From owner-pro64-support@oss.sgi.com Fri Sep 1 10:27:29 2000 Received: by oss.sgi.com id ; Fri, 1 Sep 2000 10:27:19 -0700 Received: from mail.ee.gatech.edu ([130.207.230.10]:47060 "EHLO mail.ee.gatech.edu") by oss.sgi.com with ESMTP id ; Fri, 1 Sep 2000 10:26:54 -0700 Received: from lory.ee.gatech.edu (IDENT:root@lory.ee.gatech.edu [130.207.228.43]) by mail.ee.gatech.edu (8.11.0/8.11.0) with ESMTP id e81HQh024413 for ; Fri, 1 Sep 2000 13:26:43 -0400 (EDT) Received: (from hardnett@localhost) by lory.ee.gatech.edu (8.9.3/8.8.4) id NAA00771; Fri, 1 Sep 2000 13:19:41 -0400 From: Charles Hardnett MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Message-ID: <14767.58594.107811.557276@lory.ee.gatech.edu> Date: Fri, 1 Sep 2000 13:18:26 -0400 (EDT) To: pro64-support@oss.sgi.com Subject: The .B files X-Mailer: VM 6.71 under 21.1 (patch 4) "Arches" XEmacs Lucid Reply-To: hardnett@cc.gatech.edu Sender: owner-pro64-support@oss.sgi.com Precedence: bulk Return-Path: X-Orcpt: rfc822;pro64-support-outgoing has anyone used the -keep option for the compiler. If so, there is a .B file that is generated. Is there anyone that knows at what stage during the compilation this file is generated? Charles -- -------------------------------------------------------------------- Charles R. Hardnett www.spelman.edu/~hardnett From owner-pro64-support@oss.sgi.com Fri Sep 1 10:31:59 2000 Received: by oss.sgi.com id ; Fri, 1 Sep 2000 10:31:39 -0700 Received: from deliverator.sgi.com ([204.94.214.10]:21340 "EHLO deliverator.sgi.com") by oss.sgi.com with ESMTP id ; Fri, 1 Sep 2000 10:31:28 -0700 Received: from gaea.engr.sgi.com (gaea.engr.sgi.com [130.62.180.97]) by deliverator.sgi.com (980309.SGI.8.8.8-aspam-6.2/980310.SGI-aspam) via ESMTP id KAA06524 for ; Fri, 1 Sep 2000 10:23:50 -0700 (PDT) mail_from (murthy@gaea.engr.sgi.com) Received: (from murthy@localhost) by gaea.engr.sgi.com (980427.SGI.8.8.8/970903.SGI.AUTOCF) id KAA37617; Fri, 1 Sep 2000 10:28:33 -0700 (PDT) Date: Fri, 1 Sep 2000 10:28:33 -0700 (PDT) From: murthy@gaea.engr.sgi.com (Chandrasekhar Murthy) Message-Id: <200009011728.KAA37617@gaea.engr.sgi.com> To: pro64-support@oss.sgi.com, hardnett@cc.gatech.edu Subject: Re: The .B files Sender: owner-pro64-support@oss.sgi.com Precedence: bulk Return-Path: X-Orcpt: rfc822;pro64-support-outgoing The ".B" files are generated by the frontends. Murthy From owner-pro64-support@oss.sgi.com Fri Sep 1 10:33:29 2000 Received: by oss.sgi.com id ; Fri, 1 Sep 2000 10:33:09 -0700 Received: from deliverator.sgi.com ([204.94.214.10]:57948 "EHLO deliverator.sgi.com") by oss.sgi.com with ESMTP id ; Fri, 1 Sep 2000 10:33:03 -0700 Received: from sgihud.hudson.sgi.com (sgihud.hudson.sgi.com [169.238.41.4]) by deliverator.sgi.com (980309.SGI.8.8.8-aspam-6.2/980310.SGI-aspam) via ESMTP id KAA06772 for ; Fri, 1 Sep 2000 10:25:24 -0700 (PDT) mail_from (lesniak@sgihud.hudson.sgi.com) Received: (from lesniak@localhost) by sgihud.hudson.sgi.com (980427.SGI.8.8.8/980728.SGI.AUTOCF) id NAA40733; Fri, 1 Sep 2000 13:30:20 -0400 (EDT) Date: Fri, 1 Sep 2000 13:30:20 -0400 (EDT) From: lesniak@sgihud.hudson.sgi.com (Ken Lesniak) Message-Id: <200009011730.NAA40733@sgihud.hudson.sgi.com> To: pro64-support@oss.sgi.com, hardnett@cc.gatech.edu Subject: Re: The .B files Reply-To: lesniak@sgihud.hudson.sgi.com Sender: owner-pro64-support@oss.sgi.com Precedence: bulk Return-Path: X-Orcpt: rfc822;pro64-support-outgoing >From: Charles Hardnett > >has anyone used the -keep option for the compiler. If so, there is a >.B file that is generated. Is there anyone that knows at what stage >during the compilation this file is generated? For all practical purposes, you can say it's the output of the frontends. Ken From owner-pro64-support@oss.sgi.com Fri Sep 1 11:36:59 2000 Received: by oss.sgi.com id ; Fri, 1 Sep 2000 