14fMDI-0008Mt-00; Tue, 20 Mar 2001 14:30:18 +0100
Received: (from mkp@localhost)
	by jaguar.mkp.net (8.9.3/8.9.3) id IAA13497;
	Tue, 20 Mar 2001 08:28:16 -0500
X-Authentication-Warning: jaguar.mkp.net: mkp set sender to mkp@mkp.net using -f
To: Joshua Baker-LePain <jlb17@duke.edu>
Cc: Linux xfs mailing list <linux-xfs@oss.sgi.com>
Subject: Re: Pre-production questions
References: <Pine.LNX.4.30.0103192249060.25325-100000@chaos.egr.duke.edu>
From: "Martin K. Petersen" <mkp@linuxcare.com>
Organization: Linuxcare, Inc.
Date: 20 Mar 2001 08:28:15 -0500
In-Reply-To: <Pine.LNX.4.30.0103192249060.25325-100000@chaos.egr.duke.edu>
Message-ID: <yq1g0g8r2mo.fsf@jaguar.mkp.net>
Lines: 26
User-Agent: Gnus/5.0808 (Gnus v5.8.8) XEmacs/21.1 (Canyonlands)
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

>>>>> "Joshua" == Joshua Baker-LePain <jlb17@duke.edu> writes:

Joshua> My RAID is an IDE-SCSI system -- IDE disks, looks like SCSI to
Joshua> the host.  It has 8 80GB drives in RAID 5, for about 560GB of
Joshua> usable space.  I'd like it to be just one partition.  My
Joshua> question is, are there any special parameters I should pass to
Joshua> mkfs for such a big (well, for me at least) partition?  What
Joshua> about to tailor it to the hardware RAID?

Well.  You could try and hint to the allocator how your RAID device is
set up.  Try and look up which chunk size it uses.  You can pass that
information on to mkfs using the sunit and swidth options.  See the
mkfs.xfs man page for details.

Please note that for a RAID5, choosing the right chunk size for the
workload the crucial for performance.  If you set it too high, you end
up doing a lot of I/O on every write.  If you set it too low, your
sustained I/O performance may suffer.

It all depends on your workload and the RAID array in question (How
much cache it has, etc.)  I recommend you experiment a bit.

-- 
Martin K. Petersen, Principal Linux Consultant, Linuxcare, Inc.
mkp@linuxcare.com, http://www.linuxcare.com/
SGI XFS for Linux Developer, http://oss.sgi.com/projects/xfs/

From owner-linux-xfs@oss.sgi.com Tue Mar 20 05:57:32 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2KDvWr26608
	for linux-xfs-outgoing; Tue, 20 Mar 2001 05:57:32 -0800
Received: from tux.mkp.net (tux.mkp.net [130.225.60.11])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2KDvVM26604
	for <linux-xfs@oss.sgi.com>; Tue, 20 Mar 2001 05:57:31 -0800
Received: from tux.mkp.net ([130.225.60.11] helo=jaguar.mkp.net)
	by tux.mkp.net with esmtp (Exim 3.16 #1)
	id 14fMdZ-0008Nc-00; Tue, 20 Mar 2001 14:57:26 +0100
Received: (from mkp@localhost)
	by jaguar.mkp.net (8.9.3/8.9.3) id IAA13555;
	Tue, 20 Mar 2001 08:55:23 -0500
X-Authentication-Warning: jaguar.mkp.net: mkp set sender to mkp@mkp.net using -f
To: GCS <gcs@agentsinside.com>
Cc: Tom Duffy <tduffy@engr.sgi.com>, <linux-xfs@oss.sgi.com>
Subject: Re: Bugreport - not really important
References: <Pine.LNX.4.30.0103200223470.2403-100000@dbear.engr.sgi.com>
From: "Martin K. Petersen" <mkp@linuxcare.com>
Organization: Linuxcare, Inc.
Date: 20 Mar 2001 08:55:22 -0500
In-Reply-To: <Pine.LNX.4.30.0103200223470.2403-100000@dbear.engr.sgi.com>
Message-ID: <yq1bsqwr1dh.fsf@jaguar.mkp.net>
Lines: 31
User-Agent: Gnus/5.0808 (Gnus v5.8.8) XEmacs/21.1 (Canyonlands)
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

>>>>> "Tom" == Tom Duffy <tduffy@engr.sgi.com> writes:

Tom> what I am surprised about is that xfs did not replace the magic
Tom> number that tells the kernel and other tools what type of fs it
Tom> is.  otherwise, the reiser tools do not verify this number (or
Tom> assume it could be corrupted).

The problem here is that every fs puts its superblock a different
place.  XFS has it on block 0, and reiserfs' is 64K from the beginning
of the device.

It so happens that in your case XFS doesn't write anything on that
position.  Consequently, the reiser superblock is intact after
mkfs.xfs.

There's really no smart way to avoid this as that would require
knowledge of all file systems on the planet.  Some may have their sb
at the middle or the end of the device etc.  And dd'ing zeroes to the
entire device will take forever.

I guess we could blast - say 70KB - of NULL chars to the start of the
partition.  As far as I can tell, that should get rid of any fs that
Linux currently supports (Unless they try and use backup copies in
that case.  I guess we'll have to experiment a bit).

Nathan, comments?

-- 
Martin K. Petersen, Principal Linux Consultant, Linuxcare, Inc.
mkp@linuxcare.com, http://www.linuxcare.com/
SGI XFS for Linux Developer, http://oss.sgi.com/projects/xfs/

From owner-linux-xfs@oss.sgi.com Tue Mar 20 07:14:19 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2KFEJa28401
	for linux-xfs-outgoing; Tue, 20 Mar 2001 07:14:19 -0800
Received: from thor.theseus.com (south.orl-pub.theseus.com [12.108.42.66])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2KFEIM28398
	for <linux-xfs@oss.sgi.com>; Tue, 20 Mar 2001 07:14:18 -0800
Received: from theseus.com (IDENT:thebs@fugitive.theseus.com [192.168.0.242])
	by thor.theseus.com (8.9.3/8.9.3) with ESMTP id KAA03322;
	Tue, 20 Mar 2001 10:18:37 -0500
Message-ID: <3AB773E0.E3919CE0@theseus.com>
Date: Tue, 20 Mar 2001 10:14:40 -0500
From: Bryan-TheBS-Smith <thebs@theseus.com>
Reply-To: thebs@theseus.com, b.j.smith@ieee.org
Organization: (Personal)
X-Mailer: Mozilla 4.76 [en] (X11; U; Linux 2.2.17-0.17-FUGITIVE i686)
X-Accept-Language: en
MIME-Version: 1.0
To: Tom Duffy <tduffy@engr.sgi.com>
CC: b.j.smith@ieee.org, Eric Sandeen <sandeen@jump.net>,
   Timothy Ball <timball@tux.org>, linux-xfs@oss.sgi.com
Subject: Re: mounting loop devices  Thanx for the mkinitrd-3.0.5 patch!
References: <Pine.LNX.4.30.0103200109490.2403-200000@dbear.engr.sgi.com>
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

Tom Duffy wrote:
>                                Name: mkinitrd-3.0.5-xfs.patch
>    mkinitrd-3.0.5-xfs.patch    Type: Plain Text (TEXT/PLAIN)
>                            Encoding: BASE64

Much appreciated dude!

-- TheBS

P.S.  Where would I normally find these patches?

--
Bryan "TheBS" Smith          chat:thebs413 @AOL/MSN/Yahoo
Engineer      mailto:b.j.smith@ieee.org,thebs@theseus.com
*********************************************************
"Never apply a Star Trek solution to a Babylon 5 problem"
                                    -- Nicholas C. Weaver

From owner-linux-xfs@oss.sgi.com Tue Mar 20 07:21:03 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2KFL3K28564
	for linux-xfs-outgoing; Tue, 20 Mar 2001 07:21:03 -0800
Received: from Cantor.suse.de (ns.suse.de [213.95.15.193])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2KFL2M28561
	for <linux-xfs@oss.sgi.com>; Tue, 20 Mar 2001 07:21:02 -0800
Received: from Hermes.suse.de (Hermes.suse.de [213.95.15.136])
	by Cantor.suse.de (Postfix) with ESMTP
	id 3D09A1E13E; Tue, 20 Mar 2001 16:21:01 +0100 (MET)
Date: Tue, 20 Mar 2001 16:20:43 +0100
From: Andi Kleen <ak@suse.de>
To: "Martin K. Petersen" <mkp@linuxcare.com>
Cc: GCS <gcs@agentsinside.com>, Tom Duffy <tduffy@engr.sgi.com>,
   linux-xfs@oss.sgi.com
Subject: Re: Bugreport - not really important
Message-ID: <20010320162043.A12187@gruyere.muc.suse.de>
References: <Pine.LNX.4.30.0103200223470.2403-100000@dbear.engr.sgi.com> <yq1bsqwr1dh.fsf@jaguar.mkp.net>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.2.5i
In-Reply-To: <yq1bsqwr1dh.fsf@jaguar.mkp.net>; from mkp@linuxcare.com on Tue, Mar 20, 2001 at 08:55:22AM -0500
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

On Tue, Mar 20, 2001 at 08:55:22AM -0500, Martin K. Petersen wrote:
> >>>>> "Tom" == Tom Duffy <tduffy@engr.sgi.com> writes:
> 
> Tom> what I am surprised about is that xfs did not replace the magic
> Tom> number that tells the kernel and other tools what type of fs it
> Tom> is.  otherwise, the reiser tools do not verify this number (or
> Tom> assume it could be corrupted).
> 
> The problem here is that every fs puts its superblock a different
> place.  XFS has it on block 0, and reiserfs' is 64K from the beginning
> of the device.

This means it is impossible to install lilo on a XFS partition (as opposed 
to the MBR) 


-Andi

From owner-linux-xfs@oss.sgi.com Tue Mar 20 09:00:02 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2KH02030848
	for linux-xfs-outgoing; Tue, 20 Mar 2001 09:00:02 -0800
Received: from deliverator.sgi.com (deliverator.sgi.com [204.94.214.10])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2KH01M30841
	for <linux-xfs@oss.sgi.com>; Tue, 20 Mar 2001 09:00:01 -0800
Received: from ledzep.americas.sgi.com (ledzep.americas.sgi.com [137.38.226.97]) by deliverator.sgi.com (980309.SGI.8.8.8-aspam-6.2/980310.SGI-aspam) via ESMTP id IAA06876
	for <linux-xfs@oss.sgi.com>; Tue, 20 Mar 2001 08:58:49 -0800 (PST)
	mail_from (cattelan@thebarn.com)
Received: from gibble.americas.sgi.com (gibble.americas.sgi.com [128.162.195.80]) by ledzep.americas.sgi.com (SGI-SGI-8.9.3/americas-smart-nospam1.1) with ESMTP id KAA78036; Tue, 20 Mar 2001 10:58:45 -0600 (CST)
Received: from thebarn.com (localhost [127.0.0.1])
	by gibble.americas.sgi.com (8.11.2/8.11.2) with ESMTP id f2KGnre16641;
	Tue, 20 Mar 2001 11:49:53 -0500
Message-ID: <3AB78A30.AB186508@thebarn.com>
Date: Tue, 20 Mar 2001 11:49:52 -0500
From: Russell Cattelan <cattelan@thebarn.com>
X-Mailer: Mozilla 4.76 [en] (X11; U; Linux 2.4.2-XFS i686)
X-Accept-Language: en
MIME-Version: 1.0
To: thebs@theseus.com
CC: linux-xfs@oss.sgi.com
Subject: Re: mounting loop devices  Thanx for the mkinitrd-3.0.5 patch!
References: <Pine.LNX.4.30.0103200109490.2403-200000@dbear.engr.sgi.com> <3AB773E0.E3919CE0@theseus.com>
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

Bryan-TheBS-Smith wrote:

> Tom Duffy wrote:
> >                                Name: mkinitrd-3.0.5-xfs.patch
> >    mkinitrd-3.0.5-xfs.patch    Type: Plain Text (TEXT/PLAIN)
> >                            Encoding: BASE64
>
> Much appreciated dude!
>
> -- TheBS
>
> P.S.  Where would I normally find these patches?
>

At some point mkinitrd was in the cmds tree, seems
it got dropped at some point.

I'll toss it back in there when I get a chance.



>
> --
> Bryan "TheBS" Smith          chat:thebs413 @AOL/MSN/Yahoo
> Engineer      mailto:b.j.smith@ieee.org,thebs@theseus.com
> *********************************************************
> "Never apply a Star Trek solution to a Babylon 5 problem"
>                                     -- Nicholas C. Weaver

--
Russell Cattelan
--
Digital Elves inc. -- Currently on loan to SGI
Linux XFS core developer.




From owner-linux-xfs@oss.sgi.com Tue Mar 20 10:18:43 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2KIIhf32212
	for linux-xfs-outgoing; Tue, 20 Mar 2001 10:18:43 -0800
Received: from chaos.egr.duke.edu (IDENT:root@chaos.egr.duke.edu [152.3.195.82])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2KIIgM32209
	for <linux-xfs@oss.sgi.com>; Tue, 20 Mar 2001 10:18:42 -0800
Received: from localhost (jlb@localhost)
	by chaos.egr.duke.edu (8.9.3/8.9.3) with ESMTP id NAA27517
	for <linux-xfs@oss.sgi.com>; Tue, 20 Mar 2001 13:18:41 -0500
X-Authentication-Warning: chaos.egr.duke.edu: jlb owned process doing -bs
Date: Tue, 20 Mar 2001 13:18:41 -0500 (EST)
From: Joshua Baker-LePain <jlb17@duke.edu>
X-Sender:  <jlb@chaos.egr.duke.edu>
To: Linux xfs mailing list <linux-xfs@oss.sgi.com>
Subject: Re: Pre-production questions
In-Reply-To: <yq1g0g8r2mo.fsf@jaguar.mkp.net>
Message-ID: <Pine.LNX.4.30.0103201256210.2796-100000@chaos.egr.duke.edu>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

On 20 Mar 2001 at 8:28am, Martin K. Petersen wrote

> >>>>> "Joshua" == Joshua Baker-LePain <jlb17@duke.edu> writes:
>
> Joshua> My RAID is an IDE-SCSI system -- IDE disks, looks like SCSI to
> Joshua> the host.  It has 8 80GB drives in RAID 5, for about 560GB of
> Joshua> usable space.  I'd like it to be just one partition.  My
> Joshua> question is, are there any special parameters I should pass to
> Joshua> mkfs for such a big (well, for me at least) partition?  What
> Joshua> about to tailor it to the hardware RAID?
>
> Well.  You could try and hint to the allocator how your RAID device is
> set up.  Try and look up which chunk size it uses.  You can pass that
> information on to mkfs using the sunit and swidth options.  See the
> mkfs.xfs man page for details.

The unit's default is a stripe size of 64 512 byte blocks.  The stripe
size can be set between 4 and 128 (incrementally, of course).  For the
defaults, would this be correct:

mkfs.xfs -d sunit=1,swidth=64

Or should I multiply both of those numbers by 512?  Also, should I make
the logfile any bigger than the default due to the size of the partition?

> Please note that for a RAID5, choosing the right chunk size for the
> workload the crucial for performance.  If you set it too high, you end
> up doing a lot of I/O on every write.  If you set it too low, your
> sustained I/O performance may suffer.
>
> It all depends on your workload and the RAID array in question (How
> much cache it has, etc.)  I recommend you experiment a bit.
>
The system is maxed out at 128MB of cache RAM.  The primary use will be
reading and writing (over NFS) lots of largish (250+MB) data files.

Sorry to pester everyone with my RAID naivete.  I do indeed plan on doing
lots of testing.  Thanks, all.

-- 
Joshua Baker-LePain
Department of Biomedical Engineering
Duke University


From owner-linux-xfs@oss.sgi.com Tue Mar 20 10:21:32 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2KILWR32251
	for linux-xfs-outgoing; Tue, 20 Mar 2001 10:21:32 -0800
Received: from pneumatic-tube.sgi.com (pneumatic-tube.sgi.com [204.94.214.22])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2KILWM32248
	for <linux-xfs@oss.sgi.com>; Tue, 20 Mar 2001 10:21:32 -0800
Received: from ledzep.americas.sgi.com (ledzep.americas.sgi.com [137.38.226.97]) by pneumatic-tube.sgi.com (980327.SGI.8.8.8-aspam/980310.SGI-aspam) via ESMTP id KAA00883
	for <linux-xfs@oss.sgi.com>; Tue, 20 Mar 2001 10:31:29 -0800 (PST)
	mail_from (cattelan@thebarn.com)
Received: from gibble.americas.sgi.com (gibble.americas.sgi.com [128.162.195.80]) by ledzep.americas.sgi.com (SGI-SGI-8.9.3/americas-smart-nospam1.1) with ESMTP id MAA81096; Tue, 20 Mar 2001 12:20:14 -0600 (CST)
Received: from thebarn.com (localhost [127.0.0.1])
	by gibble.americas.sgi.com (8.11.2/8.11.2) with ESMTP id f2KIBNe27420;
	Tue, 20 Mar 2001 13:11:23 -0500
Message-ID: <3AB79D4B.42DCD9A1@thebarn.com>
Date: Tue, 20 Mar 2001 13:11:23 -0500
From: Russell Cattelan <cattelan@thebarn.com>
X-Mailer: Mozilla 4.76 [en] (X11; U; Linux 2.4.2-XFS i686)
X-Accept-Language: en
MIME-Version: 1.0
To: thebs@theseus.com, b.j.smith@ieee.org, linux-xfs@oss.sgi.com
Subject: Re: mounting loop devices  Thanx for the mkinitrd-3.0.5 patch!
References: <Pine.LNX.4.30.0103200109490.2403-200000@dbear.engr.sgi.com> <3AB773E0.E3919CE0@theseus.com>
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

Bryan-TheBS-Smith wrote:

> Tom Duffy wrote:
> >                                Name: mkinitrd-3.0.5-xfs.patch
> >    mkinitrd-3.0.5-xfs.patch    Type: Plain Text (TEXT/PLAIN)
> >                            Encoding: BASE64
>
> Much appreciated dude!
>
> -- TheBS
>
> P.S.  Where would I normally find these patches?

Opps sorry it wasn't removed.
cmd/xfstests/misc/mkinitrd.xfs

Suppose we could find a better place for this; "xfstests"  is a bit
misleading.



>
>
> --
> Bryan "TheBS" Smith          chat:thebs413 @AOL/MSN/Yahoo
> Engineer      mailto:b.j.smith@ieee.org,thebs@theseus.com
> *********************************************************
> "Never apply a Star Trek solution to a Babylon 5 problem"
>                                     -- Nicholas C. Weaver

--
Russell Cattelan
--
Digital Elves inc. -- Currently on loan to SGI
Linux XFS core developer.




From owner-linux-xfs@oss.sgi.com Tue Mar 20 10:33:18 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2KIXIu32447
	for linux-xfs-outgoing; Tue, 20 Mar 2001 10:33:18 -0800
Received: from pneumatic-tube.sgi.com (pneumatic-tube.sgi.com [204.94.214.22])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2KIXHM32444
	for <linux-xfs@oss.sgi.com>; Tue, 20 Mar 2001 10:33:17 -0800
Received: from cthulhu.engr.sgi.com (gate3-relay.engr.sgi.com [130.62.1.234]) by pneumatic-tube.sgi.com (980327.SGI.8.8.8-aspam/980310.SGI-aspam) via ESMTP id KAA05859
	for <linux-xfs@oss.sgi.com>; Tue, 20 Mar 2001 10:43:15 -0800 (PST)
	mail_from (tduffy@engr.sgi.com)
Received: from dbear.engr.sgi.com (dbear.engr.sgi.com [163.154.18.85])
	by cthulhu.engr.sgi.com (SGI-8.9.3/8.9.3) with ESMTP id KAA71918;
	Tue, 20 Mar 2001 10:30:46 -0800 (PST)
Date: Tue, 20 Mar 2001 10:26:55 -0800 (PST)
From: Tom Duffy <tduffy@engr.sgi.com>
To: Russell Cattelan <cattelan@thebarn.com>
cc: <thebs@theseus.com>, <b.j.smith@ieee.org>, <linux-xfs@oss.sgi.com>
Subject: Re: mounting loop devices  Thanx for the mkinitrd-3.0.5 patch!
In-Reply-To: <3AB79D4B.42DCD9A1@thebarn.com>
Message-ID: <Pine.LNX.4.30.0103201024210.2403-100000@dbear.engr.sgi.com>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

> Opps sorry it wasn't removed.
> cmd/xfstests/misc/mkinitrd.xfs
>
> Suppose we could find a better place for this; "xfstests"  is a bit
> misleading.

right, but this is still 2.6 based.  should I check in my 3.0.9 based
mkinitrd?

btw, jeremy brown will be adding some functionality real soon to mkinitrd
to detect which filesystem you will need for boot and slap that into the
mkinitrd (we were getting problems with reiserfs as well).  it does not
make sense to include xfs and pagebuf in the initrd if you do not have to.

I guess we should check that in as well once it is available.

-tduffy


From owner-linux-xfs@oss.sgi.com Tue Mar 20 11:19:50 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2KJJol00670
	for linux-xfs-outgoing; Tue, 20 Mar 2001 11:19:50 -0800
Received: from pneumatic-tube.sgi.com (pneumatic-tube.sgi.com [204.94.214.22])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2KJJoM00667
	for <linux-xfs@oss.sgi.com>; Tue, 20 Mar 2001 11:19:50 -0800
Received: from ledzep.americas.sgi.com (ledzep.americas.sgi.com [137.38.226.97]) by pneumatic-tube.sgi.com (980327.SGI.8.8.8-aspam/980310.SGI-aspam) via ESMTP id LAA05269
	for <linux-xfs@oss.sgi.com>; Tue, 20 Mar 2001 11:29:47 -0800 (PST)
	mail_from (cattelan@thebarn.com)
Received: from gibble.americas.sgi.com (gibble.americas.sgi.com [128.162.195.80]) by ledzep.americas.sgi.com (SGI-SGI-8.9.3/americas-smart-nospam1.1) with ESMTP id NAA74964; Tue, 20 Mar 2001 13:18:32 -0600 (CST)
Received: from thebarn.com (localhost [127.0.0.1])
	by gibble.americas.sgi.com (8.11.2/8.11.2) with ESMTP id f2KJ9ee30029;
	Tue, 20 Mar 2001 14:09:40 -0500
Message-ID: <3AB7AAF2.3045DAF0@thebarn.com>
Date: Tue, 20 Mar 2001 14:09:38 -0500
From: Russell Cattelan <cattelan@thebarn.com>
X-Mailer: Mozilla 4.76 [en] (X11; U; Linux 2.4.2-XFS i686)
X-Accept-Language: en
MIME-Version: 1.0
To: Tom Duffy <tduffy@engr.sgi.com>
CC: thebs@theseus.com, b.j.smith@ieee.org, linux-xfs@oss.sgi.com
Subject: Re: mounting loop devices  Thanx for the mkinitrd-3.0.5 patch!
References: <Pine.LNX.4.30.0103201024210.2403-100000@dbear.engr.sgi.com>
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

Tom Duffy wrote:

> > Opps sorry it wasn't removed.
> > cmd/xfstests/misc/mkinitrd.xfs
> >
> > Suppose we could find a better place for this; "xfstests"  is a bit
> > misleading.
>
> right, but this is still 2.6 based.  should I check in my 3.0.9 based
> mkinitrd?

Hmm is it RedHat specific? The last time I updated mkinitrd I had revert the
mktemp portion; as Redhat seems to be the only one that supports mktemp -d
(aka temp directory)


>
>
> btw, jeremy brown will be adding some functionality real soon to mkinitrd
> to detect which filesystem you will need for boot and slap that into the
> mkinitrd (we were getting problems with reiserfs as well).  it does not
> make sense to include xfs and pagebuf in the initrd if you do not have to.

>
>
> I guess we should check that in as well once it is available.

I guess we can defer the whole thing for now.

The rpm's I'm building for prerelease 0.10 has XFS build into the kernel
which
renders the changes to mkinitrd un necessary unless somebody really wants
have
XFS as a module again.

For now I can export your is now that I know about it.


>
>
> -tduffy

--
Russell Cattelan
--
Digital Elves inc. -- Currently on loan to SGI
Linux XFS core developer.




From owner-linux-xfs@oss.sgi.com Tue Mar 20 12:14:04 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2KKE4J01970
	for linux-xfs-outgoing; Tue, 20 Mar 2001 12:14:04 -0800
Received: from tux.mkp.net (tux.mkp.net [130.225.60.11])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2KKE2M01966
	for <linux-xfs@oss.sgi.com>; Tue, 20 Mar 2001 12:14:02 -0800
Received: from tux.mkp.net ([130.225.60.11] helo=jaguar.mkp.net)
	by tux.mkp.net with esmtp (Exim 3.16 #1)
	id 14fSVw-0000Bz-00; Tue, 20 Mar 2001 21:14:01 +0100
Received: (from mkp@localhost)
	by jaguar.mkp.net (8.9.3/8.9.3) id PAA14133;
	Tue, 20 Mar 2001 15:11:56 -0500
X-Authentication-Warning: jaguar.mkp.net: mkp set sender to mkp@mkp.net using -f
To: linux-xfs@oss.sgi.com
Subject: TAKE - Upgrade LVM to 0.9beta6
From: "Martin K. Petersen" <mkp@linuxcare.com>
Organization: Linuxcare, Inc.
Date: 20 Mar 2001 15:11:55 -0500
Message-ID: <yq1ofuwp5dg.fsf@jaguar.mkp.net>
Lines: 27
User-Agent: Gnus/5.0808 (Gnus v5.8.8) XEmacs/21.1 (Canyonlands)
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk


Date:  Tue Mar 20 12:10:08 PST 2001
Workarea:  sshgate.corp.sgi.com:/export/d0/mkp/XFS/slinx-beta6

The following file(s) were checked into:
  bonnie.engr.sgi.com:/isms/slinx/2.4.x-xfs

Modid:  2.4.x-xfs:slinx:90447a
linux/drivers/md/lvm-internal.h - 1.1
linux/drivers/md/lvm-fs.c - 1.1
linux/include/linux/lvm.h - 1.7
linux/drivers/md/lvm.c - 1.8
linux/drivers/md/Makefile - 1.4
linux/drivers/md/lvm-snap.c - 1.4
cmd/xfsprogs/include/lvm_user.h - 1.3
cmd/xfsprogs/include/lvm.h - 1.3
cmd/xfsprogs/include/liblvm.h - 1.3
cmd/xfsprogs/doc/CHANGES - 1.8
cmd/xfsprogs/VERSION - 1.8
cmd/xfsprogs/include/lvm_log.h - 1.2
cmd/xfsprogs/include/lvm_config.h - 1.2
        - Upgrade LVM to 0.9beta6

-- 
Martin K. Petersen, Principal Linux Consultant, Linuxcare, Inc.
mkp@linuxcare.com, http://www.linuxcare.com/
SGI XFS for Linux Developer, http://oss.sgi.com/projects/xfs/

From owner-linux-xfs@oss.sgi.com Tue Mar 20 12:17:39 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2KKHdS02082
	for linux-xfs-outgoing; Tue, 20 Mar 2001 12:17:39 -0800
Received: from tux.mkp.net (tux.mkp.net [130.225.60.11])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2KKHbM02079
	for <linux-xfs@oss.sgi.com>; Tue, 20 Mar 2001 12:17:38 -0800
Received: from tux.mkp.net ([130.225.60.11] helo=jaguar.mkp.net)
	by tux.mkp.net with esmtp (Exim 3.16 #1)
	id 14fSZL-0000CG-00; Tue, 20 Mar 2001 21:17:32 +0100
Received: (from mkp@localhost)
	by jaguar.mkp.net (8.9.3/8.9.3) id PAA14139;
	Tue, 20 Mar 2001 15:15:26 -0500
X-Authentication-Warning: jaguar.mkp.net: mkp set sender to mkp@mkp.net using -f
To: Andi Kleen <ak@suse.de>
Cc: linux-xfs@oss.sgi.com
Subject: Re: Bugreport - not really important
References: <Pine.LNX.4.30.0103200223470.2403-100000@dbear.engr.sgi.com>
	<yq1bsqwr1dh.fsf@jaguar.mkp.net>
	<20010320162043.A12187@gruyere.muc.suse.de>
From: "Martin K. Petersen" <mkp@linuxcare.com>
Organization: Linuxcare, Inc.
Date: 20 Mar 2001 15:15:26 -0500
In-Reply-To: <20010320162043.A12187@gruyere.muc.suse.de>
Message-ID: <yq1g0g8p57l.fsf@jaguar.mkp.net>
Lines: 16
User-Agent: Gnus/5.0808 (Gnus v5.8.8) XEmacs/21.1 (Canyonlands)
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

>>>>> "Andi" == Andi Kleen <ak@suse.de> writes:

>> The problem here is that every fs puts its superblock a different
>> place.  XFS has it on block 0, and reiserfs' is 64K from the
>> beginning of the device.

Andi> This means it is impossible to install lilo on a XFS partition
Andi> (as opposed to the MBR)

Yeah, well.  IRIX people put it there long time ago, and keeping the
on-disk format the same is important to us.

-- 
Martin K. Petersen, Principal Linux Consultant, Linuxcare, Inc.
mkp@linuxcare.com, http://www.linuxcare.com/
SGI XFS for Linux Developer, http://oss.sgi.com/projects/xfs/

From owner-linux-xfs@oss.sgi.com Tue Mar 20 12:23:39 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2KKNdV02187
	for linux-xfs-outgoing; Tue, 20 Mar 2001 12:23:39 -0800
Received: from tux.mkp.net (tux.mkp.net [130.225.60.11])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2KKNbM02184
	for <linux-xfs@oss.sgi.com>; Tue, 20 Mar 2001 12:23:38 -0800
Received: from tux.mkp.net ([130.225.60.11] helo=jaguar.mkp.net)
	by tux.mkp.net with esmtp (Exim 3.16 #1)
	id 14fSfG-0000Cf-00; Tue, 20 Mar 2001 21:23:36 +0100
Received: (from mkp@localhost)
	by jaguar.mkp.net (8.9.3/8.9.3) id PAA14154;
	Tue, 20 Mar 2001 15:21:34 -0500
X-Authentication-Warning: jaguar.mkp.net: mkp set sender to mkp@mkp.net using -f
To: Joshua Baker-LePain <jlb17@duke.edu>
Cc: Linux xfs mailing list <linux-xfs@oss.sgi.com>
Subject: Re: Pre-production questions
References: <Pine.LNX.4.30.0103201256210.2796-100000@chaos.egr.duke.edu>
From: "Martin K. Petersen" <mkp@linuxcare.com>
Organization: Linuxcare, Inc.
Date: 20 Mar 2001 15:21:34 -0500
In-Reply-To: <Pine.LNX.4.30.0103201256210.2796-100000@chaos.egr.duke.edu>
Message-ID: <yq1bsqwp4xd.fsf@jaguar.mkp.net>
Lines: 35
User-Agent: Gnus/5.0808 (Gnus v5.8.8) XEmacs/21.1 (Canyonlands)
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

>>>>> "Joshua" == Joshua Baker-LePain <jlb17@duke.edu> writes:

Joshua> The unit's default is a stripe size of 64 512 byte blocks.
Joshua> The stripe size can be set between 4 and 128 (incrementally,
Joshua> of course).  For the defaults, would this be correct:

Joshua> mkfs.xfs -d sunit=1,swidth=64

sunit = stripe width / number of disks

So mkfs.xfs -d sunit=8,swidth=64 for your 8 disks.


Joshua> Also, should I make the logfile any bigger than the default
Joshua> due to the size of the partition?

Depends how log-heavy the fs will be.  But in general we tend to use a
bigger log than the 1000b default.

We've talked about adding heuristics to mkfs to make a qualified
guess.  That's on my todo list.


Joshua> The system is maxed out at 128MB of cache RAM.  The primary
Joshua> use will be reading and writing (over NFS) lots of largish
Joshua> (250+MB) data files.

In that case I'd probably up the stripe size a bit.  But do some
benchmarks for yourself before you decide.  RAID controllers vary a
lot.

-- 
Martin K. Petersen, Principal Linux Consultant, Linuxcare, Inc.
mkp@linuxcare.com, http://www.linuxcare.com/
SGI XFS for Linux Developer, http://oss.sgi.com/projects/xfs/

From owner-linux-xfs@oss.sgi.com Tue Mar 20 12:25:16 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2KKPGA02207
	for linux-xfs-outgoing; Tue, 20 Mar 2001 12:25:16 -0800
Received: from pneumatic-tube.sgi.com (pneumatic-tube.sgi.com [204.94.214.22])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2KKPGM02204
	for <linux-xfs@oss.sgi.com>; Tue, 20 Mar 2001 12:25:16 -0800
Received: from zeus-fddi.americas.sgi.com (zeus-fddi.americas.sgi.com [128.162.8.103]) by pneumatic-tube.sgi.com (980327.SGI.8.8.8-aspam/980310.SGI-aspam) via ESMTP id MAA08713
	for <linux-xfs@oss.sgi.com>; Tue, 20 Mar 2001 12:35:14 -0800 (PST)
	mail_from (lord@sgi.com)
Received: from daisy-e185.americas.sgi.com (daisy-e194.americas.sgi.com [128.162.194.214]) by zeus-fddi.americas.sgi.com (8.9.3/americas-smart-nospam1.1) with ESMTP id OAA1202193; Tue, 20 Mar 2001 14:23:58 -0600 (CST)
Received: from jen.americas.sgi.com (IDENT:root@jen.americas.sgi.com [128.162.184.86]) by daisy-e185.americas.sgi.com (SGI-8.9.3/SGI-server-1.7) with ESMTP id OAA25935; Tue, 20 Mar 2001 14:23:58 -0600 (CST)
Received: from jen.americas.sgi.com by jen.americas.sgi.com (8.11.0/SGI-client-1.7) via ESMTP id f2KKP3s07381; Tue, 20 Mar 2001 14:25:03 -0600
Message-Id: <200103202025.f2KKP3s07381@jen.americas.sgi.com>
X-Mailer: exmh version 2.2 06/23/2000 with nmh-1.0.4
To: Joshua Baker-LePain <jlb17@duke.edu>
cc: Linux xfs mailing list <linux-xfs@oss.sgi.com>
Subject: Re: Pre-production questions 
In-Reply-To: Message from Joshua Baker-LePain <jlb17@duke.edu> 
   of "Tue, 20 Mar 2001 13:18:41 EST." <Pine.LNX.4.30.0103201256210.2796-100000@chaos.egr.duke.edu> 
Date: Tue, 20 Mar 2001 14:25:03 -0600
From: Steve Lord <lord@sgi.com>
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

> On 20 Mar 2001 at 8:28am, Martin K. Petersen wrote
> 
> > >>>>> "Joshua" == Joshua Baker-LePain <jlb17@duke.edu> writes:
> >
> > Joshua> My RAID is an IDE-SCSI system -- IDE disks, looks like SCSI to
> > Joshua> the host.  It has 8 80GB drives in RAID 5, for about 560GB of
> > Joshua> usable space.  I'd like it to be just one partition.  My
> > Joshua> question is, are there any special parameters I should pass to
> > Joshua> mkfs for such a big (well, for me at least) partition?  What
> > Joshua> about to tailor it to the hardware RAID?
> >
> > Well.  You could try and hint to the allocator how your RAID device is
> > set up.  Try and look up which chunk size it uses.  You can pass that
> > information on to mkfs using the sunit and swidth options.  See the
> > mkfs.xfs man page for details.
> 
> The unit's default is a stripe size of 64 512 byte blocks.  The stripe
> size can be set between 4 and 128 (incrementally, of course).  For the
> defaults, would this be correct:
> 
> mkfs.xfs -d sunit=1,swidth=64
> 
> Or should I multiply both of those numbers by 512?  Also, should I make
> the logfile any bigger than the default due to the size of the partition?
> 

Actually this is incorrect, you need

	-d sunit=64,swidth=448

The sunit is the amount of data on each disk in 512 byte chunks, the swidth
is the total amount of data before we go back to the first disk again. You
have 8 disks in raid5 so 7 data disks and a parity disk. You need 7 * 64 as
the stripe width.

You probably want a bigger log file, you have lots of room to play with,
doing 16384b or 32768b is not going to hurt you. You should also mount
with -o logbufs=8 if you expect heavy traffic.

Steve

p.s. Let us know if these numbers make things go faster!



From owner-linux-xfs@oss.sgi.com Tue Mar 20 12:43:30 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2KKhU602487
	for linux-xfs-outgoing; Tue, 20 Mar 2001 12:43:30 -0800
Received: from tux.mkp.net (tux.mkp.net [130.225.60.11])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2KKhTM02484
	for <linux-xfs@oss.sgi.com>; Tue, 20 Mar 2001 12:43:29 -0800
Received: from tux.mkp.net ([130.225.60.11] helo=jaguar.mkp.net)
	by tux.mkp.net with esmtp (Exim 3.16 #1)
	id 14fSyR-0000E6-00; Tue, 20 Mar 2001 21:43:25 +0100
Received: (from mkp@localhost)
	by jaguar.mkp.net (8.9.3/8.9.3) id PAA14163;
	Tue, 20 Mar 2001 15:41:22 -0500
X-Authentication-Warning: jaguar.mkp.net: mkp set sender to mkp@mkp.net using -f
To: Steve Lord <lord@sgi.com>
Cc: Joshua Baker-LePain <jlb17@duke.edu>,
   Linux xfs mailing list <linux-xfs@oss.sgi.com>
Subject: Re: Pre-production questions
References: <200103202025.f2KKP3s07381@jen.americas.sgi.com>
From: "Martin K. Petersen" <mkp@linuxcare.com>
Organization: Linuxcare, Inc.
Date: 20 Mar 2001 15:41:22 -0500
In-Reply-To: <200103202025.f2KKP3s07381@jen.americas.sgi.com>
Message-ID: <yq13dc8p40d.fsf@jaguar.mkp.net>
Lines: 21
User-Agent: Gnus/5.0808 (Gnus v5.8.8) XEmacs/21.1 (Canyonlands)
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

>>>>> "Steve" == Steve Lord <lord@sgi.com> writes:

>> mkfs.xfs -d sunit=1,swidth=64
>>
>> Or should I multiply both of those numbers by 512?  Also, should I
>> make the logfile any bigger than the default due to the size of the
>> partition?

Steve> Actually this is incorrect, you need

Steve> 	-d sunit=64,swidth=448

Hmmm. I read the 64KB as across all the disks.  Usually RAID5 arrays
tend to use very small values compared to RAID0/1.

Joshua - check your manual to make sure what they mean.

-- 
Martin K. Petersen, Principal Linux Consultant, Linuxcare, Inc.
mkp@linuxcare.com, http://www.linuxcare.com/
SGI XFS for Linux Developer, http://oss.sgi.com/projects/xfs/

From owner-linux-xfs@oss.sgi.com Tue Mar 20 12:54:17 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2KKsHx02856
	for linux-xfs-outgoing; Tue, 20 Mar 2001 12:54:17 -0800
Received: from pipt.oz.cc.utah.edu (jdr1529@pipt.oz.cc.utah.edu [155.99.2.7])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2KKsFM02852
	for <linux-xfs@oss.sgi.com>; Tue, 20 Mar 2001 12:54:16 -0800
Received: from localhost (jdr1529@localhost)
	by pipt.oz.cc.utah.edu (8.9.2/8.9.2) with ESMTP id NAA03592
	for <linux-xfs@oss.sgi.com>; Tue, 20 Mar 2001 13:54:13 -0700 (MST)
Date: Tue, 20 Mar 2001 13:54:12 -0700 (MST)
From: james rich <james.rich@m.cc.utah.edu>
To: linux-xfs@oss.sgi.com
Subject: Re: Bugreport - not really important
In-Reply-To: <yq1g0g8p57l.fsf@jaguar.mkp.net>
Message-ID: <Pine.GSO.4.05.10103201351520.19852-100000@pipt.oz.cc.utah.edu>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

On 20 Mar 2001, Martin K. Petersen wrote:

> >>>>> "Andi" == Andi Kleen <ak@suse.de> writes:
> 
> >> The problem here is that every fs puts its superblock a different
> >> place.  XFS has it on block 0, and reiserfs' is 64K from the
> >> beginning of the device.
> 
> Andi> This means it is impossible to install lilo on a XFS partition
> Andi> (as opposed to the MBR)
> 
> Yeah, well.  IRIX people put it there long time ago, and keeping the
> on-disk format the same is important to us.

I have to agree here - compatibilty with IRIX is far more important,
especially considering that LILO can go on the MBR.

James Rich
james.rich@m.cc.utah.edu


From owner-linux-xfs@oss.sgi.com Tue Mar 20 13:35:43 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2KLZhA03477
	for linux-xfs-outgoing; Tue, 20 Mar 2001 13:35:43 -0800
Received: from sgi.com (sgi.SGI.COM [192.48.153.1])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2KLZgM03474
	for <linux-xfs@oss.sgi.com>; Tue, 20 Mar 2001 13:35:42 -0800
Received: from zeus-fddi.americas.sgi.com (128-162-8-103.americas.sgi.com [128.162.8.103]) 
	by sgi.com (980327.SGI.8.8.8-aspam/980304.SGI-aspam:
       SGI does not authorize the use of its proprietary
       systems or networks for unsolicited or bulk email
       from the Internet.) 
	via ESMTP id NAA06607
	for <linux-xfs@oss.sgi.com>; Tue, 20 Mar 2001 13:35:30 -0800 (PST)
	mail_from (lord@sgi.com)
Received: from daisy-e185.americas.sgi.com (daisy-e194.americas.sgi.com [128.162.194.214]) by zeus-fddi.americas.sgi.com (8.9.3/americas-smart-nospam1.1) with ESMTP id PAA1206362 for <linux-xfs@oss.sgi.com>; Tue, 20 Mar 2001 15:34:09 -0600 (CST)
Received: from jen.americas.sgi.com (IDENT:root@jen.americas.sgi.com [128.162.184.86]) by daisy-e185.americas.sgi.com (SGI-8.9.3/SGI-server-1.7) with ESMTP id PAA97828 for <linux-xfs@oss.sgi.com>; Tue, 20 Mar 2001 15:34:09 -0600 (CST)
From: Steve Lord <lord@sgi.com>
Received: by jen.americas.sgi.com (8.11.0/SGI-client-1.7) id f2KLZEl22129; Tue, 20 Mar 2001 15:35:14 -0600
Message-Id: <200103202135.f2KLZEl22129@jen.americas.sgi.com>
Date: Tue, 20 Mar 2001 15:35:14 -0600
Subject: TAKE - Fix a recovery panic
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

If you have a large and dirty log and a fast enough system you can
get into knots during recovery. This fixes the problem, and cleans up
some more pagebuf crud.

Date:  Tue Mar 20 13:33:07 PST 2001
Workarea:  jen.americas.sgi.com:/src/lord/xfs-linux.2.4

The following file(s) were checked into:
  bonnie.engr.sgi.com:/isms/slinx/2.4.x-xfs


Modid:  2.4.x-xfs:slinx:90451a
linux/include/linux/page_buf.h - 1.83
	- remove argument from lookup_pages call.

linux/fs/pagebuf/page_buf.c - 1.66
linux/fs/pagebuf/page_buf_io.c - 1.69
	- Do not automatically set the mapped flag in a pagebuf, only set it when
	  we actually map it. In the pagebuf_get path when mapping a pagebuf,
	  free some address space if there is a backlock waiting to be freed.
	  More code cleanup of paths no longer used and dead code, plus on
	  the I/O to disk path make the kiobuf code ifdef able and simplify
	  the buffer head case where possible.



From owner-linux-xfs@oss.sgi.com Tue Mar 20 15:17:12 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2KNHCF05603
	for linux-xfs-outgoing; Tue, 20 Mar 2001 15:17:12 -0800
Received: from deliverator.sgi.com (deliverator.sgi.com [204.94.214.10])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2KNHBM05599
	for <linux-xfs@oss.sgi.com>; Tue, 20 Mar 2001 15:17:11 -0800
Received: from sydney.sydney.sgi.com (sydney.sydney.sgi.com [134.14.48.2]) by deliverator.sgi.com (980309.SGI.8.8.8-aspam-6.2/980310.SGI-aspam) via SMTP id PAA10634
	for <linux-xfs@oss.sgi.com>; Tue, 20 Mar 2001 15:15:57 -0800 (PST)
	mail_from (kaos@melbourne.sgi.com)
Received: from kao2.melbourne.sgi.com by sydney.sydney.sgi.com via ESMTP (950413.SGI.8.6.12/930416.SGI)
	 id KAA01120; Wed, 21 Mar 2001 10:15:37 +1100
X-Mailer: exmh version 2.1.1 10/15/1999
From: Keith Owens <kaos@melbourne.sgi.com>
To: "Martin K. Petersen" <mkp@linuxcare.com>
cc: GCS <gcs@agentsinside.com>, Tom Duffy <tduffy@engr.sgi.com>,
   linux-xfs@oss.sgi.com
Subject: Re: Bugreport - not really important 
In-reply-to: Your message of "20 Mar 2001 08:55:22 CDT."
             <yq1bsqwr1dh.fsf@jaguar.mkp.net> 
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Date: Wed, 21 Mar 2001 10:15:42 +1100
Message-ID: <5985.985130142@kao2.melbourne.sgi.com>
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

On 20 Mar 2001 08:55:22 -0500, 
"Martin K. Petersen" <mkp@linuxcare.com> wrote:
>I guess we could blast - say 70KB - of NULL chars to the start of the
>partition.  As far as I can tell, that should get rid of any fs that
>Linux currently supports (Unless they try and use backup copies in
>that case.  I guess we'll have to experiment a bit).

EFI partitions are basically VFAT but they have a backup copy of the
FAT at the end of the partition.  People have wiped the start of EFI
partitions and it still comes back as EFI.


From owner-linux-xfs@oss.sgi.com Tue Mar 20 15:25:23 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2KNPNW05783
	for linux-xfs-outgoing; Tue, 20 Mar 2001 15:25:23 -0800
Received: from tux.mkp.net (tux.mkp.net [130.225.60.11])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2KNPMM05780
	for <linux-xfs@oss.sgi.com>; Tue, 20 Mar 2001 15:25:22 -0800
Received: from tux.mkp.net ([130.225.60.11] helo=jaguar.mkp.net)
	by tux.mkp.net with esmtp (Exim 3.16 #1)
	id 14fVV5-0000Ka-00; Wed, 21 Mar 2001 00:25:20 +0100
Received: (from mkp@localhost)
	by jaguar.mkp.net (8.9.3/8.9.3) id SAA14238;
	Tue, 20 Mar 2001 18:25:04 -0500
X-Authentication-Warning: jaguar.mkp.net: mkp set sender to mkp@mkp.net using -f
To: Keith Owens <kaos@melbourne.sgi.com>
Cc: linux-xfs@oss.sgi.com
Subject: Re: Bugreport - not really important
References: <5985.985130142@kao2.melbourne.sgi.com>
From: "Martin K. Petersen" <mkp@linuxcare.com>
Organization: Linuxcare, Inc.
Date: 20 Mar 2001 18:25:04 -0500
In-Reply-To: <5985.985130142@kao2.melbourne.sgi.com>
Message-ID: <yq1pufcnhv3.fsf@jaguar.mkp.net>
Lines: 17
User-Agent: Gnus/5.0808 (Gnus v5.8.8) XEmacs/21.1 (Canyonlands)
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

>>>>> "Keith" == Keith Owens <kaos@melbourne.sgi.com> writes:

>> I guess we could blast - say 70KB - of NULL chars to the start of
>> the partition.  As far as I can tell, that should get rid of any fs
>> that Linux currently supports (Unless they try and use backup
>> copies in that case.  I guess we'll have to experiment a bit).

Keith> EFI partitions are basically VFAT but they have a backup copy
Keith> of the FAT at the end of the partition.  People have wiped the
Keith> start of EFI partitions and it still comes back as EFI.

Bugger.  We need to clear out some space at the end too then...

-- 
Martin K. Petersen, Principal Linux Consultant, Linuxcare, Inc.
mkp@linuxcare.com, http://www.linuxcare.com/
SGI XFS for Linux Developer, http://oss.sgi.com/projects/xfs/

From owner-linux-xfs@oss.sgi.com Tue Mar 20 15:42:33 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2KNgXO05966
	for linux-xfs-outgoing; Tue, 20 Mar 2001 15:42:33 -0800
Received: from yog-sothoth.sgi.com (eugate.sgi.com [192.48.160.10])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2KNgWM05963
	for <linux-xfs@oss.sgi.com>; Tue, 20 Mar 2001 15:42:32 -0800
Received: from zeus-fddi.americas.sgi.com (zeus-fddi.americas.sgi.com [128.162.8.103]) by yog-sothoth.sgi.com (980305.SGI.8.8.8-aspam-6.2/980304.SGI-aspam-europe) via ESMTP id AAA4717510
	for <linux-xfs@oss.sgi.com>; Wed, 21 Mar 2001 00:42:31 +0100 (CET)
	mail_from (lord@sgi.com)
Received: from daisy-e185.americas.sgi.com (daisy-e194.americas.sgi.com [128.162.194.214]) by zeus-fddi.americas.sgi.com (8.9.3/americas-smart-nospam1.1) with ESMTP id RAA1170990; Tue, 20 Mar 2001 17:41:09 -0600 (CST)
Received: from jen.americas.sgi.com (IDENT:root@jen.americas.sgi.com [128.162.184.86]) by daisy-e185.americas.sgi.com (SGI-8.9.3/SGI-server-1.7) with ESMTP id RAA87409; Tue, 20 Mar 2001 17:41:08 -0600 (CST)
Received: from jen.americas.sgi.com by jen.americas.sgi.com (8.11.0/SGI-client-1.7) via ESMTP id f2KNgC421484; Tue, 20 Mar 2001 17:42:12 -0600
Message-Id: <200103202342.f2KNgC421484@jen.americas.sgi.com>
X-Mailer: exmh version 2.2 06/23/2000 with nmh-1.0.4
To: "Martin K. Petersen" <mkp@linuxcare.com>
cc: Keith Owens <kaos@melbourne.sgi.com>, linux-xfs@oss.sgi.com
Subject: Re: Bugreport - not really important 
In-Reply-To: Message from "Martin K. Petersen" <mkp@linuxcare.com> 
   of "20 Mar 2001 18:25:04 EST." <yq1pufcnhv3.fsf@jaguar.mkp.net> 
Date: Tue, 20 Mar 2001 17:42:12 -0600
From: Steve Lord <lord@sgi.com>
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

> >>>>> "Keith" == Keith Owens <kaos@melbourne.sgi.com> writes:
> 
> >> I guess we could blast - say 70KB - of NULL chars to the start of
> >> the partition.  As far as I can tell, that should get rid of any fs
> >> that Linux currently supports (Unless they try and use backup
> >> copies in that case.  I guess we'll have to experiment a bit).
> 
> Keith> EFI partitions are basically VFAT but they have a backup copy
> Keith> of the FAT at the end of the partition.  People have wiped the
> Keith> start of EFI partitions and it still comes back as EFI.
> 
> Bugger.  We need to clear out some space at the end too then...

We already write one block out there....

So does every other mkfs go to great care to trash other filesystems
superblocks? I don't see why we should if they don't. 

Steve

> 
> -- 
> Martin K. Petersen, Principal Linux Consultant, Linuxcare, Inc.
> mkp@linuxcare.com, http://www.linuxcare.com/
> SGI XFS for Linux Developer, http://oss.sgi.com/projects/xfs/



From owner-linux-xfs@oss.sgi.com Tue Mar 20 15:47:13 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2KNlDw06036
	for linux-xfs-outgoing; Tue, 20 Mar 2001 15:47:13 -0800
Received: from deliverator.sgi.com (deliverator.sgi.com [204.94.214.10])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2KNlCM06033
	for <linux-xfs@oss.sgi.com>; Tue, 20 Mar 2001 15:47:12 -0800
Received: from larry.melbourne.sgi.com (larry.melbourne.sgi.com [134.14.52.130]) by deliverator.sgi.com (980309.SGI.8.8.8-aspam-6.2/980310.SGI-aspam) via SMTP id PAA14923
	for <linux-xfs@oss.sgi.com>; Tue, 20 Mar 2001 15:45:56 -0800 (PST)
	mail_from (nathans@wobbly.melbourne.sgi.com)
Received: from wobbly.melbourne.sgi.com (wobbly.melbourne.sgi.com [134.14.55.135]) by larry.melbourne.sgi.com (950413.SGI.8.6.12/950213.SGI.AUTOCF) via ESMTP id KAA06616; Wed, 21 Mar 2001 10:45:48 +1100
Received: (from nathans@localhost) by wobbly.melbourne.sgi.com (980427.SGI.8.8.8/980728.SGI.AUTOCF) id KAA39452; Wed, 21 Mar 2001 10:45:45 +1100 (EDT)
From: "Nathan Scott" <nathans@wobbly.melbourne.sgi.com>
Message-Id: <10103211045.ZM227205@wobbly.melbourne.sgi.com>
Date: Wed, 21 Mar 2001 10:45:44 -0400
In-Reply-To: "Martin K. Petersen" <mkp@linuxcare.com>
        "Re: Bugreport - not really important" (Mar 20,  8:55am)
References: <Pine.LNX.4.30.0103200223470.2403-100000@dbear.engr.sgi.com> 
	<yq1bsqwr1dh.fsf@jaguar.mkp.net>
X-Mailer: Z-Mail (3.2.3 08feb96 MediaMail)
To: "Martin K. Petersen" <mkp@linuxcare.com>, GCS <gcs@agentsinside.com>
Subject: Re: Bugreport - not really important
Cc: Tom Duffy <tduffy@engr.sgi.com>, <linux-xfs@oss.sgi.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

hi,

On Mar 20,  8:55am, Martin K. Petersen wrote:
> Subject: Re: Bugreport - not really important
> >>>>> "Tom" == Tom Duffy <tduffy@engr.sgi.com> writes:
> 
> Tom> what I am surprised about is that xfs did not replace the magic
> Tom> number that tells the kernel and other tools what type of fs it

as Martin said, there's no one location to poke a new magic
number.  mount(8) has knowledge of known good places to go
and look for the primary superblock & magic numbers therein,
and unless given the -t arg, will basically make an educated
guess at the filesystem type.

mkfs.xfs(8) shares this probing code and will also attempt to
identify an existing filesystem before overwriting it - there
is just no support for reiser in there yet (cos its not in the
mount code - see util-linux-2.11/mount/mount_guess_fstype.c).

[be good if someone would do the reiser side of things for
mount, & we would just pick it up from Andries' mount code]

> Tom> is.  otherwise, the reiser tools do not verify this number (or
> Tom> assume it could be corrupted).
> 
> The problem here is that every fs puts its superblock a different
> place.  XFS has it on block 0, and reiserfs' is 64K from the beginning
> of the device.
> 
> It so happens that in your case XFS doesn't write anything on that
> position.  Consequently, the reiser superblock is intact after
> mkfs.xfs.
> 
> There's really no smart way to avoid this as that would require
> knowledge of all file systems on the planet.  Some may have their sb
> at the middle or the end of the device etc.  And dd'ing zeroes to the
> entire device will take forever.
> 

(nod)

> I guess we could blast - say 70KB - of NULL chars to the start of the
> partition.  As far as I can tell, that should get rid of any fs that
> Linux currently supports (Unless they try and use backup copies in
> that case.  I guess we'll have to experiment a bit).
> 
> Nathan, comments?
> 

yeah, thats probably the simplest solution (hack!)- delicate
like a sledgehammer, but I guess its the only sure-fire way.
be careful about going back from the end, I think there may
be an md superblock back there (?) and that was also how we
got into the need for BLKSETSIZE (we used to blow away last
512bytes in case an old EFS superblock was there).

cheers.

-- 
Nathan

From owner-linux-xfs@oss.sgi.com Tue Mar 20 15:54:47 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2KNslu06127
	for linux-xfs-outgoing; Tue, 20 Mar 2001 15:54:47 -0800
Received: from pneumatic-tube.sgi.com (pneumatic-tube.sgi.com [204.94.214.22])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2KNslM06124
	for <linux-xfs@oss.sgi.com>; Tue, 20 Mar 2001 15:54:47 -0800
Received: from larry.melbourne.sgi.com (larry.melbourne.sgi.com [134.14.52.130]) by pneumatic-tube.sgi.com (980327.SGI.8.8.8-aspam/980310.SGI-aspam) via SMTP id QAA02078
	for <linux-xfs@oss.sgi.com>; Tue, 20 Mar 2001 16:04:44 -0800 (PST)
	mail_from (nathans@wobbly.melbourne.sgi.com)
Received: from wobbly.melbourne.sgi.com (wobbly.melbourne.sgi.com [134.14.55.135]) by larry.melbourne.sgi.com (950413.SGI.8.6.12/950213.SGI.AUTOCF) via ESMTP id KAA06658; Wed, 21 Mar 2001 10:53:25 +1100
Received: (from nathans@localhost) by wobbly.melbourne.sgi.com (980427.SGI.8.8.8/980728.SGI.AUTOCF) id KAA39759; Wed, 21 Mar 2001 10:53:14 +1100 (EDT)
From: "Nathan Scott" <nathans@wobbly.melbourne.sgi.com>
Message-Id: <10103211053.ZM240119@wobbly.melbourne.sgi.com>
Date: Wed, 21 Mar 2001 10:53:13 -0400
In-Reply-To: Russell Cattelan <cattelan@thebarn.com>
        "Re: mounting loop devices  Thanx for the mkinitrd-3.0.5 patch!" (Mar 20,  1:11pm)
References: <Pine.LNX.4.30.0103200109490.2403-200000@dbear.engr.sgi.com> 
	<3AB773E0.E3919CE0@theseus.com> 
	<3AB79D4B.42DCD9A1@thebarn.com>
X-Mailer: Z-Mail (3.2.3 08feb96 MediaMail)
To: Russell Cattelan <cattelan@thebarn.com>, thebs@theseus.com,
   b.j.smith@ieee.org, linux-xfs@oss.sgi.com
Subject: Re: mounting loop devices  Thanx for the mkinitrd-3.0.5 patch!
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

hi, 

On Mar 20,  1:11pm, Russell Cattelan wrote:
> Subject: Re: mounting loop devices  Thanx for the mkinitrd-3.0.5 patch!
> Bryan-TheBS-Smith wrote:
> 
> > Tom Duffy wrote:
> > >                                Name: mkinitrd-3.0.5-xfs.patch
> > >    mkinitrd-3.0.5-xfs.patch    Type: Plain Text (TEXT/PLAIN)
> > >                            Encoding: BASE64
> >
> > Much appreciated dude!
> >
> > -- TheBS
> >
> > P.S.  Where would I normally find these patches?
> 
> Opps sorry it wasn't removed.
> cmd/xfstests/misc/mkinitrd.xfs
> 
> Suppose we could find a better place for this; "xfstests"  is a bit
> misleading.
> 

yeah, sorry about that - it got moved in the Great Packaging
Reorg.  I didn't find a good spot for that, so it just ended
up there with a few other odds and ends.  how about if I do
a cmd/xfsmisc/ with this sort of stuff?  (I'm going to need
it for something else later, and the quot.c code that's there
now has been taken up into the new quota utilities, so that'll
go away too).

cheers.

-- 
Nathan

From owner-linux-xfs@oss.sgi.com Tue Mar 20 16:02:08 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2L028906390
	for linux-xfs-outgoing; Tue, 20 Mar 2001 16:02:08 -0800
Received: from sgi.com (sgi.SGI.COM [192.48.153.1])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2L024M06385
	for <linux-xfs@oss.sgi.com>; Tue, 20 Mar 2001 16:02:04 -0800
Received: from zeus-fddi.americas.sgi.com (128-162-8-103.americas.sgi.com [128.162.8.103]) 
	by sgi.com (980327.SGI.8.8.8-aspam/980304.SGI-aspam:
       SGI does not authorize the use of its proprietary
       systems or networks for unsolicited or bulk email
       from the Internet.) 
	via ESMTP id QAA04818
	for <linux-xfs@oss.sgi.com>; Tue, 20 Mar 2001 16:01:38 -0800 (PST)
	mail_from (lord@sgi.com)
Received: from daisy-e185.americas.sgi.com (daisy-e194.americas.sgi.com [128.162.194.214]) by zeus-fddi.americas.sgi.com (8.9.3/americas-smart-nospam1.1) with ESMTP id RAA1205460; Tue, 20 Mar 2001 17:59:16 -0600 (CST)
Received: from jen.americas.sgi.com (IDENT:root@jen.americas.sgi.com [128.162.184.86]) by daisy-e185.americas.sgi.com (SGI-8.9.3/SGI-server-1.7) with ESMTP id RAA52130; Tue, 20 Mar 2001 17:59:16 -0600 (CST)
Received: from jen.americas.sgi.com by jen.americas.sgi.com (8.11.0/SGI-client-1.7) via ESMTP id f2L00K723322; Tue, 20 Mar 2001 18:00:20 -0600
Message-Id: <200103210000.f2L00K723322@jen.americas.sgi.com>
X-Mailer: exmh version 2.2 06/23/2000 with nmh-1.0.4
To: "Nathan Scott" <nathans@wobbly.melbourne.sgi.com>
cc: Russell Cattelan <cattelan@thebarn.com>, thebs@theseus.com,
   b.j.smith@ieee.org, linux-xfs@oss.sgi.com
Subject: Re: mounting loop devices  Thanx for the mkinitrd-3.0.5 patch! 
In-Reply-To: Message from "Nathan Scott" <nathans@wobbly.melbourne.sgi.com> 
   of "Wed, 21 Mar 2001 10:53:13 -0400." <10103211053.ZM240119@wobbly.melbourne.sgi.com> 
Date: Tue, 20 Mar 2001 18:00:20 -0600
From: Steve Lord <lord@sgi.com>
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk


> 
> yeah, sorry about that - it got moved in the Great Packaging
> Reorg.  I didn't find a good spot for that, so it just ended
> up there with a few other odds and ends.  how about if I do
> a cmd/xfsmisc/ with this sort of stuff?  (I'm going to need
> it for something else later, and the quot.c code that's there
> now has been taken up into the new quota utilities, so that'll
> go away too).

And this comment reminds me - the one thing we don't have in the
tree is an update to the Changes file indicating the lowest revision
of various command packages for use with xfs. This would of course
include the xfs commands and how to get them.

Steve

> 
> cheers.
> 
> -- 
> Nathan



From owner-linux-xfs@oss.sgi.com Tue Mar 20 16:14:30 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2L0EUn06666
	for linux-xfs-outgoing; Tue, 20 Mar 2001 16:14:30 -0800
Received: from sgi.com (sgi.SGI.COM [192.48.153.1])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2L0EUM06662
	for <linux-xfs@oss.sgi.com>; Tue, 20 Mar 2001 16:14:30 -0800
Received: from ledzep.americas.sgi.com (relay.sgi.com [137.38.226.97] (may be forged)) 
	by sgi.com (980327.SGI.8.8.8-aspam/980304.SGI-aspam:
       SGI does not authorize the use of its proprietary
       systems or networks for unsolicited or bulk email
       from the Internet.) 
	via ESMTP id QAA00931
	for <linux-xfs@oss.sgi.com>; Tue, 20 Mar 2001 16:14:29 -0800 (PST)
	mail_from (cattelan@thebarn.com)
Received: from gibble.americas.sgi.com (gibble.americas.sgi.com [128.162.195.80]) by ledzep.americas.sgi.com (SGI-SGI-8.9.3/americas-smart-nospam1.1) with ESMTP id SAA18733; Tue, 20 Mar 2001 18:13:11 -0600 (CST)
Received: from thebarn.com (localhost [127.0.0.1])
	by gibble.americas.sgi.com (8.11.2/8.11.2) with ESMTP id f2L038e32284;
	Tue, 20 Mar 2001 19:03:08 -0500
Message-ID: <3AB7EFBA.D5C7F335@thebarn.com>
Date: Tue, 20 Mar 2001 19:03:07 -0500
From: Russell Cattelan <cattelan@thebarn.com>
X-Mailer: Mozilla 4.76 [en] (X11; U; Linux 2.4.2-XFS i686)
X-Accept-Language: en
MIME-Version: 1.0
To: Nathan Scott <nathans@wobbly.melbourne.sgi.com>
CC: thebs@theseus.com, b.j.smith@ieee.org, linux-xfs@oss.sgi.com
Subject: Re: mounting loop devices  Thanx for the mkinitrd-3.0.5 patch!
References: <Pine.LNX.4.30.0103200109490.2403-200000@dbear.engr.sgi.com> 
		<3AB773E0.E3919CE0@theseus.com> 
		<3AB79D4B.42DCD9A1@thebarn.com> <10103211053.ZM240119@wobbly.melbourne.sgi.com>
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

Nathan Scott wrote:

> hi,
>
> On Mar 20,  1:11pm, Russell Cattelan wrote:
> > Subject: Re: mounting loop devices  Thanx for the mkinitrd-3.0.5 patch!
> > Bryan-TheBS-Smith wrote:
> >
> > > Tom Duffy wrote:
> > > >                                Name: mkinitrd-3.0.5-xfs.patch
> > > >    mkinitrd-3.0.5-xfs.patch    Type: Plain Text (TEXT/PLAIN)
> > > >                            Encoding: BASE64
> > >
> > > Much appreciated dude!
> > >
> > > -- TheBS
> > >
> > > P.S.  Where would I normally find these patches?
> >
> > Opps sorry it wasn't removed.
> > cmd/xfstests/misc/mkinitrd.xfs
> >
> > Suppose we could find a better place for this; "xfstests"  is a bit
> > misleading.
> >
>
> yeah, sorry about that - it got moved in the Great Packaging
> Reorg.  I didn't find a good spot for that, so it just ended
> up there with a few other odds and ends.  how about if I do
> a cmd/xfsmisc/ with this sort of stuff?  (I'm going to need
> it for something else later, and the quot.c code that's there
> now has been taken up into the new quota utilities, so that'll
> go away too).

sounds good.
Tom has an ism for mkinitrd but thinking about it it may be easiest to just
have an
additional copy in our tree.
Especially since the one he has checked in only works on RH systems.



>
>
> cheers.
>
> --
> Nathan

--
Russell Cattelan
--
Digital Elves inc. -- Currently on loan to SGI
Linux XFS core developer.




From owner-linux-xfs@oss.sgi.com Tue Mar 20 16:23:44 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2L0NiN06813
	for linux-xfs-outgoing; Tue, 20 Mar 2001 16:23:44 -0800
Received: from yog-sothoth.sgi.com (eugate.sgi.com [192.48.160.10])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2L0NfM06806
	for <linux-xfs@oss.sgi.com>; Tue, 20 Mar 2001 16:23:41 -0800
Received: from snort.melbourne.sgi.com (snort.melbourne.sgi.com [134.14.55.149]) by yog-sothoth.sgi.com (980305.SGI.8.8.8-aspam-6.2/980304.SGI-aspam-europe) via ESMTP id BAA6060830
	for <linux-xfs@oss.sgi.com>; Wed, 21 Mar 2001 01:23:37 +0100 (CET)
	mail_from (nathans@snort.melbourne.sgi.com)
Received: (from nathans@localhost)
	by snort.melbourne.sgi.com (SGI-8.9.3/8.9.3) id LAA75216
	for linux-xfs@oss.sgi.com; Wed, 21 Mar 2001 11:22:16 +1100 (EST)
Date: Wed, 21 Mar 2001 11:22:16 +1100 (EST)
From: Nathan Scott <nathans@snort.melbourne.sgi.com>
Message-Id: <200103210022.LAA75216@snort.melbourne.sgi.com>
To: linux-xfs@oss.sgi.com
Subject: TAKE - doc
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

Date:  Tue Mar 20 15:07:12 PST 2001
Workarea:  snort.melbourne.sgi.com:/diskb/build4/nathans/linux-xfs

The following file(s) were checked into:
  bonnie.engr.sgi.com:/isms/slinx/2.4.x-xfs


Modid:  2.4.x-xfs:slinx:90459a
cmd/xfsprogs/doc/CHANGES - 1.9
	- add bullet point re spec file changes, make upper/lower case
	  a little more consistent.


Date:  Tue Mar 20 16:21:29 PST 2001
Workarea:  snort.melbourne.sgi.com:/diskb/build4/nathans/linux-xfs

The following file(s) were checked into:
  bonnie.engr.sgi.com:/isms/slinx/2.4.x-xfs


Modid:  2.4.x-xfs:slinx:90467a
linux/Documentation/Changes - 1.33
	- add some verbage alongside e2fsprogs/reiserfsprogs about
	  xfsprogs, known good versions, etc.


From owner-linux-xfs@oss.sgi.com Tue Mar 20 16:27:38 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2L0Rcc06854
	for linux-xfs-outgoing; Tue, 20 Mar 2001 16:27:38 -0800
Received: from tux.mkp.net (tux.mkp.net [130.225.60.11])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2L0RbM06851
	for <linux-xfs@oss.sgi.com>; Tue, 20 Mar 2001 16:27:37 -0800
Received: from tux.mkp.net ([130.225.60.11] helo=jaguar.mkp.net)
	by tux.mkp.net with esmtp (Exim 3.16 #1)
	id 14fWT6-0000Mi-00; Wed, 21 Mar 2001 01:27:17 +0100
Received: (from mkp@localhost)
	by jaguar.mkp.net (8.9.3/8.9.3) id TAA14270;
	Tue, 20 Mar 2001 19:27:08 -0500
X-Authentication-Warning: jaguar.mkp.net: mkp set sender to mkp@mkp.net using -f
To: "Nathan Scott" <nathans@wobbly.melbourne.sgi.com>
Cc: GCS <gcs@agentsinside.com>, Tom Duffy <tduffy@engr.sgi.com>,
   <linux-xfs@oss.sgi.com>
Subject: Re: Bugreport - not really important
References: <Pine.LNX.4.30.0103200223470.2403-100000@dbear.engr.sgi.com>
	<yq1bsqwr1dh.fsf@jaguar.mkp.net>
	<10103211045.ZM227205@wobbly.melbourne.sgi.com>
From: "Martin K. Petersen" <mkp@linuxcare.com>
Organization: Linuxcare, Inc.
Date: 20 Mar 2001 19:27:07 -0500
In-Reply-To: <10103211045.ZM227205@wobbly.melbourne.sgi.com>
Message-ID: <yq1hf0onezo.fsf@jaguar.mkp.net>
Lines: 15
User-Agent: Gnus/5.0808 (Gnus v5.8.8) XEmacs/21.1 (Canyonlands)
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

>>>>> "Nathan" == Nathan Scott <nathans@wobbly.melbourne.sgi.com> writes:

Nathan> yeah, thats probably the simplest solution (hack!)- delicate
Nathan> like a sledgehammer, but I guess its the only sure-fire way.
Nathan> be careful about going back from the end, I think there may be
Nathan> an md superblock back there (?) 

Yes, but that shouldn't be exposed to userspace.  The MD device should
appear to be a few blocks shorter than the physically available space
on the disk(s).

-- 
Martin K. Petersen, Principal Linux Consultant, Linuxcare, Inc.
mkp@linuxcare.com, http://www.linuxcare.com/
SGI XFS for Linux Developer, http://oss.sgi.com/projects/xfs/

From owner-linux-xfs@oss.sgi.com Tue Mar 20 17:04:31 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2L14VP07588
	for linux-xfs-outgoing; Tue, 20 Mar 2001 17:04:31 -0800
Received: from deliverator.sgi.com (deliverator.sgi.com [204.94.214.10])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2L14UM07578;
	Tue, 20 Mar 2001 17:04:30 -0800
Received: from ledzep.americas.sgi.com (ledzep.americas.sgi.com [137.38.226.97]) by deliverator.sgi.com (980309.SGI.8.8.8-aspam-6.2/980310.SGI-aspam) via ESMTP id RAA26179; Tue, 20 Mar 2001 17:03:17 -0800 (PST)
	mail_from (cattelan@gibble.americas.sgi.com)
Received: from gibble.americas.sgi.com (gibble.americas.sgi.com [128.162.195.80]) by ledzep.americas.sgi.com (SGI-SGI-8.9.3/americas-smart-nospam1.1) with ESMTP id TAA17888; Tue, 20 Mar 2001 19:03:13 -0600 (CST)
Received: (from cattelan@localhost)
	by gibble.americas.sgi.com (8.11.2/8.11.2) id f2L0rAD32563;
	Tue, 20 Mar 2001 19:53:10 -0500
From: Russell Cattelan <cattelan@thebarn.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Message-ID: <15031.64373.271981.281791@gibble.americas.sgi.com>
Date: Tue, 20 Mar 2001 18:53:09 -0600 (CST)
To: linux-xfs-announce@oss.sgi.com, linux-xfs@oss.sgi.com
Subject: XFS prerelease 0,10 test1 images.
X-Mailer: VM 6.75 under 21.1 (patch 14) "Cuyahoga Valley" XEmacs Lucid
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

Since I have to be someplace 10 minutes ago this will be short.

The first of the test images for the up coming XFS 0.10 
release are now available on the ftp site.
ftp://linux-xfs.sgi.com/projects/xfs/download/PreRelease-0.10-test1
or the old stand by.
ftp://oss.sgi.com/projects/xfs/download/PreRelease-0.10-test1

The format is basically the same as the 0.9 release, with 
a few exceptions.
First xfs commands have been broken down into 
separate packages and are now distributed as individual rpms
and  individual src tar files.
acl-1.0.1.src.tar.gz
dmapi-0.1.1.src.tar.gz 
xfsprogs-1.1.6.src.tar.gz
attr-1.0.1.src.tar.gz
xfsdump-1.0.2.src.tar.gz


Second the kernel patches have been broken down into
two pieces 
linux-2.4.2-core-xfs-03192001.patch.gz
(core linux files changed to support XFS)
and
linux-2.4.2-xfs-03192001.patch.gz
(all original xfs files)

kdb-v1.8-2.4.2-fixed.patch.gz
loop-xfs-7C.patch.gz
linux-2.4.2-changeloop.patch.gz
are also provided as they were used to 
build the 0.10 rpm's.


In addition to the source portion of the release
the anaconda based installer has been updated to 
the latest RedHat wolverine base.
(note installer now requires Wolverine CD's for install)
(notetoo system "Upgrade" is now available)

Available as an iso image or individual files.

Running XFS systems can simple be upgraded by grabbing
the rpms from the installer and installing them manually.


Another spin is forth coming as a few bugs have
been fixed and need to be included but please
test these images and send feedback.

-- 
Russell Cattelan
--
Digital Elves inc. -- Currently on loan to SGI
Linux XFS core developer.



From owner-linux-xfs@oss.sgi.com Tue Mar 20 18:13:37 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2L2Db408768
	for linux-xfs-outgoing; Tue, 20 Mar 2001 18:13:37 -0800
Received: from pneumatic-tube.sgi.com (pneumatic-tube.sgi.com [204.94.214.22])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2L2DbM08765
	for <linux-xfs@oss.sgi.com>; Tue, 20 Mar 2001 18:13:37 -0800
Received: from snort.melbourne.sgi.com (snort.melbourne.sgi.com [134.14.55.149]) by pneumatic-tube.sgi.com (980327.SGI.8.8.8-aspam/980310.SGI-aspam) via ESMTP id SAA03889
	for <linux-xfs@oss.sgi.com>; Tue, 20 Mar 2001 18:23:34 -0800 (PST)
	mail_from (nathans@snort.melbourne.sgi.com)
Received: (from nathans@localhost)
	by snort.melbourne.sgi.com (SGI-8.9.3/8.9.3) id NAA88197
	for linux-xfs@oss.sgi.com; Wed, 21 Mar 2001 13:12:15 +1100 (EST)
Date: Wed, 21 Mar 2001 13:12:15 +1100 (EST)
From: Nathan Scott <nathans@snort.melbourne.sgi.com>
Message-Id: <200103210212.NAA88197@snort.melbourne.sgi.com>
To: linux-xfs@oss.sgi.com
Subject: TAKE - xfsmisc
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

A more sensible place to put scripts, utilities, patches, or
what-have-you for those things which have no other home.


Date:  Tue Mar 20 17:47:12 PST 2001
Workarea:  snort.melbourne.sgi.com:/diskb/build4/nathans/linux-xfs

The following file(s) were checked into:
  bonnie.engr.sgi.com:/isms/slinx/2.4.x-xfs


Modid:  2.4.x-xfs:slinx:90473a
cmd/xfstests/misc/xfs_stats.pl - 1.2
	- use pod to embed the man page in the script.


Date:  Tue Mar 20 17:52:40 PST 2001
Workarea:  snort.melbourne.sgi.com:/diskb/build4/nathans/linux-xfs

The following file(s) were checked into:
  bonnie.engr.sgi.com:/isms/slinx/2.4.x-xfs


Modid:  2.4.x-xfs:slinx:90474a
cmd/xfstests/common.quota - 1.5
	- use quot from root, misc is going away.

cmd/xfstests/Makefile - 1.3
	- remove misc & man directories, these have found new homes.


Date:  Tue Mar 20 17:56:37 PST 2001
Workarea:  snort.melbourne.sgi.com:/diskb/build4/nathans/linux-xfs

The following file(s) were checked into:
  bonnie.engr.sgi.com:/isms/slinx/2.4.x-xfs


Modid:  2.4.x-xfs:slinx:90476a
cmd/xfstests/misc/Makefile - 1.5
	- no longer needed - nothing here anymore.

cmd/xfstests/man/Makefile - 1.3
cmd/xfstests/man/man1/Makefile - 1.2
	- no longer needed - no man pages here anymore.

cmd/xfstests/man/man1/xfs_stats.1 - 1.2
	- converted into (inline) pod within the xfs_stats.pl script.

cmd/xfstests/misc/quot.c - 1.2
cmd/xfstests/man/man1/quot.1 - 1.2
	- subsumed by the new quota tools.


[moved cmd/xfstests/misc to cmd/xfsmisc here]


Date:  Tue Mar 20 18:09:40 PST 2001
Workarea:  snort.melbourne.sgi.com:/diskb/build4/nathans/linux-xfs

The following file(s) were checked into:
  bonnie.engr.sgi.com:/isms/slinx/2.4.x-xfs


Modid:  2.4.x-xfs:slinx:90479a
cmd/xfsmisc/README - 1.1
	- describes each file in the xfsmisc directory.



From owner-linux-xfs@oss.sgi.com Tue Mar 20 22:03:54 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2L63sB13116
	for linux-xfs-outgoing; Tue, 20 Mar 2001 22:03:54 -0800
Received: from sgi.com (sgi.SGI.COM [192.48.153.1])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2L63eM13106
	for <linux-xfs@oss.sgi.com>; Tue, 20 Mar 2001 22:03:40 -0800
Received: from cthulhu.engr.sgi.com (cthulhu.engr.sgi.com [192.26.80.2]) 
	by sgi.com (980327.SGI.8.8.8-aspam/980304.SGI-aspam:
       SGI does not authorize the use of its proprietary
       systems or networks for unsolicited or bulk email
       from the Internet.) 
	via ESMTP id WAA08985
	for <linux-xfs@oss.sgi.com>; Tue, 20 Mar 2001 22:03:24 -0800 (PST)
	mail_from (tduffy@engr.sgi.com)
Received: from dbear.engr.sgi.com (dbear.engr.sgi.com [163.154.18.85])
	by cthulhu.engr.sgi.com (SGI-8.9.3/8.9.3) with ESMTP id WAA44304;
	Tue, 20 Mar 2001 22:01:24 -0800 (PST)
Date: Tue, 20 Mar 2001 21:57:31 -0800 (PST)
From: Tom Duffy <tduffy@engr.sgi.com>
To: Russell Cattelan <cattelan@thebarn.com>
cc: <thebs@theseus.com>, <b.j.smith@ieee.org>, <linux-xfs@oss.sgi.com>
Subject: Re: mounting loop devices  Thanx for the mkinitrd-3.0.5 patch!
In-Reply-To: <3AB7AAF2.3045DAF0@thebarn.com>
Message-ID: <Pine.LNX.4.30.0103202155261.19348-100000@dbear.engr.sgi.com>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

> Hmm is it RedHat specific? The last time I updated mkinitrd I had revert the
> mktemp portion; as Redhat seems to be the only one that supports mktemp -d
> (aka temp directory)

well, I still have the part of the patch that takes out the mktemp stuff
in there, but I also am shipping mktemp rpms so that it will work on other
distro's.  i was going to eventually revert it back to redhat's way.

> The rpm's I'm building for prerelease 0.10 has XFS build into the kernel
> which
> renders the changes to mkinitrd un necessary unless somebody really wants
> have
> XFS as a module again.

why are you building different rpms than the one's in LBS?

this makes no sense.  we should release xfs with the lbs so we get the
added exposure on LBS as well.

-tduffy


From owner-linux-xfs@oss.sgi.com Tue Mar 20 23:27:27 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2L7RRI16661
	for linux-xfs-outgoing; Tue, 20 Mar 2001 23:27:27 -0800
Received: from lips.borg.umn.edu (lips.borg.umn.edu [160.94.232.50])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2L7RQM16658
	for <linux-xfs@oss.sgi.com>; Tue, 20 Mar 2001 23:27:26 -0800
Received: from thebarn.com (nic-31-c12-219.mn.mediaone.net [24.31.12.219])
	by lips.borg.umn.edu (8.12.0.Beta5/8.12.0.Beta5) with ESMTP id f2L7RIdE012207;
	Wed, 21 Mar 2001 01:27:18 -0600 (CST)
Message-ID: <3AB85758.7FB62EF9@thebarn.com>
Date: Wed, 21 Mar 2001 01:25:12 -0600
From: Russell Cattelan <cattelan@thebarn.com>
X-Mailer: Mozilla 4.75 [en] (X11; U; Linux 2.4.2-XFS i686)
X-Accept-Language: en
MIME-Version: 1.0
To: Tom Duffy <tduffy@engr.sgi.com>
CC: linux-xfs@oss.sgi.com
Subject: Re: mounting loop devices  Thanx for the mkinitrd-3.0.5 patch!
References: <Pine.LNX.4.30.0103202155261.19348-100000@dbear.engr.sgi.com>
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

Tom Duffy wrote:

> > Hmm is it RedHat specific? The last time I updated mkinitrd I had revert the
> > mktemp portion; as Redhat seems to be the only one that supports mktemp -d
> > (aka temp directory)
>
> well, I still have the part of the patch that takes out the mktemp stuff
> in there, but I also am shipping mktemp rpms so that it will work on other
> distro's.  i was going to eventually revert it back to redhat's way.

Hopefully most of the distro's will get  caught up  and provide the and updated
mktemp version, and the issue will go away.

Just checked my Mandrake system and it has a  -d capable  mktemp.

>
> > The rpm's I'm building for prerelease 0.10 has XFS build into the kernel
> > which
> > renders the changes to mkinitrd un necessary unless somebody really wants
> > have
> > XFS as a module again.
>
> why are you building different rpms than the one's in LBS?

The last time we tried a build it failed.
Also given the unclear  situation of what can be released in regards
to xscsi  that will need to be dropped from anything that is put on oss.

>
>
> this makes no sense.  we should release xfs with the lbs so we get the
> added exposure on LBS as well.

Three parts to the release are being done.
0.10 Patches
0.10 based rpm based on 2.4.2 + 0.10 Patches + loopfs fix.
LBS based rpms.
The addition of the LBS rpms is contingent on them building and working,
as of last friday that did not appear to be the case.


I'll fire off a build and  see how things go if things work I'll add the rpm
to the next testX push.


>
> -tduffy


From owner-linux-xfs@oss.sgi.com Wed Mar 21 04:49:14 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2LCnE025119
	for linux-xfs-outgoing; Wed, 21 Mar 2001 04:49:14 -0800
Received: from home.smithconcepts.com (ubr-35.28.151.oviedo.cfl.rr.com [65.35.28.151])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2LCnDM25115;
	Wed, 21 Mar 2001 04:49:13 -0800
Received: from ieee.org (IDENT:bjsmith@bitman.oviedo.smithconcepts.com [172.24.24.192])
	by home.smithconcepts.com (8.9.3/8.9.3) with ESMTP id HAA18216;
	Wed, 21 Mar 2001 07:43:40 -0500
Message-ID: <3AB8A573.9BD47F15@ieee.org>
Date: Wed, 21 Mar 2001 07:58:27 -0500
From: "Bryan J. Smith" <b.j.smith@ieee.org>
Reply-To: b.j.smith@ieee.org, thebs@theseus.com
Organization: SmithConcepts/Personal
X-Mailer: Mozilla 4.76 [en] (X11; U; Linux 2.2.18-0.13 i686)
X-Accept-Language: en
MIME-Version: 1.0
To: Russell Cattelan <cattelan@thebarn.com>
CC: linux-xfs-announce@oss.sgi.com, linux-xfs@oss.sgi.com
Subject: Re: XFS prerelease 0,10 test1 images.
References: <15031.64373.271981.281791@gibble.americas.sgi.com>
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

Russell Cattelan wrote:
> Another spin is forth coming as a few bugs have
> been fixed and need to be included but please
> test these images and send feedback.

Much appreciated and will do!

And the numerous updates are noted.  I scrapped my RPMset from Mar
19th as well.  Either way, I'm going to run with this one for a bit.

