From owner-devfs@oss.sgi.com Wed Mar 1 00:22:13 2000 Received: by oss.sgi.com id ; Wed, 1 Mar 2000 00:22:04 -0800 Received: from mobilix.atnf.CSIRO.AU ([130.155.191.29]:4480 "HELO mobilix.atnf.CSIRO.AU") by oss.sgi.com with SMTP id ; Wed, 1 Mar 2000 00:21:41 -0800 Received: (from rgooch@localhost) by mobilix.atnf.CSIRO.AU (8.6.12/8.6.12) id TAA00536; Wed, 1 Mar 2000 19:21:08 +1100 Date: Wed, 1 Mar 2000 19:21:08 +1100 Message-Id: <200003010821.TAA00536@mobilix.atnf.CSIRO.AU> From: Richard Gooch To: Matthew Vanecek Cc: devfs@oss.sgi.com, Douglas Gilbert Subject: Re: devfs + xinit Authentication error In-Reply-To: <38BCA8B6.149CA8B@directlink.net> References: <38BCA8B6.149CA8B@directlink.net> Sender: owner-devfs@oss.sgi.com Precedence: bulk Return-Path: X-Orcpt: rfc822;devfs-outgoing Matthew Vanecek writes: > Douglas Gilbert wrote: > > > > I'm trying to use devfs (with the default configuration of devfsd) on > > a slightly modified RH 6.0 i386 platform. When I call startx it fails > > with an authentication error in xinit. [Works with the classic /dev .] > > Any guidance? > > > > Doug Gilbert > > > I'm having the exact same problem, on 2.3.48. It complains that I might > not own the console. /dev/tty0 exists, and is a link to /dev/vc/0. The > permissions on it are a little weird: > > lr-xr-xr-x 1 root root 0 Feb 29 15:50 /dev/tty0 -> vc/0 Nothing weird about it. Symlink permissions basically don't matter. > I've tried changing them with a PERMISSIONS string in devfsd.conf and > directly, but of course chmod does not work on links--only on the file > that the link points to. And evidently, X thinks that it cannot write > to tty0, because tty0 (and *all* the symlinks, every last one of them > under /dev) do not have the write bit set. Admittedly, this is not a > problem on such devices as /dev/cdrom, or other read-only media, but > when you have apps with the old names hard-coded (like X), and want to > make symlinks to satisfy those hard-codings, but the symlinks won't let > you make them lrwxrwxrwx, you can get rather frustrated. > > root@reliant:/dev$ ls -l tty0 > lr-xr-xr-x 1 root root 0 Feb 29 15:50 tty0 -> vc/0 > root@reliant:/dev$ rm -f tty0 > root@reliant:/dev$ ls -l tty0 > ls: tty0: No such file or directory > root@reliant:/dev$ ln -sf vc/0 tty0 > root@reliant:/dev$ ls -l tty0 > lr-xr-xr-x 1 root root 0 Feb 29 23:07 tty0 -> vc/0 > > This would seem to be a fairly major failing in either devfs or > devfsd. Is it a known bug, or really a bug? How do we fix it? It > doesn't seem to be documented anywhere. I've tried playing with the > umask, and deleting/recreating symlinks, but nothing seems to work. > Is there some undocumented configuration option that I should know > about? The bug is in X. It should use stat(2) or access(2) and not lstat(2) to figure permissions. Assuming that's what X is doing. The bug isn't in devfs/devfsd. > TIA for answers. devfs has gone through some extreme changes since > 2.2.x. Once upon a time, it was fairly effortless and bug-free, and > compiling the patch into the kernel was really all that was > required. Now, it seems needlessly complicated. For systems > developers, I guess that's not an issue--as they like to spend > extraordinary amounts of time in making a system work, as opposed to > the rest of us who just want to use the system. Blame Linus for the name changes. In time, he expects distributions to see the merit in the new names and move over to devfs wholesale. Regards, Richard.... Permanent: rgooch@atnf.csiro.au Current: rgooch@ras.ucalgary.ca From owner-devfs@oss.sgi.com Wed Mar 1 05:02:00 2000 Received: by oss.sgi.com id ; Wed, 1 Mar 2000 05:01:50 -0800 Received: from smtp.interlog.com ([207.34.202.37]:62479 "EHLO smtp.interlog.com") by oss.sgi.com with ESMTP id ; Wed, 1 Mar 2000 05:01:35 -0800 Received: from interlog.com (ip213-201.cc.interlog.com [207.34.213.201]) by smtp.interlog.com (8.9.1/8.9.1) with ESMTP id IAA26414; Wed, 1 Mar 2000 08:01:20 -0500 (EST) Message-ID: <38BD152A.59B4F13B@interlog.com> Date: Wed, 01 Mar 2000 08:03:38 -0500 From: Douglas Gilbert X-Mailer: Mozilla 4.72 [en] (X11; U; Linux 2.3.48 i586) X-Accept-Language: en MIME-Version: 1.0 To: Richard Gooch CC: Matthew Vanecek , devfs@oss.sgi.com, linux-kernel@vger.rutgers.edu Subject: Re: devfs + xinit Authentication error References: <38BCA8B6.149CA8B@directlink.net> <200003010821.TAA00536@mobilix.atnf.CSIRO.AU> Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Sender: owner-devfs@oss.sgi.com Precedence: bulk Return-Path: X-Orcpt: rfc822;devfs-outgoing Richard Gooch wrote: > > Matthew Vanecek writes: > > Douglas Gilbert wrote: > > > > > > I'm trying to use devfs (with the default configuration of devfsd) on > > > a slightly modified RH 6.0 i386 platform. When I call startx it fails > > > with an authentication error in xinit. [Works with the classic /dev .] > > > Any guidance? > > > > > > Doug Gilbert > > > > > > I'm having the exact same problem, on 2.3.48. It complains that I might > > not own the console. The solution (at least on RH6.0 and RH6.1) for the X problem was to change the following line in /etc/security/console.perms from: =tty[0-9][0-9]* :[0-9]\.[0-9] :[0-9] to: =tty[0-9][0-9]* [0-9][0-9]* :[0-9]\.[0-9] :[0-9] There is a related problem which stops root logging in on a virtual console. The solution was to add the last 8 lines to /etc/securetty yielding: tty1 tty2 tty3 tty4 tty5 tty6 tty7 tty8 1 2 3 4 5 6 7 8 BTW Both 'who' and 'df' look different under the devfs regime. > [snip] > > Blame Linus for the name changes. In time, he expects distributions to > see the merit in the new names and move over to devfs wholesale. As Richard says, the interesting thing will be to see how the various distributions will jump with devfs. If that was known in advance of 2.4 being released then applications that scan for devices (e.g. cdparanoia, cdrecord and SANE) could be ready if there was to be a wholesale change. Doug Gilbert From owner-devfs@oss.sgi.com Wed Mar 1 05:45:30 2000 Received: by oss.sgi.com id ; Wed, 1 Mar 2000 05:45:20 -0800 Received: from t3-static3-206.adsl.directlink.net ([63.68.136.206]:34320 "HELO t3-static3-206.adsl.directlink.net") by oss.sgi.com with SMTP id ; Wed, 1 Mar 2000 05:45:16 -0800 Received: from directlink.net (reliant.home.pri [192.168.1.25]) by t3-static3-206.adsl.directlink.net (Postfix) with ESMTP id 8F11D35B6; Wed, 1 Mar 2000 07:45:03 -0600 (CST) Message-ID: <38BD1EDB.2152B331@directlink.net> Date: Wed, 01 Mar 2000 07:44:59 -0600 From: Matthew Vanecek Organization: University of North Texas X-Mailer: Mozilla 4.72 [en] (X11; U; Linux 2.3.48 i586) X-Accept-Language: en MIME-Version: 1.0 To: Richard Gooch Cc: devfs@oss.sgi.com, linux kernel list Subject: Re: devfs + xinit Authentication error References: <38BCA8B6.149CA8B@directlink.net> <200003010821.TAA00536@mobilix.atnf.CSIRO.AU> Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Sender: owner-devfs@oss.sgi.com Precedence: bulk Return-Path: X-Orcpt: rfc822;devfs-outgoing Richard Gooch wrote: > > Matthew Vanecek writes: > > Douglas Gilbert wrote: > > > > > > I'm trying to use devfs (with the default configuration of devfsd) on > > > a slightly modified RH 6.0 i386 platform. When I call startx it fails > > > with an authentication error in xinit. [Works with the classic /dev .] > > > Any guidance? > > > > > > Doug Gilbert > > > > > > I'm having the exact same problem, on 2.3.48. It complains that I might > > not own the console. /dev/tty0 exists, and is a link to /dev/vc/0. The > > permissions on it are a little weird: > > > > lr-xr-xr-x 1 root root 0 Feb 29 15:50 /dev/tty0 -> vc/0 > > Nothing weird about it. Symlink permissions basically don't matter. > > > I've tried changing them with a PERMISSIONS string in devfsd.conf and > > directly, but of course chmod does not work on links--only on the file > > that the link points to. And evidently, X thinks that it cannot write > > to tty0, because tty0 (and *all* the symlinks, every last one of them > > under /dev) do not have the write bit set. Admittedly, this is not a > > problem on such devices as /dev/cdrom, or other read-only media, but > > when you have apps with the old names hard-coded (like X), and want to > > make symlinks to satisfy those hard-codings, but the symlinks won't let > > you make them lrwxrwxrwx, you can get rather frustrated. > > > > root@reliant:/dev$ ls -l tty0 > > lr-xr-xr-x 1 root root 0 Feb 29 15:50 tty0 -> vc/0 > > root@reliant:/dev$ rm -f tty0 > > root@reliant:/dev$ ls -l tty0 > > ls: tty0: No such file or directory > > root@reliant:/dev$ ln -sf