11:36:49 -0700 Received: from deliverator.sgi.com ([204.94.214.10]:30837 "EHLO deliverator.sgi.com") by oss.sgi.com with ESMTP id ; Fri, 1 Sep 2000 11:36:28 -0700 Received: from rohi.engr.sgi.com (rohi.engr.sgi.com [130.62.180.74]) by deliverator.sgi.com (980309.SGI.8.8.8-aspam-6.2/980310.SGI-aspam) via ESMTP id LAA18139 for ; Fri, 1 Sep 2000 11:28:49 -0700 (PDT) mail_from (mpm@rohi.engr.sgi.com) Received: (from mpm@localhost) by rohi.engr.sgi.com (980427.SGI.8.8.8/970903.SGI.AUTOCF) id LAA18272; Fri, 1 Sep 2000 11:34:00 -0700 (PDT) Date: Fri, 1 Sep 2000 11:34:00 -0700 (PDT) From: mpm@rohi.engr.sgi.com (Michael Murphy) Message-Id: <200009011834.LAA18272@rohi.engr.sgi.com> To: pro64-support@oss.sgi.com, hardnett@cc.gatech.edu Subject: Re: The .B files Sender: owner-pro64-support@oss.sgi.com Precedence: bulk Return-Path: X-Orcpt: rfc822;pro64-support-outgoing FYI, the utility ir_b2a (in /usr/local/bin) will print out the .B (and .I) files in ascii format. -- Mike Murphy -- mpm@sgi.com -- quote of the day: -- "F x S = k. The product of Freedom and Security is a constant. -- To gain more freedom of thought and/or action you must give up -- some security, and vice versa." (Niven's Fourth) From owner-pro64-support@oss.sgi.com Tue Sep 5 06:20:45 2000 Received: by oss.sgi.com id ; Tue, 5 Sep 2000 06:20:36 -0700 Received: from beta.dmz-eu.st.com ([164.129.1.35]:27357 "HELO beta.dmz-eu.st.com") by oss.sgi.com with SMTP id ; Tue, 5 Sep 2000 06:20:13 -0700 Received: from zeta.dmz-eu.st.com (zeta.dmz-eu.st.com [164.129.230.9]) by beta.dmz-eu.st.com (STMicroelectronics) with ESMTP id 7CCBC4A24 for ; Tue, 5 Sep 2000 13:20:11 +0000 (GMT) Received: by zeta.dmz-eu.st.com (STMicroelectronics, from userid 0) id 3E13848C5; Tue, 5 Sep 2000 13:20:20 +0000 (GMT) Received: from eux100.sgp.st.com (localhost [127.0.0.1]) by zeta.dmz-eu.st.com (STMicroelectronics) with ESMTP id 19E891841 for ; Tue, 5 Sep 2000 13:20:20 +0000 (GMT) Received: from st.com (lod30.gnb.st.com [164.129.117.127]) by eux100.sgp.st.com (8.8.6 (PHNE_17190)/8.8.6) with ESMTP id PAA25570 for ; Tue, 5 Sep 2000 15:20:04 +0200 (METDST) Message-ID: <39B4F326.C695C14E@st.com> Date: Tue, 05 Sep 2000 15:20:38 +0200 From: Arthur Stoutchinin Reply-To: Arthur.Stoutchinin@st.com Organization: STMicroelectronics X-Mailer: Mozilla 4.7 [en] (X11; I; SunOS 5.5.1 sun4u) X-Accept-Language: ru, en MIME-Version: 1.0 To: pro64-support@oss.sgi.com Subject: license terms Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Sender: owner-pro64-support@oss.sgi.com Precedence: bulk Return-Path: X-Orcpt: rfc822;pro64-support-outgoing Good day, I've been looking at the Pro64 page and I could not find the terms for using the Pro64 software. Is there a license agreement clearly written somewhere ? Is it the same as the Open Source license ? We would like to use the Pro64 technology as the basis for our own development but are not sure what we can and can not do with it. thanks a lot, Arthur Stoutchinin From owner-pro64-support@oss.sgi.com Fri Sep 8 05:13:11 2000 Received: by oss.sgi.com id ; Fri, 8 Sep 2000 05:13:02 -0700 Received: from d06lmsgate-2.uk.ibm.com ([195.212.29.2]:7323 "EHLO d06lmsgate-2.uk.ibm.com") by oss.sgi.com with ESMTP id ; Fri, 8 Sep 2000 05:12:46 -0700 Received: from d06relay02.portsmouth.uk.ibm.com (d06relay02.portsmouth.uk.ibm.com [9.166.84.148]) by d06lmsgate-2.uk.ibm.com (1.0.0) with ESMTP id MAA54264 for ; Fri, 8 Sep 2000 12:59:57 +0100 From: SCHAN@uk.ibm.com Received: from d06mta07.portsmouth.uk.ibm.com (d06mta07_cs0 [9.180.35.5]) by d06relay02.portsmouth.uk.ibm.com (8.8.8m3/NCO v4.93) with SMTP id NAA75940 for ; Fri, 8 Sep 2000 13:12:29 +0100 Received: by d06mta07.portsmouth.uk.ibm.com(Lotus SMTP MTA v4.6.5 (863.2 5-20-1999)) id 80256954.00430B4F ; Fri, 8 Sep 2000 13:12:18 +0100 X-Lotus-FromDomain: IBMGB