-- TheBS

-- 
Bryan "TheBS" Smith         chat:thebs413 @AOL/MSN/Yahoo
Engineer     mailto:b.j.smith@ieee.org,thebs@theseus.com
********************************************************
"Linux will do for applications what the Internet did to
 networks" -- Sam Palmisano, IBM Chief Operating Officer

From owner-linux-xfs@oss.sgi.com Wed Mar 21 15:52:29 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2LNqTl11282
	for linux-xfs-outgoing; Wed, 21 Mar 2001 15:52:29 -0800
Received: from yog-sothoth.sgi.com (eugate.sgi.com [192.48.160.10])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2LNqRM11277
	for <linux-xfs@oss.sgi.com>; Wed, 21 Mar 2001 15:52:27 -0800
Received: from zeus-fddi.americas.sgi.com (zeus-fddi.americas.sgi.com [128.162.8.103]) by yog-sothoth.sgi.com (980305.SGI.8.8.8-aspam-6.2/980304.SGI-aspam-europe) via ESMTP id AAA6145871
	for <linux-xfs@oss.sgi.com>; Thu, 22 Mar 2001 00:52:25 +0100 (CET)
	mail_from (lord@sgi.com)
Received: from daisy-e185.americas.sgi.com (daisy-e194.americas.sgi.com [128.162.194.214]) by zeus-fddi.americas.sgi.com (8.9.3/americas-smart-nospam1.1) with ESMTP id RAA1217592 for <linux-xfs@oss.sgi.com>; Wed, 21 Mar 2001 17:51:09 -0600 (CST)
Received: from jen.americas.sgi.com (IDENT:root@jen.americas.sgi.com [128.162.184.86]) by daisy-e185.americas.sgi.com (SGI-8.9.3/SGI-server-1.7) with ESMTP id RAA16940 for <linux-xfs@oss.sgi.com>; Wed, 21 Mar 2001 17:51:08 -0600 (CST)
From: Steve Lord <lord@sgi.com>
Received: by jen.americas.sgi.com (8.11.0/SGI-client-1.7) id f2LNq2L14588; Wed, 21 Mar 2001 17:52:02 -0600
Message-Id: <200103212352.f2LNq2L14588@jen.americas.sgi.com>
Date: Wed, 21 Mar 2001 17:52:02 -0600
Subject: TAKE - fix compiler warning from pagebuf.c
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

And fix a possible oops in the process - always listen to your
compiler, it usually knows what it is doing.

Date:  Wed Mar 21 15:50:19 PST 2001
Workarea:  jen.americas.sgi.com:/src/lord/xfs-linux.2.4

The following file(s) were checked into:
  bonnie.engr.sgi.com:/isms/slinx/2.4.x-xfs


Modid:  2.4.x-xfs:slinx:90564a
linux/fs/pagebuf/page_buf.c - 1.67
	- A thinko in some code restructuring - this could cause oopses on
	  I/O errors.



From owner-linux-xfs@oss.sgi.com Wed Mar 21 17:33:04 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2M1X4013004
	for linux-xfs-outgoing; Wed, 21 Mar 2001 17:33:04 -0800
Received: from yog-sothoth.sgi.com (eugate.sgi.com [192.48.160.10])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2M1X3M12999
	for <linux-xfs@oss.sgi.com>; Wed, 21 Mar 2001 17:33:03 -0800
Received: from snort.melbourne.sgi.com (snort.melbourne.sgi.com [134.14.55.149]) by yog-sothoth.sgi.com (980305.SGI.8.8.8-aspam-6.2/980304.SGI-aspam-europe) via ESMTP id CAA6122589
	for <linux-xfs@oss.sgi.com>; Thu, 22 Mar 2001 02:32:43 +0100 (CET)
	mail_from (nathans@snort.melbourne.sgi.com)
Received: (from nathans@localhost)
	by snort.melbourne.sgi.com (SGI-8.9.3/8.9.3) id MAA08297
	for linux-xfs@oss.sgi.com; Thu, 22 Mar 2001 12:31:21 +1100 (EST)
Date: Thu, 22 Mar 2001 12:31:21 +1100 (EST)
From: Nathan Scott <nathans@snort.melbourne.sgi.com>
Message-Id: <200103220131.MAA08297@snort.melbourne.sgi.com>
To: linux-xfs@oss.sgi.com
Subject: TAKE - xfsdump/quota
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

Ivan/Tim - I think this is all you'll need from my end
wrt quota support in xfsdump.  Let me know if not.

cheers.


Date:  Wed Mar 21 17:29:53 PST 2001
Workarea:  snort.melbourne.sgi.com:/diskb/build4/nathans/linux-xfs

The following file(s) were checked into:
  bonnie.engr.sgi.com:/isms/slinx/2.4.x-xfs


Modid:  2.4.x-xfs:slinx:90586a
cmd/xfsdump/quota/Makefile - 1.1
	- build and install xfsdq and xfsrq utility programs for xfsdump.

cmd/xfsdump/man/man8/xfsdq.8 - 1.1
	- man page for xfsdq/xfsrq utility programs for xfsdump.

cmd/xfsdump/quota/xfsdq.c - 1.1
	- equivalent to repquota -e under IRIX, needed by xfsdump.

cmd/xfsdump/quota/xfsrq.sh - 1.1
	- equivalent to edquota -i under IRIX, needed by xfsdump.

cmd/xfsdump/Makefile - 1.3
	- add xfsdump quota tools subdir.



From owner-linux-xfs@oss.sgi.com Wed Mar 21 20:32:23 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2M4WNg16720
	for linux-xfs-outgoing; Wed, 21 Mar 2001 20:32:23 -0800
Received: from gwyn.tux.org (ident-user@gwyn.tux.org [207.96.122.8])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2M4WMM16717
	for <linux-xfs@oss.sgi.com>; Wed, 21 Mar 2001 20:32:22 -0800
Received: (from timball@localhost)
	by gwyn.tux.org (8.9.3/8.9.1) id XAA29379
	for linux-xfs@oss.sgi.com; Wed, 21 Mar 2001 23:32:15 -0500
Date: Wed, 21 Mar 2001 23:32:15 -0500
From: Timothy Ball <timball@tux.org>
To: XFS Mailing List <linux-xfs@oss.sgi.com>
Subject: Script to move a partition from one filesystem to xfs
Message-ID: <20010321233215.F4268@gwyn.tux.org>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.3.14i
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

It's not xfs specific but it will do it. Currently it works on my box
fine just fine. The things that are *not* working is editting /etc/fstab
(I have to stare and think about sed *shudder*) and figuring out if you
have enough space to move a partition on it's own. (Currently you have
to stare at df -h and figure it out for yourself.) Also it has no
ability to change /... cause that'd be harder. But I've been thinking
abt doing it by changing /sbin/init and using lilo -R, but this makes a
good public release. I'll add more features if more ppl use it. :)

It's available at http://www.doublewide.net/~timball

If someone could stare at it and be sure I'm not totally insane that'd
be great.

--timball

-- 
	Send mail with subject "send pgp key" for public key.
pub  1024R/CFF85605 1999-06-10 Timothy L. Ball <timball@sheergenius.com>
     Key fingerprint = 8A 8E 64 D6 21 C0 90 29  9F D6 1E DC F8 18 CB CD

From owner-linux-xfs@oss.sgi.com Wed Mar 21 22:22:09 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2M6M9L19388
	for linux-xfs-outgoing; Wed, 21 Mar 2001 22:22:09 -0800
Received: from lupo.thebarn.com (nic-31-c12-219.mn.mediaone.net [24.31.12.219])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2M6M7M19385;
	Wed, 21 Mar 2001 22:22:08 -0800
Received: (from cattelan@localhost)
	by lupo.thebarn.com (8.11.3/8.11.0) id f2M6M2X74447;
	Thu, 22 Mar 2001 00:22:02 -0600 (CST)
	(envelope-from cattelan)
From: Russell Cattelan <cattelan@thebarn.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Message-ID: <15033.39410.958392.383512@lupo.thebarn.com>
Date: Thu, 22 Mar 2001 00:21:38 -0600 (CST)
To: linux-xfs@oss.sgi.com, linux-xfs-announce@oss.sgi.com
Subject: XFS prerelease 0.10 test2 images
X-Mailer: VM 6.72 under 21.1 (patch 14) "Cuyahoga Valley" XEmacs Lucid
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk


Just a quick note:
The rpms have been regenerated to the latest state of
the XFS tree.
This was primarily to pick up the lvm 0.9beta6 stuff,
be a few bugs have also been squashed in the last few days.

Everything but the iso image has been regenerated.
Eric is working on enforcing a package order for the installer
that will reduce the number of times the CD's need to be swapped.

Once that is working a new iso image will be made available.


ftp://linux-xfs/projects/xfs/download/PreRelease-0.10-test2/

-- 
Russell Cattelan
cattelan@thebarn.com

From owner-linux-xfs@oss.sgi.com Thu Mar 22 05:49:18 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2MDnID31442
	for linux-xfs-outgoing; Thu, 22 Mar 2001 05:49:18 -0800
Received: from home.smithconcepts.com (ubr-35.28.151.oviedo.cfl.rr.com [65.35.28.151])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2MDnHM31435;
	Thu, 22 Mar 2001 05:49:17 -0800
Received: from ieee.org (IDENT:bjsmith@bitman.oviedo.smithconcepts.com [172.24.24.192])
	by home.smithconcepts.com (8.9.3/8.9.3) with ESMTP id IAA31367;
	Thu, 22 Mar 2001 08:43:37 -0500
Message-ID: <3ABA0514.80E6BE35@ieee.org>
Date: Thu, 22 Mar 2001 08:58:44 -0500
From: "Bryan J. Smith" <b.j.smith@ieee.org>
Reply-To: b.j.smith@ieee.org, thebs@theseus.com
Organization: SmithConcepts/Personal
X-Mailer: Mozilla 4.76 [en] (X11; U; Linux 2.2.18-0.13 i686)
X-Accept-Language: en
MIME-Version: 1.0
To: Russell Cattelan <cattelan@thebarn.com>
CC: linux-xfs@oss.sgi.com, linux-xfs-announce@oss.sgi.com
Subject: Re: XFS prerelease 0.10 test2 images  Thank you sirs ...
References: <15033.39410.958392.383512@lupo.thebarn.com>
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

Re: XFS prerelease 0.10 test2 images  Thank you sirs ...

Russell Cattelan wrote:
> Just a quick note:
> The rpms have been regenerated to the latest state of
> the XFS tree.
> This was primarily to pick up the lvm 0.9beta6 stuff,
> be a few bugs have also been squashed in the last few days.
> Everything but the iso image has been regenerated.
> Eric is working on enforcing a package order for the installer
> that will reduce the number of times the CD's need to be swapped.
> Once that is working a new iso image will be made available.
> ftp://linux-xfs/projects/xfs/download/PreRelease-0.10-test2/

Gentleman --

Just wanted to thank you for your time and dedication.  This release
for Wolverine will go into immediate production for some test
systems at home (I'm writing a HOWTO for a firewall device based on
2.4) as well as a few test workstations at work (we just had 6
consultants contracts expire, so I've got a nice 6 x 1GHz cluster to
build ;-).

Just let me know if I can be of any [small] assistence in the future
(packaging, documentation, etc...).  I don't have kids and need to
keep my time occupied.  ;-PPP

-- TheBS

P.S.  Okay, I'll come clean.  I've got an "ulterior motive" in
addition.  I need some meeting space for my LUG
(http://www.elug.org), of which, has detracted from my time for the
past few weeks (as I've been more active looking for a place -- as
well as getting our web site/content rebuilt to show).  I was hoping
I could build a little "influence" with some of you guys at SGI, and
maybe get you'all to host a meeting or two.  Again, I'm coming
"clean" (although understand this is NOT the reason why I was ever
interested in helping the XFS project -- which is heavily
work-related being a company that relies on Linux NFS servers). 
Anyone know anyone at the Orlando office in the central Florida
Research Park?

-- 
Bryan "TheBS" Smith         chat:thebs413 @AOL/MSN/Yahoo
Engineer     mailto:b.j.smith@ieee.org,thebs@theseus.com
********************************************************
"Linux will do for applications what the Internet did to
 networks" -- Sam Palmisano, IBM Chief Operating Officer

From owner-linux-xfs@oss.sgi.com Thu Mar 22 06:32:14 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2MEWEh00570
	for linux-xfs-outgoing; Thu, 22 Mar 2001 06:32:14 -0800
Received: from deliverator.sgi.com (deliverator.sgi.com [204.94.214.10])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2MEWDM00566
	for <linux-xfs@oss.sgi.com>; Thu, 22 Mar 2001 06:32:13 -0800
Received: from zeus-fddi.americas.sgi.com (zeus-fddi.americas.sgi.com [128.162.8.103]) by deliverator.sgi.com (980309.SGI.8.8.8-aspam-6.2/980310.SGI-aspam) via ESMTP id GAA00218
	for <linux-xfs@oss.sgi.com>; Thu, 22 Mar 2001 06:30:56 -0800 (PST)
	mail_from (lord@sgi.com)
Received: from daisy-e185.americas.sgi.com (daisy-e194.americas.sgi.com [128.162.194.214]) by zeus-fddi.americas.sgi.com (8.9.3/americas-smart-nospam1.1) with ESMTP id IAA1221926 for <linux-xfs@oss.sgi.com>; Thu, 22 Mar 2001 08:30:43 -0600 (CST)
Received: from jen.americas.sgi.com (IDENT:root@jen.americas.sgi.com [128.162.184.86]) by daisy-e185.americas.sgi.com (SGI-8.9.3/SGI-server-1.7) with ESMTP id IAA01971 for <linux-xfs@oss.sgi.com>; Thu, 22 Mar 2001 08:30:43 -0600 (CST)
From: Steve Lord <lord@sgi.com>
Received: by jen.americas.sgi.com (8.11.0/SGI-client-1.7) id f2MEVMX21621; Thu, 22 Mar 2001 08:31:22 -0600
Message-Id: <200103221431.f2MEVMX21621@jen.americas.sgi.com>
Date: Thu, 22 Mar 2001 08:31:22 -0600
Subject: TAKE - add maintainer info for xfs
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

Date:  Thu Mar 22 06:30:09 PST 2001
Workarea:  jen.americas.sgi.com:/src/lord/xfs-linux.2.4

The following file(s) were checked into:
  bonnie.engr.sgi.com:/isms/slinx/2.4.x-xfs


Modid:  2.4.x-xfs:slinx:90611a
linux/MAINTAINERS - 1.54
	- Add maintainer entry for xfs

linux/Documentation/Changes - 1.34
	- Add location of where to get xfs command source code

linux/Documentation/filesystems/xfs.txt - 1.5
	- Remove kiocluster from listed options



From owner-linux-xfs@oss.sgi.com Thu Mar 22 10:09:14 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2MI9EP05678
	for linux-xfs-outgoing; Thu, 22 Mar 2001 10:09:14 -0800
Received: from gwyn.tux.org (ident-user@gwyn.tux.org [207.96.122.8])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2MI9DM05675
	for <linux-xfs@oss.sgi.com>; Thu, 22 Mar 2001 10:09:14 -0800
Received: (from timball@localhost)
	by gwyn.tux.org (8.9.3/8.9.1) id NAA02873
	for linux-xfs@oss.sgi.com; Thu, 22 Mar 2001 13:09:13 -0500
Date: Thu, 22 Mar 2001 13:09:13 -0500
From: Timothy Ball <timball@tux.org>
To: XFS Mailing List <linux-xfs@oss.sgi.com>
Subject: migrate-parition script moved
Message-ID: <20010322130913.B1831@gwyn.tux.org>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.3.14i
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

So right when I post my script to be downloaded the machine starts
freaking out. I've moved the script to: http://www.tux.org/~timball
The link is at the top. All problems I stated abt the script before
still exist. 

--timball

-- 
	Send mail with subject "send pgp key" for public key.
pub  1024R/CFF85605 1999-06-10 Timothy L. Ball <timball@sheergenius.com>
     Key fingerprint = 8A 8E 64 D6 21 C0 90 29  9F D6 1E DC F8 18 CB CD

From owner-linux-xfs@oss.sgi.com Thu Mar 22 12:43:06 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2MKh6I10159
	for linux-xfs-outgoing; Thu, 22 Mar 2001 12:43:06 -0800
Received: from astro.umn.edu (hal.astro.umn.edu [128.101.221.100])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2MKh5M10154
	for <linux-xfs@oss.sgi.com>; Thu, 22 Mar 2001 12:43:05 -0800
Received: (qmail 3915 invoked by uid 835); 22 Mar 2001 20:42:59 -0000
Received: from localhost (sendmail-bs@127.0.0.1)
  by localhost with SMTP; 22 Mar 2001 20:42:59 -0000
Date: Thu, 22 Mar 2001 14:42:59 -0600 (CST)
From: kelley eicher <carde@astro.umn.edu>
To: linux-xfs@oss.sgi.com
Subject: spurious log header error...
Message-ID: <Pine.LNX.4.21.0103221425120.27241-100000@HAL.astro.umn.edu>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

i've been testing the 0.9 release of xfs for linux and the following error
seems to come up randomly when i run `xfs_logprint` on my xfs disks.
---

Header 0x4 wanted 0xfeedbabe

******************************************
* ERROR: header cycle=4 block=705        *
******************************************
Bad log record header

---

is this a case where the most recent log overlaps the oldest log record as
stated in the `xfs_logprint` manpage? does this command require that the
disk be unmounted before it is run?

-k


 >> kelley j eicher
<<  Sys Admin
 >> Univ. of MN Astronomy Dept.
<<  ph: (612) 626-2067 or (612) 624-3589
 >> fx: (612) 626-2029
<<  office: 385 physics
 >> carde@astro.umn.edu



From owner-linux-xfs@oss.sgi.com Thu Mar 22 13:19:54 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2MLJsO11395
	for linux-xfs-outgoing; Thu, 22 Mar 2001 13:19:54 -0800
Received: from deliverator.sgi.com (deliverator.sgi.com [204.94.214.10])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2MLJrM11392
	for <linux-xfs@oss.sgi.com>; Thu, 22 Mar 2001 13:19:53 -0800
Received: from zeus-fddi.americas.sgi.com (zeus-fddi.americas.sgi.com [128.162.8.103]) by deliverator.sgi.com (980309.SGI.8.8.8-aspam-6.2/980310.SGI-aspam) via ESMTP id NAA07492
	for <linux-xfs@oss.sgi.com>; Thu, 22 Mar 2001 13:18:40 -0800 (PST)
	mail_from (lord@sgi.com)
Received: from daisy-e185.americas.sgi.com (daisy-e194.americas.sgi.com [128.162.194.214]) by zeus-fddi.americas.sgi.com (8.9.3/americas-smart-nospam1.1) with ESMTP id PAA1228141; Thu, 22 Mar 2001 15:18:36 -0600 (CST)
Received: from jen.americas.sgi.com (IDENT:root@jen.americas.sgi.com [128.162.184.86]) by daisy-e185.americas.sgi.com (SGI-8.9.3/SGI-server-1.7) with ESMTP id PAA61352; Thu, 22 Mar 2001 15:18:36 -0600 (CST)
Received: from jen.americas.sgi.com by jen.americas.sgi.com (8.11.0/SGI-client-1.7) via ESMTP id f2MLJBo04239; Thu, 22 Mar 2001 15:19:11 -0600
Message-Id: <200103222119.f2MLJBo04239@jen.americas.sgi.com>
X-Mailer: exmh version 2.2 06/23/2000 with nmh-1.0.4
To: kelley eicher <carde@astro.umn.edu>
cc: linux-xfs@oss.sgi.com
Subject: Re: spurious log header error... 
In-Reply-To: Message from kelley eicher <carde@astro.umn.edu> 
   of "Thu, 22 Mar 2001 14:42:59 CST." <Pine.LNX.4.21.0103221425120.27241-100000@HAL.astro.umn.edu> 
Date: Thu, 22 Mar 2001 15:19:10 -0600
From: Steve Lord <lord@sgi.com>
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

> i've been testing the 0.9 release of xfs for linux and the following error
> seems to come up randomly when i run `xfs_logprint` on my xfs disks.
> ---
> 
> Header 0x4 wanted 0xfeedbabe
> 
> ******************************************
> * ERROR: header cycle=4 block=705        *
> ******************************************
> Bad log record header
> 
> ---
> 
> is this a case where the most recent log overlaps the oldest log record as
> stated in the `xfs_logprint` manpage? does this command require that the
> disk be unmounted before it is run?
> 
> -k
> 


In general xfs_logprint on a live system is going to be iffy, it is looking
at a moving target. Also there are various modes of operation where logprint
will try to interpret all the data in the log even when it is not all valid.
If you build a filesystem and do not completely fill the log for example
you will get some bad headers.

My other comment is that the 0.9 release is a little old now, there is now
a 0.10 set of images out there. The quickest way to upgrade should be to
go to this directory:

ftp://oss.sgi.com/projects/xfs/download/PreRelease-0.10-test2/Wolverine-SGI-XFS-0.10/RedHat/RPMS/

download the kernel rpm and command rpms you need and do an rpm -U

Note that xfscmds got broken into several things

xfsprogs is the main package, the rest you do not need for basic stuff.

Steve




From owner-linux-xfs@oss.sgi.com Thu Mar 22 16:43:00 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2N0h0i15604
	for linux-xfs-outgoing; Thu, 22 Mar 2001 16:43:00 -0800
Received: from mail11.jump.net (mail11.jump.net [206.196.91.11])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2N0gwM15593;
	Thu, 22 Mar 2001 16:42:58 -0800
Received: from sgi.com (aus-dsl-dhcp1-26.customer.jump.net [63.163.168.26])
	by mail11.jump.net (8.10.2/) with ESMTP id f2N0gtG27064;
	Thu, 22 Mar 2001 18:42:55 -0600 (CST)
Message-ID: <3ABA9C42.407227B6@sgi.com>
Date: Thu, 22 Mar 2001 18:43:46 -0600
From: Eric Sandeen <sandeen@sgi.com>
X-Mailer: Mozilla 4.76 [en] (X11; U; Linux 2.4.2-XFS i686)
X-Accept-Language: en
MIME-Version: 1.0
To: Russell Cattelan <cattelan@thebarn.com>
CC: linux-xfs@oss.sgi.com, linux-xfs-announce@oss.sgi.com
Subject: Re: XFS prerelease 0.10 test2 images
References: <15033.39410.958392.383512@lupo.thebarn.com>
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

Russell Cattelan wrote:
 
> Everything but the iso image has been regenerated.
> Eric is working on enforcing a package order for the installer
> that will reduce the number of times the CD's need to be swapped.

I have regenerated the ISO & uploaded it to:

ftp://linux-xfs.sgi.com/projects/xfs/download/PreRelease-0.10-test2/iso/

and

ftp://oss.sgi.com/projects/xfs/download/PreRelease-0.10-test2/iso/

The install should now have fewer CD swaps:

boot sgi disc
swap to rh1 disc
swap to sgi disc
swap to rh1 disc
swap to sgi disc | <= Only if you selected
swap to rh1 disc | <= "everything"
swap to rh2 disc

-Eric

From owner-linux-xfs@oss.sgi.com Thu Mar 22 18:29:53 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2N2Tr318842
	for linux-xfs-outgoing; Thu, 22 Mar 2001 18:29:53 -0800
Received: from mail5.mia.bellsouth.net (mail5.mia.bellsouth.net [205.152.144.17])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2N2TqM18837
	for <linux-xfs@oss.sgi.com>; Thu, 22 Mar 2001 18:29:53 -0800
Received: from cedar.aspen.com (adsl-61-5-192.mia.bellsouth.net [208.61.5.192])
	by mail5.mia.bellsouth.net (3.3.5alt/0.75.2) with SMTP id VAA13917
	for <linux-xfs@oss.sgi.com>; Thu, 22 Mar 2001 21:35:37 -0500 (EST)
From: Juan Casero <casero@bellsouth.net>
To: linux-xfs@oss.sgi.com
Subject: devFS patch for NVidia drivers
Date: Thu, 22 Mar 2001 21:24:46 -0500
X-Mailer: KMail [version 1.1.99]
Content-Type: text/plain;
  charset="iso-8859-1"
MIME-Version: 1.0
Message-Id: <01032221244600.00870@cedar.aspen.com>
Content-Transfer-Encoding: 8bit
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

Hi Folks -

Anybody have the patch for the nvidia drivers to make it devFS aware so I can
use it with XFS and Linux?  I tried a search of the web but alot of the links that 
came up are no longer available.  It's almost as if NVidia doesn't want us to have
the driver just yet.

Thanks...

-- 
Juan Casero
email:  casero@bellsouth.net
   __   _
  / /  (_)__  __ ____  __
 / /__/ / _ \/ // /\ \/ /  . . .  t h e   c h o i c e   o f   a
/____/_/_//_/\_,_/ /_/\_\              G N U   g e n e r a t i o n . . .

From owner-linux-xfs@oss.sgi.com Thu Mar 22 18:40:18 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2N2eIu19086
	for linux-xfs-outgoing; Thu, 22 Mar 2001 18:40:18 -0800
Received: from sgi.com (sgi.SGI.COM [192.48.153.1])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2N2eDM19082
	for <linux-xfs@oss.sgi.com>; Thu, 22 Mar 2001 18:40:13 -0800
Received: from cthulhu.engr.sgi.com (cthulhu.engr.sgi.com [192.26.80.2]) 
	by sgi.com (980327.SGI.8.8.8-aspam/980304.SGI-aspam:
       SGI does not authorize the use of its proprietary
       systems or networks for unsolicited or bulk email
       from the Internet.) 
	via ESMTP id SAA06846
	for <linux-xfs@oss.sgi.com>; Thu, 22 Mar 2001 18:39:59 -0800 (PST)
	mail_from (tduffy@engr.sgi.com)
Received: from dbear.engr.sgi.com (dbear.engr.sgi.com [163.154.18.85])
	by cthulhu.engr.sgi.com (SGI-8.9.3/8.9.3) with ESMTP id SAA38081;
	Thu, 22 Mar 2001 18:38:34 -0800 (PST)
Date: Thu, 22 Mar 2001 18:34:43 -0800 (PST)
From: Tom Duffy <tduffy@engr.sgi.com>
To: Juan Casero <casero@bellsouth.net>
cc: <linux-xfs@oss.sgi.com>
Subject: Re: devFS patch for NVidia drivers
In-Reply-To: <01032221244600.00870@cedar.aspen.com>
Message-ID: <Pine.LNX.4.30.0103221830520.19348-100000@dbear.engr.sgi.com>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

first off, I would check irc.openprojects.org #nvidia for the patch.

secondly, you can always add the entries in /dev/ at bootup with the
following lines added in your /etc/rc.local

        if [ ! -c /dev/nvidiactl ]; then
            nvmajor=195
            for nvminor in 0 1 2 3; do
                mknod -m 0666 /dev/nvidia$nvminor c $nvmajor $nvminor
            done
            mknod -m 0666 /dev/nvidiactl c $nvmajor 255
        fi

-tduffy

On Thu, 22 Mar 2001, Juan Casero wrote:

> Hi Folks -
>
> Anybody have the patch for the nvidia drivers to make it devFS aware so I can
> use it with XFS and Linux?  I tried a search of the web but alot of the links that
> came up are no longer available.  It's almost as if NVidia doesn't want us to have
> the driver just yet.
>
> Thanks...
>
>


From owner-linux-xfs@oss.sgi.com Thu Mar 22 23:04:00 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2N740c23407
	for linux-xfs-outgoing; Thu, 22 Mar 2001 23:04:00 -0800
Received: from mail.get2chip.com ([64.169.83.2])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2N73xM23396;
	Thu, 22 Mar 2001 23:03:59 -0800
Message-ID: <3ABAF48A.3020908@get2chip.com>
Date: Thu, 22 Mar 2001 23:00:26 -0800
From: Chris Croswhite <ccroswhite@get2chip.com>
MIME-Version: 1.0
To: Eric Sandeen <sandeen@sgi.com>
CC: Russell Cattelan <cattelan@thebarn.com>, linux-xfs@oss.sgi.com,
   linux-xfs-announce@oss.sgi.com
Subject: Re: XFS prerelease 0.10 test2 images
References: <15033.39410.958392.383512@lupo.thebarn.com> <3ABA9C42.407227B6@sgi.com>
Content-Type: text/plain; charset=us-ascii; format=flowed
Content-Transfer-Encoding: 7bit
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk


I have tried the iso image.  However, everytime time I place in disc one 
of RH7.0, the installer indicates that the wrong disc is installed.  I 
have inserted disc 2 of RH7.0 also to no avail.  Am I missing/using 
something here?

TIA,
Chris

Eric Sandeen wrote:

> Russell Cattelan wrote:
>  
> 
>> Everything but the iso image has been regenerated.
>> Eric is working on enforcing a package order for the installer
>> that will reduce the number of times the CD's need to be swapped.
> 
> 
> I have regenerated the ISO & uploaded it to:
> 
> ftp://linux-xfs.sgi.com/projects/xfs/download/PreRelease-0.10-test2/iso/
> 
> and
> 
> ftp://oss.sgi.com/projects/xfs/download/PreRelease-0.10-test2/iso/
> 
> The install should now have fewer CD swaps:
> 
> boot sgi disc
> swap to rh1 disc
> swap to sgi disc
> swap to rh1 disc
> swap to sgi disc | <= Only if you selected
> swap to rh1 disc | <= "everything"
> swap to rh2 disc
> 
> -Eric
> 
> 


From owner-linux-xfs@oss.sgi.com Thu Mar 22 23:35:12 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2N7ZCi24034
	for linux-xfs-outgoing; Thu, 22 Mar 2001 23:35:12 -0800
Received: from home.smithconcepts.com (ubr-35.28.151.oviedo.cfl.rr.com [65.35.28.151])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2N7ZAM24025;
	Thu, 22 Mar 2001 23:35:11 -0800
Received: from ieee.org (IDENT:bjsmith@bitman.oviedo.smithconcepts.com [172.24.24.192])
	by home.smithconcepts.com (8.9.3/8.9.3) with ESMTP id CAA09149;
	Fri, 23 Mar 2001 02:28:58 -0500
Message-ID: <3ABAFEC7.37CC5D3B@ieee.org>
Date: Fri, 23 Mar 2001 02:44:07 -0500
From: "Bryan J. Smith" <b.j.smith@ieee.org>
Reply-To: b.j.smith@ieee.org, thebs@theseus.com
Organization: SmithConcepts/Personal
X-Mailer: Mozilla 4.76 [en] (X11; U; Linux 2.2.18-0.13 i686)
X-Accept-Language: en
MIME-Version: 1.0
To: Chris Croswhite <ccroswhite@get2chip.com>
CC: Eric Sandeen <sandeen@sgi.com>, Russell Cattelan <cattelan@thebarn.com>,
   linux-xfs@oss.sgi.com, linux-xfs-announce@oss.sgi.com
Subject: Re: XFS prerelease 0.10 test2 images  It's for Wolverine (RH 7.0.91 
 aka RH71 beta 2)
References: <15033.39410.958392.383512@lupo.thebarn.com> <3ABA9C42.407227B6@sgi.com> <3ABAF48A.3020908@get2chip.com>
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

Re: XFS prerelease 0.10 test2 images  It's for Wolverine (7.0.91
aka 7.1 beta 2)

Chris Croswhite wrote:
> I have tried the iso image.  However, everytime time I place
> in disc one of RH7.0, the installer indicates that the wrong disc
> is installed.  I have inserted disc 2 of RH7.0 also to no avail.
> Am I missing/using something here?

>From my understanding, the PR 0.10 releases are for RedHat
"Wolverine" aka 7.0.91 aka 7.1 beta 2.

-- TheBS

P.S.  7.0.90 aka 7.1 beta [1] was "Fisher"

-- 
Bryan "TheBS" Smith         chat:thebs413 @AOL/MSN/Yahoo
Engineer     mailto:b.j.smith@ieee.org,thebs@theseus.com
********************************************************
"Linux will do for applications what the Internet did to
 networks" -- Sam Palmisano, IBM Chief Operating Officer

From owner-linux-xfs@oss.sgi.com Fri Mar 23 04:02:33 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2NC2Xl29984
	for linux-xfs-outgoing; Fri, 23 Mar 2001 04:02:33 -0800
Received: from haddock.cd.chalmers.se (haddock.cd.chalmers.se [129.16.79.2])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2NC2VM29979
	for <linux-xfs@oss.sgi.com>; Fri, 23 Mar 2001 04:02:31 -0800
Received: from haddock.cd.chalmers.se (localhost [127.0.0.1])
	by haddock.cd.chalmers.se (8.9.3+Sun/8.8.7) with ESMTP id NAA28761
	for <linux-xfs@oss.sgi.com>; Fri, 23 Mar 2001 13:02:20 +0100 (MET)
Message-Id: <200103231202.NAA28761@haddock.cd.chalmers.se>
X-Mailer: exmh version 2.1.1 10/15/1999
X-Face: ^"+mumOlNwo;yI>`\39\txuVze?eiR9uqpo2*mE!9MWXgXI0v(3ArwymNWe'.q:eLl!=guD
 x{jGFEWN6,#HoN%2qRW;7.CL]9%Ap,067"u1%!NUqS.MhV'+,6$Fj-;W2Z}Y,JUZ'L+f<?~E{ID*=T
 V\NxAOu>)|B@3k3n;gLl*#i[(J-os#fNnDJ8m["|JWNwpORh|_.MGkR#|a~QS!"4hEQ{O{[Ii14{xD
 PU/:5wuv7m1=TK=.>G8wdfpY~]{H(Qa\1`%|Hz:!)c3f9UOW|WgE"4d\E7?oDu9.
To: linux-xfs@oss.sgi.com
Subject: xfs on sparc64
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Date: Fri, 23 Mar 2001 13:02:20 +0100
From: Anders Hammarquist <iko@cd.chalmers.se>
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

Hi all!

I my search for a journaling filesystem that actually works on something
other than i386, I've landed on xfs. Initial tests look promising (it
reads, writes, and recovers from a "crash"). I had to patch it slightly
though (below) since sparc64 can't use headers from outside the kernel
tree.

--->8---
Index: linux/include/linux/grio.h
===================================================================
RCS file: /cvs/linux-2.4-xfs/linux/include/linux/grio.h,v
retrieving revision 1.4
diff -u -r1.4 grio.h
--- linux/include/linux/grio.h	2000/12/03 23:44:02	1.4
+++ linux/include/linux/grio.h	2001/03/23 11:57:12
@@ -38,7 +38,7 @@
 #ifndef _LINUX_GRIO_H_
 #define _LINUX_GRIO_H_
 
-#include <limits.h>
+#include <linux/limits.h>
 
 /* ioctl interface to GRIO on /dev/grio (10,160) */
 
--->8---

/Anders

-- 
 -- Of course I'm crazy, but that doesn't mean I'm wrong.
Anders Hammarquist                                  | iko@cd.chalmers.se
Physics student, Chalmers University of Technology, | Hem: +46 31 88 48 50
G|teborg, Sweden.           RADIO: SM6XMM and N2JGL | Mob: +46 707 27 86 87



From owner-linux-xfs@oss.sgi.com Fri Mar 23 06:32:53 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2NEWr400998
	for linux-xfs-outgoing; Fri, 23 Mar 2001 06:32:53 -0800
Received: from mail11.jump.net (mail11.jump.net [206.196.91.11])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2NEWqM00995
	for <linux-xfs@oss.sgi.com>; Fri, 23 Mar 2001 06:32:53 -0800
Received: from sgi.com (aus-dsl-dhcp1-26.customer.jump.net [63.163.168.26])
	by mail11.jump.net (8.10.2/) with ESMTP id f2NEWnG08887;
	Fri, 23 Mar 2001 08:32:49 -0600 (CST)
Message-ID: <3ABB5EC9.9C8128A@sgi.com>
Date: Fri, 23 Mar 2001 08:33:45 -0600
From: Eric Sandeen <sandeen@sgi.com>
X-Mailer: Mozilla 4.76 [en] (X11; U; Linux 2.4.2-XFS i686)
X-Accept-Language: en
MIME-Version: 1.0
To: Chris Croswhite <ccroswhite@get2chip.com>
CC: Russell Cattelan <cattelan@thebarn.com>, linux-xfs@oss.sgi.com
Subject: Re: XFS prerelease 0.10 test2 images
References: <15033.39410.958392.383512@lupo.thebarn.com> <3ABA9C42.407227B6@sgi.com> <3ABAF48A.3020908@get2chip.com>
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

Did you burn your Disc 1 and Disc 2 from the _Wolverine_ beta isos?

They should both have a file like .disc1-i386 or .disc2-i386 in the top
level of the CD, and that file should contain the text:

982439362.228105

That's how Red Hat recognises that you have the correct disc in the
drive.

-Eric

From owner-linux-xfs@oss.sgi.com Fri Mar 23 09:17:23 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2NHHN404915
	for linux-xfs-outgoing; Fri, 23 Mar 2001 09:17:23 -0800
Received: from roujin.gargoylecc.com ([65.100.85.34])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2NHHMM04912
	for <linux-xfs@oss.sgi.com>; Fri, 23 Mar 2001 09:17:22 -0800
Received: from lists.gargoylecc.com (IDENT:ringram@gargoyle.gargoylecc.com [65.100.85.35])
	by roujin.gargoylecc.com (8.9.3/8.9.3) with ESMTP id BAA24595
	for <linux-xfs@oss.sgi.com>; Sat, 24 Mar 2001 01:30:47 -0700
Date: Fri, 23 Mar 2001 10:17:59 -0700 (MST)
From: Russel Ingram <ringram@gargoylecc.com>
X-Sender:  <ringram@localhost.localdomain>
To: <linux-xfs@oss.sgi.com>
Subject: Re: XFS prerelease 0.10 test2 images
In-Reply-To: <3ABA9C42.407227B6@sgi.com>
Message-ID: <Pine.LNX.4.30.0103231015180.9167-100000@localhost.localdomain>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

Has anyone set up an ftp server to do ftp/http/nfs installs from yet.
Russell maybe?

Russ

-- 
---------------------------------------------------------------
"Bill Gates and Microsoft have ruined the computer industry for
 a long time to come by creating a class of ignorant and lazy
 computer users."  --Russel Ingram

"Mommy ... can I go out and ... KILL TONIGHT!?"
                   --Glen  Danzig, The Misfits
---------------------------------------------------------------
Russ Ingram
Gargoyle Computer Consulting
(307)742-1361
www.gargoylecc.com


From owner-linux-xfs@oss.sgi.com Fri Mar 23 09:22:15 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2NHMF305051
	for linux-xfs-outgoing; Fri, 23 Mar 2001 09:22:15 -0800
Received: from kamov.deltanet.ro (kamov.deltanet.ro [193.226.175.59])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2NHMCM05048
	for <linux-xfs@oss.sgi.com>; Fri, 23 Mar 2001 09:22:13 -0800
Received: from shiva.ppetru.net (home.ppetru.net [193.230.129.57])
	by kamov.deltanet.ro (Postfix) with ESMTP id 177D6F7D7
	for <linux-xfs@oss.sgi.com>; Fri, 23 Mar 2001 19:22:06 +0200 (EET)
Received: by shiva.ppetru.net (Postfix, from userid 1000)
	id 91E7C40B1E; Fri, 23 Mar 2001 19:21:28 +0200 (EET)
Date: Fri, 23 Mar 2001 19:21:28 +0200
To: linux-xfs@oss.sgi.com
Subject: Any sparc64 issues?
Message-ID: <20010323192128.A2028@ppetru.net>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.3.15i
From: ppetru@ppetru.net (Petru Paler)
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

Hi,

Are there any sparc64 users of xfs? Does it have any showstopper
bugs?

--
Petru Paler, mailto:ppetru@ppetru.net
http://www.ppetru.net - ICQ: 41817235

From owner-linux-xfs@oss.sgi.com Fri Mar 23 09:25:15 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2NHPFW05171
	for linux-xfs-outgoing; Fri, 23 Mar 2001 09:25:15 -0800
Received: from mail15.jump.net (mail15.jump.net [206.196.91.15])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2NHPFM05168
	for <linux-xfs@oss.sgi.com>; Fri, 23 Mar 2001 09:25:15 -0800
Received: from sgi.com (aus-dsl-dhcp1-26.customer.jump.net [63.163.168.26])
	by mail15.jump.net (8.10.2/) with ESMTP id f2NHOn810309;
	Fri, 23 Mar 2001 11:24:49 -0600 (CST)
Message-ID: <3ABB871B.308107B@sgi.com>
Date: Fri, 23 Mar 2001 11:25:47 -0600
From: Eric Sandeen <sandeen@sgi.com>
X-Mailer: Mozilla 4.76 [en] (X11; U; Linux 2.4.2-XFS i686)
X-Accept-Language: en
MIME-Version: 1.0
To: Russel Ingram <ringram@gargoylecc.com>
CC: linux-xfs@oss.sgi.com
Subject: Re: XFS prerelease 0.10 test2 images
References: <Pine.LNX.4.30.0103231015180.9167-100000@localhost.localdomain>
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

Russel Ingram wrote:
> 
> Has anyone set up an ftp server to do ftp/http/nfs installs from yet.
> Russell maybe?

Russell has the older PreRelease 0.9 / Red Hat 7.0 install on
ftp.thebarn.com/SGI, but it doesn't look like he's updated it to
PR0.10/RH7.1-beta yet.

-Eric

From owner-linux-xfs@oss.sgi.com Fri Mar 23 10:27:45 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2NIRjH06470
	for linux-xfs-outgoing; Fri, 23 Mar 2001 10:27:45 -0800
Received: from tux.mkp.net (tux.mkp.net [130.225.60.11])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2NIRiM06467
	for <linux-xfs@oss.sgi.com>; Fri, 23 Mar 2001 10:27:44 -0800
Received: from tux.mkp.net ([130.225.60.11] helo=jaguar.mkp.net)
	by tux.mkp.net with esmtp (Exim 3.16 #1)
	id 14gWHe-0002od-00; Fri, 23 Mar 2001 19:27:35 +0100
Received: (from mkp@localhost)
	by jaguar.mkp.net (8.9.3/8.9.3) id NAA15589;
	Fri, 23 Mar 2001 13:27:08 -0500
X-Authentication-Warning: jaguar.mkp.net: mkp set sender to mkp@mkp.net using -f
To: ppetru@ppetru.net (Petru Paler)
Cc: linux-xfs@oss.sgi.com
Subject: Re: Any sparc64 issues?
References: <20010323192128.A2028@ppetru.net>
From: "Martin K. Petersen" <mkp@linuxcare.com>
Organization: Linuxcare, Inc.
Date: 23 Mar 2001 13:27:08 -0500
In-Reply-To: <20010323192128.A2028@ppetru.net>
Message-ID: <yq11yronxxf.fsf@jaguar.mkp.net>
Lines: 15
User-Agent: Gnus/5.0808 (Gnus v5.8.8) XEmacs/21.1 (Canyonlands)
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

>>>>> "Petru" == Petru Paler <ppetru@ppetru.net> writes:

Petru> Are there any sparc64 users of xfs? Does it have any
Petru> showstopper bugs?

I haven't upgraded my Sun at home recently, but it worked last time I
checked.

Only thing to worry about is that you need to create your XFS
filesystems with a blocksize of 8K instead of the 4K used on ia32.

-- 
Martin K. Petersen, Principal Linux Consultant, Linuxcare, Inc.
mkp@linuxcare.com, http://www.linuxcare.com/
SGI XFS for Linux Developer, http://oss.sgi.com/projects/xfs/

From owner-linux-xfs@oss.sgi.com Fri Mar 23 10:36:32 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2NIaWR06611
	for linux-xfs-outgoing; Fri, 23 Mar 2001 10:36:32 -0800
Received: from thor.theseus.com (south.orl-pub.theseus.com [12.108.42.66])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2NIaVM06607
	for <linux-xfs@oss.sgi.com>; Fri, 23 Mar 2001 10:36:31 -0800
Received: from theseus.com (IDENT:thebs@fugitive.theseus.com [192.168.0.242])
	by thor.theseus.com (8.9.3/8.9.3) with ESMTP id NAA05003;
	Fri, 23 Mar 2001 13:41:02 -0500
Message-ID: <3ABB97B9.1EC0E52C@theseus.com>
Date: Fri, 23 Mar 2001 13:36:41 -0500
From: Bryan-TheBS-Smith <thebs@theseus.com>
Reply-To: thebs@theseus.com, b.j.smith@ieee.org
Organization: (Personal)
X-Mailer: Mozilla 4.76 [en] (X11; U; Linux 2.2.17-0.17-FUGITIVE i686)
X-Accept-Language: en
MIME-Version: 1.0
To: Russel Ingram <ringram@gargoylecc.com>
CC: linux-xfs@oss.sgi.com
Subject: Re: XFS prerelease 0.10 test2 images
References: <Pine.LNX.4.30.0103231015180.9167-100000@localhost.localdomain>
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

Russel Ingram wrote:
> Has anyone set up an ftp server to do ftp/http/nfs installs
> from yet.  Russell maybe?

???
I just did a couple at home on my private network.  I didn't know
SGI was offering this capability over the Internet?

BTW, it worked great over NFS.

-- TheBS

--
Bryan "TheBS" Smith          chat:thebs413 @AOL/MSN/Yahoo
Engineer      mailto:b.j.smith@ieee.org,thebs@theseus.com
*********************************************************
"Never apply a Star Trek solution to a Babylon 5 problem"
                                    -- Nicholas C. Weaver

From owner-linux-xfs@oss.sgi.com Fri Mar 23 10:37:28 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2NIbSS06644
	for linux-xfs-outgoing; Fri, 23 Mar 2001 10:37:28 -0800
Received: from mail.get2chip.com ([64.169.83.2])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2NIbRM06637;
	Fri, 23 Mar 2001 10:37:27 -0800
Message-ID: <3ABB96E7.9030803@get2chip.com>
Date: Fri, 23 Mar 2001 10:33:11 -0800
From: Chris Croswhite <ccroswhite@get2chip.com>
MIME-Version: 1.0
To: b.j.smith@ieee.org, thebs@theseus.com
CC: Eric Sandeen <sandeen@sgi.com>, Russell Cattelan <cattelan@thebarn.com>,
   linux-xfs@oss.sgi.com, linux-xfs-announce@oss.sgi.com
Subject: Re: XFS prerelease 0.10 test2 images  It's for Wolverine (RH 7.0.91 
	 aka RH71 beta 2)
References: <15033.39410.958392.383512@lupo.thebarn.com> <3ABA9C42.407227B6@sgi.com> <3ABAF48A.3020908@get2chip.com> <3ABAFEC7.37CC5D3B@ieee.org>
Content-Type: text/plain; charset=us-ascii; format=flowed
Content-Transfer-Encoding: 7bit
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

This would be my problem.  Could someone point me in the direction of 
the Wolverine images?

Thanks,
Chris

Bryan J. Smith wrote:

> Re: XFS prerelease 0.10 test2 images  It's for Wolverine (7.0.91
> aka 7.1 beta 2)
> 
> Chris Croswhite wrote:
> 
>> I have tried the iso image.  However, everytime time I place
>> in disc one of RH7.0, the installer indicates that the wrong disc
>> is installed.  I have inserted disc 2 of RH7.0 also to no avail.
>> Am I missing/using something here?
> 
> 
> >From my understanding, the PR 0.10 releases are for RedHat
> "Wolverine" aka 7.0.91 aka 7.1 beta 2.
> 
> -- TheBS
> 
> P.S.  7.0.90 aka 7.1 beta [1] was "Fisher"
> 


From owner-linux-xfs@oss.sgi.com Fri Mar 23 10:42:28 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2NIgSv06749
	for linux-xfs-outgoing; Fri, 23 Mar 2001 10:42:28 -0800
Received: from thor.theseus.com (south.orl-pub.theseus.com [12.108.42.66])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2NIgRM06746
	for <linux-xfs@oss.sgi.com>; Fri, 23 Mar 2001 10:42:27 -0800
Received: from theseus.com (IDENT:thebs@fugitive.theseus.com [192.168.0.242])
	by thor.theseus.com (8.9.3/8.9.3) with ESMTP id NAA05173;
	Fri, 23 Mar 2001 13:47:05 -0500
Message-ID: <3ABB9924.EFBB59CA@theseus.com>
Date: Fri, 23 Mar 2001 13:42:44 -0500
From: Bryan-TheBS-Smith <thebs@theseus.com>
Reply-To: thebs@theseus.com, b.j.smith@ieee.org
Organization: (Personal)
X-Mailer: Mozilla 4.76 [en] (X11; U; Linux 2.2.17-0.17-FUGITIVE i686)
X-Accept-Language: en
MIME-Version: 1.0
To: Chris Croswhite <ccroswhite@get2chip.com>
CC: Eric Sandeen <sandeen@sgi.com>, Russell Cattelan <cattelan@thebarn.com>,
   linux-xfs@oss.sgi.com
Subject: Re: XFS prerelease 0.10 test2 images  It's for Wolverine (RH 7.0.91 
 aka RH71 beta 2)
References: <15033.39410.958392.383512@lupo.thebarn.com> <3ABA9C42.407227B6@sgi.com> <3ABAF48A.3020908@get2chip.com> <3ABAFEC7.37CC5D3B@ieee.org> <3ABB96E7.9030803@get2chip.com>
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

Chris Croswhite wrote:
> This would be my problem.  Could someone point me in the
> direction of the Wolverine images?

Pick your favorite mirror:
   http://www.redhat.com/download/mirror.html

And it will be under the ./beta/wolverine directory if the site
mirrors those directories as well.  ./beta will be at the same level
as the release directories (e.g., ./redhat-6.2, ./redhat-7.0,
etc...) -- which should be right under the root of the redhat
mirror.

-- TheBS

--
Bryan "TheBS" Smith          chat:thebs413 @AOL/MSN/Yahoo
Engineer      mailto:b.j.smith@ieee.org,thebs@theseus.com
*********************************************************
"Never apply a Star Trek solution to a Babylon 5 problem"
                                    -- Nicholas C. Weaver

From owner-linux-xfs@oss.sgi.com Fri Mar 23 10:51:57 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2NIpvi06912
	for linux-xfs-outgoing; Fri, 23 Mar 2001 10:51:57 -0800
Received: from mail.get2chip.com ([64.169.83.2])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2NIptM06903;
	Fri, 23 Mar 2001 10:51:55 -0800
Message-ID: <3ABB9A49.2080606@get2chip.com>
Date: Fri, 23 Mar 2001 10:47:37 -0800
From: Chris Croswhite <ccroswhite@get2chip.com>
MIME-Version: 1.0
To: Chris Croswhite <ccroswhite@get2chip.com>
CC: b.j.smith@ieee.org, thebs@theseus.com, Eric Sandeen <sandeen@sgi.com>,
   Russell Cattelan <cattelan@thebarn.com>, linux-xfs@oss.sgi.com,
   linux-xfs-announce@oss.sgi.com
Subject: Re: XFS prerelease 0.10 test2 images  It's for Wolverine (RH 7.0.91 
		 aka RH71 beta 2)
References: <15033.39410.958392.383512@lupo.thebarn.com> <3ABA9C42.407227B6@sgi.com> <3ABAF48A.3020908@get2chip.com> <3ABAFEC7.37CC5D3B@ieee.org> <3ABB96E7.9030803@get2chip.com>
Content-Type: text/plain; charset=us-ascii; format=flowed
Content-Transfer-Encoding: 7bit
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

Stupid me.  I just had to look for it (next time I will open my eyes).

Thanks,
Chris

Chris Croswhite wrote:

> This would be my problem.  Could someone point me in the direction of 
> the Wolverine images?
> 
> Thanks,
> Chris
> 
> Bryan J. Smith wrote:
> 
>> Re: XFS prerelease 0.10 test2 images  It's for Wolverine (7.0.91
>> aka 7.1 beta 2)
>> 
>> Chris Croswhite wrote:
>> 
>>> I have tried the iso image.  However, everytime time I place
>>> in disc one of RH7.0, the installer indicates that the wrong disc
>>> is installed.  I have inserted disc 2 of RH7.0 also to no avail.
>>> Am I missing/using something here?
>> 
>> 
>> 
>>  >From my understanding, the PR 0.10 releases are for RedHat
>> "Wolverine" aka 7.0.91 aka 7.1 beta 2.
>> 
>> -- TheBS
>> 
>> P.S.  7.0.90 aka 7.1 beta [1] was "Fisher"
>> 
> 
> 


From owner-linux-xfs@oss.sgi.com Fri Mar 23 10:54:54 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2NIssB06979
	for linux-xfs-outgoing; Fri, 23 Mar 2001 10:54:54 -0800
Received: from roujin.gargoylecc.com ([65.100.85.34])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2NIsrM06972
	for <linux-xfs@oss.sgi.com>; Fri, 23 Mar 2001 10:54:53 -0800
Received: from roujin.gargoylecc.com (IDENT:ringram@roujin.gargoylecc.com [65.100.85.34])
	by roujin.gargoylecc.com (8.9.3/8.9.3) with ESMTP id DAA24787;
	Sat, 24 Mar 2001 03:06:11 -0700
Date: Sat, 24 Mar 2001 03:06:11 -0700 (MST)
From: Russel Ingram <ringram@gargoylecc.com>
To: thebs@theseus.com, b.j.smith@ieee.org
cc: linux-xfs@oss.sgi.com
Subject: Re: XFS prerelease 0.10 test2 images
In-Reply-To: <3ABB97B9.1EC0E52C@theseus.com>
Message-ID: <Pine.LNX.4.21.0103240301100.24774-100000@roujin.gargoylecc.com>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

On Fri, 23 Mar 2001, Bryan-TheBS-Smith wrote:

> Russel Ingram wrote:
> > Has anyone set up an ftp server to do ftp/http/nfs installs
> > from yet.  Russell maybe?
> 
> ???
> I just did a couple at home on my private network.  I didn't know
> SGI was offering this capability over the Internet?
> 
> BTW, it worked great over NFS.
> 
> -- TheBS
> 
I think SGI is not.  Russell Cattelan was kind enough with 0.9 to set
this up on one of his servers at umn and leave it open for those of us
that wanted to use it.  AFAIK, its a private thing and is probably not
advertised because Russell doesn't want his server flooded.


-- 
---------------------------------------------------------------
"Bill Gates and Microsoft have ruined the computer industry for
 a long time to come by creating a class of ignorant and lazy
 computer users."  --Russel Ingram

"Mommy ... can I go out and ... KILL TONIGHT!?"
                   --Glen  Danzig, The Misfits
---------------------------------------------------------------
Russ Ingram
Gargoyle Computer Consulting
(307)742-1361
www.gargoylecc.com


From owner-linux-xfs@oss.sgi.com Fri Mar 23 10:58:44 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2NIwiX07060
	for linux-xfs-outgoing; Fri, 23 Mar 2001 10:58:44 -0800
Received: from thor.theseus.com (south.orl-pub.theseus.com [12.108.42.66])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2NIwhM07053;
	Fri, 23 Mar 2001 10:58:43 -0800
Received: from theseus.com (IDENT:thebs@fugitive.theseus.com [192.168.0.242])
	by thor.theseus.com (8.9.3/8.9.3) with ESMTP id OAA05698;
	Fri, 23 Mar 2001 14:03:09 -0500
Message-ID: <3ABB9CE8.C539926B@theseus.com>
Date: Fri, 23 Mar 2001 13:58:48 -0500
From: Bryan-TheBS-Smith <thebs@theseus.com>
Reply-To: thebs@theseus.com, b.j.smith@ieee.org
Organization: (Personal)
X-Mailer: Mozilla 4.76 [en] (X11; U; Linux 2.2.17-0.17-FUGITIVE i686)
X-Accept-Language: en
MIME-Version: 1.0
To: Chris Croswhite <ccroswhite@get2chip.com>
CC: b.j.smith@ieee.org, Eric Sandeen <sandeen@sgi.com>,
   Russell Cattelan <cattelan@thebarn.com>, linux-xfs@oss.sgi.com,
   linux-xfs-announce@oss.sgi.com
Subject: Re: XFS prerelease 0.10 test2 images  Athlon kernels
References: <15033.39410.958392.383512@lupo.thebarn.com> <3ABA9C42.407227B6@sgi.com> <3ABAF48A.3020908@get2chip.com> <3ABAFEC7.37CC5D3B@ieee.org> <3ABB96E7.9030803@get2chip.com> <3ABB9A49.2080606@get2chip.com>
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

Re: XFS prerelease 0.10 test2 images  Athlon kernels

Chris Croswhite wrote:
> Stupid me.  I just had to look for it (next time I will open my eyes).

We'll bitchslap you silly next time.  ;-PPP

BTW, I left my house this morning with my system building some
Athlon-optimized kernels for the PR 0.10-test2 release.  Let me know
if you are interested in them (I will be putting them up to my site
tonight).  For the .config files, all I did was diff the i686/Athlon
.configs in stock Wolverine and patch them against the i686 .config
in PR 0.10-test2 (to generate the Athlon .config).

-- TheBS

--
Bryan "TheBS" Smith          chat:thebs413 @AOL/MSN/Yahoo
Engineer      mailto:b.j.smith@ieee.org,thebs@theseus.com
*********************************************************
"Never apply a Star Trek solution to a Babylon 5 problem"
                                    -- Nicholas C. Weaver

From owner-linux-xfs@oss.sgi.com Fri Mar 23 11:06:53 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2NJ6rf07303
	for linux-xfs-outgoing; Fri, 23 Mar 2001 11:06:53 -0800
Received: from porgy.srv.nld.sonera.net (mbox-01.soneraplaza.nl [195.66.15.137])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2NJ6qM07300
	for <linux-xfs@oss.sgi.com>; Fri, 23 Mar 2001 11:06:52 -0800
Received: from qn-212-58-163-36.quicknet.nl ([212.58.163.36]:64581 "EHLO
        auto-nb1.xs4all.nl") by soneramail.nl with ESMTP id <S46804AbRCWTGY>;
	Fri, 23 Mar 2001 20:06:24 +0100
Message-Id: <4.3.2.7.2.20010323200137.03a81a60@pop.xs4all.nl>
X-Sender: knuffie@pop.xs4all.nl
X-Mailer: QUALCOMM Windows Eudora Version 4.3.2
Date:   Fri, 23 Mar 2001 20:06:30 +0100
To: Chris Croswhite <ccroswhite@get2chip.com>
From: Seth Mos <knuffie@xs4all.nl>
Subject: Re: XFS prerelease 0.10 test2 images  It's for Wolverine (RH
  7.0.91  aka RH71 beta 2)
Cc: linux-xfs@oss.sgi.com
In-Reply-To: <3ABB96E7.9030803@get2chip.com>
References: <15033.39410.958392.383512@lupo.thebarn.com>
 <3ABA9C42.407227B6@sgi.com>
 <3ABAF48A.3020908@get2chip.com>
 <3ABAFEC7.37CC5D3B@ieee.org>
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"; format=flowed
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

At 10:33 23-3-2001 -0800, Chris Croswhite wrote:
>This would be my problem.  Could someone point me in the direction of the 
>Wolverine images?

try http://www.redhat.com/mirrors.html and choose something close or on the 
same continent. Also check which network they are one. My Cable ISP has a 
terrible broken uplink to UUnet (amsterdan->new-york->UK) and some others 
networks even have a peering agreement or are connected by KPN/Qwest networks.
This makes the difference between 3 or 50 KB/s stable traffic.

And you whant to test this, it can make the difference between 2 days or 5 
hours.

Good luck


>Thanks,
>Chris
>
>Bryan J. Smith wrote:
>
>>Re: XFS prerelease 0.10 test2 images  It's for Wolverine (7.0.91
>>aka 7.1 beta 2)
>>Chris Croswhite wrote:
>>
>>>I have tried the iso image.  However, everytime time I place
>>>in disc one of RH7.0, the installer indicates that the wrong disc
>>>is installed.  I have inserted disc 2 of RH7.0 also to no avail.
>>>Am I missing/using something here?
>>
>> >From my understanding, the PR 0.10 releases are for RedHat
>>"Wolverine" aka 7.0.91 aka 7.1 beta 2.
>>-- TheBS
>>P.S.  7.0.90 aka 7.1 beta [1] was "Fisher"
>

--
Seth
Has anybody seen my lightbulb?
I _really_ need some light here.


From owner-linux-xfs@oss.sgi.com Fri Mar 23 11:11:45 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2NJBjw07386
	for linux-xfs-outgoing; Fri, 23 Mar 2001 11:11:45 -0800
Received: from chiara.elte.hu (chiara.elte.hu [157.181.150.200])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2NJBhM07383
	for <linux-xfs@oss.sgi.com>; Fri, 23 Mar 2001 11:11:43 -0800
Received: by chiara.elte.hu (Postfix, from userid 17000)
	id AB291187D; Fri, 23 Mar 2001 20:11:37 +0100 (CET)
Date: Fri, 23 Mar 2001 20:11:37 +0100
From: KELEMEN Peter <fuji@elte.hu>
To: linux-xfs@oss.sgi.com
Subject: xfs_repair wipes ACL entries, maybe endian-related?
Message-ID: <20010323201137.D19598@chiara.elte.hu>
Reply-To: KELEMEN Peter <fuji@elte.hu>
Mime-Version: 1.0
Content-Type: text/plain; charset=iso-8859-1
Content-Disposition: inline
Content-Transfer-Encoding: 8bit
User-Agent: Mutt/1.2.5i
Organization: ELTE Eotvos Lorand University of Sciences, Budapest, Hungary
X-GPG-KeyID: 1024D/EE4C26E8 2000-03-20
X-GPG-Fingerprint: D402 4AF3 7488 165B CC34  4147 7F0C D922 EE4C 26E8
X-PGP-KeyID: 1024/45F83E45 1998/04/04
X-PGP-Fingerprint: 26 87 63 4B 07 28 1F AD  6D AA B5 8A D6 03 0F BF
X-Comment: Personal opinion.  Paragraphs might have been reformatted.
X-Copyright: Forwarding or publishing without permission is prohibited.
X-Accept-Language: hu,en
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

Hello,

[ I have no access to the test machine, I'm writing from my memory ]

I've been peeking at ACLs today (CVS tree as of 2001/03/22 ~16:00
CET).  After setting the ACL on a file, unmounting the filesystem
and running xfs_repair, it intends to wipe out the ACL entries.
Looking at the code, xfsprogs/repair/attr_repair.c::valid_acl()
decides it should be cleared.  Reason: acl_cnt > MAX_ACL_ENTRIES,
I sniffed around with xfs_db, SGI_ACL_FILE attr with 304 byte
value is OK, in my case it was \000\000\000\005... (normal u,g,o
plus one u and a mask); valid_acl thinks that acl_cnt is 83886080,
which just happens to be 0x5000000.  I somehow suspect that adding
some INT_GET(..., ARCH_CONVERT) macros around the obvious places
in valid_acl() is not The Right Thing To Do(tm), hence no patch.

Peter

-- 
    .+'''+.         .+'''+.         .+'''+.         .+'''+.         .+''
 Kelemen Péter     /       \       /       \       /      fuji@elte.hu
.+'         `+...+'         `+...+'         `+...+'         `+...+'

From owner-linux-xfs@oss.sgi.com Fri Mar 23 11:43:24 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2NJhOp07944
	for linux-xfs-outgoing; Fri, 23 Mar 2001 11:43:24 -0800
Received: from haddock.cd.chalmers.se (haddock.cd.chalmers.se [129.16.79.2])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2NJhNM07941
	for <linux-xfs@oss.sgi.com>; Fri, 23 Mar 2001 11:43:23 -0800
Received: from haddock.cd.chalmers.se (localhost [127.0.0.1])
	by haddock.cd.chalmers.se (8.9.3+Sun/8.8.7) with ESMTP id UAA15425;
	Fri, 23 Mar 2001 20:43:01 +0100 (MET)
Message-Id: <200103231943.UAA15425@haddock.cd.chalmers.se>
X-Face: ^"+mumOlNwo;yI>`\39\txuVze?eiR9uqpo2*mE!9MWXgXI0v(3ArwymNWe'.q:eLl!=guD
 x{jGFEWN6,#HoN%2qRW;7.CL]9%Ap,067"u1%!NUqS.MhV'+,6$Fj-;W2Z}Y,JUZ'L+f<?~E{ID*=T
 V\NxAOu>)|B@3k3n;gLl*#i[(J-os#fNnDJ8m["|JWNwpORh|_.MGkR#|a~QS!"4hEQ{O{[Ii14{xD
 PU/:5wuv7m1=TK=.>G8wdfpY~]{H(Qa\1`%|Hz:!)c3f9UOW|WgE"4d\E7?oDu9.
To: ppetru@ppetru.net (Petru Paler)
cc: linux-xfs@oss.sgi.com
Subject: Re: Any sparc64 issues? 
In-Reply-To: Message from ppetru@ppetru.net (Petru Paler) 
   of "Fri, 23 Mar 2001 19:21:28 +0200." <20010323192128.A2028@ppetru.net> 
References: <20010323192128.A2028@ppetru.net> 
MIME-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
Content-ID: <15422.985376581.1@haddock.cd.chalmers.se>
Date: Fri, 23 Mar 2001 20:43:01 +0100
From: Anders Hammarquist <iko@cd.chalmers.se>
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

> Are there any sparc64 users of xfs? Does it have any showstopper
> bugs?

I just compiled the CVS version yesterday, and apart from one small
patch to get it to find limits.h (posted earlier today) it seems to
be working fine.

/Anders

-- 
 -- Of course I'm crazy, but that doesn't mean I'm wrong.
Anders Hammarquist                                  | iko@cd.chalmers.se
Physics student, Chalmers University of Technology, | Hem: +46 31 88 48 50
G|teborg, Sweden.           RADIO: SM6XMM and N2JGL | Mob: +46 707 27 86 87

From owner-linux-xfs@oss.sgi.com Fri Mar 23 12:19:06 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2NKJ6D08517
	for linux-xfs-outgoing; Fri, 23 Mar 2001 12:19:06 -0800
Received: from ifi.informatik.uni-stuttgart.de (ifi.informatik.uni-stuttgart.de [129.69.211.1])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2NKJ4M08509
	for <linux-xfs@oss.sgi.com>; Fri, 23 Mar 2001 12:19:05 -0800
Received: from techno.informatik.uni-stuttgart.de (techno [129.69.218.24]) by ifi.informatik.uni-stuttgart.de (8.9.3/2.2) with ESMTP id VAA01528 for <linux-xfs@oss.sgi.com>; Fri, 23 Mar 2001 21:18:57 +0100 (MET)
Received: (from magallon@localhost) by techno.informatik.uni-stuttgart.de (8.11.0/2.2) id f2NKJ1Z32150 for linux-xfs@oss.sgi.com; Fri, 23 Mar 2001 21:19:01 +0100
Date: Fri, 23 Mar 2001 21:19:01 +0100
From: "Marcelo E. Magallon" <marcelo.magallon@bigfoot.com>
To: linux-xfs@oss.sgi.com
Subject: [OT] XFS on a SGI VW 230?
Message-ID: <20010323211901.A32118@techno.informatik.uni-stuttgart.de>
Mail-Followup-To: "Marcelo E. Magallon" <marcelo.magallon@bigfoot.com>,
	linux-xfs@oss.sgi.com
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.1.14i
X-Operating-System: Linux techno 2.4.2-ac8
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

Hi,

 I'm sorry about the offtopic, but I can't find the info on the SGI
 website.  We have got an SGI Visual Workstation 230 here and I'd like
 to install XFS on it (it's already installed, as a matter of fact).
 The problem I'm having now is the kernel-SGI-VPro-rpmgen package
 doesn't work out of the box with 2.4 kernels.  I have patched it, but I
 still can't get it to work properly, and I would like to know if
 there's an official upgrade from SGI somewhere.  SGI's support official
 position seems to be "we don't support 2.4 kernels on the released
 proPack VWE"

 TIA,

-- 
Marcelo

From owner-linux-xfs@oss.sgi.com Fri Mar 23 13:18:07 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2NLI7I10040
	for linux-xfs-outgoing; Fri, 23 Mar 2001 13:18:07 -0800
Received: from tux.mkp.net (tux.mkp.net [130.225.60.11])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2NLI5M10037
	for <linux-xfs@oss.sgi.com>; Fri, 23 Mar 2001 13:18:05 -0800
Received: from tux.mkp.net ([130.225.60.11] helo=jaguar.mkp.net)
	by tux.mkp.net with esmtp (Exim 3.16 #1)
	id 14gYwc-0002wq-00; Fri, 23 Mar 2001 22:18:03 +0100
Received: (from mkp@localhost)
	by jaguar.mkp.net (8.9.3/8.9.3) id QAA15688;
	Fri, 23 Mar 2001 16:17:38 -0500
X-Authentication-Warning: jaguar.mkp.net: mkp set sender to mkp@mkp.net using -f
To: linux-xfs@oss.sgi.com
Subject: TAKE - Fix grio include
From: "Martin K. Petersen" <mkp@linuxcare.com>
Organization: Linuxcare, Inc.
Date: 23 Mar 2001 16:17:38 -0500
Message-ID: <yq1k85gmbgt.fsf@jaguar.mkp.net>
Lines: 18
User-Agent: Gnus/5.0808 (Gnus v5.8.8) XEmacs/21.1 (Canyonlands)
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk


Fix from Anders Hammarquist <iko@cd.chalmers.se>.

Date:  Fri Mar 23 13:13:53 PST 2001
Workarea:  sshgate.corp.sgi.com:/home/mkp/XFS/slinx-pristine

The following file(s) were checked into:
  bonnie.engr.sgi.com:/isms/slinx/2.4.x-xfs


Modid:  2.4.x-xfs:slinx:90771a
linux/include/linux/grio.h - 1.5
        - Small fix for sparc64 builds

-- 
Martin K. Petersen, Principal Linux Consultant, Linuxcare, Inc.
mkp@linuxcare.com, http://www.linuxcare.com/
SGI XFS for Linux Developer, http://oss.sgi.com/projects/xfs/

From owner-linux-xfs@oss.sgi.com Fri Mar 23 13:31:25 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2NLVPb10231
	for linux-xfs-outgoing; Fri, 23 Mar 2001 13:31:25 -0800
Received: from wisdom.myplace.net (cc19815-a.zwoll1.ov.nl.home.com [212.204.138.247])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2NLVKM10228
	for <linux-xfs@oss.sgi.com>; Fri, 23 Mar 2001 13:31:24 -0800
Received: from ws1 (ws1.myplace.net [192.168.1.15])
	by wisdom.myplace.net (Postfix) with SMTP id D9B31155
	for <linux-xfs@oss.sgi.com>; Fri, 23 Mar 2001 22:31:17 +0100 (CET)
Message-ID: <006401c0b42c$072faea0$0f01a8c0@myplace.net>
From: "Bas" <list@showme.wox.org>
To: <linux-xfs@oss.sgi.com>
Subject: 2.4.2 - XFS kernel March 19th & quota
Date: Fri, 23 Mar 2001 22:31:15 -0800
MIME-Version: 1.0
Content-Type: text/plain;
	charset="iso-8859-1"
Content-Transfer-Encoding: 7bit
X-Priority: 3
X-MSMail-Priority: Normal
X-Mailer: Microsoft Outlook Express 5.00.2615.200
X-MimeOLE: Produced By Microsoft MimeOLE V5.00.2615.200
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

Hi,

I'm running XFS with quota support and it works, but I can't seem to find
(checked the man pages and the documentation that comes with the cvs-tree)
the block size that is used when setting quota. I thought is 1024 B now, but
am I wrong ?

Thanks again,
Bas.




From owner-linux-xfs@oss.sgi.com Fri Mar 23 13:45:44 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2NLjiq10551
	for linux-xfs-outgoing; Fri, 23 Mar 2001 13:45:44 -0800
Received: from deliverator.sgi.com (deliverator.sgi.com [204.94.214.10])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2NLjiM10548
	for <linux-xfs@oss.sgi.com>; Fri, 23 Mar 2001 13:45:44 -0800
Received: from cthulhu.engr.sgi.com (gate3-relay.engr.sgi.com [130.62.1.234]) by deliverator.sgi.com (980309.SGI.8.8.8-aspam-6.2/980310.SGI-aspam) via ESMTP id NAA19511
	for <linux-xfs@oss.sgi.com>; Fri, 23 Mar 2001 13:44:31 -0800 (PST)
	mail_from (tduffy@engr.sgi.com)
Received: from dbear.engr.sgi.com (dbear.engr.sgi.com [163.154.18.85])
	by cthulhu.engr.sgi.com (SGI-8.9.3/8.9.3) with ESMTP id NAA76406;
	Fri, 23 Mar 2001 13:44:26 -0800 (PST)
Date: Fri, 23 Mar 2001 13:40:27 -0800 (PST)
From: Tom Duffy <tduffy@engr.sgi.com>
To: "Marcelo E. Magallon" <marcelo.magallon@bigfoot.com>
cc: <linux-xfs@oss.sgi.com>
Subject: Re: [OT] XFS on a SGI VW 230?
In-Reply-To: <20010323211901.A32118@techno.informatik.uni-stuttgart.de>
Message-ID: <Pine.LNX.4.30.0103231338080.19348-100000@dbear.engr.sgi.com>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

well, if you are using 1.3-vw, expect things to *not* work under a 2.4
kernel.

if you are using vwe 3.0, you can hack it to work on 2.4 kernel, although
this is completely unsupported.

you will need to run sgivpro_rpmgen to get new driver rpms for your
kernel.

if you wait til vwe 3.1, you will get 2.4 kernel w/ xfs support officially
supported.

-tduffy

On Fri, 23 Mar 2001, Marcelo E. Magallon wrote:

> Hi,
>
>  I'm sorry about the offtopic, but I can't find the info on the SGI
>  website.  We have got an SGI Visual Workstation 230 here and I'd like
>  to install XFS on it (it's already installed, as a matter of fact).
>  The problem I'm having now is the kernel-SGI-VPro-rpmgen package
>  doesn't work out of the box with 2.4 kernels.  I have patched it, but I
>  still can't get it to work properly, and I would like to know if
>  there's an official upgrade from SGI somewhere.  SGI's support official
>  position seems to be "we don't support 2.4 kernels on the released
>  proPack VWE"
>
>  TIA,
>
>



From owner-linux-xfs@oss.sgi.com Fri Mar 23 14:11:21 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2NMBL511050
	for linux-xfs-outgoing; Fri, 23 Mar 2001 14:11:21 -0800
Received: from tux.mkp.net (tux.mkp.net [130.225.60.11])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2NMBKM11047
	for <linux-xfs@oss.sgi.com>; Fri, 23 Mar 2001 14:11:20 -0800
Received: from tux.mkp.net ([130.225.60.11] helo=jaguar.mkp.net)
	by tux.mkp.net with esmtp (Exim 3.16 #1)
	id 14gZm5-0002zI-00; Fri, 23 Mar 2001 23:11:18 +0100
Received: (from mkp@localhost)
	by jaguar.mkp.net (8.9.3/8.9.3) id RAA15699;
	Fri, 23 Mar 2001 17:10:49 -0500
X-Authentication-Warning: jaguar.mkp.net: mkp set sender to mkp@mkp.net using -f
To: linux-xfs@oss.sgi.com
Subject: TAKE - mkfs logsize heuristics
From: "Martin K. Petersen" <mkp@linuxcare.com>
Organization: Linuxcare, Inc.
Date: 23 Mar 2001 17:10:48 -0500
Message-ID: <yq1g0g4m907.fsf@jaguar.mkp.net>
Lines: 18
User-Agent: Gnus/5.0808 (Gnus v5.8.8) XEmacs/21.1 (Canyonlands)
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk


Date:  Fri Mar 23 14:09:53 PST 2001
Workarea:  sshgate.corp.sgi.com:/home/mkp/XFS/slinx-pristine

The following file(s) were checked into:
  bonnie.engr.sgi.com:/isms/slinx/2.4.x-xfs


Modid:  2.4.x-xfs:slinx:90778a
cmd/xfsprogs/mkfs/xfs_mkfs.c - 1.3
cmd/xfsprogs/man/man8/mkfs.xfs.8 - 1.2
        - Add heuristics to the mkfs command to make it guesstimate
          internal log size.

-- 
Martin K. Petersen, Principal Linux Consultant, Linuxcare, Inc.
mkp@linuxcare.com, http://www.linuxcare.com/
SGI XFS for Linux Developer, http://oss.sgi.com/projects/xfs/

From owner-linux-xfs@oss.sgi.com Fri Mar 23 15:48:08 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2NNm8F13238
	for linux-xfs-outgoing; Fri, 23 Mar 2001 15:48:08 -0800
Received: from arrakis.vcu.edu (arrakis.vcu.edu [128.172.1.132])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2NNm8M13235
	for <linux-xfs@oss.sgi.com>; Fri, 23 Mar 2001 15:48:08 -0800
Received: from 0 (arrakis.vcu.edu [127.0.0.1])
	by arrakis.vcu.edu (8.9.3/8.9.3) with ESMTP id SAA19107
	for <linux-xfs@oss.sgi.com>; Fri, 23 Mar 2001 18:48:07 -0500
Message-Id: <200103232348.SAA19107@arrakis.vcu.edu>
Date: Fri, 23 Mar 2001 18:48:07 -500
Content-Transfer-Encoding: 8bit
To: linux-xfs@oss.sgi.com
From: "K. Mitchell Russell" <kmrussel@hsc.vcu.edu>
Content-Type: text/plain; charset=iso-8859-1
Subject: Unable to mount SCSI root fs with pr0.10test1 or test2
MIME-Version: 1.0
User-Agent: IMHO/0.98 (Webmail for Roxen)
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

Dear Colleagues,

I have tried pr0.10test1 and test2, and always get a kernel panic on
boot:

VFS: Cannot open root device "803" or 08:03
Please append a correct "root=" boot option
Kernel panic: VFS: Unable to mount root fs on 08:03

This is a SCSI drive on aic7890 controller with all partitions
(/,/boot,/var,/usr,/home) set to XFS at install.

No luck changing ramdisk_size at boottime either (increased ramdisk up
to 100000k even!)

root= was specified correctly at boottime.

XFS-pr0.9 worked like a charm with the stock kernel and initrd
install, so I'm curious why the pr0.10 stock kernel and initrd isn't. 
Probably is not loading scsi module properly.  I recall seeing
something similar some time ago when i compiled a 2.4 kernel from the
xfs devel tree with the scsi_support as a module and not built in.  At
that time, despite configuring the initrd correctly to load all scsi
modules, it would panic as well.  After install i usually compile a
new kernel with SCSI built in, but this is a problem for initial
installing of XFS in pr0.10.  

Once I get my test workstation running again, I will compare the SCSI
options and modules in the pr0.9 kernel and pr0.10 kernel.  In the
mean time, anyone else having a similar problem?

Regards,

K. Mitchell Russell, M.D.
MedITAC Research Lab
kmrussel@hsc.vcu.edu
www.meditac.com 

From owner-linux-xfs@oss.sgi.com Fri Mar 23 16:07:52 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2O07qE13700
	for linux-xfs-outgoing; Fri, 23 Mar 2001 16:07:52 -0800
Received: from mail11.jump.net (mail11.jump.net [206.196.91.11])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2O07pM13697
	for <linux-xfs@oss.sgi.com>; Fri, 23 Mar 2001 16:07:51 -0800
Received: from Porter (aus-dsl-dhcp1-26.customer.jump.net [63.163.168.26])
	by mail11.jump.net (8.10.2/) with ESMTP id f2O07nv16479;
	Fri, 23 Mar 2001 18:07:49 -0600 (CST)
Subject: Re: Unable to mount SCSI root fs with pr0.10test1 or test2
From: Eric Sandeen <sandeen@sgi.com>
To: "K." Mitchell Russell <kmrussel@hsc.vcu.edu>
Cc: linux-xfs@oss.sgi.com
In-Reply-To: <200103232348.SAA19107@arrakis.vcu.edu>
References: <200103232348.SAA19107@arrakis.vcu.edu>
Content-Type: text/plain
X-Mailer: Evolution/0.9+cvs.2001.03.23.09.06 (Preview Release)
Date: 23 Mar 2001 18:08:49 -0600
Message-Id: <985392529.11122.1.camel@Porter>
Mime-Version: 1.0
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

On 23 Mar 2001 18:48:07 -0500, K. Mitchell Russell wrote:
> Dear Colleagues,
> 
> I have tried pr0.10test1 and test2, and always get a kernel panic on
> boot:
> 
> VFS: Cannot open root device "803" or 08:03
> Please append a correct "root=" boot option
> Kernel panic: VFS: Unable to mount root fs on 08:03

Hm, I've heard a couple reports of this on the PR0.9/RH7.0 installer,
but this is the first for the new bits.

I'll try doing a scsi install here and see how it goes.

Do you see the ramdisk and scsi module load before the panic?

-Eric




From owner-linux-xfs@oss.sgi.com Fri Mar 23 16:37:41 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2O0bfT15004
	for linux-xfs-outgoing; Fri, 23 Mar 2001 16:37:41 -0800
Received: from dragon.vcu.edu (dragon.vcu.edu [128.172.65.9])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2O0beM15000
	for <linux-xfs@oss.sgi.com>; Fri, 23 Mar 2001 16:37:40 -0800
Received: from comet.vcu.edu (comet.vcu.edu [128.172.1.33])
	by dragon.vcu.edu (8.9.3/8.9.3) with ESMTP id TAA2967511;
	Fri, 23 Mar 2001 19:40:17 -0500 (EST)
Date: Fri, 23 Mar 2001 19:37:37 -0500 (EST)
From: K Mitchell Russell <kmrussel@hsc.vcu.edu>
X-Sender: kmrussel@comet.vcu.edu
To: Eric Sandeen <sandeen@sgi.com>
cc: linux-xfs@oss.sgi.com
Subject: Re: Unable to mount SCSI root fs with pr0.10test1 or test2
In-Reply-To: <985392529.11122.1.camel@Porter>
Message-ID: <Pine.A41.4.10.10103231933150.34436-100000@comet.vcu.edu>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

I saw the ramdisk load, but not the scsi module, which leads me to
believe that's the problem.  I'm reverted back to the pr0.9 (which boots
fine) now, and taking a look at the initrd linuxrc.

on a separate note, i noticed the create boot floppy at the end of
install did not work, reporting an error in both the pr0.10test1 and
test2.

Thanks,
Mitchell

________________________________________________________________________
K. Mitchell Russell, M.D.                        |  kmrussel@hsc.vcu.edu
Research Fellow, MedITAC Research Lab            |  www.meditac.com

On 23 Mar 2001, Eric Sandeen wrote:

> On 23 Mar 2001 18:48:07 -0500, K. Mitchell Russell wrote:
> > Dear Colleagues,
> > 
> > I have tried pr0.10test1 and test2, and always get a kernel panic on
> > boot:
> > 
> > VFS: Cannot open root device "803" or 08:03
> > Please append a correct "root=" boot option
> > Kernel panic: VFS: Unable to mount root fs on 08:03
> 
> Hm, I've heard a couple reports of this on the PR0.9/RH7.0 installer,
> but this is the first for the new bits.
> 
> I'll try doing a scsi install here and see how it goes.
> 
> Do you see the ramdisk and scsi module load before the panic?
> 
> -Eric
> 
> 
> 
> 


From owner-linux-xfs@oss.sgi.com Fri Mar 23 17:10:55 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2O1Atq15626
	for linux-xfs-outgoing; Fri, 23 Mar 2001 17:10:55 -0800
Received: from mail.get2chip.com ([64.169.83.2])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2O1AsM15623
	for <linux-xfs@oss.sgi.com>; Fri, 23 Mar 2001 17:10:54 -0800
Message-ID: <3ABBF32E.30704@get2chip.com>
Date: Fri, 23 Mar 2001 17:06:54 -0800
From: Chris Croswhite <ccroswhite@get2chip.com>
MIME-Version: 1.0
To: K Mitchell Russell <kmrussel@hsc.vcu.edu>
CC: Eric Sandeen <sandeen@sgi.com>, linux-xfs@oss.sgi.com
Subject: Re: Unable to mount SCSI root fs with pr0.10test1 or test2
References: <Pine.A41.4.10.10103231933150.34436-100000@comet.vcu.edu>
Content-Type: text/plain; charset=us-ascii; format=flowed
Content-Transfer-Encoding: 7bit
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

I am also receiving the same error.  My system also has a scsi drive.  
As for the scsi driver being loaded, I can not tell, the system moves 
too quickly, however, I would asume so.