vc/0 tty0 > > root@reliant:/dev$ ls -l tty0 > > lr-xr-xr-x 1 root root 0 Feb 29 23:07 tty0 -> vc/0 > > > > This would seem to be a fairly major failing in either devfs or > > devfsd. Is it a known bug, or really a bug? How do we fix it? It > > doesn't seem to be documented anywhere. I've tried playing with the > > umask, and deleting/recreating symlinks, but nothing seems to work. > > Is there some undocumented configuration option that I should know > > about? > > The bug is in X. It should use stat(2) or access(2) and not lstat(2) > to figure permissions. Assuming that's what X is doing. The bug isn't > in devfs/devfsd. > Well, I'm not real sure *what* X is doing, but it evidently expects to find the w bit set, regardless if it's a link or not. vc/* is of course owned by the logged-in user, and set to rw-rw-rw-, so that's not a problem. However, I think for the short term (at the very least on my own box) it would be easier to adjust devfs to create symlinks as lrwxrwxrwx (instead of lr-xr-xr-x). Where in the code would I look to adjust that? Not looking to submit a patch--just want to get it to work here, until everyone else decides to use good coding practices. > > TIA for answers. devfs has gone through some extreme changes since > > 2.2.x. Once upon a time, it was fairly effortless and bug-free, and > > compiling the patch into the kernel was really all that was > > required. Now, it seems needlessly complicated. For systems > > developers, I guess that's not an issue--as they like to spend > > extraordinary amounts of time in making a system work, as opposed to > > the rest of us who just want to use the system. > > Blame Linus for the name changes. In time, he expects distributions to > see the merit in the new names and move over to devfs wholesale. > > Regards, > > Richard.... Linus: Can we have the old name system back? It was really easier to work with, and to me it was just as expressive. Or how about this--let's pick a naming convention and STICK WITH IT!! devfs is stable enough to be in the stable thread (where's the guy that was backporting to 2.2.x? I miss him!), and as such, it's naming convention needs to be solidified and stabilized. I want the old names back (by default--not by devfsd), but if that's not possible, how about some assurance that you're not going to change the 'standards' again? (Like he'll ever read this anyhow. Oh, well, that's MHO, anyhow). -- Matthew Vanecek Visit my Website at http://mysite.directlink.net/linuxguy For answers type: perl -e 'print $i=pack(c5,(41*2),sqrt(7056),(unpack(c,H)-2),oct(115),10);' ***************************************************************** For 93 million miles, there is nothing between the sun and my shadow except me. I'm always getting in the way of something... From owner-devfs@oss.sgi.com Wed Mar 1 16:04:31 2000 Received: by oss.sgi.com id ; Wed, 1 Mar 2000 16:04:21 -0800 Received: from hibernia.clubi.ie ([212.17.32.129]:36106 "EHLO hibernia.clubi.ie") by oss.sgi.com with ESMTP id ; Wed, 1 Mar 2000 16:04:15 -0800 Received: from localhost (paul@localhost) by hibernia.clubi.ie (8.9.3/8.9.1) with ESMTP id AAA27663; Thu, 2 Mar 2000 00:04:41 GMT X-Authentication-Warning: hibernia.spin.ie: paul owned process doing -bs Date: Thu, 2 Mar 2000 00:04:40 +0000 (GMT) From: Paul Jakma X-Sender: paul@hibernia.spin.ie To: Douglas Gilbert cc: Richard Gooch , Matthew Vanecek , devfs@oss.sgi.com, linux-kernel@vger.rutgers.edu Subject: Re: devfs + xinit Authentication error In-Reply-To: <38BD152A.59B4F13B@interlog.com> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-devfs@oss.sgi.com Precedence: bulk Return-Path: X-Orcpt: rfc822;devfs-outgoing On Wed, 1 Mar 2000, Douglas Gilbert wrote: The solution (at least on RH6.0 and RH6.1) for the X problem was to change the following line in /etc/security/console.perms from: =tty[0-9][0-9]* :[0-9]\.[0-9] :[0-9] to: =tty[0-9][0-9]* [0-9][0-9]* :[0-9]\.[0-9] :[0-9] There is a related problem which stops root logging in on a virtual console. The solution was to add the last 8 lines to /etc/securetty yielding: tty1 tty2 tty3 tty4 tty5 tty6 tty7 tty8 1 2 3 4 5 6 7 8 ick... this could cause /dev/pty/[1-8] to be deemed secure for root login. What's really needed is to fix the PAM securetty module. At the moment it won't parse full paths like /dev/vc/6 - which imo means pam securetty is broken. regards, -- Paul Jakma paul@clubi.ie PGP5 key: http://www.clubi.ie/jakma/publickey.txt ------------------------------------------- Fortune: A New York City ordinance prohibits the shooting of rabbits from the rear of a Third Avenue street car -- if the car is in motion. From owner-devfs@oss.sgi.com Wed Mar 1 23:05:12 2000 Received: by oss.sgi.com id ; Wed, 1 Mar 2000 23:05:03 -0800 Received: from firewall-in.sch57.msk.ru ([195.178.195.6]:26637 "EHLO dell.sch57.msk.ru") by oss.sgi.com with ESMTP id ; Wed, 1 Mar 2000 23:04:44 -0800 Received: from khim.UUCP (uucp@localhost) by dell.sch57.msk.ru (8.8.8/8.8.8) with UUCP id JAA15093; Thu, 2 Mar 2000 09:48:46 +0300 Received: by khim.sch57.msk.ru (dMail for DOS v2.07a2, 12Jun98); Thu, 2 Mar 2000 09:08:57 +0300 To: dgilbert@interlog.com, rgooch@atnf.csiro.au Cc: linuxguy@directlink.net, devfs@oss.sgi.com, linux-kernel@vger.rutgers.edu References: <38BD152A.59B4F13B@interlog.com> <38BCA8B6.149CA8B@directlink.net> <200003010821.TAA00536@mobilix.atnf.CSIRO.AU> Message-Id: Organization: MCCME From: "Khimenko Victor" Date: Thu, 2 Mar 2000 09:08:57 +0300 (MSK) X-Mailer: dMail [Demos Mail for DOS v2.07a2] Subject: Re: devfs + xinit Authentication error Lines: 63 MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Sender: owner-devfs@oss.sgi.com Precedence: bulk Return-Path: X-Orcpt: rfc822;devfs-outgoing In <38BD152A.59B4F13B@interlog.com> Douglas Gilbert (dgilbert@interlog.com) wrote: > Richard Gooch wrote: >> >> Matthew Vanecek writes: >> > Douglas Gilbert wrote: >> > > >> > > I'm trying to use devfs (with the default configuration of devfsd) on >> > > a slightly modified RH 6.0 i386 platform. When I call startx it fails >> > > with an authentication error in xinit. [Works with the classic /dev .] >> > > Any guidance? >> > > >> > > Doug Gilbert >> > >> > >> > I'm having the exact same problem, on 2.3.48. It complains that I might >> > not own the console. > The solution (at least on RH6.0 and RH6.1) for the X problem was > to change the following line in /etc/security/console.perms from: > =tty[0-9][0-9]* :[0-9]\.[0-9] :[0-9] > to: > =tty[0-9][0-9]* [0-9][0-9]* :[0-9]\.[0-9] :[0-9] > There is a related problem which stops root logging in on > a virtual console. The solution was to add the last 8 > lines to /etc/securetty yielding: > tty1 > tty2 > tty3 > tty4 > tty5 > tty6 > tty7 > tty8 > 1 > 2 > 3 > 4 > 5 > 6 > 7 > 8 Don't do this !!! You just added first few telnet connections to allowed root consoles !!! If you can not fix linux-utils (two line fix) then use su or sudo ... > BTW Both 'who' and 'df' look different under the devfs regime. >> [snip] >> >> Blame Linus for the name changes. In time, he expects distributions to >> see the merit in the new names and move over to devfs wholesale. > As Richard says, the interesting thing will be to see how the > various distributions will jump with devfs. If that was known > in advance of 2.4 being released then applications that scan > for devices (e.g. cdparanoia, cdrecord and SANE) could be ready > if there was to be a wholesale change. From owner-devfs@oss.sgi.com Wed Mar 1 23:06:03 2000 Received: by oss.sgi.com id ; Wed, 1 Mar 2000 23:05:53 -0800 Received: from firewall-in.sch57.msk.ru ([195.178.195.6]:28429 "EHLO dell.sch57.msk.ru") by oss.sgi.com with ESMTP id ; Wed, 1 Mar 2000 23:05:48 -0800 Received: from khim.UUCP (uucp@localhost) by dell.sch57.msk.ru (8.8.8/8.8.8) with UUCP id JAA15094; Thu, 2 Mar 2000 09:48:48 +0300 Received: by khim.sch57.msk.ru (dMail for DOS v2.07a2, 12Jun98); Thu, 2 Mar 2000 09:30:20 +0300 To: paul@clubi.ie, dgilbert@interlog.com Cc: rgooch@atnf.csiro.au, linuxguy@directlink.net, devfs@oss.sgi.com, linux-kernel@vger.rutgers.edu References: Message-Id: Organization: MCCME From: "Khimenko Victor" Date: Thu, 2 Mar 2000 09:30:20 +0300 (MSK) X-Mailer: dMail [Demos Mail for DOS v2.07a2] Subject: Re: devfs + xinit Authentication error Lines: 73 MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Sender: owner-devfs@oss.sgi.com Precedence: bulk Return-Path: X-Orcpt: rfc822;devfs-outgoing In Paul Jakma (paul@clubi.ie) wrote: > On Wed, 1 Mar 2000, Douglas Gilbert wrote: > The solution (at least on RH6.0 and RH6.1) for the X problem was > to change the following line in /etc/security/console.perms from: > =tty[0-9][0-9]* :[0-9]\.[0-9] :[0-9] > to: > =tty[0-9][0-9]* [0-9][0-9]* :[0-9]\.[0-9] :[0-9] > There is a related problem which stops root logging in on > a virtual console. The solution was to add the last 8 > lines to /etc/securetty yielding: > tty1 > tty2 > tty3 > tty4 > tty5 > tty6 > tty7 > tty8 > 1 > 2 > 3 > 4 > 5 > 6 > 7 > 8 > ick... this could cause /dev/pty/[1-8] to be deemed secure for root > login. Exactly. And it's NOT what you want. > What's really needed is to fix the PAM securetty module. At the moment it > won't parse full paths like /dev/vc/6 - which imo means pam securetty is > broken. Oh, yeah. Of course PAM is broken. We need to embed telepathy in PAM IMMEDEATELY ! Or you can fix PAM so it will use full name after THAT -- cut -- if ((tty = rindex(ttyn, '/'))) ++tty; else tty = ttyn; -- cut -- snippet from login.c ? PAM module was NEVER supplied with full device name to begin with. Add tiny patch to your login.c (from unix-utils) and stop blaming the innocent PAM. --- util-linux-2.9x/login-utils/login.c Sat Dec 11 17:00:45 1999 +++ util-linux-2.9x/login-utils/login.c Sat Dec 11 17:03:16 1999 @@ -515,10 +515,10 @@ tcsetattr(0,TCSAFLUSH,&tt); } - if ((tty = rindex(ttyn, '/'))) - ++tty; - else + if (strncmp(ttyn, "/dev/", 5)) tty = ttyn; + else + tty = ttyn+5; #if 0 /* other than iso-8859-1 */ From owner-devfs@oss.sgi.com Thu Mar 2 01:47:35 2000 Received: by oss.sgi.com id ; Thu, 2 Mar 2000 01:47:15 -0800 Received: from hibernia.clubi.ie ([212.17.32.129]:45579 "EHLO hibernia.clubi.ie") by oss.sgi.com with ESMTP id ; Thu, 2 Mar 2000 01:47:09 -0800 Received: from localhost (paul@localhost) by hibernia.clubi.ie (8.9.3/8.9.1) with ESMTP id JAA00500; Thu, 2 Mar 2000 09:47:27 GMT X-Authentication-Warning: hibernia.spin.ie: paul owned process doing -bs Date: Thu, 2 Mar 2000 09:47:27 +0000 (GMT) From: Paul Jakma X-Sender: paul@hibernia.spin.ie To: Khimenko Victor cc: dgilbert@interlog.com, rgooch@atnf.csiro.au, linuxguy@directlink.net, devfs@oss.sgi.com, linux-kernel@vger.rutgers.edu Subject: Re: devfs + xinit Authentication error In-Reply-To: Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-devfs@oss.sgi.com Precedence: bulk Return-Path: X-Orcpt: rfc822;devfs-outgoing On Thu, 2 Mar 2000, Khimenko Victor wrote: In Paul Jakma (paul@clubi.ie) wrote: > On Wed, 1 Mar 2000, Douglas Gilbert wrote: Exactly. And it's NOT what you want. > What's really needed is to fix the PAM securetty module. At the moment it > won't parse full paths like /dev/vc/6 - which imo means pam securetty is > broken. Oh, yeah. Of course PAM is broken. We need to embed telepathy in PAM IMMEDEATELY ! :) Or you can fix PAM so it will use full name after THAT -- cut -- if ((tty = rindex(ttyn, '/'))) ++tty; else tty = ttyn; -- cut -- snippet from login.c ? PAM module was NEVER supplied with full device name to begin with. Add tiny patch to your login.c (from unix-utils) and stop blaming the innocent PAM. sorry, i just presumed the brokenness was in PAM. thanks for pointing the correct source of the problem. You should that submit that patch to the utils-linux maintainer. regards, -- Paul Jakma paul@clubi.ie PGP5 key: http://www.clubi.ie/jakma/publickey.txt ------------------------------------------- Fortune: A national debt, if it is not excessive, will be to us a national blessing. -- Alexander Hamilton From owner-devfs@oss.sgi.com Thu Mar 2 15:53:07 2000 Received: by oss.sgi.com id ; Thu, 2 Mar 2000 15:52:48 -0800 Received: from hera.cwi.nl ([192.16.191.1]:58279 "EHLO hera.cwi.nl") by oss.sgi.com with ESMTP id ; Thu, 2 Mar 2000 15:52:27 -0800 Received: from arend.cwi.nl (arend.cwi.nl [192.16.184.162]) by hera.cwi.nl with ESMTP id AAA05687 for ; Fri, 3 Mar 2000 00:50:18 +0100 (MET) Received: by arend.cwi.nl id AAA03569; Fri, 3 Mar 2000 00:50:17 +0100 (MET) Date: Fri, 3 Mar 2000 00:50:17 +0100 (MET) From: Andries.Brouwer@cwi.nl Message-Id: To: dgilbert@interlog.com, khim@sch57.msk.ru, paul@clubi.ie Subject: Re: [util-linux] Re: devfs + xinit Authentication error Cc: devfs@oss.sgi.com, linux-kernel@vger.rutgers.edu, linuxguy@directlink.net, rgooch@atnf.csiro.au Sender: owner-devfs@oss.sgi.com Precedence: bulk Return-Path: X-Orcpt: rfc822;devfs-outgoing - if ((tty = rindex(ttyn, '/'))) - ++tty; - else + if (strncmp(ttyn, "/dev/", 5)) tty = ttyn; + else + tty = ttyn+5; Thanks! Applied Andries From owner-devfs@oss.sgi.com Thu Mar 2 20:43:17 2000 Received: by oss.sgi.com id ; Thu, 2 Mar 2000 20:42:58 -0800 Received: from tinylinux.tip.CSIRO.AU ([130.155.192.102]:35201 "HELO mobilix.atnf.CSIRO.AU") by oss.sgi.com with SMTP id ; Thu, 2 Mar 2000 20:42:38 -0800 Received: (from rgooch@localhost) by mobilix.atnf.CSIRO.AU (8.6.12/8.6.12) id PAA03148; Fri, 3 Mar 2000 15:42:15 +1100 Date: Fri, 3 Mar 2000 15:42:15 +1100 Message-Id: <200003030442.PAA03148@mobilix.atnf.CSIRO.AU> From: Richard Gooch To: Khimenko Victor Cc: reese@isn.net, devfs@oss.sgi.com Subject: Re: Congrats to Richard Gooch In-Reply-To: References: <200002172226.JAA07210@mobilix.atnf.CSIRO.AU> Sender: owner-devfs@oss.sgi.com Precedence: bulk Return-Path: X-Orcpt: rfc822;devfs-outgoing Khimenko Victor writes: > On Fri, 18 Feb 2000, Richard Gooch wrote: > > Let me get this straight: when no medium is inserted, the "disc" entry > > exists? That's not what should happen. > > > Ok. I just checked once more. No, if there are no medium is inserted > "disc" entry is not created. Now I recovered the whole situation: I > inserted ZIP drive without partition table => ide-floppy loaded, > disc entry created. Then I tried to mount is as zip-win (i.e. not as > whole disc but as part4) and failed. Then I replaced it with proper > zip disc with partiton table and tried to access part4... Yes, most > common problem is fixed in recent versions of devfs. This one does > not :-(( Of course it's suble problem and so solution can wait :-) So the problem is that while the "disc" entry is created (zip-win) medium inserted, no partition entry is created? >From what you say, it appears that the medium doesn't have a valid partition table. When you insert and access the win-zip medium, what partition table information appears in the kernel logs? Perhaps the problem lies in the Linux partition table reading. If that's the case, devfs won't see any partitions either. Regards, Richard.... Permanent: rgooch@atnf.csiro.au Current: rgooch@ras.ucalgary.ca From owner-devfs@oss.sgi.com Fri Mar 3 00:20:28 2000 Received: by oss.sgi.com id ; Fri, 3 Mar 2000 00:20:19 -0800 Received: from firewall-in.sch57.msk.ru ([195.178.195.6]:11017 "EHLO dell.sch57.msk.ru") by oss.sgi.com with ESMTP id ; Fri, 3 Mar 2000 00:19:55 -0800 Received: from khim.UUCP (uucp@localhost) by dell.sch57.msk.ru (8.8.8/8.8.8) with UUCP id LAA22128; Fri, 3 Mar 2000 11:02:22 +0300 Received: by khim.sch57.msk.ru (dMail for DOS v2.07a2, 12Jun98); Fri, 3 Mar 2000 11:07:20 +0300 To: rgooch@atnf.csiro.au, khim@dell.sch57.msk.ru Cc: reese@isn.net, devfs@oss.sgi.com References: <200003030442.PAA03148@mobilix.atnf.CSIRO.AU> <200002172226.JAA07210@mobilix.atnf.CSIRO.AU> Message-Id: Organization: MCCME From: "Khimenko Victor" Date: Fri, 3 Mar 2000 11:07:20 +0300 (MSK) X-PM-Encryptor: PM-CRYPTPGP, 1 X-Mailer: dMail [Demos Mail for DOS v2.07a2] Subject: Re: Congrats to Richard Gooch Lines: 174 MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Sender: owner-devfs@oss.sgi.com Precedence: bulk Return-Path: X-Orcpt: rfc822;devfs-outgoing In <200003030442.PAA03148@mobilix.atnf.CSIRO.AU> Richard Gooch (rgooch@atnf.csiro.au) wrote: RG> Khimenko Victor writes: >> On Fri, 18 Feb 2000, Richard Gooch wrote: >> > Let me get this straight: when no medium is inserted, the "disc" entry >> > exists? That's not what should happen. >> > >> Ok. I just checked once more. No, if there are no medium is inserted >> "disc" entry is not created. Now I recovered the whole situation: I >> inserted ZIP drive without partition table => ide-floppy loaded, >> disc entry created. Then I tried to mount is as zip-win (i.e. not as >> whole disc but as part4) and failed. Then I replaced it with proper >> zip disc with partiton table and tried to access part4... Yes, most >> common problem is fixed in recent versions of devfs. This one does >> not :-(( Of course it's suble problem and so solution can wait :-) RG> So the problem is that while the "disc" entry is created (zip-win) RG> medium inserted, no partition entry is created? No. No partition entry was created when I tried to access partion entry ! >>From what you say, it appears that the medium