Chris

K Mitchell Russell wrote:

> I saw the ramdisk load, but not the scsi module, which leads me to
> believe that's the problem.  I'm reverted back to the pr0.9 (which boots
> fine) now, and taking a look at the initrd linuxrc.
> 
> on a separate note, i noticed the create boot floppy at the end of
> install did not work, reporting an error in both the pr0.10test1 and
> test2.
> 
> Thanks,
> Mitchell
> 
> ________________________________________________________________________
> K. Mitchell Russell, M.D.                        |  kmrussel@hsc.vcu.edu
> Research Fellow, MedITAC Research Lab            |  www.meditac.com
> 
> On 23 Mar 2001, Eric Sandeen wrote:
> 
>> On 23 Mar 2001 18:48:07 -0500, K. Mitchell Russell wrote:
>> 
>>> Dear Colleagues,
>>> 
>>> I have tried pr0.10test1 and test2, and always get a kernel panic on
>>> boot:
>>> 
>>> VFS: Cannot open root device "803" or 08:03
>>> Please append a correct "root=" boot option
>>> Kernel panic: VFS: Unable to mount root fs on 08:03
>> 
>> Hm, I've heard a couple reports of this on the PR0.9/RH7.0 installer,
>> but this is the first for the new bits.
>> 
>> I'll try doing a scsi install here and see how it goes.
>> 
>> Do you see the ramdisk and scsi module load before the panic?
>> 
>> -Eric
>> 
>> 
>> 
>> 


From owner-linux-xfs@oss.sgi.com Fri Mar 23 17:14:18 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2O1EIQ15685
	for linux-xfs-outgoing; Fri, 23 Mar 2001 17:14:18 -0800
Received: from dragon.vcu.edu (dragon.vcu.edu [128.172.65.9])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2O1EHM15682
	for <linux-xfs@oss.sgi.com>; Fri, 23 Mar 2001 17:14:18 -0800
Received: from comet.vcu.edu (comet.vcu.edu [128.172.1.33])
	by dragon.vcu.edu (8.9.3/8.9.3) with ESMTP id UAA3022046;
	Fri, 23 Mar 2001 20:16:56 -0500 (EST)
Date: Fri, 23 Mar 2001 20:14:16 -0500 (EST)
From: K Mitchell Russell <kmrussel@hsc.vcu.edu>
X-Sender: kmrussel@comet.vcu.edu
To: Eric Sandeen <sandeen@sgi.com>
cc: linux-xfs@oss.sgi.com
Subject: Re: Unable to mount SCSI root fs with pr0.10test1 or test2
In-Reply-To: <985392529.11122.1.camel@Porter>
Message-ID: <Pine.A41.4.10.10103232007550.41756-100000@comet.vcu.edu>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

Interestingly, i reinstalled pr0.9, booted up fine.  Then upgraded
mkinitrd to 3.0.6, e2fsprogs to 1.19, and installed the
kernel-smp-2.4.2-SGI_XFS_0.10.i686.rpm from the pr0.10test2.  Then
created a new initrd from the 2.4.2-SGI_XFS_0.10smp modules, and this
boots fine!  Perhaps there is a problem with the Wolverine installer
creating the initrd...

________________________________________________________________________
K. Mitchell Russell, M.D.                        |  kmrussel@hsc.vcu.edu
Research Fellow, MedITAC Research Lab            |  www.meditac.com

On 23 Mar 2001, Eric Sandeen wrote:

> On 23 Mar 2001 18:48:07 -0500, K. Mitchell Russell wrote:
> > Dear Colleagues,
> > 
> > I have tried pr0.10test1 and test2, and always get a kernel panic on
> > boot:
> > 
> > VFS: Cannot open root device "803" or 08:03
> > Please append a correct "root=" boot option
> > Kernel panic: VFS: Unable to mount root fs on 08:03
> 
> Hm, I've heard a couple reports of this on the PR0.9/RH7.0 installer,
> but this is the first for the new bits.
> 
> I'll try doing a scsi install here and see how it goes.
> 
> Do you see the ramdisk and scsi module load before the panic?
> 
> -Eric
> 
> 
> 
> 


From owner-linux-xfs@oss.sgi.com Fri Mar 23 17:26:35 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2O1QZs16116
	for linux-xfs-outgoing; Fri, 23 Mar 2001 17:26:35 -0800
Received: from mail11.jump.net (mail11.jump.net [206.196.91.11])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2O1QYM16113
	for <linux-xfs@oss.sgi.com>; Fri, 23 Mar 2001 17:26:34 -0800
Received: from Porter (aus-dsl-dhcp1-26.customer.jump.net [63.163.168.26])
	by mail11.jump.net (8.10.2/) with ESMTP id f2O1QXv00943;
	Fri, 23 Mar 2001 19:26:33 -0600 (CST)
Subject: Re: Unable to mount SCSI root fs with pr0.10test1 or test2
From: Eric Sandeen <sandeen@sgi.com>
To: K Mitchell Russell <kmrussel@hsc.vcu.edu>
Cc: linux-xfs@oss.sgi.com
In-Reply-To: <Pine.A41.4.10.10103232007550.41756-100000@comet.vcu.edu>
References: <Pine.A41.4.10.10103232007550.41756-100000@comet.vcu.edu>
Content-Type: text/plain
X-Mailer: Evolution/0.9+cvs.2001.03.23.09.06 (Preview Release)
Date: 23 Mar 2001 19:27:34 -0600
Message-Id: <985397254.11121.2.camel@Porter>
Mime-Version: 1.0
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

On 23 Mar 2001 20:14:16 -0500, K Mitchell Russell wrote:
> Interestingly, i reinstalled pr0.9, booted up fine.  Then upgraded
> mkinitrd to 3.0.6, e2fsprogs to 1.19, and installed the
> kernel-smp-2.4.2-SGI_XFS_0.10.i686.rpm from the pr0.10test2.  Then
> created a new initrd from the 2.4.2-SGI_XFS_0.10smp modules, and this
> boots fine!  Perhaps there is a problem with the Wolverine installer
> creating the initrd...

Hm.... well, I did a full scsi install here on an adaptec controller
(aic7xxx driver) and it's a good news / bad news sort of thing - the
good news is that it all went fine, and booted right up.  The bad news
is I didn't reproduce your error.  :)

I'll take a look at what anaconda is doing... perhaps there is a problem
there.

Also, I'm not surprised that the boot floppy creation failed - kernel +
xfs + scsi driver is getting pretty big...  I guess we should disable
that or at least offer a warning if we can't make it fit.

-Eric

 



From owner-linux-xfs@oss.sgi.com Fri Mar 23 17:34:58 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2O1YwO16309
	for linux-xfs-outgoing; Fri, 23 Mar 2001 17:34:58 -0800
Received: from home.smithconcepts.com (ubr-35.28.151.oviedo.cfl.rr.com [65.35.28.151])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2O1YvM16306
	for <linux-xfs@oss.sgi.com>; Fri, 23 Mar 2001 17:34:58 -0800
Received: from ieee.org (IDENT:bjsmith@bitman.oviedo.smithconcepts.com [172.24.24.192])
	by home.smithconcepts.com (8.9.3/8.9.3) with ESMTP id UAA09170;
	Fri, 23 Mar 2001 20:29:00 -0500
Message-ID: <3ABBFBEA.23BA0D17@ieee.org>
Date: Fri, 23 Mar 2001 20:44:10 -0500
From: "Bryan J. Smith" <b.j.smith@ieee.org>
Reply-To: b.j.smith@ieee.org, thebs@theseus.com
Organization: SmithConcepts/Personal
X-Mailer: Mozilla 4.76 [en] (X11; U; Linux 2.2.18-0.13 i686)
X-Accept-Language: en
MIME-Version: 1.0
To: Eric Sandeen <sandeen@sgi.com>
CC: K Mitchell Russell <kmrussel@hsc.vcu.edu>, linux-xfs@oss.sgi.com
Subject: Re: Unable to mount SCSI root fs with pr0.10test1 or test2
References: <Pine.A41.4.10.10103232007550.41756-100000@comet.vcu.edu> <985397254.11121.2.camel@Porter>
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

Eric Sandeen wrote:
> Hm.... well, I did a full scsi install here on an adaptec controller
> (aic7xxx driver) and it's a good news / bad news sort of thing - the
> good news is that it all went fine, and booted right up.  The bad news
> is I didn't reproduce your error.  :)
> I'll take a look at what anaconda is doing... perhaps there is a problem
> there.
> Also, I'm not surprised that the boot floppy creation failed - kernel +
> xfs + scsi driver is getting pretty big...  I guess we should disable
> that or at least offer a warning if we can't make it fit.

I'm golden with a pair of IDE disks on the Iwill SideRAID66 (HPT368
chipset).  It booted just fine with all filesystems XFS except
/boot.  Ananaconda not create an initrd though.

I'm seeming to have a serious issue with PAM and NIS -- which is a
Wolverine issue.  Anyone else here having issues?  The second I try
to bind to my NIS server, _all_ authentication goes down (even
local).  I have to shutdown ypbind completely (even if it didn't
bind -- which it is not anyway) before any authentication is
possible.

Time to check Bugzilla I guess ...

-- TheBS

P.S.  I've had nothing but problems with aic7xxx and more recent
kernels anyway -- especially 2940's with older BIOSes.  I've given
up on Adaptec controllers and stick with AdvanSys and Symbios
Logic-based chipsets for general cards (and have for a long time),
and Mylex for high-end RAID controllers.  On the cheap (for just
peripherials), I favor the $20 TekRam DC-315U (TekRam S1040
chipset).

-- 
Bryan "TheBS" Smith         chat:thebs413 @AOL/MSN/Yahoo
Engineer     mailto:b.j.smith@ieee.org,thebs@theseus.com
********************************************************
"Linux will do for applications what the Internet did to
 networks" -- Sam Palmisano, IBM Chief Operating Officer

From owner-linux-xfs@oss.sgi.com Fri Mar 23 17:38:03 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2O1c3v16351
	for linux-xfs-outgoing; Fri, 23 Mar 2001 17:38:03 -0800
Received: from mail.get2chip.com ([64.169.83.2])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2O1c2M16348
	for <linux-xfs@oss.sgi.com>; Fri, 23 Mar 2001 17:38:02 -0800
Message-ID: <3ABBF990.3060201@get2chip.com>
Date: Fri, 23 Mar 2001 17:34:08 -0800
From: Chris Croswhite <ccroswhite@get2chip.com>
MIME-Version: 1.0
To: Eric Sandeen <sandeen@sgi.com>
CC: K Mitchell Russell <kmrussel@hsc.vcu.edu>, linux-xfs@oss.sgi.com
Subject: Re: Unable to mount SCSI root fs with pr0.10test1 or test2
References: <Pine.A41.4.10.10103232007550.41756-100000@comet.vcu.edu> <985397254.11121.2.camel@Porter>
Content-Type: text/plain; charset=us-ascii; format=flowed
Content-Transfer-Encoding: 7bit
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

I have tried the 10test on another machine with scsi drive.  Again, I 
get the same panic of can not open root device.  Is there a fix for this?

Chris

Eric Sandeen wrote:

> On 23 Mar 2001 20:14:16 -0500, K Mitchell Russell wrote:
> 
>> Interestingly, i reinstalled pr0.9, booted up fine.  Then upgraded
>> mkinitrd to 3.0.6, e2fsprogs to 1.19, and installed the
>> kernel-smp-2.4.2-SGI_XFS_0.10.i686.rpm from the pr0.10test2.  Then
>> created a new initrd from the 2.4.2-SGI_XFS_0.10smp modules, and this
>> boots fine!  Perhaps there is a problem with the Wolverine installer
>> creating the initrd...
> 
> 
> Hm.... well, I did a full scsi install here on an adaptec controller
> (aic7xxx driver) and it's a good news / bad news sort of thing - the
> good news is that it all went fine, and booted right up.  The bad news
> is I didn't reproduce your error.  :)
> 
> I'll take a look at what anaconda is doing... perhaps there is a problem
> there.
> 
> Also, I'm not surprised that the boot floppy creation failed - kernel +
> xfs + scsi driver is getting pretty big...  I guess we should disable
> that or at least offer a warning if we can't make it fit.
> 
> -Eric
> 
>  
> 
> 
> 


From owner-linux-xfs@oss.sgi.com Fri Mar 23 18:27:09 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2O2R9M17070
	for linux-xfs-outgoing; Fri, 23 Mar 2001 18:27:09 -0800
Received: from home.smithconcepts.com (ubr-35.28.151.oviedo.cfl.rr.com [65.35.28.151])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2O2R8M17067
	for <linux-xfs@oss.sgi.com>; Fri, 23 Mar 2001 18:27:08 -0800
Received: from ieee.org (IDENT:bjsmith@bitman.oviedo.smithconcepts.com [172.24.24.192])
	by home.smithconcepts.com (8.9.3/8.9.3) with ESMTP id VAA09699;
	Fri, 23 Mar 2001 21:20:42 -0500
Message-ID: <3ABC0808.805BAE27@ieee.org>
Date: Fri, 23 Mar 2001 21:35:52 -0500
From: "Bryan J. Smith" <b.j.smith@ieee.org>
Reply-To: b.j.smith@ieee.org, thebs@theseus.com
Organization: SmithConcepts/Personal
X-Mailer: Mozilla 4.76 [en] (X11; U; Linux 2.2.18-0.13 i686)
X-Accept-Language: en
MIME-Version: 1.0
To: thebs@theseus.com
CC: Eric Sandeen <sandeen@sgi.com>, K Mitchell Russell <kmrussel@hsc.vcu.edu>,
   linux-xfs@oss.sgi.com
Subject: Re: Unable to mount SCSI root fs with pr0.10test1 or test2  PAM/NIS 
 problems fixed
References: <Pine.A41.4.10.10103232007550.41756-100000@comet.vcu.edu> <985397254.11121.2.camel@Porter> <3ABBFBEA.23BA0D17@ieee.org>
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

"Bryan J. Smith" wrote:
> I'm seeming to have a serious issue with PAM and NIS -- which is a
> Wolverine issue.  Anyone else here having issues?  The second I try
> to bind to my NIS server, _all_ authentication goes down (even
> local).  I have to shutdown ypbind completely (even if it didn't
> bind -- which it is not anyway) before any authentication is
> possible.
> Time to check Bugzilla I guess ...

FYI, I just pulled the latest PAM and YP* SRPMs from Rawhide, built
and installed them and my NIS problem was fixed.  Even an unbound
domain will now authenticate local users correctly.

-- TheBS

P.S.  FYI, I'm getting header issues with the Athlon target kernel
build.  That'll teach me to announce something before it's done.

-- 
Bryan "TheBS" Smith         chat:thebs413 @AOL/MSN/Yahoo
Engineer     mailto:b.j.smith@ieee.org,thebs@theseus.com
********************************************************
"Linux will do for applications what the Internet did to
 networks" -- Sam Palmisano, IBM Chief Operating Officer

From owner-linux-xfs@oss.sgi.com Fri Mar 23 18:31:59 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2O2VxO17251
	for linux-xfs-outgoing; Fri, 23 Mar 2001 18:31:59 -0800
Received: from mail.get2chip.com ([64.169.83.2])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2O2VwM17244
	for <linux-xfs@oss.sgi.com>; Fri, 23 Mar 2001 18:31:58 -0800
Message-ID: <3ABC05FA.7030605@get2chip.com>
Date: Fri, 23 Mar 2001 18:27:06 -0800
From: Chris Croswhite <ccroswhite@get2chip.com>
MIME-Version: 1.0
CC: Eric Sandeen <sandeen@sgi.com>, K Mitchell Russell <kmrussel@hsc.vcu.edu>,
   linux-xfs@oss.sgi.com
Subject: Re: Unable to mount SCSI root fs with pr0.10test1 or test2
References: <Pine.A41.4.10.10103232007550.41756-100000@comet.vcu.edu> <985397254.11121.2.camel@Porter> <3ABBF990.3060201@get2chip.com>
Content-Type: text/plain; charset=us-ascii; format=flowed
Content-Transfer-Encoding: 7bit
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk


I just ran a clean Wolverine install on an Adaptec equiped machine and 
ran into NO PROBLEMS.  This would lead me to believe that the XFS 
install has in someway changed what anaconda is doing or this has 
something to do with DFS?  I have not a clue.  Is there anyone who can 
clear this up?

TIA,
Chris


Chris Croswhite wrote:

> I have tried the 10test on another machine with scsi drive.  Again, I 
> get the same panic of can not open root device.  Is there a fix for this?
> 
> Chris
> 
> Eric Sandeen wrote:
> 
>> On 23 Mar 2001 20:14:16 -0500, K Mitchell Russell wrote:
>> 
>>> Interestingly, i reinstalled pr0.9, booted up fine.  Then upgraded
>>> mkinitrd to 3.0.6, e2fsprogs to 1.19, and installed the
>>> kernel-smp-2.4.2-SGI_XFS_0.10.i686.rpm from the pr0.10test2.  Then
>>> created a new initrd from the 2.4.2-SGI_XFS_0.10smp modules, and this
>>> boots fine!  Perhaps there is a problem with the Wolverine installer
>>> creating the initrd...
>> 
>> 
>> 
>> Hm.... well, I did a full scsi install here on an adaptec controller
>> (aic7xxx driver) and it's a good news / bad news sort of thing - the
>> good news is that it all went fine, and booted right up.  The bad news
>> is I didn't reproduce your error.  :)
>> 
>> I'll take a look at what anaconda is doing... perhaps there is a problem
>> there.
>> 
>> Also, I'm not surprised that the boot floppy creation failed - kernel +
>> xfs + scsi driver is getting pretty big...  I guess we should disable
>> that or at least offer a warning if we can't make it fit.
>> 
>> -Eric
>> 
>>  
>> 
>> 
>> 
> 
> 


From owner-linux-xfs@oss.sgi.com Sat Mar 24 03:14:10 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2OBEAV24678
	for linux-xfs-outgoing; Sat, 24 Mar 2001 03:14:10 -0800
Received: from kamov.deltanet.ro (kamov.deltanet.ro [193.226.175.59])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2OBE7M24671
	for <linux-xfs@oss.sgi.com>; Sat, 24 Mar 2001 03:14:08 -0800
Received: from shiva.ppetru.net (home.ppetru.net [193.230.129.57])
	by kamov.deltanet.ro (Postfix) with ESMTP id 13BEBF7CB
	for <linux-xfs@oss.sgi.com>; Sat, 24 Mar 2001 13:14:03 +0200 (EET)
Received: by shiva.ppetru.net (Postfix, from userid 1000)
	id 90A6E40B15; Sat, 24 Mar 2001 13:13:36 +0200 (EET)
Date: Sat, 24 Mar 2001 13:13:36 +0200
To: linux-xfs@oss.sgi.com
Subject: Effect of kiobufs?
Message-ID: <20010324131336.A3369@ppetru.net>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.3.15i
From: ppetru@ppetru.net (Petru Paler)
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

Hi,

What are the gains of using the "kio" mount options? Are
kiobufs faster than buffer heads, or they just allow more
efficient merging of requests?

Thanks,

--
Petru Paler, mailto:ppetru@ppetru.net
http://www.ppetru.net - ICQ: 41817235

From owner-linux-xfs@oss.sgi.com Sat Mar 24 03:31:48 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2OBVmi24832
	for linux-xfs-outgoing; Sat, 24 Mar 2001 03:31:48 -0800
Received: from virtualhost.dk (ns.virtualhost.dk [195.184.98.160])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2OBVlM24829
	for <linux-xfs@oss.sgi.com>; Sat, 24 Mar 2001 03:31:47 -0800
Received: from burns.home.kernel.dk ([192.168.0.2] ident=root)
	by virtualhost.dk with esmtp (Exim 3.22 #4)
	id 14gmGL-0004RL-00; Sat, 24 Mar 2001 12:31:17 +0100
Received: from axboe by burns.home.kernel.dk with local (Exim 3.13 #1 (Debian))
	id 14gmGM-00009Z-00; Sat, 24 Mar 2001 12:31:18 +0100
Date: Sat, 24 Mar 2001 12:31:18 +0100
From: Jens Axboe <axboe@suse.de>
To: Petru Paler <ppetru@ppetru.net>
Cc: linux-xfs@oss.sgi.com
Subject: Re: Effect of kiobufs?
Message-ID: <20010324123118.C418@suse.de>
References: <20010324131336.A3369@ppetru.net>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <20010324131336.A3369@ppetru.net>; from ppetru@ppetru.net on Sat, Mar 24, 2001 at 01:13:36PM +0200
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

On Sat, Mar 24 2001, Petru Paler wrote:
> Hi,
> 
> What are the gains of using the "kio" mount options? Are
> kiobufs faster than buffer heads, or they just allow more
> efficient merging of requests?

The kiobufs allow efficient "merging" from the fs side, so that
it can submit single large chunks of I/O. The gains can be pretty
good, I know Steve Lord had some numbers on IDE and SCSI bh vs kiobuf
that showed this nicely.

-- 
Jens Axboe


From owner-linux-xfs@oss.sgi.com Sat Mar 24 11:00:20 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2OJ0KW00974
	for linux-xfs-outgoing; Sat, 24 Mar 2001 11:00:20 -0800
Received: from arrakis.vcu.edu (arrakis.vcu.edu [128.172.1.132])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2OJ0DM00971
	for <linux-xfs@oss.sgi.com>; Sat, 24 Mar 2001 11:00:19 -0800
Received: from 0 (arrakis.vcu.edu [127.0.0.1])
	by arrakis.vcu.edu (8.9.3/8.9.3) with ESMTP id OAA16477
	for <linux-xfs@oss.sgi.com>; Sat, 24 Mar 2001 14:00:10 -0500
Message-Id: <200103241900.OAA16477@arrakis.vcu.edu>
Date: Sat, 24 Mar 2001 14:00:10 -500
Content-Transfer-Encoding: 8bit
To: linux-xfs@oss.sgi.com
From: "K. Mitchell Russell" <kmrussel@hsc.vcu.edu>
Content-Type: text/plain; charset=iso-8859-1
Subject: Re: Unable to mount SCSI root fs with pr0.10test1 or test2
MIME-Version: 1.0
User-Agent: IMHO/0.98 (Webmail for Roxen)
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

I have tried the same thing Chris, installing the base Wolverine, and
it boots fine also.  But still no luck with the XFS installer booting
either SMP or UP.  I'm glad I'm not the only one, or I'd doubt my
sanity.

Thanks for the info Eric.  I don't know what kind of magic words you
used at boottime, but I'm still reproducing the problem on my
hardware.  Are you using any bootprompt commands?  I'll keep trying,
but since I can't even get the system up after running the XFS
installer, I can't very well look at the results of the installer to
determine the problem.  I do know that the kernel panic is at the
point where the VFS is loading root from the initrd, after the ramdisk
is loaded, and before the scsi modules are loaded, not the point of
where it's loading the XFS scsi disk root.

Thanks,
Mitchell

-------------------
> 
> I just ran a clean Wolverine install on an Adaptec equiped machine
and 
> ran into NO PROBLEMS.  This would lead me to believe that the XFS 
> install has in someway changed what anaconda is doing or this has 
> something to do with DFS?  I have not a clue.  Is there anyone who
can 
> clear this up?
> 
> TIA,
> Chris
> 
> 
> Chris Croswhite wrote:
> 
> > I have tried the 10test on another machine with scsi drive. 
Again, I 
> > get the same panic of can not open root device.  Is there a fix
for this?
> > 
> > Chris
> > 
> > Eric Sandeen wrote:
> > 
> >> On 23 Mar 2001 20:14:16 -0500, K Mitchell Russell wrote:
> >> 
> >>> Interestingly, i reinstalled pr0.9, booted up fine.  Then
upgraded
> >>> mkinitrd to 3.0.6, e2fsprogs to 1.19, and installed the
> >>> kernel-smp-2.4.2-SGI_XFS_0.10.i686.rpm from the pr0.10test2. 
Then
> >>> created a new initrd from the 2.4.2-SGI_XFS_0.10smp modules, and
this
> >>> boots fine!  Perhaps there is a problem with the Wolverine
installer
> >>> creating the initrd...
> >> 
> >> 
> >> 
> >> Hm.... well, I did a full scsi install here on an adaptec
controller
> >> (aic7xxx driver) and it's a good news / bad news sort of thing -
the
> >> good news is that it all went fine, and booted right up.  The bad
news
> >> is I didn't reproduce your error.  :)
> >> 
> >> I'll take a look at what anaconda is doing... perhaps there is a
problem
> >> there.
> >> 
> >> Also, I'm not surprised that the boot floppy creation failed -
kernel +
> >> xfs + scsi driver is getting pretty big...  I guess we should
disable
> >> that or at least offer a warning if we can't make it fit.
> >> 
> >> -Eric
> >> 
> >>  
> >> 
> >> 
> >> 
> > 
> > 
> 
> 


From owner-linux-xfs@oss.sgi.com Sat Mar 24 11:10:27 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2OJARv01162
	for linux-xfs-outgoing; Sat, 24 Mar 2001 11:10:27 -0800
Received: from mail11.jump.net (mail11.jump.net [206.196.91.11])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2OJAQM01159
	for <linux-xfs@oss.sgi.com>; Sat, 24 Mar 2001 11:10:26 -0800
Received: from Porter (aus-dsl-dhcp1-26.customer.jump.net [63.163.168.26])
	by mail11.jump.net (8.10.2/) with ESMTP id f2OJAOv29683;
	Sat, 24 Mar 2001 13:10:24 -0600 (CST)
Subject: Re: Unable to mount SCSI root fs with pr0.10test1 or test2
From: Eric Sandeen <sandeen@sgi.com>
To: "K." Mitchell Russell <kmrussel@hsc.vcu.edu>
Cc: linux-xfs@oss.sgi.com
In-Reply-To: <200103241900.OAA16477@arrakis.vcu.edu>
References: <200103241900.OAA16477@arrakis.vcu.edu>
Content-Type: text/plain
X-Mailer: Evolution/0.9+cvs.2001.03.23.09.06 (Preview Release)
Date: 24 Mar 2001 13:11:32 -0600
Message-Id: <985461092.12058.5.camel@Porter>
Mime-Version: 1.0
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

On 24 Mar 2001 14:00:10 -0500, K. Mitchell Russell wrote:
> I have tried the same thing Chris, installing the base Wolverine, and
> it boots fine also.  But still no luck with the XFS installer booting
> either SMP or UP.  I'm glad I'm not the only one, or I'd doubt my
> sanity.

Just talked to someone else with the same problem, and it looks like the
installer did not create an initial ramdisk in his case - can you guys
run "linux rescue" off the SGI boot CD and poke around to see if the
ramdisk image exists in /boot, and whether it's called out on the
initrd= line in /etc/lilo.conf?

If that turns out to be the problem, instructions for re-creating the
initrd image are at
http://linux-xfs.sgi.com/projects/xfs/installcavs.html

 
> Thanks for the info Eric.  I don't know what kind of magic words you
> used at boottime, but I'm still reproducing the problem on my
> hardware.  

I just realized that although I did the install on SCSI, the actual boot
device is still /dev/hda1 (i.e. lilo is installed on /dev/hda1) so
that's kind of cheating.  I'll turn off the IDE controllers and re-try
it.

> Are you using any bootprompt commands?  I'll keep trying,
> but since I can't even get the system up after running the XFS
> installer, I can't very well look at the results of the installer to
> determine the problem.  

You can try running "linux rescue" off the CD to poke around your system
a bit...

> I do know that the kernel panic is at the
> point where the VFS is loading root from the initrd, after the ramdisk
> is loaded, and before the scsi modules are loaded, not the point of
> where it's loading the XFS scsi disk root.

Hm, then perhaps the initrd image _is_ getting created... maybe you can
verify this with "linux rescue"

Thanks for looking into this!

-Eric


From owner-linux-xfs@oss.sgi.com Sat Mar 24 15:31:38 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2ONVcB04982
	for linux-xfs-outgoing; Sat, 24 Mar 2001 15:31:38 -0800
Received: from dragon.vcu.edu (dragon.vcu.edu [128.172.65.9])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2ONVbM04979
	for <linux-xfs@oss.sgi.com>; Sat, 24 Mar 2001 15:31:37 -0800
Received: from comet.vcu.edu (comet.vcu.edu [128.172.1.33])
	by dragon.vcu.edu (8.9.3/8.9.3) with ESMTP id SAA3054544;
	Sat, 24 Mar 2001 18:33:58 -0500 (EST)
Date: Sat, 24 Mar 2001 18:31:17 -0500 (EST)
From: K Mitchell Russell <kmrussel@hsc.vcu.edu>
X-Sender: kmrussel@comet.vcu.edu
To: Eric Sandeen <sandeen@sgi.com>
cc: linux-xfs@oss.sgi.com
Subject: Re: Unable to mount SCSI root fs with pr0.10test1 or test2
In-Reply-To: <985461092.12058.5.camel@Porter>
Message-ID: <Pine.A41.4.10.10103241816420.35546-100000@comet.vcu.edu>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

On 24 Mar 2001, Eric Sandeen wrote:

> On 24 Mar 2001 14:00:10 -0500, K. Mitchell Russell wrote:
> > I have tried the same thing Chris, installing the base Wolverine, and
> > it boots fine also.  But still no luck with the XFS installer booting
> > either SMP or UP.  I'm glad I'm not the only one, or I'd doubt my
> > sanity.
> 
> Just talked to someone else with the same problem, and it looks like the
> installer did not create an initial ramdisk in his case - can you guys
> run "linux rescue" off the SGI boot CD and poke around to see if the
> ramdisk image exists in /boot, and whether it's called out on the
> initrd= line in /etc/lilo.conf?
> 
> If that turns out to be the problem, instructions for re-creating the
> initrd image are at
> http://linux-xfs.sgi.com/projects/xfs/installcavs.html

Brilliant!  I booted in 'linux rescue' mode off the SGI boot CD, created
nodes for the SCSI partitions (using block major 8), and chroot'd to my
installation. Indeed, there are NO initrd's on my system whatsoever, and
no entries for them in /etc/lilo.conf.  I created the initrd's and
booted perfectly into my new installation.

> 
>  
> > Thanks for the info Eric.  I don't know what kind of magic words you
> > used at boottime, but I'm still reproducing the problem on my
> > hardware.  
> 
> I just realized that although I did the install on SCSI, the actual boot
> device is still /dev/hda1 (i.e. lilo is installed on /dev/hda1) so
> that's kind of cheating.  I'll turn off the IDE controllers and re-try
> it.
> 
> > Are you using any bootprompt commands?  I'll keep trying,
> > but since I can't even get the system up after running the XFS
> > installer, I can't very well look at the results of the installer to
> > determine the problem.  
> 
> You can try running "linux rescue" off the CD to poke around your system
> a bit...

Never had the occasion to use it until now.  Learn something new every
day :)

> 
> > I do know that the kernel panic is at the
> > point where the VFS is loading root from the initrd, after the ramdisk
> > is loaded, and before the scsi modules are loaded, not the point of
> > where it's loading the XFS scsi disk root.
> 
> Hm, then perhaps the initrd image _is_ getting created... maybe you can
> verify this with "linux rescue"

Indeed they are not being created.  I'm a relative newbie, and not
familiar enough with the anaconda to see why, but I did notice at the
end of the installation, it reported problems unmounting sysimage,
sysimage/usr, and tmp/loop0 (failed: 16).  Is loop0 being used to create
the initrd perhaps?

Thanks for the help,
Mitchell

K. Mitchell Russell, M.D.
kmrussel@hsc.vcu.edu
MedITAC Research Lab
www.meditac.com


From owner-linux-xfs@oss.sgi.com Sun Mar 25 08:55:09 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2PGt9520269
	for linux-xfs-outgoing; Sun, 25 Mar 2001 08:55:09 -0800
Received: from dittos.yi.org (mkc-65-28-13-11.kc.rr.com [65.28.13.11])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2PGshM20253
	for <linux-xfs@oss.sgi.com>; Sun, 25 Mar 2001 08:54:58 -0800
Received: by dittos.yi.org (Postfix, from userid 500)
	id 492F31C00A3D; Sun, 25 Mar 2001 10:54:37 -0600 (CST)
Subject: Re: Effect of kiobufs?
To: ppetru@ppetru.net (Petru Paler)
Date: Sun, 25 Mar 2001 10:54:37 -0600 (CST)
Cc: linux-xfs@oss.sgi.com
In-Reply-To: <20010324131336.A3369@ppetru.net> from "Petru Paler" at Mar 24, 2001 01:13:36 PM
X-Mailer: ELM [version 2.5 PL3]
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Message-Id: <20010325165437.492F31C00A3D@dittos.yi.org>
From: dbl@dittos.yi.org (Dave Lounsberry)
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk


Here is what I am seeing on the following system:

	Athlon 1GHZ
	Ultra2 LVD SCSI 10K RPM disk
	256MB Memory

To be sure, I ran it several times. I used a package called bonnie to get these numbers. 

- XFS mounted with default mount options
              ---------Sequential Output------------ 
              -Per Char-    --Block---    -Rewrite-- 
              K/sec %CPU    K/sec %CPU    K/sec %CPU 
               6736 84.7    43180 24.9     9902  8.5 

               -----Sequential Input---     --Random--
               -Per Char-    --Block---     --Seeks---
               K/sec %CPU    K/sec %CPU      /sec %CPU
                6284 76.0    57735 26.5     871.6  3.1

- XFS mounted with kio mount option
              -------Sequential Output--------------- 
              -Per Char-     --Block---    -Rewrite-- 
              K/sec %CPU     K/sec %CPU   K/sec %CPU 
               7454 92.0    143541 81.3   101492 79.3

               -----Sequential Input---   --Random--
              -Per Char-    --Block---    --Seeks---
              K/sec %CPU    K/sec %CPU     /sec %CPU
               7785 93.6    262545 92.3    24547.3 98.2

The block level I/O is amazing and actually beyond the capabilities of Ultra2 SCSI. 
Are these numbers for real????? Ultra2 LVD runs at 80MB/s not 260MB/s. 

Dave Lounsberry
dbl@dittos.yi.org

> 
> Hi,
> 
> What are the gains of using the "kio" mount options? Are
> kiobufs faster than buffer heads, or they just allow more
> efficient merging of requests?
> 
> Thanks,
> 
> --
> Petru Paler, mailto:ppetru@ppetru.net
> http://www.ppetru.net - ICQ: 41817235
> 


From owner-linux-xfs@oss.sgi.com Sun Mar 25 11:52:50 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2PJqo123076
	for linux-xfs-outgoing; Sun, 25 Mar 2001 11:52:50 -0800
Received: from antares.cedar.buffalo.edu (antares.cedar.Buffalo.EDU [128.205.33.2])
	by oss.sgi.com (8.11.3/8.11.3) with SMTP id f2PJqnM23073
	for <linux-xfs@oss.sgi.com>; Sun, 25 Mar 2001 11:52:49 -0800
Received: (qmail 9871 invoked from network); 25 Mar 2001 19:52:49 -0000
Received: from zaurak.cedar.buffalo.edu (128.205.33.110)
  by antares.cedar.buffalo.edu with SMTP; 25 Mar 2001 19:52:49 -0000
Received: (from ajay@localhost)
	by zaurak.cedar.buffalo.edu (8.9.3+Sun/8.9.3) id OAA23098
	for linux-xfs@oss.sgi.com; Sun, 25 Mar 2001 14:52:48 -0500 (EST)
Date: Sun, 25 Mar 2001 14:52:48 -0500
From: Ajay Shekhawat <ajay@cedar.Buffalo.EDU>
To: linux-xfs@oss.sgi.com
Subject: LOCKUP with XFS and NFS
Message-ID: <20010325145248.H16131@zaurak.cedar.buffalo.edu>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.2.5i
Organization: Center for Document Analysis and Recognition
X-OfficePhone: +1 (716)-645-6164 ext. 101
X-Fax-Number: +1 (716)-645-6176
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

Background: I'm trying to evaluate the suitability of XFS -vs- ReiserFS for
    an NFS server. I'm leaning towards XFS because we have a few IRIX machines,
    and the admins are familiar with XFS (backup/restore, etc.).
    The "test" server is a dual P-II 450 machine w/ 256MB RAM and a couple 
    of SCSI disks on an Adaptex 2940. One of these disks was formatted as XFS. 
    This disk is NFS exported to 4-5 client machines.  The base installation is
    RedHat Wolverine; we upgraded the kernel to the latest XFS patched kernel,
    0.10-test2 using the kernel source RPM on the OSS ftp site.

Testing:  We are beating on this exported filesystem from 4-5 clients. All
    clients are reading random 3MB files each, in a tight loop. One of the
    clients is writing 0-1MB size files (randomly), again in a tight loop.
    This homespun testing methodology tries to simulate the kind of use the
    real server would expect to get.

Problem:  After about 3-4 hours of continuous beating, the system experienced
    a lockup. The message on the console is duplicated below. Also given
    below is a "function call trace" which I attempted to figure out using
    disassembled output of the kernel, my kernel hacking skills being
    severely lacking  :-)

Question: Is this a knfsd problem, or an XFS problem? It appears like the
    problem is with XFS, but I've been easily mistaken before...

Thanks for any help in this regard,


Ajay


Listing #1: Console message -----------------------------------------------

NMI watchdog detected LOCKUP on CPU0, registers:
CPU: 	0
EIP: 	0010:[<c02b12a8>]
EFLAGS: 00000086
eax: 00000000	ebx: c1453c00 	ecx: cf025cd0 edx: c1453c00
esi: 0000007c   edi: c1447d10   ebp: cf024000 esp: cf025be8
ds: 0018  es: 0018   ss: 0018
Process: nfsd (pid: 645, stackpage=cf025000)
Stack: c1447d10 00000286 00000003 c012a2a3 c1447d10 00000003 00000000 00000000
       00000001 00001000 c1447d78 c0134e34 c1447d10 00000003 00000000 00000001
       c0134f11 00000001 00000000 c12f5e1c 00000811 cf9d69e0 00000000 00000001
Call Trace: [<c012a2a3>] [<c0134e34>] [<c0134f11>] [<c01351e4>] [<c01895cb>] [<c018926d>] [<c012d834>]
       [<c0124a8d>] [<c01f1cd8>] [<c01f1b2c>] [<c012514b>] [<c01253e1>] [<c0125807>] [<c0125750>] [<c01f1fcf>]
       [<c01eeee7>] [<c01eee40>] [<c017316d>] [<c01eee40>] [<c0170335>] [<c016fa33>] [<c02a9a78>] [<c016f7da>]
       [<c0107503>]

Code: 7e f8 e9 6c 8e e7 ff 90 7e 18 00 f3 90 7e f8 e9 d3 8e e7 ff
Console shuts up...




Listing #2: Function call trace (done by hand) --------------------------------

address         in function
-------         ------------------------
c012a258        <kmem_cache_alloc>
c0134df4        <get_unused_buffer_head>
c0134eec        <create_buffers>
c01351cc        <create_empty_buffers>
c0189568        <hook_buffers_to_page>
c0189178        <pagebuf_read_full_page>
c012d6f8        <__alloc_pages>
c01249d4        <add_to_page_cache_unique>
c01f1cc4        <linvfs_read_full_page>
c01f1b2c        <linvfs_pb_bmap>
c0124f28        <generic_file_readahead>
c01251c8        <do_generic_file_read>
c01257a4        <generic_file_read>
c0125750        <file_read_actor>
c01f1d90        <xfs_read>
c01eee40        <linvfs_read>
c01eee40        <linvfs_read>
c0172f40        <nfsd_read>
c01eee40        <linvfs_read>
c0170214        <nfsd_proc_read>
c016f968        <nfsd_dispatch>
c02a97cc        <svc_process>
c016f610        <nfsd>
c01074e0        <kernel_thread>


From owner-linux-xfs@oss.sgi.com Sun Mar 25 13:21:08 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2PLL8h24739
	for linux-xfs-outgoing; Sun, 25 Mar 2001 13:21:08 -0800
Received: from babel.spoiled.org (babel.spoiled.org [212.84.234.227])
	by oss.sgi.com (8.11.3/8.11.3) with SMTP id f2PLL7M24736
	for <linux-xfs@oss.sgi.com>; Sun, 25 Mar 2001 13:21:07 -0800
Received: (qmail 8622 invoked by uid 8); 25 Mar 2001 21:21:05 -0000
From: "thomas graichen" <list-linux.sgi.xfs@spoiled.org>
Reply-To: "thomas graichen" <tgr@spoiled.org>
X-Newsgroups: spoiled.linux.sgi.xfs
Subject: Re: Effect of kiobufs?
Date: Sun, 25 Mar 2001 23:26:27 +0200
Organization: spoiled dot org
Lines: 26
Distribution: local
Message-ID: <news2mail-99lng0$69n$2@babel.spoiled.org>
References: <20010324131336.A3369@ppetru.net> <20010325165437.492F31C00A3D@dittos.yi.org>
X-Complaints-To: newsmaster@spoiled.org
User-Agent: Pan/0.9.1 (Unix)
X-No-Productlinks: Yes
To: linux-xfs@oss.sgi.com
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

always make shure to at least use a testfilesize (-s option
of bonnie) twice as big as your main memory to avoid cache
effects

t

In article <20010325165437.492F31C00A3D@dittos.yi.org>,
dbl@dittos.yi.org  wrote:

> 
> Here is what I am seeing on the following system:
> 
> 	Athlon 1GHZ Ultra2 LVD SCSI 10K RPM disk
> 	256MB Memory
> 
> To be sure, I ran it several times. I used a package called bonnie to
> get these numbers. 
> 
> - XFS mounted with default mount options
>               ---------Sequential Output----------


-- 
thomas graichen <tgr@spoiled.org> ... perfection is reached, not
when there is no longer anything to add, but when there is no
longer anything to take away. --- antoine de saint-exupery

From owner-linux-xfs@oss.sgi.com Sun Mar 25 13:55:20 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2PLtKw25850
	for linux-xfs-outgoing; Sun, 25 Mar 2001 13:55:20 -0800
Received: from mail.ocs.com.au (ppp0.ocs.com.au [203.34.97.3])
	by oss.sgi.com (8.11.3/8.11.3) with SMTP id f2PLtGM25847
	for <linux-xfs@oss.sgi.com>; Sun, 25 Mar 2001 13:55:17 -0800
Received: (qmail 25023 invoked from network); 25 Mar 2001 21:55:13 -0000
Received: from ocs3.ocs-net (192.168.255.3)
  by mail.ocs.com.au with SMTP; 25 Mar 2001 21:55:13 -0000
X-Mailer: exmh version 2.1.1 10/15/1999
From: Keith Owens <kaos@melbourne.sgi.com>
To: Ajay Shekhawat <ajay@cedar.Buffalo.EDU>
cc: linux-xfs@oss.sgi.com
Subject: Re: LOCKUP with XFS and NFS 
In-reply-to: Your message of "Sun, 25 Mar 2001 14:52:48 EST."
             <20010325145248.H16131@zaurak.cedar.buffalo.edu> 
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Date: Mon, 26 Mar 2001 07:55:13 +1000
Message-ID: <5872.985557313@ocs3.ocs-net>
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

On Sun, 25 Mar 2001 14:52:48 -0500, 
Ajay Shekhawat <ajay@cedar.Buffalo.EDU> wrote:
>NMI watchdog detected LOCKUP on CPU0, registers:
>Listing #2: Function call trace (done by hand) --------------------------------

Any reason you did the conversion by hand instead of using ksymoops
which is designed to automatically decode these oops reports?


From owner-linux-xfs@oss.sgi.com Sun Mar 25 14:07:51 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2PM7pP26009
	for linux-xfs-outgoing; Sun, 25 Mar 2001 14:07:51 -0800
Received: from antares.cedar.buffalo.edu (antares.cedar.Buffalo.EDU [128.205.33.2])
	by oss.sgi.com (8.11.3/8.11.3) with SMTP id f2PM7oM26006
	for <linux-xfs@oss.sgi.com>; Sun, 25 Mar 2001 14:07:50 -0800
Received: (qmail 10297 invoked from network); 25 Mar 2001 22:07:50 -0000
Received: from zaurak.cedar.buffalo.edu (128.205.33.110)
  by antares.cedar.buffalo.edu with SMTP; 25 Mar 2001 22:07:50 -0000
Received: (from ajay@localhost)
	by zaurak.cedar.buffalo.edu (8.9.3+Sun/8.9.3) id RAA23128
	for linux-xfs@oss.sgi.com; Sun, 25 Mar 2001 17:07:49 -0500 (EST)
Date: Sun, 25 Mar 2001 17:07:49 -0500
From: Ajay Shekhawat <ajay@cedar.Buffalo.EDU>
To: linux-xfs@oss.sgi.com
Subject: Re: LOCKUP with XFS and NFS
Message-ID: <20010325170749.K16131@zaurak.cedar.buffalo.edu>
References: <20010325145248.H16131@zaurak.cedar.buffalo.edu> <5872.985557313@ocs3.ocs-net>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.2.5i
In-Reply-To: <5872.985557313@ocs3.ocs-net>; from kaos@melbourne.sgi.com on Mon, Mar 26, 2001 at 07:55:13AM +1000
Organization: Center for Document Analysis and Recognition
X-OfficePhone: +1 (716)-645-6164 ext. 101
X-Fax-Number: +1 (716)-645-6176
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

On Mon, Mar 26, 2001 at 07:55:13AM +1000, Keith Owens wrote:
> Any reason you did the conversion by hand instead of using ksymoops
> which is designed to automatically decode these oops reports?

My bad. I didn't know I could feed the stuff that I typed into ksymoops  :-)

Here's what ksymoops says:

------------------------------------------------------------------------------
ksymoops 2.4.0 on i686 2.4.2-SGI_XFS_0.10.  Options used
     -V (specified)
     -k /proc/ksyms (specified)
     -l /proc/modules (specified)
     -o /lib/modules/2.4.2-SGI_XFS_0.10/ (specified)
     -m /boot/System.map-2.4.2-SGI_XFS_0.10 (specified)

NMI watchdog detected LOCKUP on CPU0, registers:
CPU:    0
EIP:    0010:[<c02b12a8>]
Using defaults from ksymoops -t elf32-i386 -a i386
EFLAGS: 00000086
eax: 00000000   ebx: c1453c00     ecx: cf025cd0 edx: c1453c00
esi: 0000007c   edi: c1447d10   ebp: cf024000 esp: cf025be8
ds: 0018  es: 0018   ss: 0018
Stack: c1447d10 00000286 00000003 c012a2a3 c1447d10 00000003 00000000 00000000
       00000001 00001000 c1447d78 c0134e34 c1447d10 00000003 00000000 00000001
       c0134f11 00000001 00000000 c12f5e1c 00000811 cf9d69e0 00000000 00000001
Call Trace: [<c012a2a3>] [<c0134e34>] [<c0134f11>] [<c01351e4>] [<c01895cb>] [<c018926d>] [<c012d834>]
       [<c0124a8d>] [<c01f1cd8>] [<c01f1b2c>] [<c012514b>] [<c01253e1>] [<c0125807>] [<c0125750>] [<c01f1fcf>]
       [<c01eeee7>] [<c01eee40>] [<c017316d>] [<c01eee40>] [<c0170335>] [<c016fa33>] [<c02a9a78>] [<c016f7da>]
       [<c0107503>]
Code: 7e f8 e9 6c 8e e7 ff 90 7e 18 00 f3 90 7e f8 e9 d3 8e e7 ff

>>EIP; c02b12a8 <stext_lock+14e4/7468>   <=====
Trace; c012a2a3 <kmem_cache_alloc+4b/bc>
Trace; c0134e34 <get_unused_buffer_head+40/b8>
Trace; c0134f11 <create_buffers+25/1d0>
Trace; c01351e4 <create_empty_buffers+18/74>
Trace; c01895cb <hook_buffers_to_page+63/e8>
Trace; c018926d <pagebuf_read_full_page+f5/21c>
Trace; c012d834 <__alloc_pages+13c/2ec>
Trace; c0124a8d <add_to_page_cache_unique+b9/cc>
Trace; c01f1cd8 <linvfs_read_full_page+14/18>
Trace; c01f1b2c <linvfs_pb_bmap+0/88>
Trace; c012514b <generic_file_readahead+223/2a0>
Trace; c01253e1 <do_generic_file_read+219/588>
Trace; c0125807 <generic_file_read+63/80>
Trace; c0125750 <file_read_actor+0/54>
Trace; c01f1fcf <xfs_read+23f/270>
Trace; c01eeee7 <linvfs_read+a7/c4>
Trace; c01eee40 <linvfs_read+0/c4>
Trace; c017316d <nfsd_read+22d/2d8>
Trace; c01eee40 <linvfs_read+0/c4>
Trace; c0170335 <nfsd_proc_read+121/12c>
Trace; c016fa33 <nfsd_dispatch+cb/168>
Trace; c02a9a78 <svc_process+2ac/544>
Trace; c016f7da <nfsd+1ca/358>
Trace; c0107503 <kernel_thread+23/30>
Code;  c02b12a8 <stext_lock+14e4/7468>
00000000 <_EIP>:
Code;  c02b12a8 <stext_lock+14e4/7468>   <=====
   0:   7e f8                     jle    fffffffa <_EIP+0xfffffffa> c02b12a2 <stext_lock+14de/7468>   <=====
Code;  c02b12aa <stext_lock+14e6/7468>
   2:   e9 6c 8e e7 ff            jmp    ffe78e73 <_EIP+0xffe78e73> c012a11b <kmem_cache_alloc_batch+1b/80>
Code;  c02b12af <stext_lock+14eb/7468>
   7:   90                        nop    
Code;  c02b12b0 <stext_lock+14ec/7468>
   8:   7e 18                     jle    22 <_EIP+0x22> c02b12ca <stext_lock+1506/7468>
Code;  c02b12b2 <stext_lock+14ee/7468>
   a:   00 f3                     add    %dh,%bl
Code;  c02b12b4 <stext_lock+14f0/7468>
   c:   90                        nop    
Code;  c02b12b5 <stext_lock+14f1/7468>
   d:   7e f8                     jle    7 <_EIP+0x7> c02b12af <stext_lock+14eb/7468>
Code;  c02b12b7 <stext_lock+14f3/7468>
   f:   e9 d3 8e e7 ff            jmp    ffe78ee7 <_EIP+0xffe78ee7> c012a18f <free_block+f/d8>


From owner-linux-xfs@oss.sgi.com Sun Mar 25 15:34:21 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2PNYLj28399
	for linux-xfs-outgoing; Sun, 25 Mar 2001 15:34:21 -0800
Received: from sgi.com (sgi.SGI.COM [192.48.153.1])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2PNYJM28396
	for <linux-xfs@oss.sgi.com>; Sun, 25 Mar 2001 15:34:19 -0800
Received: from larry.melbourne.sgi.com ([134.14.52.130]) 
	by sgi.com (980327.SGI.8.8.8-aspam/980304.SGI-aspam:
       SGI does not authorize the use of its proprietary
       systems or networks for unsolicited or bulk email
       from the Internet.) 
	via SMTP id PAA09810
	for <linux-xfs@oss.sgi.com>; Sun, 25 Mar 2001 15:34:12 -0800 (PST)
	mail_from (nathans@wobbly.melbourne.sgi.com)
Received: from wobbly.melbourne.sgi.com (wobbly.melbourne.sgi.com [134.14.55.135]) by larry.melbourne.sgi.com (950413.SGI.8.6.12/950213.SGI.AUTOCF) via ESMTP id JAA13122; Mon, 26 Mar 2001 09:32:50 +1000
Received: (from nathans@localhost) by wobbly.melbourne.sgi.com (980427.SGI.8.8.8/980728.SGI.AUTOCF) id JAA53857; Mon, 26 Mar 2001 09:32:48 +1000 (EST)
From: "Nathan Scott" <nathans@wobbly.melbourne.sgi.com>
Message-Id: <10103260932.ZM235393@wobbly.melbourne.sgi.com>
Date: Mon, 26 Mar 2001 09:32:46 -0500
In-Reply-To: "Bas" <list@showme.wox.org>
        "2.4.2 - XFS kernel March 19th & quota" (Mar 23, 10:31pm)
References: <006401c0b42c$072faea0$0f01a8c0@myplace.net>
X-Mailer: Z-Mail (3.2.3 08feb96 MediaMail)
To: "Bas" <list@showme.wox.org>, <linux-xfs@oss.sgi.com>
Subject: Re: 2.4.2 - XFS kernel March 19th & quota
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

hi,

On Mar 23, 10:31pm, Bas wrote:
> Subject: 2.4.2 - XFS kernel March 19th & quota
> Hi,
> 
> I'm running XFS with quota support and it works, but I can't seem to find
> (checked the man pages and the documentation that comes with the cvs-tree)
> the block size that is used when setting quota. I thought is 1024 B now, but
> am I wrong ?
> 

That's correct - the user tools work with a blocksize of
1024 bytes.  The quotactl system call, once it gets into
the XFS quota manager code, manipulates 512 byte blocks,
so there are conversions done in the user tools.  I have
sometimes found it useful to use the "quot" tool to cross
reference the block counts reported by the tools - this
tool doesn't use the quotactl syscall but instead walks
the filesystem and totals user block counts, then reports
in 1024 byte units also.

There was also a recent change (couple of weeks ago) to the
kernel and userspace tools in this regard, so if you're
seeing unexpected block count values, its likely that your
tools are out of sync with your kernel.

Before you grab new tools, wait for my next message (later
today) on the new version of the quota tools.  These will
include the "quot" utility I referred to above.

cheers.

-- 
Nathan

From owner-linux-xfs@oss.sgi.com Sun Mar 25 16:20:41 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2Q0KfI28918
	for linux-xfs-outgoing; Sun, 25 Mar 2001 16:20:41 -0800
Received: from roujin.gargoylecc.com ([65.100.85.34])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2Q0KeM28915
	for <linux-xfs@oss.sgi.com>; Sun, 25 Mar 2001 16:20:40 -0800
Received: from lists.gargoylecc.com (IDENT:ringram@gargoyle.gargoylecc.com [65.100.85.35])
	by roujin.gargoylecc.com (8.9.3/8.9.3) with ESMTP id IAA03198
	for <linux-xfs@oss.sgi.com>; Mon, 26 Mar 2001 08:34:15 -0700
Date: Sun, 25 Mar 2001 17:20:43 -0700 (MST)
From: Russel Ingram <ringram@gargoylecc.com>
X-Sender:  <ringram@localhost.localdomain>
To: <linux-xfs@oss.sgi.com>
Subject: kinda off topic
Message-ID: <Pine.LNX.4.30.0103251714450.20493-100000@localhost.localdomain>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

This is kinda off topic, but...

Is there some way we can get an rpm with the correct setup in
/etc/security/console.perms so a normal user can use X?  I don't even know
what rpm that file is part of but its a simple enough thing that it seems
to me it would be an easy fix to make and include on the sgi iso as a
replacement for the RH rpm.

Russ
-- 
---------------------------------------------------------------
"Bill Gates and Microsoft have ruined the computer industry for
 a long time to come by creating a class of ignorant and lazy
 computer users."  --Russel Ingram

"Mommy ... can I go out and ... KILL TONIGHT!?"
                   --Glen  Danzig, The Misfits
---------------------------------------------------------------
Russ Ingram
Gargoyle Computer Consulting
(307)742-1361
www.gargoylecc.com


From owner-linux-xfs@oss.sgi.com Sun Mar 25 16:27:57 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2Q0Rvq29010
	for linux-xfs-outgoing; Sun, 25 Mar 2001 16:27:57 -0800
Received: from sgi.com (sgi.SGI.COM [192.48.153.1])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2Q0RuM29007
	for <linux-xfs@oss.sgi.com>; Sun, 25 Mar 2001 16:27:56 -0800
Received: from cthulhu.engr.sgi.com (cthulhu.engr.sgi.com [192.26.80.2]) 
	by sgi.com (980327.SGI.8.8.8-aspam/980304.SGI-aspam:
       SGI does not authorize the use of its proprietary
       systems or networks for unsolicited or bulk email
       from the Internet.) 
	via ESMTP id QAA08940
	for <linux-xfs@oss.sgi.com>; Sun, 25 Mar 2001 16:27:56 -0800 (PST)
	mail_from (tduffy@engr.sgi.com)
Received: from dbear.engr.sgi.com (dbear.engr.sgi.com [163.154.18.85])
	by cthulhu.engr.sgi.com (SGI-8.9.3/8.9.3) with ESMTP id QAA20031;
	Sun, 25 Mar 2001 16:26:39 -0800 (PST)
Date: Sun, 25 Mar 2001 16:22:35 -0800 (PST)
From: Tom Duffy <tduffy@engr.sgi.com>
To: Russel Ingram <ringram@gargoylecc.com>
cc: <linux-xfs@oss.sgi.com>
Subject: Re: kinda off topic
In-Reply-To: <Pine.LNX.4.30.0103251714450.20493-100000@localhost.localdomain>
Message-ID: <Pine.LNX.4.30.0103251620570.19348-100000@dbear.engr.sgi.com>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

well, the final version of the iso should include some rpms we stick in
the ProPack which fix some of these issues...namely sgi-extra and
sgi-initscripts...

-tduffy

On Sun, 25 Mar 2001, Russel Ingram wrote:

> This is kinda off topic, but...
>
> Is there some way we can get an rpm with the correct setup in
> /etc/security/console.perms so a normal user can use X?  I don't even know
> what rpm that file is part of but its a simple enough thing that it seems
> to me it would be an easy fix to make and include on the sgi iso as a
> replacement for the RH rpm.
>
> Russ
>



From owner-linux-xfs@oss.sgi.com Sun Mar 25 16:42:04 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2Q0g4u29606
	for linux-xfs-outgoing; Sun, 25 Mar 2001 16:42:04 -0800
Received: from roujin.gargoylecc.com ([65.100.85.34])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2Q0g3M29603
	for <linux-xfs@oss.sgi.com>; Sun, 25 Mar 2001 16:42:03 -0800
Received: from lists.gargoylecc.com (IDENT:ringram@gargoyle.gargoylecc.com [65.100.85.35])
	by roujin.gargoylecc.com (8.9.3/8.9.3) with ESMTP id IAA03214;
	Mon, 26 Mar 2001 08:56:07 -0700
Date: Sun, 25 Mar 2001 17:42:34 -0700 (MST)
From: Russel Ingram <ringram@gargoylecc.com>
X-Sender:  <ringram@localhost.localdomain>
To: Tom Duffy <tduffy@engr.sgi.com>
cc: <linux-xfs@oss.sgi.com>
Subject: Re: kinda off topic
In-Reply-To: <Pine.LNX.4.30.0103251620570.19348-100000@dbear.engr.sgi.com>
Message-ID: <Pine.LNX.4.30.0103251739590.20493-100000@localhost.localdomain>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

Cool.  I realize that a lot of people probably use the graphical login so
it doesn't even effect them, but since the xfs install by default includes
devfs it seems that it should work for everyone out-of-box.

On Sun, 25 Mar 2001, Tom Duffy wrote:

> well, the final version of the iso should include some rpms we stick in
> the ProPack which fix some of these issues...namely sgi-extra and
> sgi-initscripts...
>
> -tduffy
>
> On Sun, 25 Mar 2001, Russel Ingram wrote:
>
> > This is kinda off topic, but...
> >
> > Is there some way we can get an rpm with the correct setup in
> > /etc/security/console.perms so a normal user can use X?  I don't even know
> > what rpm that file is part of but its a simple enough thing that it seems
> > to me it would be an easy fix to make and include on the sgi iso as a
> > replacement for the RH rpm.
> >
> > Russ
> >
>
>

-- 
---------------------------------------------------------------
"Bill Gates and Microsoft have ruined the computer industry for
 a long time to come by creating a class of ignorant and lazy
 computer users."  --Russel Ingram

"Mommy ... can I go out and ... KILL TONIGHT!?"
                   --Glen  Danzig, The Misfits
---------------------------------------------------------------
Russ Ingram
Gargoyle Computer Consulting
(307)742-1361
www.gargoylecc.com


From owner-linux-xfs@oss.sgi.com Sun Mar 25 16:48:55 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2Q0mte29816
	for linux-xfs-outgoing; Sun, 25 Mar 2001 16:48:55 -0800
Received: from pneumatic-tube.sgi.com (pneumatic-tube.sgi.com [204.94.214.22])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2Q0mtM29813
	for <linux-xfs@oss.sgi.com>; Sun, 25 Mar 2001 16:48:55 -0800
Received: from cthulhu.engr.sgi.com (gate3-relay.engr.sgi.com [130.62.1.234]) by pneumatic-tube.sgi.com (980327.SGI.8.8.8-aspam/980310.SGI-aspam) via ESMTP id QAA05339
	for <linux-xfs@oss.sgi.com>; Sun, 25 Mar 2001 16:58:58 -0800 (PST)
	mail_from (tduffy@engr.sgi.com)
Received: from dbear.engr.sgi.com (dbear.engr.sgi.com [163.154.18.85])
	by cthulhu.engr.sgi.com (SGI-8.9.3/8.9.3) with ESMTP id QAA56173;
	Sun, 25 Mar 2001 16:47:37 -0800 (PST)
Date: Sun, 25 Mar 2001 16:43:34 -0800 (PST)
From: Tom Duffy <tduffy@engr.sgi.com>
To: Russel Ingram <ringram@gargoylecc.com>
cc: <linux-xfs@oss.sgi.com>
Subject: Re: kinda off topic
In-Reply-To: <Pine.LNX.4.30.0103251743340.20493-100000@localhost.localdomain>
Message-ID: <Pine.LNX.4.30.0103251641550.19348-100000@dbear.engr.sgi.com>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

On Sun, 25 Mar 2001, Russel Ingram wrote:

> One other semi-related thing.  Could we get the devfsd.conf file to
> include entries to set up /dev/cdrom and /dev/mouse?

the way we solve this is that we create symlinks for them on boot up.
basically, if you create a symlink, when the machine goes down (nicely),
it records where your symlinks are in a file called device_catalog and
then recreates them when you boot back up.

not very elegant, but it works.  this should be in the rpms when we
include them on the iso..

-tduffy



From owner-linux-xfs@oss.sgi.com Sun Mar 25 17:08:03 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2Q183c30361
	for linux-xfs-outgoing; Sun, 25 Mar 2001 17:08:03 -0800
Received: from roujin.gargoylecc.com ([65.100.85.34])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2Q182M30358
	for <linux-xfs@oss.sgi.com>; Sun, 25 Mar 2001 17:08:02 -0800
Received: from lists.gargoylecc.com (IDENT:ringram@gargoyle.gargoylecc.com [65.100.85.35])
	by roujin.gargoylecc.com (8.9.3/8.9.3) with ESMTP id JAA03238;
	Mon, 26 Mar 2001 09:22:00 -0700
Date: Sun, 25 Mar 2001 18:08:27 -0700 (MST)
From: Russel Ingram <ringram@gargoylecc.com>
X-Sender:  <ringram@localhost.localdomain>
To: Tom Duffy <tduffy@engr.sgi.com>
cc: <linux-xfs@oss.sgi.com>
Subject: Re: kinda off topic
In-Reply-To: <Pine.LNX.4.30.0103251641550.19348-100000@dbear.engr.sgi.com>
Message-ID: <Pine.LNX.4.30.0103251807400.20552-100000@localhost.localdomain>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

Again, very cool.  I did know how its done, just thought it ought to be
done at install time.

On Sun, 25 Mar 2001, Tom Duffy wrote:

> On Sun, 25 Mar 2001, Russel Ingram wrote:
>
> > One other semi-related thing.  Could we get the devfsd.conf file to
> > include entries to set up /dev/cdrom and /dev/mouse?
>
> the way we solve this is that we create symlinks for them on boot up.
> basically, if you create a symlink, when the machine goes down (nicely),
> it records where your symlinks are in a file called device_catalog and
> then recreates them when you boot back up.
>
> not very elegant, but it works.  this should be in the rpms when we
> include them on the iso..
>
> -tduffy
>
>

-- 
---------------------------------------------------------------
"Bill Gates and Microsoft have ruined the computer industry for
 a long time to come by creating a class of ignorant and lazy
 computer users."  --Russel Ingram

"Mommy ... can I go out and ... KILL TONIGHT!?"
                   --Glen  Danzig, The Misfits
---------------------------------------------------------------
Russ Ingram
Gargoyle Computer Consulting
(307)742-1361
www.gargoylecc.com


From owner-linux-xfs@oss.sgi.com Sun Mar 25 20:28:30 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2Q4SUv02931
	for linux-xfs-outgoing; Sun, 25 Mar 2001 20:28:30 -0800
Received: from mail11.jump.net (mail11.jump.net [206.196.91.11])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2Q4STM02928
	for <linux-xfs@oss.sgi.com>; Sun, 25 Mar 2001 20:28:29 -0800
Received: from Porter (aus-dsl-dhcp1-26.customer.jump.net [63.163.168.26])
	by mail11.jump.net (8.10.2/) with ESMTP id f2Q4RHG20524;
	Sun, 25 Mar 2001 22:27:17 -0600 (CST)
Subject: Re: kinda off topic
From: Eric Sandeen <sandeen@sgi.com>
To: Tom Duffy <tduffy@engr.sgi.com>
Cc: Russel Ingram <ringram@gargoylecc.com>, linux-xfs@oss.sgi.com
In-Reply-To: <Pine.LNX.4.30.0103251620570.19348-100000@dbear.engr.sgi.com>
References: <Pine.LNX.4.30.0103251620570.19348-100000@dbear.engr.sgi.com>
Content-Type: text/plain
X-Mailer: Evolution/0.9+cvs.2001.03.23.09.06 (Preview Release)
Date: 25 Mar 2001 22:28:09 -0600
Message-Id: <985580890.2789.1.camel@Porter>
Mime-Version: 1.0
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

On 25 Mar 2001 16:22:35 -0800, Tom Duffy wrote:
> well, the final version of the iso should include some rpms we stick in
> the ProPack which fix some of these issues...namely sgi-extra and
> sgi-initscripts...



Hm, might be worth submitting this as a bug to Red Hat so that 

they include the vc/* stuff in the pam RPM 
(which is where console.perms comes from...)

I have another bug to submit, I'll go ahead and do that - easier than
writing scripts for the installer.  :)

-Eric


From owner-linux-xfs@oss.sgi.com Sun Mar 25 20:46:04 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2Q4k4d03126
	for linux-xfs-outgoing; Sun, 25 Mar 2001 20:46:04 -0800
Received: from mail15.jump.net (mail15.jump.net [206.196.91.15])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2Q4k3M03123
	for <linux-xfs@oss.sgi.com>; Sun, 25 Mar 2001 20:46:03 -0800
Received: from Porter (aus-dsl-dhcp1-26.customer.jump.net [63.163.168.26])
	by mail15.jump.net (8.10.2/) with ESMTP id f2Q4iHu12497;
	Sun, 25 Mar 2001 22:44:17 -0600 (CST)
Subject: Re: kinda off topic
From: Eric Sandeen <sandeen@sgi.com>
To: Eric Sandeen <sandeen@sgi.com>
Cc: Tom Duffy <tduffy@engr.sgi.com>, Russel Ingram <ringram@gargoylecc.com>,
   linux-xfs@oss.sgi.com
In-Reply-To: <985580890.2789.1.camel@Porter>
References: <Pine.LNX.4.30.0103251620570.19348-100000@dbear.engr.sgi.com> 
	<985580890.2789.1.camel@Porter>
Content-Type: text/plain
X-Mailer: Evolution/0.9+cvs.2001.03.23.09.06 (Preview Release)
Date: 25 Mar 2001 22:45:08 -0600
Message-Id: <985581908.2923.0.camel@Porter>
Mime-Version: 1.0
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

On 25 Mar 2001 22:28:09 -0600, Eric Sandeen wrote:
 
> Hm, might be worth submitting this as a bug to Red Hat so that 


Already fixed in Rawhide, in pam-0.74-18.i386.rpm

Changelog
<snip>
- console.perms: logins on /dev/vc/* are also console logins, from
Douglas
    Gilbert via Tim Waugh

-Eric


From owner-linux-xfs@oss.sgi.com Sun Mar 25 21:11:57 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2Q5BvF03556
	for linux-xfs-outgoing; Sun, 25 Mar 2001 21:11:57 -0800
Received: from lips.borg.umn.edu (lips.borg.umn.edu [160.94.232.50])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2Q5BuM03553
	for <linux-xfs@oss.sgi.com>; Sun, 25 Mar 2001 21:11:56 -0800
Received: from thebarn.com (nic-31-c12-219.mn.mediaone.net [24.31.12.219])
	by lips.borg.umn.edu (8.12.0.Beta5/8.12.0.Beta5) with ESMTP id f2Q5BadE053853;
	Sun, 25 Mar 2001 23:11:36 -0600 (CST)
Message-ID: <3ABECF03.21A0F832@thebarn.com>
Date: Sun, 25 Mar 2001 23:09:23 -0600
From: Russell Cattelan <cattelan@thebarn.com>
X-Mailer: Mozilla 4.75 [en] (X11; U; Linux 2.4.2-XFS i686)
X-Accept-Language: en
MIME-Version: 1.0
To: Russel Ingram <ringram@gargoylecc.com>
CC: thebs@theseus.com, b.j.smith@ieee.org, linux-xfs@oss.sgi.com
Subject: Re: XFS prerelease 0.10 test2 images
References: <Pine.LNX.4.21.0103240301100.24774-100000@roujin.gargoylecc.com>
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

Russel Ingram wrote:

> On Fri, 23 Mar 2001, Bryan-TheBS-Smith wrote:
>
> > Russel Ingram wrote:
> > > Has anyone set up an ftp server to do ftp/http/nfs installs
> > > from yet.  Russell maybe?
> >
> > ???
> > I just did a couple at home on my private network.  I didn't know
> > SGI was offering this capability over the Internet?
> >
> > BTW, it worked great over NFS.
> >
> > -- TheBS
> >
> I think SGI is not.  Russell Cattelan was kind enough with 0.9 to set
> this up on one of his servers at umn and leave it open for those of us
> that wanted to use it.  AFAIK, its a private thing and is probably not
> advertised because Russell doesn't want his server flooded.

The reason SGI doesn't have a full redhat mirror is due to cryto code
in the redhat distribution.

I'll look into setting up a full mirror at thebarn sometime this week.

>
> --
> ---------------------------------------------------------------
> "Bill Gates and Microsoft have ruined the computer industry for
>  a long time to come by creating a class of ignorant and lazy
>  computer users."  --Russel Ingram
>
> "Mommy ... can I go out and ... KILL TONIGHT!?"
>                    --Glen  Danzig, The Misfits
> ---------------------------------------------------------------
> Russ Ingram
> Gargoyle Computer Consulting
> (307)742-1361
> www.gargoylecc.com


From owner-linux-xfs@oss.sgi.com Sun Mar 25 23:09:03 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2Q793O05667
	for linux-xfs-outgoing; Sun, 25 Mar 2001 23:09:03 -0800
Received: from pneumatic-tube.sgi.com (pneumatic-tube.sgi.com [204.94.214.22])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2Q792M05662
	for <linux-xfs@oss.sgi.com>; Sun, 25 Mar 2001 23:09:02 -0800
Received: from snort.melbourne.sgi.com (snort.melbourne.sgi.com [134.14.55.149]) by pneumatic-tube.sgi.com (980327.SGI.8.8.8-aspam/980310.SGI-aspam) via ESMTP id XAA08517
	for <linux-xfs@oss.sgi.com>; Sun, 25 Mar 2001 23:19:05 -0800 (PST)
	mail_from (nathans@snort.melbourne.sgi.com)
Received: (from nathans@localhost)
	by snort.melbourne.sgi.com (SGI-8.9.3/8.9.3) id RAA82590
	for linux-xfs@oss.sgi.com; Mon, 26 Mar 2001 17:07:38 +1000 (EST)
Date: Mon, 26 Mar 2001 17:07:38 +1000 (EST)
From: Nathan Scott <nathans@snort.melbourne.sgi.com>
Message-Id: <200103260707.RAA82590@snort.melbourne.sgi.com>
To: linux-xfs@oss.sgi.com
Subject: TAKE - qa
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

Date:  Sun Mar 25 23:07:33 PST 2001
Workarea:  snort.melbourne.sgi.com:/diskb/build4/nathans/linux-xfs

The following file(s) were checked into:
  bonnie.engr.sgi.com:/isms/slinx/2.4.x-xfs


Modid:  2.4.x-xfs:slinx:90845a
cmd/xfstests/src/nametest.c - 1.2
cmd/xfstests/src/lstat64.c - 1.2
cmd/xfstests/src/alloc.c - 1.2
	- fix warnings when compiled against newer versions of glibc.



From owner-linux-xfs@oss.sgi.com Mon Mar 26 02:41:29 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2QAfTw16377
	for linux-xfs-outgoing; Mon, 26 Mar 2001 02:41:29 -0800
Received: from lucy.physik.tu-cottbus.de (lucy.physik.TU-Cottbus.De [141.43.75.1])
	by oss.sgi.com (8.11.3/8.11.3) with SMTP id f2QAfSM16374
	for <linux-xfs@oss.sgi.com>; Mon, 26 Mar 2001 02:41:28 -0800
Received: (qmail 9035 invoked from network); 26 Mar 2001 10:41:20 -0000
Received: from wynton.physik.tu-cottbus.de (george@141.43.75.16)
  by lucy.physik.tu-cottbus.de with SMTP; 26 Mar 2001 10:41:20 -0000
Date: Mon, 26 Mar 2001 12:41:20 +0200
From: Ionut Georgescu <george@physik.tu-cottbus.de>
To: linux-xfs@oss.sgi.com
Subject: crash with bonnie++
Message-ID: <Pine.SGI.4.20.0103261221260.84003-100000@wynton.physik.tu-cottbus.de>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

Hi,

I did some simple tests with bonnie++ on the CVS tree from Sunday evening.
The command was
	bonnie++ -f -s 512
The system hung up. The error message was:

XFS assertion failed: iclog->ic_state != XLOG_STATE_DO_CALLBACK,
file: xfs_log.c, line 2832
kernel BUG at debug.c: 48!
...
...
Kernel panic: Aiee, killing interrupt handler!
In interrupt handler - not syncing

I tried Alt+SysRq+S -> got the same error message. Rebooted with
Alt+SysRq+B. After reboot the filesystem looked OK.


The crash is reproduceable. The box is a 1GHz Athlon with 256MB RAM. The
partition was ~5GB on a 10 GB SCSI Drive. Controler: aic7xxx. The mkfs
options were the defaults. The mount options were logbufs=4,kio .

I don't have experience on collecting data in such situations, so please
excuse the 'unproper' format of the 'report' :)).

Regards,
Ionut


***************
* Ionut Georgescu      
* http://www.physik.tu-cottbus.de/~george/
* ICQ: 38973105
* "In Windows you can do everything Microsoft wants you to do; in Unix you
*                can do anything the computer is able to do."



From owner-linux-xfs@oss.sgi.com Mon Mar 26 04:57:59 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2QCvxc19821
	for linux-xfs-outgoing; Mon, 26 Mar 2001 04:57:59 -0800
Received: from pmail.now-india.com (pmail.now-india.com [203.195.128.72])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2QCvPM19809
	for <linux-xfs@oss.sgi.com>; Mon, 26 Mar 2001 04:57:56 -0800
Received: from infinity ([203.195.159.115]) by
          pmail.now-india.com (Netscape Messaging Server 4.1) with SMTP id
          GAT2TB00.R5W for <linux-xfs@oss.sgi.com>; Mon, 26 Mar 2001
          18:30:47 +0530 
Message-ID: <006601c0b5f4$5ec91ce0$3a06a8c0@nowindia.net.in>
From: "Mehul Vora" <mehul@now-india.com>
To: <linux-xfs@oss.sgi.com>
Subject: xfs pre-release 0.9 on redhat 6.2
Date: Mon, 26 Mar 2001 18:27:52 +0530
MIME-Version: 1.0
Content-Type: multipart/alternative;
	boundary="----=_NextPart_000_0063_01C0B622.784A4350"
X-Priority: 3
X-MSMail-Priority: Normal
X-Mailer: Microsoft Outlook Express 5.00.2919.6700
X-MimeOLE: Produced By Microsoft MimeOLE V5.00.2919.6700
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

This is a multi-part message in MIME format.

------=_NextPart_000_0063_01C0B622.784A4350
Content-Type: text/plain;
	charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable

Hii,

i was trying to upgrade my redhat 6.2 kernel 2.2.14-12 system to run =
kernel 2.4.0 with xfs pre-release 0.9... i could go through the =
upgradation process (well not actually upgradation cos i didnt overwrite =
2.4.0 over 2.2.14-12) but then when i boot using the new kernel image, =
it says root partition doesnt contain ext2 filesystem or the superblock =
is corrupt and takes me to maintenance prompt... interestingly, i can =
boot through using 2.2.14-12 kernel image just fine... can you advise???

T&R
mehul

------=_NextPart_000_0063_01C0B622.784A4350
Content-Type: text/html;
	charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable

<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.0 Transitional//EN">
<HTML><HEAD>
<META content=3D"text/html; charset=3Diso-8859-1" =
http-equiv=3DContent-Type>
<META content=3D"MSHTML 5.00.2920.0" name=3DGENERATOR>
<STYLE></STYLE>
</HEAD>
<BODY bgColor=3D#ffffff>
<DIV><FONT face=3DArial size=3D2>Hii,</FONT></DIV>
<DIV>&nbsp;</DIV>
<DIV><FONT face=3DArial size=3D2>i was trying to upgrade my redhat 6.2 =
kernel=20
2.2.14-12 system to run kernel 2.4.0 with xfs pre-release 0.9... i could =
go=20
through the upgradation process (well not actually upgradation cos i=20
didnt&nbsp;overwrite 2.4.0 over 2.2.14-12) but then when i boot using =
the new=20
kernel image, it says root partition doesnt contain ext2 filesystem or =
the=20
superblock is corrupt and takes me to maintenance prompt... =
interestingly, i can=20
boot through using 2.2.14-12 kernel image just fine... can you=20
advise???</FONT></DIV>
<DIV>&nbsp;</DIV>
<DIV><FONT face=3DArial size=3D2>T&amp;R</FONT></DIV>
<DIV><FONT face=3DArial size=3D2>mehul</FONT></DIV></BODY></HTML>

------=_NextPart_000_0063_01C0B622.784A4350--


From owner-linux-xfs@oss.sgi.com Mon Mar 26 06:52:10 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2QEqAB21948
	for linux-xfs-outgoing; Mon, 26 Mar 2001 06:52:10 -0800
Received: from yog-sothoth.sgi.com (eugate.sgi.com [192.48.160.10])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2QEq4M21945
	for <linux-xfs@oss.sgi.com>; Mon, 26 Mar 2001 06:52:05 -0800
Received: from zeus-fddi.americas.sgi.com (zeus-fddi.americas.sgi.com [128.162.8.103]) by yog-sothoth.sgi.com (980305.SGI.8.8.8-aspam-6.2/980304.SGI-aspam-europe) via ESMTP id QAA6397755
	for <linux-xfs@oss.sgi.com>; Mon, 26 Mar 2001 16:52:02 +0200 (CEST)
	mail_from (lord@sgi.com)
Received: from daisy-e185.americas.sgi.com (daisy-e194.americas.sgi.com [128.162.194.214]) by zeus-fddi.americas.sgi.com (8.9.3/americas-smart-nospam1.1) with ESMTP id IAA1261630; Mon, 26 Mar 2001 08:50:45 -0600 (CST)
Received: from jen.americas.sgi.com (IDENT:root@jen.americas.sgi.com [128.162.184.86]) by daisy-e185.americas.sgi.com (SGI-8.9.3/SGI-server-1.7) with ESMTP id IAA79367; Mon, 26 Mar 2001 08:50:45 -0600 (CST)
Received: from jen.americas.sgi.com by jen.americas.sgi.com (8.11.0/SGI-client-1.7) via ESMTP id f2QEoFt19250; Mon, 26 Mar 2001 08:50:15 -0600
Message-Id: <200103261450.f2QEoFt19250@jen.americas.sgi.com>
X-Mailer: exmh version 2.2 06/23/2000 with nmh-1.0.4
To: Ionut Georgescu <george@physik.tu-cottbus.de>
cc: linux-xfs@oss.sgi.com
Subject: Re: crash with bonnie++ 
In-Reply-To: Message from Ionut Georgescu <george@physik.tu-cottbus.de> 
   of "Mon, 26 Mar 2001 12:41:20 +0200." <Pine.SGI.4.20.0103261221260.84003-100000@wynton.physik.tu-cottbus.de> 
Date: Mon, 26 Mar 2001 08:50:15 -0600
From: Steve Lord <lord@sgi.com>
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

> Hi,
> 
> I did some simple tests with bonnie++ on the CVS tree from Sunday evening.
> The command was
> 	bonnie++ -f -s 512
> The system hung up. The error message was:
> 
> XFS assertion failed: iclog->ic_state != XLOG_STATE_DO_CALLBACK,
> file: xfs_log.c, line 2832
> kernel BUG at debug.c: 48!
> ...
> ...
> Kernel panic: Aiee, killing interrupt handler!
> In interrupt handler - not syncing
> 
> I tried Alt+SysRq+S -> got the same error message. Rebooted with
> Alt+SysRq+B. After reboot the filesystem looked OK.


I presume you really mean line 2032 - there is no assert anywere near
the line you reported, but this one appears at 2832. Anyway, I will go an
build a debug kernel and see if I cannot reproduce this - this code was
changed recently by a merge of a fix from Irix, you may have found a hole
in this code, or an error in the merging of the fix.

In the meantime, it should be safe to turn the debug compile option off and
run without it - it does make a major performance difference.

You could also try running without the kio option, although I dount this will
make a difference.

Steve


> 
> 
> The crash is reproduceable. The box is a 1GHz Athlon with 256MB RAM. The
> partition was ~5GB on a 10 GB SCSI Drive. Controler: aic7xxx. The mkfs
> options were the defaults. The mount options were logbufs=4,kio .
> 
> I don't have experience on collecting data in such situations, so please
> excuse the 'unproper' format of the 'report' :)).
> 
> Regards,
> Ionut
> 
> 
> ***************
> * Ionut Georgescu      
> * http://www.physik.tu-cottbus.de/~george/
> * ICQ: 38973105
> * "In Windows you can do everything Microsoft wants you to do; in Unix you
> *                can do anything the computer is able to do."
> 



From owner-linux-xfs@oss.sgi.com Mon Mar 26 07:24:56 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2QFOuX23542
	for linux-xfs-outgoing; Mon, 26 Mar 2001 07:24:56 -0800
Received: from yog-sothoth.sgi.com (eugate.sgi.com [192.48.160.10])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2QFOqM23533
	for <linux-xfs@oss.sgi.com>; Mon, 26 Mar 2001 07:24:52 -0800
Received: from zeus-fddi.americas.sgi.com (zeus-fddi.americas.sgi.com [128.162.8.103]) by yog-sothoth.sgi.com (980305.SGI.8.8.8-aspam-6.2/980304.SGI-aspam-europe) via ESMTP id RAA4609999
	for <linux-xfs@oss.sgi.com>; Mon, 26 Mar 2001 17:24:49 +0200 (CEST)
	mail_from (lord@sgi.com)
Received: from daisy-e185.americas.sgi.com (daisy-e194.americas.sgi.com [128.162.194.214]) by zeus-fddi.americas.sgi.com (8.9.3/americas-smart-nospam1.1) with ESMTP id JAA1258688; Mon, 26 Mar 2001 09:23:24 -0600 (CST)
Received: from jen.americas.sgi.com (IDENT:root@jen.americas.sgi.com [128.162.184.86]) by daisy-e185.americas.sgi.com (SGI-8.9.3/SGI-server-1.7) with ESMTP id JAA39610; Mon, 26 Mar 2001 09:23:24 -0600 (CST)
Received: from jen.americas.sgi.com by jen.americas.sgi.com (8.11.0/SGI-client-1.7) via ESMTP id f2QFMso19425; Mon, 26 Mar 2001 09:22:54 -0600
Message-Id: <200103261522.f2QFMso19425@jen.americas.sgi.com>
X-Mailer: exmh version 2.2 06/23/2000 with nmh-1.0.4
To: KELEMEN Peter <fuji@elte.hu>
cc: linux-xfs@oss.sgi.com
Subject: Re: xfs_repair wipes ACL entries, maybe endian-related? 
In-Reply-To: Message from KELEMEN Peter <fuji@elte.hu> 
   of "Fri, 23 Mar 2001 20:11:37 +0100." <20010323201137.D19598@chiara.elte.hu> 
Content-Transfer-Encoding: 8bit
Date: Mon, 26 Mar 2001 09:22:54 -0600
From: Steve Lord <lord@sgi.com>
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

> Hello,
> 
> [ I have no access to the test machine, I'm writing from my memory ]
> 
> I've been peeking at ACLs today (CVS tree as of 2001/03/22 ~16:00
> CET).  After setting the ACL on a file, unmounting the filesystem
> and running xfs_repair, it intends to wipe out the ACL entries.
> Looking at the code, xfsprogs/repair/attr_repair.c::valid_acl()
> decides it should be cleared.  Reason: acl_cnt > MAX_ACL_ENTRIES,
> I sniffed around with xfs_db, SGI_ACL_FILE attr with 304 byte
> value is OK, in my case it was \000\000\000\005... (normal u,g,o
> plus one u and a mask); valid_acl thinks that acl_cnt is 83886080,
> which just happens to be 0x5000000.  I somehow suspect that adding
> some INT_GET(..., ARCH_CONVERT) macros around the obvious places
> in valid_acl() is not The Right Thing To Do(tm), hence no patch.
> 
> Peter
> 


Looks like you are about correct, the kernel code has endian conversion for
on disk acls, the repair code does not.

xfs_acl_get_endian needs to be ported to user space I think, try this
change to cmd/xfsprogs/repair/attr_repair.c

Steve

===========================================================================
Index: cmd/xfsprogs/repair/attr_repair.c
===========================================================================

--- /usr/tmp/TmpDir.19411-0/cmd/xfsprogs/repair/attr_repair.c_1.2       Mon Mar 26 09:22:59 2001
+++ cmd/xfsprogs/repair/attr_repair.c   Mon Mar 26 09:22:10 2001
@@ -921,6 +921,23 @@
        return (err);  /* and repair */
 }
 
+static void
+xfs_acl_get_endian(struct acl *aclp)
+{
+    struct acl_entry *ace, *end;
+
+    /* do the endian conversion */
+    INT_SET(aclp->acl_cnt, ARCH_CONVERT, aclp->acl_cnt);
+
+    /* loop thru ACEs of ACL */
+    end = &aclp->acl_entry[0]+aclp->acl_cnt;
+    for (ace=&aclp->acl_entry[0]; ace < end; ace++) {
+        INT_SET(ace->ae_tag, ARCH_CONVERT, ace->ae_tag);
+        INT_SET(ace->ae_id, ARCH_CONVERT, ace->ae_id);
+        INT_SET(ace->ae_perm, ARCH_CONVERT, ace->ae_perm);
+    }
+}
+
 /* 
  * Validate an ACL
  */
@@ -933,6 +950,8 @@
 
        if (aclp == NULL)
                goto acl_invalid;
+
+       xfs_acl_get_endian(aclp);
 
        if (aclp->acl_cnt > ACL_MAX_ENTRIES)
                goto acl_invalid;



From owner-linux-xfs@oss.sgi.com Mon Mar 26 07:50:03 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2QFo3624773
	for linux-xfs-outgoing; Mon, 26 Mar 2001 07:50:03 -0800
Received: from sgi.com (sgi.SGI.COM [192.48.153.1])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2QFo1M24763
	for <linux-xfs@oss.sgi.com>; Mon, 26 Mar 2001 07:50:01 -0800
Received: from zeus-fddi.americas.sgi.com (128-162-8-103.americas.sgi.com [128.162.8.103]) 
	by sgi.com (980327.SGI.8.8.8-aspam/980304.SGI-aspam:
       SGI does not authorize the use of its proprietary
       systems or networks for unsolicited or bulk email
       from the Internet.) 
	via ESMTP id HAA02594
	for <linux-xfs@oss.sgi.com>; Mon, 26 Mar 2001 07:49:52 -0800 (PST)
	mail_from (lord@sgi.com)
Received: from daisy-e185.americas.sgi.com (daisy-e194.americas.sgi.com [128.162.194.214]) by zeus-fddi.americas.sgi.com (8.9.3/americas-smart-nospam1.1) with ESMTP id JAA1262327; Mon, 26 Mar 2001 09:48:35 -0600 (CST)
Received: from jen.americas.sgi.com (IDENT:root@jen.americas.sgi.com [128.162.184.86]) by daisy-e185.americas.sgi.com (SGI-8.9.3/SGI-server-1.7) with ESMTP id JAA30231; Mon, 26 Mar 2001 09:48:35 -0600 (CST)
Received: from jen.americas.sgi.com by jen.americas.sgi.com (8.11.0/SGI-client-1.7) via ESMTP id f2QFm5Y19448; Mon, 26 Mar 2001 09:48:05 -0600
Message-Id: <200103261548.f2QFm5Y19448@jen.americas.sgi.com>
X-Mailer: exmh version 2.2 06/23/2000 with nmh-1.0.4
To: Ionut Georgescu <george@physik.tu-cottbus.de>, linux-xfs@oss.sgi.com
Subject: Re: crash with bonnie++ 
In-Reply-To: Message from Steve Lord <lord@sgi.com> 
   of "Mon, 26 Mar 2001 08:50:15 CST." <200103261450.f2QEoFt19250@jen.americas.sgi.com> 
Date: Mon, 26 Mar 2001 09:48:05 -0600
From: Steve Lord <lord@sgi.com>
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk


> 
> I presume you really mean line 2032 - there is no assert anywere near
> the line you reported, but this one appears at 2832. Anyway, I will go an
> build a debug kernel and see if I cannot reproduce this - this code was
> changed recently by a merge of a fix from Irix, you may have found a hole
> in this code, or an error in the merging of the fix.
> 
> In the meantime, it should be safe to turn the debug compile option off and
> run without it - it does make a major performance difference.
> 
> You could also try running without the kio option, although I dount this will
> make a difference.
> 
> Steve
> 

The diagnosis is that this is a bogus assert from the new fix, the code is an
incorrect sanity check in the debug path, fix on its way once I get it built.