doesn't have a valid RG> partition table. When you insert and access the win-zip medium, what RG> partition table information appears in the kernel logs? When I insert -- no. When I access -- yes. RG> Perhaps the problem lies in the Linux partition table reading. If RG> that's the case, devfs won't see any partitions either. No. There ARE problems in Linux partition table reading. But it's other issue. Here we have DESIGN flaw, not IMPLEMENTATION flaw. Kernel will NOT rescan partition table when you'll try to load already loaded driver. Kernel WILL rescan partition table if you'll access medium. But when you'll try to use mount mount will not try to access medium via "disc" entry and "part4" will not apper there since there was no actual access to medium. Chicken-egg problem. Here is transcript (not that kernel will correctly read partion table only on second pass; this is kernel problem and orthogonal issue) -- cut -- [root@localhost /root]# dd if=/dev/ide/host0/bus1/target1/lun0/disc of=/dev/null bs=512 count=1 Dev 16:40 SGI disklabel: bad magic 00000000 unknown partition table 1+0 records in 1+0 records out [root@localhost /root]# dmesg | tail NTFS version 990411 MSDOS FS: Using codepage 866 MSDOS FS: IO charset koi8-r MSDOS FS: Using codepage 866 MSDOS FS: IO charset koi8-r hdd: 98304kB, 196608 blocks, 512 sector size hdd: 98304kB, 32/64/96 CHS, 4096 kBps, 512 sector size, 2941 rpm /dev/ide/host0/bus1/target1/lun0:Dev 16:40 Sun disklabel: bad magic 0000 Dev 16:40 SGI disklabel: bad magic 00000000 unknown partition table [root@localhost /root]# dd if=/dev/ide/host0/bus1/target1/lun0/disc of=/dev/null bs=512 count=1 Dev 16:40 SGI disklabel: bad magic 00000000 unknown partition table 1+0 records in 1+0 records out [root@localhost /root]# dmesg | tail MSDOS FS: IO charset koi8-r hdd: 98304kB, 196608 blocks, 512 sector size hdd: 98304kB, 32/64/96 CHS, 4096 kBps, 512 sector size, 2941 rpm /dev/ide/host0/bus1/target1/lun0:Dev 16:40 Sun disklabel: bad magic 0000 Dev 16:40 SGI disklabel: bad magic 00000000 unknown partition table VFS: Disk change detected on device ide1(22,64) /dev/ide/host0/bus1/target1/lun0:Dev 16:40 Sun disklabel: bad magic 0000 Dev 16:40 SGI disklabel: bad magic 00000000 unknown partition table [root@localhost /root]# ls -al /dev/ide/host0/bus1/target1/lun0 total 0 drwxr-xr-x 1 root root 0 Jan 1 1970 . drwxr-xr-x 1 root root 0 Jan 1 1970 .. brw------- 1 root root 22, 64 Jan 1 1970 disc [root@localhost /root]# dd if=/dev/ide/host0/bus1/target1/lun0/part4 of=/dev/null bs=512 count=1 dd: /dev/ide/host0/bus1/target1/lun0/part4: No such file or directory [root@localhost /root]# dmesg | tail MSDOS FS: IO charset koi8-r hdd: 98304kB, 196608 blocks, 512 sector size hdd: 98304kB, 32/64/96 CHS, 4096 kBps, 512 sector size, 2941 rpm /dev/ide/host0/bus1/target1/lun0:Dev 16:40 Sun disklabel: bad magic 0000 Dev 16:40 SGI disklabel: bad magic 00000000 unknown partition table VFS: Disk change detected on device ide1(22,64) /dev/ide/host0/bus1/target1/lun0:Dev 16:40 Sun disklabel: bad magic 0000 Dev 16:40 SGI disklabel: bad magic 00000000 unknown partition table [root@localhost /root]# ls -al /dev/ide/host0/bus1/target1/lun0 total 0 drwxr-xr-x 1 root root 0 Jan 1 1970 . drwxr-xr-x 1 root root 0 Jan 1 1970 .. brw------- 1 root root 22, 64 Jan 1 1970 disc no partition table reload> [root@localhost /root]# dd if=/dev/ide/host0/bus1/target1/lun0/disc of=/dev/null bs=512 count=1 Dev 16:40 SGI disklabel: bad magic 50492eeb unknown partition table 1+0 records in 1+0 records out [root@localhost /root]# dmesg | tail Dev 16:40 SGI disklabel: bad magic 00000000 unknown partition table VFS: Disk change detected on device ide1(22,64) /dev/ide/host0/bus1/target1/lun0:Dev 16:40 Sun disklabel: bad magic 0000 Dev 16:40 SGI disklabel: bad magic 00000000 unknown partition table VFS: Disk change detected on device ide1(22,64) /dev/ide/host0/bus1/target1/lun0:Dev 16:40 Sun disklabel: bad magic 0000 Dev 16:40 SGI disklabel: bad magic 50492eeb unknown partition table [root@localhost /root]# ls -al /dev/ide/host0/bus1/target1/lun0 total 0 drwxr-xr-x 1 root root 0 Jan 1 1970 . drwxr-xr-x 1 root root 0 Jan 1 1970 .. brw------- 1 root root 22, 64 Jan 1 1970 disc [root@localhost /root]# dd if=/dev/ide/host0/bus1/target1/lun0/disc of=/dev/null bs=512 count=1 1+0 records in 1+0 records out [root@localhost /root]# dmesg | tail VFS: Disk change detected on device ide1(22,64) /dev/ide/host0/bus1/target1/lun0:Dev 16:40 Sun disklabel: bad magic 0000 Dev 16:40 SGI disklabel: bad magic 00000000 unknown partition table VFS: Disk change detected on device ide1(22,64) /dev/ide/host0/bus1/target1/lun0:Dev 16:40 Sun disklabel: bad magic 0000 Dev 16:40 SGI disklabel: bad magic 50492eeb unknown partition table VFS: Disk change detected on device ide1(22,64) /dev/ide/host0/bus1/target1/lun0: p4 [root@localhost /root]# ls -al /dev/ide/host0/bus1/target1/lun0 total 0 drwxr-xr-x 1 root root 0 Jan 1 1970 . drwxr-xr-x 1 root root 0 Jan 1 1970 .. brw------- 1 root root 22, 64 Jan 1 1970 disc brw------- 1 root root 22, 68 Jan 1 1970 part4 [root@localhost /root]# dd if=/dev/ide/host0/bus1/target1/lun0/part4 of=/dev/null bs=512 count=1 1+0 records in 1+0 records out [root@localhost /root]# dd if=/dev/ide/host0/bus1/target1/lun0/disc of=/dev/null bs=512 count=1 1+0 records in 1+0 records out [root@localhost /root]# dd if=/dev/ide/host0/bus1/target1/lun0/part4 of=/dev/null bs=512 count=1 1+0 records in 1+0 records out [root@localhost /root]# dmesg | tail 1+0 records in 1+0 records out [root@localhost /root]# dmesg | tail VFS: Disk change detected on device ide1(22,64) /dev/ide/host0/bus1/target1/lun0:Dev 16:40 Sun disklabel: bad magic 0000 Dev 16:40 SGI disklabel: bad magic 50492eeb unknown partition table VFS: Disk change detected on device ide1(22,64) /dev/ide/host0/bus1/target1/lun0: p4 VFS: Disk change detected on device ide1(22,64) /dev/ide/host0/bus1/target1/lun0: p4 VFS: Disk change detected on device ide1(22,64) /dev/ide/host0/bus1/target1/lun0: p4 [root@localhost /root]# rmmod ide-floppy [root@localhost /root]# ls -al /dev/ide/host0/bus1/target1/lun0 total 0 drwxr-xr-x 1 root root 0 Jan 1 1970 . drwxr-xr-x 1 root root 0 Jan 1 1970 .. brw------- 1 root root 22, 64 Jan 1 1970 disc brw------- 1 root root 22, 68 Jan 1 1970 part4 -- cut -- From owner-devfs@oss.sgi.com Sun Mar 12 19:52:36 2000 Received: by oss.sgi.com id ; Sun, 12 Mar 2000 19:52:26 -0800 Received: from mercury.mv.net ([199.125.85.40]:19728 "EHLO mercury.mv.net") by oss.sgi.com with ESMTP id ; Sun, 12 Mar 2000 19:52:16 -0800 Received: from wa1hco (nnh-3-36.mv.com [207.22.38.36]) by mercury.mv.net (8.8.8/mem-971025) with SMTP id WAA07287 for ; Sun, 12 Mar 2000 22:51:55 -0500 (EST) Message-ID: <001d01bf8c9f$4453e220$0201a8c0@home> From: "Jeff Millar" To: Subject: sorry for the stupid subscribe message Date: Sun, 12 Mar 2000 22:49:39 -0500 MIME-Version: 1.0 Content-Type: text/plain; charset="Windows-1252" Content-Transfer-Encoding: 7bit X-Priority: 3 X-MSMail-Priority: Normal X-Mailer: Microsoft Outlook Express 5.00.2919.6600 X-MimeOLE: Produced By Microsoft MimeOLE V5.00.2919.6600 Sender: owner-devfs@oss.sgi.com Precedence: bulk Return-Path: X-Orcpt: rfc822;devfs-outgoing I'm too numb to click on the right link From owner-devfs@oss.sgi.com Mon Mar 13 05:51:34 2000 Received: by oss.sgi.com id ; Mon, 13 Mar 2000 05:51:14 -0800 Received: from mirror1.wittsend.com ([130.205.0.28]:34310 "EHLO alcove.wittsend.com") by oss.sgi.com with ESMTP id ; Mon, 13 Mar 2000 05:50:53 -0800 Received: (from mhw@localhost) by alcove.wittsend.com (8.9.3/8.9.3) id IAA07914; Mon, 13 Mar 2000 08:50:26 -0500 Date: Mon, 13 Mar 2000 08:50:26 -0500 From: "Michael H. Warfield" To: Richard Gooch Cc: devfs@oss.sgi.com, mhw@wittsend.com Subject: Re: Linux-2.3.51, and the pre-2.4 series.. Message-ID: <20000313085026.C2010@alcove.wittsend.com> References: <20000311112556.A8346@alcove.wittsend.com> <200003130145.MAA00687@mobilix.atnf.CSIRO.AU> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline User-Agent: Mutt/1.1.5i In-Reply-To: <200003130145.MAA00687@mobilix.atnf.CSIRO.AU>; from rgooch@atnf.csiro.au on Mon, Mar 13, 2000 at 12:45:57PM +1100 Sender: owner-devfs@oss.sgi.com Precedence: bulk Return-Path: X-Orcpt: rfc822;devfs-outgoing On Mon, Mar 13, 2000 at 12:45:57PM +1100, Richard Gooch wrote: > Michael H. Warfield writes: > > On Sat, Mar 11, 2000 at 02:38:11PM +0000, Alan Cox wrote: > > > > did you enable devfs and forget to edit /etc/inittab to use 'vc/n' instead > > > > of 'ttyn' ? > > > If that is the case then devfs needs some tweaking > > No... That's the way it's designed to work and how it's been > > documented... Sigh... > > He's got a user space daemon, devfsd, that manages some symlinks > > from old names to devfs names, but most of the name space is orthogonal > > to the "conventional names". So you either switch to what Richard decided > > were more rational names, or you create the symlinks by hand, or you run > > his user space daemon. It's in the devfs/README file in the doco directory. > > Unfortunately, the devfsd doesn't cover everything (it does cover this, > > though). He's got a set of fixed links that it knows about (vc's are in > > the list). I haven't looked into it deeply enough to see if it can be > > configured for other devices (like the Computone devices) that he didn't > > code into it. The Computone devices are not in there, in spite of being > > in the devices.txt file: > If there's problems with devfsd not putting in all the right > compatibility symlinks, that's a bug that should be fixed. Send > patches. I promise to intend to provide all compatibility links. I > don't promise to be infallible (that incurs a surcharge). You need the following mappings to support the Computone device driver names as described in devices.txt. /dev/ip2ipl[n] -> /dev/ip2/ipl[n] n = 0 - 3 /dev/ip2stat[n] -> /dev/ip2/stat[n] n = 0 - 3 /dev/ttyF[n] -> /dev/ttf/[n] n = 0 - 255 /dev/cuf[n] -> /dev/cuf/[n] n = 0 - 255 (I followed the convention set by ttyS[n] -> tts/[n] and cua[n] -> cua/[n]). ===== From devices.txt ===== ] 71 char Computone IntelliPort II serial card ] 0 = /dev/ttyF0 IntelliPort II board 0, port 0 ] 1 = /dev/ttyF1 IntelliPort II board 0, port 1 ] ... ] 63 = /dev/ttyF63 IntelliPort II board 0, port 63 ] 64 = /dev/ttyF64 IntelliPort II board 1, port 0 ] 65 = /dev/ttyF65 IntelliPort II board 1, port 1 ] ... ] 127 = /dev/ttyF127 IntelliPort II board 1, port 63 ] 128 = /dev/ttyF128 IntelliPort II board 2, port 0 ] 129 = /dev/ttyF129 IntelliPort II board 2, port 1 ] ... ] 191 = /dev/ttyF191 IntelliPort II board 2, port 63 ] 192 = /dev/ttyF192 IntelliPort II board 3, port 0 ] 193 = /dev/ttyF193 IntelliPort II board 3, port 1 ] ... ] 255 = /dev/ttyF255 IntelliPort II board 3, port 63 ] ] 72 char Computone IntelliPort II serial card - alternate devices ] 0 = /dev/cuf0 Callout device for ttyF0 ] 1 = /dev/cuf1 Callout device for ttyF1 ] ... ] 63 = /dev/cuf63 Callout device for ttyF63 ] 64 = /dev/cuf64 Callout device for ttyF64 ] 65 = /dev/cuf65 Callout device for ttyF65 ] ... ] 127 = /dev/cuf127 Callout device for ttyF127 ] 128 = /dev/cuf128 Callout device for ttyF128 ] 129 = /dev/cuf129 Callout device for ttyF129 ] ... ] 191 = /dev/cuf191 Callout device for ttyF191 ] 192 = /dev/cuf192 Callout device for ttyF192 ] 193 = /dev/cuf193 Callout device for ttyF193 ] ... ] 255 = /dev/cuf255 Callout device for ttyF255 ] ] 73 char Computone IntelliPort II serial card - control devices ] 0 = /dev/ip2ipl0 Loadware device for board 0 ] 1 = /dev/ip2stat0 Status device for board 0 ] 4 = /dev/ip2ipl1 Loadware device for board 1 ] 5 = /dev/ip2stat1 Status device for board 1 ] 8 = /dev/ip2ipl2 Loadware device for board 2 ] 9 = /dev/ip2stat2 Status device for board 2 ] 12 = /dev/ip2ipl3 Loadware device for board 3 ] 13 = /dev/ip2stat3 Status device for board 3 > P.S. please send devfs-related messages to devfs@oss.sgi.com instead > of the kernel list. Just send "subscribe devfs" in the body to > majordomo@oss.sgi.com Added to Cc list. > Regards, > Richard.... > Permanent: rgooch@atnf.csiro.au > Current: rgooch@ras.ucalgary.ca Mike -- Michael H. Warfield | (770) 985-6132 | mhw@WittsEnd.com (The Mad Wizard) | (770) 331-2437 | http://www.wittsend.com/mhw/ NIC whois: MHW9 | An optimist believes we live in the best of all PGP Key: 0xDF1DD471 | possible worlds. A pessimist is sure of it! From owner-devfs@oss.sgi.com Wed Mar 15 06:38:01 2000 Received: by oss.sgi.com id ; Wed, 15 Mar 2000 06:37:51 -0800 Received: from mirror1.wittsend.com ([130.205.0.28]:51212 "EHLO alcove.wittsend.com") by oss.sgi.com with ESMTP id ; Wed, 15 Mar 2000 06:37:38 -0800 Received: (from mhw@localhost) by alcove.wittsend.com (8.9.3/8.9.3) id JAA23476; Wed, 15 Mar 2000 09:36:59 -0500 Date: Wed, 15 Mar 2000 09:36:59 -0500 From: "Michael H. Warfield" To: linux-kernel@vger.rutgers.edu, linux-usb@suse.com, devfs@oss.sgi.com Cc: mhw@wittsend.com Subject: DevFS and USB... Message-ID: <20000315093659.C15481@alcove.wittsend.com> Mail-Followup-To: "Michael H. Warfield" , linux-kernel@vger.rutgers.edu, linux-usb@suse.com, devfs@oss.sgi.com, mhw@wittsend.com Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline User-Agent: Mutt/1.1.5i Sender: owner-devfs@oss.sgi.com Precedence: bulk Return-Path: X-Orcpt: rfc822;devfs-outgoing Hello all, I just got involved in USB a few months ago (mostly for a camera memory interface) and am now in love with it. The fact that I can resolve a lot of interrupt conflicts and problems with running out of interrupts just by moving stuff to USB is wonderful. Skeptical at first, I'm now finding that it's solving a number of problems (like add on serial interfaces, parallel interfaces, multiple mice, and tablets). Considering that DevFS is now in the 2.3 kernels and is likely to be a fixture in the 2.4 kernels and USB is becoming quite popular, is anyone looking at DevFS support in USB? Given the comings and goings of devices and drivers and what not in USB, I would guess that DevFS support is going to involve more that a little bit of work (reasonably straight forward, just a lot of it). I just got done adding DevFS support to the Computone Multiport drivers and someone remarked to me that now that DevFS in in the kernel sources we are now cleaning up some loose ends (referring to the Computone Drivers as one such loose end). Seems to me like USB is a whole ball of loose ends for DevFS. Anyone even know what the USB hierarchy in the DevFS namespace should look like? Mike -- Michael H. Warfield | (770) 985-6132 | mhw@WittsEnd.com (The Mad Wizard) | (770) 331-2437 | http://www.wittsend.com/mhw/ NIC whois: MHW9 | An optimist believes we live in the best of all PGP Key: 0xDF1DD471 | possible worlds. A pessimist is sure of it! From owner-devfs@oss.sgi.com Wed Mar 15 08:24:52 2000 Received: by oss.sgi.com id ; Wed, 15 Mar 2000 08:24:42 -0800 Received: from ganymede.or.intel.com ([134.134.248.3]:10503 "EHLO ganymede.or.intel.com") by oss.sgi.com with ESMTP id ; Wed, 15 Mar 2000 08:24:18 -0800 Received: from SMTP (orsmsxvs01-1.jf.intel.com [192.168.65.200]) by ganymede.or.intel.com (8.9.1a+p1/8.9.1/d: relay.m4,v 1.19 2000/01/29 00:15:43 dmccart Exp $) with SMTP id IAA08193; Wed, 15 Mar 2000 08:24:09 -0800 (PST) Received: from orsmsx28.jf.intel.com ([192.168.70.28]) by 192.168.70.200 (Norton AntiVirus for Internet Email Gateways 1.0) ; Wed, 15 Mar 2000 16:24:08 0000 (GMT) Received: by orsmsx28.jf.intel.com with Internet Mail Service (5.5.2448.0) id ; Wed, 15 Mar 2000 08:24:06 -0800 Message-ID: From: "Dunlap, Randy" To: "'Michael H. Warfield'" , linux-kernel@vger.rutgers.edu, linux-usb@suse.com, devfs@oss.sgi.com Subject: RE: [linux-usb] DevFS and USB... Date: Wed, 15 Mar 2000 08:23:35 -0800 MIME-Version: 1.0 X-Mailer: Internet Mail Service (5.5.2448.0) Content-Type: text/plain; charset="ISO-8859-1" Sender: owner-devfs@oss.sgi.com Precedence: bulk Return-Path: X-Orcpt: rfc822;devfs-outgoing Hi Mike, I haven't looked at it and I don't recall seeing any email from anyone else on the linux-usb mailing list who has looked at DevFS. I don't disagree with your (guesstimate) evaluation of USB and DevFS. > Anyone even know what the USB hierarchy in the > DevFS namespace should look like? I'm no expert (or even User yet) on DevFS. My guess is simply /dev/usb/something. Where would I look for some examples of other devices/buses? in linux/Documentation/filesystems/devfs/mk-devlinks or ...................................../README ? I'm interested in your feedback and help on this. Thanks, ~Randy [www.linux-usb.org] ___________________________________________________ |Randy Dunlap Intel Corp., DAL Sr. SW Engr.