Steve



From owner-linux-xfs@oss.sgi.com Mon Mar 26 07:53:48 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2QFrms24905
	for linux-xfs-outgoing; Mon, 26 Mar 2001 07:53:48 -0800
Received: from sgi.com (sgi.SGI.COM [192.48.153.1])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2QFrlM24902
	for <linux-xfs@oss.sgi.com>; Mon, 26 Mar 2001 07:53:47 -0800
Received: from zeus-fddi.americas.sgi.com (128-162-8-103.americas.sgi.com [128.162.8.103]) 
	by sgi.com (980327.SGI.8.8.8-aspam/980304.SGI-aspam:
       SGI does not authorize the use of its proprietary
       systems or networks for unsolicited or bulk email
       from the Internet.) 
	via ESMTP id HAA04007
	for <linux-xfs@oss.sgi.com>; Mon, 26 Mar 2001 07:53:11 -0800 (PST)
	mail_from (lord@sgi.com)
Received: from daisy-e185.americas.sgi.com (daisy-e194.americas.sgi.com [128.162.194.214]) by zeus-fddi.americas.sgi.com (8.9.3/americas-smart-nospam1.1) with ESMTP id JAA1261265 for <linux-xfs@oss.sgi.com>; Mon, 26 Mar 2001 09:51:54 -0600 (CST)
Received: from jen.americas.sgi.com (IDENT:root@jen.americas.sgi.com [128.162.184.86]) by daisy-e185.americas.sgi.com (SGI-8.9.3/SGI-server-1.7) with ESMTP id JAA55780 for <linux-xfs@oss.sgi.com>; Mon, 26 Mar 2001 09:51:54 -0600 (CST)
From: Steve Lord <lord@sgi.com>
Received: by jen.americas.sgi.com (8.11.0/SGI-client-1.7) id f2QFpOd20275; Mon, 26 Mar 2001 09:51:24 -0600
Message-Id: <200103261551.f2QFpOd20275@jen.americas.sgi.com>
Date: Mon, 26 Mar 2001 09:51:24 -0600
Subject: TAKE - fix assert failure in debug kernels
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

Bad assertion added with fix merged from irix, the change will only matter
if you turn on the xfs debug code.

Date:  Mon Mar 26 07:50:52 PST 2001
Workarea:  jen.americas.sgi.com:/src/lord/xfs-linux.2.4

The following file(s) were checked into:
  bonnie.engr.sgi.com:/isms/slinx/2.4.x-xfs


Modid:  2.4.x-xfs:slinx:90850a
linux/fs/xfs/xfs_log.c - 1.232
	- Fix incorrect debug code, there is an assert for log buffers not being
	  in the correct state, the assert is valid, but it was being applied to
	  too many log buffers. Make the whole check debug specific, and add logic
	  to drop out of the loop at the correct point.



From owner-linux-xfs@oss.sgi.com Mon Mar 26 07:58:19 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2QFwJO25031
	for linux-xfs-outgoing; Mon, 26 Mar 2001 07:58:19 -0800
Received: from thor.theseus.com (south.orl-pub.theseus.com [12.108.42.66])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2QFwIM25026
	for <linux-xfs@oss.sgi.com>; Mon, 26 Mar 2001 07:58:18 -0800
Received: from theseus.com (IDENT:thebs@fugitive.theseus.com [192.168.0.242])
	by thor.theseus.com (8.9.3/8.9.3) with ESMTP id LAA16418;
	Mon, 26 Mar 2001 11:03:12 -0500
Message-ID: <3ABF6725.7FB6FDED@theseus.com>
Date: Mon, 26 Mar 2001 10:58:29 -0500
From: Bryan-TheBS-Smith <thebs@theseus.com>
Reply-To: thebs@theseus.com, b.j.smith@ieee.org
Organization: (Personal)
X-Mailer: Mozilla 4.76 [en] (X11; U; Linux 2.2.17-0.17-FUGITIVE i686)
X-Accept-Language: en
MIME-Version: 1.0
To: linux-xfs@oss.sgi.com
Subject: Test2:  Sometimes she initrd's, sometimes she doesn't ...
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

Test2:  Sometimes she initrd's, sometimes she doesn't ...

I made two "test installs" on the same system.  One was with
partitions:

   hda2   /boot    Ext2
   hda3   swap
   hda7   /        XFS
   hda8   /tmp     XFS
   hda9   /var     XFS
   hda10  /usr     XFS

And one with partitions:

   hda2   /        Ext2
   hda3   swap
   hda7   /tmp     XFS
   hda8   /var     XFS
   hda9   /usr     XFS

In the first case, with the separate /boot, it did NOT install and
enable an "initrd.img" file.  *BUT*, in the second case, it *DID*! 
Interesting ...

-- TheBS

P.S.  In both cases, I used a 3rd party boot manager with LILO
installed at the begining of the boot/root, respectively,
partitions.  Both booted fine.

--
Bryan "TheBS" Smith          chat:thebs413 @AOL/MSN/Yahoo
Engineer      mailto:b.j.smith@ieee.org,thebs@theseus.com
*********************************************************
"Never apply a Star Trek solution to a Babylon 5 problem"
                                    -- Nicholas C. Weaver

From owner-linux-xfs@oss.sgi.com Mon Mar 26 08:14:04 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2QGE4725318
	for linux-xfs-outgoing; Mon, 26 Mar 2001 08:14:04 -0800
Received: from mail11.jump.net (mail11.jump.net [206.196.91.11])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2QGE3M25315
	for <linux-xfs@oss.sgi.com>; Mon, 26 Mar 2001 08:14:03 -0800
Received: from sgi.com (aus-dsl-dhcp1-26.customer.jump.net [63.163.168.26])
	by mail11.jump.net (8.10.2/) with ESMTP id f2QGDvK24799;
	Mon, 26 Mar 2001 10:13:58 -0600 (CST)
Message-ID: <3ABF6AFF.C6B19AEB@sgi.com>
Date: Mon, 26 Mar 2001 10:14:55 -0600
From: Eric Sandeen <sandeen@sgi.com>
X-Mailer: Mozilla 4.76 [en] (X11; U; Linux 2.4.2-XFS i686)
X-Accept-Language: en
MIME-Version: 1.0
To: thebs@theseus.com, b.j.smith@ieee.org
CC: linux-xfs@oss.sgi.com
Subject: Re: Test2:  Sometimes she initrd's, sometimes she doesn't ...
References: <3ABF6725.7FB6FDED@theseus.com>
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

Yep, I've got to look into this, the XFS installer has been plagued with
mkinitrd problems for a while...

Although - I'm not surprised that there is no initrd in your first case,
since IDE, EXT2, and XFS are all built into the kernel, and no initrd is
needed.

But unless I'm missing something (only one cup of coffee so far today),
I don't see why the second case built an initrd - can you take a look at
what's in the initrd image?

So in this case, you got an extra initrd you didn't need - as opposed to
the unlucky folks who _didn't_ get one that they _did_ need.  :)

-Eric

Bryan-TheBS-Smith wrote:
> 
> Test2:  Sometimes she initrd's, sometimes she doesn't ...
> 
> I made two "test installs" on the same system.  One was with
> partitions:
> 
>    hda2   /boot    Ext2
>    hda3   swap
>    hda7   /        XFS
>    hda8   /tmp     XFS
>    hda9   /var     XFS
>    hda10  /usr     XFS
> 
> And one with partitions:
> 
>    hda2   /        Ext2
>    hda3   swap
>    hda7   /tmp     XFS
>    hda8   /var     XFS
>    hda9   /usr     XFS
> 
> In the first case, with the separate /boot, it did NOT install and
> enable an "initrd.img" file.  *BUT*, in the second case, it *DID*!
> Interesting ...

From owner-linux-xfs@oss.sgi.com Mon Mar 26 08:17:38 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2QGHcX25400
	for linux-xfs-outgoing; Mon, 26 Mar 2001 08:17:38 -0800
Received: from thor.theseus.com (south.orl-pub.theseus.com [12.108.42.66])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2QGHcM25397
	for <linux-xfs@oss.sgi.com>; Mon, 26 Mar 2001 08:17:38 -0800
Received: from theseus.com (IDENT:thebs@fugitive.theseus.com [192.168.0.242])
	by thor.theseus.com (8.9.3/8.9.3) with ESMTP id LAA16945;
	Mon, 26 Mar 2001 11:22:09 -0500
Message-ID: <3ABF6B96.A126F907@theseus.com>
Date: Mon, 26 Mar 2001 11:17:26 -0500
From: Bryan-TheBS-Smith <thebs@theseus.com>
Reply-To: thebs@theseus.com, b.j.smith@ieee.org
Organization: (Personal)
X-Mailer: Mozilla 4.76 [en] (X11; U; Linux 2.2.17-0.17-FUGITIVE i686)
X-Accept-Language: en
MIME-Version: 1.0
To: Eric Sandeen <sandeen@sgi.com>
CC: b.j.smith@ieee.org, linux-xfs@oss.sgi.com
Subject: Re: Test2:  Sometimes she initrd's, sometimes she doesn't ...
References: <3ABF6725.7FB6FDED@theseus.com> <3ABF6AFF.C6B19AEB@sgi.com>
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

Eric Sandeen wrote:
> But unless I'm missing something (only one cup of coffee so far today),
> I don't see why the second case built an initrd - can you take a look at
> what's in the initrd image?

Actually, come to think of it, it is only ~300KB.  Hmmm, I bet it
put aic7xxx in there (I have a BIOS-less AVA-2906 for just a CD-R). 
I'll let you know what is in it when I get home tonight.

-- TheBS

--
Bryan "TheBS" Smith          chat:thebs413 @AOL/MSN/Yahoo
Engineer      mailto:b.j.smith@ieee.org,thebs@theseus.com
*********************************************************
"Never apply a Star Trek solution to a Babylon 5 problem"
                                    -- Nicholas C. Weaver

From owner-linux-xfs@oss.sgi.com Mon Mar 26 08:32:25 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2QGWP525782
	for linux-xfs-outgoing; Mon, 26 Mar 2001 08:32:25 -0800
Received: from pneumatic-tube.sgi.com (pneumatic-tube.sgi.com [204.94.214.22])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2QGWPM25779
	for <linux-xfs@oss.sgi.com>; Mon, 26 Mar 2001 08:32:25 -0800
Received: from zeus-fddi.americas.sgi.com (zeus-fddi.americas.sgi.com [128.162.8.103]) by pneumatic-tube.sgi.com (980327.SGI.8.8.8-aspam/980310.SGI-aspam) via ESMTP id IAA03528
	for <linux-xfs@oss.sgi.com>; Mon, 26 Mar 2001 08:42:29 -0800 (PST)
	mail_from (lord@sgi.com)
Received: from daisy-e185.americas.sgi.com (daisy-e194.americas.sgi.com [128.162.194.214]) by zeus-fddi.americas.sgi.com (8.9.3/americas-smart-nospam1.1) with ESMTP id KAA1263497 for <linux-xfs@oss.sgi.com>; Mon, 26 Mar 2001 10:31:09 -0600 (CST)
Received: from jen.americas.sgi.com (IDENT:root@jen.americas.sgi.com [128.162.184.86]) by daisy-e185.americas.sgi.com (SGI-8.9.3/SGI-server-1.7) with ESMTP id KAA24384 for <linux-xfs@oss.sgi.com>; Mon, 26 Mar 2001 10:31:09 -0600 (CST)
From: Steve Lord <lord@sgi.com>
Received: by jen.americas.sgi.com (8.11.0/SGI-client-1.7) id f2QGUdo20413; Mon, 26 Mar 2001 10:30:39 -0600
Message-Id: <200103261630.f2QGUdo20413@jen.americas.sgi.com>
Date: Mon, 26 Mar 2001 10:30:39 -0600
Subject: TAKE - fix race condition under very heavy load
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

There was a case where two cpus can be looking up the same inode at once,
they both allocate the inode, one of them gets to tear it down. The
problem is the teardown code was incorrect on Linux. 

Date:  Mon Mar 26 08:16:56 PST 2001
Workarea:  jen.americas.sgi.com:/src/lord/xfs-linux.2.4

The following file(s) were checked into:
  bonnie.engr.sgi.com:/isms/slinx/2.4.x-xfs


Modid:  2.4.x-xfs:slinx:90851a
linux/fs/xfs/xfs_iget.c - 1.132
	- Change xfs_iget to call vn_free rather than vn_rele in a race case
	  where two threads create the same inode at once.

linux/fs/xfs/linux/xfs_vnode.c - 1.52
	- Change vn_free to clear the superblock pointer in the inode and call iput,
	  this is called with a partitially initialized inode and we do not want to
	  call the super methods to free it, that causes us to die in xfs.



From owner-linux-xfs@oss.sgi.com Mon Mar 26 08:37:32 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2QGbW225966
	for linux-xfs-outgoing; Mon, 26 Mar 2001 08:37:32 -0800
Received: from sgi.com (sgi.SGI.COM [192.48.153.1])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2QGbVM25962
	for <linux-xfs@oss.sgi.com>; Mon, 26 Mar 2001 08:37:31 -0800
Received: from ledzep.americas.sgi.com (relay.sgi.com [137.38.226.97] (may be forged)) 
	by sgi.com (980327.SGI.8.8.8-aspam/980304.SGI-aspam:
       SGI does not authorize the use of its proprietary
       systems or networks for unsolicited or bulk email
       from the Internet.) 
	via ESMTP id IAA07938
	for <linux-xfs@oss.sgi.com>; Mon, 26 Mar 2001 08:37:27 -0800 (PST)
	mail_from (cattelan@thebarn.com)
Received: from gibble.americas.sgi.com (gibble.americas.sgi.com [128.162.195.80]) by ledzep.americas.sgi.com (SGI-SGI-8.9.3/americas-smart-nospam1.1) with ESMTP id KAA67313; Mon, 26 Mar 2001 10:36:08 -0600 (CST)
Received: from thebarn.com (localhost [127.0.0.1])
	by gibble.americas.sgi.com (8.11.2/8.11.2) with ESMTP id f2QGQ0e22065;
	Mon, 26 Mar 2001 11:26:00 -0500
Message-ID: <3ABF6D98.774B0874@thebarn.com>
Date: Mon, 26 Mar 2001 11:26:00 -0500
From: Russell Cattelan <cattelan@thebarn.com>
X-Mailer: Mozilla 4.76 [en] (X11; U; Linux 2.4.2-XFS i686)
X-Accept-Language: en
MIME-Version: 1.0
To: Eric Sandeen <sandeen@sgi.com>
CC: thebs@theseus.com, b.j.smith@ieee.org, linux-xfs@oss.sgi.com
Subject: Re: Test2:  Sometimes she initrd's, sometimes she doesn't ...
References: <3ABF6725.7FB6FDED@theseus.com> <3ABF6AFF.C6B19AEB@sgi.com>
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

Eric Sandeen wrote:

> Yep, I've got to look into this, the XFS installer has been plagued with
> mkinitrd problems for a while...

This doesn't appear to be an installer problem but rather a kernel problem.

For some reason the loop module either isn't loading or is refusing to
mount
and more devices.
Since the install images is mounted via loopfs the modules does appear to
be
loading but mkinitrd is failing.

I need to find a box that actually fails to make an initrd and try and
diagnose
the error.

The installer could be modified to call mkinitrd with the -x option and
then
dump the output to syslog or a console... that might give us an idea
of where the failure occurs.


>
>
> Although - I'm not surprised that there is no initrd in your first case,
> since IDE, EXT2, and XFS are all built into the kernel, and no initrd is
> needed.
>
> But unless I'm missing something (only one cup of coffee so far today),
> I don't see why the second case built an initrd - can you take a look at
> what's in the initrd image?
>
> So in this case, you got an extra initrd you didn't need - as opposed to
> the unlucky folks who _didn't_ get one that they _did_ need.  :)
>
> -Eric
>
> Bryan-TheBS-Smith wrote:
> >
> > Test2:  Sometimes she initrd's, sometimes she doesn't ...
> >
> > I made two "test installs" on the same system.  One was with
> > partitions:
> >
> >    hda2   /boot    Ext2
> >    hda3   swap
> >    hda7   /        XFS
> >    hda8   /tmp     XFS
> >    hda9   /var     XFS
> >    hda10  /usr     XFS
> >
> > And one with partitions:
> >
> >    hda2   /        Ext2
> >    hda3   swap
> >    hda7   /tmp     XFS
> >    hda8   /var     XFS
> >    hda9   /usr     XFS
> >
> > In the first case, with the separate /boot, it did NOT install and
> > enable an "initrd.img" file.  *BUT*, in the second case, it *DID*!
> > Interesting ...

--
Russell Cattelan
--
Digital Elves inc. -- Currently on loan to SGI
Linux XFS core developer.




From owner-linux-xfs@oss.sgi.com Mon Mar 26 09:10:21 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2QHALP27419
	for linux-xfs-outgoing; Mon, 26 Mar 2001 09:10:21 -0800
Received: from sgi.com (sgi.SGI.COM [192.48.153.1])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2QHA7M27398
	for <linux-xfs@oss.sgi.com>; Mon, 26 Mar 2001 09:10:07 -0800
Received: from zeus-fddi.americas.sgi.com (128-162-8-103.americas.sgi.com [128.162.8.103]) 
	by sgi.com (980327.SGI.8.8.8-aspam/980304.SGI-aspam:
       SGI does not authorize the use of its proprietary
       systems or networks for unsolicited or bulk email
       from the Internet.) 
	via ESMTP id JAA00740
	for <linux-xfs@oss.sgi.com>; Mon, 26 Mar 2001 09:09:55 -0800 (PST)
	mail_from (lord@sgi.com)
Received: from daisy-e185.americas.sgi.com (daisy-e194.americas.sgi.com [128.162.194.214]) by zeus-fddi.americas.sgi.com (8.9.3/americas-smart-nospam1.1) with ESMTP id LAA1264736; Mon, 26 Mar 2001 11:08:21 -0600 (CST)
Received: from jen.americas.sgi.com (IDENT:root@jen.americas.sgi.com [128.162.184.86]) by daisy-e185.americas.sgi.com (SGI-8.9.3/SGI-server-1.7) with ESMTP id LAA07427; Mon, 26 Mar 2001 11:08:19 -0600 (CST)
Received: from jen.americas.sgi.com by jen.americas.sgi.com (8.11.0/SGI-client-1.7) via ESMTP id f2QH7ni20800; Mon, 26 Mar 2001 11:07:49 -0600
Message-Id: <200103261707.f2QH7ni20800@jen.americas.sgi.com>
X-Mailer: exmh version 2.2 06/23/2000 with nmh-1.0.4
To: Jens Axboe <axboe@suse.de>
cc: Petru Paler <ppetru@ppetru.net>, linux-xfs@oss.sgi.com
Subject: Re: Effect of kiobufs? 
In-Reply-To: Message from Jens Axboe <axboe@suse.de> 
   of "Sat, 24 Mar 2001 12:31:18 +0100." <20010324123118.C418@suse.de> 
Date: Mon, 26 Mar 2001 11:07:49 -0600
From: Steve Lord <lord@sgi.com>
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

> On Sat, Mar 24 2001, Petru Paler wrote:
> > Hi,
> > 
> > What are the gains of using the "kio" mount options? Are
> > kiobufs faster than buffer heads, or they just allow more
> > efficient merging of requests?
> 
> The kiobufs allow efficient "merging" from the fs side, so that
> it can submit single large chunks of I/O. The gains can be pretty
> good, I know Steve Lord had some numbers on IDE and SCSI bh vs kiobuf
> that showed this nicely.
> 
> -- 
> Jens Axboe


Hmm, not sure I have those numbers anymore Jens. The other point is that
merging of kiobuf requests within the elevator has never been implemented.

Kiobufs and xfs is a long story, but XFS comes from an environment where the
filesystem and buffer cache do request merging and the elevator does not. 
For the raw I/O path we have shown that not having to setup buffer heads
on requests, but being able to pass the kiobuf directly to the block
layer was a win, I do not have numbers handy for this myself. There is
definitely a cpu penalty to pay in some cases by using buffer heads when
you start out with large chunks of data you want to do I/O on.

XFS itself has chunks of metadata larger than a page, and writes to the
log are upto 32K at once, the work of setting up buffer_head in these
cases would be worth avoiding, but is not a huge benefit.

We used to have code in the xfs write path where we could cluster together
several hundred pages in one I/O request. This gave good throughput, but 
did not play well with the rest of the kernel in terms of resource
consumption, getting pages out of xfs when there was high memory pressure
was a problem, the ordering of writes to the disk was somewhat random as well.

We changed the the algorithm used in the write path to fix these problems
and removed the code which did the clustering in this matter. Our write path
now clusters in the elevator using buffer heads. Thanks to the work Jens
has done there, the elevator is doing a pretty good job nowadays. This
leaves the only things in xfs capable of requests bigger than a page
being the metadata component - which is going to use 1 page I/O in most
places, and the direct I/O path.

Long term I would still like to see some mechanism where the filesystem can
issue requests of larger than a page and have them dealt with correctly. This
will probably be something lighter weight than a kiobuf.

We now return to our regularly scheduled programming!

Steve
 




From owner-linux-xfs@oss.sgi.com Mon Mar 26 09:36:16 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2QHaG028441
	for linux-xfs-outgoing; Mon, 26 Mar 2001 09:36:16 -0800
Received: from mail15.jump.net (mail15.jump.net [206.196.91.15])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2QHaGM28438
	for <linux-xfs@oss.sgi.com>; Mon, 26 Mar 2001 09:36:16 -0800
Received: from sgi.com (aus-dsl-dhcp1-26.customer.jump.net [63.163.168.26])
	by mail15.jump.net (8.10.2/) with ESMTP id f2QHa7S08833;
	Mon, 26 Mar 2001 11:36:07 -0600 (CST)
Message-ID: <3ABF7E42.D634D15F@sgi.com>
Date: Mon, 26 Mar 2001 11:37:06 -0600
From: Eric Sandeen <sandeen@sgi.com>
X-Mailer: Mozilla 4.76 [en] (X11; U; Linux 2.4.2-XFS i686)
X-Accept-Language: en
MIME-Version: 1.0
To: Mehul Vora <mehul@now-india.com>
CC: linux-xfs@oss.sgi.com
Subject: Re: xfs pre-release 0.9 on redhat 6.2
References: <006601c0b5f4$5ec91ce0$3a06a8c0@nowindia.net.in>
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

Mehul Vora wrote:
> Hii,

> i was trying to upgrade my redhat 6.2 kernel 2.2.14-12 system 
> to run kernel 2.4.0 with xfs pre-release 0.9... i could go 
> through the upgradation process (well not actually upgradation 
> cos i didnt overwrite 2.4.0 over 2.2.14-12) but then when i 
> boot using the new kernel image, it says root partition doesnt 
> contain ext2 filesystem or the superblock is corrupt and takes 
> me to maintenance prompt... interestingly, i can boot through
> using 2.2.14-12 kernel image just fine... can you advise???

How did you do the upgrade?  Did you make an initial ramdisk for the new
kernel?  The PR-0.9 kernel has both IDE and EXT2 built as modules.

You might take a look at our latest semi-formal release,
PreRelease-0.10-test2, as it has many updates since 0.9, and is based on
a 2.4.2 kernel.

-Eric

From owner-linux-xfs@oss.sgi.com Mon Mar 26 10:58:44 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2QIwi830903
	for linux-xfs-outgoing; Mon, 26 Mar 2001 10:58:44 -0800
Received: from sws5.ctd.ornl.gov (sws5.ctd.ornl.gov [160.91.68.105])
	by oss.sgi.com (8.11.3/8.11.3) with SMTP id f2QIwiM30900
	for <linux-xfs@oss.sgi.com>; Mon, 26 Mar 2001 10:58:44 -0800
Received: (qmail 31798 invoked by uid 3995); 26 Mar 2001 18:58:43 -0000
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Message-ID: <15039.37219.881717.250160@sws5.ctd.ornl.gov>
Date: Mon, 26 Mar 2001 13:58:43 -0500 (EST)
From: Dave Sill <ds-linux-xfs@sws5.ctd.ornl.gov>
To: linux-xfs@oss.sgi.com
Subject: Moving IRIX disk to Linux
X-Mailer: VM 6.75 under 21.1 (patch 14) "Cuyahoga Valley" XEmacs Lucid
Organization: Oak Ridge National Lab, Oak Ridge, Tenn., USA
X-Face: "p~Q]mg{;e*}YR|)&Q/&Q\*~5UWfZX34;5M<gHyq0KvGt<$qi6=B7;$3~1p?Jc7-g?1vPVAc%YX$E6]%1jvk:,"f*jA|?~Cxb?WJ5gt}L?qsMAjROM~rHbK27yx=t,L/?IHb8@|cYg8"Y)~0IpU~J.^w,VW)u?M3q-AS{f`@RZ]Wll
X-Disclaimer: My opinions do not necessarily represent those of my employer
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

I installed the kernel and other updates from RPM on my Athlon Red Hat
7.0 system:

  Linux malachite 2.4.0-SGI_XFS_PR #1 Fri Jan 12 00:05:18 CST 2001 i686 unknown

It boots OK, except something seems to blow /dev/mouse away during the
boot.

I hooked up a drive with an XFS filesystem created on an IRIX
system. This is just a test; the FS contains junk files.

During the boot, the kernel reports:

  Detected scsi disk sda at scsi0, channel 0, id 1, lun 0
  SCSI device sda: 91923356 512-byte hdwr sectors (47065 MB)
  Partition check:
   /dev/scsi/host0/bus0/target1/lun0: unknown partition table

But fdisk sees the partition table OK:

  Disk /dev/sda (SGI disk label): 28 heads, 328 sectors, 10009 cylinders
  Units = cylinders of 9184 * 512 bytes
  
  ----- partitions -----
     Device  Info      Start       End   Sectors  Id  System
  /dev/sda8               1     10009  91919260   a  SGI xfs
  /dev/sda9               0         0      4096   0  SGI volhdr
  /dev/sda11              0     10009  91923356   6  SGI volume

And the only device file associated with the disk is /dev/sda--no
sda8, sda9, or sda11.

I tried mounting /dev/sda, and got:

  # mount -t xfs /dev/sda /mnt
  mount: wrong fs type, bad option, bad superblock on /dev/sda,
         or too many mounted file systems

Anyone have any ideas? I'm a linux-xfs newbie, but I'm familar with
XFS and Linux.

-Dave

From owner-linux-xfs@oss.sgi.com Mon Mar 26 11:07:56 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2QJ7uP31442
	for linux-xfs-outgoing; Mon, 26 Mar 2001 11:07:56 -0800
Received: from deliverator.sgi.com (deliverator.sgi.com [204.94.214.10])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2QJ7tM31439
	for <linux-xfs@oss.sgi.com>; Mon, 26 Mar 2001 11:07:55 -0800
Received: from zeus-fddi.americas.sgi.com (zeus-fddi.americas.sgi.com [128.162.8.103]) by deliverator.sgi.com (980309.SGI.8.8.8-aspam-6.2/980310.SGI-aspam) via ESMTP id LAA16352
	for <linux-xfs@oss.sgi.com>; Mon, 26 Mar 2001 11:06:41 -0800 (PST)
	mail_from (lord@sgi.com)
Received: from daisy-e185.americas.sgi.com (daisy-e194.americas.sgi.com [128.162.194.214]) by zeus-fddi.americas.sgi.com (8.9.3/americas-smart-nospam1.1) with ESMTP id NAA1265482; Mon, 26 Mar 2001 13:06:37 -0600 (CST)
Received: from jen.americas.sgi.com (IDENT:root@jen.americas.sgi.com [128.162.184.86]) by daisy-e185.americas.sgi.com (SGI-8.9.3/SGI-server-1.7) with ESMTP id NAA45660; Mon, 26 Mar 2001 13:06:37 -0600 (CST)
Received: from jen.americas.sgi.com by jen.americas.sgi.com (8.11.0/SGI-client-1.7) via ESMTP id f2QJ66t21060; Mon, 26 Mar 2001 13:06:06 -0600
Message-Id: <200103261906.f2QJ66t21060@jen.americas.sgi.com>
X-Mailer: exmh version 2.2 06/23/2000 with nmh-1.0.4
To: Dave Sill <ds-linux-xfs@sws5.ctd.ornl.gov>
cc: linux-xfs@oss.sgi.com
Subject: Re: Moving IRIX disk to Linux 
In-Reply-To: Message from Dave Sill <ds-linux-xfs@sws5.ctd.ornl.gov> 
   of "Mon, 26 Mar 2001 13:58:43 EST." <15039.37219.881717.250160@sws5.ctd.ornl.gov> 
Date: Mon, 26 Mar 2001 13:06:06 -0600
From: Steve Lord <lord@sgi.com>
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk


Maybe you need to add irix partition table support - I am not sure this
was done in the pre-release kernels.

Steve

> I installed the kernel and other updates from RPM on my Athlon Red Hat
> 7.0 system:
> 
>   Linux malachite 2.4.0-SGI_XFS_PR #1 Fri Jan 12 00:05:18 CST 2001 i686 unkno
> wn
> 
> It boots OK, except something seems to blow /dev/mouse away during the
> boot.
> 
> I hooked up a drive with an XFS filesystem created on an IRIX
> system. This is just a test; the FS contains junk files.
> 
> During the boot, the kernel reports:
> 
>   Detected scsi disk sda at scsi0, channel 0, id 1, lun 0
>   SCSI device sda: 91923356 512-byte hdwr sectors (47065 MB)
>   Partition check:
>    /dev/scsi/host0/bus0/target1/lun0: unknown partition table
> 
> But fdisk sees the partition table OK:
> 
>   Disk /dev/sda (SGI disk label): 28 heads, 328 sectors, 10009 cylinders
>   Units = cylinders of 9184 * 512 bytes
>   
>   ----- partitions -----
>      Device  Info      Start       End   Sectors  Id  System
>   /dev/sda8               1     10009  91919260   a  SGI xfs
>   /dev/sda9               0         0      4096   0  SGI volhdr
>   /dev/sda11              0     10009  91923356   6  SGI volume
> 
> And the only device file associated with the disk is /dev/sda--no
> sda8, sda9, or sda11.
> 
> I tried mounting /dev/sda, and got:
> 
>   # mount -t xfs /dev/sda /mnt
>   mount: wrong fs type, bad option, bad superblock on /dev/sda,
>          or too many mounted file systems
> 
> Anyone have any ideas? I'm a linux-xfs newbie, but I'm familar with
> XFS and Linux.
> 
> -Dave



From owner-linux-xfs@oss.sgi.com Mon Mar 26 11:14:31 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2QJEVY31929
	for linux-xfs-outgoing; Mon, 26 Mar 2001 11:14:31 -0800
Received: from sws5.ctd.ornl.gov (sws5.ctd.ornl.gov [160.91.68.105])
	by oss.sgi.com (8.11.3/8.11.3) with SMTP id f2QJEUM31925
	for <linux-xfs@oss.sgi.com>; Mon, 26 Mar 2001 11:14:30 -0800
Received: (qmail 32315 invoked by uid 3995); 26 Mar 2001 19:14:30 -0000
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Message-ID: <15039.38166.487732.622471@sws5.ctd.ornl.gov>
Date: Mon, 26 Mar 2001 14:14:30 -0500 (EST)
From: Dave Sill <ds-linux-xfs@sws5.ctd.ornl.gov>
To: linux-xfs@oss.sgi.com
Subject: Re: Moving IRIX disk to Linux 
In-Reply-To: <200103261906.f2QJ66t21060@jen.americas.sgi.com>
References: <ds-linux-xfs@sws5.ctd.ornl.gov>
	<15039.37219.881717.250160@sws5.ctd.ornl.gov>
	<200103261906.f2QJ66t21060@jen.americas.sgi.com>
X-Mailer: VM 6.75 under 21.1 (patch 14) "Cuyahoga Valley" XEmacs Lucid
Organization: Oak Ridge National Lab, Oak Ridge, Tenn., USA
X-Face: "p~Q]mg{;e*}YR|)&Q/&Q\*~5UWfZX34;5M<gHyq0KvGt<$qi6=B7;$3~1p?Jc7-g?1vPVAc%YX$E6]%1jvk:,"f*jA|?~Cxb?WJ5gt}L?qsMAjROM~rHbK27yx=t,L/?IHb8@|cYg8"Y)~0IpU~J.^w,VW)u?M3q-AS{f`@RZ]Wll
X-Disclaimer: My opinions do not necessarily represent those of my employer
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

[Please reply only to the list.]

Steve Lord <lord@sgi.com> wrote:

>Maybe you need to add irix partition table support - I am not sure this
>was done in the pre-release kernels.

OK, how would I go about doing that?

-Dave

From owner-linux-xfs@oss.sgi.com Mon Mar 26 11:16:30 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2QJGU932171
	for linux-xfs-outgoing; Mon, 26 Mar 2001 11:16:30 -0800
Received: from thor.theseus.com (south.orl-pub.theseus.com [12.108.42.66])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2QJGUM32164
	for <linux-xfs@oss.sgi.com>; Mon, 26 Mar 2001 11:16:30 -0800
Received: from theseus.com (IDENT:thebs@fugitive.theseus.com [192.168.0.242])
	by thor.theseus.com (8.9.3/8.9.3) with ESMTP id OAA21611;
	Mon, 26 Mar 2001 14:21:23 -0500
Message-ID: <3ABF9597.B54A4A0E@theseus.com>
Date: Mon, 26 Mar 2001 14:16:39 -0500
From: Bryan-TheBS-Smith <thebs@theseus.com>
Reply-To: thebs@theseus.com, b.j.smith@ieee.org
Organization: (Personal)
X-Mailer: Mozilla 4.76 [en] (X11; U; Linux 2.2.17-0.17-FUGITIVE i686)
X-Accept-Language: en
MIME-Version: 1.0
To: Dave Sill <ds-linux-xfs@sws5.ctd.ornl.gov>
CC: linux-xfs@oss.sgi.com
Subject: Re: Moving IRIX disk to Linux
References: <15039.37219.881717.250160@sws5.ctd.ornl.gov>
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

Dave Sill wrote:
> It boots OK, except something seems to blow /dev/mouse away
> during the boot.

That's the new DevFS -- /dev is created on-the-fly.  Symlinks are
not preserved between boots.  I just had my first experiences with
it too (see http://www.zepa.net/hypermail/elug/2001/03/0164.html).

The best way to fix it is to use the actual device in GPM and X.

E.g.:

   $ cat /etc/sysconfig/gpm 
   # Additional options for gpm (e.g. acceleration)
   OPTIONS="-m /dev/psaux"

   $ cat /etc/X11/XFree86-4
   ...
   Driver          "mouse"
   Option          "Device"                "/dev/psaux"
   ...

> I hooked up a drive with an XFS filesystem created on an IRIX
> system. This is just a test; the FS contains junk files.
> During the boot, the kernel reports:
>   Detected scsi disk sda at scsi0, channel 0, id 1, lun 0
>   SCSI device sda: 91923356 512-byte hdwr sectors (47065 MB)
>   Partition check:
>    /dev/scsi/host0/bus0/target1/lun0: unknown partition table
> But fdisk sees the partition table OK:
>   Disk /dev/sda (SGI disk label): 28 heads, 328 sectors, 10009 cylinders
>   Units = cylinders of 9184 * 512 bytes
> 
>   ----- partitions -----
>      Device  Info      Start       End   Sectors  Id  System
>   /dev/sda8               1     10009  91919260   a  SGI xfs
>   /dev/sda9               0         0      4096   0  SGI volhdr
>   /dev/sda11              0     10009  91923356   6  SGI volume
> And the only device file associated with the disk is /dev/sda--no
> sda8, sda9, or sda11.

Was the disk from a SGI Irix system?

The pre-release kernels don't have the partition support for SGI
disk labels compiled in by default.

-- TheBS

--
Bryan "TheBS" Smith          chat:thebs413 @AOL/MSN/Yahoo
Engineer      mailto:b.j.smith@ieee.org,thebs@theseus.com
*********************************************************
"Never apply a Star Trek solution to a Babylon 5 problem"
                                    -- Nicholas C. Weaver

From owner-linux-xfs@oss.sgi.com Mon Mar 26 11:23:07 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2QJN7N32609
	for linux-xfs-outgoing; Mon, 26 Mar 2001 11:23:07 -0800
Received: from mail15.jump.net (mail15.jump.net [206.196.91.15])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2QJN6M32606
	for <linux-xfs@oss.sgi.com>; Mon, 26 Mar 2001 11:23:06 -0800
Received: from sgi.com (aus-dsl-dhcp1-26.customer.jump.net [63.163.168.26])
	by mail15.jump.net (8.10.2/) with ESMTP id f2QJN6J10059
	for <linux-xfs@oss.sgi.com>; Mon, 26 Mar 2001 13:23:06 -0600 (CST)
Message-ID: <3ABF9754.13107384@sgi.com>
Date: Mon, 26 Mar 2001 13:24:04 -0600
From: Eric Sandeen <sandeen@sgi.com>
X-Mailer: Mozilla 4.76 [en] (X11; U; Linux 2.4.2-XFS i686)
X-Accept-Language: en
MIME-Version: 1.0
To: linux-xfs@oss.sgi.com
Subject: Re: Moving IRIX disk to Linux
References: <ds-linux-xfs@sws5.ctd.ornl.gov>
		<15039.37219.881717.250160@sws5.ctd.ornl.gov>
		<200103261906.f2QJ66t21060@jen.americas.sgi.com> <15039.38166.487732.622471@sws5.ctd.ornl.gov>
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

Dave Sill wrote:
> 
> [Please reply only to the list.]
> 
> Steve Lord <lord@sgi.com> wrote:
> 
> >Maybe you need to add irix partition table support - I am not sure this
> >was done in the pre-release kernels.

It is in fact turned off in the kernel RPMS... we'll turn it on in the
next spin.

> OK, how would I go about doing that?

Go to /usr/src/linux and run make menuconfig, or make xconfig, and
configure your kernel with Advanced Partition Selection and SGI
partition support turned on... then recompile your kernel & install. 
(http://www.linuxdoc.org/HOWTO/Kernel-HOWTO.html)

Of course you'll have to go through all the other config options as
well... I can send you a .config to start with if you'd like.

-Eric

From owner-linux-xfs@oss.sgi.com Mon Mar 26 11:23:51 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2QJNpf32761
	for linux-xfs-outgoing; Mon, 26 Mar 2001 11:23:51 -0800
Received: from pneumatic-tube.sgi.com (pneumatic-tube.sgi.com [204.94.214.22])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2QJNoM32757
	for <linux-xfs@oss.sgi.com>; Mon, 26 Mar 2001 11:23:50 -0800
Received: from zeus-fddi.americas.sgi.com (zeus-fddi.americas.sgi.com [128.162.8.103]) by pneumatic-tube.sgi.com (980327.SGI.8.8.8-aspam/980310.SGI-aspam) via ESMTP id LAA08018
	for <linux-xfs@oss.sgi.com>; Mon, 26 Mar 2001 11:33:55 -0800 (PST)
	mail_from (lord@sgi.com)
Received: from daisy-e185.americas.sgi.com (daisy-e194.americas.sgi.com [128.162.194.214]) by zeus-fddi.americas.sgi.com (8.9.3/americas-smart-nospam1.1) with ESMTP id NAA1265491; Mon, 26 Mar 2001 13:22:26 -0600 (CST)
Received: from jen.americas.sgi.com (IDENT:root@jen.americas.sgi.com [128.162.184.86]) by daisy-e185.americas.sgi.com (SGI-8.9.3/SGI-server-1.7) with ESMTP id NAA12250; Mon, 26 Mar 2001 13:22:26 -0600 (CST)
Received: from jen.americas.sgi.com by jen.americas.sgi.com (8.11.0/SGI-client-1.7) via ESMTP id f2QJLt721201; Mon, 26 Mar 2001 13:21:55 -0600
Message-Id: <200103261921.f2QJLt721201@jen.americas.sgi.com>
X-Mailer: exmh version 2.2 06/23/2000 with nmh-1.0.4
To: Dave Sill <ds-linux-xfs@sws5.ctd.ornl.gov>
cc: linux-xfs@oss.sgi.com
Subject: Re: Moving IRIX disk to Linux 
In-Reply-To: Message from Dave Sill <ds-linux-xfs@sws5.ctd.ornl.gov> 
   of "Mon, 26 Mar 2001 14:14:30 EST." <15039.38166.487732.622471@sws5.ctd.ornl.gov> 
Date: Mon, 26 Mar 2001 13:21:55 -0600
From: Steve Lord <lord@sgi.com>
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

> [Please reply only to the list.]
> 
> Steve Lord <lord@sgi.com> wrote:
> 
> >Maybe you need to add irix partition table support - I am not sure this
> >was done in the pre-release kernels.
> 
> OK, how would I go about doing that?
> 
> -Dave

You have to rebuild the kernel, in make xconfig at the bottom of the
filesystems menu is a partition types option, within this you need to
enable advanced partition selection, and add sgi partition support
which is three up from the bottom.

The new kernel should then recognize the partitions OK.

This all presumes you are OK with configuring, building and installing kernels.

Steve


Steve



From owner-linux-xfs@oss.sgi.com Mon Mar 26 11:39:38 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2QJdcm01206
	for linux-xfs-outgoing; Mon, 26 Mar 2001 11:39:38 -0800
Received: from sgi.com (sgi.SGI.COM [192.48.153.1])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2QJdPM01197
	for <linux-xfs@oss.sgi.com>; Mon, 26 Mar 2001 11:39:25 -0800
Received: from info.engr.sgi.com ([192.26.80.216]) 
	by sgi.com (980327.SGI.8.8.8-aspam/980304.SGI-aspam:
       SGI does not authorize the use of its proprietary
       systems or networks for unsolicited or bulk email
       from the Internet.) 
	via ESMTP id LAA09708; Mon, 26 Mar 2001 11:38:39 -0800 (PST)
	mail_from (nobody@info.engr.sgi.com)
Received: (from nobody@localhost)
	by info.engr.sgi.com (SGI-8.9.3/8.9.3) id LAA00233;
	Mon, 26 Mar 2001 11:38:38 -0800 (PST)
Date: Mon, 26 Mar 2001 11:38:38 -0800 (PST)
Message-Id: <200103261938.LAA00233@info.engr.sgi.com>
X-Pv-Incident: 819118 
webPV: fsgi158.americas.sgi.com 
webExec: webpvsubmit,PvProjectIncident 
Reply-To: sgi.bugs.xfs@fido.engr.sgi.com
From: pv@relay.sgi.com (tbd@sgi.com)
Subject: BUG 819118 - pwd on sun NFS client does not return entire path
To: nb@sgi.com
Cc: linux-xfs@oss.sgi.com
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

View Incident: http://co-op.engr.sgi.com/BugWorks/code/bwxquery.cgi?search=Search&wlong=1&view_type=Bug&wi=819118

Submitter : tbd                       Submitter Domain : sgi.com            
Assigned Engineer : nb                Assigned Domain : sgi.com             
Assigned Group : xfs-linux            Category : software                   
Customer Reported : F                 Priority : 2                          
Project : xfs-linux                   Status : open                         
Description :
pwd does not always return the entire path on an NFS mounted directory on 
a sun client (i.e. the local mount point is not included). The server 
is running 2.4.2 with the equivalent of XFS release 0.10. This happens when
using version 2 or version 3 NFS. Does NOT happen with with ext2 file
system or with linux client.

latte 31% /etc/mount | grep tarpon
  /tmp/tarpon_v2 on tarpon:/mnt1 vers=2/remote on Mon Mar 26 09:48:56 2001
  /tmp/tarpon_v3 on tarpon:/mnt1 vers=3/remote on Mon Mar 26 09:49:04 2001
  latte 32% cd /tmp/tarpon_v3
  latte 33% pwd
  /tmp/tarpon_v3
  latte 34% cd cthon00
  latte 35% pwd
  pwd: cannot determine current directory!

Note that a similar problem also occurs using IRIX clients but that is
because of a known problem in IRIX which is documented in PV 815265.

Note that it only happens with pwd under csh and /bin/pwd.
  latte 28% ksh
  $ cd /tmp/tarpon_v3/cthon00
  $ pwd
  /tmp/tarpon_v3/cthon00
  $ csh
  latte 21% pwd
  /cthon00
  latte 22% ksh   
  $ pwd
  /cthon00
  $ cd /tmp/tarpon_v3/cthon00
  $ pwd
  /tmp/tarpon_v3/cthon00

latte 29% truss /bin/pwd
execve("/bin/pwd", 0xDFFFFCE8, 0xDFFFFCF0)  argc = 1
open("/dev/zero", O_RDONLY)                     = 3
mmap(0x00000000, 4096, PROT_READ|PROT_WRITE|PROT_EXEC, MAP_PRIVATE, 3, 0) = 0xDF7C0000
open("/usr/lib/libc.so.1", O_RDONLY)            = 4
fstat(4, 0xDFFFF88C)                            = 0
mmap(0x00000000, 4096, PROT_READ|PROT_EXEC, MAP_PRIVATE, 4, 0) = 0xDF7B0000
mmap(0x00000000, 700416, PROT_READ|PROT_EXEC, MAP_PRIVATE, 4, 0) = 0xDF700000
munmap(0xDF793000, 61440)                       = 0
mmap(0xDF7A2000, 27744, PROT_READ|PROT_WRITE|PROT_EXEC, MAP_PRIVATE|MAP_FIXED, 4, 598016) = 0xDF7A2000
mmap(0xDF7A9000, 6000, PROT_READ|PROT_WRITE|PROT_EXEC, MAP_PRIVATE|MAP_FIXED, 3, 0) = 0xDF7A9000
close(4)                                        = 0
open("/usr/lib/libdl.so.1", O_RDONLY)           = 4
fstat(4, 0xDFFFF88C)                            = 0
mmap(0xDF7B0000, 4096, PROT_READ|PROT_EXEC, MAP_PRIVATE|MAP_FIXED, 4, 0) = 0xDF7B0000
close(4)                                        = 0
open("/usr/platform/SUNW,SPARCserver-1000/lib/libc_psr.so.1", O_RDONLY) Err#2 ENOENT
close(3)                                        = 0
pathconf(".", _PC_PATH_MAX)                     = 1024
stat64("./", 0xDFFFFB98)                        = 0
brk(0x00021178)                                 = 0
brk(0x00023178)                                 = 0
stat64("/", 0xDFFFFB00)                         = 0
open64("./../", O_RDONLY|O_NDELAY)              = 3
fcntl(3, F_SETFD, 0x00000001)                   = 0
fstat64(3, 0xDFFFF1A0)                          = 0
fstat64(3, 0xDFFFFB98)                          = 0
getdents64(3, 0x00021998, 1048)                 = 1040
close(3)                                        = 0
open64("./../../", O_RDONLY|O_NDELAY)           = 3
fcntl(3, F_SETFD, 0x00000001)                   = 0
fstat64(3, 0xDFFFF1A0)                          = 0
fstat64(3, 0xDFFFFB98)                          = 0
close(3)                                        = 0
/cthon00
write(1, " / c t h o n 0 0\n", 9)               = 9
llseek(0, 0, SEEK_CUR)                          = 7329
_exit(0)
latte 30%

From owner-linux-xfs@oss.sgi.com Mon Mar 26 11:45:57 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2QJjvk01522
	for linux-xfs-outgoing; Mon, 26 Mar 2001 11:45:57 -0800
Received: from pneumatic-tube.sgi.com (pneumatic-tube.sgi.com [204.94.214.22])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2QJjvM01519
	for <linux-xfs@oss.sgi.com>; Mon, 26 Mar 2001 11:45:57 -0800
Received: from info.engr.sgi.com (info.engr.sgi.com [192.26.80.216]) by pneumatic-tube.sgi.com (980327.SGI.8.8.8-aspam/980310.SGI-aspam) via ESMTP id LAA08015; Mon, 26 Mar 2001 11:56:01 -0800 (PST)
	mail_from (nobody@info.engr.sgi.com)
Received: (from nobody@localhost)
	by info.engr.sgi.com (SGI-8.9.3/8.9.3) id LAA02447;
	Mon, 26 Mar 2001 11:45:56 -0800 (PST)
Date: Mon, 26 Mar 2001 11:45:56 -0800 (PST)
Message-Id: <200103261945.LAA02447@info.engr.sgi.com>
X-Pv-Incident: 819118 
webPV: jen.americas.sgi.com 
webExec: webpvupdate,pvincident 
Reply-To: sgi.bugs.xfs@fido.engr.sgi.com
From: pv@relay.sgi.com (lord@sgi.com)
Subject: ADD 819118 - pwd on sun NFS client does not return entire path
To: nb@sgi.com
Cc: linux-xfs@oss.sgi.com
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

View Incident: http://co-op.engr.sgi.com/BugWorks/code/bwxquery.cgi?search=Search&wlong=1&view_type=Bug&wi=819118

 Status : open                         Priority : 2                         
 Assigned Engineer : nb                Submitter : tbd                      
*Modified User : lord                 *Modified User Domain : sgi.com       
*Description :
pwd does not always return the entire path on an NFS mounted directory on 
a sun client (i.e. the local mount point is not included). The server 
is running 2.4.2 with the equivalent of XFS release 0.10. This happens when
using version 2 or version 3 NFS. Does NOT happen with with ext2 file
system or with linux client.

latte 31% /etc/mount | grep tarpon
  /tmp/tarpon_v2 on tarpon:/mnt1 vers=2/remote on Mon Mar 26 09:48:56 2001
  /tmp/tarpon_v3 on tarpon:/mnt1 vers=3/remote on Mon Mar 26 09:49:04 2001
  latte 32% cd /tmp/tarpon_v3

.....


==========================
ADDITIONAL INFORMATION (ADD)
From: lord@sgi.com (BugWorks)
Date: Mar 26 2001 11:45:55AM
==========================

Please do the following on both nfs over ext2 and nfs over xfs:

on the sun run this from a directory one level down within the
filesystem.

truss -v stat64,getdents64,fstat64 /usr/bin/pwd

and put the results in the pv.

Thanks

   Steve

From owner-linux-xfs@oss.sgi.com Mon Mar 26 11:58:50 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2QJwoQ01929
	for linux-xfs-outgoing; Mon, 26 Mar 2001 11:58:50 -0800
Received: from pneumatic-tube.sgi.com (pneumatic-tube.sgi.com [204.94.214.22])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2QJwmM01926
	for <linux-xfs@oss.sgi.com>; Mon, 26 Mar 2001 11:58:49 -0800
Received: from info.engr.sgi.com (info.engr.sgi.com [192.26.80.216]) by pneumatic-tube.sgi.com (980327.SGI.8.8.8-aspam/980310.SGI-aspam) via ESMTP id MAA05651; Mon, 26 Mar 2001 12:08:53 -0800 (PST)
	mail_from (nobody@info.engr.sgi.com)
Received: (from nobody@localhost)
	by info.engr.sgi.com (SGI-8.9.3/8.9.3) id LAA72437;
	Mon, 26 Mar 2001 11:58:47 -0800 (PST)
Date: Mon, 26 Mar 2001 11:58:47 -0800 (PST)
Message-Id: <200103261958.LAA72437@info.engr.sgi.com>
X-Pv-Incident: 819118 
webPV: fsgi158.americas.sgi.com 
webExec: webpvupdate,pvincident 
Reply-To: sgi.bugs.xfs@fido.engr.sgi.com
From: pv@relay.sgi.com (tbd@sgi.com)
Subject: ADD 819118 - pwd on sun NFS client does not return entire path
To: nb@sgi.com
Cc: linux-xfs@oss.sgi.com
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

View Incident: http://co-op.engr.sgi.com/BugWorks/code/bwxquery.cgi?search=Search&wlong=1&view_type=Bug&wi=819118

 Status : open                         Priority : 2                         
 Assigned Engineer : nb                Submitter : tbd                      
*Modified User : tbd                  *Modified User Domain : sgi.com       
*Description :
pwd does not always return the entire path on an NFS mounted directory on 
a sun client (i.e. the local mount point is not included). The server 
is running 2.4.2 with the equivalent of XFS release 0.10. This happens when
using version 2 or version 3 NFS. Does NOT happen with with ext2 file
system or with linux client.

latte 31% /etc/mount | grep tarpon
  /tmp/tarpon_v2 on tarpon:/mnt1 vers=2/remote on Mon Mar 26 09:48:56 2001
  /tmp/tarpon_v3 on tarpon:/mnt1 vers=3/remote on Mon Mar 26 09:49:04 2001
  latte 32% cd /tmp/tarpon_v3

.....


==========================
ADDITIONAL INFORMATION (ADD)
From: tbd@sgi.com (BugWorks)
Date: Mar 26 2001 11:58:47AM
==========================

>Please do the following on both nfs over ext2 and nfs over xfs:
>
>on the sun run this from a directory one level down within the
>filesystem.
>
>truss -v stat64,getdents64,fstat64 /usr/bin/pwd
>
>and put the results in the pv.

------------
----ext2----
------------
latte 27% truss -v stat64,getdents64,fstat64 /usr/bin/pwd
execve("/usr/bin/pwd", 0xDFFFFCC0, 0xDFFFFCC8)  argc = 1
open("/dev/zero", O_RDONLY)                     = 3
mmap(0x00000000, 4096, PROT_READ|PROT_WRITE|PROT_EXEC, MAP_PRIVATE, 3, 0) = 0xDF7C0000
open("/usr/lib/libc.so.1", O_RDONLY)            = 4
fstat(4, 0xDFFFF884)                            = 0
    d=0x00800006 i=2314  m=0100755 l=1  u=2     g=2     sz=1013292
        at = Mar 26 13:50:40 CST 2001  [ 985636240 ]
        mt = Apr 21 03:31:13 CDT 1999  [ 924683473 ]
        ct = Apr 28 15:29:42 CDT 1999  [ 925331382 ]
    bsz=8192  blks=2000  fs=ufs
mmap(0x00000000, 4096, PROT_READ|PROT_EXEC, MAP_PRIVATE, 4, 0) = 0xDF7B0000
mmap(0x00000000, 700416, PROT_READ|PROT_EXEC, MAP_PRIVATE, 4, 0) = 0xDF700000
munmap(0xDF793000, 61440)                       = 0
mmap(0xDF7A2000, 27744, PROT_READ|PROT_WRITE|PROT_EXEC, MAP_PRIVATE|MAP_FIXED, 4, 598016) = 0xDF7A2000
mmap(0xDF7A9000, 6000, PROT_READ|PROT_WRITE|PROT_EXEC, MAP_PRIVATE|MAP_FIXED, 3, 0) = 0xDF7A9000
close(4)                                        = 0
open("/usr/lib/libdl.so.1", O_RDONLY)           = 4
fstat(4, 0xDFFFF884)                            = 0
    d=0x00800006 i=2321  m=0100755 l=1  u=2     g=2     sz=4280
        at = Mar 26 13:50:40 CST 2001  [ 985636240 ]
        mt = Apr 21 03:31:41 CDT 1999  [ 924683501 ]
        ct = Apr 28 15:44:59 CDT 1999  [ 925332299 ]
    bsz=8192  blks=10    fs=ufs
mmap(0xDF7B0000, 4096, PROT_READ|PROT_EXEC, MAP_PRIVATE|MAP_FIXED, 4, 0) = 0xDF7B0000
close(4)                                        = 0
open("/usr/platform/SUNW,SPARCserver-1000/lib/libc_psr.so.1", O_RDONLY) Err#2 ENOENT
close(3)                                        = 0
pathconf(".", _PC_PATH_MAX)                     = 1024
stat64("./", 0xDFFFFB90)                        = 0
    d=0x02B000BC i=820082 m=0040755 l=7  u=24241 g=1047  sz=4096
        at = Mar 26 08:48:55 CST 2001  [ 985618135 ]
        mt = Mar 26 08:48:55 CST 2001  [ 985618135 ]
        ct = Mar 26 08:48:55 CST 2001  [ 985618135 ]
    bsz=8192  blks=8     fs=nfs
brk(0x00021178)                                 = 0
brk(0x00023178)                                 = 0
stat64("/", 0xDFFFFAF8)                         = 0
    d=0x00800000 i=2     m=0040755 l=33 u=0     g=0     sz=1024
        at = Mar 25 17:02:06 CST 2001  [ 985561326 ]
        mt = Jan 24 13:41:35 CST 2001  [ 980365295 ]
        ct = Jan 24 13:41:35 CST 2001  [ 980365295 ]
    bsz=8192  blks=2     fs=ufs
open64("./../", O_RDONLY|O_NDELAY)              = 3
fcntl(3, F_SETFD, 0x00000001)                   = 0
fstat64(3, 0xDFFFF198)                          = 0
    d=0x02B000BC i=180225 m=0041777 l=12 u=0     g=0     sz=4096
        at = Mar 26 08:51:23 CST 2001  [ 985618283 ]
        mt = Mar 26 08:48:55 CST 2001  [ 985618135 ]
        ct = Mar 26 08:48:55 CST 2001  [ 985618135 ]
    bsz=8192  blks=8     fs=nfs
fstat64(3, 0xDFFFFB90)                          = 0
    d=0x02B000BC i=180225 m=0041777 l=12 u=0     g=0     sz=4096
        at = Mar 26 08:51:23 CST 2001  [ 985618283 ]
        mt = Mar 26 08:48:55 CST 2001  [ 985618135 ]
        ct = Mar 26 08:48:55 CST 2001  [ 985618135 ]
    bsz=8192  blks=8     fs=nfs
getdents64(3, 0x00021998, 1048)                 = 552
    ino=180225 off=12   rlen=24  "."
    ino=2     off=24   rlen=24  ".."
    ino=921011 off=44   rlen=32  ".font-unix"
    ino=804348 off=56   rlen=24  "x"
    ino=182682 off=80   rlen=32  "outputfile"
    ino=99934 off=104  rlen=32  ".sawmill-root"
    ino=182685 off=124  rlen=32  "rpctorture.c"
    ino=181695 off=140  rlen=24  "disk"
    ino=557889 off=160  rlen=32  ".ICE-unix"
    ino=885242 off=180  rlen=32  "orbit-root"
    ino=99775 off=200  rlen=32  "SANavigator"
    ino=607263 off=228  rlen=24  "smbfs"
    ino=869460 off=244  rlen=24  "test"
    ino=67020 off=256  rlen=24  "temp"
    ino=820082 off=384  rlen=32  "cthon00"
    ino=182683 off=404  rlen=32  "file_13.out"
    ino=182686 off=424  rlen=32  "includes.h"
    ino=183530 off=4096 rlen=64  "samba-debugtools-2.0.4b-19990519.i386.rpm"
close(3)                                        = 0
open64("./../../", O_RDONLY|O_NDELAY)           = 3
fcntl(3, F_SETFD, 0x00000001)                   = 0
fstat64(3, 0xDFFFF198)                          = 0
    d=0x00000001 i=503623994 m=0041777 l=20 u=3     g=3     sz=1030
        at = Mar 26 09:48:27 CST 2001  [ 985621707 ]
        mt = Mar 26 13:48:53 CST 2001  [ 985636133 ]
        ct = Mar 26 13:48:53 CST 2001  [ 985636133 ]
    bsz=4096  blks=8     fs=tmpfs
fstat64(3, 0xDFFFFB90)                          = 0
    d=0x00000001 i=503623994 m=0041777 l=20 u=3     g=3     sz=1030
        at = Mar 26 09:48:27 CST 2001  [ 985621707 ]
        mt = Mar 26 13:48:53 CST 2001  [ 985636133 ]
        ct = Mar 26 13:48:53 CST 2001  [ 985636133 ]
    bsz=4096  blks=8     fs=tmpfs
open("/etc/mnttab", O_RDONLY)                   = 4
fstat64(4, 0xDFFFEB50)                          = 0
    d=0x00800000 i=36051 m=0100644 l=1  u=0     g=3     sz=6434
        at = Mar 26 13:50:04 CST 2001  [ 985636204 ]
        mt = Mar 26 13:49:34 CST 2001  [ 985636174 ]
        ct = Mar 26 13:49:34 CST 2001  [ 985636174 ]
    bsz=8192  blks=14    fs=ufs
brk(0x00023178)                                 = 0
brk(0x00025178)                                 = 0
ioctl(4, TCGETA, 0xDFFFEADC)                    Err#25 ENOTTY
read(4, " / d e v / d s k / c 0 t".., 8192)     = 6434
lstat64("/opt/dcelocal/var/dfs/aggrs/lfs1", 0xDFFFEC78) = 0
lstat64("/opt/dcelocal/var/dfs/aggrs/lfs.root", 0xDFFFEC78) = 0
lstat64("/...", 0xDFFFEC78)                     = 0
lstat64("/tmp/tarpon_ext2_v3", 0xDFFFED78)      = 0
lstat64("/tmp/tarpon_ext2_v3/..", 0xDFFFED78)   = 0
llseek(4, 0, SEEK_CUR)                          = 6434
close(4)                                        = 0
close(3)                                        = 0
/tmp/tarpon_ext2_v3/cthon00
write(1, " / t m p / t a r p o n _".., 28)      = 28
llseek(0, 0, SEEK_CUR)                          = 5053
_exit(0)
latte 28% 


--------------
--XFS---------
--------------
latte 27% truss -v stat64,getdents64,fstat64 /usr/bin/pwd
execve("/usr/bin/pwd", 0xDFFFFCC0, 0xDFFFFCC8)  argc = 1
open("/dev/zero", O_RDONLY)                     = 3
mmap(0x00000000, 4096, PROT_READ|PROT_WRITE|PROT_EXEC, MAP_PRIVATE, 3, 0) = 0xDF7C0000
open("/usr/lib/libc.so.1", O_RDONLY)            = 4
fstat(4, 0xDFFFF884)                            = 0
    d=0x00800006 i=2314  m=0100755 l=1  u=2     g=2     sz=1013292
        at = Mar 26 13:50:40 CST 2001  [ 985636240 ]
        mt = Apr 21 03:31:13 CDT 1999  [ 924683473 ]
        ct = Apr 28 15:29:42 CDT 1999  [ 925331382 ]
    bsz=8192  blks=2000  fs=ufs
mmap(0x00000000, 4096, PROT_READ|PROT_EXEC, MAP_PRIVATE, 4, 0) = 0xDF7B0000
mmap(0x00000000, 700416, PROT_READ|PROT_EXEC, MAP_PRIVATE, 4, 0) = 0xDF700000
munmap(0xDF793000, 61440)                       = 0
mmap(0xDF7A2000, 27744, PROT_READ|PROT_WRITE|PROT_EXEC, MAP_PRIVATE|MAP_FIXED, 4, 598016) = 0xDF7A2000
mmap(0xDF7A9000, 6000, PROT_READ|PROT_WRITE|PROT_EXEC, MAP_PRIVATE|MAP_FIXED, 3, 0) = 0xDF7A9000
close(4)                                        = 0
open("/usr/lib/libdl.so.1", O_RDONLY)           = 4
fstat(4, 0xDFFFF884)                            = 0
    d=0x00800006 i=2321  m=0100755 l=1  u=2     g=2     sz=4280
        at = Mar 26 13:50:40 CST 2001  [ 985636240 ]
        mt = Apr 21 03:31:41 CDT 1999  [ 924683501 ]
        ct = Apr 28 15:44:59 CDT 1999  [ 925332299 ]
    bsz=8192  blks=10    fs=ufs
mmap(0xDF7B0000, 4096, PROT_READ|PROT_EXEC, MAP_PRIVATE|MAP_FIXED, 4, 0) = 0xDF7B0000
close(4)                                        = 0
open("/usr/platform/SUNW,SPARCserver-1000/lib/libc_psr.so.1", O_RDONLY) Err#2 ENOENT
close(3)                                        = 0
pathconf(".", _PC_PATH_MAX)                     = 1024
stat64("./", 0xDFFFFB90)                        = 0
    d=0x02B000BC i=820082 m=0040755 l=7  u=24241 g=1047  sz=4096
        at = Mar 26 08:48:55 CST 2001  [ 985618135 ]
        mt = Mar 26 08:48:55 CST 2001  [ 985618135 ]
        ct = Mar 26 08:48:55 CST 2001  [ 985618135 ]
    bsz=8192  blks=8     fs=nfs
brk(0x00021178)                                 = 0
brk(0x00023178)                                 = 0
stat64("/", 0xDFFFFAF8)                         = 0
    d=0x00800000 i=2     m=0040755 l=33 u=0     g=0     sz=1024
        at = Mar 25 17:02:06 CST 2001  [ 985561326 ]
        mt = Jan 24 13:41:35 CST 2001  [ 980365295 ]
        ct = Jan 24 13:41:35 CST 2001  [ 980365295 ]
    bsz=8192  blks=2     fs=ufs
open64("./../", O_RDONLY|O_NDELAY)              = 3
fcntl(3, F_SETFD, 0x00000001)                   = 0
fstat64(3, 0xDFFFF198)                          = 0
    d=0x02B000BC i=180225 m=0041777 l=12 u=0     g=0     sz=4096
        at = Mar 26 08:51:23 CST 2001  [ 985618283 ]
        mt = Mar 26 08:48:55 CST 2001  [ 985618135 ]
        ct = Mar 26 08:48:55 CST 2001  [ 985618135 ]
    bsz=8192  blks=8     fs=nfs
fstat64(3, 0xDFFFFB90)                          = 0
    d=0x02B000BC i=180225 m=0041777 l=12 u=0     g=0     sz=4096
        at = Mar 26 08:51:23 CST 2001  [ 985618283 ]
        mt = Mar 26 08:48:55 CST 2001  [ 985618135 ]
        ct = Mar 26 08:48:55 CST 2001  [ 985618135 ]
    bsz=8192  blks=8     fs=nfs
getdents64(3, 0x00021998, 1048)                 = 552
    ino=180225 off=12   rlen=24  "."
    ino=2     off=24   rlen=24  ".."
    ino=921011 off=44   rlen=32  ".font-unix"
    ino=804348 off=56   rlen=24  "x"
    ino=182682 off=80   rlen=32  "outputfile"
    ino=99934 off=104  rlen=32  ".sawmill-root"
    ino=182685 off=124  rlen=32  "rpctorture.c"
    ino=181695 off=140  rlen=24  "disk"
    ino=557889 off=160  rlen=32  ".ICE-unix"
    ino=885242 off=180  rlen=32  "orbit-root"
    ino=99775 off=200  rlen=32  "SANavigator"
    ino=607263 off=228  rlen=24  "smbfs"
    ino=869460 off=244  rlen=24  "test"
    ino=67020 off=256  rlen=24  "temp"
    ino=820082 off=384  rlen=32  "cthon00"
    ino=182683 off=404  rlen=32  "file_13.out"
    ino=182686 off=424  rlen=32  "includes.h"
    ino=183530 off=4096 rlen=64  "samba-debugtools-2.0.4b-19990519.i386.rpm"
close(3)                                        = 0
open64("./../../", O_RDONLY|O_NDELAY)           = 3
fcntl(3, F_SETFD, 0x00000001)                   = 0
fstat64(3, 0xDFFFF198)                          = 0
    d=0x00000001 i=503623994 m=0041777 l=20 u=3     g=3     sz=1030
        at = Mar 26 09:48:27 CST 2001  [ 985621707 ]
        mt = Mar 26 13:48:53 CST 2001  [ 985636133 ]
        ct = Mar 26 13:48:53 CST 2001  [ 985636133 ]
    bsz=4096  blks=8     fs=tmpfs
fstat64(3, 0xDFFFFB90)                          = 0
    d=0x00000001 i=503623994 m=0041777 l=20 u=3     g=3     sz=1030
        at = Mar 26 09:48:27 CST 2001  [ 985621707 ]
        mt = Mar 26 13:48:53 CST 2001  [ 985636133 ]
        ct = Mar 26 13:48:53 CST 2001  [ 985636133 ]
    bsz=4096  blks=8     fs=tmpfs
open("/etc/mnttab", O_RDONLY)                   = 4
fstat64(4, 0xDFFFEB50)                          = 0
    d=0x00800000 i=36051 m=0100644 l=1  u=0     g=3     sz=6434
        at = Mar 26 13:50:04 CST 2001  [ 985636204 ]
        mt = Mar 26 13:49:34 CST 2001  [ 985636174 ]
        ct = Mar 26 13:49:34 CST 2001  [ 985636174 ]
    bsz=8192  blks=14    fs=ufs
brk(0x00023178)                                 = 0
brk(0x00025178)                                 = 0
ioctl(4, TCGETA, 0xDFFFEADC)                    Err#25 ENOTTY
read(4, " / d e v / d s k / c 0 t".., 8192)     = 6434
lstat64("/opt/dcelocal/var/dfs/aggrs/lfs1", 0xDFFFEC78) = 0
lstat64("/opt/dcelocal/var/dfs/aggrs/lfs.root", 0xDFFFEC78) = 0
lstat64("/...", 0xDFFFEC78)                     = 0
lstat64("/tmp/tarpon_ext2_v3", 0xDFFFED78)      = 0
lstat64("/tmp/tarpon_ext2_v3/..", 0xDFFFED78)   = 0
llseek(4, 0, SEEK_CUR)                          = 6434
close(4)                                        = 0
close(3)                                        = 0
/tmp/tarpon_ext2_v3/cthon00
write(1, " / t m p / t a r p o n _".., 28)      = 28
llseek(0, 0, SEEK_CUR)                          = 5053
_exit(0)
latte 28% cd /tmp/tarpon_v3/cthon00
latte 29% truss -v stat64,getdents64,fstat64 /usr/bin/pwd
execve("/usr/bin/pwd", 0xDFFFFCC8, 0xDFFFFCD0)  argc = 1
open("/dev/zero", O_RDONLY)                     = 3
mmap(0x00000000, 4096, PROT_READ|PROT_WRITE|PROT_EXEC, MAP_PRIVATE, 3, 0) = 0xDF7C0000
open("/usr/lib/libc.so.1", O_RDONLY)            = 4
fstat(4, 0xDFFFF884)                            = 0
    d=0x00800006 i=2314  m=0100755 l=1  u=2     g=2     sz=1013292
        at = Mar 26 13:53:16 CST 2001  [ 985636396 ]
        mt = Apr 21 03:31:13 CDT 1999  [ 924683473 ]
        ct = Apr 28 15:29:42 CDT 1999  [ 925331382 ]
    bsz=8192  blks=2000  fs=ufs
mmap(0x00000000, 4096, PROT_READ|PROT_EXEC, MAP_PRIVATE, 4, 0) = 0xDF7B0000
mmap(0x00000000, 700416, PROT_READ|PROT_EXEC, MAP_PRIVATE, 4, 0) = 0xDF700000
munmap(0xDF793000, 61440)                       = 0
mmap(0xDF7A2000, 27744, PROT_READ|PROT_WRITE|PROT_EXEC, MAP_PRIVATE|MAP_FIXED, 4, 598016) = 0xDF7A2000
mmap(0xDF7A9000, 6000, PROT_READ|PROT_WRITE|PROT_EXEC, MAP_PRIVATE|MAP_FIXED, 3, 0) = 0xDF7A9000
close(4)                                        = 0
open("/usr/lib/libdl.so.1", O_RDONLY)           = 4
fstat(4, 0xDFFFF884)                            = 0
    d=0x00800006 i=2321  m=0100755 l=1  u=2     g=2     sz=4280
        at = Mar 26 13:53:16 CST 2001  [ 985636396 ]
        mt = Apr 21 03:31:41 CDT 1999  [ 924683501 ]
        ct = Apr 28 15:44:59 CDT 1999  [ 925332299 ]
    bsz=8192  blks=10    fs=ufs
mmap(0xDF7B0000, 4096, PROT_READ|PROT_EXEC, MAP_PRIVATE|MAP_FIXED, 4, 0) = 0xDF7B0000
close(4)                                        = 0
open("/usr/platform/SUNW,SPARCserver-1000/lib/libc_psr.so.1", O_RDONLY) Err#2 ENOENT
close(3)                                        = 0
pathconf(".", _PC_PATH_MAX)                     = 1024
stat64("./", 0xDFFFFB90)                        = 0
    d=0x02B000AC i=4194432 m=0040755 l=7  u=0     g=0     sz=4096
        at = Mar 26 05:55:34 CST 2001  [ 985607734 ]
        mt = Mar 26 04:48:05 CST 2001  [ 985603685 ]
        ct = Mar 26 04:48:05 CST 2001  [ 985603685 ]
    bsz=8192  blks=8     fs=nfs
brk(0x00021178)                                 = 0
brk(0x00023178)                                 = 0
stat64("/", 0xDFFFFAF8)                         = 0
    d=0x00800000 i=2     m=0040755 l=33 u=0     g=0     sz=1024
        at = Mar 25 17:02:06 CST 2001  [ 985561326 ]
        mt = Jan 24 13:41:35 CST 2001  [ 980365295 ]
        ct = Jan 24 13:41:35 CST 2001  [ 980365295 ]
    bsz=8192  blks=2     fs=ufs
open64("./../", O_RDONLY|O_NDELAY)              = 3
fcntl(3, F_SETFD, 0x00000001)                   = 0
fstat64(3, 0xDFFFF198)                          = 0
    d=0x02B000AC i=128   m=0040777 l=7  u=0     g=0     sz=4096
        at = Mar 26 08:30:22 CST 2001  [ 985617022 ]
        mt = Mar 26 05:57:36 CST 2001  [ 985607856 ]
        ct = Mar 26 05:57:36 CST 2001  [ 985607856 ]
    bsz=8192  blks=8     fs=nfs
fstat64(3, 0xDFFFFB90)                          = 0
    d=0x02B000AC i=128   m=0040777 l=7  u=0     g=0     sz=4096
        at = Mar 26 08:30:22 CST 2001  [ 985617022 ]
        mt = Mar 26 05:57:36 CST 2001  [ 985607856 ]
        ct = Mar 26 05:57:36 CST 2001  [ 985607856 ]
    bsz=8192  blks=8     fs=nfs
getdents64(3, 0x00021998, 1048)                 = 1040
    ino=128   off=4    rlen=48  "."
    ino=128   off=9    rlen=48  ".."
    ino=132   off=11   rlen=48  "junk"
    ino=131   off=17   rlen=56  "file.66748"
    ino=134   off=20   rlen=56  "nfstest.66964"
    ino=135   off=26   rlen=56  "tmpfile66978"
    ino=4194432 off=33   rlen=56  "cthon00"
    ino=12583139 off=35   rlen=48  "xxxxx"
    ino=136   off=39   rlen=64  "testfile.67051"
    ino=133   off=51   rlen=56  "file.67110"
    ino=191   off=57   rlen=56  "nfstest.67341"
    ino=65888 off=70   rlen=56  "tmpfile67348"
    ino=16797312 off=75   rlen=48  "xtbd1"
    ino=190   off=85   rlen=64  "testfile.67395"
    ino=65889 off=94   rlen=56  "file.67506"
    ino=65891 off=97   rlen=56  "nfstest.68109"
    ino=65892 off=100  rlen=56  "tmpfile68113"
    ino=65893 off=106  rlen=56  "file.68165"
    ino=65894 off=109  rlen=56  "nfstest.68780"
close(3)                                        = 0
open64("./../../", O_RDONLY|O_NDELAY)           = 3
fcntl(3, F_SETFD, 0x00000001)                   = 0
fstat64(3, 0xDFFFF198)                          = 0
    d=0x02B000AC i=128   m=0040777 l=7  u=0     g=0     sz=4096
        at = Mar 26 08:30:22 CST 2001  [ 985617022 ]
        mt = Mar 26 05:57:36 CST 2001  [ 985607856 ]
        ct = Mar 26 05:57:36 CST 2001  [ 985607856 ]
    bsz=8192  blks=8     fs=nfs
fstat64(3, 0xDFFFFB90)                          = 0
    d=0x02B000AC i=128   m=0040777 l=7  u=0     g=0     sz=4096
        at = Mar 26 08:30:22 CST 2001  [ 985617022 ]
        mt = Mar 26 05:57:36 CST 2001  [ 985607856 ]
        ct = Mar 26 05:57:36 CST 2001  [ 985607856 ]
    bsz=8192  blks=8     fs=nfs
close(3)                                        = 0
/cthon00
write(1, " / c t h o n 0 0\n", 9)               = 9
llseek(0, 0, SEEK_CUR)                          = 9391
_exit(0)

From owner-linux-xfs@oss.sgi.com Mon Mar 26 13:13:24 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2QLDOu04085
	for linux-xfs-outgoing; Mon, 26 Mar 2001 13:13:24 -0800
Received: from sgi.com (sgi.SGI.COM [192.48.153.1])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2QLDEM04079
	for <linux-xfs@oss.sgi.com>; Mon, 26 Mar 2001 13:13:14 -0800
Received: from cthulhu.engr.sgi.com (cthulhu.engr.sgi.com [192.26.80.2]) 
	by sgi.com (980327.SGI.8.8.8-aspam/980304.SGI-aspam:
       SGI does not authorize the use of its proprietary
       systems or networks for unsolicited or bulk email
       from the Internet.) 
	via ESMTP id NAA09687
	for <linux-xfs@oss.sgi.com>; Mon, 26 Mar 2001 13:13:12 -0800 (PST)
	mail_from (tduffy@engr.sgi.com)
Received: from dbear.engr.sgi.com (dbear.engr.sgi.com [163.154.18.85])
	by cthulhu.engr.sgi.com (SGI-8.9.3/8.9.3) with ESMTP id NAA54628;
	Mon, 26 Mar 2001 13:11:54 -0800 (PST)
Date: Mon, 26 Mar 2001 13:07:49 -0800 (PST)
From: Tom Duffy <tduffy@engr.sgi.com>
To: Eric Sandeen <sandeen@sgi.com>
cc: <linux-xfs@oss.sgi.com>
Subject: Re: Moving IRIX disk to Linux
In-Reply-To: <3ABF9754.13107384@sgi.com>
Message-ID: <Pine.LNX.4.30.0103261306530.19348-100000@dbear.engr.sgi.com>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

the LBS kernel's have this turned on in their configs.  btw, we should use
these rpms for the next spin.  they are finally working...

and all the xscsi stuff that cannot be GPL'ed has been pulled out to a
separate rpm.  so, these images are clean to go out to the world.

-tduffy

On Mon, 26 Mar 2001, Eric Sandeen wrote:

> Dave Sill wrote:
> >
> > [Please reply only to the list.]
> >
> > Steve Lord <lord@sgi.com> wrote:
> >
> > >Maybe you need to add irix partition table support - I am not sure this
> > >was done in the pre-release kernels.
>
> It is in fact turned off in the kernel RPMS... we'll turn it on in the
> next spin.
>
> > OK, how would I go about doing that?
>
> Go to /usr/src/linux and run make menuconfig, or make xconfig, and
> configure your kernel with Advanced Partition Selection and SGI
> partition support turned on... then recompile your kernel & install.
> (http://www.linuxdoc.org/HOWTO/Kernel-HOWTO.html)
>
> Of course you'll have to go through all the other config options as
> well... I can send you a .config to start with if you'd like.
>
> -Eric
>


From owner-linux-xfs@oss.sgi.com Mon Mar 26 13:51:57 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2QLpvL04921
	for linux-xfs-outgoing; Mon, 26 Mar 2001 13:51:57 -0800
Received: from deliverator.sgi.com (deliverator.sgi.com [204.94.214.10])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2QLptM04918
	for <linux-xfs@oss.sgi.com>; Mon, 26 Mar 2001 13:51:56 -0800
Received: from ledzep.americas.sgi.com (ledzep.americas.sgi.com [137.38.226.97]) by deliverator.sgi.com (980309.SGI.8.8.8-aspam-6.2/980310.SGI-aspam) via ESMTP id NAA11732
	for <linux-xfs@oss.sgi.com>; Mon, 26 Mar 2001 13:50:41 -0800 (PST)
	mail_from (cattelan@thebarn.com)
Received: from gibble.americas.sgi.com (gibble.americas.sgi.com [128.162.195.80]) by ledzep.americas.sgi.com (SGI-SGI-8.9.3/americas-smart-nospam1.1) with ESMTP id PAA78932; Mon, 26 Mar 2001 15:50:38 -0600 (CST)
Received: from thebarn.com (localhost [127.0.0.1])
	by gibble.americas.sgi.com (8.11.2/8.11.2) with ESMTP id f2QLeUe23589;
	Mon, 26 Mar 2001 16:40:30 -0500
Message-ID: <3ABFB74D.3A24ED@thebarn.com>
Date: Mon, 26 Mar 2001 16:40:29 -0500
From: Russell Cattelan <cattelan@thebarn.com>
X-Mailer: Mozilla 4.76 [en] (X11; U; Linux 2.4.2-XFS i686)
X-Accept-Language: en
MIME-Version: 1.0
To: Tom Duffy <tduffy@engr.sgi.com>
CC: Eric Sandeen <sandeen@sgi.com>, linux-xfs@oss.sgi.com
Subject: Re: Moving IRIX disk to Linux
References: <Pine.LNX.4.30.0103261306530.19348-100000@dbear.engr.sgi.com>
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

Tom Duffy wrote:

> the LBS kernel's have this turned on in their configs.  btw, we should use
> these rpms for the next spin.  they are finally working...

Hmm sorry no dice:


plusb.c:985: warning: initialization from incompatible pointer type
qlogicfc.c: In function `isp2x00_detect':
qlogicfc.c:734: warning: `hostdata' might be used uninitialized in this
function
qlogicfc.c:768: warning: `s' might be used uninitialized in this function
qlogicfc.c:768: warning: `s' might be used uninitialized in this function
qlogicfc.c:768: warning: `s' might be used uninitialized in this function
qlogicfc.c:768: warning: `s' might be used uninitialized in this function
NCR5380.c:795: warning: `NCR5380_print_options' defined but not used
uhci.c:2848: warning: `uhci_pci_suspend' defined but not used
uhci.c:2853: warning: `uhci_pci_resume' defined but not used
aicasm_gram.y:1443: warning: type mismatch with previous implicit declaration
/usr/lib/bison.simple:643: warning: previous implicit declaration of `yyerror'
aicasm_gram.y:1443: warning: `yyerror' was previously implicitly declared to
ret
urn `int'
In file included from aic7xxx_osm.h:347,
                 from aic7xxx_linux.c:131:
aic7xxx.h:40: aic7xxx_reg.h: No such file or directory
aic7xxx_osm.h: In function `ahc_flush_device_writes':
In file included from aic7xxx_linux.c:131:
aic7xxx_osm.h:893: `INTSTAT' undeclared (first use in this function)
aic7xxx_osm.h:893: (Each undeclared identifier is reported only once
aic7xxx_osm.h:893: for each function it appears in.)
aic7xxx_inline.h: In function `ahc_pause_bug_fix':
In file included from aic7xxx_linux.c:132:
aic7xxx_inline.h:57: `CCSCBCTL' undeclared (first use in this function)
aic7xxx_inline.h: In function `sequencer_paused':
aic7xxx_inline.h:67: `HCNTRL' undeclared (first use in this function)
aic7xxx_inline.h:67: `PAUSE' undeclared (first use in this function)
aic7xxx_inline.h:68: warning: control reaches end of non-void function
aic7xxx_inline.h: In function `pause_sequencer':
aic7xxx_inline.h:80: `HCNTRL' undeclared (first use in this function)
aic7xxx_inline.h: In function `unpause_sequencer':
aic7xxx_inline.h:105: `INTSTAT' undeclared (first use in this function)
aic7xxx_inline.h:105: `SCSIINT' undeclared (first use in this function)
aic7xxx_inline.h:105: `SEQINT' undeclared (first use in this function)
aic7xxx_inline.h:105: `BRKADRINT' undeclared (first use in this function)
aic7xxx_inline.h:106: `HCNTRL' undeclared (first use in this function)
aic7xxx_inline.h: In function `ahc_check_residual':
aic7xxx_inline.h:220: `SG_RESID_VALID' undeclared (first use in this function)
aic7xxx_inline.h: In function `ahc_queue_scb':
aic7xxx_inline.h:336: `SCB_LIST_NULL' undeclared (first use in this function)
aic7xxx_inline.h:345: `HNSCB_QOFF' undeclared (first use in this function)
aic7xxx_inline.h:349: `KERNEL_QINPOS' undeclared (first use in this function)
aic7xxx_inline.h: In function `ahc_check_cmdcmpltqueues':
aic7xxx_inline.h:390: `SCB_LIST_NULL' undeclared (first use in this function)
aic7xxx_inline.h: In function `ahc_intr':
aic7xxx_inline.h:417: `CMDCMPLT' undeclared (first use in this function)
aic7xxx_inline.h:419: `INT_PEND' undeclared (first use in this function)
aic7xxx_inline.h:422: `INTSTAT' undeclared (first use in this function)
aic7xxx_inline.h:426: `ERROR' undeclared (first use in this function)
aic7xxx_inline.h:426: `PCIERRSTAT' undeclared (first use in this function)
aic7xxx_inline.h:442: `CLRINT' undeclared (first use in this function)
aic7xxx_inline.h:442: `CLRCMDINT' undeclared (first use in this function)
aic7xxx_inline.h:462: `BRKADRINT' undeclared (first use in this function)
aic7xxx_inline.h:468: `SEQINT' undeclared (first use in this function)
aic7xxx_inline.h:468: `SCSIINT' undeclared (first use in this function)
aic7xxx_linux.c: In function `ahc_print_path':
aic7xxx_linux.c:286: `TWIN_CHNLB' undeclared (first use in this function)
aic7xxx_linux.c:287: `TWIN_TID' undeclared (first use in this function)
aic7xxx_linux.c:287: `TID' undeclared (first use in this function)
aic7xxx_linux.c:287: `TID_SHIFT' undeclared (first use in this function)
aic7xxx_linux.c: In function `ahc_platform_freeze_devq':
aic7xxx_linux.c:1246: `TWIN_TID' undeclared (first use in this function)
aic7xxx_linux.c:1246: `TID' undeclared (first use in this function)
aic7xxx_linux.c:1246: `TID_SHIFT' undeclared (first use in this function)
aic7xxx_linux.c:1247: `TWIN_CHNLB' undeclared (first use in this function)
aic7xxx_linux.c:1248: `SCB_LIST_NULL' undeclared (first use in this function)
aic7xxx_linux.c: In function `ahc_platform_abort_scbs':
aic7xxx_linux.c:1294: `SCB_LIST_NULL' undeclared (first use in this function)
aic7xxx_linux.c: In function `ahc_run_device_queue':
aic7xxx_linux.c:1518: `TID_SHIFT' undeclared (first use in this function)
aic7xxx_linux.c:1518: `TID' undeclared (first use in this function)
aic7xxx_linux.c:1518: `TWIN_CHNLB' undeclared (first use in this function)
aic7xxx_linux.c:1520: `TWIN_TID' undeclared (first use in this function)
aic7xxx_linux.c:1522: `OID' undeclared (first use in this function)
aic7xxx_linux.c:1527: `ULTRAENB' undeclared (first use in this function)
aic7xxx_linux.c:1530: `DISCENB' undeclared (first use in this function)
aic7xxx_linux.c:1577: `SG_FULL_RESID' undeclared (first use in this function)
aic7xxx_linux.c:1624: `SG_LIST_NULL' undeclared (first use in this function)
aic7xxx_linux.c:1641: `TARGET_SCB' undeclared (first use in this function)
aic7xxx_linux.c:1641: `TAG_ENB' undeclared (first use in this function)
aic7xxx_linux.c:1475: warning: `mask' might be used uninitialized in this
functi
on
aic7xxx_linux.c:1644: warning: `target_offset' might be used uninitialized in
th
is function
aic7xxx_linux.c: In function `ahc_done':
aic7xxx_linux.c:1895: `TWIN_TID' undeclared (first use in this function)
aic7xxx_linux.c:1895: `TID' undeclared (first use in this function)
aic7xxx_linux.c:1895: `TID_SHIFT' undeclared (first use in this function)
aic7xxx_linux.c:1895: `TWIN_CHNLB' undeclared (first use in this function)
aic7xxx_linux.c:1893: warning: `target_offset' might be used uninitialized in
th
is function
aic7xxx_linux.c: In function `ahc_filter_command':
aic7xxx_linux.c:2094: `TID_SHIFT' undeclared (first use in this function)
aic7xxx_linux.c:2094: `TID' undeclared (first use in this function)
aic7xxx_linux.c:2094: `TWIN_CHNLB' undeclared (first use in this function)
aic7xxx_linux.c:2095: `OID' undeclared (first use in this function)
aic7xxx_linux.c: In function `aic7xxx_queue_recovery_cmd':
aic7xxx_linux.c:2288: `SCB_LIST_NULL' undeclared (first use in this function)
aic7xxx_linux.c:2318: `INTSTAT' undeclared (first use in this function)
aic7xxx_linux.c:2318: `INT_PEND' undeclared (first use in this function)
aic7xxx_linux.c:2358: `LASTPHASE' undeclared (first use in this function)
aic7xxx_linux.c:2359: `SCBPTR' undeclared (first use in this function)
aic7xxx_linux.c:2360: `SCB_TAG' undeclared (first use in this function)
aic7xxx_linux.c:2362: `P_BUSFREE' undeclared (first use in this function)
aic7xxx_linux.c:2364: `SAVED_SCSIID' undeclared (first use in this function)
aic7xxx_linux.c:2364: `TWIN_TID' undeclared (first use in this function)
aic7xxx_linux.c:2364: `TID' undeclared (first use in this function)
aic7xxx_linux.c:2364: `TID_SHIFT' undeclared (first use in this function)
aic7xxx_linux.c:2372: `MSG_OUT' undeclared (first use in this function)
aic7xxx_linux.c:2372: `HOST_MSG' undeclared (first use in this function)
aic7xxx_linux.c:2373: `SCSISIGO' undeclared (first use in this function)
aic7xxx_linux.c:2373: `ATNO' undeclared (first use in this function)
aic7xxx_linux.c:2395: `MK_MESSAGE' undeclared (first use in this function)
aic7xxx_linux.c:2395: `DISCONNECTED' undeclared (first use in this function)
aic7xxx_linux.c:2420: `SCB_CONTROL' undeclared (first use in this function)
make[3]: *** [aic7xxx_linux.o] Error 1
make[2]: *** [_modsubdir_aic7xxx] Error 2
make[2]: *** Waiting for unfinished jobs....
make[1]: *** [_modsubdir_scsi] Error 2
make: *** [_mod_drivers] Error 2
error: Bad exit status from /var/tmp/rpm-tmp.27602 (%build)


RPM build errors:
    Bad exit status from /var/tmp/rpm-tmp.27602 (%build)
Mon Mar 26 15:46:12 CST 2001


>
>
> and all the xscsi stuff that cannot be GPL'ed has been pulled out to a
> separate rpm.  so, these images are clean to go out to the world.
>
> -tduffy
>
> On Mon, 26 Mar 2001, Eric Sandeen wrote:
>
> > Dave Sill wrote:
> > >
> > > [Please reply only to the list.]
> > >
> > > Steve Lord <lord@sgi.com> wrote:
> > >
> > > >Maybe you need to add irix partition table support - I am not sure this
> > > >was done in the pre-release kernels.
> >
> > It is in fact turned off in the kernel RPMS... we'll turn it on in the
> > next spin.
> >
> > > OK, how would I go about doing that?
> >
> > Go to /usr/src/linux and run make menuconfig, or make xconfig, and
> > configure your kernel with Advanced Partition Selection and SGI
> > partition support turned on... then recompile your kernel & install.
> > (http://www.linuxdoc.org/HOWTO/Kernel-HOWTO.html)
> >
> > Of course you'll have to go through all the other config options as
> > well... I can send you a .config to start with if you'd like.
> >
> > -Eric
> >

--
Russell Cattelan
--
Digital Elves inc. -- Currently on loan to SGI
Linux XFS core developer.




From owner-linux-xfs@oss.sgi.com Mon Mar 26 13:59:03 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2QLx3f05251
	for linux-xfs-outgoing; Mon, 26 Mar 2001 13:59:03 -0800
Received: from deliverator.sgi.com (deliverator.sgi.com [204.94.214.10])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2QLx3M05248
	for <linux-xfs@oss.sgi.com>; Mon, 26 Mar 2001 13:59:03 -0800
Received: from zeus-fddi.americas.sgi.com (zeus-fddi.americas.sgi.com [128.162.8.103]) by deliverator.sgi.com (980309.SGI.8.8.8-aspam-6.2/980310.SGI-aspam) via ESMTP id NAA12745
	for <linux-xfs@oss.sgi.com>; Mon, 26 Mar 2001 13:57:49 -0800 (PST)
	mail_from (lord@sgi.com)
Received: from daisy-e185.americas.sgi.com (daisy-e194.americas.sgi.com [128.162.194.214]) by zeus-fddi.americas.sgi.com (8.9.3/americas-smart-nospam1.1) with ESMTP id PAA1269051 for <linux-xfs@oss.sgi.com>; Mon, 26 Mar 2001 15:57:46 -0600 (CST)
Received: from jen.americas.sgi.com (IDENT:root@jen.americas.sgi.com [128.162.184.86]) by daisy-e185.americas.sgi.com (SGI-8.9.3/SGI-server-1.7) with ESMTP id PAA11535 for <linux-xfs@oss.sgi.com>; Mon, 26 Mar 2001 15:57:46 -0600 (CST)
From: Steve Lord <lord@sgi.com>
Received: by jen.americas.sgi.com (8.11.0/SGI-client-1.7) id f2QLvEF31599; Mon, 26 Mar 2001 15:57:14 -0600
Message-Id: <200103262157.f2QLvEF31599@jen.americas.sgi.com>
Date: Mon, 26 Mar 2001 15:57:14 -0600
Subject: TAKE - more of the xfs_iget fix
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

I missed a case....

Date:  Mon Mar 26 13:57:23 PST 2001
Workarea:  jen.americas.sgi.com:/src/lord/xfs-linux.2.4

The following file(s) were checked into:
  bonnie.engr.sgi.com:/isms/slinx/2.4.x-xfs


Modid:  2.4.x-xfs:slinx:90877a
linux/fs/xfs/xfs_iget.c - 1.133
	- Fix the previous fix - forgot about a couple of cases, aim7 is now
	  up to a load of 300 on my puny little test box.



From owner-linux-xfs@oss.sgi.com Mon Mar 26 14:10:36 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2QMAau05804
	for linux-xfs-outgoing; Mon, 26 Mar 2001 14:10:36 -0800
Received: from deliverator.sgi.com (deliverator.sgi.com [204.94.214.10])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2QMAZM05801
	for <linux-xfs@oss.sgi.com>; Mon, 26 Mar 2001 14:10:35 -0800
Received: from cthulhu.engr.sgi.com (gate3-relay.engr.sgi.com [130.62.1.234]) by deliverator.sgi.com (980309.SGI.8.8.8-aspam-6.2/980310.SGI-aspam) via ESMTP id OAA15478
	for <linux-xfs@oss.sgi.com>; Mon, 26 Mar 2001 14:09:22 -0800 (PST)
	mail_from (tduffy@engr.sgi.com)
Received: from dbear.engr.sgi.com (dbear.engr.sgi.com [163.154.18.85])
	by cthulhu.engr.sgi.com (SGI-8.9.3/8.9.3) with ESMTP id OAA93648;
	Mon, 26 Mar 2001 14:09:06 -0800 (PST)
Date: Mon, 26 Mar 2001 14:05:01 -0800 (PST)
From: Tom Duffy <tduffy@engr.sgi.com>
To: Russell Cattelan <cattelan@thebarn.com>
cc: Eric Sandeen <sandeen@sgi.com>, <linux-xfs@oss.sgi.com>
Subject: Re: Moving IRIX disk to Linux
In-Reply-To: <3ABFB74D.3A24ED@thebarn.com>
Message-ID: <Pine.LNX.4.30.0103261404180.19348-100000@dbear.engr.sgi.com>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

this is a problem with parallell building...try setting NPROC to 1

-tduffy

On Mon, 26 Mar 2001, Russell Cattelan wrote:

> Tom Duffy wrote:
>
> > the LBS kernel's have this turned on in their configs.  btw, we should use
> > these rpms for the next spin.  they are finally working...
>
> Hmm sorry no dice:
>
>
> plusb.c:985: warning: initialization from incompatible pointer type
> qlogicfc.c: In function `isp2x00_detect':
> qlogicfc.c:734: warning: `hostdata' might be used uninitialized in this
> function
> qlogicfc.c:768: warning: `s' might be used uninitialized in this function
> qlogicfc.c:768: warning: `s' might be used uninitialized in this function
> qlogicfc.c:768: warning: `s' might be used uninitialized in this function
> qlogicfc.c:768: warning: `s' might be used uninitialized in this function
> NCR5380.c:795: warning: `NCR5380_print_options' defined but not used
> uhci.c:2848: warning: `uhci_pci_suspend' defined but not used
> uhci.c:2853: warning: `uhci_pci_resume' defined but not used
> aicasm_gram.y:1443: warning: type mismatch with previous implicit declaration
> /usr/lib/bison.simple:643: warning: previous implicit declaration of `yyerror'
> aicasm_gram.y:1443: warning: `yyerror' was previously implicitly declared to
> ret
> urn `int'
> In file included from aic7xxx_osm.h:347,
>                  from aic7xxx_linux.c:131:
> aic7xxx.h:40: aic7xxx_reg.h: No such file or directory
> aic7xxx_osm.h: In function `ahc_flush_device_writes':
> In file included from aic7xxx_linux.c:131:
> aic7xxx_osm.h:893: `INTSTAT' undeclared (first use in this function)
> aic7xxx_osm.h:893: (Each undeclared identifier is reported only once
> aic7xxx_osm.h:893: for each function it appears in.)
> aic7xxx_inline.h: In function `ahc_pause_bug_fix':
> In file included from aic7xxx_linux.c:132:
> aic7xxx_inline.h:57: `CCSCBCTL' undeclared (first use in this function)
> aic7xxx_inline.h: In function `sequencer_paused':
> aic7xxx_inline.h:67: `HCNTRL' undeclared (first use in this function)
> aic7xxx_inline.h:67: `PAUSE' undeclared (first use in this function)
> aic7xxx_inline.h:68: warning: control reaches end of non-void function
> aic7xxx_inline.h: In function `pause_sequencer':
> aic7xxx_inline.h:80: `HCNTRL' undeclared (first use in this function)
> aic7xxx_inline.h: In function `unpause_sequencer':
> aic7xxx_inline.h:105: `INTSTAT' undeclared (first use in this function)
> aic7xxx_inline.h:105: `SCSIINT' undeclared (first use in this function)
> aic7xxx_inline.h:105: `SEQINT' undeclared (first use in this function)
> aic7xxx_inline.h:105: `BRKADRINT' undeclared (first use in this function)
> aic7xxx_inline.h:106: `HCNTRL' undeclared (first use in this function)
> aic7xxx_inline.h: In function `ahc_check_residual':
> aic7xxx_inline.h:220: `SG_RESID_VALID' undeclared (first use in this function)
> aic7xxx_inline.h: In function `ahc_queue_scb':
> aic7xxx_inline.h:336: `SCB_LIST_NULL' undeclared (first use in this function)
> aic7xxx_inline.h:345: `HNSCB_QOFF' undeclared (first use in this function)
> aic7xxx_inline.h:349: `KERNEL_QINPOS' undeclared (first use in this function)
> aic7xxx_inline.h: In function `ahc_check_cmdcmpltqueues':
> aic7xxx_inline.h:390: `SCB_LIST_NULL' undeclared (first use in this function)
> aic7xxx_inline.h: In function `ahc_intr':
> aic7xxx_inline.h:417: `CMDCMPLT' undeclared (first use in this function)
> aic7xxx_inline.h:419: `INT_PEND' undeclared (first use in this function)
> aic7xxx_inline.h:422: `INTSTAT' undeclared (first use in this function)
> aic7xxx_inline.h:426: `ERROR' undeclared (first use in this function)
> aic7xxx_inline.h:426: `PCIERRSTAT' undeclared (first use in this function)
> aic7xxx_inline.h:442: `CLRINT' undeclared (first use in this function)
> aic7xxx_inline.h:442: `CLRCMDINT' undeclared (first use in this function)
> aic7xxx_inline.h:462: `BRKADRINT' undeclared (first use in this function)
> aic7xxx_inline.h:468: `SEQINT' undeclared (first use in this function)
> aic7xxx_inline.h:468: `SCSIINT' undeclared (first use in this function)
> aic7xxx_linux.c: In function `ahc_print_path':
> aic7xxx_linux.c:286: `TWIN_CHNLB' undeclared (first use in this function)
> aic7xxx_linux.c:287: `TWIN_TID' undeclared (first use in this function)
> aic7xxx_linux.c:287: `TID' undeclared (first use in this function)
> aic7xxx_linux.c:287: `TID_SHIFT' undeclared (first use in this function)
> aic7xxx_linux.c: In function `ahc_platform_freeze_devq':
> aic7xxx_linux.c:1246: `TWIN_TID' undeclared (first use in this function)
> aic7xxx_linux.c:1246: `TID' undeclared (first use in this function)
> aic7xxx_linux.c:1246: `TID_SHIFT' undeclared (first use in this function)
> aic7xxx_linux.c:1247: `TWIN_CHNLB' undeclared (first use in this function)
> aic7xxx_linux.c:1248: `SCB_LIST_NULL' undeclared (first use in this function)
> aic7xxx_linux.c: In function `ahc_platform_abort_scbs':
> aic7xxx_linux.c:1294: `SCB_LIST_NULL' undeclared (first use in this function)
> aic7xxx_linux.c: In function `ahc_run_device_queue':
> aic7xxx_linux.c:1518: `TID_SHIFT' undeclared (first use in this function)
> aic7xxx_linux.c:1518: `TID' undeclared (first use in this function)
> aic7xxx_linux.c:1518: `TWIN_CHNLB' undeclared (first use in this function)
> aic7xxx_linux.c:1520: `TWIN_TID' undeclared (first use in this function)
> aic7xxx_linux.c:1522: `OID' undeclared (first use in this function)
> aic7xxx_linux.c:1527: `ULTRAENB' undeclared (first use in this function)
> aic7xxx_linux.c:1530: `DISCENB' undeclared (first use in this function)
> aic7xxx_linux.c:1577: `SG_FULL_RESID' undeclared (first use in this function)
> aic7xxx_linux.c:1624: `SG_LIST_NULL' undeclared (first use in this function)
> aic7xxx_linux.c:1641: `TARGET_SCB' undeclared (first use in this function)
> aic7xxx_linux.c:1641: `TAG_ENB' undeclared (first use in this function)
> aic7xxx_linux.c:1475: warning: `mask' might be used uninitialized in this
> functi
> on
> aic7xxx_linux.c:1644: warning: `target_offset' might be used uninitialized in
> th
> is function
> aic7xxx_linux.c: In function `ahc_done':
> aic7xxx_linux.c:1895: `TWIN_TID' undeclared (first use in this function)
> aic7xxx_linux.c:1895: `TID' undeclared (first use in this function)
> aic7xxx_linux.c:1895: `TID_SHIFT' undeclared (first use in this function)
> aic7xxx_linux.c:1895: `TWIN_CHNLB' undeclared (first use in this function)
> aic7xxx_linux.c:1893: warning: `target_offset' might be used uninitialized in
> th
> is function
> aic7xxx_linux.c: In function `ahc_filter_command':
> aic7xxx_linux.c:2094: `TID_SHIFT' undeclared (first use in this function)
> aic7xxx_linux.c:2094: `TID' undeclared (first use in this function)
> aic7xxx_linux.c:2094: `TWIN_CHNLB' undeclared (first use in this function)
> aic7xxx_linux.c:2095: `OID' undeclared (first use in this function)
> aic7xxx_linux.c: In function `aic7xxx_queue_recovery_cmd':
> aic7xxx_linux.c:2288: `SCB_LIST_NULL' undeclared (first use in this function)
> aic7xxx_linux.c:2318: `INTSTAT' undeclared (first use in this function)
> aic7xxx_linux.c:2318: `INT_PEND' undeclared (first use in this function)
> aic7xxx_linux.c:2358: `LASTPHASE' undeclared (first use in this function)
> aic7xxx_linux.c:2359: `SCBPTR' undeclared (first use in this function)
> aic7xxx_linux.c:2360: `SCB_TAG' undeclared (first use in this function)
> aic7xxx_linux.c:2362: `P_BUSFREE' undeclared (first use in this function)
> aic7xxx_linux.c:2364: `SAVED_SCSIID' undeclared (first use in this function)
> aic7xxx_linux.c:2364: `TWIN_TID' undeclared (first use in this function)
> aic7xxx_linux.c:2364: `TID' undeclared (first use in this function)
> aic7xxx_linux.c:2364: `TID_SHIFT' undeclared (first use in this function)
> aic7xxx_linux.c:2372: `MSG_OUT' undeclared (first use in this function)
> aic7xxx_linux.c:2372: `HOST_MSG' undeclared (first use in this function)
> aic7xxx_linux.c:2373: `SCSISIGO' undeclared (first use in this function)
> aic7xxx_linux.c:2373: `ATNO' undeclared (first use in this function)
> aic7xxx_linux.c:2395: `MK_MESSAGE' undeclared (first use in this function)
> aic7xxx_linux.c:2395: `DISCONNECTED' undeclared (first use in this function)
> aic7xxx_linux.c:2420: `SCB_CONTROL' undeclared (first use in this function)
> make[3]: *** [aic7xxx_linux.o] Error 1
> make[2]: *** [_modsubdir_aic7xxx] Error 2
> make[2]: *** Waiting for unfinished jobs....
> make[1]: *** [_modsubdir_scsi] Error 2
> make: *** [_mod_drivers] Error 2
> error: Bad exit status from /var/tmp/rpm-tmp.27602 (%build)
>
>
> RPM build errors:
>     Bad exit status from /var/tmp/rpm-tmp.27602 (%build)
> Mon Mar 26 15:46:12 CST 2001
>
>
> >
> >
> > and all the xscsi stuff that cannot be GPL'ed has been pulled out to a
> > separate rpm.  so, these images are clean to go out to the world.
> >
> > -tduffy
> >
> > On Mon, 26 Mar 2001, Eric Sandeen wrote:
> >
> > > Dave Sill wrote:
> > > >
> > > > [Please reply only to the list.]
> > > >
> > > > Steve Lord <lord@sgi.com> wrote:
> > > >
> > > > >Maybe you need to add irix partition table support - I am not sure this
> > > > >was done in the pre-release kernels.
> > >
> > > It is in fact turned off in the kernel RPMS... we'll turn it on in the
> > > next spin.
> > >
> > > > OK, how would I go about doing that?
> > >
> > > Go to /usr/src/linux and run make menuconfig, or make xconfig, and
> > > configure your kernel with Advanced Partition Selection and SGI
> > > partition support turned on... then recompile your kernel & install.
> > > (http://www.linuxdoc.org/HOWTO/Kernel-HOWTO.html)
> > >
> > > Of course you'll have to go through all the other config options as
> > > well... I can send you a .config to start with if you'd like.
> > >
> > > -Eric
> > >
>
> --
> Russell Cattelan
> --
> Digital Elves inc. -- Currently on loan to SGI
> Linux XFS core developer.
>
>
>


From owner-linux-xfs@oss.sgi.com Mon Mar 26 14:14:33 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2QMEXT06087
	for linux-xfs-outgoing; Mon, 26 Mar 2001 14:14:33 -0800
Received: from yog-sothoth.sgi.com (eugate.sgi.com [192.48.160.10])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2QMEWM06084
	for <linux-xfs@oss.sgi.com>; Mon, 26 Mar 2001 14:14:32 -0800
Received: from cthulhu.engr.sgi.com (cthulhu.engr.sgi.com [192.26.80.2]) by yog-sothoth.sgi.com (980305.SGI.8.8.8-aspam-6.2/980304.SGI-aspam-europe) via ESMTP id AAA2861404
	for <linux-xfs@oss.sgi.com>; Tue, 27 Mar 2001 00:14:30 +0200 (CEST)
	mail_from (tduffy@engr.sgi.com)
Received: from dbear.engr.sgi.com (dbear.engr.sgi.com [163.154.18.85])
	by cthulhu.engr.sgi.com (SGI-8.9.3/8.9.3) with ESMTP id OAA33749;
	Mon, 26 Mar 2001 14:13:12 -0800 (PST)
Date: Mon, 26 Mar 2001 14:09:07 -0800 (PST)
From: Tom Duffy <tduffy@engr.sgi.com>
To: Russell Cattelan <cattelan@thebarn.com>
cc: Eric Sandeen <sandeen@sgi.com>, <linux-xfs@oss.sgi.com>
Subject: Re: Moving IRIX disk to Linux
In-Reply-To: <Pine.LNX.4.30.0103261404180.19348-100000@dbear.engr.sgi.com>
Message-ID: <Pine.LNX.4.30.0103261408120.19348-100000@dbear.engr.sgi.com>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

dude...why don't you just use the kernels that jalal built on:

dist:/test/ProPack1.5-ia32/latest/RPMS

that kernel is now running on waco.engr.

-tduffy



From owner-linux-xfs@oss.sgi.com Mon Mar 26 14:50:10 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2QMoAG07851
	for linux-xfs-outgoing; Mon, 26 Mar 2001 14:50:10 -0800
Received: from groucho.maths.monash.edu.au (groucho.maths.monash.edu.au [130.194.160.211])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2QMo9M07848
	for <linux-xfs@oss.sgi.com>; Mon, 26 Mar 2001 14:50:09 -0800
Received: (from rjh@localhost)
	by groucho.maths.monash.edu.au (8.8.8/8.8.8) id WAA22086
	for linux-xfs@oss.sgi.com; Mon, 26 Mar 2001 22:50:07 GMT
From: Robin Humble <rjh@groucho.maths.monash.edu.au>
Message-Id: <200103262250.WAA22086@groucho.maths.monash.edu.au>
Subject: Re: ADD 819118 - pwd on sun NFS client does not return entire path
To: linux-xfs@oss.sgi.com
Date: Tue, 27 Mar 2001 08:50:06 +1000 (EST)
In-Reply-To:  <200103261958.LAA72437@info.engr.sgi.com> from "tbd@sgi.com" at Mar 26, 2001 11:58:47 AM
X-Mailer: ELM [version 2.5 PL1]
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk


> Status : open                         Priority : 2
> Assigned Engineer : nb                Submitter : tbd
>*Modified User : tbd                  *Modified User Domain : sgi.com
>*Description :
>pwd does not always return the entire path on an NFS mounted directory on
>a sun client (i.e. the local mount point is not included). The server

This happens with Tru64 unix as well as Solaris. ext2 behaves fine,
XFS doesn't. I didn't try NFS v2, only v3.  2.4.2-XFS cvs from mar18
+ 'wolverine'

IRIX behaves weirdly no matter what... :-/

cheers,
robin

ps. sorry - I don't know how to use the bug tracking system so I'm just
sending to the list.

From owner-linux-xfs@oss.sgi.com Mon Mar 26 15:13:24 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2QNDOa08315
	for linux-xfs-outgoing; Mon, 26 Mar 2001 15:13:24 -0800
Received: from deliverator.sgi.com (deliverator.sgi.com [204.94.214.10])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2QNDOM08312
	for <linux-xfs@oss.sgi.com>; Mon, 26 Mar 2001 15:13:24 -0800
Received: from zeus-fddi.americas.sgi.com (zeus-fddi.americas.sgi.com [128.162.8.103]) by deliverator.sgi.com (980309.SGI.8.8.8-aspam-6.2/980310.SGI-aspam) via ESMTP id PAA25651
	for <linux-xfs@oss.sgi.com>; Mon, 26 Mar 2001 15:12:10 -0800 (PST)
	mail_from (lord@sgi.com)
Received: from daisy-e185.americas.sgi.com (daisy-e194.americas.sgi.com [128.162.194.214]) by zeus-fddi.americas.sgi.com (8.9.3/americas-smart-nospam1.1) with ESMTP id RAA1267561; Mon, 26 Mar 2001 17:12:06 -0600 (CST)
Received: from jen.americas.sgi.com (IDENT:root@jen.americas.sgi.com [128.162.184.86]) by daisy-e185.americas.sgi.com (SGI-8.9.3/SGI-server-1.7) with ESMTP id RAA67945; Mon, 26 Mar 2001 17:12:06 -0600 (CST)
Received: from jen.americas.sgi.com by jen.americas.sgi.com (8.11.0/SGI-client-1.7) via ESMTP id f2QNBX802452; Mon, 26 Mar 2001 17:11:33 -0600
Message-Id: <200103262311.f2QNBX802452@jen.americas.sgi.com>
X-Mailer: exmh version 2.2 06/23/2000 with nmh-1.0.4
To: Robin Humble <rjh@groucho.maths.monash.edu.au>
cc: linux-xfs@oss.sgi.com
Subject: Re: ADD 819118 - pwd on sun NFS client does not return entire path 
In-Reply-To: Message from Robin Humble <rjh@groucho.maths.monash.edu.au> 
   of "Tue, 27 Mar 2001 08:50:06 +1000." <200103262250.WAA22086@groucho.maths.monash.edu.au> 
Date: Mon, 26 Mar 2001 17:11:33 -0600
From: Steve Lord <lord@sgi.com>
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk


> 
> This happens with Tru64 unix as well as Solaris. ext2 behaves fine,
> XFS doesn't. I didn't try NFS v2, only v3.  2.4.2-XFS cvs from mar18
> + 'wolverine'
> 
> IRIX behaves weirdly no matter what... :-/
> 
> cheers,
> robin
> 
> ps. sorry - I don't know how to use the bug tracking system so I'm just
> sending to the list.

This is related to the specific algorithm used to work out pwd, solaris
walks up the directory tree doing readdirs and stating files, it stops
when it finds the root of a filesystem. The problem is that for xfs for
some reason . and .. in the root dir are both the same inode when looked 
at via an nfs mount, for ext2 they are not. This makes the top of the xfs
filesystem look like the root of the whole machine. Even opening .. is
pushing us back into the same inode at this point. I would have thought
that traversing out of a filesystem via .. would be an nfs client only
operation.

That is what I think is happening, working out why this is happening is
another matter! 

Steve





From owner-linux-xfs@oss.sgi.com Mon Mar 26 16:05:22 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2R05Mh09500
	for linux-xfs-outgoing; Mon, 26 Mar 2001 16:05:22 -0800
Received: from sun.rhrk.uni-kl.de (sun.rhrk.uni-kl.de [131.246.137.50])
	by oss.sgi.com (8.11.3/8.11.3) with SMTP id f2R05LM09495
	for <linux-xfs@oss.sgi.com>; Mon, 26 Mar 2001 16:05:21 -0800
Received: from aixs1.rhrk.uni-kl.de
           ( exim@aixs1.rhrk.uni-kl.de [131.246.137.3] ) by sun.rhrk.uni-kl.de
          id aa04945 for <linux-xfs@oss.sgi.com>; 27 Mar 2001 01:05 MET
Received: from trip210.wohnheim.uni-kl.de ([131.246.188.10] helo=student.uni-kl.de)
	by aixs1.rhrk.uni-kl.de with esmtp (Exim 3.03 #2)
	id 14hgz7-0004TA-00
	for linux-xfs@oss.sgi.com; Tue, 27 Mar 2001 02:05:17 +0200
Message-ID: <3ABFD93E.373322A6@student.uni-kl.de>
Date: Tue, 27 Mar 2001 02:05:18 +0200
From: "R. Sinoradzki" <sinoradz@student.uni-kl.de>
X-Mailer: Mozilla 4.76 [en] (X11; U; Linux 2.4.1-XFS i686)
X-Accept-Language: en
MIME-Version: 1.0
To: linux-xfs@oss.sgi.com
Subject: linker error on Debian/unstable
Content-Type: text/plain; charset=iso-8859-2
Content-Transfer-Encoding: 7bit
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

Hi,

  I have compiled the source on my Debian/i686 box and had some
problems with 'ld'. The linking failed.

There are some lines in ./arch/i386/boot/Makefile like

bootsect: bootsect.o
        $(LD) -Ttext 0x0 -s -oformat binary -o $@ $<
			    ^^^^^^^^
I use GNU ld version 2.11.90.0.1 (with BFD 2.11.90.0.1)
I think ld option '-oformat' has changed to '--oformat'.
After changing this I had no problems to compile the source.
So perhaps this helps somebody with the same problem ...

bye Ralf

From owner-linux-xfs@oss.sgi.com Mon Mar 26 16:11:04 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2R0B4d09677
	for linux-xfs-outgoing; Mon, 26 Mar 2001 16:11:04 -0800
Received: from sgi.com (sgi.SGI.COM [192.48.153.1])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2R0B3M09673
	for <linux-xfs@oss.sgi.com>; Mon, 26 Mar 2001 16:11:03 -0800
Received: from larry.melbourne.sgi.com ([134.14.52.130]) 
	by sgi.com (980327.SGI.8.8.8-aspam/980304.SGI-aspam:
       SGI does not authorize the use of its proprietary
       systems or networks for unsolicited or bulk email
       from the Internet.) 
	via SMTP id QAA09516
	for <linux-xfs@oss.sgi.com>; Mon, 26 Mar 2001 16:10:23 -0800 (PST)
	mail_from (nathans@wobbly.melbourne.sgi.com)
Received: from wobbly.melbourne.sgi.com (wobbly.melbourne.sgi.com [134.14.55.135]) by larry.melbourne.sgi.com (950413.SGI.8.6.12/950213.SGI.AUTOCF) via ESMTP id KAA22644; Tue, 27 Mar 2001 10:08:54 +1000
Received: (from nathans@localhost) by wobbly.melbourne.sgi.com (980427.SGI.8.8.8/980728.SGI.AUTOCF) id KAA58004; Tue, 27 Mar 2001 10:08:44 +1000 (EST)
From: "Nathan Scott" <nathans@wobbly.melbourne.sgi.com>
Message-Id: <10103271008.ZM246419@wobbly.melbourne.sgi.com>
Date: Tue, 27 Mar 2001 10:08:42 -0500
In-Reply-To: Steve Lord <lord@sgi.com>
        "Re: Moving IRIX disk to Linux" (Mar 26,  1:06pm)
References: <200103261906.f2QJ66t21060@jen.americas.sgi.com>
X-Mailer: Z-Mail (3.2.3 08feb96 MediaMail)
To: Dave Sill <ds-linux-xfs@sws5.ctd.ornl.gov>
Subject: Re: Moving IRIX disk to Linux
Cc: linux-xfs@oss.sgi.com
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

hi,

On Mar 26,  1:06pm, Steve Lord wrote:
> Subject: Re: Moving IRIX disk to Linux
> > ...
> > During the boot, the kernel reports:
> > 
> >   Detected scsi disk sda at scsi0, channel 0, id 1, lun 0
> >   SCSI device sda: 91923356 512-byte hdwr sectors (47065 MB)
> >   Partition check:
> >    /dev/scsi/host0/bus0/target1/lun0: unknown partition table
> > 

As Steve pointed out, your kernel doesn't support the SGI
disk partitioning scheme used in IRIX.  The other thing
worth noting is...

> > But fdisk sees the partition table OK:
> > 

fdisk knows what the ondisk format for an SGI disk label
is, so doesn't require any kernel support in order to
print the partition table.  However, ...

> >   Disk /dev/sda (SGI disk label): 28 heads, 328 sectors, 10009 cylinders
> >   Units = cylinders of 9184 * 512 bytes
> >   
> >   ----- partitions -----
> >      Device  Info      Start       End   Sectors  Id  System
> >   /dev/sda8               1     10009  91919260   a  SGI xfs
> >   /dev/sda9               0         0      4096   0  SGI volhdr
> >   /dev/sda11              0     10009  91923356   6  SGI volume
> > 
> > And the only device file associated with the disk is /dev/sda--no
> > sda8, sda9, or sda11.
> > 

Once you have built a kernel with SGI partition support,
you should refer to /proc/partitions to figure out the
device names to use when mounting the filesystem, and not
the fdisk output - the kernel uses a different partition
numbering scheme and the two do not match up (fdisk gets
it even more wrong when devfs is in use, but it is "wrong"
without devfs too).

I haven't had time to try to fix this as yet, but shouldn't
be too hard (just needs someone to understand the algorithm
used in linux/fs/partitions/sgi.c, then make fdisk use the
same algorithm, rather than using the partition IDs from
the label - that code is in the util-linux package -> eg.
util-linux-2.11/fdisk/fdisksgilabel.c).

cheers.

-- 
Nathan

From owner-linux-xfs@oss.sgi.com Mon Mar 26 16:28:17 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2R0SHU10218
	for linux-xfs-outgoing; Mon, 26 Mar 2001 16:28:17 -0800
Received: from yog-sothoth.sgi.com (eugate.sgi.com [192.48.160.10])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2R0SDM10215
	for <linux-xfs@oss.sgi.com>; Mon, 26 Mar 2001 16:28:13 -0800
Received: from sydney.sydney.sgi.com (sydney.sydney.sgi.com [134.14.48.2]) by yog-sothoth.sgi.com (980305.SGI.8.8.8-aspam-6.2/980304.SGI-aspam-europe) via SMTP id CAA6342319
	for <linux-xfs@oss.sgi.com>; Tue, 27 Mar 2001 02:28:10 +0200 (CEST)
	mail_from (kaos@melbourne.sgi.com)
Received: from kao2.melbourne.sgi.com by sydney.sydney.sgi.com via ESMTP (950413.SGI.8.6.12/930416.SGI)
	 id KAA01641; Tue, 27 Mar 2001 10:26:41 +1000
X-Mailer: exmh version 2.1.1 10/15/1999
From: Keith Owens <kaos@melbourne.sgi.com>
To: "R. Sinoradzki" <sinoradz@student.uni-kl.de>
cc: linux-xfs@oss.sgi.com
Subject: Re: linker error on Debian/unstable 
In-reply-to: Your message of "Tue, 27 Mar 2001 02:05:18 +0200."
             <3ABFD93E.373322A6@student.uni-kl.de> 
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Date: Tue, 27 Mar 2001 10:26:45 +1000
Message-ID: <7846.985652805@kao2.melbourne.sgi.com>
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

On Tue, 27 Mar 2001 02:05:18 +0200, 
"R. Sinoradzki" <sinoradz@student.uni-kl.de> wrote:
>bootsect: bootsect.o
>        $(LD) -Ttext 0x0 -s -oformat binary -o $@ $<
>			    ^^^^^^^^
>I use GNU ld version 2.11.90.0.1 (with BFD 2.11.90.0.1)
>I think ld option '-oformat' has changed to '--oformat'.

Already fixed in 2.4.3-pre8 and 2.4.2-ac25.  XFS was back level, thanks
for pointing this out.


From owner-linux-xfs@oss.sgi.com Mon Mar 26 16:32:22 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2R0WMY10262
	for linux-xfs-outgoing; Mon, 26 Mar 2001 16:32:22 -0800
Received: from deliverator.sgi.com (deliverator.sgi.com [204.94.214.10])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2R0WMM10259
	for <linux-xfs@oss.sgi.com>; Mon, 26 Mar 2001 16:32:22 -0800
Received: from sherman.melbourne.sgi.com (sherman.melbourne.sgi.com [134.14.55.175]) by deliverator.sgi.com (980309.SGI.8.8.8-aspam-6.2/980310.SGI-aspam) via ESMTP id QAA11018
	for <linux-xfs@oss.sgi.com>; Mon, 26 Mar 2001 16:30:56 -0800 (PST)
	mail_from (kaos@sherman.melbourne.sgi.com)
Received: (from kaos@localhost)
	by sherman.melbourne.sgi.com (8.9.3/8.9.3) id KAA18671;
	Tue, 27 Mar 2001 10:29:31 +1000
Date: Tue, 27 Mar 2001 10:29:31 +1000
From: Keith Owens <kaos@sherman.melbourne.sgi.com>
Message-Id: <200103270029.KAA18671@sherman.melbourne.sgi.com>
Subject: TAKE - Change -oformat to --oformat for newer binutils
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

Change -oformat to --oformat for newer binutils

Date:  Mon Mar 26 16:28:12 PST 2001
Workarea:  sherman.melbourne.sgi.com:/build/kaos/2.4.x-xfs

The following file(s) were checked into:
  bonnie.engr.sgi.com:/isms/slinx/2.4.x-xfs


Modid:  2.4.x-xfs:slinx:90916a
linux/arch/i386/boot/Makefile - 1.9


From owner-linux-xfs@oss.sgi.com Mon Mar 26 17:05:53 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2R15rG10901
	for linux-xfs-outgoing; Mon, 26 Mar 2001 17:05:53 -0800
Received: from mail.intercomp-sys.com (internal.intercomp-sys.com [194.90.152.9])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2R15pM10898
	for <linux-xfs@oss.sgi.com>; Mon, 26 Mar 2001 17:05:51 -0800
Received: from ladyboy ([213.8.132.94])
	by mail.intercomp-sys.com (8.9.3/8.9.3) with SMTP id EAA20162
	for <linux-xfs@oss.sgi.com>; Tue, 27 Mar 2001 04:03:48 +0200
Message-Id: <200103270203.EAA20162@mail.intercomp-sys.com>
Content-Type: text/plain;
  charset="iso-8859-1"
From: Hetz Ben Hamo <hetz-home@cobol2java.com>
To: linux-xfs@oss.sgi.com
Subject: regarding XFS
Date: Tue, 27 Mar 2001 03:07:42 +0000
X-Mailer: KMail [version 1.2.1]
MIME-Version: 1.0
Content-Transfer-Encoding: 8bit
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

Hi,

I would like to ask 2 questions about XFS:

1. I have tried it on 4 machines and I really like the results. I wanted to 
push it on few servers at work, but our sys-admin says only version 1.0 will 
be allowed.. When 1.0 is expected to be out?

2. Could you tell me when a coversion tool to convert ext2 to XFS will be 
out? I have 2 machines here at home that I want to convert them to XFS...

Thanks,
Hetz

From owner-linux-xfs@oss.sgi.com Mon Mar 26 17:24:25 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2R1OPv11123
	for linux-xfs-outgoing; Mon, 26 Mar 2001 17:24:25 -0800
Received: from pneumatic-tube.sgi.com (pneumatic-tube.sgi.com [204.94.214.22])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2R1OOM11120
	for <linux-xfs@oss.sgi.com>; Mon, 26 Mar 2001 17:24:24 -0800
Received: from ledzep.americas.sgi.com (ledzep.americas.sgi.com [137.38.226.97]) by pneumatic-tube.sgi.com (980327.SGI.8.8.8-aspam/980310.SGI-aspam) via ESMTP id RAA03469
	for <linux-xfs@oss.sgi.com>; Mon, 26 Mar 2001 17:34:29 -0800 (PST)
	mail_from (cattelan@thebarn.com)
Received: from gibble.americas.sgi.com (gibble.americas.sgi.com [128.162.195.80]) by ledzep.americas.sgi.com (SGI-SGI-8.9.3/americas-smart-nospam1.1) with ESMTP id TAA96159; Mon, 26 Mar 2001 19:23:08 -0600 (CST)
Received: from thebarn.com (localhost [127.0.0.1])
	by gibble.americas.sgi.com (8.11.2/8.11.2) with ESMTP id f2R1D0e24180;
	Mon, 26 Mar 2001 20:13:00 -0500
Message-ID: <3ABFE91B.E87A60EF@thebarn.com>
Date: Mon, 26 Mar 2001 20:12:59 -0500
From: Russell Cattelan <cattelan@thebarn.com>
X-Mailer: Mozilla 4.76 [en] (X11; U; Linux 2.4.2-XFS i686)
X-Accept-Language: en
MIME-Version: 1.0
To: Hetz Ben Hamo <hetz-home@cobol2java.com>
CC: linux-xfs@oss.sgi.com
Subject: Re: regarding XFS
References: <200103270203.EAA20162@mail.intercomp-sys.com>
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

Hetz Ben Hamo wrote:

> Hi,
>
> I would like to ask 2 questions about XFS:
>
> 1. I have tried it on 4 machines and I really like the results. I wanted to
> push it on few servers at work, but our sys-admin says only version 1.0 will
> be allowed.. When 1.0 is expected to be out?

Well if you want I can slap a 1.0 version string on an rpm and send it your way
:-)
Seriously though the tentative date is April 6th barring the discovery of any
major bugs.


>
>
> 2. Could you tell me when a coversion tool to convert ext2 to XFS will be
> out? I have 2 machines here at home that I want to convert them to XFS...

Probably the best conversion tool you will find looks like this:
% dump
% mkfs.xfs
% restore

Any in-place conversion tool would be very difficult to implement and wouldn't
give much bang for the buck.
Some brave soul may want to take on such a programming challenge but the  SGI
XFS team
have no plans to do such a utility.


>
>
> Thanks,
> Hetz

--
Russell Cattelan
--
Digital Elves inc. -- Currently on loan to SGI
Linux XFS core developer.




From owner-linux-xfs@oss.sgi.com Mon Mar 26 17:29:07 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2R1T7n11273
	for linux-xfs-outgoing; Mon, 26 Mar 2001 17:29:07 -0800
Received: from home.smithconcepts.com (ubr-35.28.151.oviedo.cfl.rr.com [65.35.28.151])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2R1T5M11270
	for <linux-xfs@oss.sgi.com>; Mon, 26 Mar 2001 17:29:06 -0800
Received: from ieee.org (IDENT:bjsmith@bitman.oviedo.smithconcepts.com [172.24.24.192])
	by home.smithconcepts.com (8.9.3/8.9.3) with ESMTP id UAA30734;
	Mon, 26 Mar 2001 20:23:35 -0500
Message-ID: <3ABFEF23.92FFB24E@ieee.org>
Date: Mon, 26 Mar 2001 20:38:43 -0500
From: "Bryan J. Smith" <b.j.smith@ieee.org>
Reply-To: b.j.smith@ieee.org, thebs@theseus.com
Organization: SmithConcepts/Personal
X-Mailer: Mozilla 4.76 [en] (X11; U; Linux 2.2.18-0.13 i686)
X-Accept-Language: en
MIME-Version: 1.0
To: Hetz Ben Hamo <hetz-home@cobol2java.com>
CC: linux-xfs@oss.sgi.com
Subject: Re: regarding XFS
References: <200103270203.EAA20162@mail.intercomp-sys.com>
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

Hetz Ben Hamo wrote:
> 1. I have tried it on 4 machines and I really like the results.
> I wanted to push it on few servers at work, but our sys-admin
> says only version 1.0 will be allowed.. When 1.0 is expected to
> be out?

<rant>
First off, bitch slap him silly.  Is he a Windows sysadmin?  God, I
cannot believe the hypocracy that goes on sometimes in corporate
environments.  Tell you what, I'll rebuild everything with a 1.0
release tagged for 'ya, okay?  Version numbers don't mean crap from
a "stability" standpoint in any world.  Some people use 0.99, some
people just go linear as revisions are made and jump to 1.0 from
0.13 when they consider it "1.0", etc...

Secondly, do you support the servers or does he?  That's where the
real "decision making" comes in.  I'm also good to solicit and
review all the opinions, but *I* make the final decision on the
servers I support.

If it makes any difference, I've using Ext3 version 0.0.2f on my
main server for almost 9 months now.  Version 0.0.6b is out, but
0.0.2f works fine for me (0 data loss).  
</rant>

*NOW SERIOUSLY*, what you should be doing is adopting anything new
on servers of lesser importance.  This is how I started with Ext3. 
First was the Squid web/proxy server.  Then it was the print
server.  Finally the main NFS/SMB, over a 3 month period.  You do it
in steps, give each system a month or two of uptime.  Then you know
you can trust it.

> 2. Could you tell me when a coversion tool to convert ext2 to XFS will be
> out? I have 2 machines here at home that I want to convert them to XFS...

Here's the procedure:
  - Buy a 80GB hard disk
  - ...

*SERIOUSLY NOW*, it is always best to "start clean" anyway.  I've
heard of a number of people getting toasted by ReiserFS'
conversion.  If anything, "augment" your storage in your system and
slowly migrate the data off of the Ext2 partitions.  Once they are
all off the old Ext2 partitions, you can reformat them XFS.

It's not like Windows NT/2000 where you've got drive letters (and
using "volume sets" to extend drivers isn't exactly my idea of
"mission critical") -- you can mount where you need the storage, and
symlink as necessary.

Just my $0.02 and suggestions.

-- TheBS

P.S.  I'd be interested in hearing more on your "sysadmin" off-list.

-- 
Bryan "TheBS" Smith         chat:thebs413 @AOL/MSN/Yahoo
Engineer     mailto:b.j.smith@ieee.org,thebs@theseus.com
********************************************************
"Linux will do for applications what the Internet did to
 networks" -- Sam Palmisano, IBM Chief Operating Officer

From owner-linux-xfs@oss.sgi.com Mon Mar 26 18:21:55 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2R2LtP12549
	for linux-xfs-outgoing; Mon, 26 Mar 2001 18:21:55 -0800
Received: from mac.gargoylecc.com ([65.100.85.38])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2R2LsM12545
	for <linux-xfs@oss.sgi.com>; Mon, 26 Mar 2001 18:21:54 -0800
Received: from gargoylecc.com (localhost.localdomain [127.0.0.1])
	by mac.gargoylecc.com (8.11.2/8.11.2) with ESMTP id f2R2Gdf04252;
	Mon, 26 Mar 2001 19:16:43 -0700
Message-ID: <3ABFF7FA.6090902@gargoylecc.com>
Date: Mon, 26 Mar 2001 19:16:26 -0700
From: Russ Ingram <ringram@gargoylecc.com>
User-Agent: Mozilla/5.0 (X11; U; Linux 2.4.2-XFS_0.10 i686; en-US; 0.7) Gecko/20010203
X-Accept-Language: en
MIME-Version: 1.0
To: Hetz Ben Hamo <hetz-home@cobol2java.com>
CC: linux-xfs@oss.sgi.com
Subject: Re: regarding XFS
References: <200103270203.EAA20162@mail.intercomp-sys.com>
Content-Type: text/plain; charset=us-ascii; format=flowed
Content-Transfer-Encoding: 7bit
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

Hetz Ben Hamo wrote:

> 2. Could you tell me when a coversion tool to convert ext2 to XFS will be
> out? I have 2 machines here at home that I want to convert them to XFS...


There is a HOWTO that give the general steps for moving an existing 
ext2 based system to xfs at http://www.gargoylecc.com/Linux+XFS-HOWTO.html

This is not conversion, but rather migration.  Since there isn't (and 
probably won't ever be) a conversion utility the only way to switch an 
existing system over to xfs is by playing a sort of "musical chairs" 
with your filesystems.  Hope this is helpfull.

Russ
--
---------------------------------------------------------------
"Bill Gates and Microsoft have ruined the computer industry for
  a long time to come by creating a class of ignorant and lazy
  computer users."  --Russel Ingram

"Mommy ... can I go out and ... KILL TONIGHT!?"
                    --Glen  Danzig, The Misfits
---------------------------------------------------------------
Russ Ingram
Gargoyle Computer Consulting
(307)742-1361
www.gargoylecc.com




From owner-linux-xfs@oss.sgi.com Mon Mar 26 20:26:52 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2R4QqZ15081
	for linux-xfs-outgoing; Mon, 26 Mar 2001 20:26:52 -0800
Received: from tux.mkp.net (tux.mkp.net [130.225.60.11])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2R4QpM15077
	for <linux-xfs@oss.sgi.com>; Mon, 26 Mar 2001 20:26:51 -0800
Received: from tux.mkp.net ([130.225.60.11] helo=jaguar.mkp.net)
	by tux.mkp.net with esmtp (Exim 3.16 #1)
	id 14hl46-0006WN-00; Tue, 27 Mar 2001 06:26:44 +0200
Received: (from mkp@localhost)
	by jaguar.mkp.net (8.9.3/8.9.3) id XAA16817;
	Mon, 26 Mar 2001 23:25:59 -0500
X-Authentication-Warning: jaguar.mkp.net: mkp set sender to mkp@mkp.net using -f
To: linux-xfs@oss.sgi.com
Subject: TAKE - MD stripe extraction
From: "Martin K. Petersen" <mkp@linuxcare.com>
Organization: Linuxcare, Inc.
Date: 26 Mar 2001 23:25:59 -0500
Message-ID: <yq166gv3kiw.fsf@jaguar.mkp.net>
Lines: 20
User-Agent: Gnus/5.0808 (Gnus v5.8.8) XEmacs/21.1 (Canyonlands)
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk


Date:  Mon Mar 26 20:24:31 PST 2001
Workarea:  sshgate.corp.sgi.com:/export/d0/mkp/XFS/slinx-md

The following file(s) were checked into:
  bonnie.engr.sgi.com:/isms/slinx/2.4.x-xfs


Modid:  2.4.x-xfs:slinx:90938a
cmd/xfsprogs/include/md-int.h - 1.1
cmd/xfsprogs/mkfs/xfs_mkfs.c - 1.4
cmd/xfsprogs/doc/CHANGES - 1.11
cmd/xfsprogs/VERSION - 1.10
        - Automagic stripe unit/stripe width extraction for MD RAID
          devices

-- 
Martin K. Petersen, Principal Linux Consultant, Linuxcare, Inc.
mkp@linuxcare.com, http://www.linuxcare.com/
SGI XFS for Linux Developer, http://oss.sgi.com/projects/xfs/

From owner-linux-xfs@oss.sgi.com Mon Mar 26 20:38:53 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2R4cr315330
	for linux-xfs-outgoing; Mon, 26 Mar 2001 20:38:53 -0800
Received: from tux.mkp.net (tux.mkp.net [130.225.60.11])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2R4cqM15327
	for <linux-xfs@oss.sgi.com>; Mon, 26 Mar 2001 20:38:52 -0800
Received: from tux.mkp.net ([130.225.60.11] helo=jaguar.mkp.net)
	by tux.mkp.net with esmtp (Exim 3.16 #1)
	id 14hlFp-0006Ww-00; Tue, 27 Mar 2001 06:38:51 +0200
Received: (from mkp@localhost)
	by jaguar.mkp.net (8.9.3/8.9.3) id XAA16822;
	Mon, 26 Mar 2001 23:38:07 -0500
X-Authentication-Warning: jaguar.mkp.net: mkp set sender to mkp@mkp.net using -f
To: linux-xfs@oss.sgi.com
Subject: TAKE - mkfs man page
From: "Martin K. Petersen" <mkp@linuxcare.com>
Organization: Linuxcare, Inc.
Date: 26 Mar 2001 23:38:06 -0500
Message-ID: <yq11yrj3jyp.fsf@jaguar.mkp.net>
Lines: 16
User-Agent: Gnus/5.0808 (Gnus v5.8.8) XEmacs/21.1 (Canyonlands)
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk


Date:  Mon Mar 26 20:37:58 PST 2001
Workarea:  sshgate.corp.sgi.com:/export/d0/mkp/XFS/slinx-md

The following file(s) were checked into:
  bonnie.engr.sgi.com:/isms/slinx/2.4.x-xfs


Modid:  2.4.x-xfs:slinx:90940a
cmd/xfsprogs/man/man8/mkfs.xfs.8 - 1.3
        - Updated man page section about sunit and swidth.

-- 
Martin K. Petersen, Principal Linux Consultant, Linuxcare, Inc.
mkp@linuxcare.com, http://www.linuxcare.com/
SGI XFS for Linux Developer, http://oss.sgi.com/projects/xfs/

From owner-linux-xfs@oss.sgi.com Mon Mar 26 22:57:16 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2R6vGN18499
	for linux-xfs-outgoing; Mon, 26 Mar 2001 22:57:16 -0800
Received: from pneumatic-tube.sgi.com (pneumatic-tube.sgi.com [204.94.214.22])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2R6vFM18496
	for <linux-xfs@oss.sgi.com>; Mon, 26 Mar 2001 22:57:15 -0800
Received: from snort.melbourne.sgi.com (snort.melbourne.sgi.com [134.14.55.149]) by pneumatic-tube.sgi.com (980327.SGI.8.8.8-aspam/980310.SGI-aspam) via ESMTP id XAA06608
	for <linux-xfs@oss.sgi.com>; Mon, 26 Mar 2001 23:07:19 -0800 (PST)
	mail_from (nathans@snort.melbourne.sgi.com)
Received: (from nathans@localhost)
	by snort.melbourne.sgi.com (SGI-8.9.3/8.9.3) id QAA68292
	for linux-xfs@oss.sgi.com; Tue, 27 Mar 2001 16:55:49 +1000 (EST)
Date: Tue, 27 Mar 2001 16:55:49 +1000 (EST)
From: Nathan Scott <nathans@snort.melbourne.sgi.com>
Message-Id: <200103270655.QAA68292@snort.melbourne.sgi.com>
To: linux-xfs@oss.sgi.com
Subject: TAKE - xfsdump
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

Date:  Mon Mar 26 22:55:40 PST 2001
Workarea:  snort.melbourne.sgi.com:/diskb/build4/nathans/linux-xfs

The following file(s) were checked into:
  bonnie.engr.sgi.com:/isms/slinx/2.4.x-xfs


Modid:  2.4.x-xfs:slinx:90944a
cmd/xfsdump/quota/xfsdq.c - 1.2
	- fix return value on success - thanks Ivan.



From owner-linux-xfs@oss.sgi.com Mon Mar 26 22:59:13 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2R6xDx18587
	for linux-xfs-outgoing; Mon, 26 Mar 2001 22:59:13 -0800
Received: from pneumatic-tube.sgi.com (pneumatic-tube.sgi.com [204.94.214.22])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2R6xDM18584
	for <linux-xfs@oss.sgi.com>; Mon, 26 Mar 2001 22:59:13 -0800
Received: from snort.melbourne.sgi.com (snort.melbourne.sgi.com [134.14.55.149]) by pneumatic-tube.sgi.com (980327.SGI.8.8.8-aspam/980310.SGI-aspam) via ESMTP id XAA07207
	for <linux-xfs@oss.sgi.com>; Mon, 26 Mar 2001 23:09:17 -0800 (PST)
	mail_from (nathans@snort.melbourne.sgi.com)
Received: (from nathans@localhost)
	by snort.melbourne.sgi.com (SGI-8.9.3/8.9.3) id QAA01515
	for linux-xfs@oss.sgi.com; Tue, 27 Mar 2001 16:57:48 +1000 (EST)
Date: Tue, 27 Mar 2001 16:57:48 +1000 (EST)
From: Nathan Scott <nathans@snort.melbourne.sgi.com>
Message-Id: <200103270657.QAA01515@snort.melbourne.sgi.com>
To: linux-xfs@oss.sgi.com
Subject: TAKE - qa
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

Date:  Mon Mar 26 22:57:37 PST 2001
Workarea:  snort.melbourne.sgi.com:/diskb/build4/nathans/linux-xfs

The following file(s) were checked into:
  bonnie.engr.sgi.com:/isms/slinx/2.4.x-xfs


Modid:  2.4.x-xfs:slinx:90945a
cmd/xfstests/src/Makefile - 1.4
cmd/xfstests/010 - 1.2
cmd/xfstests/configure.in - 1.4
cmd/xfstests/include/builddefs.in - 1.2
	- be more flexible with a lack of -ldb on a system (just skip test 010).



From owner-linux-xfs@oss.sgi.com Mon Mar 26 23:35:24 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2R7ZOD19453
	for linux-xfs-outgoing; Mon, 26 Mar 2001 23:35:24 -0800
Received: from lupo.thebarn.com (nic-31-c12-219.mn.mediaone.net [24.31.12.219])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2R7ZNM19450
	for <linux-xfs@oss.sgi.com>; Mon, 26 Mar 2001 23:35:23 -0800
Received: (from cattelan@localhost)
	by lupo.thebarn.com (8.11.3/8.11.0) id f2R7Z7889916;
	Tue, 27 Mar 2001 01:35:07 -0600 (CST)
	(envelope-from cattelan)
From: Russell Cattelan <cattelan@thebarn.com>
MIME-Version: 1.0
Message-ID: <15040.17067.255930.228018@lupo.thebarn.com>
Date: Tue, 27 Mar 2001 01:35:07 -0600 (CST)
To: linux-xfs@oss.sgi.com
Subject: PreRelease tag added to tree.
X-Mailer: VM 6.72 under 21.1 (patch 14) "Cuyahoga Valley" XEmacs Lucid
User-Agent: SEMI/1.13.7 (Awazu) FLIM/1.13.2 (Kasanui) MULE XEmacs/21.1 (patch 14) (Cuyahoga Valley) (i386--freebsd)
Content-Type: text/plain; charset=US-ASCII
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

Just a quick note:
The tag "PreRelease-0_10" has been added to cvs tree.

This is a global tag that represents the state of the
tree as of prerelease 0.10 test2.

This tag can be feed to cvs e.g.
% cvs update -r PreRelease-0_10
to checkout a copy of the prerelease tree.


-Russell

From owner-linux-xfs@oss.sgi.com Mon Mar 26 23:37:40 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2R7be319494
	for linux-xfs-outgoing; Mon, 26 Mar 2001 23:37:40 -0800
Received: from deliverator.sgi.com (deliverator.sgi.com [204.94.214.10])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2R7bdM19491
	for <linux-xfs@oss.sgi.com>; Mon, 26 Mar 2001 23:37:39 -0800
Received: from larry.melbourne.sgi.com (larry.melbourne.sgi.com [134.14.52.130]) by deliverator.sgi.com (980309.SGI.8.8.8-aspam-6.2/980310.SGI-aspam) via SMTP id XAA18866
	for <linux-xfs@oss.sgi.com>; Mon, 26 Mar 2001 23:36:22 -0800 (PST)
	mail_from (nathans@wobbly.melbourne.sgi.com)
Received: from wobbly.melbourne.sgi.com (wobbly.melbourne.sgi.com [134.14.55.135]) by larry.melbourne.sgi.com (950413.SGI.8.6.12/950213.SGI.AUTOCF) via ESMTP id RAA26597; Tue, 27 Mar 2001 17:36:13 +1000
Received: (from nathans@localhost) by wobbly.melbourne.sgi.com (980427.SGI.8.8.8/980728.SGI.AUTOCF) id RAA49349; Tue, 27 Mar 2001 17:36:10 +1000 (EST)
From: "Nathan Scott" <nathans@wobbly.melbourne.sgi.com>
Message-Id: <10103271736.ZM214128@wobbly.melbourne.sgi.com>
Date: Tue, 27 Mar 2001 17:36:09 -0500
In-Reply-To: "Nathan Scott" <nathans@wobbly.melbourne.sgi.com>
        "Re: 2.4.2 - XFS kernel March 19th & quota" (Mar 26,  9:32am)
References: <006401c0b42c$072faea0$0f01a8c0@myplace.net> 
	<10103260932.ZM235393@wobbly.melbourne.sgi.com>
X-Mailer: Z-Mail (3.2.3 08feb96 MediaMail)
To: "Bas" <list@showme.wox.org>, <linux-xfs@oss.sgi.com>
Subject: Re: 2.4.2 - XFS kernel March 19th & quota
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

hi,

On Mar 26,  9:32am, Nathan Scott wrote:
> Subject: Re: 2.4.2 - XFS kernel March 19th & quota
> > 
> > I'm running XFS with quota support ...
> 
> Before you grab new tools, wait for my next message (later
> today) on the new version of the quota tools.  These will
> include the "quot" utility I referred to above.
> 

OK, I didn't manage to make "later today" ;-)  - I've
held off a bit until I could more thoroughly test the
new version of the tools, which are now pretty solid.

The initial version of the new tools is now available
(via cvs only, currently) from the cvs link below:
    http://sourceforge.net/projects/linuxquota/

These tools support:
	- 2.4.X quota (VFS, version 1)
	- 2.4.X-acX quota (VFS, version 2)
	- XFS quota (journaled, nifty)

To checkout via cvs, I'm using the commands:

[initial login..]
$ cvs -d:pserver:anonymous@cvs.linuxquota.sourceforge.net:/cvsroot/linuxquota login
[subsequent checkouts..]
$ cvs -z3 -d:pserver:anonymous@cvs.linuxquota.sourceforge.net:/cvsroot/linuxquota co quota-tools

I'll send some more comprehensive mail once there's a
source tarball for download and once the Sourceforge
project page has been updated.  The quota tools which
currently reside in the XFS cvs repository will soon
be "downsized", in preference for the new version.

cheers.

-- 
Nathan

From owner-linux-xfs@oss.sgi.com Tue Mar 27 03:10:47 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2RBAlD25205
	for linux-xfs-outgoing; Tue, 27 Mar 2001 03:10:47 -0800
Received: from chiara.elte.hu (chiara.elte.hu [157.181.150.200])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2RBAjM25202
	for <linux-xfs@oss.sgi.com>; Tue, 27 Mar 2001 03:10:46 -0800
Received: by chiara.elte.hu (Postfix, from userid 17000)
	id 233B4187C; Tue, 27 Mar 2001 13:10:25 +0200 (CEST)
Date: Tue, 27 Mar 2001 13:10:25 +0200
From: KELEMEN Peter <fuji@elte.hu>
To: linux-xfs@oss.sgi.com
Subject: xfs_log.c typo?
Message-ID: <20010327131025.A8661@chiara.elte.hu>
Reply-To: KELEMEN Peter <fuji@elte.hu>
Mime-Version: 1.0
Content-Type: multipart/mixed; boundary="HcAYCG3uE/tztfnV"
Content-Disposition: inline
Content-Transfer-Encoding: 8bit
User-Agent: Mutt/1.2.5i
Organization: ELTE Eotvos Lorand University of Sciences, Budapest, Hungary
X-GPG-KeyID: 1024D/EE4C26E8 2000-03-20
X-GPG-Fingerprint: D402 4AF3 7488 165B CC34  4147 7F0C D922 EE4C 26E8
X-PGP-KeyID: 1024/45F83E45 1998/04/04
X-PGP-Fingerprint: 26 87 63 4B 07 28 1F AD  6D AA B5 8A D6 03 0F BF
X-Comment: Personal opinion.  Paragraphs might have been reformatted.
X-Copyright: Forwarding or publishing without permission is prohibited.
X-Accept-Language: hu,en
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk


--HcAYCG3uE/tztfnV
Content-Type: text/plain; charset=iso-8859-1
Content-Disposition: inline
Content-Transfer-Encoding: 8bit

Hello,

While browsing around the code at 4 a.m., I came across this what
I suspect to be a typo.

Peter

-- 
    .+'''+.         .+'''+.         .+'''+.         .+'''+.         .+''
 Kelemen Péter     /       \       /       \       /      fuji@elte.hu
.+'         `+...+'         `+...+'         `+...+'         `+...+'

--HcAYCG3uE/tztfnV
Content-Type: text/plain; charset=us-ascii
Content-Disposition: attachment; filename="xfs_log.c-typo20010327.patch"

--- linux/fs/xfs/xfs_log.c.orig	Tue Mar 27 12:17:57 2001
+++ linux/fs/xfs/xfs_log.c	Tue Mar 27 12:18:59 2001
@@ -164,7 +164,7 @@
 		     (void *)((unsigned long)log->l_curr_cycle),
 		     (void *)((unsigned long)log->l_curr_block),
 		     (void *)((unsigned long)CYCLE_LSN(log->l_tail_lsn, ARCH_NOCONVERT)),
-		     (void *)((unsigned long)BLOCK_LSN(log->l_tail_lsn, ARCH_NOCONVER)),
+		     (void *)((unsigned long)BLOCK_LSN(log->l_tail_lsn, ARCH_NOCONVERT)),
 		     (void *)string,
 		     (void *)((unsigned long)13),
 		     (void *)((unsigned long)14),

--HcAYCG3uE/tztfnV--

From owner-linux-xfs@oss.sgi.com Tue Mar 27 04:33:52 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2RCXq326798
	for linux-xfs-outgoing; Tue, 27 Mar 2001 04:33:52 -0800
Received: from yog-sothoth.sgi.com (eugate.sgi.com [192.48.160.10])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2RCXbM26793
	for <linux-xfs@oss.sgi.com>; Tue, 27 Mar 2001 04:33:37 -0800
Received: from zeus-fddi.americas.sgi.com (zeus-fddi.americas.sgi.com [128.162.8.103]) by yog-sothoth.sgi.com (980305.SGI.8.8.8-aspam-6.2/980304.SGI-aspam-europe) via ESMTP id OAA6483116
	for <linux-xfs@oss.sgi.com>; Tue, 27 Mar 2001 14:33:34 +0200 (CEST)
	mail_from (lord@sgi.com)
Received: from daisy-e185.americas.sgi.com (daisy-e194.americas.sgi.com [128.162.194.214]) by zeus-fddi.americas.sgi.com (8.9.3/americas-smart-nospam1.1) with ESMTP id GAA1231643; Tue, 27 Mar 2001 06:30:53 -0600 (CST)
Received: from jen.americas.sgi.com (IDENT:root@jen.americas.sgi.com [128.162.184.86]) by daisy-e185.americas.sgi.com (SGI-8.9.3/SGI-server-1.7) with ESMTP id GAA01080; Tue, 27 Mar 2001 06:30:49 -0600 (CST)
Received: from jen.americas.sgi.com by jen.americas.sgi.com (8.11.0/SGI-client-1.7) via ESMTP id f2RCU3d03153; Tue, 27 Mar 2001 06:30:04 -0600
Message-Id: <200103271230.f2RCU3d03153@jen.americas.sgi.com>
X-Mailer: exmh version 2.2 06/23/2000 with nmh-1.0.4
To: KELEMEN Peter <fuji@elte.hu>
cc: linux-xfs@oss.sgi.com
Subject: Re: xfs_log.c typo? 
In-Reply-To: Message from KELEMEN Peter <fuji@elte.hu> 
   of "Tue, 27 Mar 2001 13:10:25 +0200." <20010327131025.A8661@chiara.elte.hu> 
Content-Transfer-Encoding: 8bit
Date: Tue, 27 Mar 2001 06:30:03 -0600
From: Steve Lord <lord@sgi.com>
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk


> Hello,
> 
> While browsing around the code at 4 a.m., I came across this what
> I suspect to be a typo.


Yes, this is a typo, it is in some debug code which probably has not been
enabled for a couple of years, but definitely a typo.

Steve



From owner-linux-xfs@oss.sgi.com Tue Mar 27 05:01:35 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2RD1Zg27869
	for linux-xfs-outgoing; Tue, 27 Mar 2001 05:01:35 -0800
Received: from ns.tecosim.de (ns.tecosim.de [194.24.222.9])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2RD1VM27866
	for <linux-xfs@oss.sgi.com>; Tue, 27 Mar 2001 05:01:32 -0800
Received: from donner.tecosim.de (root@donner.tecosim.de [194.24.222.109])
	by ns.tecosim.de (8.9.3/8.9.3/SuSE Linux 8.9.3-0.1) with ESMTP id PAA11570;
	Tue, 27 Mar 2001 15:01:28 +0200
Received: (from leh@localhost)
	by donner.tecosim.de (8.11.2/8.11.2/SuSE Linux 8.11.1-0.5) id f2RD1R103423;
	Tue, 27 Mar 2001 15:01:27 +0200
Date: Tue, 27 Mar 2001 15:01:27 +0200
From: Utz Lehmann <leh@tecosim.de>
To: Russell Cattelan <cattelan@thebarn.com>
Cc: linux-xfs@oss.sgi.com
Subject: Re: PreRelease tag added to tree.
Message-ID: <20010327150127.A1388@tecosim.de>
References: <15040.17067.255930.228018@lupo.thebarn.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.2.5i
In-Reply-To: <15040.17067.255930.228018@lupo.thebarn.com>; from cattelan@thebarn.com on Tue, Mar 27, 2001 at 01:35:07AM -0600
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

Hi

I tried this:

leh@donner:xfs\> cvs update -r PreRelease-0_10
cvs [server aborted]: cannot write /cvs/CVSROOT/val-tags: Permission denied

Is this my fault, or the cvs server?


Regards

utz lehmann

Russell Cattelan [cattelan@thebarn.com] wrote:
> Just a quick note:
> The tag "PreRelease-0_10" has been added to cvs tree.
> 
> This is a global tag that represents the state of the
> tree as of prerelease 0.10 test2.
> 
> This tag can be feed to cvs e.g.
> % cvs update -r PreRelease-0_10
> to checkout a copy of the prerelease tree.
> 
> 
> -Russell

From owner-linux-xfs@oss.sgi.com Tue Mar 27 07:16:05 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2RFG5V32281
	for linux-xfs-outgoing; Tue, 27 Mar 2001 07:16:05 -0800
Received: from lips.borg.umn.edu (lips.borg.umn.edu [160.94.232.50])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2RFG4M32277
	for <linux-xfs@oss.sgi.com>; Tue, 27 Mar 2001 07:16:04 -0800
Received: from thebarn.com (nic-31-c12-219.mn.mediaone.net [24.31.12.219])
	by lips.borg.umn.edu (8.12.0.Beta5/8.12.0.Beta5) with ESMTP id f2RFFkdE071088;
	Tue, 27 Mar 2001 09:15:46 -0600 (CST)
Message-ID: <3AC0AE0F.A7506220@thebarn.com>
Date: Tue, 27 Mar 2001 09:13:19 -0600
From: Russell Cattelan <cattelan@thebarn.com>
X-Mailer: Mozilla 4.75 [en] (X11; U; Linux 2.4.2-XFS i686)
X-Accept-Language: en
MIME-Version: 1.0
To: Utz Lehmann <leh@tecosim.de>
CC: linux-xfs@oss.sgi.com
Subject: Re: PreRelease tag added to tree.
References: <15040.17067.255930.228018@lupo.thebarn.com> <20010327150127.A1388@tecosim.de>
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

Utz Lehmann wrote:

> Hi
>
> I tried this:
>
> leh@donner:xfs\> cvs update -r PreRelease-0_10
> cvs [server aborted]: cannot write /cvs/CVSROOT/val-tags: Permission denied
>
> Is this my fault, or the cvs server?

I suspect it's the cvs server ... I'll look into it.


>
>
> Regards
>
> utz lehmann
>
> Russell Cattelan [cattelan@thebarn.com] wrote:
> > Just a quick note:
> > The tag "PreRelease-0_10" has been added to cvs tree.
> >
> > This is a global tag that represents the state of the
> > tree as of prerelease 0.10 test2.
> >
> > This tag can be feed to cvs e.g.
> > % cvs update -r PreRelease-0_10
> > to checkout a copy of the prerelease tree.
> >
> >
> > -Russell


From owner-linux-xfs@oss.sgi.com Tue Mar 27 07:53:13 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2RFrDw00585
	for linux-xfs-outgoing; Tue, 27 Mar 2001 07:53:13 -0800
Received: from dfwns01.algx.net (ns1.algx.net [216.99.225.30])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2RFrCM00580
	for <linux-xfs@oss.sgi.com>; Tue, 27 Mar 2001 07:53:12 -0800
Received: from jtsdell (66-2-81-26.customer.algx.net [66.2.81.26])
	by dfwns01.algx.net (8.9.1/8.9.1) with ESMTP id JAA16527
	for <linux-xfs@oss.sgi.com>; Tue, 27 Mar 2001 09:53:00 -0600 (CST)
Message-ID: <XFMail.20010327105142.jtrostel@connex.com>
X-Mailer: XFMail 1.4.8 on Linux
X-Priority: 3 (Normal)
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 8bit
MIME-Version: 1.0
Date: Tue, 27 Mar 2001 10:51:42 -0500 (EST)
Reply-To: jtrostel@connex.com
Organization: Connex
From: John Trostel <jtrostel@connex.com>
To: linux-xfs@oss.sgi.com
Subject: Problems with ACL inheritance and chacl
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

I am seeing some strange behavior with both ACL inheritance and chacl
operation.  Does anyone else see this or is my build faulty?

I started with a freshly formatted xfs partition on /mnt/xfs_part.

I set the access, default and mask ACLs for this parition as follows:

  [root@jtsdell /mnt]# mount /mnt/xfs_part/

  [root@jtsdell /mnt]# chacl -l xfs_part/
  xfs_part/ []
        
  [root@jtsdell /mnt]# chacl -b u::rwx,g::rwx,o::rwx,m::rwx u::rwx,g::rwx,o::rwx
                           xfs_part

  [root@jtsdell /mnt]# chacl -l xfs_part/
  xfs_part/ [u::rwx,g::rwx,o::rwx,m::rwx/u::rwx,g::rwx,o::rwx]


So far... so good!


  [root@jtsdell /mnt]# exit
  exit

  [jt@jtsdell xfs_backup]$ cd ..

  [jt@jtsdell /mnt]$ cd xfs_part

  [jt@jtsdell xfs_part]$ ls

  [jt@jtsdell xfs_part]$ touch jt.txt

  [jt@jtsdell xfs_part]$ chacl -l jt.txt
  jt.txt [u::rw-,g::rw-,o::r--]

Not good... Shouldn't this get the default ACLs u::rwx,g::rwx,o::rwx ?

  [jt@jtsdell xfs_part]$ su
  Password:

  [root@jtsdell xfs_part]# touch root.txt
 
  [root@jtsdell xfs_part]# chacl -l root.txt
  root.txt [u::rw-,g::rw-,o::r--]

Hmm... root has the same problem!

  [root@jtsdell xfs_part]# chacl -B root.txt
  chacl: error removing access acl on "root.txt": Invalid argument
  chacl: error removing default acl on "root.txt": Invalid argument

  [root@jtsdell
  xfs_part]# chacl -D root.txt
  chacl: error removing default acl on "root.txt": Invalid argument

Shouldn't 'chacl' be removing my ACLs here?

I can go back and manually set the ACLs without problem, but they don't seem to
propogate down correctly and chacl removal seems flawed.  I have downloaded a
fresh copy of the CVS tree and will try with an updated kernel and cmd tools
later today.  I noticed that chacl hasn't been changed in a while. Maybe it
bears looking into.... (I'll take a peek while the system is hoisting itself by
it's own petard!)

-- 
John M. Trostel
Linux OS Engineer
Connex
jtrostel@connex.com

From owner-linux-xfs@oss.sgi.com Tue Mar 27 08:20:52 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2RGKqM01041
	for linux-xfs-outgoing; Tue, 27 Mar 2001 08:20:52 -0800
Received: from chaos.egr.duke.edu (IDENT:root@chaos.egr.duke.edu [152.3.195.82])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2RGKpM01036
	for <linux-xfs@oss.sgi.com>; Tue, 27 Mar 2001 08:20:51 -0800
Received: from localhost (jlb@localhost)
	by chaos.egr.duke.edu (8.9.3/8.9.3) with ESMTP id LAA07308;
	Tue, 27 Mar 2001 11:20:43 -0500
X-Authentication-Warning: chaos.egr.duke.edu: jlb owned process doing -bs
Date: Tue, 27 Mar 2001 11:20:43 -0500 (EST)
From: Joshua Baker-LePain <jlb17@duke.edu>
X-Sender:  <jlb@chaos.egr.duke.edu>
Reply-To: Linux xfs mailing list <linux-xfs@oss.sgi.com>
To: Steve Lord <lord@sgi.com>
cc: Linux xfs mailing list <linux-xfs@oss.sgi.com>
Subject: Re: Pre-production questions 
In-Reply-To: <200103202025.f2KKP3s07381@jen.americas.sgi.com>
Message-ID: <Pine.LNX.4.30.0103271100050.1152-100000@chaos.egr.duke.edu>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

On Tue, 20 Mar 2001 at 2:25pm, Steve Lord wrote

> p.s. Let us know if these numbers make things go faster!
>
OK.  ;)  The answer is a bit, but not really.

I verified with the vendor that, indeed, the stripe size specified on the
RAID system is blocks/disk.  So, I upped the stripe size to 128
(512byte) blocks (I'll be dealing mostly with large files), made two
filesystems, and formatted them as following:

[jlb@philip jlb]$ sudo mkfs.xfs -f /dev/sdb1
meta-data=/dev/sdb1              isize=256    agcount=134, agsize=261319
blks
data     =                       bsize=4096   blocks=35016700, imaxpct=25
         =                       sunit=0      swidth=0 blks, unwritten=0
naming   =version 2              bsize=4096
log      =internal log           bsize=4096   blocks=1200
realtime =none                   extsz=65536  blocks=0, rtextents=0
[jlb@philip jlb]$ sudo mkfs.xfs -f -d sunit=128,swidth=896 /dev/sdb2
meta-data=/dev/sdb2              isize=256    agcount=134, agsize=261328
blks
data     =                       bsize=4096   blocks=35016704, imaxpct=25
         =                       sunit=16     swidth=112 blks, unwritten=0
naming   =version 2              bsize=4096
log      =internal log           bsize=4096   blocks=1200
realtime =none                   extsz=65536  blocks=0, rtextents=0
[jlb@philip /mnt]$ sudo mount -t xfs /dev/sdb1 /mnt/raid1
[jlb@philip /mnt]$ sudo mount -t xfs /dev/sdb2 /mnt/raid2

I benchmarked with bonnie++ and lmdd.  If you want me to run any more
tests, just let me know.  I'll still be playi^Wtesting for most of the
rest of this week.  It's going to be a shame to let the grad students at
this thing...

For the record, the RAID is attached to a Dell Precision 410 with one PII
450 and 512MB of RAM.  It's attached via an Initio INI-A100U2W card (Domex
branded).  I'm still running 2.4.2-XFS pulled out of CVS on 3/13.  For
those who don't recall the rest of this (week-old) thread, the RAID is
made up of 8 5400RPM 80GB Maxtor disks in RAID5 and appearing as a SCSI
disk to the host.

Here are the results (averaged over three runs):

[jlb@philip jlb]$ pwd
/mnt/raid1/jlb
[jlb@philip jlb]$ bonnie++ -r 512 -n 64
Version  1.01       ------Sequential Output------ --Sequential Input- --Random-
                    -Per Chr- --Block-- -Rewrite- -Per Chr- --Block-- --Seeks--
Machine        Size K/sec %CP K/sec %CP K/sec %CP K/sec %CP K/sec %CP  /sec %CP
philip.egr.duke. 1G  5058  98 30302  35 10498  16  4609  84 26341  22 133.8   1
                    ------Sequential Create------ --------Random Create--------
                    -Create-- --Read--- -Delete-- -Create-- --Read--- -Delete--
              files  /sec %CP  /sec %CP  /sec %CP  /sec %CP  /sec %CP  /sec %CP
                 64  1299  89 88532  98  2185  90  1239  84 99707  98   782  44

[jlb@philip jlb]$ pwd
/mnt/raid2/jlb
[jlb@philip jlb]$ bonnie++ -r 512 -n 64
Version  1.01       ------Sequential Output------ --Sequential Input- --Random-
                    -Per Chr- --Block-- -Rewrite- -Per Chr- --Block-- --Seeks--
Machine        Size K/sec %CP K/sec %CP K/sec %CP K/sec %CP K/sec %CP  /sec %CP
philip.egr.duke. 1G  4962  97 32775  38  9669  15  4498  84 26304  21 134.6   1
                    ------Sequential Create------ --------Random Create--------
                    -Create-- --Read--- -Delete-- -Create-- --Read--- -Delete--
              files  /sec %CP  /sec %CP  /sec %CP  /sec %CP  /sec %CP  /sec %CP
                 64  1144  79 88914  96  2196  92  1029  70 101379 95   585  32

[jlb@philip jlb]$ pwd
/mnt/raid1/jlb
[jlb@philip jlb]$ lmdd if=internal of=tmp bs=32k count=25000 fsync=1
819.2000 MB in 31.2588 secs, 26.2070 MB/sec
[jlb@philip jlb]$ lmdd if=tmp of=internal
819.2000 MB in 32.1173 secs, 25.5065 MB/sec

[jlb@philip jlb]$ pwd
/mnt/raid2/jlb
[jlb@philip jlb]$ lmdd if=internal of=tmp bs=32k count=25000 fsync=1
819.2000 MB in 31.3305 secs, 26.1470 MB/sec
[jlb@philip jlb]$ lmdd if=tmp of=internal
819.2000 MB in 32.3775 secs, 25.3015 MB/sec

Please note that I am *not* complaining about these numbers.  I'll
take 26-30MB/sec any day.  But if you have any more suggestions or want me
to do any other tests, let me know.  Thanks!