| |randy.dunlap.at.intel.com 503-696-2055| |NOTE: Any views presented here are mine alone | |and may not represent the views of my employer. | |_________________________________________________| > -----Original Message----- > From: Michael H. Warfield [mailto:mhw@wittsend.com] > Sent: Wednesday, March 15, 2000 6:37 AM > To: linux-kernel@vger.rutgers.edu; linux-usb@suse.com; > devfs@oss.sgi.com > Cc: mhw@wittsend.com > Subject: [linux-usb] DevFS and USB... > > > Hello all, > > I just got involved in USB a few months ago (mostly for a camera > memory interface) and am now in love with it. The fact that > I can resolve > a lot of interrupt conflicts and problems with running out of > interrupts > just by moving stuff to USB is wonderful. Skeptical at first, I'm > now finding that it's solving a number of problems (like add on serial > interfaces, parallel interfaces, multiple mice, and tablets). > > Considering that DevFS is now in the 2.3 kernels and is likely > to be a fixture in the 2.4 kernels and USB is becoming quite popular, > is anyone looking at DevFS support in USB? Given the comings > and goings > of devices and drivers and what not in USB, I would guess that DevFS > support is going to involve more that a little bit of work (reasonably > straight forward, just a lot of it). > > I just got done adding DevFS support to the Computone Multiport > drivers and someone remarked to me that now that DevFS in in > the kernel > sources we are now cleaning up some loose ends (referring to > the Computone > Drivers as one such loose end). Seems to me like USB is a > whole ball of > loose ends for DevFS. Anyone even know what the USB hierarchy in the > DevFS namespace should look like? > > Mike > -- > Michael H. Warfield | (770) 985-6132 | mhw@WittsEnd.com > (The Mad Wizard) | (770) 331-2437 | http://www.wittsend.com/mhw/ NIC whois: MHW9 | An optimist believes we live in the best of all PGP Key: 0xDF1DD471 | possible worlds. A pessimist is sure of it! --------------------------------------------------------------------- To unsubscribe, e-mail: linux-usb-unsubscribe@suse.com For additional commands, e-mail: linux-usb-help@suse.com From owner-devfs@oss.sgi.com Wed Mar 15 09:43:42 2000 Received: by oss.sgi.com id ; Wed, 15 Mar 2000 09:43:22 -0800 Received: from cm-24-142-61-146.cableco-op.ispchannel.com ([24.142.61.146]:27898 convert rfc822-to-8bit "EHLO nova.botz.org") by oss.sgi.com with ESMTP id ; Wed, 15 Mar 2000 09:42:52 -0800 Received: from nova.botz.org (IDENT:jbotz@localhost [127.0.0.1]) by nova.botz.org (8.9.3/8.9.3) with ESMTP id JAA17128 for ; Wed, 15 Mar 2000 09:42:43 -0800 Message-Id: <200003151742.JAA17128@nova.botz.org> X-Mailer: exmh version 2.1.1 10/15/1999 To: devfs@oss.sgi.com Subject: ttyS1, S2, etc... Mime-Version: 1.0 Content-Type: text/plain; charset=iso-8859-1 Mime-Version: 1.0 Content-Transfer-Encoding: 8BIT Date: Wed, 15 Mar 2000 09:42:43 -0800 From: Jurgen Botz Sender: owner-devfs@oss.sgi.com Precedence: bulk Return-Path: X-Orcpt: rfc822;devfs-outgoing I have a six-port BOCA serial card. It is supported by the standard serial driver with multi-port extension and share IRQ. devfs doesn't do anything with this... I have to manually create the devices. I think that may be inevitable, since even the serial driver can't really detect these kinds of cards and you have to use 'setserial' configure them. What's the best way to deal with this? Maybe devfsd special case for them? Or should the serial driver have devfs code to do the right thing once setserial 'activates' a port? - Jürgen From owner-devfs@oss.sgi.com Wed Mar 15 16:31:43 2000 Received: by oss.sgi.com id ; Wed, 15 Mar 2000 16:31:24 -0800 Received: from linuxcare.canberra.net.au ([203.29.91.49]:9206 "HELO mobilix.atnf.CSIRO.AU") by oss.sgi.com with SMTP id ; Wed, 15 Mar 2000 16:30:56 -0800 Received: (from rgooch@localhost) by mobilix.atnf.CSIRO.AU (8.6.12/8.6.12) id LAA21959; Thu, 16 Mar 2000 11:30:42 +1100 Date: Thu, 16 Mar 2000 11:30:42 +1100 Message-Id: <200003160030.LAA21959@mobilix.atnf.CSIRO.AU> From: Richard Gooch To: Jurgen Botz Cc: devfs@oss.sgi.com Subject: Re: ttyS1, S2, etc... In-Reply-To: <200003151742.JAA17128@nova.botz.org> References: <200003151742.JAA17128@nova.botz.org> Sender: owner-devfs@oss.sgi.com Precedence: bulk Return-Path: X-Orcpt: rfc822;devfs-outgoing Jurgen Botz writes: > I have a six-port BOCA serial card. It is supported by the standard > serial driver with multi-port extension and share IRQ. devfs doesn't > do anything with this... I have to manually create the devices. I > think that may be inevitable, since even the serial driver can't > really detect these kinds of cards and you have to use 'setserial' > configure them. What's the best way to deal with this? Maybe > devfsd special case for them? Or should the serial driver have > devfs code to do the right thing once setserial 'activates' a port? In drivers/char/serial.c:register_serial() there are calls which create devfs entries. So the low-level serial driver should call register_serial() and everything should be fine. It certainly works for PCMCIA serial cards. Insert the card and the devfs entry magically appears. Regards, Richard.... Permanent: rgooch@atnf.csiro.au Current: rgooch@ras.ucalgary.ca From owner-devfs@oss.sgi.com Thu Mar 16 09:52:48 2000 Received: by oss.sgi.com id ; Thu, 16 Mar 2000 09:52:28 -0800 Received: from cm-24-142-61-146.cableco-op.ispchannel.com ([24.142.61.146]:49392 convert rfc822-to-8bit "EHLO nova.botz.org") by oss.sgi.com with ESMTP id ; Thu, 16 Mar 2000 09:52:09 -0800 Received: from nova.botz.org (IDENT:jbotz@localhost [127.0.0.1]) by nova.botz.org (8.9.3/8.9.3) with ESMTP id JAA24104; Thu, 16 Mar 2000 09:51:18 -0800 Message-Id: <200003161751.JAA24104@nova.botz.org> X-Mailer: exmh version 2.1.1 10/15/1999 To: Richard Gooch cc: devfs@oss.sgi.com, tytso@mit.edu Subject: Re: ttyS1, S2, etc... In-Reply-To: Message from Richard Gooch of "Thu, 16 Mar 2000 11:30:42 +1100." <200003160030.LAA21959@mobilix.atnf.CSIRO.AU> Mime-Version: 1.0 Content-Type: text/plain; charset=iso-8859-1 Mime-Version: 1.0 Content-Transfer-Encoding: 8BIT Date: Thu, 16 Mar 2000 09:51:18 -0800 From: Jurgen Botz Sender: owner-devfs@oss.sgi.com Precedence: bulk Return-Path: X-Orcpt: rfc822;devfs-outgoing Richard Gooch wrote: > In drivers/char/serial.c:register_serial() there are calls which > create devfs entries. So the low-level serial driver should call > register_serial() and everything should be fine. It certainly works > for PCMCIA serial cards. Insert the card and the devfs entry magically > appears. The problem is that ISA multiport serial cards don't appear to be detected by the kernel at all and register_serial never gets called. To use the ports you set the IO base and IRQ with setserial which sets them in the driver with ioctls on the device. This bypasses register_serial in the driver, too, so even if I create /dev/ttyS2 and then do the setserial thing, /dev/tts/2 still doesn't appear (but /dev/ttyS2 then works fine). Not that that would make a difference for devfs, of course. Looking through serial.c I see that PCI and ISAPNP cards probably don't have this problem since the driver detects them and presumably calls register_serial for their ports. There is probably no way around this chick-and-egg problem without providing some alternate run-time or init-time configuration mechanism for the serial driver (i.e. module/kernel flags maybe). That may not be worth the trouble to provide automatic devfs support for the diminishing number of non-PNP ISA multiport boards out there. Since I currently have to run an rc script to configure the ports anyway, I might as well create the device files there with mknod. I'm copying tytso in case in case he has any thoughts on this... - Jürgen From owner-devfs@oss.sgi.com Thu Mar 16 21:41:22 2000 Received: by oss.sgi.com id ; Thu, 16 Mar 2000 21:41:02 -0800 Received: from linuxcare.canberra.net.au ([203.29.91.49]:16380 "EHLO trampoline.thunk.org") by oss.sgi.com with ESMTP id ; Thu, 16 Mar 2000 21:40:38 -0800 Received: (from tytso@localhost) by trampoline.thunk.org (8.9.3/8.9.3) id AAA02886; Fri, 17 Mar 2000 00:43:39 -0500 Date: Fri, 17 Mar 2000 00:43:39 -0500 Message-Id: <200003170543.AAA02886@trampoline.thunk.org> To: jurgen@botz.org CC: rgooch@atnf.csiro.au, devfs@oss.sgi.com In-reply-to: <200003161751.JAA24104@nova.botz.org> (message from Jurgen Botz on Thu, 16 Mar 2000 09:51:18 -0800) Subject: Re: ttyS1, S2, etc... From: tytso@mit.edu Phone: (781) 391-3464 References: <200003161751.JAA24104@nova.botz.org> Sender: owner-devfs@oss.sgi.com Precedence: bulk Return-Path: X-Orcpt: rfc822;devfs-outgoing Date: Thu, 16 Mar 2000 09:51:18 -0800 From: Jurgen Botz Richard Gooch wrote: > In drivers/char/serial.c:register_serial() there are calls which > create devfs entries. So the low-level serial driver should call > register_serial() and everything should be fine. It certainly works > for PCMCIA serial cards. Insert the card and the devfs entry magically > appears. The problem is that ISA multiport serial cards don't appear to be detected