-- 
Joshua Baker-LePain
Department of Biomedical Engineering
Duke University


From owner-linux-xfs@oss.sgi.com Tue Mar 27 13:46:59 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2RLkxE11600
	for linux-xfs-outgoing; Tue, 27 Mar 2001 13:46:59 -0800
Received: from yog-sothoth.sgi.com (eugate.sgi.com [192.48.160.10])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2RLkvM11597
	for <linux-xfs@oss.sgi.com>; Tue, 27 Mar 2001 13:46:58 -0800
Received: from snort.melbourne.sgi.com (snort.melbourne.sgi.com [134.14.55.149]) by yog-sothoth.sgi.com (980305.SGI.8.8.8-aspam-6.2/980304.SGI-aspam-europe) via ESMTP id XAA3065399
	for <linux-xfs@oss.sgi.com>; Tue, 27 Mar 2001 23:46:55 +0200 (CEST)
	mail_from (nathans@snort.melbourne.sgi.com)
Received: (from nathans@localhost)
	by snort.melbourne.sgi.com (SGI-8.9.3/8.9.3) id HAA88251
	for linux-xfs@oss.sgi.com; Wed, 28 Mar 2001 07:45:29 +1000 (EST)
Date: Wed, 28 Mar 2001 07:45:29 +1000 (EST)
From: Nathan Scott <nathans@snort.melbourne.sgi.com>
Message-Id: <200103272145.HAA88251@snort.melbourne.sgi.com>
To: linux-xfs@oss.sgi.com
Subject: TAKE - fix a warning
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

Date:  Tue Mar 27 13:44:21 PST 2001
Workarea:  snort.melbourne.sgi.com:/diskb/build4/nathans/linux-xfs

The following file(s) were checked into:
  bonnie.engr.sgi.com:/isms/slinx/2.4.x-xfs


Modid:  2.4.x-xfs:slinx:90986a
linux/fs/xfs/linux/xfs_lrw.c - 1.85
	- fix a compiler warning in some dmapi code (unused variable).


From owner-linux-xfs@oss.sgi.com Tue Mar 27 13:56:04 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2RLu4R11769
	for linux-xfs-outgoing; Tue, 27 Mar 2001 13:56:04 -0800
Received: from pneumatic-tube.sgi.com (pneumatic-tube.sgi.com [204.94.214.22])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2RLu4M11766
	for <linux-xfs@oss.sgi.com>; Tue, 27 Mar 2001 13:56:04 -0800
Received: from zeus-fddi.americas.sgi.com (zeus-fddi.americas.sgi.com [128.162.8.103]) by pneumatic-tube.sgi.com (980327.SGI.8.8.8-aspam/980310.SGI-aspam) via ESMTP id OAA08117
	for <linux-xfs@oss.sgi.com>; Tue, 27 Mar 2001 14:06:09 -0800 (PST)
	mail_from (lord@sgi.com)
Received: from daisy-e185.americas.sgi.com (daisy-e194.americas.sgi.com [128.162.194.214]) by zeus-fddi.americas.sgi.com (8.9.3/americas-smart-nospam1.1) with ESMTP id PAA1281800 for <linux-xfs@oss.sgi.com>; Tue, 27 Mar 2001 15:54:48 -0600 (CST)
Received: from jen.americas.sgi.com (IDENT:root@jen.americas.sgi.com [128.162.184.86]) by daisy-e185.americas.sgi.com (SGI-8.9.3/SGI-server-1.7) with ESMTP id PAA48532 for <linux-xfs@oss.sgi.com>; Tue, 27 Mar 2001 15:54:47 -0600 (CST)
From: Steve Lord <lord@sgi.com>
Received: by jen.americas.sgi.com (8.11.0/SGI-client-1.7) id f2RLrwG20980; Tue, 27 Mar 2001 15:53:58 -0600
Message-Id: <200103272153.f2RLrwG20980@jen.americas.sgi.com>
Date: Tue, 27 Mar 2001 15:53:58 -0600
Subject: TAKE - fix small typo in debug code
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

Date:  Tue Mar 27 13:54:29 PST 2001
Workarea:  jen.americas.sgi.com:/src/lord/xfs-linux.2.4

The following file(s) were checked into:
  bonnie.engr.sgi.com:/isms/slinx/2.4.x-xfs


Modid:  2.4.x-xfs:slinx:90991a
linux/fs/xfs/xfs_log.c - 1.233
	- fix typo in some debug code - would cause build problem if we
	  turned on log tracing.



From owner-linux-xfs@oss.sgi.com Tue Mar 27 15:28:37 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2RNSbS14470
	for linux-xfs-outgoing; Tue, 27 Mar 2001 15:28:37 -0800
Received: from deliverator.sgi.com (deliverator.sgi.com [204.94.214.10])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2RNSbM14467
	for <linux-xfs@oss.sgi.com>; Tue, 27 Mar 2001 15:28:37 -0800
Received: from snort.melbourne.sgi.com (snort.melbourne.sgi.com [134.14.55.149]) by deliverator.sgi.com (980309.SGI.8.8.8-aspam-6.2/980310.SGI-aspam) via ESMTP id PAA29453
	for <linux-xfs@oss.sgi.com>; Tue, 27 Mar 2001 15:27:22 -0800 (PST)
	mail_from (nathans@snort.melbourne.sgi.com)
Received: (from nathans@localhost)
	by snort.melbourne.sgi.com (SGI-8.9.3/8.9.3) id JAA53100
	for linux-xfs@oss.sgi.com; Wed, 28 Mar 2001 09:27:10 +1000 (EST)
Date: Wed, 28 Mar 2001 09:27:10 +1000 (EST)
From: Nathan Scott <nathans@snort.melbourne.sgi.com>
Message-Id: <200103272327.JAA53100@snort.melbourne.sgi.com>
To: linux-xfs@oss.sgi.com
Subject: TAKE - Makefile
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

Russell, you'll want this for 0.10 too.

cheers.

Date:  Tue Mar 27 15:26:40 PST 2001
Workarea:  snort.melbourne.sgi.com:/diskb/build4/nathans/linux-xfs

The following file(s) were checked into:
  bonnie.engr.sgi.com:/isms/slinx/2.4.x-xfs


Modid:  2.4.x-xfs:slinx:91017a
cmd/xfsprogs/include/Makefile - 1.3
	- need md-int.h in the list of lsrcfiles in order that it makes
	  it into the package.



From owner-linux-xfs@oss.sgi.com Tue Mar 27 16:16:40 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2S0Ge515150
	for linux-xfs-outgoing; Tue, 27 Mar 2001 16:16:40 -0800
Received: from deliverator.sgi.com (deliverator.sgi.com [204.94.214.10])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2S0GdM15147
	for <linux-xfs@oss.sgi.com>; Tue, 27 Mar 2001 16:16:39 -0800
Received: from larry.melbourne.sgi.com (larry.melbourne.sgi.com [134.14.52.130]) by deliverator.sgi.com (980309.SGI.8.8.8-aspam-6.2/980310.SGI-aspam) via SMTP id QAA06140
	for <linux-xfs@oss.sgi.com>; Tue, 27 Mar 2001 16:15:24 -0800 (PST)
	mail_from (ajag@fudge.melbourne.sgi.com)
Received: from fudge.melbourne.sgi.com (fudge.melbourne.sgi.com [134.14.55.184]) by larry.melbourne.sgi.com (950413.SGI.8.6.12/950213.SGI.AUTOCF) via ESMTP id KAA03195; Wed, 28 Mar 2001 10:15:13 +1000
Received: (from ajag@localhost) by fudge.melbourne.sgi.com (980427.SGI.8.8.8/980728.SGI.AUTOCF) id KAA14066; Wed, 28 Mar 2001 10:15:12 +1000 (EST)
Date: Wed, 28 Mar 2001 10:15:12 +1000
From: Andrew Gildfind <ajag@fudge.melbourne.sgi.com>
To: John Trostel <jtrostel@connex.com>
Cc: linux-xfs@oss.sgi.com
Subject: Re: Problems with ACL inheritance and chacl
Message-ID: <20010328101511.B112360@fudge.melbourne.sgi.com>
References: <XFMail.20010327105142.jtrostel@connex.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
X-Mailer: Mutt 1.0us
In-Reply-To: <XFMail.20010327105142.jtrostel@connex.com>; from jtrostel@connex.com on Tue, Mar 27, 2001 at 10:51:42AM -0500
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

On Tue, Mar 27, 2001 at 10:51:42AM -0500, John Trostel wrote:
> I am seeing some strange behavior with both ACL inheritance and chacl
> operation.  Does anyone else see this or is my build faulty?
> 
> I started with a freshly formatted xfs partition on /mnt/xfs_part.
> 
> I set the access, default and mask ACLs for this parition as follows:
> 
>   [root@jtsdell /mnt]# mount /mnt/xfs_part/
> 
>   [root@jtsdell /mnt]# chacl -l xfs_part/
>   xfs_part/ []
>         
>   [root@jtsdell /mnt]# chacl -b u::rwx,g::rwx,o::rwx,m::rwx u::rwx,g::rwx,o::rwx
>                            xfs_part
> 
>   [root@jtsdell /mnt]# chacl -l xfs_part/
>   xfs_part/ [u::rwx,g::rwx,o::rwx,m::rwx/u::rwx,g::rwx,o::rwx]
> 
> 
> So far... so good!
> 
> 
>   [root@jtsdell /mnt]# exit
>   exit
> 
>   [jt@jtsdell xfs_backup]$ cd ..
> 
>   [jt@jtsdell /mnt]$ cd xfs_part
> 
>   [jt@jtsdell xfs_part]$ ls
> 
>   [jt@jtsdell xfs_part]$ touch jt.txt
> 
>   [jt@jtsdell xfs_part]$ chacl -l jt.txt
>   jt.txt [u::rw-,g::rw-,o::r--]
> 
> Not good... Shouldn't this get the default ACLs u::rwx,g::rwx,o::rwx ?

I'm not precisely sure of the semantics of ACL inheritance, but I noticed
when playing around with this that the inherited ACL was modified by
different umasks... I'll have to look into this further.

> 
>   [jt@jtsdell xfs_part]$ su
>   Password:
> 
>   [root@jtsdell xfs_part]# touch root.txt
>  
>   [root@jtsdell xfs_part]# chacl -l root.txt
>   root.txt [u::rw-,g::rw-,o::r--]
> 
> Hmm... root has the same problem!
> 
>   [root@jtsdell xfs_part]# chacl -B root.txt
>   chacl: error removing access acl on "root.txt": Invalid argument
>   chacl: error removing default acl on "root.txt": Invalid argument
> 
>   [root@jtsdell
>   xfs_part]# chacl -D root.txt
>   chacl: error removing default acl on "root.txt": Invalid argument
> 
> Shouldn't 'chacl' be removing my ACLs here?

It should be, I stumbled across this bug yesterday, I have a fix that should
go in today.

Andrew

-- 
Andrew Gildfind - R&D Software Engineer - SGI Melbourne Australia
email: ajag@sgi.com - work: +61.3.9834.8200 mobile: 0412.834.183

From owner-linux-xfs@oss.sgi.com Tue Mar 27 17:03:06 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2S136216554
	for linux-xfs-outgoing; Tue, 27 Mar 2001 17:03:06 -0800
Received: from pneumatic-tube.sgi.com (pneumatic-tube.sgi.com [204.94.214.22])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2S134M16551
	for <linux-xfs@oss.sgi.com>; Tue, 27 Mar 2001 17:03:04 -0800
Received: from larry.melbourne.sgi.com (larry.melbourne.sgi.com [134.14.52.130]) by pneumatic-tube.sgi.com (980327.SGI.8.8.8-aspam/980310.SGI-aspam) via SMTP id RAA00601
	for <linux-xfs@oss.sgi.com>; Tue, 27 Mar 2001 17:13:09 -0800 (PST)
	mail_from (nathans@wobbly.melbourne.sgi.com)
Received: from wobbly.melbourne.sgi.com (wobbly.melbourne.sgi.com [134.14.55.135]) by larry.melbourne.sgi.com (950413.SGI.8.6.12/950213.SGI.AUTOCF) via ESMTP id LAA03709; Wed, 28 Mar 2001 11:01:31 +1000
Received: (from nathans@localhost) by wobbly.melbourne.sgi.com (980427.SGI.8.8.8/980728.SGI.AUTOCF) id LAA27441; Wed, 28 Mar 2001 11:01:24 +1000 (EST)
From: "Nathan Scott" <nathans@wobbly.melbourne.sgi.com>
Message-Id: <10103281101.ZM257143@wobbly.melbourne.sgi.com>
Date: Wed, 28 Mar 2001 11:01:22 -0500
X-Mailer: Z-Mail (3.2.3 08feb96 MediaMail)
To: linux-xfs@oss.sgi.com
Subject: [announce] Quota tools update
Cc: Jan Kara <jack@ucw.cz>, Marco van Wieringen <mvw@planets.elm.net>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

hi,

For those with an interest in using quota with XFS...

A new version of the quota userspace tools has now been
uploaded to the linuxquota Sourceforge project, at:

    http://sourceforge.net/projects/linuxquota/

The current version is 3.01-pre2 and has full support
for XFS.  It also supports both of the current native
Linux quota formats - i.e. the version in 2.4.X as well
as the version in 2.4.X-acX.

Let me know if you have any problems with these tools,
or alternatively you could send reports to the lists on
the Sourceforge project page above.

The version of the quota tools currently on oss.sgi.com
in the XFS cvs repository will soon be removed.

thanks.

-- 
Nathan

From owner-linux-xfs@oss.sgi.com Wed Mar 28 07:40:50 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2SFeoA06635
	for linux-xfs-outgoing; Wed, 28 Mar 2001 07:40:50 -0800
Received: from sws5.ctd.ornl.gov (sws5.ctd.ornl.gov [160.91.68.105])
	by oss.sgi.com (8.11.3/8.11.3) with SMTP id f2SFeoM06632
	for <linux-xfs@oss.sgi.com>; Wed, 28 Mar 2001 07:40:50 -0800
Received: (qmail 13494 invoked by uid 3995); 28 Mar 2001 15:40:49 -0000
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Message-ID: <15042.1537.463368.336615@sws5.ctd.ornl.gov>
Date: Wed, 28 Mar 2001 10:40:49 -0500 (EST)
From: Dave Sill <ds-linux-xfs@sws5.ctd.ornl.gov>
To: linux-xfs@oss.sgi.com
Subject: Module loading problems
X-Mailer: VM 6.75 under 21.1 (patch 14) "Cuyahoga Valley" XEmacs Lucid
Organization: Oak Ridge National Lab, Oak Ridge, Tenn., USA
X-Face: "p~Q]mg{;e*}YR|)&Q/&Q\*~5UWfZX34;5M<gHyq0KvGt<$qi6=B7;$3~1p?Jc7-g?1vPVAc%YX$E6]%1jvk:,"f*jA|?~Cxb?WJ5gt}L?qsMAjROM~rHbK27yx=t,L/?IHb8@|cYg8"Y)~0IpU~J.^w,VW)u?M3q-AS{f`@RZ]Wll
X-Disclaimer: My opinions do not necessarily represent those of my employer
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

I recently asked about moving an IRIX disk to a Linux box, and learned
that I'd have to build a kernel with support for IRIX partitions.

I checked out the current kernel source using CVS, copied over a
.config file provided by Eric Sandeen, configured in the IRIX
partition support, then did:

  make dep
  make bzImage
  make modules
  make modules_install
  make install
  mkinitrd /boot/initrd-2.4.2-XFS.img 2.4.2-XFS

Everything seemed to go pretty smoothly during the build. When I boot
the new kernel, it's unable to load any modules. For example, if I
try:

# insmod /lib/modules/2.4.2-XFS/kernel/drivers/block/loop.o 
/lib/modules/2.4.2-XFS/kernel/drivers/block/loop.o: unresolved symbol dput
/lib/modules/2.4.2-XFS/kernel/drivers/block/loop.o: unresolved symbol register_disk
/lib/modules/2.4.2-XFS/kernel/drivers/block/loop.o: unresolved symbol __wake_up
/lib/modules/2.4.2-XFS/kernel/drivers/block/loop.o: unresolved symbol blk_cleanup_queue
/lib/modules/2.4.2-XFS/kernel/drivers/block/loop.o: unresolved symbol kmalloc
/lib/modules/2.4.2-XFS/kernel/drivers/block/loop.o: unresolved symbol kdevname
/lib/modules/2.4.2-XFS/kernel/drivers/block/loop.o: unresolved symbol __wait_on_buffer
/lib/modules/2.4.2-XFS/kernel/drivers/block/loop.o: unresolved symbol devfs_mk_dir
/lib/modules/2.4.2-XFS/kernel/drivers/block/loop.o: unresolved symbol end_that_request_last
/lib/modules/2.4.2-XFS/kernel/drivers/block/loop.o: unresolved symbol blkdev_put/lib/modules/2.4.2-XFS/kernel/drivers/block/loop.o: unresolved symbol getblk
/lib/modules/2.4.2-XFS/kernel/drivers/block/loop.o: unresolved symbol try_inc_mod_count
/lib/modules/2.4.2-XFS/kernel/drivers/block/loop.o: unresolved symbol file_moveto
/lib/modules/2.4.2-XFS/kernel/drivers/block/loop.o: unresolved symbol panic
/lib/modules/2.4.2-XFS/kernel/drivers/block/loop.o: unresolved symbol ll_rw_bloc
/lib/modules/2.4.2-XFS/kernel/drivers/block/loop.o: unresolved symbol devfs_unregister_blkdev
/lib/modules/2.4.2-XFS/kernel/drivers/block/loop.o: unresolved symbol devfs_unregister
/lib/modules/2.4.2-XFS/kernel/drivers/block/loop.o: unresolved symbol end_that_request_first
/lib/modules/2.4.2-XFS/kernel/drivers/block/loop.o: unresolved symbol blk_queue_headactive
/lib/modules/2.4.2-XFS/kernel/drivers/block/loop.o: unresolved symbol blk_queue_pluggable
/lib/modules/2.4.2-XFS/kernel/drivers/block/loop.o: unresolved symbol blk_size
/lib/modules/2.4.2-XFS/kernel/drivers/block/loop.o: unresolved symbol is_read_only
/lib/modules/2.4.2-XFS/kernel/drivers/block/loop.o: unresolved symbol mark_buffer_dirty
/lib/modules/2.4.2-XFS/kernel/drivers/block/loop.o: unresolved symbol kfree
/lib/modules/2.4.2-XFS/kernel/drivers/block/loop.o: unresolved symbol do_generic_file_read
/lib/modules/2.4.2-XFS/kernel/drivers/block/loop.o: unresolved symbol blkdev_get/lib/modules/2.4.2-XFS/kernel/drivers/block/loop.o: unresolved symbol devfs_register_series
/lib/modules/2.4.2-XFS/kernel/drivers/block/loop.o: unresolved symbol blk_dev
/lib/modules/2.4.2-XFS/kernel/drivers/block/loop.o: unresolved symbol fput
/lib/modules/2.4.2-XFS/kernel/drivers/block/loop.o: unresolved symbol blk_init_queue
/lib/modules/2.4.2-XFS/kernel/drivers/block/loop.o: unresolved symbol set_device_ro
/lib/modules/2.4.2-XFS/kernel/drivers/block/loop.o: unresolved symbol grab_cache_page
/lib/modules/2.4.2-XFS/kernel/drivers/block/loop.o: unresolved symbol devfs_register_blkdev
/lib/modules/2.4.2-XFS/kernel/drivers/block/loop.o: unresolved symbol __invalidate_buffers
/lib/modules/2.4.2-XFS/kernel/drivers/block/loop.o: unresolved symbol __brelse
/lib/modules/2.4.2-XFS/kernel/drivers/block/loop.o: unresolved symbol fget
/lib/modules/2.4.2-XFS/kernel/drivers/block/loop.o: unresolved symbol __free_pages
/lib/modules/2.4.2-XFS/kernel/drivers/block/loop.o: unresolved symbol get_write_access
/lib/modules/2.4.2-XFS/kernel/drivers/block/loop.o: unresolved symbol printk
/lib/modules/2.4.2-XFS/kernel/drivers/block/loop.o: unresolved symbol get_empty_filp
/lib/modules/2.4.2-XFS/kernel/drivers/block/loop.o: unresolved symbol blksize_size

What am I missing?

-Dave

From owner-linux-xfs@oss.sgi.com Wed Mar 28 07:52:14 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2SFqEb07048
	for linux-xfs-outgoing; Wed, 28 Mar 2001 07:52:14 -0800
Received: from mail.ocs.com.au (ppp0.ocs.com.au [203.34.97.3])
	by oss.sgi.com (8.11.3/8.11.3) with SMTP id f2SFqCM07043
	for <linux-xfs@oss.sgi.com>; Wed, 28 Mar 2001 07:52:12 -0800
Received: (qmail 2253 invoked from network); 28 Mar 2001 15:52:08 -0000
Received: from ocs3.ocs-net (192.168.255.3)
  by mail.ocs.com.au with SMTP; 28 Mar 2001 15:52:08 -0000
X-Mailer: exmh version 2.1.1 10/15/1999
From: Keith Owens <kaos@melbourne.sgi.com>
To: Dave Sill <ds-linux-xfs@sws5.ctd.ornl.gov>
cc: linux-xfs@oss.sgi.com
Subject: Re: Module loading problems 
In-reply-to: Your message of "Wed, 28 Mar 2001 10:40:49 EST."
             <15042.1537.463368.336615@sws5.ctd.ornl.gov> 
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Date: Thu, 29 Mar 2001 01:52:07 +1000
Message-ID: <13262.985794727@ocs3.ocs-net>
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

On Wed, 28 Mar 2001 10:40:49 -0500 (EST), 
Dave Sill <ds-linux-xfs@sws5.ctd.ornl.gov> wrote:
>I checked out the current kernel source using CVS, copied over a
>.config file provided by Eric Sandeen, configured in the IRIX
>partition support, then did:
>
>  make dep
>  make bzImage
>  make modules
>  make modules_install
>  make install
>  mkinitrd /boot/initrd-2.4.2-XFS.img 2.4.2-XFS

Unless you are building a kernel that has to run on multiple machines
with different hardware (i.e. a distribution) I strongly recommend that
you do not use mkinitrd.  Build your root driver and filesystem into
the kernel, load other objects as modules after boot and forget initrd.
It is good for distributors but a pain in the neck for everybody else.

>Everything seemed to go pretty smoothly during the build. When I boot
>the new kernel, it's unable to load any modules. For example, if I
>try:
>
># insmod /lib/modules/2.4.2-XFS/kernel/drivers/block/loop.o 
>/lib/modules/2.4.2-XFS/kernel/drivers/block/loop.o: unresolved symbol dput
>/lib/modules/2.4.2-XFS/kernel/drivers/block/loop.o: unresolved symbol register_disk

Use modprobe loop, not insmod /full/path/name.o.  modprobe is a high
level command which automatically loads any other modules that are
required to satisfy missing symbols, insmod is a low level command
which does exactly what you say and no more.


From owner-linux-xfs@oss.sgi.com Wed Mar 28 07:57:56 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2SFvuB07838
	for linux-xfs-outgoing; Wed, 28 Mar 2001 07:57:56 -0800
Received: from thor.theseus.com (south.orl-pub.theseus.com [12.108.42.66])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2SFvuM07834
	for <linux-xfs@oss.sgi.com>; Wed, 28 Mar 2001 07:57:56 -0800
Received: from theseus.com (IDENT:thebs@fugitive.theseus.com [192.168.0.242])
	by thor.theseus.com (8.9.3/8.9.3) with ESMTP id LAA09240;
	Wed, 28 Mar 2001 11:02:55 -0500
Message-ID: <3AC20A06.59E80D4A@theseus.com>
Date: Wed, 28 Mar 2001 10:57:58 -0500
From: Bryan-TheBS-Smith <thebs@theseus.com>
Reply-To: thebs@theseus.com, b.j.smith@ieee.org
Organization: (Personal)
X-Mailer: Mozilla 4.76 [en] (X11; U; Linux 2.2.17-0.17-FUGITIVE i686)
X-Accept-Language: en
MIME-Version: 1.0
To: Dave Sill <ds-linux-xfs@sws5.ctd.ornl.gov>
CC: linux-xfs@oss.sgi.com, elg-eluglist@elug.org
Subject: Re: Module loading problems
References: <15042.1537.463368.336615@sws5.ctd.ornl.gov>
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

Dave Sill wrote:
> Everything seemed to go pretty smoothly during the build. When I boot
> the new kernel, it's unable to load any modules.  <... cut ...>

Hmmm, try running "depmod -a" -- but I think your "make
modules_install" may have failed to overwrite older modules.

Hence why I always recommend the use of "EXTRAVERSION" in an
alternate "makefile" to "tag" a new kernel with a label _different_
than the existing kernel.  That way you can boot either kernel (new
or old) and not have to worry about modules conflicting with each
other.

Next time, create a file called "makefile" in your /usr/src/linux
directory:

   # Include main Makefile -- add kernel tag to EXTRAVERSION
   include Makefile
   EXTRAVERSION:=$(EXTRAVERSION)-MYBUILD

Then run "make config ..." as normal.  Your kernel
files/modules/dires will be tagged with "-MYBUILD" at the end (or
replace "MYBUILD" with whatever you want).

I have a set of scripts to automate this.  You can find them here:
http://lists.leap-cf.org/pipermail/leaplist/2001-February/009641.html

[ They are a little out of date.  I'll send updated ones directly to
you if you'd like ]

I'm also working on a ncurses-based menuing system that automates
this.  Here's a preliminary screenshot:
http://www.smithconcepts.com/files/conmenu/conmenu_kernel.jpg

-- TheBS

CC: ELUG

--
Bryan "TheBS" Smith          chat:thebs413 @AOL/MSN/Yahoo
Engineer      mailto:b.j.smith@ieee.org,thebs@theseus.com
*********************************************************
"Never apply a Star Trek solution to a Babylon 5 problem"
                                    -- Nicholas C. Weaver

From owner-linux-xfs@oss.sgi.com Wed Mar 28 08:07:07 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2SG77D08273
	for linux-xfs-outgoing; Wed, 28 Mar 2001 08:07:07 -0800
Received: from sws5.ctd.ornl.gov (sws5.ctd.ornl.gov [160.91.68.105])
	by oss.sgi.com (8.11.3/8.11.3) with SMTP id f2SG76M08270
	for <linux-xfs@oss.sgi.com>; Wed, 28 Mar 2001 08:07:06 -0800
Received: (qmail 14310 invoked by uid 3995); 28 Mar 2001 16:07:06 -0000
From: Dave Sill <ds-linux-xfs@sws5.ctd.ornl.gov>
To: linux-xfs@oss.sgi.com
Subject: Re: Module loading problems
References: <13262.985794727@ocs3.ocs-net>
Content-Type: text/plain; charset=US-ASCII
Date: 28 Mar 2001 11:07:05 -0500
In-Reply-To: Keith Owens's message of "Thu, 29 Mar 2001 01:52:07 +1000"
Message-ID: <wx0elvhyj12.fsf@sws5.ctd.ornl.gov>
Lines: 21
User-Agent: Gnus/5.0807 (Gnus v5.8.7) XEmacs/21.1 (Cuyahoga Valley)
MIME-Version: 1.0
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

Keith Owens <kaos@melbourne.sgi.com> writes:

> Unless you are building a kernel that has to run on multiple machines
> with different hardware (i.e. a distribution) I strongly recommend that
> you do not use mkinitrd.  Build your root driver and filesystem into
> the kernel,

How? Via "make xconfig"? Do I just do "lsmod" and configure the listed
modules into the kernel?

> load other objects as modules after boot and forget initrd.
> It is good for distributors but a pain in the neck for everybody else.

So what could be causing the modules not to load?

> Use modprobe loop, not insmod /full/path/name.o.

OK, will do. I presume the boot process does the right thing, though,
and it gets the same errors.

-Dave

From owner-linux-xfs@oss.sgi.com Wed Mar 28 08:16:54 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2SGGsG08844
	for linux-xfs-outgoing; Wed, 28 Mar 2001 08:16:54 -0800
Received: from dsl-squash.corp.sgi.com (nic-30-c48-217.mw.mediaone.net [24.30.48.217])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2SGGrM08841
	for <linux-xfs@oss.sgi.com>; Wed, 28 Mar 2001 08:16:53 -0800
Received: from sgi.com (localhost [127.0.0.1])
	by dsl-squash.corp.sgi.com (Postfix) with ESMTP
	id 5167B46020; Wed, 28 Mar 2001 11:16:28 -0500 (EST)
Message-ID: <3AC20E5C.80D1B236@sgi.com>
Date: Wed, 28 Mar 2001 11:16:28 -0500
From: J Landman <landman@sgi.com>
X-Mailer: Mozilla 4.76 [en] (X11; U; Linux 2.2.18 i686)
X-Accept-Language: en
MIME-Version: 1.0
To: Dave Sill <ds-linux-xfs@sws5.ctd.ornl.gov>
Cc: linux-xfs@oss.sgi.com
Subject: Re: Module loading problems
References: <13262.985794727@ocs3.ocs-net> <wx0elvhyj12.fsf@sws5.ctd.ornl.gov>
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

Dave Sill wrote:
> 
> Keith Owens <kaos@melbourne.sgi.com> writes:
> 
> > Unless you are building a kernel that has to run on multiple machines
> > with different hardware (i.e. a distribution) I strongly recommend that
> > you do not use mkinitrd.  Build your root driver and filesystem into
> > the kernel,
> 
> How? Via "make xconfig"? Do I just do "lsmod" and configure the listed
> modules into the kernel?

Actually you are prompted for a yes/module/no (via radio buttons) from
xconfig.  Select the yes option instead of the module option.

> 
> > load other objects as modules after boot and forget initrd.
> > It is good for distributors but a pain in the neck for everybody else.
> 
> So what could be causing the modules not to load?

Unresolved dependencies with regards to other modules which failed to
load.  Look at your logs (/var/log/syslog) to see what failed to load
and what succeeded.  Try insmoding the failed stuff after booting to
single user mode (type "linux single" at the lilo prompt if your kernel
image is named "linux")

-- 
 Joseph Landman, Ph.D.,   | SGI Americas Technology Center
 email:  landman@sgi.com  | Sr. Systems Engineering Specialist
 voice:  +1 248 427 4469  | Bioinformatics Application Development
   fax:  +1 248 427 5600  | and Linux Cluster based solutions.
--------------------------+---------------------------------------  
http://www.sgi.com/chembio  http://reality.sgi.com/landman  
------------------------------------------------------------------

From owner-linux-xfs@oss.sgi.com Wed Mar 28 08:20:09 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2SGK9w08976
	for linux-xfs-outgoing; Wed, 28 Mar 2001 08:20:09 -0800
Received: from mail.ocs.com.au (ppp0.ocs.com.au [203.34.97.3])
	by oss.sgi.com (8.11.3/8.11.3) with SMTP id f2SGK6M08972
	for <linux-xfs@oss.sgi.com>; Wed, 28 Mar 2001 08:20:07 -0800
Received: (qmail 2535 invoked from network); 28 Mar 2001 16:20:04 -0000
Received: from ocs3.ocs-net (192.168.255.3)
  by mail.ocs.com.au with SMTP; 28 Mar 2001 16:20:04 -0000
X-Mailer: exmh version 2.1.1 10/15/1999
From: Keith Owens <kaos@melbourne.sgi.com>
To: thebs@theseus.com, b.j.smith@ieee.org
cc: Dave Sill <ds-linux-xfs@sws5.ctd.ornl.gov>, linux-xfs@oss.sgi.com,
   elg-eluglist@elug.org
Subject: Re: Module loading problems 
In-reply-to: Your message of "Wed, 28 Mar 2001 10:57:58 EST."
             <3AC20A06.59E80D4A@theseus.com> 
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Date: Thu, 29 Mar 2001 02:20:02 +1000
Message-ID: <13790.985796402@ocs3.ocs-net>
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

On Wed, 28 Mar 2001 10:57:58 -0500, 
Bryan-TheBS-Smith <thebs@theseus.com> wrote:
>Dave Sill wrote:
>> Everything seemed to go pretty smoothly during the build. When I boot
>> the new kernel, it's unable to load any modules.  <... cut ...>
>
>Hmmm, try running "depmod -a" -- but I think your "make
>modules_install" may have failed to overwrite older modules.

Not a problem since 2.4 kernels around August 2000.  Since then, 2.4
make modules_install erases the old modules first, to avoid old
versions being left around.


From owner-linux-xfs@oss.sgi.com Wed Mar 28 08:24:03 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2SGO3e09133
	for linux-xfs-outgoing; Wed, 28 Mar 2001 08:24:03 -0800
Received: from mail.ocs.com.au (ppp0.ocs.com.au [203.34.97.3])
	by oss.sgi.com (8.11.3/8.11.3) with SMTP id f2SGO1M09130
	for <linux-xfs@oss.sgi.com>; Wed, 28 Mar 2001 08:24:01 -0800
Received: (qmail 2600 invoked from network); 28 Mar 2001 16:23:59 -0000
Received: from ocs3.ocs-net (192.168.255.3)
  by mail.ocs.com.au with SMTP; 28 Mar 2001 16:23:59 -0000
X-Mailer: exmh version 2.1.1 10/15/1999
From: Keith Owens <kaos@melbourne.sgi.com>
To: Dave Sill <ds-linux-xfs@sws5.ctd.ornl.gov>
cc: linux-xfs@oss.sgi.com
Subject: Re: Module loading problems 
In-reply-to: Your message of "28 Mar 2001 11:07:05 EST."
             <wx0elvhyj12.fsf@sws5.ctd.ornl.gov> 
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Date: Thu, 29 Mar 2001 02:23:58 +1000
Message-ID: <13823.985796638@ocs3.ocs-net>
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

On 28 Mar 2001 11:07:05 -0500, 
Dave Sill <ds-linux-xfs@sws5.ctd.ornl.gov> wrote:
>How? Via "make xconfig"? Do I just do "lsmod" and configure the listed
>modules into the kernel?

Identify your root driver (IDE or SCSI and which type of SCSI
controller) and your root filesystem.  Select those as Y, not M, using
any of the config utilities.  For example, on my IDE system with ext2
as the root file system I build IDE and ext2 into the kernel.  On my
SCSI system I build SCSI controller NCR53C8XX into the kernel, again
with ext2.

>> Use modprobe loop, not insmod /full/path/name.o.
>
>OK, will do. I presume the boot process does the right thing, though,
>and it gets the same errors.

As root, depmod -ae to get a list of problem modules and symbols.  Then
take a look at http://www.tux.org/lkml/#s8-8, it probably applies.


From owner-linux-xfs@oss.sgi.com Wed Mar 28 08:36:08 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2SGa8n09526
	for linux-xfs-outgoing; Wed, 28 Mar 2001 08:36:08 -0800
Received: from sws5.ctd.ornl.gov (sws5.ctd.ornl.gov [160.91.68.105])
	by oss.sgi.com (8.11.3/8.11.3) with SMTP id f2SGa7M09523
	for <linux-xfs@oss.sgi.com>; Wed, 28 Mar 2001 08:36:07 -0800
Received: (qmail 15290 invoked by uid 3995); 28 Mar 2001 16:36:07 -0000
From: Dave Sill <ds-linux-xfs@sws5.ctd.ornl.gov>
To: linux-xfs@oss.sgi.com
Subject: Re: Module loading problems
References: <13262.985794727@ocs3.ocs-net> <wx0elvhyj12.fsf@sws5.ctd.ornl.gov>
	<3AC20E5C.80D1B236@sgi.com>
Content-Type: text/plain; charset=US-ASCII
Date: 28 Mar 2001 11:36:07 -0500
In-Reply-To: J Landman's message of "Wed, 28 Mar 2001 11:16:28 -0500"
Message-ID: <wx08zlpyhoo.fsf@sws5.ctd.ornl.gov>
Lines: 95
User-Agent: Gnus/5.0807 (Gnus v5.8.7) XEmacs/21.1 (Cuyahoga Valley)
MIME-Version: 1.0
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

J Landman <landman@sgi.com> writes:

> Dave Sill wrote:
> > 
> > Keith Owens <kaos@melbourne.sgi.com> writes:
> > 
> > > Unless you are building a kernel that has to run on multiple machines
> > > with different hardware (i.e. a distribution) I strongly recommend that
> > > you do not use mkinitrd.  Build your root driver and filesystem into
> > > the kernel,
> > 
> > How? Via "make xconfig"? Do I just do "lsmod" and configure the listed
> > modules into the kernel?
> 
> Actually you are prompted for a yes/module/no (via radio buttons) from
> xconfig.  Select the yes option instead of the module option.

OK, let's get one thing out of the way: I'm not a clueless newbie, so
please spare me the newbie answers. Yes, I know xconfig allows you to
configure things in/out/module. My question was how do I determine
exactly which things need to be configured in, and which can be left
as modules, in order to boot without a ram disk.

> > So what could be causing the modules not to load?
> 
> Unresolved dependencies with regards to other modules which failed to
> load.  Look at your logs (/var/log/syslog) to see what failed to load
> and what succeeded.

No modules successfully locaded. Here are the log entries:

Mar 28 10:05:53 malachite modprobe: /lib/modules/2.4.2-XFS/kernel/drivers/usb/usbcore.o: unresolved symbol devfs_unregister_chrdev 
Mar 28 10:05:53 malachite modprobe: /lib/modules/2.4.2-XFS/kernel/drivers/usb/usbcore.o: unresolved symbol schedule_timeout 
Mar 28 10:05:53 malachite modprobe: /lib/modules/2.4.2-XFS/kernel/drivers/usb/usbcore.o: unresolved symbol __wake_up 
Mar 28 10:05:53 malachite modprobe: /lib/modules/2.4.2-XFS/kernel/drivers/usb/usbcore.o: unresolved symbol __generic_copy_from_user 
Mar 28 10:05:53 malachite modprobe: /lib/modules/2.4.2-XFS/kernel/drivers/usb/usbcore.o: unresolved symbol schedule 
Mar 28 10:05:53 malachite modprobe: /lib/modules/2.4.2-XFS/kernel/drivers/usb/usbcore.o: unresolved symbol kmalloc 
Mar 28 10:05:53 malachite modprobe: /lib/modules/2.4.2-XFS/kernel/drivers/usb/usbcore.o: unresolved symbol devfs_mk_dir 
Mar 28 10:05:53 malachite modprobe: /lib/modules/2.4.2-XFS/kernel/drivers/usb/usbcore.o: unresolved symbol proc_bus 
Mar 28 10:05:53 malachite modprobe: /lib/modules/2.4.2-XFS/kernel/drivers/usb/usbcore.o: unresolved symbol __get_free_pages 
Mar 28 10:05:53 malachite modprobe: /lib/modules/2.4.2-XFS/kernel/drivers/usb/usbcore.o: unresolved symbol unregister_filesystem 
Mar 28 10:05:53 malachite modprobe: /lib/modules/2.4.2-XFS/kernel/drivers/usb/usbcore.o: unresolved symbol iput 
Mar 28 10:05:53 malachite modprobe: /lib/modules/2.4.2-XFS/kernel/drivers/usb/usbcore.o: unresolved symbol send_sig_info 
Mar 28 10:05:53 malachite modprobe: /lib/modules/2.4.2-XFS/kernel/drivers/usb/usbcore.o: unresolved symbol remove_wait_queue 
Mar 28 10:05:53 malachite modprobe: /lib/modules/2.4.2-XFS/kernel/drivers/usb/usbcore.o: unresolved symbol try_inc_mod_count 
Mar 28 10:05:53 malachite modprobe: /lib/modules/2.4.2-XFS/kernel/drivers/usb/usbcore.o: unresolved symbol kill_proc 
Mar 28 10:05:53 malachite modprobe: /lib/modules/2.4.2-XFS/kernel/drivers/usb/usbcore.o: unresolved symbol d_rehash 
Mar 28 10:05:53 malachite modprobe: /lib/modules/2.4.2-XFS/kernel/drivers/usb/usbcore.o: unresolved symbol devfs_unregister 
Mar 28 10:05:53 malachite modprobe: /lib/modules/2.4.2-XFS/kernel/drivers/usb/usbcore.o: unresolved symbol register_filesystem 
Mar 28 10:05:53 malachite modprobe: /lib/modules/2.4.2-XFS/kernel/drivers/usb/usbcore.o: unresolved symbol free_pages 
Mar 28 10:05:53 malachite modprobe: /lib/modules/2.4.2-XFS/kernel/drivers/usb/usbcore.o: unresolved symbol proc_mkdir 
Mar 28 10:05:53 malachite modprobe: /lib/modules/2.4.2-XFS/kernel/drivers/usb/usbcore.o: unresolved symbol d_alloc_root 
Mar 28 10:05:53 malachite modprobe: /lib/modules/2.4.2-XFS/kernel/drivers/usb/usbcore.o: unresolved symbol interruptible_sleep_on 
Mar 28 10:05:53 malachite modprobe: /lib/modules/2.4.2-XFS/kernel/drivers/usb/usbcore.o: unresolved symbol __pollwait 
Mar 28 10:05:53 malachite modprobe: /lib/modules/2.4.2-XFS/kernel/drivers/usb/usbcore.o: unresolved symbol kfree 
Mar 28 10:05:53 malachite modprobe: /lib/modules/2.4.2-XFS/kernel/drivers/usb/usbcore.o: unresolved symbol remove_proc_entry 
Mar 28 10:05:53 malachite modprobe: /lib/modules/2.4.2-XFS/kernel/drivers/usb/usbcore.o: unresolved symbol call_usermodehelper 
Mar 28 10:05:53 malachite modprobe: /lib/modules/2.4.2-XFS/kernel/drivers/usb/usbcore.o: unresolved symbol devfs_register_chrdev 
Mar 28 10:05:53 malachite modprobe: /lib/modules/2.4.2-XFS/kernel/drivers/usb/usbcore.o: unresolved symbol hotplug_path 
Mar 28 10:05:53 malachite modprobe: /lib/modules/2.4.2-XFS/kernel/drivers/usb/usbcore.o: unresolved symbol add_wait_queue 
Mar 28 10:05:53 malachite modprobe: /lib/modules/2.4.2-XFS/kernel/drivers/usb/usbcore.o: unresolved symbol xtime 
Mar 28 10:05:53 malachite modprobe: /lib/modules/2.4.2-XFS/kernel/drivers/usb/usbcore.o: unresolved symbol simple_strtoul 
Mar 28 10:05:53 malachite modprobe: /lib/modules/2.4.2-XFS/kernel/drivers/usb/usbcore.o: unresolved symbol sprintf 
Mar 28 10:05:53 malachite modprobe: /lib/modules/2.4.2-XFS/kernel/drivers/usb/usbcore.o: unresolved symbol daemonize 
Mar 28 10:05:53 malachite modprobe: /lib/modules/2.4.2-XFS/kernel/drivers/usb/usbcore.o: unresolved symbol d_instantiate 
Mar 28 10:05:53 malachite modprobe: /lib/modules/2.4.2-XFS/kernel/drivers/usb/usbcore.o: unresolved symbol strtok 
Mar 28 10:05:53 malachite modprobe: /lib/modules/2.4.2-XFS/kernel/drivers/usb/usbcore.o: unresolved symbol iget4 
Mar 28 10:05:53 malachite modprobe: /lib/modules/2.4.2-XFS/kernel/drivers/usb/usbcore.o: unresolved symbol printk 
Mar 28 10:05:53 malachite modprobe: /lib/modules/2.4.2-XFS/kernel/drivers/usb/usbcore.o: unresolved symbol irq_stat 
Mar 28 10:05:53 malachite modprobe: /lib/modules/2.4.2-XFS/kernel/drivers/usb/usbcore.o: unresolved symbol kernel_thread 
Mar 28 10:05:53 malachite modprobe: /lib/modules/2.4.2-XFS/kernel/drivers/usb/usbcore.o: unresolved symbol __const_udelay 
Mar 28 10:05:53 malachite modprobe: /lib/modules/2.4.2-XFS/kernel/drivers/usb/usbcore.o: unresolved symbol __generic_copy_to_user 
Mar 28 10:05:53 malachite modprobe: /lib/modules/2.4.2-XFS/kernel/drivers/usb/usbcore.o: unresolved symbol up_and_exit 
Mar 28 10:05:53 malachite modprobe: /lib/modules/2.4.2-XFS/kernel/drivers/usb/usbcore.o: insmod /lib/modules/2.4.2-XFS/kernel/drivers/usb/usbcore.o failed 
Mar 28 10:05:53 malachite modprobe: /lib/modules/2.4.2-XFS/kernel/drivers/usb/usbcore.o: insmod usb-uhci failed 
Mar 28 10:05:54 malachite rc.sysinit: Loading sound module (emu10k1):  failed 
Mar 28 10:05:54 malachite modprobe: /lib/modules/2.4.2-XFS/kernel/drivers/sound/soundcore.o: unresolved symbol devfs_unregister_chrdev 
Mar 28 10:05:54 malachite modprobe: /lib/modules/2.4.2-XFS/kernel/drivers/sound/soundcore.o: unresolved symbol devfs_register 
Mar 28 10:05:54 malachite modprobe: /lib/modules/2.4.2-XFS/kernel/drivers/sound/soundcore.o: unresolved symbol kmalloc 
Mar 28 10:05:54 malachite modprobe: /lib/modules/2.4.2-XFS/kernel/drivers/sound/soundcore.o: unresolved symbol devfs_mk_dir 
Mar 28 10:05:54 malachite modprobe: /lib/modules/2.4.2-XFS/kernel/drivers/sound/soundcore.o: unresolved symbol vfree 
Mar 28 10:05:54 malachite modprobe: /lib/modules/2.4.2-XFS/kernel/drivers/sound/soundcore.o: unresolved symbol try_inc_mod_count 
Mar 28 10:05:54 malachite modprobe: /lib/modules/2.4.2-XFS/kernel/drivers/sound/soundcore.o: unresolved symbol sys_close 
Mar 28 10:05:54 malachite modprobe: /lib/modules/2.4.2-XFS/kernel/drivers/sound/soundcore.o: unresolved symbol devfs_unregister 
Mar 28 10:05:54 malachite modprobe: /lib/modules/2.4.2-XFS/kernel/drivers/sound/soundcore.o: unresolved symbol request_module 
Mar 28 10:05:54 malachite modprobe: /lib/modules/2.4.2-XFS/kernel/drivers/sound/soundcore.o: unresolved symbol devfs_register_chrdev 
Mar 28 10:05:54 malachite modprobe: /lib/modules/2.4.2-XFS/kernel/drivers/sound/soundcore.o: unresolved symbol sprintf 
Mar 28 10:05:54 malachite modprobe: /lib/modules/2.4.2-XFS/kernel/drivers/sound/soundcore.o: unresolved symbol __vmalloc 
Mar 28 10:05:54 malachite modprobe: /lib/modules/2.4.2-XFS/kernel/drivers/sound/soundcore.o: unresolved symbol printk 
Mar 28 10:05:54 malachite modprobe: /lib/modules/2.4.2-XFS/kernel/drivers/sound/soundcore.o: insmod /lib/modules/2.4.2-XFS/kernel/drivers/sound/soundcore.o failed 
Mar 28 10:05:54 malachite modprobe: /lib/modules/2.4.2-XFS/kernel/drivers/sound/soundcore.o: insmod emu10k1 failed 

I don't know why only usb and sound were logged.

-Dave

From owner-linux-xfs@oss.sgi.com Wed Mar 28 09:49:49 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2SHnnc10970
	for linux-xfs-outgoing; Wed, 28 Mar 2001 09:49:49 -0800
Received: from hub-slc.firsthealth.com (hub-slc.firsthealth.com [209.180.88.35])
	by oss.sgi.com (8.11.3/8.11.3) with SMTP id f2SHnmM10967
	for <linux-xfs@oss.sgi.com>; Wed, 28 Mar 2001 09:49:48 -0800
Received: from 10.1.114.34 by hub-slc.firsthealth.com with SMTP ( SMTP
 Relay (MMS v4.7)); Wed, 28 Mar 2001 10:53:37 -0700
X-Server-Uuid: 4bf5d58d-3e8a-11d3-966e-00508b4fb619
Received: from xnote.com (mcpherve1.firsthealth.com [10.1.17.190]) by
 SLCM02.firsthealth.com; Wed, 28 Mar 2001 10:49:38 -0700
Message-ID: <3AC2245B.4040805@xnote.com>
Date: Wed, 28 Mar 2001 10:50:19 -0700
From: "Vernon McPherron" <vernon@xnote.com>
Reply-to: vernon@xnote.com
User-Agent: Mozilla/5.0 (Windows; U; WinNT4.0; en-US; 0.8)
 Gecko/20010215
X-Accept-Language: en
MIME-Version: 1.0
To: linux-xfs@oss.sgi.com
Subject: Re: Module loading problems
X-WSS-ID: 16DCFAAB132846-01-01
Content-Type: text/plain; 
 charset=us-ascii; 
 format=flowed
Content-Transfer-Encoding: 7bit
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

 >>> So what could be causing the modules not to load?
 >>
 >> Unresolved dependencies with regards to other modules which failed to
 >> load.  Look at your logs (/var/log/syslog) to see what failed to load
 >> and what succeeded.
 >
 >
 > No modules successfully locaded. Here are the log entries:


Well actually you may have to do a
make mrpropper
and then copy your .config over.

I had problems with modules (with fresh CVS tree) until I did a make 
mrpropper and then the make menuconfig, etc...
-- 


-=/Vernon McPherron/=-



From owner-linux-xfs@oss.sgi.com Wed Mar 28 10:46:14 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2SIkE012458
	for linux-xfs-outgoing; Wed, 28 Mar 2001 10:46:14 -0800
Received: from sws5.ctd.ornl.gov (sws5.ctd.ornl.gov [160.91.68.105])
	by oss.sgi.com (8.11.3/8.11.3) with SMTP id f2SIkDM12455
	for <linux-xfs@oss.sgi.com>; Wed, 28 Mar 2001 10:46:13 -0800
Received: (qmail 19468 invoked by uid 3995); 28 Mar 2001 18:46:13 -0000
From: Dave Sill <ds-linux-xfs@sws5.ctd.ornl.gov>
To: linux-xfs@oss.sgi.com
Subject: Re: Module loading problems
References: <13823.985796638@ocs3.ocs-net>
Content-Type: text/plain; charset=US-ASCII
Date: 28 Mar 2001 13:46:13 -0500
In-Reply-To: Keith Owens's message of "Thu, 29 Mar 2001 02:23:58 +1000"
Message-ID: <wx0vgotratm.fsf@sws5.ctd.ornl.gov>
Lines: 12
User-Agent: Gnus/5.0807 (Gnus v5.8.7) XEmacs/21.1 (Cuyahoga Valley)
MIME-Version: 1.0
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

Keith Owens <kaos@melbourne.sgi.com> writes:

> As root, depmod -ae to get a list of problem modules and symbols.  Then
> take a look at http://www.tux.org/lkml/#s8-8, it probably applies.

Bingo.

I've now successfully mounted the IRIX filesystem.

Thanks, everybody!

-Dave

From owner-linux-xfs@oss.sgi.com Wed Mar 28 11:31:56 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2SJVus13551
	for linux-xfs-outgoing; Wed, 28 Mar 2001 11:31:56 -0800
Received: from pneumatic-tube.sgi.com (pneumatic-tube.sgi.com [204.94.214.22])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2SJVsM13548
	for <linux-xfs@oss.sgi.com>; Wed, 28 Mar 2001 11:31:54 -0800
Received: from ledzep.americas.sgi.com (ledzep.americas.sgi.com [137.38.226.97]) by pneumatic-tube.sgi.com (980327.SGI.8.8.8-aspam/980310.SGI-aspam) via ESMTP id LAA05115
	for <linux-xfs@oss.sgi.com>; Wed, 28 Mar 2001 11:41:53 -0800 (PST)
	mail_from (cattelan@thebarn.com)
Received: from gibble.americas.sgi.com (gibble.americas.sgi.com [128.162.195.80]) by ledzep.americas.sgi.com (SGI-SGI-8.9.3/americas-smart-nospam1.1) with ESMTP id NAA09843; Wed, 28 Mar 2001 13:30:00 -0600 (CST)
Received: from thebarn.com (localhost [127.0.0.1])
	by gibble.americas.sgi.com (8.11.2/8.11.2) with ESMTP id f2SJJpe09937;
	Wed, 28 Mar 2001 14:19:51 -0500
Message-ID: <3AC23956.598060DA@thebarn.com>
Date: Wed, 28 Mar 2001 14:19:50 -0500
From: Russell Cattelan <cattelan@thebarn.com>
X-Mailer: Mozilla 4.76 [en] (X11; U; Linux 2.4.2-XFS i686)
X-Accept-Language: en
MIME-Version: 1.0
To: Keith Owens <kaos@melbourne.sgi.com>
CC: Dave Sill <ds-linux-xfs@sws5.ctd.ornl.gov>, linux-xfs@oss.sgi.com
Subject: Re: Module loading problems
References: <13262.985794727@ocs3.ocs-net>
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

Keith Owens wrote:

> On Wed, 28 Mar 2001 10:40:49 -0500 (EST),
> Dave Sill <ds-linux-xfs@sws5.ctd.ornl.gov> wrote:
> >I checked out the current kernel source using CVS, copied over a
> >.config file provided by Eric Sandeen, configured in the IRIX
> >partition support, then did:
> >
> >  make dep
> >  make bzImage
> >  make modules
> >  make modules_install
> >  make install
> >  mkinitrd /boot/initrd-2.4.2-XFS.img 2.4.2-XFS
>
> Unless you are building a kernel that has to run on multiple machines
> with different hardware (i.e. a distribution) I strongly recommend that
> you do not use mkinitrd.  Build your root driver and filesystem into
> the kernel, load other objects as modules after boot and forget initrd.
> It is good for distributors but a pain in the neck for everybody else.

I run almost all my system with XFS and the scsi drivers as modules loaded via initrd
and have very few problems with the setup. The only drawback is that on RH systems it
does take a bit of time to load each module, most of my systems are Mandrake systems
and they load modules almost instantly... I have no idea what the differences might be.

But Keith is correct if your building custom kernels it's easier  to just compile
everything in.

>
>
> >Everything seemed to go pretty smoothly during the build. When I boot
> >the new kernel, it's unable to load any modules. For example, if I
> >try:
> >
> ># insmod /lib/modules/2.4.2-XFS/kernel/drivers/block/loop.o
> >/lib/modules/2.4.2-XFS/kernel/drivers/block/loop.o: unresolved symbol dput
> >/lib/modules/2.4.2-XFS/kernel/drivers/block/loop.o: unresolved symbol register_disk
>
> Use modprobe loop, not insmod /full/path/name.o.  modprobe is a high
> level command which automatically loads any other modules that are
> required to satisfy missing symbols, insmod is a low level command
> which does exactly what you say and no more.

--
Russell Cattelan
--
Digital Elves inc. -- Currently on loan to SGI
Linux XFS core developer.




From owner-linux-xfs@oss.sgi.com Wed Mar 28 11:32:36 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2SJWaR13575
	for linux-xfs-outgoing; Wed, 28 Mar 2001 11:32:36 -0800
Received: from wiley.ceo.com (66-2-81-26.customer.algx.net [66.2.81.26])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2SJWGM13562
	for <linux-xfs@oss.sgi.com>; Wed, 28 Mar 2001 11:32:16 -0800
Received: from mindspring.com (IDENT:danny@localhost [127.0.0.1])
	by wiley.ceo.com (8.9.3/8.9.3) with ESMTP id OAA02631
	for <linux-xfs@oss.sgi.com>; Wed, 28 Mar 2001 14:45:49 -0500
Message-ID: <3AC23F6D.60E56609@mindspring.com>
Date: Wed, 28 Mar 2001 14:45:49 -0500
From: Danny <danscox@mindspring.com>
Reply-To: dcox@connex.com
Organization: Connex Inc
X-Mailer: Mozilla 4.76 [en] (X11; U; Linux 2.4.2-XFS i686)
X-Accept-Language: en
MIME-Version: 1.0
To: Linux-XFS <linux-xfs@oss.sgi.com>
Subject: Unmount Didn't; Clues?
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

All,

	Updated from CVS this AM, this just happened to me:

======================================================================================
Mar 29 03:12:26 dsc_proto_1 kernel: Start mounting filesystem: md(9,1) 
Mar 29 03:12:27 dsc_proto_1 kernel: Ending clean XFS mount for
filesystem: md(9,1) 
Mar 29 03:19:15 dsc_proto_1 kernel: XFS unmount got error 38 
Mar 29 03:19:15 dsc_proto_1 kernel: linvfs_put_super: vfsp/0xc3bc2ae0
left dangling! 
Mar 29 03:19:15 dsc_proto_1 kernel: VFS: Busy inodes after unmount.
Self-destruct in 5 seconds.  Have a nice day... 
Mar 29 03:19:56 dsc_proto_1 kernel: XFS: Filesystem has duplicate UUID -
can't mount 
Mar 29 03:21:35 dsc_proto_1 last message repeated 2 times
=====================================================================================

	Now: I'm pretty sure that a reboot would allow the mount to succeed; in
other words, /dev/md1 is still "mounted" in some way in the kernel;
hence the dup UUID.

	This machine is a test machine only.  Is there anything you'd like me
to try before I reboot?

	Have a good 'un!

-- 
"Men occasionally stumble over the truth, but most of them pick 
themselves up and hurry off as if nothing had happened." 
   -- Winston Churchill 

Danny

From owner-linux-xfs@oss.sgi.com Wed Mar 28 11:52:28 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2SJqSB14288
	for linux-xfs-outgoing; Wed, 28 Mar 2001 11:52:28 -0800
Received: from wiley.ceo.com (66-2-81-26.customer.algx.net [66.2.81.26])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2SJqMM14283
	for <linux-xfs@oss.sgi.com>; Wed, 28 Mar 2001 11:52:23 -0800
Received: from mindspring.com (IDENT:danny@localhost [127.0.0.1])
	by wiley.ceo.com (8.9.3/8.9.3) with ESMTP id PAA02657;
	Wed, 28 Mar 2001 15:02:42 -0500
Message-ID: <3AC24362.90A7B8AF@mindspring.com>
Date: Wed, 28 Mar 2001 15:02:42 -0500
From: Danny <danscox@mindspring.com>
Reply-To: dcox@connex.com
Organization: Connex Inc
X-Mailer: Mozilla 4.76 [en] (X11; U; Linux 2.4.2-XFS i686)
X-Accept-Language: en
MIME-Version: 1.0
To: Russell Cattelan <cattelan@thebarn.com>
CC: linux-xfs@oss.sgi.com
Subject: Re: Module loading problems
References: <13262.985794727@ocs3.ocs-net> <3AC23956.598060DA@thebarn.com>
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

Russell,

Russell Cattelan wrote:
> But Keith is correct if your building custom kernels it's easier  to just compile
> everything in.

	And safer!  During one testing phase, I almost locked myself out: the
xfs modules wouldn't load, and guess where the source was?  On an XFS
fs!  Whups!  I've built it in from then on....

-- 
"Men occasionally stumble over the truth, but most of them pick 
themselves up and hurry off as if nothing had happened." 
   -- Winston Churchill 

Danny

From owner-linux-xfs@oss.sgi.com Wed Mar 28 12:41:47 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2SKfl115558
	for linux-xfs-outgoing; Wed, 28 Mar 2001 12:41:47 -0800
Received: from as104.tel.hr (as104.tel.hr [195.29.150.2])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2SKfjM15555
	for <linux-xfs@oss.sgi.com>; Wed, 28 Mar 2001 12:41:46 -0800
Received: from kc.hinet.hr (ad9-m15.net.hinet.hr [195.29.137.15])
	by as104.tel.hr (0.0.0/0.0.0) with ESMTP id WAA300974
	for <linux-xfs@oss.sgi.com>; Wed, 28 Mar 2001 22:41:43 +0200 (MEST)
Message-ID: <3AC24D27.DAF2A23B@kc.hinet.hr>
Date: Wed, 28 Mar 2001 22:44:23 +0200
From: Mario <mario.kurtovic@kc.hinet.hr>
X-Mailer: Mozilla 4.75 [en] (X11; U; Linux 2.4.0-SGI_XFS.0 i686)
X-Accept-Language: en
MIME-Version: 1.0
To: linux-xfs@oss.sgi.com
Subject: gcc 3.0 prerelease
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

Successfully compile latest cvs tree with  
gcc 3.0 prerelease (Codesourcery build 20010318).
Kernel is up and running, no crash etc. My standard testing query on
Interbase database (2 million records) run very well. I try more
testing, Sybase enterprise server, samba (big access and clarion
database files), iozone benchmark.

Best regards,

Mario Kurtovic

From owner-linux-xfs@oss.sgi.com Wed Mar 28 13:35:02 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2SLZ2V16630
	for linux-xfs-outgoing; Wed, 28 Mar 2001 13:35:02 -0800
Received: from pneumatic-tube.sgi.com (pneumatic-tube.sgi.com [204.94.214.22])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2SLYrM16621
	for <linux-xfs@oss.sgi.com>; Wed, 28 Mar 2001 13:34:53 -0800
Received: from ledzep.americas.sgi.com (ledzep.americas.sgi.com [137.38.226.97]) by pneumatic-tube.sgi.com (980327.SGI.8.8.8-aspam/980310.SGI-aspam) via ESMTP id NAA09898
	for <linux-xfs@oss.sgi.com>; Wed, 28 Mar 2001 13:44:59 -0800 (PST)
	mail_from (cattelan@thebarn.com)
Received: from gibble.americas.sgi.com (gibble.americas.sgi.com [128.162.195.80]) by ledzep.americas.sgi.com (SGI-SGI-8.9.3/americas-smart-nospam1.1) with ESMTP id PAA64183; Wed, 28 Mar 2001 15:33:36 -0600 (CST)
Received: from thebarn.com (localhost [127.0.0.1])
	by gibble.americas.sgi.com (8.11.2/8.11.2) with ESMTP id f2SLNRe26167;
	Wed, 28 Mar 2001 16:23:27 -0500
Message-ID: <3AC2564F.9FA3C2CD@thebarn.com>
Date: Wed, 28 Mar 2001 16:23:27 -0500
From: Russell Cattelan <cattelan@thebarn.com>
X-Mailer: Mozilla 4.76 [en] (X11; U; Linux 2.4.2-XFS i686)
X-Accept-Language: en
MIME-Version: 1.0
To: Mario <mario.kurtovic@kc.hinet.hr>
CC: linux-xfs@oss.sgi.com
Subject: Re: gcc 3.0 prerelease
References: <3AC24D27.DAF2A23B@kc.hinet.hr>
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

Mario wrote:

> Successfully compile latest cvs tree with
> gcc 3.0 prerelease (Codesourcery build 20010318).
> Kernel is up and running, no crash etc. My standard testing query on
> Interbase database (2 million records) run very well. I try more
> testing, Sybase enterprise server, samba (big access and clarion
> database files), iozone benchmark.
>
> Best regards,
>
> Mario Kurtovic

good news... keep us updated.


--
Russell Cattelan
--
Digital Elves inc. -- Currently on loan to SGI
Linux XFS core developer.




From owner-linux-xfs@oss.sgi.com Wed Mar 28 14:38:09 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2SMc9n18108
	for linux-xfs-outgoing; Wed, 28 Mar 2001 14:38:09 -0800
Received: from sgi.com (sgi.SGI.COM [192.48.153.1])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2SMc7M18105
	for <linux-xfs@oss.sgi.com>; Wed, 28 Mar 2001 14:38:08 -0800
Received: from larry.melbourne.sgi.com ([134.14.52.130]) 
	by sgi.com (980327.SGI.8.8.8-aspam/980304.SGI-aspam:
       SGI does not authorize the use of its proprietary
       systems or networks for unsolicited or bulk email
       from the Internet.) 
	via SMTP id OAA06274
	for <linux-xfs@oss.sgi.com>; Wed, 28 Mar 2001 14:38:05 -0800 (PST)
	mail_from (dxm@clouds.melbourne.sgi.com)
Received: from clouds.melbourne.sgi.com (clouds.melbourne.sgi.com [134.14.55.166]) by larry.melbourne.sgi.com (950413.SGI.8.6.12/950213.SGI.AUTOCF) via ESMTP id IAA11311; Thu, 29 Mar 2001 08:36:39 +1000
Received: from clouds.melbourne.sgi.com (localhost [127.0.0.1]) by clouds.melbourne.sgi.com (980427.SGI.8.8.8/980728.SGI.AUTOCF) via ESMTP id IAA19159; Thu, 29 Mar 2001 08:36:16 +1000 (EST)
Message-Id: <200103282236.IAA19159@clouds.melbourne.sgi.com>
X-Mailer: exmh version 2.0.2 2/24/98
To: dcox@connex.com
cc: linux-xfs@oss.sgi.com
Subject: Re: Unmount Didn't; Clues? 
In-reply-to: Your message of "Wed, 28 Mar 2001 14:45:49 EST."
             <3AC23F6D.60E56609@mindspring.com> 
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Date: Thu, 29 Mar 2001 08:36:16 +1000
From: Daniel Moore <dxm@clouds.melbourne.sgi.com>
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk


Danny writes:
 => All,
 => Mar 29 03:19:15 dsc_proto_1 kernel: XFS unmount got error 38 
 => Mar 29 03:19:15 dsc_proto_1 kernel: linvfs_put_super: vfsp/0xc3bc2ae0
 => left dangling! 
 => Mar 29 03:19:15 dsc_proto_1 kernel: VFS: Busy inodes after unmount.
 => Self-destruct in 5 seconds.  Have a nice day... 
 => Mar 29 03:19:56 dsc_proto_1 kernel: XFS: Filesystem has duplicate UUID -
 => can't mount 

 => 	Now: I'm pretty sure that a reboot would allow the mount to succeed; in
 => other words, /dev/md1 is still "mounted" in some way in the kernel;
 => hence the dup UUID.

If xfs_unmountfs doesn't get called, the uuid for the mounted xfs fs won't
get removed from the uuid table, and further mounts of that filesystem
will be rejected. There are a couple of paths in which the unmount can
fail to call xfs_unmountfs, but they shouldn't happen unless something
is really wrong anyway.

What I'm really curious about is why the unmount failed in the first
place - that definitely shouldn't be happening.

Can you give us some vitals on your setup? Are you using highmem, quota, 
acls, dmapi etc? Running any specific tests?

 => 	This machine is a test machine only.  Is there anything you'd like me
 => to try before I reboot?

I think it's too late now. If you can work out how to repeat this problem,
we should be able to suggest some ways to debug it.

Regards,
-----------------------------------------------------
 Daniel Moore                  dxm@sgi.com
 R&D Software Engineer         Phone: +61-3-98348209
 SGI Performance Tools Group   Fax:   +61-3-98132378
-----------------------------------------------------


From owner-linux-xfs@oss.sgi.com Wed Mar 28 15:07:48 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2SN7mt18770
	for linux-xfs-outgoing; Wed, 28 Mar 2001 15:07:48 -0800
Received: from sgi.com (sgi.SGI.COM [192.48.153.1])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2SN7lM18767
	for <linux-xfs@oss.sgi.com>; Wed, 28 Mar 2001 15:07:47 -0800
Received: from zeus-fddi.americas.sgi.com (128-162-8-103.americas.sgi.com [128.162.8.103]) 
	by sgi.com (980327.SGI.8.8.8-aspam/980304.SGI-aspam:
       SGI does not authorize the use of its proprietary
       systems or networks for unsolicited or bulk email
       from the Internet.) 
	via ESMTP id PAA02003
	for <linux-xfs@oss.sgi.com>; Wed, 28 Mar 2001 15:07:43 -0800 (PST)
	mail_from (mann@sgi.com)
Received: from daisy-e185.americas.sgi.com (daisy-e194.americas.sgi.com [128.162.194.214]) by zeus-fddi.americas.sgi.com (8.9.3/americas-smart-nospam1.1) with ESMTP id RAA1294768 for <linux-xfs@oss.sgi.com>; Wed, 28 Mar 2001 17:06:24 -0600 (CST)
Received: from fsgi632.americas.sgi.com (fsgi632.americas.sgi.com [128.162.184.134]) by daisy-e185.americas.sgi.com (SGI-8.9.3/SGI-server-1.7) with ESMTP id RAA24591 for <linux-xfs@oss.sgi.com>; Wed, 28 Mar 2001 17:06:24 -0600 (CST)
From: Mark Nordstrand <mann@sgi.com>
Received: by fsgi632.americas.sgi.com (SGI-8.9.3/SGI-client-1.7) id RAA11098; Wed, 28 Mar 2001 17:06:23 -0600 (CST)
Message-Id: <200103282306.RAA11098@fsgi632.americas.sgi.com>
Subject: Re: Unmount Didn't; Clues?
To: linux-xfs@oss.sgi.com
Date: Wed, 28 Mar 2001 17:06:23 -0600 (CST)
X-Mailer: ELM [version 2.5 PL2]
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

Since I malformed the address this didn't make it to the list:

>
>
> Danny writes:
>  => All,
>  => Mar 29 03:19:15 dsc_proto_1 kernel: XFS unmount got error 38
>  => Mar 29 03:19:15 dsc_proto_1 kernel: linvfs_put_super: vfsp/0xc3bc2ae0
>  => left dangling!
>  => Mar 29 03:19:15 dsc_proto_1 kernel: VFS: Busy inodes after unmount.
>  => Self-destruct in 5 seconds.  Have a nice day...
>  => Mar 29 03:19:56 dsc_proto_1 kernel: XFS: Filesystem has duplicate UUID -
>  => can't mount
>
>  =>   Now: I'm pretty sure that a reboot would allow the mount to succeed; in
>  => other words, /dev/md1 is still "mounted" in some way in the kernel;
>  => hence the dup UUID.
>
> If xfs_unmountfs doesn't get called, the uuid for the mounted xfs fs won't
> get removed from the uuid table, and further mounts of that filesystem
> will be rejected. There are a couple of paths in which the unmount can
> fail to call xfs_unmountfs, but they shouldn't happen unless something
> is really wrong anyway.
>
> What I'm really curious about is why the unmount failed in the first
> place - that definitely shouldn't be happening.
>
Curious, I just hit this running a forced shutdown test.  Different
error, same effect.  In my case, it definitely won't follow the code
path to call xfs_unmountfs().  In either event, it doesn't look like
the filesystem is mounted in any useful sense.

I am curious to know how it got here.  Any information would be
appreciated.

> Can you give us some vitals on your setup? Are you using highmem, quota,
> acls, dmapi etc? Running any specific tests?
>
>  =>   This machine is a test machine only.  Is there anything you'd like me
>  => to try before I reboot?
>
> I think it's too late now. If you can work out how to repeat this problem,
> we should be able to suggest some ways to debug it.
>
If I read the code right, there isn't much else which can be done besides
rebooting.....

Mark



From owner-linux-xfs@oss.sgi.com Wed Mar 28 15:56:19 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2SNuJA19578
	for linux-xfs-outgoing; Wed, 28 Mar 2001 15:56:19 -0800
Received: from deliverator.sgi.com (deliverator.sgi.com [204.94.214.10])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2SNuHM19575
	for <linux-xfs@oss.sgi.com>; Wed, 28 Mar 2001 15:56:17 -0800
Received: from zeus-fddi.americas.sgi.com (zeus-fddi.americas.sgi.com [128.162.8.103]) by deliverator.sgi.com (980309.SGI.8.8.8-aspam-6.2/980310.SGI-aspam) via ESMTP id PAA04600
	for <linux-xfs@oss.sgi.com>; Wed, 28 Mar 2001 15:55:03 -0800 (PST)
	mail_from (lord@sgi.com)
Received: from daisy-e185.americas.sgi.com (daisy-e194.americas.sgi.com [128.162.194.214]) by zeus-fddi.americas.sgi.com (8.9.3/americas-smart-nospam1.1) with ESMTP id RAA1260366 for <linux-xfs@oss.sgi.com>; Wed, 28 Mar 2001 17:55:01 -0600 (CST)
Received: from jen.americas.sgi.com (IDENT:root@jen.americas.sgi.com [128.162.184.86]) by daisy-e185.americas.sgi.com (SGI-8.9.3/SGI-server-1.7) with ESMTP id RAA73144; Wed, 28 Mar 2001 17:55:01 -0600 (CST)
Received: from jen.americas.sgi.com by jen.americas.sgi.com (8.11.0/SGI-client-1.7) via ESMTP id f2SNrvO02630; Wed, 28 Mar 2001 17:53:57 -0600
Message-Id: <200103282353.f2SNrvO02630@jen.americas.sgi.com>
X-Mailer: exmh version 2.2 06/23/2000 with nmh-1.0.4
To: Mark Nordstrand <mann@sgi.com>
cc: linux-xfs@oss.sgi.com
Subject: Re: Unmount Didn't; Clues? 
In-Reply-To: Message from Mark Nordstrand <mann@sgi.com> 
   of "Wed, 28 Mar 2001 17:06:23 CST." <200103282306.RAA11098@fsgi632.americas.sgi.com> 
Date: Wed, 28 Mar 2001 17:53:57 -0600
From: Steve Lord <lord@sgi.com>
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk


So we noticed that EWRONGFS got remapped to ENOSYS, but that is purely
a mount error, not an unmount. But, could this be to do with the errno
remapping changes in xfs_linux.h on Mar 13th?

Steve

> Since I malformed the address this didn't make it to the list:
> 
> >
> >
> > Danny writes:
> >  => All,
> >  => Mar 29 03:19:15 dsc_proto_1 kernel: XFS unmount got error 38
> >  => Mar 29 03:19:15 dsc_proto_1 kernel: linvfs_put_super: vfsp/0xc3bc2ae0
> >  => left dangling!
> >  => Mar 29 03:19:15 dsc_proto_1 kernel: VFS: Busy inodes after unmount.
> >  => Self-destruct in 5 seconds.  Have a nice day...
> >  => Mar 29 03:19:56 dsc_proto_1 kernel: XFS: Filesystem has duplicate UUID 
> -
> >  => can't mount
> >
> >  =>   Now: I'm pretty sure that a reboot would allow the mount to succeed; 
> in
> >  => other words, /dev/md1 is still "mounted" in some way in the kernel;
> >  => hence the dup UUID.
> >
> > If xfs_unmountfs doesn't get called, the uuid for the mounted xfs fs won't
> > get removed from the uuid table, and further mounts of that filesystem
> > will be rejected. There are a couple of paths in which the unmount can
> > fail to call xfs_unmountfs, but they shouldn't happen unless something
> > is really wrong anyway.
> >
> > What I'm really curious about is why the unmount failed in the first
> > place - that definitely shouldn't be happening.
> >
> Curious, I just hit this running a forced shutdown test.  Different
> error, same effect.  In my case, it definitely won't follow the code
> path to call xfs_unmountfs().  In either event, it doesn't look like
> the filesystem is mounted in any useful sense.
> 
> I am curious to know how it got here.  Any information would be
> appreciated.
> 
> > Can you give us some vitals on your setup? Are you using highmem, quota,
> > acls, dmapi etc? Running any specific tests?
> >
> >  =>   This machine is a test machine only.  Is there anything you'd like me
> >  => to try before I reboot?
> >
> > I think it's too late now. If you can work out how to repeat this problem,
> > we should be able to suggest some ways to debug it.
> >
> If I read the code right, there isn't much else which can be done besides
> rebooting.....
> 
> Mark
> 



From owner-linux-xfs@oss.sgi.com Wed Mar 28 21:02:45 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2T52jn23970
	for linux-xfs-outgoing; Wed, 28 Mar 2001 21:02:45 -0800
Received: from deliverator.sgi.com (deliverator.sgi.com [204.94.214.10])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2T52jM23967
	for <linux-xfs@oss.sgi.com>; Wed, 28 Mar 2001 21:02:45 -0800
Received: from snort.melbourne.sgi.com (snort.melbourne.sgi.com [134.14.55.149]) by deliverator.sgi.com (980309.SGI.8.8.8-aspam-6.2/980310.SGI-aspam) via ESMTP id VAA03380
	for <linux-xfs@oss.sgi.com>; Wed, 28 Mar 2001 21:01:30 -0800 (PST)
	mail_from (nathans@snort.melbourne.sgi.com)
Received: (from nathans@localhost)
	by snort.melbourne.sgi.com (SGI-8.9.3/8.9.3) id PAA28652
	for linux-xfs@oss.sgi.com; Thu, 29 Mar 2001 15:01:15 +1000 (EST)
Date: Thu, 29 Mar 2001 15:01:15 +1000 (EST)
From: Nathan Scott <nathans@snort.melbourne.sgi.com>
Message-Id: <200103290501.PAA28652@snort.melbourne.sgi.com>
To: linux-xfs@oss.sgi.com
Subject: TAKE - xfsdump version
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

Date:  Wed Mar 28 20:51:18 PST 2001
Workarea:  snort.melbourne.sgi.com:/diskb/build4/nathans/linux-xfs

The following file(s) were checked into:
  bonnie.engr.sgi.com:/isms/slinx/2.4.x-xfs


Modid:  2.4.x-xfs:slinx:91138a
cmd/xfsdump/VERSION - 1.4
cmd/xfsdump/doc/CHANGES - 1.4
	- bump the version number after those recent spec file changes.



From owner-linux-xfs@oss.sgi.com Wed Mar 28 21:42:18 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2T5gI524682
	for linux-xfs-outgoing; Wed, 28 Mar 2001 21:42:18 -0800
Received: from sgi.com (sgi.SGI.COM [192.48.153.1])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2T5gHM24679
	for <linux-xfs@oss.sgi.com>; Wed, 28 Mar 2001 21:42:17 -0800
Received: from ledzep.americas.sgi.com (relay.sgi.com [137.38.226.97] (may be forged)) 
	by sgi.com (980327.SGI.8.8.8-aspam/980304.SGI-aspam:
       SGI does not authorize the use of its proprietary
       systems or networks for unsolicited or bulk email
       from the Internet.) 
	via ESMTP id VAA08088
	for <linux-xfs@oss.sgi.com>; Wed, 28 Mar 2001 21:42:16 -0800 (PST)
	mail_from (cattelan@gibble.americas.sgi.com)
Received: from gibble.americas.sgi.com (gibble.americas.sgi.com [128.162.195.80]) by ledzep.americas.sgi.com (SGI-SGI-8.9.3/americas-smart-nospam1.1) with ESMTP id XAA87355 for <linux-xfs@oss.sgi.com>; Wed, 28 Mar 2001 23:40:59 -0600 (CST)
Received: (from cattelan@localhost)
	by gibble.americas.sgi.com (8.11.2/8.11.2) id f2T5UoM08793
	for linux-xfs@oss.sgi.com; Thu, 29 Mar 2001 00:30:50 -0500
Date: Thu, 29 Mar 2001 00:30:50 -0500
From: Russell Cattelan <cattelan@sgi.com>
Message-Id: <200103290530.f2T5UoM08793@gibble.americas.sgi.com>
Subject: TAKE - 
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

Date:  Wed Mar 28 21:39:40 PST 2001
Workarea:  gibble.americas.sgi.com:/export/extra/x2.4-xfs-devel

The following file(s) were checked into:
  bonnie.engr.sgi.com:/isms/slinx/2.4.x-xfs


Modid:  2.4.x-xfs:slinx:91142a
linux/fs/pagebuf/page_buf_io.c - 1.70
	- Cast page->index to 64 bit before multiplying by page size.
	  This was causing the offset passed to bmap to overflow for file
	  indexes over 4gig.



From owner-linux-xfs@oss.sgi.com Wed Mar 28 22:27:26 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2T6RQY25514
	for linux-xfs-outgoing; Wed, 28 Mar 2001 22:27:26 -0800
Received: from sgi.com (sgi.SGI.COM [192.48.153.1])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2T6RPM25511
	for <linux-xfs@oss.sgi.com>; Wed, 28 Mar 2001 22:27:25 -0800
Received: from snort.melbourne.sgi.com ([134.14.55.149]) 
	by sgi.com (980327.SGI.8.8.8-aspam/980304.SGI-aspam:
       SGI does not authorize the use of its proprietary
       systems or networks for unsolicited or bulk email
       from the Internet.) 
	via ESMTP id WAA05970
	for <linux-xfs@oss.sgi.com>; Wed, 28 Mar 2001 22:27:23 -0800 (PST)
	mail_from (nathans@snort.melbourne.sgi.com)
Received: (from nathans@localhost)
	by snort.melbourne.sgi.com (SGI-8.9.3/8.9.3) id QAA89488
	for linux-xfs@oss.sgi.com; Thu, 29 Mar 2001 16:25:56 +1000 (EST)
Date: Thu, 29 Mar 2001 16:25:56 +1000 (EST)
From: Nathan Scott <nathans@snort.melbourne.sgi.com>
Message-Id: <200103290625.QAA89488@snort.melbourne.sgi.com>
To: linux-xfs@oss.sgi.com
Subject: TAKE - qa & doc
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

Date:  Wed Mar 28 22:25:45 PST 2001
Workarea:  snort.melbourne.sgi.com:/diskb/build4/nathans/linux-xfs

The following file(s) were checked into:
  bonnie.engr.sgi.com:/isms/slinx/2.4.x-xfs


Modid:  2.4.x-xfs:slinx:91146a
cmd/xfstests/common.quota - 1.6
cmd/xfstests/tools/auto-qa - 1.12
	- do qa testing on the new quota userspace.

cmd/xfsprogs/doc/README.quota - 1.2
	- update to document the new quota userspace version & changes
	  in xfsdump.



From owner-linux-xfs@oss.sgi.com Wed Mar 28 22:33:46 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2T6Xkg25583
	for linux-xfs-outgoing; Wed, 28 Mar 2001 22:33:46 -0800
Received: from yog-sothoth.sgi.com (eugate.sgi.com [192.48.160.10])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2T6XiM25580
	for <linux-xfs@oss.sgi.com>; Wed, 28 Mar 2001 22:33:45 -0800
Received: from snort.melbourne.sgi.com (snort.melbourne.sgi.com [134.14.55.149]) by yog-sothoth.sgi.com (980305.SGI.8.8.8-aspam-6.2/980304.SGI-aspam-europe) via ESMTP id IAA6597734
	for <linux-xfs@oss.sgi.com>; Thu, 29 Mar 2001 08:33:41 +0200 (CEST)
	mail_from (nathans@snort.melbourne.sgi.com)
Received: (from nathans@localhost)
	by snort.melbourne.sgi.com (SGI-8.9.3/8.9.3) id QAA58466
	for linux-xfs@oss.sgi.com; Thu, 29 Mar 2001 16:32:11 +1000 (EST)
Date: Thu, 29 Mar 2001 16:32:11 +1000 (EST)
From: Nathan Scott <nathans@snort.melbourne.sgi.com>
Message-Id: <200103290632.QAA58466@snort.melbourne.sgi.com>
To: linux-xfs@oss.sgi.com
Subject: TAKE - build+clean
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

Date:  Wed Mar 28 22:30:35 PST 2001
Workarea:  snort.melbourne.sgi.com:/diskb/build4/nathans/linux-xfs

The following file(s) were checked into:
  bonnie.engr.sgi.com:/isms/slinx/2.4.x-xfs


Modid:  2.4.x-xfs:slinx:91148a
cmd/attr/Makepkgs - 1.3
cmd/xfsdump/Makepkgs - 1.2
cmd/xfsprogs/Makepkgs - 1.3
cmd/acl/Makepkgs - 1.2
cmd/dmapi/Makepkgs - 1.3
	- make clean before proceeding to do a build to ensure no leftovers
	  corrupt subsequent invocations (for rpm builds).

cmd/xfstests/tools/srctest - 1.2
	- needed updating - still referred to xfs-cmds explicitly.  now takes
	  package as an argument before testing whether package can rebuild
	  itself successfully (from src tarball).



From owner-linux-xfs@oss.sgi.com Wed Mar 28 22:36:12 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2T6aCi25604
	for linux-xfs-outgoing; Wed, 28 Mar 2001 22:36:12 -0800
Received: from yog-sothoth.sgi.com (eugate.sgi.com [192.48.160.10])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2T6aBM25601
	for <linux-xfs@oss.sgi.com>; Wed, 28 Mar 2001 22:36:11 -0800
Received: from snort.melbourne.sgi.com (snort.melbourne.sgi.com [134.14.55.149]) by yog-sothoth.sgi.com (980305.SGI.8.8.8-aspam-6.2/980304.SGI-aspam-europe) via ESMTP id IAA2350817
	for <linux-xfs@oss.sgi.com>; Thu, 29 Mar 2001 08:36:07 +0200 (CEST)
	mail_from (ajag@snort.melbourne.sgi.com)
Received: (from ajag@localhost)
	by snort.melbourne.sgi.com (SGI-8.9.3/8.9.3) id QAA66974
	for linux-xfs@oss.sgi.com; Thu, 29 Mar 2001 16:34:38 +1000 (EST)
Date: Thu, 29 Mar 2001 16:34:38 +1000 (EST)
From: Andrew Gildfind <ajag@snort.melbourne.sgi.com>
Message-Id: <200103290634.QAA66974@snort.melbourne.sgi.com>
To: linux-xfs@oss.sgi.com
Subject: TAKE - xfs_repair removing acls
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

This checks in Steve Lord's fix that adds endian conversion to the acl
code in repair. A qa test that checks this stuff is coming in a 
subsequent checkin.

Date:  Wed Mar 28 22:30:18 PST 2001
Workarea:  snort.melbourne.sgi.com:/home/ajag/isms/slinx

The following file(s) were checked into:
  bonnie.engr.sgi.com:/isms/slinx/2.4.x-xfs


Modid:  2.4.x-xfs:slinx:91149a
cmd/xfsprogs/repair/attr_repair.c - 1.3
	- Add endian conversion to xfs_repair to prevent removal of valid acls.



From owner-linux-xfs@oss.sgi.com Thu Mar 29 05:46:19 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2TDkJH01236
	for linux-xfs-outgoing; Thu, 29 Mar 2001 05:46:19 -0800
Received: from sws5.ctd.ornl.gov (sws5.ctd.ornl.gov [160.91.68.105])
	by oss.sgi.com (8.11.3/8.11.3) with SMTP id f2TDkJM01233
	for <linux-xfs@oss.sgi.com>; Thu, 29 Mar 2001 05:46:19 -0800
Received: (qmail 19894 invoked by uid 3995); 29 Mar 2001 13:46:18 -0000
From: ds-linux-xfs@sws5.ctd.ornl.gov
To: linux-xfs@oss.sgi.com
Subject: Documentation suggestions
Content-Type: text/plain; charset=US-ASCII
Date: 29 Mar 2001 08:46:18 -0500
In-Reply-To: Andrew Gildfind's message of "Thu, 29 Mar 2001 16:34:38 +1000 (EST)"
Message-ID: <wx0g0fw1ydx.fsf@sws5.ctd.ornl.gov>
Lines: 12
User-Agent: Gnus/5.0807 (Gnus v5.8.7) XEmacs/21.1 (Cuyahoga Valley)
MIME-Version: 1.0
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

In http://linux-xfs.sgi.com/projects/xfs/faq.html, under:

  Q: Will I be able to use my old IRIX XFS disks on linux? 

Mention that SGI partition support must be built into the kernel.

In http://linux-xfs.sgi.com/projects/xfs/cvs_download.html, recommend
"make mrproper". This page should also have build instructions or a
pointer to instructions somewhere else, like
http://www.linuxdoc.org/HOWTO/Kernel-HOWTO.html.

-Dave

From owner-linux-xfs@oss.sgi.com Thu Mar 29 05:54:51 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2TDspI01373
	for linux-xfs-outgoing; Thu, 29 Mar 2001 05:54:51 -0800
Received: from esparrall.udg.es (esparrall.udg.es [130.206.124.16])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2TDsZM01362
	for <linux-xfs@oss.sgi.com>; Thu, 29 Mar 2001 05:54:35 -0800
Received: from gcs by esparrall.udg.es with local (Exim 3.22 #1 (Debian))
	id 14idng-0000hp-00
	for <linux-xfs@oss.sgi.com>; Thu, 29 Mar 2001 16:53:24 +0200
Date: Thu, 29 Mar 2001 16:53:24 +0200
From: GCS <gcs@agentsinside.com>
To: linux-xfs@oss.sgi.com
Subject: Re: Documentation suggestions
Message-ID: <20010329165324.A2673@esparrall.udg.es>
References: <wx0g0fw1ydx.fsf@sws5.ctd.ornl.gov>
Mime-Version: 1.0
Content-Type: text/plain; charset=iso-8859-1
Content-Disposition: inline
Content-Transfer-Encoding: 8bit
User-Agent: Mutt/1.3.15i
In-Reply-To: =?iso-8859-1?Q?ds-linux-xfs=40sws5=2Ectd?=
	=?iso-8859-1?Q?=2Eornl=2Egov_wrote_on_CS=2C_M=C1?=
	=?iso-8859-1?Q?R?= 29, 2001 at 08:46:18 -0500:
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk



Hello all,

On CS, MÁR 29, 2001 at 08:46:18 -0500, ds-linux-xfs@sws5.ctd.ornl.gov wrote:

> Mention that SGI partition support must be built into the kernel.
> In http://linux-xfs.sgi.com/projects/xfs/cvs_download.html, recommend
> "make mrproper".
 Totally agree, if you are asking for my word. I have never succeded to build
the kernel from CVS without make mrproper. Also, running klogd with -x switch
is a good advice.

Regards, Laszlo

From owner-linux-xfs@oss.sgi.com Thu Mar 29 06:32:11 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2TEWBq02070
	for linux-xfs-outgoing; Thu, 29 Mar 2001 06:32:11 -0800
Received: from ns.tecosim.de (ns.tecosim.de [194.24.222.9])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2TEW7M02066
	for <linux-xfs@oss.sgi.com>; Thu, 29 Mar 2001 06:32:08 -0800
Received: from donner.tecosim.de (root@donner.tecosim.de [194.24.222.109])
	by ns.tecosim.de (8.9.3/8.9.3/SuSE Linux 8.9.3-0.1) with ESMTP id QAA11757;
	Thu, 29 Mar 2001 16:31:48 +0200
Received: (from leh@localhost)
	by donner.tecosim.de (8.11.2/8.11.2/SuSE Linux 8.11.1-0.5) id f2TEVmh07231;
	Thu, 29 Mar 2001 16:31:48 +0200
Date: Thu, 29 Mar 2001 16:31:48 +0200
From: Utz Lehmann <leh@tecosim.de>
To: ds-linux-xfs@sws5.ctd.ornl.gov
Cc: linux-xfs@oss.sgi.com
Subject: Re: Documentation suggestions
Message-ID: <20010329163148.C4321@tecosim.de>
References: <wx0g0fw1ydx.fsf@sws5.ctd.ornl.gov>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.2.5i
In-Reply-To: <wx0g0fw1ydx.fsf@sws5.ctd.ornl.gov>; from ds-linux-xfs@sws5.ctd.ornl.gov on Thu, Mar 29, 2001 at 08:46:18AM -0500
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

ds-linux-xfs@sws5.ctd.ornl.gov [ds-linux-xfs@sws5.ctd.ornl.gov] wrote:
> In http://linux-xfs.sgi.com/projects/xfs/faq.html, under:
> 
>   Q: Will I be able to use my old IRIX XFS disks on linux? 
> 
> Mention that SGI partition support must be built into the kernel.

And IRIX XLV disk are not mountable due lack of XLV support in linux.


utz

From owner-linux-xfs@oss.sgi.com Thu Mar 29 06:55:23 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2TEtNo02397
	for linux-xfs-outgoing; Thu, 29 Mar 2001 06:55:23 -0800
Received: from ns.tecosim.de (ns.tecosim.de [194.24.222.9])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2TEtLM02394
	for <linux-xfs@oss.sgi.com>; Thu, 29 Mar 2001 06:55:21 -0800
Received: from donner.tecosim.de (root@donner.tecosim.de [194.24.222.109])
	by ns.tecosim.de (8.9.3/8.9.3/SuSE Linux 8.9.3-0.1) with ESMTP id QAA12162
	for <linux-xfs@oss.sgi.com>; Thu, 29 Mar 2001 16:55:20 +0200
Received: (from leh@localhost)
	by donner.tecosim.de (8.11.2/8.11.2/SuSE Linux 8.11.1-0.5) id f2TEtJ307783
	for linux-xfs@oss.sgi.com; Thu, 29 Mar 2001 16:55:19 +0200
Date: Thu, 29 Mar 2001 16:55:19 +0200
From: Utz Lehmann <leh@tecosim.de>
To: linux-xfs@oss.sgi.com
Subject: shutdown umount hangs
Message-ID: <20010329165519.A7386@tecosim.de>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.2.5i
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

Hi

With newer cvs kernel (cvs from this week, before i had holidays) i have
problems with my linux workstation (suse 7.1) at work.

Umounting the filesystems while shutdown hangs.

Following partitions are xfs:

/dev/sda1 on / type xfs (rw)
/dev/vg00/usr on /usr type xfs (rw,kio,logbufs=4,logbsize=32768)
/dev/vg00/var on /var type xfs (rw,kio,logbufs=4,logbsize=32768)
/dev/vg00/opt on /opt type xfs (rw,kio,logbufs=4,logbsize=32768)
/dev/vg00/tmp on /tmp type xfs (rw,kio,logbufs=4,logbsize=32768)



After umounting /tmp and /opt it hangs. Normally /var and /usr followed.

It hangs in the /etc/init.d/halt script:

[...]
echo "Unmounting file systems"
umount -avt noproc,nonfs,nosmbfs || {
    rc_status
    UMOUNT_FAILED=true
}
rc_status -v1 -r


With a cvs kernel from march 2nd i have no problems.


regards.

utz lehmann

From owner-linux-xfs@oss.sgi.com Thu Mar 29 07:04:49 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2TF4nA02637
	for linux-xfs-outgoing; Thu, 29 Mar 2001 07:04:49 -0800
Received: from ns.tecosim.de (ns.tecosim.de [194.24.222.9])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2TF4lM02633
	for <linux-xfs@oss.sgi.com>; Thu, 29 Mar 2001 07:04:48 -0800
Received: from donner.tecosim.de (root@donner.tecosim.de [194.24.222.109])
	by ns.tecosim.de (8.9.3/8.9.3/SuSE Linux 8.9.3-0.1) with ESMTP id RAA12397
	for <linux-xfs@oss.sgi.com>; Thu, 29 Mar 2001 17:04:46 +0200
Received: (from leh@localhost)
	by donner.tecosim.de (8.11.2/8.11.2/SuSE Linux 8.11.1-0.5) id f2TF4kI08030
	for linux-xfs@oss.sgi.com; Thu, 29 Mar 2001 17:04:46 +0200
Date: Thu, 29 Mar 2001 17:04:46 +0200
From: Utz Lehmann <leh@tecosim.de>
To: linux-xfs@oss.sgi.com
Subject: Re: shutdown umount hangs
Message-ID: <20010329170446.B7386@tecosim.de>
References: <20010329165519.A7386@tecosim.de>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.2.5i
In-Reply-To: <20010329165519.A7386@tecosim.de>; from leh@tecosim.de on Thu, Mar 29, 2001 at 04:55:19PM +0200
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

btw: the kernel is compiled with the kgcc from redhat 7.0.

Utz Lehmann [leh@tecosim.de] wrote:
> Hi
> 
> With newer cvs kernel (cvs from this week, before i had holidays) i have
> problems with my linux workstation (suse 7.1) at work.
> 
> Umounting the filesystems while shutdown hangs.
> 
> Following partitions are xfs:
> 
> /dev/sda1 on / type xfs (rw)
> /dev/vg00/usr on /usr type xfs (rw,kio,logbufs=4,logbsize=32768)
> /dev/vg00/var on /var type xfs (rw,kio,logbufs=4,logbsize=32768)
> /dev/vg00/opt on /opt type xfs (rw,kio,logbufs=4,logbsize=32768)
> /dev/vg00/tmp on /tmp type xfs (rw,kio,logbufs=4,logbsize=32768)
> 
> 
> 
> After umounting /tmp and /opt it hangs. Normally /var and /usr followed.
> 
> It hangs in the /etc/init.d/halt script:
> 
> [...]
> echo "Unmounting file systems"
> umount -avt noproc,nonfs,nosmbfs || {
>     rc_status
>     UMOUNT_FAILED=true
> }
> rc_status -v1 -r
> 
> 
> With a cvs kernel from march 2nd i have no problems.
> 
> 
> regards.
> 
> utz lehmann

From owner-linux-xfs@oss.sgi.com Thu Mar 29 07:18:23 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2TFINJ02881
	for linux-xfs-outgoing; Thu, 29 Mar 2001 07:18:23 -0800
Received: from sgi.com (sgi.SGI.COM [192.48.153.1])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2TFIMM02878
	for <linux-xfs@oss.sgi.com>; Thu, 29 Mar 2001 07:18:23 -0800
Received: from zeus-fddi.americas.sgi.com (128-162-8-103.americas.sgi.com [128.162.8.103]) 
	by sgi.com (980327.SGI.8.8.8-aspam/980304.SGI-aspam:
       SGI does not authorize the use of its proprietary
       systems or networks for unsolicited or bulk email
       from the Internet.) 
	via ESMTP id HAA09238
	for <linux-xfs@oss.sgi.com>; Thu, 29 Mar 2001 07:18:22 -0800 (PST)
	mail_from (mann@sgi.com)
Received: from daisy-e185.americas.sgi.com (daisy-e194.americas.sgi.com [128.162.194.214]) by zeus-fddi.americas.sgi.com (8.9.3/americas-smart-nospam1.1) with ESMTP id JAA1302274 for <linux-xfs@oss.sgi.com>; Thu, 29 Mar 2001 09:17:06 -0600 (CST)
Received: from fsgi632.americas.sgi.com (fsgi632.americas.sgi.com [128.162.184.134]) by daisy-e185.americas.sgi.com (SGI-8.9.3/SGI-server-1.7) with ESMTP id JAA90191; Thu, 29 Mar 2001 09:17:06 -0600 (CST)
From: Mark Nordstrand <mann@sgi.com>
Received: by fsgi632.americas.sgi.com (SGI-8.9.3/SGI-client-1.7) id JAA12057; Thu, 29 Mar 2001 09:17:05 -0600 (CST)
Message-Id: <200103291517.JAA12057@fsgi632.americas.sgi.com>
Subject: Re: Unmount Didn't; Clues?
To: lord@sgi.com (Steve Lord)
Date: Thu, 29 Mar 2001 09:17:05 -0600 (CST)
Cc: linux-xfs@oss.sgi.com
In-Reply-To: <200103282353.f2SNrvO02630@jen.americas.sgi.com> from "Steve Lord" at Mar 28, 2001 05:53:57 PM
X-Mailer: ELM [version 2.5 PL2]
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

> 
> 
> So we noticed that EWRONGFS got remapped to ENOSYS, but that is purely
> a mount error, not an unmount. But, could this be to do with the errno
> remapping changes in xfs_linux.h on Mar 13th?
> 

There is a code path in xfs_unmount_flush() which has problems not
being able to destinguish between EFSCORRUPTED and EIO.  I'm in the 
process of moving these two, and re-testing.

Mark

> Steve
> 
> > Since I malformed the address this didn't make it to the list:
> > 
> > >
> > >
> > > Danny writes:
> > >  => All,
> > >  => Mar 29 03:19:15 dsc_proto_1 kernel: XFS unmount got error 38
> > >  => Mar 29 03:19:15 dsc_proto_1 kernel: linvfs_put_super: vfsp/0xc3bc2ae0
> > >  => left dangling!
> > >  => Mar 29 03:19:15 dsc_proto_1 kernel: VFS: Busy inodes after unmount.
> > >  => Self-destruct in 5 seconds.  Have a nice day...
> > >  => Mar 29 03:19:56 dsc_proto_1 kernel: XFS: Filesystem has duplicate UUID 
> > -
> > >  => can't mount
> > >
> > >  =>   Now: I'm pretty sure that a reboot would allow the mount to succeed; 
> > in
> > >  => other words, /dev/md1 is still "mounted" in some way in the kernel;
> > >  => hence the dup UUID.
> > >
> > > If xfs_unmountfs doesn't get called, the uuid for the mounted xfs fs won't
> > > get removed from the uuid table, and further mounts of that filesystem
> > > will be rejected. There are a couple of paths in which the unmount can
> > > fail to call xfs_unmountfs, but they shouldn't happen unless something
> > > is really wrong anyway.
> > >
> > > What I'm really curious about is why the unmount failed in the first
> > > place - that definitely shouldn't be happening.
> > >
> > Curious, I just hit this running a forced shutdown test.  Different
> > error, same effect.  In my case, it definitely won't follow the code
> > path to call xfs_unmountfs().  In either event, it doesn't look like
> > the filesystem is mounted in any useful sense.
> > 
> > I am curious to know how it got here.  Any information would be
> > appreciated.
> > 
> > > Can you give us some vitals on your setup? Are you using highmem, quota,
> > > acls, dmapi etc? Running any specific tests?
> > >
> > >  =>   This machine is a test machine only.  Is there anything you'd like me
> > >  => to try before I reboot?
> > >
> > > I think it's too late now. If you can work out how to repeat this problem,
> > > we should be able to suggest some ways to debug it.
> > >
> > If I read the code right, there isn't much else which can be done besides
> > rebooting.....
> > 
> > Mark
> > 
> 
> 


From owner-linux-xfs@oss.sgi.com Thu Mar 29 07:41:06 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2TFf6Z03306
	for linux-xfs-outgoing; Thu, 29 Mar 2001 07:41:06 -0800
Received: from heimdal.iee.lu (heimdal.iee.lu [194.154.201.2])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2TFf4M03302
	for <linux-xfs@oss.sgi.com>; Thu, 29 Mar 2001 07:41:04 -0800
Received: from picard.findel.ieeintern (picard.findel.ieeintern [10.1.0.3])
	by heimdal.iee.lu (8.9.3/8.9.3/SuSE Linux 8.9.3-0.1) with ESMTP id RAA21447;
	Thu, 29 Mar 2001 17:40:58 +0200
Received: from iee.lu (cgs.findel.ieeintern [10.1.4.9])
	by picard.findel.ieeintern (8.11.0/8.10.2/SuSE Linux 8.10.0-0.3) with ESMTP id f2TFeva21178;
	Thu, 29 Mar 2001 17:40:58 +0200
X-Authentication-Warning: picard.findel.ieeintern: Host cgs.findel.ieeintern [10.1.4.9] claimed to be iee.lu
Message-ID: <3AC35789.FF9BBCA@iee.lu>
Date: Thu, 29 Mar 2001 17:40:57 +0200
From: Carlos Gamboa Dos Santos <Carlos.Gamboa@iee.lu>
Organization: IEE (Application Development)
X-Mailer: Mozilla 4.76 [en] (X11; U; Linux 2.4.2-XFS i686)
X-Accept-Language: en, de-DE
MIME-Version: 1.0
To: Andrew Gildfind <ajag@fudge.melbourne.sgi.com>
CC: linux-xfs@oss.sgi.com
Subject: Re: Problems with ACL inheritance and chacl
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

Hi...

Andrew Gildfind wrote:
>
> On Tue, Mar 27, 2001 at 10:51:42AM -0500, John Trostel wrote:
> > I am seeing some strange behavior with both ACL inheritance and chacl
> > operation.  Does anyone else see this or is my build faulty?
> > 
> > I started with a freshly formatted xfs partition on /mnt/xfs_part.
> > 
> > I set the access, default and mask ACLs for this parition as follows:
> > 
> > [...]
> > 
> > Not good... Shouldn't this get the default ACLs u::rwx,g::rwx,o::rwx ?
> 
> I'm not precisely sure of the semantics of ACL inheritance, but I noticed
> when playing around with this that the inherited ACL was modified by
> different umasks... I'll have to look into this further.

We saw the same behaviour with our builts. After a few tests, we confirmed that
"umask" was changing the active mask of the ACL and therefore limiting the real
accesses.

But... shouldn't "umask" be ignored in ACL where a mask is present?
Otherwise it seems that the flexibility of the ACLs is lost...

For example:

[cgs@cgs] umask
022

[cgs@cgs] id
uid=716(cgs) gid=100(users) groups=100(users),506(ad_design)

[cgs@cgs] chacl -l design
design
[u::rwx,g:ad_design:rwx,g::---,o::---,m::rwx/u::rwx,g:ad_design:rwx,g::---,o::---,m::rwx]

[cgs@cgs] mkdir design/test
[cgs@cgs] chacl -l design/test
design/test
[u::rwx,g:ad_design:rwx,g::---,o::---,m::r-x/u::rwx,g:ad_design:rwx,g::---,o::---,m::rwx]


Note that now the mask has the value m::r-x! And of course another user from 
geoup ad_design can't write in design/test...

By setting umask to 002 we get the correct mask m::rwx (and access rights),
but it seems "wrong"... Or am I completely mistaken?

Thanx,

Carlos.

-- 
Carlos Gamboa Dos Santos                   Carlos.Gamboa@iee.lu
International Electronics & Engineering

Tel: (+352) 42 47 37-280                   Fax: (+352) 42 47 37-200

From owner-linux-xfs@oss.sgi.com Thu Mar 29 08:09:43 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2TG9hN04588
	for linux-xfs-outgoing; Thu, 29 Mar 2001 08:09:43 -0800
Received: from mail11.jump.net (mail11.jump.net [206.196.91.11])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2TG9gM04585
	for <linux-xfs@oss.sgi.com>; Thu, 29 Mar 2001 08:09:42 -0800
Received: from sgi.com (aus-dsl-dhcp1-26.customer.jump.net [63.163.168.26])
	by mail11.jump.net (8.10.2/) with ESMTP id f2TG9Pc22575;
	Thu, 29 Mar 2001 10:09:25 -0600 (CST)
Message-ID: <3AC35E8D.711963C5@sgi.com>
Date: Thu, 29 Mar 2001 10:10:53 -0600
From: Eric Sandeen <sandeen@sgi.com>
X-Mailer: Mozilla 4.76 [en] (X11; U; Linux 2.4.2-XFS i686)
X-Accept-Language: en
MIME-Version: 1.0
To: Utz Lehmann <leh@tecosim.de>
CC: linux-xfs@oss.sgi.com
Subject: Re: shutdown umount hangs
References: <20010329165519.A7386@tecosim.de> <20010329170446.B7386@tecosim.de>
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

Utz Lehmann wrote:

> > With newer cvs kernel (cvs from this week, before i had holidays) i have
> > problems with my linux workstation (suse 7.1) at work.
> >
> > Umounting the filesystems while shutdown hangs.

Are there any errors related to unmount in the syslog?

-Eric

From owner-linux-xfs@oss.sgi.com Thu Mar 29 08:14:27 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2TGERn04647
	for linux-xfs-outgoing; Thu, 29 Mar 2001 08:14:27 -0800
Received: from ns.tecosim.de (ns.tecosim.de [194.24.222.9])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2TGENM04644
	for <linux-xfs@oss.sgi.com>; Thu, 29 Mar 2001 08:14:24 -0800
Received: from donner.tecosim.de (root@donner.tecosim.de [194.24.222.109])
	by ns.tecosim.de (8.9.3/8.9.3/SuSE Linux 8.9.3-0.1) with ESMTP id SAA14528;
	Thu, 29 Mar 2001 18:14:19 +0200
Received: (from leh@localhost)
	by donner.tecosim.de (8.11.2/8.11.2/SuSE Linux 8.11.1-0.5) id f2TGEJ211369;
	Thu, 29 Mar 2001 18:14:19 +0200
Date: Thu, 29 Mar 2001 18:14:19 +0200
From: Utz Lehmann <leh@tecosim.de>
To: Eric Sandeen <sandeen@sgi.com>
Cc: linux-xfs@oss.sgi.com
Subject: Re: shutdown umount hangs
Message-ID: <20010329181419.G7386@tecosim.de>
References: <20010329165519.A7386@tecosim.de> <20010329170446.B7386@tecosim.de> <3AC35E8D.711963C5@sgi.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.2.5i
In-Reply-To: <3AC35E8D.711963C5@sgi.com>; from sandeen@sgi.com on Thu, Mar 29, 2001 at 10:10:53AM -0600
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

Eric Sandeen [sandeen@sgi.com] wrote:
> Utz Lehmann wrote:
> 
> > > With newer cvs kernel (cvs from this week, before i had holidays) i have
> > > problems with my linux workstation (suse 7.1) at work.
> > >
> > > Umounting the filesystems while shutdown hangs.
> 
> Are there any errors related to unmount in the syslog?

No, but syslogd was killed before umounting.
There were no messages on console, it just hangs. sysreq works.


From owner-linux-xfs@oss.sgi.com Thu Mar 29 08:19:19 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2TGJJ004704
	for linux-xfs-outgoing; Thu, 29 Mar 2001 08:19:19 -0800
Received: from dkp.com ([204.191.16.3])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2TGJFM04701
	for <linux-xfs@oss.sgi.com>; Thu, 29 Mar 2001 08:19:19 -0800
Received: (from ak@localhost)
	by dkp.com (8.9.3/8.9.3) id LAA21983
	for linux-xfs@oss.sgi.com; Thu, 29 Mar 2001 11:19:06 -0500
Date: Thu, 29 Mar 2001 11:19:06 -0500
From: Andrew Klaassen <ak@dkp.com>
To: linux-xfs@oss.sgi.com
Subject: RH7, RAID5, XFS
Message-ID: <20010329111905.A21645@key.dkp.com>
Mail-Followup-To: linux-xfs@oss.sgi.com
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.2i
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

A question about the XFS pre-release kernel rpm and RAID-5:

Does kernel in the rpm have RAID-5 enabled?  If it does, were
whatever necessary patches applied before build to make RAID-5
and XFS get along?  (Or are those patches no longer necessary?)

Thanks.

Andrew Klaassen


From owner-linux-xfs@oss.sgi.com Thu Mar 29 08:20:49 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2TGKni04795
	for linux-xfs-outgoing; Thu, 29 Mar 2001 08:20:49 -0800
Received: from ns.tecosim.de (ns.tecosim.de [194.24.222.9])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2TGKkM04782
	for <linux-xfs@oss.sgi.com>; Thu, 29 Mar 2001 08:20:47 -0800
Received: from donner.tecosim.de (root@donner.tecosim.de [194.24.222.109])
	by ns.tecosim.de (8.9.3/8.9.3/SuSE Linux 8.9.3-0.1) with ESMTP id SAA14769;
	Thu, 29 Mar 2001 18:20:44 +0200
Received: (from leh@localhost)
	by donner.tecosim.de (8.11.2/8.11.2/SuSE Linux 8.11.1-0.5) id f2TGKiV11532;
	Thu, 29 Mar 2001 18:20:44 +0200
Date: Thu, 29 Mar 2001 18:20:44 +0200
From: Utz Lehmann <leh@tecosim.de>
To: Eric Sandeen <sandeen@sgi.com>
Cc: linux-xfs@oss.sgi.com
Subject: Re: shutdown umount hangs
Message-ID: <20010329182044.H7386@tecosim.de>
References: <20010329165519.A7386@tecosim.de> <20010329170446.B7386@tecosim.de> <3AC35E8D.711963C5@sgi.com> <20010329181419.G7386@tecosim.de>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.2.5i
In-Reply-To: <20010329181419.G7386@tecosim.de>; from leh@tecosim.de on Thu, Mar 29, 2001 at 06:14:19PM +0200
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

Utz Lehmann [leh@tecosim.de] wrote:
> Eric Sandeen [sandeen@sgi.com] wrote:
> > Utz Lehmann wrote:
> > 
> > > > With newer cvs kernel (cvs from this week, before i had holidays) i have
> > > > problems with my linux workstation (suse 7.1) at work.
> > > >
> > > > Umounting the filesystems while shutdown hangs.
> > 
> > Are there any errors related to unmount in the syslog?
> 
> No, but syslogd was killed before umounting.
> There were no messages on console, it just hangs. sysreq works.

btw: is there a way to trap it into the kdb?

From owner-linux-xfs@oss.sgi.com Thu Mar 29 11:01:04 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2TJ14S09399
	for linux-xfs-outgoing; Thu, 29 Mar 2001 11:01:04 -0800
Received: from tux.mkp.net (tux.mkp.net [130.225.60.11])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2TJ12M09396
	for <linux-xfs@oss.sgi.com>; Thu, 29 Mar 2001 11:01:03 -0800
Received: from tux.mkp.net ([130.225.60.11] helo=jaguar.mkp.net)
	by tux.mkp.net with esmtp (Exim 3.16 #1)
	id 14ihfI-0000bM-00; Thu, 29 Mar 2001 21:01:00 +0200
Received: (from mkp@localhost)
	by jaguar.mkp.net (8.9.3/8.9.3) id JAA20843;
	Thu, 29 Mar 2001 09:00:49 -0500
X-Authentication-Warning: jaguar.mkp.net: mkp set sender to mkp@mkp.net using -f
To: Andrew Klaassen <ak@dkp.com>
Cc: linux-xfs@oss.sgi.com
Subject: Re: RH7, RAID5, XFS
References: <20010329111905.A21645@key.dkp.com>
From: "Martin K. Petersen" <mkp@linuxcare.com>
Organization: Linuxcare, Inc.
Date: 29 Mar 2001 09:00:49 -0500
In-Reply-To: <20010329111905.A21645@key.dkp.com>
Message-ID: <yq1k858oesu.fsf@jaguar.mkp.net>
Lines: 18
User-Agent: Gnus/5.0808 (Gnus v5.8.8) XEmacs/21.2 (Urania)
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

>>>>> "Andrew" == Andrew Klaassen <ak@dkp.com> writes:

Andrew> A question about the XFS pre-release kernel rpm and RAID-5:
Andrew> Does kernel in the rpm have RAID-5 enabled?  

I would think so, but I don't know for sure.  Russell/Eric?


Andrew> If it does, were whatever necessary patches applied before
Andrew> build to make RAID-5 and XFS get along?  (Or are those patches
Andrew> no longer necessary?)

No patches are necessary.

-- 
Martin K. Petersen, Principal Linux Consultant, Linuxcare, Inc.
mkp@linuxcare.com, http://www.linuxcare.com/
SGI XFS for Linux Developer, http://oss.sgi.com/projects/xfs/

From owner-linux-xfs@oss.sgi.com Thu Mar 29 11:13:59 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2TJDxj09588
	for linux-xfs-outgoing; Thu, 29 Mar 2001 11:13:59 -0800
Received: from pneumatic-tube.sgi.com (pneumatic-tube.sgi.com [204.94.214.22])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2TJDwM09585
	for <linux-xfs@oss.sgi.com>; Thu, 29 Mar 2001 11:13:58 -0800
Received: from ledzep.americas.sgi.com (ledzep.americas.sgi.com [137.38.226.97]) by pneumatic-tube.sgi.com (980327.SGI.8.8.8-aspam/980310.SGI-aspam) via ESMTP id LAA02155
	for <linux-xfs@oss.sgi.com>; Thu, 29 Mar 2001 11:24:05 -0800 (PST)
	mail_from (cattelan@thebarn.com)
Received: from gibble.americas.sgi.com (gibble.americas.sgi.com [128.162.195.80]) by ledzep.americas.sgi.com (SGI-SGI-8.9.3/americas-smart-nospam1.1) with ESMTP id NAA21545; Thu, 29 Mar 2001 13:12:41 -0600 (CST)
Received: from thebarn.com (localhost [127.0.0.1])
	by gibble.americas.sgi.com (8.11.2/8.11.2) with ESMTP id f2TJ2Wb12024;
	Thu, 29 Mar 2001 14:02:32 -0500
Message-ID: <3AC386C8.BB218C5@thebarn.com>
Date: Thu, 29 Mar 2001 14:02:32 -0500
From: Russell Cattelan <cattelan@thebarn.com>
X-Mailer: Mozilla 4.76 [en] (X11; U; Linux 2.4.2-XFS i686)
X-Accept-Language: en
MIME-Version: 1.0
To: "Martin K. Petersen" <mkp@linuxcare.com>
CC: Andrew Klaassen <ak@dkp.com>, linux-xfs@oss.sgi.com
Subject: Re: RH7, RAID5, XFS
References: <20010329111905.A21645@key.dkp.com> <yq1k858oesu.fsf@jaguar.mkp.net>
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

"Martin K. Petersen" wrote:

> >>>>> "Andrew" == Andrew Klaassen <ak@dkp.com> writes:
>
> Andrew> A question about the XFS pre-release kernel rpm and RAID-5:
> Andrew> Does kernel in the rpm have RAID-5 enabled?
>
> I would think so, but I don't know for sure.  Russell/Eric?

The 0.10 kernel has all the updates to md and lvm included and is enabled

The  prerelease aka 0.9  kernel is rather out of date...  don't remember
what state it
was in off hand, suggest not using it.


>
>
> Andrew> If it does, were whatever necessary patches applied before
> Andrew> build to make RAID-5 and XFS get along?  (Or are those patches
> Andrew> no longer necessary?)
>
> No patches are necessary.
>
> --
> Martin K. Petersen, Principal Linux Consultant, Linuxcare, Inc.
> mkp@linuxcare.com, http://www.linuxcare.com/
> SGI XFS for Linux Developer, http://oss.sgi.com/projects/xfs/

--
Russell Cattelan
--
Digital Elves inc. -- Currently on loan to SGI
Linux XFS core developer.




From owner-linux-xfs@oss.sgi.com Thu Mar 29 11:15:34 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2TJFYK09623
	for linux-xfs-outgoing; Thu, 29 Mar 2001 11:15:34 -0800
Received: from pneumatic-tube.sgi.com (pneumatic-tube.sgi.com [204.94.214.22])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2TJFXM09620
	for <linux-xfs@oss.sgi.com>; Thu, 29 Mar 2001 11:15:33 -0800
Received: from ledzep.americas.sgi.com (ledzep.americas.sgi.com [137.38.226.97]) by pneumatic-tube.sgi.com (980327.SGI.8.8.8-aspam/980310.SGI-aspam) via ESMTP id LAA08763
	for <linux-xfs@oss.sgi.com>; Thu, 29 Mar 2001 11:25:40 -0800 (PST)
	mail_from (cattelan@thebarn.com)
Received: from gibble.americas.sgi.com (gibble.americas.sgi.com [128.162.195.80]) by ledzep.americas.sgi.com (SGI-SGI-8.9.3/americas-smart-nospam1.1) with ESMTP id NAA10498; Thu, 29 Mar 2001 13:14:16 -0600 (CST)
Received: from thebarn.com (localhost [127.0.0.1])
	by gibble.americas.sgi.com (8.11.2/8.11.2) with ESMTP id f2TJ47b12030;
	Thu, 29 Mar 2001 14:04:07 -0500
Message-ID: <3AC38727.FC9C4FC8@thebarn.com>
Date: Thu, 29 Mar 2001 14:04:07 -0500
From: Russell Cattelan <cattelan@thebarn.com>
X-Mailer: Mozilla 4.76 [en] (X11; U; Linux 2.4.2-XFS i686)
X-Accept-Language: en
MIME-Version: 1.0
To: Utz Lehmann <leh@tecosim.de>
CC: Eric Sandeen <sandeen@sgi.com>, linux-xfs@oss.sgi.com
Subject: Re: shutdown umount hangs
References: <20010329165519.A7386@tecosim.de> <20010329170446.B7386@tecosim.de> <3AC35E8D.711963C5@sgi.com> <20010329181419.G7386@tecosim.de> <20010329182044.H7386@tecosim.de>
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

Utz Lehmann wrote:

> Utz Lehmann [leh@tecosim.de] wrote:
> > Eric Sandeen [sandeen@sgi.com] wrote:
> > > Utz Lehmann wrote:
> > >
> > > > > With newer cvs kernel (cvs from this week, before i had holidays) i have
> > > > > problems with my linux workstation (suse 7.1) at work.
> > > > >
> > > > > Umounting the filesystems while shutdown hangs.
> > >
> > > Are there any errors related to unmount in the syslog?
> >
> > No, but syslogd was killed before umounting.
> > There were no messages on console, it just hangs. sysreq works.
>
> btw: is there a way to trap it into the kdb?

You should still be able to drop into kdb, a back trace would be really helpful if
you can get it.


ctrl-a from serial line "break" from the keyboard or was is ctrl break...
don't remember always use serial line.



--
Russell Cattelan
--
Digital Elves inc. -- Currently on loan to SGI
Linux XFS core developer.




From owner-linux-xfs@oss.sgi.com Thu Mar 29 11:15:41 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2TJFfI09635
	for linux-xfs-outgoing; Thu, 29 Mar 2001 11:15:41 -0800
Received: from mail11.jump.net (mail11.jump.net [206.196.91.11])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2TJFfM09630
	for <linux-xfs@oss.sgi.com>; Thu, 29 Mar 2001 11:15:41 -0800
Received: from sgi.com (aus-dsl-dhcp1-26.customer.jump.net [63.163.168.26])
	by mail11.jump.net (8.10.2/) with ESMTP id f2TJFbu15184;
	Thu, 29 Mar 2001 13:15:37 -0600 (CST)
Message-ID: <3AC38A10.CADDE68D@sgi.com>
Date: Thu, 29 Mar 2001 13:16:32 -0600
From: Eric Sandeen <sandeen@sgi.com>
X-Mailer: Mozilla 4.76 [en] (X11; U; Linux 2.4.2-XFS i686)
X-Accept-Language: en
MIME-Version: 1.0
To: "Martin K. Petersen" <mkp@linuxcare.com>
CC: Andrew Klaassen <ak@dkp.com>, linux-xfs@oss.sgi.com
Subject: Re: RH7, RAID5, XFS
References: <20010329111905.A21645@key.dkp.com> <yq1k858oesu.fsf@jaguar.mkp.net>
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

"Martin K. Petersen" wrote:
> 
> >>>>> "Andrew" == Andrew Klaassen <ak@dkp.com> writes:
> 
> Andrew> A question about the XFS pre-release kernel rpm and RAID-5:
> Andrew> Does kernel in the rpm have RAID-5 enabled?
> 
> I would think so, but I don't know for sure.  Russell/Eric?


CONFIG_MD_RAID5 is enabled as a module.

-Eric

From owner-linux-xfs@oss.sgi.com Thu Mar 29 14:31:06 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2TMV6J13015
	for linux-xfs-outgoing; Thu, 29 Mar 2001 14:31:06 -0800
Received: from dfwns01.algx.net (ns1.algx.net [216.99.225.30])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2TMV4M13012
	for <linux-xfs@oss.sgi.com>; Thu, 29 Mar 2001 14:31:05 -0800
Received: from jtsdell (66-2-81-26.customer.algx.net [66.2.81.26])
	by dfwns01.algx.net (8.9.1/8.9.1) with ESMTP id QAA24732;
	Thu, 29 Mar 2001 16:30:48 -0600 (CST)
Message-ID: <XFMail.20010329172932.jtrostel@connex.com>
X-Mailer: XFMail 1.4.8 on Linux
X-Priority: 3 (Normal)
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 8bit
MIME-Version: 1.0
In-Reply-To: <3AC35789.FF9BBCA@iee.lu>
Date: Thu, 29 Mar 2001 17:29:32 -0500 (EST)
Reply-To: jtrostel@connex.com
Organization: Connex
From: John Trostel <jtrostel@connex.com>
To: Carlos Gamboa Dos Santos <Carlos.Gamboa@iee.lu>
Subject: Re: Problems with ACL inheritance and chacl & maybe a BUG
Cc: linux-xfs@oss.sgi.com, Andrew Gildfind <ajag@fudge.melbourne.sgi.com>
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

This appears to be the 'correct' behavior as per IEEE posix specs (IEEE Draft
P1003.2c). (Although I agree that it is counter-intuitive!)

Section 8.1.7

"A mask ACL entry specifies the maximum access which can be granted by any ACL
entry except the user entry for the file owner and the other entry".

This implies that with
        u::abc,g:abc,o::abc,m::123 ,

the mask will only grant beneath it
        u::xxx,g::123,o::xxx.

That is, it will affect the GROUP_OBJ but it won't affect the USER_OBJ or OTHER
bits AT ALL! (It will constrain the USER and GROUP entries though).

When I create I file using 'touch',the file first get the permissions as
determined by the open() function here (gnu file-utils source).

  if (! no_create)
    {
      /* Try to open FILE, creating it if necessary.  */
      fd = open (file, O_WRONLY | O_CREAT | O_NONBLOCK | O_NOCTTY,
                 S_IRUSR | S_IWUSR | S_IRGRP | S_IWGRP | S_IROTH | S_IWOTH);
      if (fd == -1)
        open_errno = errno;
    }

The initial mode described above is rw-rw-rw- (00666). According to 'open'
behavior, it should get permissions based on mode & ~umask. This translates to
666 & ~002 (i.e., 666 & 775) = 664... rw-rw-r--.

If there is a default ACL, the created file permission should then be
constrained by these.  So... if I have a default ACL u::rwx,g::rwx,o::rwx,
nothing will change.

When I try this on an XFS partition I do indeed get rw-rw-r-- permission with a
umask of 002 and a rw-rw-rw- with a umask of 000.

If I have a default ACL u::rwx,g::r--,o::---, the 'touched' file should get
u::rw-,g::r--,o::--- perms (with either umask).  It turns out that it actually
gets u::rw-,g::rw-,o::---.  THIS SEEMS WRONG... The OTHER ACL was created
correctly, but not the GROUP_OBJ ACL.  

See more comments embedded in your text below:

On 29-Mar-2001 Carlos Gamboa Dos Santos wrote:
> Hi...
> 
> Andrew Gildfind wrote:
>>
>> On Tue, Mar 27, 2001 at 10:51:42AM -0500, John Trostel wrote:
>> > I am seeing some strange behavior with both ACL inheritance and chacl
>> > operation.  Does anyone else see this or is my build faulty?
>> > 
>> > I started with a freshly formatted xfs partition on /mnt/xfs_part.
>> > 
>> > I set the access, default and mask ACLs for this parition as follows:
>> > 
>> > [...]
>> > 
>> > Not good... Shouldn't this get the default ACLs u::rwx,g::rwx,o::rwx ?
>> 
>> I'm not precisely sure of the semantics of ACL inheritance, but I noticed
>> when playing around with this that the inherited ACL was modified by
>> different umasks... I'll have to look into this further.
> 
> We saw the same behaviour with our builts. After a few tests, we confirmed
> that
> "umask" was changing the active mask of the ACL and therefore limiting the
> real
> accesses.
> 
> But... shouldn't "umask" be ignored in ACL where a mask is present?
> Otherwise it seems that the flexibility of the ACLs is lost...
> 
> For example:
> 
> [cgs@cgs] umask
> 022
> 
> [cgs@cgs] id
> uid=716(cgs) gid=100(users) groups=100(users),506(ad_design)
> 
> [cgs@cgs] chacl -l design
> design
> [u::rwx,g:ad_design:rwx,g::---,o::---,m::rwx/u::rwx,g:ad_design:rwx,g::---,o::
> ---,m::rwx]
> 
> [cgs@cgs] mkdir design/test
> [cgs@cgs] chacl -l design/test
> design/test
> [u::rwx,g:ad_design:rwx,g::---,o::---,m::r-x/u::rwx,g:ad_design:rwx,g::---,o::
> ---,m::rwx]
> 
> 
> Note that now the mask has the value m::r-x! And of course another user from 
> geoup ad_design can't write in design/test...
> 
> By setting umask to 002 we get the correct mask m::rwx (and access rights),
> but it seems "wrong"... Or am I completely mistaken?
> 

No... it seems like what I would think it would do too, but it is contrary to
the (withdrawn) Posix ACL Standard,  Specifically, see sections 23.1.2, 23.1.4,
and B.23.5

But... they've (almost) got it right.

Andrew:  Can you comment on the behavior I percieved as a bug?  Are the ACLs
being applied wrong, or am I misinterpreting the 'standard'?

-- 
John M. Trostel
Linux OS Engineer
Connex
jtrostel@connex.com

From owner-linux-xfs@oss.sgi.com Thu Mar 29 17:42:09 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2U1g9u17227
	for linux-xfs-outgoing; Thu, 29 Mar 2001 17:42:09 -0800
Received: from deliverator.sgi.com (deliverator.sgi.com [204.94.214.10])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2U1g8M17223
	for <linux-xfs@oss.sgi.com>; Thu, 29 Mar 2001 17:42:08 -0800
Received: from snort.melbourne.sgi.com (snort.melbourne.sgi.com [134.14.55.149]) by deliverator.sgi.com (980309.SGI.8.8.8-aspam-6.2/980310.SGI-aspam) via ESMTP id RAA22424
	for <linux-xfs@oss.sgi.com>; Thu, 29 Mar 2001 17:40:53 -0800 (PST)
	mail_from (nathans@snort.melbourne.sgi.com)
Received: (from nathans@localhost)
	by snort.melbourne.sgi.com (SGI-8.9.3/8.9.3) id LAA38615
	for linux-xfs@oss.sgi.com; Fri, 30 Mar 2001 11:40:38 +1000 (EST)
Date: Fri, 30 Mar 2001 11:40:38 +1000 (EST)
From: Nathan Scott <nathans@snort.melbourne.sgi.com>
Message-Id: <200103300140.LAA38615@snort.melbourne.sgi.com>
To: linux-xfs@oss.sgi.com
Subject: TAKE - quota 3.01
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

Date:  Thu Mar 29 17:38:24 PST 2001
Workarea:  snort.melbourne.sgi.com:/diskb/build4/nathans/linux-xfs

The following file(s) were checked into:
  bonnie.engr.sgi.com:/isms/slinx/2.4.x-xfs


Modid:  2.4.x-xfs:slinx:91222a
SPECS/quota.spec - 1.1
	- the new, improved quota spec file for 3.01.

SOURCES/quota-3.01-pre2.tar.gz - 1.1
	- move up to quota-3.01 user tools.

Makefile - 1.43
	- use the new quota spec file for 3.01.


[The following files were deleted..]

cmd/quota/Changelog - 1.3
cmd/quota/Makefile.in - 1.8
cmd/quota/README.gettext - 1.2
cmd/quota/README.xfs - 1.9
cmd/quota/bylabel.c - 1.3
cmd/quota/bylabel.h - 1.2
cmd/quota/configure - 1.2
cmd/quota/configure.in - 1.6
cmd/quota/doc/edquota(8).html - 1.2
cmd/quota/doc/fstab(5).html - 1.2
cmd/quota/doc/quota(1).html - 1.2
cmd/quota/doc/quota.html - 1.2
cmd/quota/doc/quota4th.fig - 1.2
cmd/quota/doc/quotacheck(8).html - 1.2
cmd/quota/doc/quotactl(2).html - 1.2
cmd/quota/doc/quotaon(8).html - 1.2
cmd/quota/doc/quotas-1.eps - 1.2
cmd/quota/doc/quotas.ms - 1.2
cmd/quota/doc/quotas.preformated - 1.2
cmd/quota/doc/repquota(8).html - 1.2
cmd/quota/doc/rquotad(8).html - 1.2
cmd/quota/dqblk.h - 1.2
cmd/quota/edquota.8 - 1.4
cmd/quota/edquota.c - 1.7
cmd/quota/hasquota.c - 1.7
cmd/quota/install-sh - 1.2
cmd/quota/mntent.h - 1.3
cmd/quota/po/pl.mo - 1.3
cmd/quota/po/pl.po - 1.2
cmd/quota/pot.c - 1.2
cmd/quota/pot.h - 1.2
cmd/quota/quota.1 - 1.5
cmd/quota/quota.c - 1.7
cmd/quota/quotacheck.8 - 1.5
cmd/quota/quotacheck.c - 1.6
cmd/quota/quotactl.2 - 1.3
cmd/quota/quotactl.c - 1.3
cmd/quota/quotaio.c - 1.5
cmd/quota/quotaon.8 - 1.6
cmd/quota/quotaon.c - 1.12
cmd/quota/quotaops.c - 1.11
cmd/quota/quotaops.h - 1.5
cmd/quota/quotastats.c - 1.4
cmd/quota/repquota.8 - 1.6
cmd/quota/repquota.c - 1.17
cmd/quota/rquota.3 - 1.2
cmd/quota/rquota.h - 1.3
cmd/quota/rquota.x - 1.3
cmd/quota/rquota_client.c - 1.3
cmd/quota/rquota_clnt.c - 1.3
cmd/quota/rquota_server.c - 1.7
cmd/quota/rquota_svc.c - 1.2
cmd/quota/rquota_xdr.c - 1.3
cmd/quota/rquotad.8 - 1.4
cmd/quota/set_limits_example.c - 1.2
cmd/quota/setquota.8 - 1.6
cmd/quota/setquota.c - 1.8
cmd/quota/setup_quota_group - 1.2
cmd/quota/warnquota.c - 1.6
cmd/quota/warnquota.conf - 1.2
cmd/quota/xqm.h - 1.4
cmd/quota/hasquota.h - 1.2
	- nuke this, move up to quota-3.01 user tools on sourceforge.net
	  as these support XFS quota natively now.

SPECS/quota-xfs.spec - 1.5
SOURCES/quota-2.00.tar.gz - 1.2
	- nuke this, move up to quota-3.01 user tools.

SOURCES/quota-2.00-xfs.patch - 1.6
	- we no longer need a patch, code is in 3.01.



From owner-linux-xfs@oss.sgi.com Thu Mar 29 17:55:33 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2U1tXk17418
	for linux-xfs-outgoing; Thu, 29 Mar 2001 17:55:33 -0800
Received: from cube.gelatinous.com (cube.gelatinous.com [207.82.194.150])
	by oss.sgi.com (8.11.3/8.11.3) with SMTP id f2U1tXM17415
	for <linux-xfs@oss.sgi.com>; Thu, 29 Mar 2001 17:55:33 -0800
Received: (qmail 85502 invoked by uid 1000); 30 Mar 2001 01:55:26 -0000
Date: Thu, 29 Mar 2001 17:55:26 -0800
From: Aaron Smith <aaron@mutex.org>
To: linux-xfs@oss.sgi.com
Subject: "user data" in transactions / logged writes
Message-ID: <20010329175526.B58595@gelatinous.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.2.5i
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

hello,

i've been reading over some of the xfs papers, and i noticed the following
in "xFS project architecture, October 8, 1993":

   3.3 xFS Growth Path

   In the long run, we expect to add user-level transactions to the file
   system, at which time selected user data will be logged along with the
   filesystem metadata. This feature will make it practical for user
   applications to easily implement databases and database-like systems on
   top of this file system.

is this essentially logged writes? does xfs have any support for
journalling written file data? please let me know if you know of anyone
who's working on it, or if you understand what pitfalls would need to be
addressed to make it possible.

thanks,
aaron

From owner-linux-xfs@oss.sgi.com Thu Mar 29 20:39:28 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2U4dSh20465
	for linux-xfs-outgoing; Thu, 29 Mar 2001 20:39:28 -0800
Received: from yog-sothoth.sgi.com (eugate.sgi.com [192.48.160.10])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2U4dQM20462
	for <linux-xfs@oss.sgi.com>; Thu, 29 Mar 2001 20:39:26 -0800
Received: from ledzep.americas.sgi.com (ledzep.americas.sgi.com [137.38.226.97]) by yog-sothoth.sgi.com (980305.SGI.8.8.8-aspam-6.2/980304.SGI-aspam-europe) via ESMTP id GAA6517231
	for <linux-xfs@oss.sgi.com>; Fri, 30 Mar 2001 06:39:24 +0200 (CEST)
	mail_from (cattelan@gibble.americas.sgi.com)
Received: from gibble.americas.sgi.com (gibble.americas.sgi.com [128.162.195.80]) by ledzep.americas.sgi.com (SGI-SGI-8.9.3/americas-smart-nospam1.1) with ESMTP id WAA67515 for <linux-xfs@oss.sgi.com>; Thu, 29 Mar 2001 22:38:08 -0600 (CST)
Received: (from cattelan@localhost)
	by gibble.americas.sgi.com (8.11.2/8.11.2) id f2U4QaU15613
	for linux-xfs@oss.sgi.com; Thu, 29 Mar 2001 23:26:36 -0500
Date: Thu, 29 Mar 2001 23:26:36 -0500
From: Russell Cattelan <cattelan@sgi.com>
Message-Id: <200103300426.f2U4QaU15613@gibble.americas.sgi.com>
Subject: TAKE - 
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

Date:  Thu Mar 29 20:35:55 PST 2001
Workarea:  gibble.americas.sgi.com:/export/extra/x2.4-xfs-r1.0
Merged by:  cattelan
Merged mods:  2.4.x-xfs:slinx:91138a,2.4.x-xfs:slinx:91142a,2.4.x-xfs:slinx:91146a
              2.4.x-xfs:slinx:91145a,2.4.x-xfs:slinx:91149a
              2.4.x-xfs:slinx:91148a,2.4.x-xfs:slinx:91204a
              2.4.x-xfs:slinx:91222a

The following file(s) were checked into:
  bonnie.engr.sgi.com:/isms/slinx/2.4.x-xfs-r1.0


Modid:  2.4.x-xfs-r1.0:slinx:91232a
SPECS/quota.spec - 1.1
	- Merge of 2.4.x-xfs:slinx:91222a originally by nathans on 03/29/01
	  the new, improved quota spec file for 3.01.

SOURCES/quota-3.01-pre2.tar.gz - 1.1
	- Merge of 2.4.x-xfs:slinx:91222a originally by nathans on 03/29/01
	  move up to quota-3.01 user tools.
	  move up to quota-3.01 user tools.

linux/fs/xfs/linux/xfs_linux.h - 1.44
	- Merge of 2.4.x-xfs:slinx:91204a originally by mann on 03/29/01
	  EFSCORRUPTED needs to be different from EIO.

Makefile - 1.43
	- Merge of 2.4.x-xfs:slinx:91222a originally by nathans on 03/29/01
	  use the new, improved quota spec file for 3.01.

linux/fs/pagebuf/page_buf_io.c - 1.70
	- Merge of 2.4.x-xfs:slinx:91142a originally by cattelan on 03/28/01
	  Cast page->index to 64 bit before multiplying by page size.
	  This was causing the offset passed to bmap to overflow for file
	  indexes over 4gig.

cmd/quota/Changelog - 1.3
cmd/quota/Makefile.in - 1.8
cmd/quota/README.gettext - 1.2
cmd/quota/README.xfs - 1.9
cmd/quota/bylabel.c - 1.3
cmd/quota/bylabel.h - 1.2
cmd/quota/configure - 1.2
cmd/quota/configure.in - 1.6
cmd/quota/doc/edquota(8).html - 1.2
cmd/quota/doc/fstab(5).html - 1.2
cmd/quota/doc/quota(1).html - 1.2
cmd/quota/doc/quota.html - 1.2
cmd/quota/doc/quota4th.fig - 1.2
cmd/quota/doc/quotacheck(8).html - 1.2
cmd/quota/doc/quotactl(2).html - 1.2
cmd/quota/doc/quotaon(8).html - 1.2
cmd/quota/doc/quotas-1.eps - 1.2
cmd/quota/doc/quotas.ms - 1.2
cmd/quota/doc/quotas.preformated - 1.2
cmd/quota/doc/repquota(8).html - 1.2
cmd/quota/doc/rquotad(8).html - 1.2
cmd/quota/dqblk.h - 1.2
cmd/quota/edquota.8 - 1.4
cmd/quota/edquota.c - 1.7
cmd/quota/hasquota.c - 1.7
cmd/quota/install-sh - 1.2
cmd/quota/mntent.h - 1.3
cmd/quota/po/pl.mo - 1.3
cmd/quota/po/pl.po - 1.2
cmd/quota/pot.c - 1.2
cmd/quota/pot.h - 1.2
cmd/quota/quota.1 - 1.5
cmd/quota/quota.c - 1.7
cmd/quota/quotacheck.8 - 1.5
cmd/quota/quotacheck.c - 1.6
cmd/quota/quotactl.2 - 1.3
cmd/quota/quotactl.c - 1.3
cmd/quota/quotaio.c - 1.5
cmd/quota/quotaon.8 - 1.6
cmd/quota/quotaon.c - 1.12
cmd/quota/quotaops.c - 1.11
cmd/quota/quotaops.h - 1.5
cmd/quota/quotastats.c - 1.4
cmd/quota/repquota.8 - 1.6
cmd/quota/repquota.c - 1.17
cmd/quota/rquota.3 - 1.2
cmd/quota/rquota.h - 1.3
cmd/quota/rquota.x - 1.3
cmd/quota/rquota_client.c - 1.3
cmd/quota/rquota_clnt.c - 1.3
cmd/quota/rquota_server.c - 1.7
cmd/quota/rquota_svc.c - 1.2
cmd/quota/rquota_xdr.c - 1.3
cmd/quota/rquotad.8 - 1.4
cmd/quota/set_limits_example.c - 1.2
cmd/quota/setquota.8 - 1.6
cmd/quota/setquota.c - 1.8
cmd/quota/setup_quota_group - 1.2
cmd/quota/warnquota.c - 1.6
cmd/quota/warnquota.conf - 1.2
cmd/quota/xqm.h - 1.4
cmd/quota/hasquota.h - 1.2
	- Merge of 2.4.x-xfs:slinx:91222a originally by nathans on 03/29/01
	  nuke this, move up to quota-3.01 user tools on sourceforge.net as these
	  support XFS quota natively now.

cmd/attr/libattr/attr.c - 1.5
	- Merge of 2.4.x-xfs:slinx:91145a originally by ajag on 03/28/01
	  change over to ia64 friendly syscall() instead of _syscall* macros

cmd/attr/Makepkgs - 1.3
	- Merge of 2.4.x-xfs:slinx:91148a originally by nathans on 03/28/01
	  make clean before proceeding to do a build to ensure no leftovers
	  corrupt subsequent invocations (for rpm builds).

cmd/xfsdump/common/getdents.c - 1.2
	- Merge of 2.4.x-xfs:slinx:91145a originally by ajag on 03/28/01
	  change over to ia64 friendly syscall() instead of _syscall* macros

cmd/xfsdump/Makepkgs - 1.2
	- Merge of 2.4.x-xfs:slinx:91148a originally by nathans on 03/28/01
	  make clean before proceeding to do a build to ensure no leftovers
	  corrupt subsequent invocations (for rpm builds).

cmd/xfsdump/VERSION - 1.4
cmd/xfsdump/doc/CHANGES - 1.4
	- Merge of 2.4.x-xfs:slinx:91138a originally by nathans on 03/28/01
	  bump the version number after recent spec file changes.

cmd/xfstests/common.quota - 1.6
	- Merge of 2.4.x-xfs:slinx:91146a originally by nathans on 03/28/01
	  do qa testing on the new quota userspace.

cmd/xfstests/tools/srctest - 1.2
	- Merge of 2.4.x-xfs:slinx:91148a originally by nathans on 03/28/01
	  needed updating - still referred to xfs-cmds explicitly.  no takes
	  package as an argument before testing whether package can rebuild
	  itself successfully.

cmd/xfstests/tools/auto-qa - 1.12
	- Merge of 2.4.x-xfs:slinx:91146a originally by nathans on 03/28/01
	  do qa testing on the new quota userspace.

cmd/xfsprogs/repair/attr_repair.c - 1.3
	- Merge of 2.4.x-xfs:slinx:91149a originally by ajag on 03/28/01
	  Add endian conversion to xfs_repair to prevent removal of valid acls.

cmd/xfsprogs/logprint/logprint.h - 1.2
cmd/xfsprogs/libxfs/xfs.h - 1.3
	- Merge of 2.4.x-xfs:slinx:91204a originally by mann on 03/29/01
	  Syncing with xfs_linux.h

cmd/xfsprogs/Makepkgs - 1.3
cmd/acl/Makepkgs - 1.2
	- Merge of 2.4.x-xfs:slinx:91148a originally by nathans on 03/28/01
	  make clean before proceeding to do a build to ensure no leftovers
	  corrupt subsequent invocations (for rpm builds).

cmd/acl/libacl/acl.c - 1.5
	- Merge of 2.4.x-xfs:slinx:91145a originally by ajag on 03/28/01
	  change over to ia64 friendly syscall() instead of _syscall* macros
	  also fix acl_valid so that it accepts acls with ACL_NOT_PRESENT - this
	  was preventing chacl from removing acls

cmd/dmapi/libdm/dm_handle2path.c - 1.4
	- Merge of 2.4.x-xfs:slinx:91145a originally by ajag on 03/28/01
	  change over to ia64 friendly syscall() instead of _syscall* macros

cmd/dmapi/Makepkgs - 1.3
	- Merge of 2.4.x-xfs:slinx:91148a originally by nathans on 03/28/01
	  make clean before proceeding to do a build to ensure no leftovers
	  corrupt subsequent invocations (for rpm builds).

SPECS/quota-xfs.spec - 1.5
SOURCES/quota-2.00.tar.gz - 1.2
	- Merge of 2.4.x-xfs:slinx:91222a originally by nathans on 03/29/01
	  nuke this, move up to quota-3.01 user tools.

SOURCES/quota-2.00-xfs.patch - 1.6
	- Merge of 2.4.x-xfs:slinx:91222a originally by nathans on 03/29/01
	  we no longer need a patch, code is in 3.01.

cmd/xfsprogs/doc/README.quota - 1.2
	- Merge of 2.4.x-xfs:slinx:91146a originally by nathans on 03/28/01
	  update to document the new quota userspace version & changes in xfsdump.



From owner-linux-xfs@oss.sgi.com Thu Mar 29 23:03:27 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2U73R922936
	for linux-xfs-outgoing; Thu, 29 Mar 2001 23:03:27 -0800
Received: from lupo.thebarn.com (nic-25-c96-008.mn.mediaone.net [24.25.96.8])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2U73PM22927;
	Thu, 29 Mar 2001 23:03:25 -0800
Received: from lupo.thebarn.com (localhost [127.0.0.1])
	by lupo.thebarn.com (8.11.3/8.11.0) with ESMTP id f2U73Ie95094;
	Fri, 30 Mar 2001 01:03:19 -0600 (CST)
	(envelope-from cattelan@thebarn.com)
Date: Fri, 30 Mar 2001 01:03:16 -0600
Message-ID: <ume4rwbbux7.wl@lupo.thebarn.com>
From: cattelan@thebarn.com
To: linux-xfs@oss.sgi.com, linux-xfs-announce@oss.sgi.com
Subject: Prerelease 0.10.1 images now available.
User-Agent: Wanderlust/2.4.0 (Rio) SEMI/1.13.7 (Awazu) FLIM/1.13.2 (Kasanui) MULE XEmacs/21.1 (patch 14) (Cuyahoga Valley) (i386--freebsd)
MIME-Version: 1.0 (generated by SEMI 1.13.7 - "Awazu")
Content-Type: text/plain; charset=US-ASCII
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

Seems like it was just yesterday that that 0.10 images came out...
ohhh it was. :-)

Well several bugs have been fixed over the past few day so
guess it is time to update things.

The most significant potential show stopper that has been fixed
was the reading of a file past 4gig resulted in the bmaps wrapping
back to the beginning.

errno EFSCORRUPTED has been reverted back to it's own value
using EIO was resulting in certain portions of the code
incorrectly handling the shut down case.

Along with the bug fixes a few rpm/config issues have been
changed:
o SGI partition support is now on by default.

o RPMs have been generated on a 6.2 rpm 3.0.5 based system
	rather than a Wolverine rpm 4.0.x based system.
	This should provide better compatibility with RH 6.2 on up 
	and possibly other rpm based distributions.

o kernel rpm requirements on ppp, kudzu, e2fsprogs, isdn.
	have been removed. These were hold overs from the RH Rawhide 
	stuff, and probably shouldn't cause problems. 

o Conflicts that were added xfsprogs < 1.1.4 and xfsdump < 1.0.2
	due to changes in the SETBLKSIZE ioctl.

The installer has not been rev'd to this release level.

No changes to the installer have been made at this time
so in an effort to save time and effort, it will not be
updated until which time is has something significant to 
say.

As always for the brave outdoories type the likes to chop their own
wood and drive their own compilers  tar file of the commands and
kernel patch files have been generated.

ftp://linux-xfs.sgi.com/projects/xfs/download/PreRelease-0.10.1-test1/

Ohh one final note I haven't actually booted these images so 
please don't shoot me if by some obscure reason they don't work.

Please do drop us a note although.

-Russell "time for a beer" Cattelan

From owner-linux-xfs@oss.sgi.com Thu Mar 29 23:05:27 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2U75RM23034
	for linux-xfs-outgoing; Thu, 29 Mar 2001 23:05:27 -0800
Received: from lupo.thebarn.com (nic-25-c96-008.mn.mediaone.net [24.25.96.8])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2U75QM23017;
	Thu, 29 Mar 2001 23:05:26 -0800
Received: from lupo.thebarn.com (localhost [127.0.0.1])
	by lupo.thebarn.com (8.11.3/8.11.0) with ESMTP id f2U75Je95100;
	Fri, 30 Mar 2001 01:05:20 -0600 (CST)
	(envelope-from cattelan@thebarn.com)
Date: Fri, 30 Mar 2001 01:05:19 -0600
Message-ID: <ume3dbvbuts.wl@lupo.thebarn.com>
From: cattelan@thebarn.com
To: linux-xfs@oss.sgi.com, linux-xfs-announce@oss.sgi.com
Subject: linux-2.4-xfs-beta tree updated
User-Agent: Wanderlust/2.4.0 (Rio) SEMI/1.13.7 (Awazu) FLIM/1.13.2 (Kasanui) MULE XEmacs/21.1 (patch 14) (Cuyahoga Valley) (i386--freebsd)
MIME-Version: 1.0 (generated by SEMI 1.13.7 - "Awazu")
Content-Type: text/plain; charset=US-ASCII
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

Oh and BTW the cvs beta tree has been to the 0.10.1 tree.


From owner-linux-xfs@oss.sgi.com Fri Mar 30 01:57:01 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2U9v1s29195
	for linux-xfs-outgoing; Fri, 30 Mar 2001 01:57:01 -0800
Received: from ns.tecosim.de (ns.tecosim.de [194.24.222.9])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2U9ueM29173
	for <linux-xfs@oss.sgi.com>; Fri, 30 Mar 2001 01:56:49 -0800
Received: from donner.tecosim.de (root@donner.tecosim.de [194.24.222.109])
	by ns.tecosim.de (8.9.3/8.9.3/SuSE Linux 8.9.3-0.1) with ESMTP id LAA23359;
	Fri, 30 Mar 2001 11:56:30 +0200
Received: (from leh@localhost)
	by donner.tecosim.de (8.11.2/8.11.2/SuSE Linux 8.11.1-0.5) id f2U9uSg01395;
	Fri, 30 Mar 2001 11:56:28 +0200
Date: Fri, 30 Mar 2001 11:56:28 +0200
From: Utz Lehmann <leh@tecosim.de>
To: Russell Cattelan <cattelan@thebarn.com>
Cc: Eric Sandeen <sandeen@sgi.com>, linux-xfs@oss.sgi.com
Subject: Re: shutdown umount hangs
Message-ID: <20010330115628.A1146@tecosim.de>
References: <20010329165519.A7386@tecosim.de> <20010329170446.B7386@tecosim.de> <3AC35E8D.711963C5@sgi.com> <20010329181419.G7386@tecosim.de> <20010329182044.H7386@tecosim.de> <3AC38727.FC9C4FC8@thebarn.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.2.5i
In-Reply-To: <3AC38727.FC9C4FC8@thebarn.com>; from cattelan@thebarn.com on Thu, Mar 29, 2001 at 02:04:07PM -0500
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

Russell Cattelan [cattelan@thebarn.com] wrote:
> Utz Lehmann wrote:
> 
> > Utz Lehmann [leh@tecosim.de] wrote:
> > > Eric Sandeen [sandeen@sgi.com] wrote:
> > > > Utz Lehmann wrote:
> > > >
> > > > > > With newer cvs kernel (cvs from this week, before i had holidays) i have
> > > > > > problems with my linux workstation (suse 7.1) at work.
> > > > > >
> > > > > > Umounting the filesystems while shutdown hangs.
> > > >
> > > > Are there any errors related to unmount in the syslog?
> > >
> > > No, but syslogd was killed before umounting.
> > > There were no messages on console, it just hangs. sysreq works.
> >
> > btw: is there a way to trap it into the kdb?
> 
> You should still be able to drop into kdb, a back trace would be really helpful if
> you can get it.

ok, here it is:

Unmounting file systems
shmfs umounted
/dev/sda2 umounted
/dev/vg00/tmp umounted
/dev/vg00/opt umounted

[wait some time]

Entering kdb (current=0xc035a000, pid 0) due to Keyboard Entry
kdb> bt
    EBP       EIP         Function(args)
0x0008e000 0xc0107193 default_idle+0x23
                               kernel .text 0xc0100000 0xc0107170 0xc01071a0
           0xc01071f8 cpu_idle+0x38
                               kernel .text 0xc0100000 0xc01071c0 0xc0107210
           0xc035c9ba start_kernel+0x13a
                               kernel .text.init 0xc035c000 0xc035c880 0xc035c9c0
kdb> go

[wait some time]

Entering kdb (current=0xc035a000, pid 0) due to Keyboard Entry
kdb> bt
    EBP       EIP         Function(args)
0x0008e000 0xc0107193 default_idle+0x23
                               kernel .text 0xc0100000 0xc0107170 0xc01071a0
           0xc01071f8 cpu_idle+0x38
                               kernel .text 0xc0100000 0xc01071c0 0xc0107210
           0xc035c9ba start_kernel+0x13a
                               kernel .text.init 0xc035c000 0xc035c880 0xc035c9c0
kdb> reboot



a working shutdown looks like this (with a cvs kernel from march 2th):

Unmounting file systems
shmfs umounted
/dev/sda2 umounted
/dev/vg00/tmp umounted
/dev/vg00/opt umounted
/dev/vg00/var umounted
/dev/vg00/usr umounted
devpts umounted
/dev/sda1 umounted                                                    done
proc umounted
vgchange -- volume group "vg00" successfully deactivated

Please stand by while rebooting the system...
stopping all md devices.
Restarting system.


btw: i test it without the kio,logbufs=4,logbsize=32768 mount option, it
hangs also.


utz

From owner-linux-xfs@oss.sgi.com Fri Mar 30 02:17:48 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2UAHme29752
	for linux-xfs-outgoing; Fri, 30 Mar 2001 02:17:48 -0800
Received: from main.braxis.co.uk (root@main.braxis.co.uk [212.160.232.26])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2UAHKM29749
	for <linux-xfs@oss.sgi.com>; Fri, 30 Mar 2001 02:17:24 -0800
Received: (from kszysiu@localhost)
	by main.braxis.co.uk (8.9.3/8.9.3) id MAA04572;
	Fri, 30 Mar 2001 12:11:59 +0200
Date: Fri, 30 Mar 2001 12:11:59 +0200
From: Krzysztof Rusocki <kszysiu@braxis.co.uk>
To: Utz Lehmann <leh@tecosim.de>
Cc: linux-xfs@oss.sgi.com
Subject: Re: shutdown umount hangs
Message-ID: <20010330121159.B3568@main.braxis.co.uk>
References: <20010329165519.A7386@tecosim.de> <20010329170446.B7386@tecosim.de> <3AC35E8D.711963C5@sgi.com> <20010329181419.G7386@tecosim.de> <20010329182044.H7386@tecosim.de> <3AC38727.FC9C4FC8@thebarn.com> <20010330115628.A1146@tecosim.de>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.2.5i
In-Reply-To: <20010330115628.A1146@tecosim.de>; from leh@tecosim.de on Fri, Mar 30, 2001 at 11:56:28AM +0200
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk


Hi,

I've also had similar problem to yours ..

When such (let's call it) "hang" occured i tried Emergency Sync
and Emergency Remount R/O  - it did not work...

After Alt+SysRQ+B .. system rebooted but i noticed that
one of my xfs partitions was NOT clearly unmounted
 - during mount it was recovered...

Then I tried to reboot ...  yet again "hang" occured..
And after reboot  the same  partition was 'recovered'..

I mounted it R/O and xfs_repair'ed ... then mounted R/W..

And during all next reboots/shutdowns such "hang" has NOT happened any
more...

seems strange....

Cheers

Krzysztof

From owner-linux-xfs@oss.sgi.com Fri Mar 30 02:38:39 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2UAcdZ30017
	for linux-xfs-outgoing; Fri, 30 Mar 2001 02:38:39 -0800
Received: from ns.tecosim.de (ns.tecosim.de [194.24.222.9])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2UAcXM30012
	for <linux-xfs@oss.sgi.com>; Fri, 30 Mar 2001 02:38:34 -0800
Received: from donner.tecosim.de (root@donner.tecosim.de [194.24.222.109])
	by ns.tecosim.de (8.9.3/8.9.3/SuSE Linux 8.9.3-0.1) with ESMTP id MAA24104;
	Fri, 30 Mar 2001 12:38:26 +0200
Received: (from leh@localhost)
	by donner.tecosim.de (8.11.2/8.11.2/SuSE Linux 8.11.1-0.5) id f2UAcP406563;
	Fri, 30 Mar 2001 12:38:25 +0200
Date: Fri, 30 Mar 2001 12:38:25 +0200
From: Utz Lehmann <leh@tecosim.de>
To: Krzysztof Rusocki <kszysiu@braxis.co.uk>
Cc: linux-xfs@oss.sgi.com
Subject: Re: shutdown umount hangs
Message-ID: <20010330123825.A1254@tecosim.de>
References: <20010329165519.A7386@tecosim.de> <20010329170446.B7386@tecosim.de> <3AC35E8D.711963C5@sgi.com> <20010329181419.G7386@tecosim.de> <20010329182044.H7386@tecosim.de> <3AC38727.FC9C4FC8@thebarn.com> <20010330115628.A1146@tecosim.de> <20010330121159.B3568@main.braxis.co.uk>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.2.5i
In-Reply-To: <20010330121159.B3568@main.braxis.co.uk>; from kszysiu@braxis.co.uk on Fri, Mar 30, 2001 at 12:11:59PM +0200
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

Are you using lvm partitions?


Krzysztof Rusocki [kszysiu@braxis.co.uk] wrote:
> 
> Hi,
> 
> I've also had similar problem to yours ..
> 
> When such (let's call it) "hang" occured i tried Emergency Sync
> and Emergency Remount R/O  - it did not work...
> 
> After Alt+SysRQ+B .. system rebooted but i noticed that
> one of my xfs partitions was NOT clearly unmounted
>  - during mount it was recovered...
> 
> Then I tried to reboot ...  yet again "hang" occured..
> And after reboot  the same  partition was 'recovered'..
> 
> I mounted it R/O and xfs_repair'ed ... then mounted R/W..
> 
> And during all next reboots/shutdowns such "hang" has NOT happened any
> more...
> 
> seems strange....
> 
> Cheers
> 
> Krzysztof

From owner-linux-xfs@oss.sgi.com Fri Mar 30 02:54:32 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2UAsWs30280
	for linux-xfs-outgoing; Fri, 30 Mar 2001 02:54:32 -0800
Received: from main.braxis.co.uk (root@main.braxis.co.uk [212.160.232.26])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2UAqPM30233
	for <linux-xfs@oss.sgi.com>; Fri, 30 Mar 2001 02:52:25 -0800
Received: (from kszysiu@localhost)
	by main.braxis.co.uk (8.9.3/8.9.3) id MAA06034;
	Fri, 30 Mar 2001 12:44:59 +0200
Date: Fri, 30 Mar 2001 12:44:59 +0200
From: Krzysztof Rusocki <kszysiu@braxis.co.uk>
To: Utz Lehmann <leh@tecosim.de>
Cc: linux-xfs@oss.sgi.com
Subject: Re: shutdown umount hangs
Message-ID: <20010330124459.A5994@main.braxis.co.uk>
References: <20010329165519.A7386@tecosim.de> <20010329170446.B7386@tecosim.de> <3AC35E8D.711963C5@sgi.com> <20010329181419.G7386@tecosim.de> <20010329182044.H7386@tecosim.de> <3AC38727.FC9C4FC8@thebarn.com> <20010330115628.A1146@tecosim.de> <20010330121159.B3568@main.braxis.co.uk> <20010330123825.A1254@tecosim.de>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.2.5i
In-Reply-To: <20010330123825.A1254@tecosim.de>; from leh@tecosim.de on Fri, Mar 30, 2001 at 12:38:25PM +0200
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

On Fri, Mar 30, 2001 at 12:38:25PM +0200, Utz Lehmann wrote:
> Are you using lvm partitions?

nope. i do not..

Krzysztof


> 
> 
> Krzysztof Rusocki [kszysiu@braxis.co.uk] wrote:
> > 
> > Hi,
> > 
> > I've also had similar problem to yours ..
> > 
> > When such (let's call it) "hang" occured i tried Emergency Sync
> > and Emergency Remount R/O  - it did not work...
> > 
> > After Alt+SysRQ+B .. system rebooted but i noticed that
> > one of my xfs partitions was NOT clearly unmounted
> >  - during mount it was recovered...
> > 
> > Then I tried to reboot ...  yet again "hang" occured..
> > And after reboot  the same  partition was 'recovered'..
> > 
> > I mounted it R/O and xfs_repair'ed ... then mounted R/W..
> > 
> > And during all next reboots/shutdowns such "hang" has NOT happened any
> > more...
> > 
> > seems strange....
> > 
> > Cheers
> > 
> > Krzysztof

From owner-linux-xfs@oss.sgi.com Fri Mar 30 05:46:44 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2UDkic00909
	for linux-xfs-outgoing; Fri, 30 Mar 2001 05:46:44 -0800
Received: from mail.coltex.nl (edge.coltex.nl [194.151.97.115])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2UDkdM00906
	for <linux-xfs@oss.sgi.com>; Fri, 30 Mar 2001 05:46:43 -0800
Received: from auto-nb1.xs4all.nl (auto-nb1.coltex.nl [10.0.1.171])
	by  mail.coltex.nl (8.11.2/8.11.2) with ESMTP id f2UCjkK25242
	for <linux-xfs@oss.sgi.com>; Fri, 30 Mar 2001 14:45:46 +0200
Message-Id: <4.3.2.7.2.20010330154455.02bb1640@pop.xs4all.nl>
X-Sender: knuffie@pop.xs4all.nl (Unverified)
X-Mailer: QUALCOMM Windows Eudora Version 4.3.2
Date: Fri, 30 Mar 2001 15:46:37 +0200
To: linux-xfs@oss.sgi.com
From: Seth Mos <knuffie@xs4all.nl>
Subject: Kernel 2.4.3 released.
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"; format=flowed
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

Hi,

I don't think anyone missed it but hey.
See your favorite mirror.

Bye
--
Seth
Has anybody seen my lightbulb?
I _really_ need some light here.


From owner-linux-xfs@oss.sgi.com Fri Mar 30 06:39:50 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2UEdoJ01867
	for linux-xfs-outgoing; Fri, 30 Mar 2001 06:39:50 -0800
Received: from wiley.ceo.com (66-2-81-26.customer.algx.net [66.2.81.26])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2UEdcM01863
	for <linux-xfs@oss.sgi.com>; Fri, 30 Mar 2001 06:39:39 -0800
Received: from mindspring.com (IDENT:danny@localhost [127.0.0.1])
	by wiley.ceo.com (8.9.3/8.9.3) with ESMTP id JAA01913
	for <linux-xfs@oss.sgi.com>; Fri, 30 Mar 2001 09:53:21 -0500
Message-ID: <3AC49DE0.AFFB6A75@mindspring.com>
Date: Fri, 30 Mar 2001 09:53:20 -0500
From: Danny <danscox@mindspring.com>
Reply-To: dcox@connex.com
Organization: Connex Inc
X-Mailer: Mozilla 4.76 [en] (X11; U; Linux 2.4.2-XFS i686)
X-Accept-Language: en
MIME-Version: 1.0
To: Linux-XFS <linux-xfs@oss.sgi.com>
Subject: [Fwd: Unmount Didn't; Clues?] Now Reproducible
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

All,

I wrote:
>         Updated from CVS this AM, this just happened to me:
> 
> ======================================================================================
> Mar 29 03:12:26 dsc_proto_1 kernel: Start mounting filesystem: md(9,1)
> Mar 29 03:12:27 dsc_proto_1 kernel: Ending clean XFS mount for
> filesystem: md(9,1)
> Mar 29 03:19:15 dsc_proto_1 kernel: XFS unmount got error 38
> Mar 29 03:19:15 dsc_proto_1 kernel: linvfs_put_super: vfsp/0xc3bc2ae0
> left dangling!
> Mar 29 03:19:15 dsc_proto_1 kernel: VFS: Busy inodes after unmount.
> Self-destruct in 5 seconds.  Have a nice day...
> Mar 29 03:19:56 dsc_proto_1 kernel: XFS: Filesystem has duplicate UUID -
> can't mount
> Mar 29 03:21:35 dsc_proto_1 last message repeated 2 times
> =====================================================================================
> 

	I (in a Diet-Coke induced fog) tried the "dmapi" mount option, and the
unmount always fails in this case.  I'm using XFS on software-raid5 or
raid1.  I havn't tried it on a plain block dev yet.  Since I don't have
DMAPI support turned on in the kernel config, this shouldn't have come
as a surprise, no?

	Thanks, and sorry for the bandwidth abuse.

-- 
"Men occasionally stumble over the truth, but most of them pick 
themselves up and hurry off as if nothing had happened." 
   -- Winston Churchill 

Danny

From owner-linux-xfs@oss.sgi.com Fri Mar 30 07:26:56 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2UFQu802646
	for linux-xfs-outgoing; Fri, 30 Mar 2001 07:26:56 -0800
Received: from lips.borg.umn.edu (lips.borg.umn.edu [160.94.232.50])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2UFQqM02643
	for <linux-xfs@oss.sgi.com>; Fri, 30 Mar 2001 07:26:52 -0800
Received: from thebarn.com (nic-25-c96-008.mn.mediaone.net [24.25.96.8])
	by lips.borg.umn.edu (8.12.0.Beta5/8.12.0.Beta5) with ESMTP id f2UFQadE099891;
	Fri, 30 Mar 2001 09:26:40 -0600 (CST)
Message-ID: <3AC4A50F.B141110B@thebarn.com>
Date: Fri, 30 Mar 2001 09:23:59 -0600
From: Russell Cattelan <cattelan@thebarn.com>
X-Mailer: Mozilla 4.75 [en] (X11; U; Linux 2.4.2-XFS i686)
X-Accept-Language: en
MIME-Version: 1.0
To: Utz Lehmann <leh@tecosim.de>
CC: Eric Sandeen <sandeen@sgi.com>, linux-xfs@oss.sgi.com
Subject: Re: shutdown umount hangs
References: <20010329165519.A7386@tecosim.de> <20010329170446.B7386@tecosim.de> <3AC35E8D.711963C5@sgi.com> <20010329181419.G7386@tecosim.de> <20010329182044.H7386@tecosim.de> <3AC38727.FC9C4FC8@thebarn.com> <20010330115628.A1146@tecosim.de>
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

Utz Lehmann wrote:

> Russell Cattelan [cattelan@thebarn.com] wrote:
>

Ok sorry this was my fault I should have explained this better.
We need the back trace of the umount process.
to do this simply type ps from the kdb prompt  find the umount pid
the type
kdb> btp <umount pid>



>
>
> ok, here it is:
>
> Unmounting file systems
> shmfs umounted
> /dev/sda2 umounted
> /dev/vg00/tmp umounted
> /dev/vg00/opt umounted
>
> [wait some time]
>
> Entering kdb (current=0xc035a000, pid 0) due to Keyboard Entry
> kdb> bt
>     EBP       EIP         Function(args)
> 0x0008e000 0xc0107193 default_idle+0x23
>                                kernel .text 0xc0100000 0xc0107170 0xc01071a0
>            0xc01071f8 cpu_idle+0x38
>                                kernel .text 0xc0100000 0xc01071c0 0xc0107210
>            0xc035c9ba start_kernel+0x13a
>                                kernel .text.init 0xc035c000 0xc035c880 0xc035c9c0
> kdb> go
>
> [wait some time]
>
> Entering kdb (current=0xc035a000, pid 0) due to Keyboard Entry
> kdb> bt
>     EBP       EIP         Function(args)
> 0x0008e000 0xc0107193 default_idle+0x23
>                                kernel .text 0xc0100000 0xc0107170 0xc01071a0
>            0xc01071f8 cpu_idle+0x38
>                                kernel .text 0xc0100000 0xc01071c0 0xc0107210
>            0xc035c9ba start_kernel+0x13a
>                                kernel .text.init 0xc035c000 0xc035c880 0xc035c9c0
> kdb> reboot
>
> a working shutdown looks like this (with a cvs kernel from march 2th):
>
> Unmounting file systems
> shmfs umounted
> /dev/sda2 umounted
> /dev/vg00/tmp umounted
> /dev/vg00/opt umounted
> /dev/vg00/var umounted
> /dev/vg00/usr umounted
> devpts umounted
> /dev/sda1 umounted                                                    done
> proc umounted
> vgchange -- volume group "vg00" successfully deactivated
>
> Please stand by while rebooting the system...
> stopping all md devices.
> Restarting system.
>
> btw: i test it without the kio,logbufs=4,logbsize=32768 mount option, it
> hangs also.
>
> utz


From owner-linux-xfs@oss.sgi.com Fri Mar 30 07:36:33 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2UFaXB02834
	for linux-xfs-outgoing; Fri, 30 Mar 2001 07:36:33 -0800
Received: from thor.theseus.com (south.orl-pub.theseus.com [12.108.42.66])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2UFaWM02831
	for <linux-xfs@oss.sgi.com>; Fri, 30 Mar 2001 07:36:32 -0800
Received: from theseus.com (IDENT:thebs@fugitive.theseus.com [192.168.0.242])
	by thor.theseus.com (8.9.3/8.9.3) with ESMTP id KAA13997;
	Fri, 30 Mar 2001 10:41:39 -0500
Message-ID: <3AC4A84D.308B33AF@theseus.com>
Date: Fri, 30 Mar 2001 10:37:49 -0500
From: Bryan-TheBS-Smith <thebs@theseus.com>
Reply-To: thebs@theseus.com, b.j.smith@ieee.org
Organization: (Personal)
X-Mailer: Mozilla 4.76 [en] (X11; U; Linux 2.2.17-0.17-FUGITIVE i686)
X-Accept-Language: en
MIME-Version: 1.0
To: cattelan@thebarn.com
CC: linux-xfs@oss.sgi.com
Subject: Re: Prerelease 0.10.1 images now available.
References: <ume4rwbbux7.wl@lupo.thebarn.com>
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

cattelan@thebarn.com wrote:
> o RPMs have been generated on a 6.2 rpm 3.0.5 based system
>         rather than a Wolverine rpm 4.0.x based system.
>         This should provide better compatibility with RH 6.2 on up
>         and possibly other rpm based distributions.