by the kernel at all and register_serial never gets called. To use the ports you set the IO base and IRQ with setserial which sets them in the driver with ioctls on the device. This bypasses register_serial in the driver, too, so even if I create /dev/ttyS2 and then do the setserial thing, /dev/tts/2 still doesn't appear (but /dev/ttyS2 then works fine). Not that that would make a difference for devfs, of course. Right. There are actually two parts to this problem. The first is that when setserial calls TIOCSSERIAL to change the uart from "none" to "non-none" and vice versa, tty_register_devfs() and tty_unregister_devfs() needs to be called. I can fix this in the serial driver fairly easily, and this will take care of the problem which you reported. The second problem is how do you actually get setserial to set the port, irq, and uart settings using only devfs in the first place. The issue here is that you can't autodetect ISA multiport serial cards, at least not those cards that don't have PNP support. The port and irq settings either have to manually configured such as is the case for COM 1/2/3/4, or the user has to manually set them using the setserial command. But, in a devfs only world, the device file doesn't appear until the port is configured, but setserial needs the device file in order to configure the port. There's no real way around this.... That may not be worth the trouble to provide automatic devfs support for the diminishing number of non-PNP ISA multiport boards out there. Since I currently have to run an rc script to configure the ports anyway, I might as well create the device files there with mknod. Perhaps, but this isn't very satisfying; it's certainly very awkward for users, and for people who were accustomed to things Just Working (tm) in Linux 2.2, it's really, really, really ugly to tell them that they have to manually mknod the device in devfs. However, short of telling them (a) don't use devfs, or (b) hacking setserial to automatically do the mknod if /dev/ttyS* is passed in and /dev/ttyS* doesn't exist. Both aren't pretty solutions, though. What makes this worse is that I've been encouraging people to use rc scripts for configuring ISA multiport cards, since there are so many of them and using compiled in, hard-coded configuration parameters is evil. The /devfs model doesn't seem to allow this terribly easily, unfortunately. I suppose we could shovel all of this dirt under the devfsd rug, and make devfsd responsible for parsing some new serial configuration file which then does the mknod and TIOCSERIAL ioctl. But that's not that great of a solution either. I'll need to think about this one some more. I'm not particularly happy with any of these solutions. Richard, what do you think? - Ted From owner-devfs@oss.sgi.com Sat Mar 18 03:35:12 2000 Received: by oss.sgi.com id ; Sat, 18 Mar 2000 03:35:02 -0800 Received: from firewall-in.sch57.msk.ru ([195.178.195.6]:35338 "EHLO dell.sch57.msk.ru") by oss.sgi.com with ESMTP id ; Sat, 18 Mar 2000 03:34:39 -0800 Received: from khim.UUCP (uucp@localhost) by dell.sch57.msk.ru (8.8.8/8.8.8) with UUCP id OAA19382; Sat, 18 Mar 2000 14:12:20 +0300 Received: by khim.sch57.msk.ru (dMail for DOS v2.07a2, 12Jun98); Sat, 18 Mar 2000 14:18:56 +0300 To: tytso@mit.edu, jurgen@botz.org Cc: rgooch@atnf.csiro.au, devfs@oss.sgi.com References: <200003170543.AAA02886@trampoline.thunk.org> <200003161751.JAA24104@nova.botz.org> Message-Id: Organization: MCCME From: "Khimenko Victor" Date: Sat, 18 Mar 2000 14:18:56 +0300 (MSK) X-Mailer: dMail [Demos Mail for DOS v2.07a2] Subject: Re: ttyS1, S2, etc... Lines: 77 MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Sender: owner-devfs@oss.sgi.com Precedence: bulk Return-Path: X-Orcpt: rfc822;devfs-outgoing 17-Mar-00 00:43 you wrote: All this setserial stuff is just a hack. Why you can not just use normal way: command-line kernel parameters and/or module parameters like other devices (like IDE or SCSI) are doing ? What's so special with serial ports anyway ? It's onething to change settings of device via ioctl (number of stop-bits, speed of port, etc) and it's other task to specify IRQ or IO port number. For latter most other drivers are using module command-lines, not magic ioctls > Date: Thu, 16 Mar 2000 09:51:18 -0800 > From: Jurgen Botz > Richard Gooch wrote: >> In drivers/char/serial.c:register_serial() there are calls which >> create devfs entries. So the low-level serial driver should call >> register_serial() and everything should be fine. It certainly works >> for PCMCIA serial cards. Insert the card and the devfs entry magically >> appears. > The problem is that ISA multiport serial cards don't appear to be > detected by the kernel at all and register_serial never gets called. > To use the ports you set the IO base and IRQ with setserial which > sets them in the driver with ioctls on the device. This bypasses > register_serial in the driver, too, so even if I create /dev/ttyS2 > and then do the setserial thing, /dev/tts/2 still doesn't appear > (but /dev/ttyS2 then works fine). Not that that would make a > difference for devfs, of course. > Right. There are actually two parts to this problem. The first is that > when setserial calls TIOCSSERIAL to change the uart from "none" to > "non-none" and vice versa, tty_register_devfs() and > tty_unregister_devfs() needs to be called. I can fix this in the serial > driver fairly easily, and this will take care of the problem which you > reported. > The second problem is how do you actually get setserial to set the port, > irq, and uart settings using only devfs in the first place. The issue > here is that you can't autodetect ISA multiport serial cards, at least > not those cards that don't have PNP support. The port and irq settings > either have to manually configured such as is the case for COM 1/2/3/4, > or the user has to manually set them using the setserial command. > But, in a devfs only world, the device file doesn't appear until the > port is configured, but setserial needs the device file in order to > configure the port. There's no real way around this.... > That may not be worth the trouble to provide automatic devfs > support for the diminishing number of non-PNP ISA multiport > boards out there. Since I currently have to run an rc script > to configure the ports anyway, I might as well create the device > files there with mknod. > Perhaps, but this isn't very satisfying; it's certainly very awkward for > users, and for people who were accustomed to things Just Working (tm) in > Linux 2.2, it's really, really, really ugly to tell them that they have > to manually mknod the device in devfs. However, short of telling them > (a) don't use devfs, or (b) hacking setserial to automatically do the > mknod if /dev/ttyS* is passed in and /dev/ttyS* doesn't exist. Both > aren't pretty solutions, though. > What makes this worse is that I've been encouraging people to use rc > scripts for configuring ISA multiport cards, since there are so many of > them and using compiled in, hard-coded configuration parameters is > evil. The /devfs model doesn't seem to allow this terribly easily, > unfortunately. > I suppose we could shovel all of this dirt under the devfsd rug, and > make devfsd responsible for parsing some new serial configuration file > which then does the mknod and TIOCSERIAL ioctl. But that's not that > great of a solution either. > I'll need to think about this one some more. I'm not particularly > happy with any of these solutions. Richard, what do you think? > - Ted From owner-devfs@oss.sgi.com Sun Mar 19 16:07:45 2000 Received: by oss.sgi.com id ; Sun, 19 Mar 2000 16:07:36 -0800 Received: from rennes-231.abo.wanadoo.fr ([193.252.125.231]:20228 "EHLO rennes-231.abo.wanadoo.fr") by oss.sgi.com with ESMTP id ; Sun, 19 Mar 2000 16:07:22 -0800 Received: from wanadoo.fr (IDENT:martin@loopback [127.0.0.1]) by rennes-231.abo.wanadoo.fr (8.9.3/8.9.3) with ESMTP id BAA00916 for ; Mon, 20 Mar 2000 01:07:24 +0100 Message-ID: <38D56BBC.3F4B01F6@wanadoo.fr> Date: Mon, 20 Mar 2000 01:07:24 +0100 From: Martin Costabel Organization: Chez moi X-Mailer: Mozilla 4.7 [en] (X11; U; Linux 2.3.99-pre2 ppc) X-Accept-Language: en MIME-Version: 1.0 To: devfs@oss.sgi.com Subject: Some basic problems with devfs References: <200003151742.JAA17128@nova.botz.org> Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Sender: owner-devfs@oss.sgi.com Precedence: bulk Return-Path: X-Orcpt: rfc822;devfs-outgoing Hi, I have only recently started to use kernels with devfs enabled, and I have some newbie questions. I don't know if this list is the right place to ask them, but this is already my first question: Is there a faq or howto for devfs, other than the far too short and technical kernel source filesystems/devfs/README? By wading through thousands of linux-kernel messages, I found some informations, so I am now starting "devfsd /dev" at the beginning of rc.sysinit, and I have learned from Doug Gilbert's page at http://www.torque.net/sg/devfs_scsi.html how to open some security holes so that I can login as root or start X as non-root. But a couple of things still don't work, in particular my printer, floppy, cd, and gpm, and this leads to 3 different questions: 1. My printer uses /dev/ttyS1, and when I used devfs for the first time with kernel 2.3.51, there was no such device. Instead of ttyS0-3 there was only a /dev/ttyS with major 4, minor 64. 