_Dumb_ question:  Doesn't compiling on an older platform take issue
(or not take advantage of) the newer GLibC and other capabilities?

-- TheBS

--
Bryan "TheBS" Smith          chat:thebs413 @AOL/MSN/Yahoo
Engineer      mailto:b.j.smith@ieee.org,thebs@theseus.com
*********************************************************
"Never apply a Star Trek solution to a Babylon 5 problem"
                                    -- Nicholas C. Weaver

From owner-linux-xfs@oss.sgi.com Fri Mar 30 07:59:59 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2UFxxn03260
	for linux-xfs-outgoing; Fri, 30 Mar 2001 07:59:59 -0800
Received: from lips.borg.umn.edu (lips.borg.umn.edu [160.94.232.50])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2UFxwM03253
	for <linux-xfs@oss.sgi.com>; Fri, 30 Mar 2001 07:59:59 -0800
Received: from thebarn.com (nic-25-c96-008.mn.mediaone.net [24.25.96.8])
	by lips.borg.umn.edu (8.12.0.Beta5/8.12.0.Beta5) with ESMTP id f2UFxrdE000333;
	Fri, 30 Mar 2001 09:59:53 -0600 (CST)
Message-ID: <3AC4AD74.72EA793@thebarn.com>
Date: Fri, 30 Mar 2001 09:59:48 -0600
From: Russell Cattelan <cattelan@thebarn.com>
X-Mailer: Mozilla 4.74 [en] (X11; U; Linux 2.2.12 i386)
X-Accept-Language: en
MIME-Version: 1.0
To: thebs@theseus.com, b.j.smith@ieee.org
CC: linux-xfs@oss.sgi.com
Subject: Re: Prerelease 0.10.1 images now available.
References: <ume4rwbbux7.wl@lupo.thebarn.com> <3AC4A84D.308B33AF@theseus.com>
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

Bryan-TheBS-Smith wrote:

> cattelan@thebarn.com wrote:
> > o RPMs have been generated on a 6.2 rpm 3.0.5 based system
> >         rather than a Wolverine rpm 4.0.x based system.
> >         This should provide better compatibility with RH 6.2 on up
> >         and possibly other rpm based distributions.
>
> _Dumb_ question:  Doesn't compiling on an older platform take issue
> (or not take advantage of) the newer GLibC and other capabilities?

This shouldn't change the interaction of the kernel and glibc,
unless there is a type change between the two versions.

But hey lets find out.... that is the reason these kernel are labeled
test.


>
>
> -- TheBS
>
> --
> Bryan "TheBS" Smith          chat:thebs413 @AOL/MSN/Yahoo
> Engineer      mailto:b.j.smith@ieee.org,thebs@theseus.com
> *********************************************************
> "Never apply a Star Trek solution to a Babylon 5 problem"
>                                     -- Nicholas C. Weaver

--
Russell Cattelan
cattelan@thebarn.com




From owner-linux-xfs@oss.sgi.com Fri Mar 30 09:10:42 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2UHAgZ04685
	for linux-xfs-outgoing; Fri, 30 Mar 2001 09:10:42 -0800
Received: from sws5.ctd.ornl.gov (sws5.ctd.ornl.gov [160.91.68.105])
	by oss.sgi.com (8.11.3/8.11.3) with SMTP id f2UHAfM04682
	for <linux-xfs@oss.sgi.com>; Fri, 30 Mar 2001 09:10:41 -0800
Received: (qmail 4095 invoked by uid 3995); 30 Mar 2001 17:10:40 -0000
Message-ID: <20010330171040.4094.qmail@sws5.ctd.ornl.gov>
From: de5@sws5.ctd.ornl.gov
To: linux-xfs@oss.sgi.com
Reply-To: de5-dated-cef51442f276d23e@sws5.ctd.ornl.gov
Subject: Won't mount
Content-Type: text/plain; charset=US-ASCII
Date: 30 Mar 2001 12:10:40 -0500
Lines: 22
User-Agent: Gnus/5.0807 (Gnus v5.8.7) XEmacs/21.1 (Cuyahoga Valley)
MIME-Version: 1.0
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

I unmounted the IRIX filesystem I mounted a couple days ago, but it
won't mount now:

# mount -t xfs /dev/sda /mnt
mount: wrong fs type, bad option, bad superblock on /dev/sda,
       or too many mounted file systems

The fs type is right, the options worked before, the fs is definitely
unmounted:

# df
Filesystem           1k-blocks      Used Available Use% Mounted on
/dev/ide/host0/bus0/target0/lun0/part6
                       1011928    146032    814492  16% /
/dev/ide/host0/bus0/target0/lun0/part1
                         23302     14324      7775  65% /boot
/dev/ide/host0/bus0/target0/lun0/part8
                      24825848   6514916  17049832  28% /home
/dev/ide/host0/bus0/target0/lun0/part5
                       3028080   2104668    769592  74% /usr

-Dave

From owner-linux-xfs@oss.sgi.com Fri Mar 30 09:17:25 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2UHHPc04756
	for linux-xfs-outgoing; Fri, 30 Mar 2001 09:17:25 -0800
Received: from deliverator.sgi.com (deliverator.sgi.com [204.94.214.10])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2UHHPM04753
	for <linux-xfs@oss.sgi.com>; Fri, 30 Mar 2001 09:17:25 -0800
Received: from cthulhu.engr.sgi.com (gate3-relay.engr.sgi.com [130.62.1.234]) by deliverator.sgi.com (980309.SGI.8.8.8-aspam-6.2/980310.SGI-aspam) via ESMTP id JAA09899
	for <linux-xfs@oss.sgi.com>; Fri, 30 Mar 2001 09:16:10 -0800 (PST)
	mail_from (tduffy@engr.sgi.com)
Received: from dbear.engr.sgi.com (dbear.engr.sgi.com [163.154.18.85])
	by cthulhu.engr.sgi.com (SGI-8.9.3/8.9.3) with ESMTP id JAA32115;
	Fri, 30 Mar 2001 09:16:07 -0800 (PST)
Date: Fri, 30 Mar 2001 09:11:53 -0800 (PST)
From: Tom Duffy <tduffy@engr.sgi.com>
To: <de5-dated-cef51442f276d23e@sws5.ctd.ornl.gov>
cc: <linux-xfs@oss.sgi.com>
Subject: Re: Won't mount
In-Reply-To: <20010330171040.4094.qmail@sws5.ctd.ornl.gov>
Message-ID: <Pine.LNX.4.30.0103300911120.10515-100000@dbear.engr.sgi.com>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

> # mount -t xfs /dev/sda /mnt
> mount: wrong fs type, bad option, bad superblock on /dev/sda,
>        or too many mounted file systems

do you really want to be mounting /dev/sda or some partition on /dev/sda?

-tduffy



From owner-linux-xfs@oss.sgi.com Fri Mar 30 09:19:55 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2UHJtN04783
	for linux-xfs-outgoing; Fri, 30 Mar 2001 09:19:55 -0800
Received: from relay02.sportsline.com (relay02.sportsline.com [63.72.118.49])
	by oss.sgi.com (8.11.3/8.11.3) with SMTP id f2UHJsM04780
	for <linux-xfs@oss.sgi.com>; Fri, 30 Mar 2001 09:19:54 -0800
Received: (qmail 31371 invoked from network); 30 Mar 2001 17:19:09 -0000
Received: from chipsworld.llamas.net (63.77.33.226)
  by relay02.sportsline.com with SMTP; 30 Mar 2001 17:19:09 -0000
Received: from localhost (chipper@localhost)
	by chipsworld.llamas.net (8.11.3/8.11.3) with ESMTP id f2UHJCZ29735;
	Fri, 30 Mar 2001 12:19:12 -0500
Date: Fri, 30 Mar 2001 12:19:12 -0500 (EST)
From: "Chris 'Chipper' Chiapusio" <chipper@llamas.net>
To: <de5-dated-cef51442f276d23e@sws5.ctd.ornl.gov>
cc: <linux-xfs@oss.sgi.com>
Subject: Re: Won't mount
In-Reply-To: <20010330171040.4094.qmail@sws5.ctd.ornl.gov>
Message-ID: <Pine.LNX.4.33.0103301218350.27433-100000@chipsworld.llamas.net>
X-Files: Resist or serve
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk


perhaps you should specify a partition instead of the whole disk

eg.

mount -t xfs /dev/sda1 /mnt

Chipper

On 30 Mar 2001 de5@sws5.ctd.ornl.gov wrote:

>I unmounted the IRIX filesystem I mounted a couple days ago, but it
>won't mount now:
>
># mount -t xfs /dev/sda /mnt
>mount: wrong fs type, bad option, bad superblock on /dev/sda,
>       or too many mounted file systems
>
>The fs type is right, the options worked before, the fs is definitely
>unmounted:
>
># df
>Filesystem           1k-blocks      Used Available Use% Mounted on
>/dev/ide/host0/bus0/target0/lun0/part6
>                       1011928    146032    814492  16% /
>/dev/ide/host0/bus0/target0/lun0/part1
>                         23302     14324      7775  65% /boot
>/dev/ide/host0/bus0/target0/lun0/part8
>                      24825848   6514916  17049832  28% /home
>/dev/ide/host0/bus0/target0/lun0/part5
>                       3028080   2104668    769592  74% /usr
>
>-Dave
>

------
                    Please encrypt anything important.
PGP Key: http://wwwkeys.pgp.net:11371/pks/lookup?op=get&search=0x6CFA486D


From owner-linux-xfs@oss.sgi.com Fri Mar 30 09:20:15 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2UHKFD04801
	for linux-xfs-outgoing; Fri, 30 Mar 2001 09:20:15 -0800
Received: from ns.tecosim.de (ns.tecosim.de [194.24.222.9])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2UHKCM04798
	for <linux-xfs@oss.sgi.com>; Fri, 30 Mar 2001 09:20:13 -0800
Received: from donner.tecosim.de (root@donner.tecosim.de [194.24.222.109])
	by ns.tecosim.de (8.9.3/8.9.3/SuSE Linux 8.9.3-0.1) with ESMTP id TAA32481;
	Fri, 30 Mar 2001 19:20:08 +0200
Received: (from leh@localhost)
	by donner.tecosim.de (8.11.2/8.11.2/SuSE Linux 8.11.1-0.5) id f2UHK6G02035;
	Fri, 30 Mar 2001 19:20:06 +0200
Date: Fri, 30 Mar 2001 19:20:06 +0200
From: Utz Lehmann <leh@tecosim.de>
To: Russell Cattelan <cattelan@thebarn.com>
Cc: Eric Sandeen <sandeen@sgi.com>, linux-xfs@oss.sgi.com
Subject: Re: shutdown umount hangs
Message-ID: <20010330192006.A1149@tecosim.de>
References: <20010329165519.A7386@tecosim.de> <20010329170446.B7386@tecosim.de> <3AC35E8D.711963C5@sgi.com> <20010329181419.G7386@tecosim.de> <20010329182044.H7386@tecosim.de> <3AC38727.FC9C4FC8@thebarn.com> <20010330115628.A1146@tecosim.de> <3AC4A50F.B141110B@thebarn.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.2.5i
In-Reply-To: <3AC4A50F.B141110B@thebarn.com>; from cattelan@thebarn.com on Fri, Mar 30, 2001 at 09:23:59AM -0600
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

Russell Cattelan [cattelan@thebarn.com] wrote:
> Utz Lehmann wrote:
> 
> > Russell Cattelan [cattelan@thebarn.com] wrote:
> >
> 
> Ok sorry this was my fault I should have explained this better.
> We need the back trace of the umount process.
> to do this simply type ps from the kdb prompt  find the umount pid
> the type
> kdb> btp <umount pid>

No problem:

Unmounting file systems
shmfs umounted
/dev/sda2 umounted
/dev/vg00/tmp umounted
/dev/vg00/opt umounted

Entering kdb (current=0xc035a000, pid 0) due to Keyboard Entry
kdb> ps
Task Addr  Pid      Parent   [*] cpu  State Thread     Command
0xc15fe000 00000001 00000000  0  000  stop  0xc15fe260 init
0xc15f0000 00000002 00000001  0  000  stop  0xc15f0260 keventd
0xc15ec000 00000003 00000001  0  000  stop  0xc15ec260 kswapd
0xc15ea000 00000004 00000001  0  000  stop  0xc15ea260 kreclaimd
0xc15e8000 00000005 00000001  0  000  stop  0xc15e8260 bdflush
0xc15e6000 00000006 00000001  0  000  stop  0xc15e6260 kupdate
0xc1580000 00000007 00000001  0  000  stop  0xc1580260 mdrecoveryd
0xc1562000 00000008 00000001  0  000  stop  0xc1562260 pagebuf_daemon
0xce0b8000 00001130 00000001  0  000  stop  0xce0b8260 rc
0xce268000 00001520 00001130  0  000  stop  0xce268260 S20reboot
0xcf83c000 00001541 00001520  0  000  stop  0xcf83c260 umount
kdb> btp 1541
    EBP       EIP         Function(args)
0xcf83de58 0xc0112b18 schedule+0x2d8
                               kernel .text 0xc0100000 0xc0112840 0xc0112c70
           0xc015fb13 pagebuf_iorequest+0x103 (0xcf6ca480)
                               kernel .text 0xc0100000 0xc015fa10 0xc015fba0
           0xc01c7c47 xfs_bdstrat_cb+0x27 (0xcf6ca480)
                               kernel .text 0xc0100000 0xc01c7c20 0xc01c7c70
           0xc0160763 pagebuf_delwri_flush+0xd3 (0xcf7b5ac0, 0x1, 0xcf83dec8)
                               kernel .text 0xc0100000 0xc0160690 0xc0160880
           0xc01c7d2d XFS_bflush+0x1d (0xcf7b5ac0, 0x3a01)
                               kernel .text 0xc0100000 0xc01c7d10 0xc01c7d40
           0xc01b6873 xfs_unmount+0xd3 (0xcf7aa400, 0x0, 0xc03b16a0)
                               kernel .text 0xc0100000 0xc01b67a0 0xc01b6920
           0xc01c189a fs_dounmount+0x5a (0xcf7aa400, 0x0, 0x0, 0xc03b16a0, 0xcf7b5dc4)
                               kernel .text 0xc0100000 0xc01c1840 0xc01c18c0
           0xc01c8d38 linvfs_put_super+0x58 (0xcf8eb800)
                               kernel .text 0xc0100000 0xc01c8ce0 0xc01c8db0
           0xc0136867 kill_super+0x87 (0xcf8eb800, 0x0, 0xc156cec0, 0xffffffff, 0xcfb29ac0)
                               kernel .text 0xc0100000 0xc01367e0 0xc0136920
           0xc0136c71 do_umount+0x1c1 (0xc156cec0, 0x0, 0x0)
                               kernel .text 0xc0100000 0xc0136ab0 0xc0136c80
           0xc0136d46 sys_umount+0xc6 (0x8052430, 0x0)
more> 
                               kernel .text 0xc0100000 0xc0136c80 0xc0136d80
           0xc0136d8c sys_oldumount+0xc (0x8052430, 0x804ee27, 0x8052478, 0x8052431, 0x804ee20)
                               kernel .text 0xc0100000 0xc0136d80 0xc0136d90
           0xc0108f77 system_call+0x33
                               kernel .text 0xc0100000 0xc0108f44 0xc0108f7c
kdb> reboot




btw: 1 of ca. 7 tries the umount works.


utz

From owner-linux-xfs@oss.sgi.com Fri Mar 30 09:40:28 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2UHeSZ05546
	for linux-xfs-outgoing; Fri, 30 Mar 2001 09:40:28 -0800
Received: from sws5.ctd.ornl.gov (sws5.ctd.ornl.gov [160.91.68.105])
	by oss.sgi.com (8.11.3/8.11.3) with SMTP id f2UHeSM05543
	for <linux-xfs@oss.sgi.com>; Fri, 30 Mar 2001 09:40:28 -0800
Received: (qmail 5064 invoked by uid 3995); 30 Mar 2001 17:40:28 -0000
Message-ID: <20010330174028.5063.qmail@sws5.ctd.ornl.gov>
From: de5@sws5.ctd.ornl.gov
To: linux-xfs@oss.sgi.com
Subject: Re: Won't mount
References: <Pine.LNX.4.30.0103300911120.10515-100000@dbear.engr.sgi.com>
Content-Type: text/plain; charset=US-ASCII
Date: 30 Mar 2001 12:40:28 -0500
In-Reply-To: Tom Duffy's message of "Fri, 30 Mar 2001 09:11:53 -0800 (PST)"
Lines: 14
User-Agent: Gnus/5.0807 (Gnus v5.8.7) XEmacs/21.1 (Cuyahoga Valley)
MIME-Version: 1.0
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

Tom Duffy <tduffy@engr.sgi.com> writes:

> > # mount -t xfs /dev/sda /mnt
> > mount: wrong fs type, bad option, bad superblock on /dev/sda,
> >        or too many mounted file systems
> 
> do you really want to be mounting /dev/sda or some partition on /dev/sda?

D'oh. I thought I'd ^R'd the previous mount command, but obviously
not.

Sorry, and thanks.

-Dave

From owner-linux-xfs@oss.sgi.com Fri Mar 30 10:36:46 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2UIaki06745
	for linux-xfs-outgoing; Fri, 30 Mar 2001 10:36:46 -0800
Received: from babel.spoiled.org (babel.spoiled.org [212.84.234.227])
	by oss.sgi.com (8.11.3/8.11.3) with SMTP id f2UIaiM06742
	for <linux-xfs@oss.sgi.com>; Fri, 30 Mar 2001 10:36:45 -0800
Received: (qmail 14369 invoked by uid 8); 30 Mar 2001 18:36:42 -0000
From: thomas graichen <list-linux.sgi.xfs@spoiled.org>
Reply-To: thomas graichen <tgr@spoiled.org>
X-Newsgroups: spoiled.linux.sgi.xfs
Subject: Re: Documentation suggestions
Date: Fri, 30 Mar 2001 20:36:46 +0200
Organization: spoiled dot org
Lines: 20
Distribution: local
Message-ID: <news2mail-20010330183646.B60FDB.NOFFLE@gray.bln.innominate.de>
References: <wx0g0fw1ydx.fsf@sws5.ctd.ornl.gov> <20010329163148.C4321@tecosim.de>
Reply-To: thomas graichen <tgr@spoiled.org>
X-Complaints-To: newsmaster@spoiled.org
User-Agent: tin/1.4.4-20000803 ("Vet for the Insane") (UNIX) (Linux/2.4.2-XFS (i586))
To: linux-xfs@oss.sgi.com
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

Utz Lehmann <leh@tecosim.de> wrote:
> ds-linux-xfs@sws5.ctd.ornl.gov [ds-linux-xfs@sws5.ctd.ornl.gov] wrote:
>> In http://linux-xfs.sgi.com/projects/xfs/faq.html, under:
>> 
>>   Q: Will I be able to use my old IRIX XFS disks on linux? 
>> 
>> Mention that SGI partition support must be built into the kernel.

> And IRIX XLV disk are not mountable due lack of XLV support in linux.

added to the FAQ - the same for the SGI partition support hint

thanks

t

-- 
thomas graichen <tgr@spoiled.org> ... perfection is reached, not
when there is no longer anything to add, but when there is no
longer anything to take away. --- antoine de saint-exupery

From owner-linux-xfs@oss.sgi.com Fri Mar 30 11:35:49 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2UJZnP07944
	for linux-xfs-outgoing; Fri, 30 Mar 2001 11:35:49 -0800
Received: from antares.cedar.buffalo.edu (antares.cedar.Buffalo.EDU [128.205.33.2])
	by oss.sgi.com (8.11.3/8.11.3) with SMTP id f2UJZmM07939
	for <linux-xfs@oss.sgi.com>; Fri, 30 Mar 2001 11:35:48 -0800
Received: (qmail 4777 invoked from network); 30 Mar 2001 19:35:43 -0000
Received: from zaurak.cedar.buffalo.edu (128.205.33.110)
  by antares.cedar.buffalo.edu with SMTP; 30 Mar 2001 19:35:43 -0000
Received: (from ajay@localhost)
	by zaurak.cedar.buffalo.edu (8.9.3+Sun/8.9.3) id OAA24603;
	Fri, 30 Mar 2001 14:35:42 -0500 (EST)
Date: Fri, 30 Mar 2001 14:35:42 -0500
From: Ajay Shekhawat <ajay@cedar.Buffalo.EDU>
To: cattelan@thebarn.com
Cc: linux-xfs@oss.sgi.com
Subject: Re: Prerelease 0.10.1 images now available.
Message-ID: <20010330143542.G16131@zaurak.cedar.buffalo.edu>
References: <ume4rwbbux7.wl@lupo.thebarn.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.2.5i
In-Reply-To: <ume4rwbbux7.wl@lupo.thebarn.com>; from cattelan@thebarn.com on Fri, Mar 30, 2001 at 01:03:16AM -0600
Organization: Center for Document Analysis and Recognition
X-OfficePhone: +1 (716)-645-6164 ext. 101
X-Fax-Number: +1 (716)-645-6176
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

On Fri, Mar 30, 2001 at 01:03:16AM -0600, cattelan@thebarn.com wrote:
> Seems like it was just yesterday that that 0.10 images came out...
> ohhh it was. :-)
> 
> Well several bugs have been fixed over the past few day so
> guess it is time to update things.
  . . . .

I've successfully patched the 2.4.3 kernel with these patches (almost everything
worked, except for a few minor  rejects).

So far, it seems to be OK. NFS serving of XFS works, and I'm beating on it
heavily.

The reason for the switch: I'd heard that the 3C905B ethernet card is better
supported in 2.4.3. I've been having problems with this card under a heavy
load; I just thought I'd try out the new kernel and see.

Ajay


From owner-linux-xfs@oss.sgi.com Fri Mar 30 11:45:54 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2UJjsA08130
	for linux-xfs-outgoing; Fri, 30 Mar 2001 11:45:54 -0800
Received: from chaos.egr.duke.edu (IDENT:root@chaos.egr.duke.edu [152.3.195.82])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2UJjrM08127
	for <linux-xfs@oss.sgi.com>; Fri, 30 Mar 2001 11:45:53 -0800
Received: from localhost (jlb@localhost)
	by chaos.egr.duke.edu (8.9.3/8.9.3) with ESMTP id OAA15928;
	Fri, 30 Mar 2001 14:45:42 -0500
X-Authentication-Warning: chaos.egr.duke.edu: jlb owned process doing -bs
Date: Fri, 30 Mar 2001 14:45:42 -0500 (EST)
From: Joshua Baker-LePain <jlb17@duke.edu>
X-Sender:  <jlb@chaos.egr.duke.edu>
To: Ajay Shekhawat <ajay@cedar.Buffalo.EDU>
cc: Linux xfs mailing list <linux-xfs@oss.sgi.com>
Subject: Re: Prerelease 0.10.1 images now available.
In-Reply-To: <20010330143542.G16131@zaurak.cedar.buffalo.edu>
Message-ID: <Pine.LNX.4.30.0103301443590.13729-100000@chaos.egr.duke.edu>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

On Fri, 30 Mar 2001 at 2:35pm, Ajay Shekhawat wrote

> I've successfully patched the 2.4.3 kernel with these patches (almost everything
> worked, except for a few minor  rejects).
>
> So far, it seems to be OK. NFS serving of XFS works, and I'm beating on it
> heavily.
>
> The reason for the switch: I'd heard that the 3C905B ethernet card is better
> supported in 2.4.3. I've been having problems with this card under a heavy
> load; I just thought I'd try out the new kernel and see.

Are you running on any SMP machines?  I had to add the 'noapic' option to
my server running the 2.4.2-xfs kernel, otherwise eth0 disappeared due to
lost interrupts after very little activity.

-- 
Joshua Baker-LePain
Department of Biomedical Engineering
Duke University


From owner-linux-xfs@oss.sgi.com Fri Mar 30 12:17:49 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2UKHnl08633
	for linux-xfs-outgoing; Fri, 30 Mar 2001 12:17:49 -0800
Received: from antares.cedar.buffalo.edu (antares.cedar.Buffalo.EDU [128.205.33.2])
	by oss.sgi.com (8.11.3/8.11.3) with SMTP id f2UKHmM08630
	for <linux-xfs@oss.sgi.com>; Fri, 30 Mar 2001 12:17:49 -0800
Received: (qmail 5114 invoked from network); 30 Mar 2001 20:17:48 -0000
Received: from zaurak.cedar.buffalo.edu (128.205.33.110)
  by antares.cedar.buffalo.edu with SMTP; 30 Mar 2001 20:17:48 -0000
Received: (from ajay@localhost)
	by zaurak.cedar.buffalo.edu (8.9.3+Sun/8.9.3) id PAA24613;
	Fri, 30 Mar 2001 15:17:47 -0500 (EST)
Date: Fri, 30 Mar 2001 15:17:47 -0500
From: Ajay Shekhawat <ajay@cedar.Buffalo.EDU>
To: Joshua Baker-LePain <jlb17@duke.edu>
Cc: Linux xfs mailing list <linux-xfs@oss.sgi.com>
Subject: Re: Prerelease 0.10.1 images now available.
Message-ID: <20010330151747.H16131@zaurak.cedar.buffalo.edu>
References: <20010330143542.G16131@zaurak.cedar.buffalo.edu> <Pine.LNX.4.30.0103301443590.13729-100000@chaos.egr.duke.edu>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.2.5i
In-Reply-To: <Pine.LNX.4.30.0103301443590.13729-100000@chaos.egr.duke.edu>; from jlb17@duke.edu on Fri, Mar 30, 2001 at 02:45:42PM -0500
Organization: Center for Document Analysis and Recognition
X-OfficePhone: +1 (716)-645-6164 ext. 101
X-Fax-Number: +1 (716)-645-6176
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

On Fri, Mar 30, 2001 at 02:45:42PM -0500, Joshua Baker-LePain wrote:
> On Fri, 30 Mar 2001 at 2:35pm, Ajay Shekhawat wrote
> 
> > I've successfully patched the 2.4.3 kernel with these patches (almost everything
> > worked, except for a few minor  rejects).
> >
> > So far, it seems to be OK. NFS serving of XFS works, and I'm beating on it
> > heavily.
> >
> > The reason for the switch: I'd heard that the 3C905B ethernet card is better
> > supported in 2.4.3. I've been having problems with this card under a heavy
> > load; I just thought I'd try out the new kernel and see.
> 
> Are you running on any SMP machines?  I had to add the 'noapic' option to
> my server running the 2.4.2-xfs kernel, otherwise eth0 disappeared due to
> lost interrupts after very little activity.

Yes, it is an SMP system, with dual P-II 450MHz CPUs on an ASUS P2B-DS
motherboard.  I didn't have to use the "noapic" option, but maybe I should 
have tried that with my 2.4.2-xfs when I encountered the problems with 
the ethernet card... Oh well, lets see how this fares.

Ajay


From owner-linux-xfs@oss.sgi.com Fri Mar 30 14:54:41 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2UMsfB11183
	for linux-xfs-outgoing; Fri, 30 Mar 2001 14:54:41 -0800
Received: from deliverator.sgi.com (deliverator.sgi.com [204.94.214.10])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2UMsfM11180
	for <linux-xfs@oss.sgi.com>; Fri, 30 Mar 2001 14:54:41 -0800
Received: from nodin.corp.sgi.com (nodin.corp.sgi.com [192.26.51.193]) by deliverator.sgi.com (980309.SGI.8.8.8-aspam-6.2/980310.SGI-aspam) via ESMTP id OAA26725
	for <linux-xfs@oss.sgi.com>; Fri, 30 Mar 2001 14:53:27 -0800 (PST)
	mail_from (cbryant@sgi.com)
Received: from la--amsa003e--n.losangeles.sgi.com (la--amsa003e--n.losangeles.sgi.com [169.238.108.17])
	by nodin.corp.sgi.com (8.11.2/8.11.2/nodin-1.0) with ESMTP id f2UMrik6779532
	for <linux-xfs@oss.sgi.com>; Fri, 30 Mar 2001 14:53:44 -0800 (PST)
Received: by la--amsa003e--n.losangeles.sgi.com with Internet Mail Service (5.5.2653.19)
	id <HJCKNXJ0>; Fri, 30 Mar 2001 14:53:13 -0800
Message-ID: <B514AE3D149DD411A0890004ACE8420D46BE62@la--amsa003e--n.losangeles.sgi.com>
From: Charlie  Bryant <cbryant@sgi.com>
To: "'linux-xfs@oss.sgi.com'" <linux-xfs@oss.sgi.com>
Subject: max xfs file sizes vs ext2?
Date: Fri, 30 Mar 2001 14:53:03 -0800
MIME-Version: 1.0
X-Mailer: Internet Mail Service (5.5.2653.19)
Content-Type: text/plain;
	charset="iso-8859-1"
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

perhaps this info exists but i'd like to boil things down in one place.

can someone provide data summarizing the maximum file size and file system
size using a ia-32 host running linux 2.4?

				ia-32 host	
				file system	
				ext2 				xfs
				----				----
max file size		2gb (lseek32)		?
max partition size	2tb (512b page size)	?

in advance, thanks!

From owner-linux-xfs@oss.sgi.com Fri Mar 30 15:47:48 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2UNlmN12359
	for linux-xfs-outgoing; Fri, 30 Mar 2001 15:47:48 -0800
Received: from sgi.com (sgi.SGI.COM [192.48.153.1])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2UNllM12356
	for <linux-xfs@oss.sgi.com>; Fri, 30 Mar 2001 15:47:47 -0800
Received: from ledzep.americas.sgi.com (relay.sgi.com [137.38.226.97] (may be forged)) 
	by sgi.com (980327.SGI.8.8.8-aspam/980304.SGI-aspam:
       SGI does not authorize the use of its proprietary
       systems or networks for unsolicited or bulk email
       from the Internet.) 
	via ESMTP id PAA06839
	for <linux-xfs@oss.sgi.com>; Fri, 30 Mar 2001 15:47:43 -0800 (PST)
	mail_from (cattelan@thebarn.com)
Received: from gibble.americas.sgi.com (gibble.americas.sgi.com [128.162.195.80]) by ledzep.americas.sgi.com (SGI-SGI-8.9.3/americas-smart-nospam1.1) with ESMTP id RAA49875; Fri, 30 Mar 2001 17:46:24 -0600 (CST)
Received: from thebarn.com (localhost [127.0.0.1])
	by gibble.americas.sgi.com (8.11.2/8.11.2) with ESMTP id f2UNYqb04941;
	Fri, 30 Mar 2001 18:34:52 -0500
Message-ID: <3AC51819.FF8A067B@thebarn.com>
Date: Fri, 30 Mar 2001 18:34:50 -0500
From: Russell Cattelan <cattelan@thebarn.com>
X-Mailer: Mozilla 4.76 [en] (X11; U; Linux 2.4.2-XFS i686)
X-Accept-Language: en
MIME-Version: 1.0
To: Charlie Bryant <cbryant@sgi.com>
CC: "'linux-xfs@oss.sgi.com'" <linux-xfs@oss.sgi.com>
Subject: Re: max xfs file sizes vs ext2?
References: <B514AE3D149DD411A0890004ACE8420D46BE62@la--amsa003e--n.losangeles.sgi.com>
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

Charlie Bryant wrote:

> perhaps this info exists but i'd like to boil things down in one place.
>
> can someone provide data summarizing the maximum file size and file system
> size using a ia-32 host running linux 2.4?
>
>                                 ia-32 host
>                                 file system
>                                 ext2                            xfs
>                                 ----                            ----
> max file size           2gb (lseek32)           ?

XFS and ext2 will have the same file size limitation.
namely 2^32 * PAGESIZE
an ia32 system with a 4k page
limit: 16384 gig.
16k page
limit; 65536 gig

>
> max partition size      2tb (512b page size)    ?

True for all FS on linux right now, it's a block device limit.



>
>
> in advance, thanks!

--
Russell Cattelan
--
Digital Elves inc. -- Currently on loan to SGI
Linux XFS core developer.




From owner-linux-xfs@oss.sgi.com Fri Mar 30 15:59:40 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2UNxeq12588
	for linux-xfs-outgoing; Fri, 30 Mar 2001 15:59:40 -0800
Received: from sgi.com (sgi.SGI.COM [192.48.153.1])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2UNxcM12585
	for <linux-xfs@oss.sgi.com>; Fri, 30 Mar 2001 15:59:38 -0800
Received: from ledzep.americas.sgi.com (relay.sgi.com [137.38.226.97] (may be forged)) 
	by sgi.com (980327.SGI.8.8.8-aspam/980304.SGI-aspam:
       SGI does not authorize the use of its proprietary
       systems or networks for unsolicited or bulk email
       from the Internet.) 
	via ESMTP id PAA06486
	for <linux-xfs@oss.sgi.com>; Fri, 30 Mar 2001 15:59:36 -0800 (PST)
	mail_from (cattelan@thebarn.com)
Received: from gibble.americas.sgi.com (gibble.americas.sgi.com [128.162.195.80]) by ledzep.americas.sgi.com (SGI-SGI-8.9.3/americas-smart-nospam1.1) with ESMTP id RAA39513; Fri, 30 Mar 2001 17:58:17 -0600 (CST)
Received: from thebarn.com (localhost [127.0.0.1])
	by gibble.americas.sgi.com (8.11.2/8.11.2) with ESMTP id f2UNkjb04979;
	Fri, 30 Mar 2001 18:46:45 -0500
Message-ID: <3AC51AE5.5E396D14@thebarn.com>
Date: Fri, 30 Mar 2001 18:46:45 -0500
From: Russell Cattelan <cattelan@thebarn.com>
X-Mailer: Mozilla 4.76 [en] (X11; U; Linux 2.4.2-XFS i686)
X-Accept-Language: en
MIME-Version: 1.0
To: Utz Lehmann <leh@tecosim.de>
CC: Eric Sandeen <sandeen@sgi.com>, linux-xfs@oss.sgi.com
Subject: Re: shutdown umount hangs
References: <20010329165519.A7386@tecosim.de> <20010329170446.B7386@tecosim.de> <3AC35E8D.711963C5@sgi.com> <20010329181419.G7386@tecosim.de> <20010329182044.H7386@tecosim.de> <3AC38727.FC9C4FC8@thebarn.com> <20010330115628.A1146@tecosim.de> <3AC4A50F.B141110B@thebarn.com> <20010330192006.A1149@tecosim.de>
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

Utz Lehmann wrote:

Well drat...  seems to be a really problem here.

Not sure we are going to be able to debug this remotely, hopefully we will be able
to reproduce the problem here.
But if you could extract a bit more info....

load the modules kdbm_pg and xfsidbg.
once the system is hung break into kdb backtrace
the unmount command, then grab the first address in
the pagebuf_iostart function and feed it to  kdb
kdb> pb <address>

We are looking for the pincount  and the lock status.
It would seem somehow the pg_delwri_flush is waiting for
the pagebuf to get unpinned, which should have already happened
before this point.
My guess is that somehow a pagebuf is getting re-inserted into the delwri queue
while the flush code in trying to work on pushing everything out.


> Russell Cattelan [cattelan@thebarn.com] wrote:
> > Utz Lehmann wrote:
> >
> > > Russell Cattelan [cattelan@thebarn.com] wrote:
> > >
> >
> > Ok sorry this was my fault I should have explained this better.
> > We need the back trace of the umount process.
> > to do this simply type ps from the kdb prompt  find the umount pid
> > the type
> > kdb> btp <umount pid>

>
> No problem:
>
> Unmounting file systems
> shmfs umounted
> /dev/sda2 umounted
> /dev/vg00/tmp umounted
> /dev/vg00/opt umounted
>
> Entering kdb (current=0xc035a000, pid 0) due to Keyboard Entry
> kdb> ps
> Task Addr  Pid      Parent   [*] cpu  State Thread     Command
> 0xc15fe000 00000001 00000000  0  000  stop  0xc15fe260 init
> 0xc15f0000 00000002 00000001  0  000  stop  0xc15f0260 keventd
> 0xc15ec000 00000003 00000001  0  000  stop  0xc15ec260 kswapd
> 0xc15ea000 00000004 00000001  0  000  stop  0xc15ea260 kreclaimd
> 0xc15e8000 00000005 00000001  0  000  stop  0xc15e8260 bdflush
> 0xc15e6000 00000006 00000001  0  000  stop  0xc15e6260 kupdate
> 0xc1580000 00000007 00000001  0  000  stop  0xc1580260 mdrecoveryd
> 0xc1562000 00000008 00000001  0  000  stop  0xc1562260 pagebuf_daemon
> 0xce0b8000 00001130 00000001  0  000  stop  0xce0b8260 rc
> 0xce268000 00001520 00001130  0  000  stop  0xce268260 S20reboot
> 0xcf83c000 00001541 00001520  0  000  stop  0xcf83c260 umount
> kdb> btp 1541
>     EBP       EIP         Function(args)
> 0xcf83de58 0xc0112b18 schedule+0x2d8
>                                kernel .text 0xc0100000 0xc0112840 0xc0112c70
>            0xc015fb13 pagebuf_iorequest+0x103 (0xcf6ca480)
>                                kernel .text 0xc0100000 0xc015fa10 0xc015fba0
>            0xc01c7c47 xfs_bdstrat_cb+0x27 (0xcf6ca480)
>                                kernel .text 0xc0100000 0xc01c7c20 0xc01c7c70
>            0xc0160763 pagebuf_delwri_flush+0xd3 (0xcf7b5ac0, 0x1, 0xcf83dec8)
>                                kernel .text 0xc0100000 0xc0160690 0xc0160880
>            0xc01c7d2d XFS_bflush+0x1d (0xcf7b5ac0, 0x3a01)
>                                kernel .text 0xc0100000 0xc01c7d10 0xc01c7d40
>            0xc01b6873 xfs_unmount+0xd3 (0xcf7aa400, 0x0, 0xc03b16a0)
>                                kernel .text 0xc0100000 0xc01b67a0 0xc01b6920
>            0xc01c189a fs_dounmount+0x5a (0xcf7aa400, 0x0, 0x0, 0xc03b16a0, 0xcf7b5dc4)
>                                kernel .text 0xc0100000 0xc01c1840 0xc01c18c0
>            0xc01c8d38 linvfs_put_super+0x58 (0xcf8eb800)
>                                kernel .text 0xc0100000 0xc01c8ce0 0xc01c8db0
>            0xc0136867 kill_super+0x87 (0xcf8eb800, 0x0, 0xc156cec0, 0xffffffff, 0xcfb29ac0)
>                                kernel .text 0xc0100000 0xc01367e0 0xc0136920
>            0xc0136c71 do_umount+0x1c1 (0xc156cec0, 0x0, 0x0)
>                                kernel .text 0xc0100000 0xc0136ab0 0xc0136c80
>            0xc0136d46 sys_umount+0xc6 (0x8052430, 0x0)
> more>
>                                kernel .text 0xc0100000 0xc0136c80 0xc0136d80
>            0xc0136d8c sys_oldumount+0xc (0x8052430, 0x804ee27, 0x8052478, 0x8052431, 0x804ee20)
>                                kernel .text 0xc0100000 0xc0136d80 0xc0136d90
>            0xc0108f77 system_call+0x33
>                                kernel .text 0xc0100000 0xc0108f44 0xc0108f7c
> kdb> reboot
>
> btw: 1 of ca. 7 tries the umount works.
>
> utz

--
Russell Cattelan
--
Digital Elves inc. -- Currently on loan to SGI
Linux XFS core developer.




From owner-linux-xfs@oss.sgi.com Fri Mar 30 16:21:52 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2V0LqW13015
	for linux-xfs-outgoing; Fri, 30 Mar 2001 16:21:52 -0800
Received: from protactinium (protactinium.btinternet.com [194.73.73.176])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2V0LpM13012
	for <linux-xfs@oss.sgi.com>; Fri, 30 Mar 2001 16:21:51 -0800
Received: from [213.1.203.252] (helo=pyewacket.nic.uklinux.net)
	by protactinium with esmtp (Exim 3.03 #83)
	id 14j99H-00012A-00
	for linux-xfs@oss.sgi.com; Sat, 31 Mar 2001 01:21:48 +0100
Received: from ndoye by pyewacket.nic.uklinux.net with local (Exim 3.22 #1)
	id 14j97P-0003ci-00
	for linux-xfs@oss.sgi.com; Sat, 31 Mar 2001 01:19:51 +0100
From: <nic@uklinux.net>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Message-ID: <15045.8871.297381.856149@localhost.localdomain>
Date: Sat, 31 Mar 2001 01:19:51 +0100
To: linux-xfs@oss.sgi.com
Subject: How to make an XFS compatible boot floppy?
X-Mailer: VM 6.90 under 21.2  (beta46) "Urania" XEmacs Lucid
Organisation: NISS
X-URL: http://www.nic.uklinux.net
X-Cabaret-Voltaire-URL: http://www.brainwashed.com/cv
X-Face:  &-8(x5K]<0/|dXSmxL9\p/,6*,C16]W8k1Elf.\e~pa~ASI57X9+eDm^Rkv'?}-bT=o]Hz{
 eMQXn
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

Hi all,

I realise that this isn't really completely xfs related, and as such I
should be burned in hell, but you all must have done it to be able to
repair your machines in dire emergencies. (It looks like a devfs
problem to me).

How can I make a boot floppy, just in case I kill my whole machine,
(or so that I can boot from floppy, delete my root filesystem and make
it an xfs partition)?

My machine is ia32 running RedHat 7.0.90 (Fisher) with devfs 1.3.11,
mkinitrd-3.0.6-1 and the lovely 2.4.2-XFS booked out from CVS a couple
of days ago. Yes, I am running devfsd as '/sbin/devfsd /dev'.

'/sbin/mkinitrd initrd-2.4.2-XFS.img 2.4.2-XFS' just hangs and I can't
get anywhere. A strace -f on that command shows it hangs with:
[pid 13931] open("/dev/loop/0", O_RDWR|O_LARGEFILE

I don't view myself as a linux (or unix for that matter) novice, or I
wouldn't have gone down this path, but I appear to have got into just
beyond my depth here, and was wondering if any of you kind folk could
either help me or point me in the correct direction for assistance.

Cheers,
nic
-- 
www.nic.uklinux.net

From owner-linux-xfs@oss.sgi.com Fri Mar 30 16:35:10 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2V0ZAF13227
	for linux-xfs-outgoing; Fri, 30 Mar 2001 16:35:10 -0800
Received: from astro.umn.edu (hal.astro.umn.edu [128.101.221.100])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2V0ZAM13224
	for <linux-xfs@oss.sgi.com>; Fri, 30 Mar 2001 16:35:10 -0800
Received: (qmail 16671 invoked by uid 835); 31 Mar 2001 00:35:04 -0000
Received: from localhost (sendmail-bs@127.0.0.1)
  by localhost with SMTP; 31 Mar 2001 00:35:04 -0000
Date: Fri, 30 Mar 2001 18:35:02 -0600 (CST)
From: kelley eicher <carde@astro.umn.edu>
To: nic@uklinux.net
cc: linux-xfs@oss.sgi.com
Subject: Re: How to make an XFS compatible boot floppy?
In-Reply-To: <15045.8871.297381.856149@localhost.localdomain>
Message-ID: <Pine.LNX.4.21.0103301833130.16410-100000@HAL.astro.umn.edu>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

i have made some boot/root/utility disks and have them available for
download off of my website. 

www.astro.umn.edu/~carde

follow the development link...

|<

ps: www.linuxdoc.org should have a HOWTO on making bootdisks

 >> kelley j eicher
<<  Sys Admin
 >> Univ. of MN Astronomy Dept.
<<  ph: (612) 626-2067 or (612) 624-3589
 >> fx: (612) 626-2029
<<  office: 385 physics
 >> carde@astro.umn.edu



From owner-linux-xfs@oss.sgi.com Fri Mar 30 16:53:08 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2V0r8W13508
	for linux-xfs-outgoing; Fri, 30 Mar 2001 16:53:08 -0800
Received: from roujin.gargoylecc.com ([65.100.85.34])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2V0r6M13503
	for <linux-xfs@oss.sgi.com>; Fri, 30 Mar 2001 16:53:07 -0800
Received: from roujin.gargoylecc.com (IDENT:ringram@roujin.gargoylecc.com [65.100.85.34])
	by roujin.gargoylecc.com (8.9.3/8.9.3) with ESMTP id JAA07481;
	Sat, 31 Mar 2001 09:07:52 -0700
Date: Sat, 31 Mar 2001 09:07:52 -0700 (MST)
From: Russel Ingram <ringram@gargoylecc.com>
To: nic@uklinux.net
cc: linux-xfs@oss.sgi.com
Subject: Re: How to make an XFS compatible boot floppy?
In-Reply-To: <15045.8871.297381.856149@localhost.localdomain>
Message-ID: <Pine.LNX.4.21.0103310900340.7472-100000@roujin.gargoylecc.com>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

On Sat, 31 Mar 2001 nic@uklinux.net wrote:

> Date: Sat, 31 Mar 2001 01:19:51 +0100
> From: nic@uklinux.net
> To: linux-xfs@oss.sgi.com
> Subject: How to make an XFS compatible boot floppy?
> 
> Hi all,
> 
> I realise that this isn't really completely xfs related, and as such I
> should be burned in hell, but you all must have done it to be able to
> repair your machines in dire emergencies. (It looks like a devfs
> problem to me).
> 
> How can I make a boot floppy, just in case I kill my whole machine,
> (or so that I can boot from floppy, delete my root filesystem and make
> it an xfs partition)?
> 

I've found that the easiest way to do this is just to use the rescue
fuctionality of the installation boot disk.  Just grab the boot.img from
the sgi download site and write it to a floppy just like you would if you
were gonna do a floppy boot installation.  Then when it gives you the
boot: prompt type in "linux rescue".  It will go through the language and
probably keyboard stuff and then drop you into a root prompt with a
ramdisk as the root filesystem.  From there you can mount your hard disk
filesystems and make whatever changes you need to to get the system workin
again.

Cheers,
Russ

-- 
---------------------------------------------------------------
"Bill Gates and Microsoft have ruined the computer industry for
 a long time to come by creating a class of ignorant and lazy
 computer users."  --Russel Ingram

"Mommy ... can I go out and ... KILL TONIGHT!?"
                   --Glen  Danzig, The Misfits
---------------------------------------------------------------
Russ Ingram
Gargoyle Computer Consulting
(307)742-1361
www.gargoylecc.com


From owner-linux-xfs@oss.sgi.com Fri Mar 30 17:18:12 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2V1ICo14092
	for linux-xfs-outgoing; Fri, 30 Mar 2001 17:18:12 -0800
Received: from mail15.jump.net (mail15.jump.net [206.196.91.15])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2V1IBM14089
	for <linux-xfs@oss.sgi.com>; Fri, 30 Mar 2001 17:18:11 -0800
Received: from sgi.com (aus-dsl-dhcp1-26.customer.jump.net [63.163.168.26])
	by mail15.jump.net (8.10.2/) with ESMTP id f2V1I8J12657;
	Fri, 30 Mar 2001 19:18:08 -0600 (CST)
Message-ID: <3AC53092.B75961FE@sgi.com>
Date: Fri, 30 Mar 2001 19:19:14 -0600
From: Eric Sandeen <sandeen@sgi.com>
X-Mailer: Mozilla 4.76 [en] (X11; U; Linux 2.4.2-XFS i686)
X-Accept-Language: en
MIME-Version: 1.0
To: nic@uklinux.net
CC: linux-xfs@oss.sgi.com
Subject: Re: How to make an XFS compatible boot floppy?
References: <15045.8871.297381.856149@localhost.localdomain>
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

nic@uklinux.net wrote:

<snip>

> How can I make a boot floppy, just in case I kill my whole machine,
> (or so that I can boot from floppy, delete my root filesystem and make
> it an xfs partition)?

<snip>
 
> '/sbin/mkinitrd initrd-2.4.2-XFS.img 2.4.2-XFS' just hangs and I can't
> get anywhere. A strace -f on that command shows it hangs with:
> [pid 13931] open("/dev/loop/0", O_RDWR|O_LARGEFILE

Odds are the reason for the hang is that mkinitrd requires the loopback
device, which is still broken in 2.4.2 (although patched & fixed in our
2.4.2-based xfs RPMS).  So, you weren't necessarily doing anything
wrong.  However, even if the loopback device worked, 2.4 + xfs is
getting big enough that fitting everything you need on a floppy is going
to be tight.

Russel's suggestion of using the rescue mode of one of our installer
images might be the easiest way to go at this point.

I haven't yet looked at Kelley's images, but thanks!  Maybe you can
tackle making the Linuxcare rescue CD xfs capable!  ;)

-Eric

From owner-linux-xfs@oss.sgi.com Fri Mar 30 17:19:27 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2V1JRD14129
	for linux-xfs-outgoing; Fri, 30 Mar 2001 17:19:27 -0800
Received: from deliverator.sgi.com (deliverator.sgi.com [204.94.214.10])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2V1JRM14126
	for <linux-xfs@oss.sgi.com>; Fri, 30 Mar 2001 17:19:27 -0800
Received: from zeus-fddi.americas.sgi.com (zeus-fddi.americas.sgi.com [128.162.8.103]) by deliverator.sgi.com (980309.SGI.8.8.8-aspam-6.2/980310.SGI-aspam) via ESMTP id RAA12815
	for <linux-xfs@oss.sgi.com>; Fri, 30 Mar 2001 17:18:12 -0800 (PST)
	mail_from (lord@sgi.com)
Received: from daisy-e185.americas.sgi.com (daisy.americas.sgi.com [128.162.185.214]) by zeus-fddi.americas.sgi.com (8.9.3/americas-smart-nospam1.1) with ESMTP id TAA1323342; Fri, 30 Mar 2001 19:18:09 -0600 (CST)
Received: from jen.americas.sgi.com (IDENT:root@jen.americas.sgi.com [128.162.184.86]) by daisy-e185.americas.sgi.com (SGI-8.9.3/SGI-server-1.7) with ESMTP id TAA35967; Fri, 30 Mar 2001 19:18:09 -0600 (CST)
Received: from jen.americas.sgi.com by jen.americas.sgi.com (8.11.0/SGI-client-1.7) via ESMTP id f2V1Gjg07741; Fri, 30 Mar 2001 19:16:45 -0600
Message-Id: <200103310116.f2V1Gjg07741@jen.americas.sgi.com>
X-Mailer: exmh version 2.2 06/23/2000 with nmh-1.0.4
To: Aaron Smith <aaron@mutex.org>
cc: linux-xfs@oss.sgi.com
Subject: Re: "user data" in transactions / logged writes 
In-Reply-To: Message from Aaron Smith <aaron@mutex.org> 
   of "Thu, 29 Mar 2001 17:55:26 PST." <20010329175526.B58595@gelatinous.com> 
Date: Fri, 30 Mar 2001 19:16:45 -0600
From: Steve Lord <lord@sgi.com>
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

> hello,
> 
> i've been reading over some of the xfs papers, and i noticed the following
> in "xFS project architecture, October 8, 1993":
> 
>    3.3 xFS Growth Path
> 
>    In the long run, we expect to add user-level transactions to the file
>    system, at which time selected user data will be logged along with the
>    filesystem metadata. This feature will make it practical for user
>    applications to easily implement databases and database-like systems on
>    top of this file system.

I think you have found a mention of something long since lost in the mists
of time here at SGI. No on is thinking about this at all so far as I know.

Steve

> 
> is this essentially logged writes? does xfs have any support for
> journalling written file data? please let me know if you know of anyone
> who's working on it, or if you understand what pitfalls would need to be
> addressed to make it possible.
> 
> thanks,
> aaron



From owner-linux-xfs@oss.sgi.com Fri Mar 30 17:53:06 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2V1r6x14894
	for linux-xfs-outgoing; Fri, 30 Mar 2001 17:53:06 -0800
Received: from roujin.gargoylecc.com ([65.100.85.34])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2V1r5M14891
	for <linux-xfs@oss.sgi.com>; Fri, 30 Mar 2001 17:53:05 -0800
Received: from roujin.gargoylecc.com (IDENT:ringram@roujin.gargoylecc.com [65.100.85.34])
	by roujin.gargoylecc.com (8.9.3/8.9.3) with ESMTP id KAA07552
	for <linux-xfs@oss.sgi.com>; Sat, 31 Mar 2001 10:08:08 -0700
Date: Sat, 31 Mar 2001 10:08:08 -0700 (MST)
From: Russel Ingram <ringram@gargoylecc.com>
To: linux-xfs@oss.sgi.com
Subject: Kernel update on CVS?
Message-ID: <Pine.LNX.4.21.0103311005590.7549-100000@roujin.gargoylecc.com>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

Has the base kernel been updated to 2.4.3 for the cvs tree yet?

-- 
---------------------------------------------------------------
"Bill Gates and Microsoft have ruined the computer industry for
 a long time to come by creating a class of ignorant and lazy
 computer users."  --Russel Ingram

"Mommy ... can I go out and ... KILL TONIGHT!?"
                   --Glen  Danzig, The Misfits
---------------------------------------------------------------
Russ Ingram
Gargoyle Computer Consulting
(307)742-1361
www.gargoylecc.com



From owner-linux-xfs@oss.sgi.com Fri Mar 30 18:03:05 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2V235B15029
	for linux-xfs-outgoing; Fri, 30 Mar 2001 18:03:05 -0800
Received: from pneumatic-tube.sgi.com (pneumatic-tube.sgi.com [204.94.214.22])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2V234M15026
	for <linux-xfs@oss.sgi.com>; Fri, 30 Mar 2001 18:03:04 -0800
Received: from zeus-fddi.americas.sgi.com (zeus-fddi.americas.sgi.com [128.162.8.103]) by pneumatic-tube.sgi.com (980327.SGI.8.8.8-aspam/980310.SGI-aspam) via ESMTP id SAA05433
	for <linux-xfs@oss.sgi.com>; Fri, 30 Mar 2001 18:13:11 -0800 (PST)
	mail_from (lord@sgi.com)
Received: from daisy-e185.americas.sgi.com (daisy.americas.sgi.com [128.162.185.214]) by zeus-fddi.americas.sgi.com (8.9.3/americas-smart-nospam1.1) with ESMTP id TAA1323515; Fri, 30 Mar 2001 19:59:13 -0600 (CST)
Received: from jen.americas.sgi.com (IDENT:root@jen.americas.sgi.com [128.162.184.86]) by daisy-e185.americas.sgi.com (SGI-8.9.3/SGI-server-1.7) with ESMTP id TAA62489; Fri, 30 Mar 2001 19:59:13 -0600 (CST)
Received: from jen.americas.sgi.com by jen.americas.sgi.com (8.11.0/SGI-client-1.7) via ESMTP id f2V1vnm07874; Fri, 30 Mar 2001 19:57:49 -0600
Message-Id: <200103310157.f2V1vnm07874@jen.americas.sgi.com>
To: Russel Ingram <ringram@gargoylecc.com>
cc: linux-xfs@oss.sgi.com
Subject: Re: Kernel update on CVS? 
References: <Pine.LNX.4.21.0103311005590.7549-100000@roujin.gargoylecc.com>
Comments: In-reply-to Russel Ingram <ringram@gargoylecc.com>
   message dated "Sat, 31 Mar 2001 10:08:08 -0700."
Date: Fri, 30 Mar 2001 19:57:49 -0600
From: Steve Lord <lord@sgi.com>
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

> Has the base kernel been updated to 2.4.3 for the cvs tree yet?
> 

Russell has done a merge, I am sitting in the 2.5 kernel summit and
just did the same thing. Should not be too long.

Steve


From owner-linux-xfs@oss.sgi.com Sat Mar 31 10:34:00 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2VIY0j28453
	for linux-xfs-outgoing; Sat, 31 Mar 2001 10:34:00 -0800
Received: from kernelpanix.aura.of.mankind (pD901EDC6.dip.t-dialin.net [217.1.237.198])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2VIXwM28448
	for <linux-xfs@oss.sgi.com>; Sat, 31 Mar 2001 10:33:58 -0800
Received: (from utz@localhost)
	by kernelpanix.aura.of.mankind (8.11.0/8.11.0) id f2VIrFI29384;
	Sat, 31 Mar 2001 20:53:15 +0200
X-Authentication-Warning: kernelpanix.aura.of.mankind: utz set sender to xfs@s2y4n2c.de using -f
Date: Sat, 31 Mar 2001 20:53:14 +0200
From: utz lehmann <xfs@s2y4n2c.de>
To: Russell Cattelan <cattelan@thebarn.com>
Cc: Utz Lehmann <leh@tecosim.de>, Eric Sandeen <sandeen@sgi.com>,
   linux-xfs@oss.sgi.com
Subject: Re: shutdown umount hangs
Message-ID: <20010331205314.A29369@s2y4n2c.de>
References: <20010329165519.A7386@tecosim.de> <20010329170446.B7386@tecosim.de> <3AC35E8D.711963C5@sgi.com> <20010329181419.G7386@tecosim.de> <20010329182044.H7386@tecosim.de> <3AC38727.FC9C4FC8@thebarn.com> <20010330115628.A1146@tecosim.de> <3AC4A50F.B141110B@thebarn.com> <20010330192006.A1149@tecosim.de> <3AC51AE5.5E396D14@thebarn.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.2.5i
In-Reply-To: <3AC51AE5.5E396D14@thebarn.com>
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

Hi

i will do this on monday.

utz

Russell Cattelan [cattelan@thebarn.com] wrote:
> Utz Lehmann wrote:
> 
> Well drat...  seems to be a really problem here.
> 
> Not sure we are going to be able to debug this remotely, hopefully we will be able
> to reproduce the problem here.
> But if you could extract a bit more info....
> 
> load the modules kdbm_pg and xfsidbg.
> once the system is hung break into kdb backtrace
> the unmount command, then grab the first address in
> the pagebuf_iostart function and feed it to  kdb
> kdb> pb <address>
> 
> We are looking for the pincount  and the lock status.
> It would seem somehow the pg_delwri_flush is waiting for
> the pagebuf to get unpinned, which should have already happened
> before this point.
> My guess is that somehow a pagebuf is getting re-inserted into the delwri queue
> while the flush code in trying to work on pushing everything out.

From owner-linux-xfs@oss.sgi.com Sat Mar 31 15:07:10 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f2VN7AL31986
	for linux-xfs-outgoing; Sat, 31 Mar 2001 15:07:10 -0800
Received: from thor.theseus.com (south.orl-pub.theseus.com [12.108.42.66])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f2VN79M31983
	for <linux-xfs@oss.sgi.com>; Sat, 31 Mar 2001 15:07:09 -0800
Received: from theseus.com (IDENT:thebs@fugitive.theseus.com [192.168.0.242])
	by thor.theseus.com (8.9.3/8.9.3) with ESMTP id SAA23214;
	Sat, 31 Mar 2001 18:12:21 -0500
Message-ID: <3AC66366.DFD31872@theseus.com>
Date: Sat, 31 Mar 2001 18:08:22 -0500
From: Bryan-TheBS-Smith <thebs@theseus.com>
Reply-To: thebs@theseus.com, b.j.smith@ieee.org
Organization: (Personal)
X-Mailer: Mozilla 4.76 [en] (X11; U; Linux 2.2.17-0.17-FUGITIVE i686)
X-Accept-Language: en
MIME-Version: 1.0
To: linux-xfs@oss.sgi.com
CC: elug-eluglist@elug.org
Subject: Steve Lord "'hanging with the man" (or men) of Linux ... Linux 2.5 
 Kernel Summit
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

Steve Lord "'hanging with the man" (or men) of Linux ... Linux 2.5
Kernel Summit

In case you didn't catch this on NewsForge:
   http://lwn.net/2001/features/KernelSummit/

Good to see Steve Lord pushing some of the XFS technologies into the
VFS space (see about half-way down the article).  If you've got any
insights on the summit, I'd love to hear them.

-- TheBS

P.S.  Question for you I _know_ you canNOT answer ;-P :  What is the
timeframe Linus "wishes" to see for 2.5?  I'm sure we'll have
feature creep once again (which should have had Linux releasing
"2.4" as "3.0" ;-), but it would be interesting to see what
everyone's view is at this point.  I'm assuming with all the changes
plan, this is 1-2 year jobbie in my view.

--
Bryan "TheBS" Smith          chat:thebs413 @AOL/MSN/Yahoo
Engineer      mailto:b.j.smith@ieee.org,thebs@theseus.com
*********************************************************
"Never apply a Star Trek solution to a Babylon 5 problem"
                                    -- Nicholas C. Weaver

From owner-linux-xfs@oss.sgi.com Sat Mar 31 16:19:03 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f310J3L00321
	for linux-xfs-outgoing; Sat, 31 Mar 2001 16:19:03 -0800
Received: from lips.borg.umn.edu (lips.borg.umn.edu [160.94.232.50])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f310J2M00318
	for <linux-xfs@oss.sgi.com>; Sat, 31 Mar 2001 16:19:02 -0800
Received: from thebarn.com (nic-25-c96-008.mn.mediaone.net [24.25.96.8])
	by lips.borg.umn.edu (8.12.0.Beta5/8.12.0.Beta5) with ESMTP id f310IviP015326;
	Sat, 31 Mar 2001 18:18:58 -0600 (CST)
Message-ID: <3AC6733F.3B62AC5E@thebarn.com>
Date: Sat, 31 Mar 2001 18:15:59 -0600
From: Russell Cattelan <cattelan@thebarn.com>
X-Mailer: Mozilla 4.75 [en] (X11; U; Linux 2.4.2-XFS i686)
X-Accept-Language: en
MIME-Version: 1.0
To: linux-xfs@oss.sgi.com
Subject: XFS 2.4.3 patch
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

A preliminary XFS patch for 2.4.3 has been placed on
oss.sgi.com/linux-xfs.sgi.com
ftp://linux-xfs.sgi.com/projects/xfs/download/patches/

This has passed basic bonnie and dbench runs, but has
not been extensivly tested, but due to the limited number
of changes things should work just fine.

The CVS tree will be updated in the next few days.


-Russell




From owner-linux-xfs@oss.sgi.com Sat Mar 31 16:23:57 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f310NvA00394
	for linux-xfs-outgoing; Sat, 31 Mar 2001 16:23:57 -0800
Received: from antares.cedar.buffalo.edu (antares.cedar.Buffalo.EDU [128.205.33.2])
	by oss.sgi.com (8.11.3/8.11.3) with SMTP id f310NuM00391
	for <linux-xfs@oss.sgi.com>; Sat, 31 Mar 2001 16:23:56 -0800
Received: (qmail 14043 invoked from network); 1 Apr 2001 00:23:55 -0000
Received: from zaurak.cedar.buffalo.edu (128.205.33.110)
  by antares.cedar.buffalo.edu with SMTP; 1 Apr 2001 00:23:55 -0000
Received: (from ajay@localhost)
	by zaurak.cedar.buffalo.edu (8.9.3+Sun/8.9.3) id TAA24767
	for linux-xfs@oss.sgi.com; Sat, 31 Mar 2001 19:23:55 -0500 (EST)
Date: Sat, 31 Mar 2001 19:23:55 -0500
From: Ajay Shekhawat <ajay@cedar.Buffalo.EDU>
To: Linux xfs mailing list <linux-xfs@oss.sgi.com>
Subject: Re: Prerelease 0.10.1 images now available.
Message-ID: <20010331192355.S16131@zaurak.cedar.buffalo.edu>
References: <20010330143542.G16131@zaurak.cedar.buffalo.edu> <Pine.LNX.4.30.0103301443590.13729-100000@chaos.egr.duke.edu> <20010330151747.H16131@zaurak.cedar.buffalo.edu>
Mime-Version: 1.0
Content-Type: multipart/mixed; boundary="LiQwW4YX+w4axhAx"
Content-Disposition: inline
User-Agent: Mutt/1.2.5i
In-Reply-To: <20010330151747.H16131@zaurak.cedar.buffalo.edu>; from ajay@cedar.Buffalo.EDU on Fri, Mar 30, 2001 at 03:17:47PM -0500
Organization: Center for Document Analysis and Recognition
X-OfficePhone: +1 (716)-645-6164 ext. 101
X-Fax-Number: +1 (716)-645-6176
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk


--LiQwW4YX+w4axhAx
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline

On Fri, Mar 30, 2001 at 03:17:47PM -0500, I wrote:
> Yes, it is an SMP system, with dual P-II 450MHz CPUs on an ASUS P2B-DS
> motherboard.  I didn't have to use the "noapic" option, but maybe I should 
> have tried that with my 2.4.2-xfs when I encountered the problems with 
> the ethernet card... Oh well, lets see how this fares.

The NFSD crashed after about 4 hours of relentless hammering. The ksymoops
outputs are attached below, in case anyone is interested. There were
several "oops"; one set from the KNFSD, and the other when I tried to
resurrect it.  
The silver lining is that the ethernet card didn't croak  :-}

Ajay


--LiQwW4YX+w4axhAx
Content-Type: text/plain; charset=us-ascii
Content-Disposition: attachment; filename="ksymoops.out.1"

ksymoops 2.4.0 on i686 2.4.3-XFS.  Options used
     -V (specified)
     -k /proc/ksyms (default)
     -l /proc/modules (default)
     -o /lib/modules/2.4.3-XFS/ (specified)
     -m /boot/System.map-2.4.3-XFS (specified)

Mar 30 18:40:55 mymachine kernel: Unable to handle kernel NULL pointer dereference at virtual address 00000010
Mar 30 18:40:55 mymachine kernel: c01712a0
Mar 30 18:40:55 mymachine kernel: *pde = 00000000
Mar 30 18:40:55 mymachine kernel: Oops: 0000
Mar 30 18:40:55 mymachine kernel: CPU:    0
Mar 30 18:40:55 mymachine kernel: EIP:    0010:[nfsd_cache_lookup+132/676]
Mar 30 18:40:55 mymachine kernel: EIP:    0010:[<c01712a0>]
Using defaults from ksymoops -t elf32-i386 -a i386
Mar 30 18:40:55 mymachine kernel: EFLAGS: 00010217
Mar 30 18:40:55 mymachine kernel: eax: 00000031   ebx: 00000000   ecx: 2f0d371e   edx: cf0be788
Mar 30 18:40:55 mymachine kernel: esi: c031b380   edi: cf028014   ebp: cf04e800   esp: cf027f38
Mar 30 18:40:55 mymachine kernel: ds: 0018   es: 0018   ss: 0018
Mar 30 18:40:55 mymachine kernel: Process nfsd (pid: 643, stackpage=cf027000)
Mar 30 18:40:55 mymachine kernel: Stack: cf04e800 c031b380 cf028014 cf04e800 00000009 00000002 00000011 2f0d371e 
Mar 30 18:40:55 mymachine kernel:        c016a774 cf04e800 00000002 cf04e938 c031b380 cf04e490 c02992e8 cf04e800 
Mar 30 18:40:55 mymachine kernel:        cf028014 cf026000 00172ed0 cf338ce0 cf026000 cf04e400 c031b700 00000000 
Mar 30 18:40:55 mymachine kernel: Call Trace: [nfsd_dispatch+60/360] [svc_process+684/1348] [nfsd+427/824] [kernel_thread+35/48] 
Mar 30 18:40:55 mymachine kernel: Call Trace: [<c016a774>] [<c02992e8>] [<c016a5ab>] [<c0105503>] 
Mar 30 18:40:55 mymachine kernel: Code: 80 7b 10 00 74 49 8b 4c 24 1c 3b 4b 24 75 40 8b 4c 24 10 3b 

>>EIP; c01712a0 <nfsd_cache_lookup+84/2a4>   <=====
Trace; c016a774 <nfsd_dispatch+3c/168>
Trace; c02992e8 <svc_process+2ac/544>
Trace; c016a5ab <nfsd+1ab/338>
Trace; c0105503 <kernel_thread+23/30>
Code;  c01712a0 <nfsd_cache_lookup+84/2a4>
00000000 <_EIP>:
Code;  c01712a0 <nfsd_cache_lookup+84/2a4>   <=====
   0:   80 7b 10 00               cmpb   $0x0,0x10(%ebx)   <=====
Code;  c01712a4 <nfsd_cache_lookup+88/2a4>
   4:   74 49                     je     4f <_EIP+0x4f> c01712ef <nfsd_cache_lookup+d3/2a4>
Code;  c01712a6 <nfsd_cache_lookup+8a/2a4>
   6:   8b 4c 24 1c               mov    0x1c(%esp,1),%ecx
Code;  c01712aa <nfsd_cache_lookup+8e/2a4>
   a:   3b 4b 24                  cmp    0x24(%ebx),%ecx
Code;  c01712ad <nfsd_cache_lookup+91/2a4>
   d:   75 40                     jne    4f <_EIP+0x4f> c01712ef <nfsd_cache_lookup+d3/2a4>
Code;  c01712af <nfsd_cache_lookup+93/2a4>
   f:   8b 4c 24 10               mov    0x10(%esp,1),%ecx
Code;  c01712b3 <nfsd_cache_lookup+97/2a4>
  13:   3b 00                     cmp    (%eax),%eax


--LiQwW4YX+w4axhAx
Content-Type: text/plain; charset=us-ascii
Content-Disposition: attachment; filename="ksymoops.out.3"

ksymoops 2.4.0 on i686 2.4.3-XFS.  Options used
     -V (specified)
     -k /proc/ksyms (default)
     -l /proc/modules (default)
     -o /lib/modules/2.4.3-XFS/ (specified)
     -m /boot/System.map-2.4.3-XFS (specified)

Mar 30 18:59:01 mymachine kernel: Unable to handle kernel paging request at virtual address 46454a7f
Mar 30 18:59:01 mymachine kernel: c011b0f0
Mar 30 18:59:01 mymachine kernel: *pde = 00000000
Mar 30 18:59:01 mymachine kernel: Oops: 0000
Mar 30 18:59:01 mymachine kernel: CPU:    0
Mar 30 18:59:01 mymachine kernel: EIP:    0010:[send_sig_info+56/176]
Mar 30 18:59:01 mymachine kernel: EIP:    0010:[<c011b0f0>]
Using defaults from ksymoops -t elf32-i386 -a i386
Mar 30 18:59:01 mymachine kernel: EFLAGS: 00010206
Mar 30 18:59:01 mymachine kernel: eax: 00000000   ebx: 46454443   ecx: 46454443   edx: 00000001
Mar 30 18:59:01 mymachine kernel: esi: 00000009   edi: 00000000   ebp: 00000000   esp: cd593f0c
Mar 30 18:59:01 mymachine kernel: ds: 0018   es: 0018   ss: 0018
Mar 30 18:59:01 mymachine kernel: Process rpc.nfsd (pid: 1102, stackpage=cd593000)
Mar 30 18:59:01 mymachine kernel: Stack: 37363534 fffffff1 00000000 00000000 c011b43d 00000009 00000001 46454443 
Mar 30 18:59:01 mymachine kernel:        c016a3c9 00000009 46454443 00000001 0000000c ffffffea 00000801 c016aa71 
Mar 30 18:59:01 mymachine kernel:        00000801 00000001 cd592000 bffff2cc 0000070b bffff0d8 00000001 cbabe000 
Mar 30 18:59:01 mymachine kernel: Call Trace: [send_sig+29/36] [nfsd_svc+249/304] [sys_nfsservctl+373/1060] [sys_munmap+49/76] [error_code+52/60] [system_call+51/56] 
Mar 30 18:59:01 mymachine kernel: Call Trace: [<c011b43d>] [<c016a3c9>] [<c016aa71>] [<c012162d>] [<c0107120>] [<c0106feb>] 
Mar 30 18:59:01 mymachine kernel: Code: 83 bb 3c 06 00 00 00 74 67 9c 5f fa f0 fe 8b 38 06 00 00 0f 

>>EIP; c011b0f0 <send_sig_info+38/b0>   <=====
Trace; c011b43d <send_sig+1d/24>
Trace; c016a3c9 <nfsd_svc+f9/130>
Trace; c016aa71 <sys_nfsservctl+175/424>
Trace; c012162d <sys_munmap+31/4c>
Trace; c0107120 <error_code+34/3c>
Trace; c0106feb <system_call+33/38>
Code;  c011b0f0 <send_sig_info+38/b0>
00000000 <_EIP>:
Code;  c011b0f0 <send_sig_info+38/b0>   <=====
   0:   83 bb 3c 06 00 00 00      cmpl   $0x0,0x63c(%ebx)   <=====
Code;  c011b0f7 <send_sig_info+3f/b0>
   7:   74 67                     je     70 <_EIP+0x70> c011b160 <send_sig_info+a8/b0>
Code;  c011b0f9 <send_sig_info+41/b0>
   9:   9c                        pushf  
Code;  c011b0fa <send_sig_info+42/b0>
   a:   5f                        pop    %edi
Code;  c011b0fb <send_sig_info+43/b0>
   b:   fa                        cli    
Code;  c011b0fc <send_sig_info+44/b0>
   c:   f0 fe 8b 38 06 00 00      lock decb 0x638(%ebx)
Code;  c011b103 <send_sig_info+4b/b0>
  13:   0f 00 00                  sldt   (%eax)


--LiQwW4YX+w4axhAx--

From owner-linux-xfs@oss.sgi.com Sat Mar 31 22:16:23 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f316GNo04629
	for linux-xfs-outgoing; Sat, 31 Mar 2001 22:16:23 -0800
Received: from groucho.maths.monash.edu.au (groucho.maths.monash.edu.au [130.194.160.211])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f316GMM04626
	for <linux-xfs@oss.sgi.com>; Sat, 31 Mar 2001 22:16:22 -0800
Received: (from rjh@localhost)
	by groucho.maths.monash.edu.au (8.8.8/8.8.8) id GAA28360
	for linux-xfs@oss.sgi.com; Sun, 1 Apr 2001 06:16:15 GMT
From: Robin Humble <rjh@groucho.maths.monash.edu.au>
Message-Id: <200104010616.GAA28360@groucho.maths.monash.edu.au>
Subject: Re: XFS 2.4.3 patch
To: linux-xfs@oss.sgi.com
Date: Sun, 1 Apr 2001 16:16:14 +1000 (EST)
In-Reply-To: <3AC6733F.3B62AC5E@thebarn.com> from "Russell Cattelan" at Mar 31, 2001 06:15:59 PM
X-Mailer: ELM [version 2.5 PL1]
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk


I downloaded the xfs and xfs-core patches and applied them both to a
straight linux-2.4.3.tar.bz  - it all builds just fine except for the
loop module:

make[2]: Entering directory
`/usr/src/linux-2.4.3-xfs-mar31/drivers/block'
gcc -D__KERNEL__ -I/usr/src/linux-2.4.3-xfs-mar31/include  -Wall
-Wstrict-prototypes -O2 -fno-strict-aliasing -fomit-frame-pointer -pipe
-march=i686 -malign-functions=4  -DMODULE -DMODVERSIONS -include
/usr/src/linux-2.4.3-xfs-mar31/include/linux/modversions.h
-DEXPORT_SYMTAB -c loop.c
loop.c: In function `loop_make_request':
loop.c:476: too few arguments to function
`generic_make_request_Ra58ea7b0'
loop.c: In function `loop_init':
loop.c:983: warning: passing arg 2 of `blk_queue_make_request_Rdfd55e90'
from incompatible pointer type
make[2]: *** [loop.o] Error 1
make[2]: Leaving directory
`/usr/src/linux-2.4.3-xfs-mar31/drivers/block'
make[1]: *** [_modsubdir_block] Error 2
make[1]: Leaving directory `/usr/src/linux-2.4.3-xfs-mar31/drivers'
make: *** [_mod_drivers] Error 2
[root@prego linux-2.4.3-xfs-mar31]# 

Am I doing something wrong?
stock 2.4.3 builds just fine except for a v4l buz module which I then
de-selected.

cheers,
robin

From owner-linux-xfs@oss.sgi.com Sat Mar 31 22:24:30 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f316OUZ04685
	for linux-xfs-outgoing; Sat, 31 Mar 2001 22:24:30 -0800
Received: from groucho.maths.monash.edu.au (groucho.maths.monash.edu.au [130.194.160.211])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f316OTM04682
	for <linux-xfs@oss.sgi.com>; Sat, 31 Mar 2001 22:24:29 -0800
Received: (from rjh@localhost)
	by groucho.maths.monash.edu.au (8.8.8/8.8.8) id GAA28512
	for linux-xfs@oss.sgi.com; Sun, 1 Apr 2001 06:24:25 GMT
From: Robin Humble <rjh@groucho.maths.monash.edu.au>
Message-Id: <200104010624.GAA28512@groucho.maths.monash.edu.au>
Subject: Re: XFS 2.4.3 patch
To: linux-xfs@oss.sgi.com
Date: Sun, 1 Apr 2001 16:24:25 +1000 (EST)
In-Reply-To: <3AC6733F.3B62AC5E@thebarn.com> from "Russell Cattelan" at Mar 31, 2001 06:15:59 PM
X-Mailer: ELM [version 2.5 PL1]
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk


I forgot to mention (oops, too many vodkas) that the primary reason
for upgrading to 2.4.3 is to get a working loop module - so this is
kinda important... Strange that it's the only broken thing - I did a
make -k modules, and the loop was the only thing that barfed :-/

Secondary:
Working 3c905b is the next most important thing... is this broken in
2.4.2 (as reported on this list) and fixed in 2.4.3 do you think?
We're getting problems with dual 3c905b's... I thought it was IRQ
issues, but we tried workarounds for that - maybe it's the 2.4.2 kernel?

cheers,
robin

From owner-linux-xfs@oss.sgi.com Sat Mar 31 22:29:22 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f316TMn04736
	for linux-xfs-outgoing; Sat, 31 Mar 2001 22:29:22 -0800
Received: from virtualhost.dk (ns.virtualhost.dk [195.184.98.160])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f316TLM04733
	for <linux-xfs@oss.sgi.com>; Sat, 31 Mar 2001 22:29:21 -0800
Received: from burns.home.kernel.dk ([192.168.0.2] ident=root)
	by virtualhost.dk with esmtp (Exim 3.22 #4)
	id 14jbLn-0006en-00; Sun, 01 Apr 2001 08:28:35 +0200
Received: from axboe by burns.home.kernel.dk with local (Exim 3.13 #1 (Debian))
	id 14jbLl-0006jp-00; Sun, 01 Apr 2001 08:28:33 +0200
Date: Sun, 1 Apr 2001 08:28:33 +0200
From: Jens Axboe <axboe@suse.de>
To: Robin Humble <rjh@groucho.maths.monash.edu.au>
Cc: linux-xfs@oss.sgi.com
Subject: Re: XFS 2.4.3 patch
Message-ID: <20010401082833.C25356@suse.de>
References: <3AC6733F.3B62AC5E@thebarn.com> <200104010616.GAA28360@groucho.maths.monash.edu.au>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <200104010616.GAA28360@groucho.maths.monash.edu.au>; from rjh@groucho.maths.monash.edu.au on Sun, Apr 01, 2001 at 04:16:14PM +1000
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

On Sun, Apr 01 2001, Robin Humble wrote:
> 
> I downloaded the xfs and xfs-core patches and applied them both to a
> straight linux-2.4.3.tar.bz  - it all builds just fine except for the
> loop module:

The 2.4.3 patch probably undid the make_request_fn changes to loop
or something like that, peek at drivers/block/rd.c to see how to
change the prototype. I'd do it, but I'm not at my machines atm.

-- 
Jens Axboe


From owner-linux-xfs@oss.sgi.com Sat Mar 31 23:06:17 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f3176Hg05100
	for linux-xfs-outgoing; Sat, 31 Mar 2001 23:06:17 -0800
Received: from lips.borg.umn.edu (lips.borg.umn.edu [160.94.232.50])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f3176GM05097
	for <linux-xfs@oss.sgi.com>; Sat, 31 Mar 2001 23:06:16 -0800
Received: from thebarn.com (nic-25-c96-008.mn.mediaone.net [24.25.96.8])
	by lips.borg.umn.edu (8.12.0.Beta5/8.12.0.Beta5) with ESMTP id f3175RiP016867;
	Sun, 1 Apr 2001 01:05:27 -0600 (CST)
Message-ID: <3AC6D32C.122FF4BE@thebarn.com>
Date: Sun, 01 Apr 2001 01:05:16 -0600
From: Russell Cattelan <cattelan@thebarn.com>
X-Mailer: Mozilla 4.74 [en] (X11; U; Linux 2.2.12 i386)
X-Accept-Language: en
MIME-Version: 1.0
To: Robin Humble <rjh@groucho.maths.monash.edu.au>
CC: linux-xfs@oss.sgi.com
Subject: Re: XFS 2.4.3 patch
References: <200104010616.GAA28360@groucho.maths.monash.edu.au>
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

Robin Humble wrote:

> I downloaded the xfs and xfs-core patches and applied them both to a
> straight linux-2.4.3.tar.bz  - it all builds just fine except for the
> loop module:

Oppps my fault I probably forgot to add loop.c to patch gen script.

Try this is should fix the build. I'll redo the patch it the morning.



--- drivers/block/loop.c        Sat Mar 31 01:19:30 2001
+++ /export/extra/lxfs-cvs/linux-2.4.3/linux/drivers/block/loop.c       Tue
Mar  6 21:35:36 2001
@@ -473,7 +473,7 @@
                        goto err;
        }

-       generic_make_request(rw, bh,NULL,0,0,0);
+       generic_make_request(rw, bh);
        return 0;

 err:



>
>
> make[2]: Entering directory
> `/usr/src/linux-2.4.3-xfs-mar31/drivers/block'
> gcc -D__KERNEL__ -I/usr/src/linux-2.4.3-xfs-mar31/include  -Wall
> -Wstrict-prototypes -O2 -fno-strict-aliasing -fomit-frame-pointer -pipe
> -march=i686 -malign-functions=4  -DMODULE -DMODVERSIONS -include
> /usr/src/linux-2.4.3-xfs-mar31/include/linux/modversions.h
> -DEXPORT_SYMTAB -c loop.c
> loop.c: In function `loop_make_request':
> loop.c:476: too few arguments to function
> `generic_make_request_Ra58ea7b0'
> loop.c: In function `loop_init':
> loop.c:983: warning: passing arg 2 of `blk_queue_make_request_Rdfd55e90'
> from incompatible pointer type
> make[2]: *** [loop.o] Error 1
> make[2]: Leaving directory
> `/usr/src/linux-2.4.3-xfs-mar31/drivers/block'
> make[1]: *** [_modsubdir_block] Error 2
> make[1]: Leaving directory `/usr/src/linux-2.4.3-xfs-mar31/drivers'
> make: *** [_mod_drivers] Error 2
> [root@prego linux-2.4.3-xfs-mar31]#
>
> Am I doing something wrong?
> stock 2.4.3 builds just fine except for a v4l buz module which I then
> de-selected.
>
> cheers,
> robin

--
Russell Cattelan
cattelan@thebarn.com




From owner-linux-xfs@oss.sgi.com Sat Mar 31 23:09:57 2001
Received: (from majordomo@localhost)
	by oss.sgi.com (8.11.3/8.11.3) id f3179vT05129
	for linux-xfs-outgoing; Sat, 31 Mar 2001 23:09:57 -0800
Received: from virtualhost.dk (ns.virtualhost.dk [195.184.98.160])
	by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f3179tM05126
	for <linux-xfs@oss.sgi.com>; Sat, 31 Mar 2001 23:09:56 -0800
Received: from burns.home.kernel.dk ([192.168.0.2] ident=root)
	by virtualhost.dk with esmtp (Exim 3.22 #4)
	id 14jbza-0006nT-00; Sun, 01 Apr 2001 09:09:42 +0200
Received: from axboe by burns.home.kernel.dk with local (Exim 3.13 #1 (Debian))
	id 14jbzX-0006oL-00; Sun, 01 Apr 2001 09:09:39 +0200
Date: Sun, 1 Apr 2001 09:09:39 +0200
From: Jens Axboe <axboe@suse.de>
To: Russell Cattelan <cattelan@thebarn.com>
Cc: Robin Humble <rjh@groucho.maths.monash.edu.au>, linux-xfs@oss.sgi.com
Subject: Re: XFS 2.4.3 patch
Message-ID: <20010401090939.B26093@suse.de>
References: <200104010616.GAA28360@groucho.maths.monash.edu.au> <3AC6D32C.122FF4BE@thebarn.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <3AC6D32C.122FF4BE@thebarn.com>; from cattelan@thebarn.com on Sun, Apr 01, 2001 at 01:05:16AM -0600
Sender: owner-linux-xfs@oss.sgi.com
Precedence: bulk

On Sun, Apr 01 2001, Russell Cattelan wrote:
> Robin Humble wrote:
> 
> > I downloaded the xfs and xfs-core patches and applied them both to a
> > straight linux-2.4.3.tar.bz  - it all builds just fine except for the
> > loop module:
> 
> Oppps my fault I probably forgot to add loop.c to patch gen script.
> 
> Try this is should fix the build. I'll redo the patch it the morning.

It would not appear to be enough:

> > loop.c: In function `loop_init':
> > loop.c:983: warning: passing arg 2 of `blk_queue_make_request_Rdfd55e90'

I'll bet you that you also didn't change the loop_make_request prototype.

-- 
Jens Axboe