3 days ago, I booted with a kernel 2.3.52, and suddenly the devices ttyS0 and ttyS1 (which correspond to the 2 serial ports I have) were back. Printing worked. Now yesterday I booted with 2.3.99-pre2, and I was back to square one: no ttyS1, only a ttyS. All 3 kernels used basically the same config options and the same kernel command line, and devfsd was started automatically in the same way. Where does this difference come from and how can I make sure that devfsd creates ttyS0 and ttyS1 as, according to README, it is supposed to do? Where does devfsd get its list of devices from? How does it decide which ones to create? 2. There is no /dev/floppy nor fd0 or whatever. ToDo mentions that swim3 is not yet supported. I have a Pmac 6400, and swim3 is its floppy controller. Is there any way to use this floppy with devfs? How hard is it to add devfs support to drivers/block/swim3.c? 3. Gpm doesn't start because /dev/mouse is missing. Normally, this is a symlink to /dev/adbmouse. If I make such a link, it will disappear upon reboot. Same problem with /dev/cdrom, /dev/modem and many others that are created either by some programs or by myself using mknod or symlinks. Is there a way to tell devfsd about such things so that devfsd recreates them at reboot? The documentation on devfsd (man devfsd) is too rudimentary to know. Does one have to use the rc.devfs script and store these informations in some tar file? Or is the recommended way after all to use 2 directories, /devfs for devfs to play with, and /dev for the use of the rest of the kernel with symlinks between the two? These may be dumb questions, but if they don't have simple answers, I shall have to say good-bye to devfs. -- Martin From owner-devfs@oss.sgi.com Sun Mar 19 18:35:45 2000 Received: by oss.sgi.com id ; Sun, 19 Mar 2000 18:35:36 -0800 Received: from linuxcare.canberra.net.au ([203.29.91.49]:52474 "EHLO trampoline.thunk.org") by oss.sgi.com with ESMTP id ; Sun, 19 Mar 2000 18:35:19 -0800 Received: (from tytso@localhost) by trampoline.thunk.org (8.9.3/8.9.3) id VAA16730; Sun, 19 Mar 2000 21:38:38 -0500 Date: Sun, 19 Mar 2000 21:38:38 -0500 Message-Id: <200003200238.VAA16730@trampoline.thunk.org> To: devfs@khim.sch57.msk.ru CC: jurgen@botz.org, rgooch@atnf.csiro.au, devfs@oss.sgi.com In-reply-to: (devfs@khim.sch57.msk.ru) Subject: Re: ttyS1, S2, etc... From: tytso@mit.edu Phone: (781) 391-3464 References: <200003170543.AAA02886@trampoline.thunk.org> <200003161751.JAA24104@nova.botz.org> Sender: owner-devfs@oss.sgi.com Precedence: bulk Return-Path: X-Orcpt: rfc822;devfs-outgoing From: "Khimenko Victor" Date: Sat, 18 Mar 2000 14:18:56 +0300 (MSK) All this setserial stuff is just a hack. Why you can not just use normal way: command-line kernel parameters and/or module parameters like other devices (like IDE or SCSI) are doing ? What's so special with serial ports anyway ? It's onething to change settings of device via ioctl (number of stop-bits, speed of port, etc) and it's other task to specify IRQ or IO port number. For latter most other drivers are using module command-lines, not magic ioctls. Some people have as many as 32-64 dumb serial ports, and so configuring them on the command-line can get very ugly. Personally, if you can't reliably autodetect devices (which is the real problem; the ISA bus is a hack), I consider being able to configure and reconfigure devices from a booted system to be much cleaner than boot-time only configuration via a command-line, which to me seems to be infinitely more kludgy. - Ted From owner-devfs@oss.sgi.com Sun Mar 19 18:46:15 2000 Received: by oss.sgi.com id ; Sun, 19 Mar 2000 18:46:06 -0800 Received: from cm-24-142-61-146.cableco-op.ispchannel.com ([24.142.61.146]:17401 convert rfc822-to-8bit "EHLO nova.botz.org") by oss.sgi.com with ESMTP id ; Sun, 19 Mar 2000 18:45:46 -0800 Received: from nova.botz.org (IDENT:jbotz@localhost [127.0.0.1]) by nova.botz.org (8.9.3/8.9.3) with ESMTP id SAA21891; Sun, 19 Mar 2000 18:44:56 -0800 Message-Id: <200003200244.SAA21891@nova.botz.org> X-Mailer: exmh version 2.1.1 10/15/1999 To: tytso@mit.edu cc: devfs@khim.sch57.msk.ru, rgooch@atnf.csiro.au, devfs@oss.sgi.com Subject: Re: ttyS1, S2, etc... In-Reply-To: Message from tytso@mit.edu of "Sun, 19 Mar 2000 21:38:38 EST." <200003200238.VAA16730@trampoline.thunk.org> Mime-Version: 1.0 Content-Type: text/plain; charset=iso-8859-1 Mime-Version: 1.0 Content-Transfer-Encoding: 8BIT Date: Sun, 19 Mar 2000 18:44:56 -0800 From: Jurgen Botz Sender: owner-devfs@oss.sgi.com Precedence: bulk Return-Path: X-Orcpt: rfc822;devfs-outgoing tytso@mit.edu wrote: > them on the command-line can get very ugly. Personally, if you can't > reliably autodetect devices (which is the real problem; the ISA bus is a > hack), I consider being able to configure and reconfigure devices from a > booted system to be much cleaner than boot-time only configuration via a > command-line, which to me seems to be infinitely more kludgy. I would agree with that, and add that device configuration shouldn't work differently for modules vs. compiled-in drivers as is the case now. Maybe it /would/ be best to provide some kind of configuration mechanism for undetectable devices through devfs/devfsd, so that we can move toward a single config file (currently devfsd.conf, maybe to be called devices.conf) for handeling all device-related stuff? - Jürgen From owner-devfs@oss.sgi.com Wed Mar 22 02:46:53 2000 Received: by oss.sgi.com id ; Wed, 22 Mar 2000 02:46:44 -0800 Received: from rennes-231.abo.wanadoo.fr ([193.252.125.231]:19716 "EHLO rennes-231.abo.wanadoo.fr") by oss.sgi.com with ESMTP id ; Wed, 22 Mar 2000 02:46:26 -0800 Received: from wanadoo.fr (IDENT:martin@loopback [127.0.0.1]) by rennes-231.abo.wanadoo.fr (8.9.3/8.9.3) with ESMTP id LAA02125 for ; Wed, 22 Mar 2000 11:46:27 +0100 Message-ID: <38D8A481.9E2D1CDA@wanadoo.fr> Date: Wed, 22 Mar 2000 11:46:25 +0100 From: Martin Costabel Organization: Chez moi X-Mailer: Mozilla 4.7 [en] (X11; U; Linux 2.3.99-pre3 ppc) X-Accept-Language: en MIME-Version: 1.0 To: devfs@oss.sgi.com Subject: Re: Some basic problems with devfs References: <200003151742.JAA17128@nova.botz.org> <38D56BBC.3F4B01F6@wanadoo.fr> Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Sender: owner-devfs@oss.sgi.com Precedence: bulk Return-Path: X-Orcpt: rfc822;devfs-outgoing Nothing like answering your own questions (no one else volunteered) :-) Martin Costabel wrote: > 1. My printer uses /dev/ttyS1, and when I used devfs for the first time > with kernel 2.3.51, there was no such device. Instead of ttyS0-3 there > was only a /dev/ttyS with major 4, minor 64. 3 days ago, I booted with a > kernel 2.3.52, and suddenly the devices ttyS0 and ttyS1 (which > correspond to the 2 serial ports I have) were back. Printing worked. Now > yesterday I booted with 2.3.99-pre2, and I was back to square one: no > ttyS1, only a ttyS. All 3 kernels used basically the same config options > and the same kernel command line, and devfsd was started automatically > in the same way. Where does this difference come from and how can I make > sure that devfsd creates ttyS0 and ttyS1 as, according to README, it is > supposed to do? Where does devfsd get its list of devices from? How does > it decide which ones to create? I got it working now. The problem was that I am using Macintosh serial ports (compiled in with the MAC_SERIAL config option), and drivers/macintosh/macserial.c was not yet devfs aware. I patched it with some lines from drivers/char/serial.c, and I have now /dev/tts/ and /dev/cua/ directories that were not there before. The 2.3.52 kernel which worked also came from a different tree and apparently had a patched version of the macserial.c file, too. -- Martin