To: pro64-support@oss.sgi.com Message-ID: <80256954.0042ED01.00@d06mta07.portsmouth.uk.ibm.com> Date: Fri, 8 Sep 2000 13:10:56 +0100 Subject: Building cross compiler Mime-Version: 1.0 Content-type: text/plain; charset=us-ascii Content-Disposition: inline Sender: owner-pro64-support@oss.sgi.com Precedence: bulk Return-Path: X-Orcpt: rfc822;pro64-support-outgoing Hello, I am trying to build the pro64 compiler as a cross compiler from the source - I don't have the nue installed but I have a proper IA64 machine running Linux working, thus I can get all the headers using NFS mounts. When I tried to build the compiler (extracted the source file, do make build -f Make.cross as suggested by the README.src) it fails and seems to be looking for a fortran compiler under /usr/ia32-sgi-linux/bin/f90 - anyone can give me any hint what's missing? I don't really need the fortran compiler - can I in fact switch it off? Where in the makefiles does it specify the system header directory, or where does the makefiles look for the system header? I suppose I can nfs mount the header files from my Linux/IA64 box into that directory instead of using the nue? (It takes a while to d/l nue around here.....) Thanks for your help. -x- Sunny Chan Java Technology Center, Mail Point 146, IBM UK, Hursley Park, Winchester, England SO21 2JN Notes ID: Sunny Chan/UK/IBM Email: schan@uk.ibm.com Telephones: Internal: 246762 External: 01962 816762 Mobile: 07887-704088 From owner-pro64-support@oss.sgi.com Fri Sep 8 07:18:33 2000 Received: by oss.sgi.com id ; Fri, 8 Sep 2000 07:18:23 -0700 Received: from mail.virtualunlimited.com ([195.109.158.2]:56986 "EHLO mail.virtualunlimited.com") by oss.sgi.com with ESMTP id ; Fri, 8 Sep 2000 07:18:13 -0700 Received: from virtual-unlimited.com (ns2.virtualunlimited.com [195.109.158.31]) by mail.virtualunlimited.com (8.10.0/8.10.0) with ESMTP id e88EI9f11934; Fri, 8 Sep 2000 16:18:10 +0200 (MET DST) Message-ID: <39B8F4A1.A00FE2D2@virtual-unlimited.com> Date: Fri, 08 Sep 2000 16:16:01 +0200 From: Bob Deblier Organization: Virtual Unlimited X-Mailer: Mozilla 4.75 [en] (X11; U; SunOS 5.7 sun4u) X-Accept-Language: en MIME-Version: 1.0 To: pro64-support@oss.sgi.com CC: schwab@suse.de Subject: SuSE IA-64 compile farm installed Pro64 compilers Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Sender: owner-pro64-support@oss.sgi.com Precedence: bulk Return-Path: X-Orcpt: rfc822;pro64-support-outgoing Dear, SuSE has set up an Itanium compile farm, on which I have been granted access as a developer. At my suggestion they have installed your Pro64 compiler. They did the install from the RPM on your site. I've already figured out that the compiler won't work without specifying the '-64' flag, but when trying to compile a simple file I run into the following problem: deblier@ia64-1:~/beecrypt-1.1.1 > sgicc -64 -o test test.c > gfec: Invalid option `el' For your information, the uname and version info: deblier@ia64-1:~ > uname -a Linux ia64-1 2.4.0-test6-SMP #1 SMP Wed Aug 16 12:35:54 GMT 2000 ia64 unknown deblier@ia64-1:~ > sgicc -version SGIcc Compilers: Version 0.01.0-10 Can you offer any suggestions as to what may be going wrong? Is there any way to set the compiler up in such a way that it doesn't require the -64 flag (i.e. make that mode the default)? Many thanks Bob Deblier System Engineer Virtual Unlimited From owner-pro64-support@oss.sgi.com Fri Sep 8 09:47:44 2000 Received: by oss.sgi.com id ; Fri, 8 Sep 2000 09:47:24 -0700 Received: from deliverator.sgi.com ([204.94.214.10]:60021 "EHLO deliverator.sgi.com") by oss.sgi.com with ESMTP id ; Fri, 8 Sep 2000 09:47:02 -0700 Received: from cchkms.engr.sgi.com (cchkms.engr.sgi.com [130.62.180.48]) by deliverator.sgi.com (980309.SGI.8.8.8-aspam-6.2/980310.SGI-aspam) via ESMTP id JAA02035 for ; Fri, 8 Sep 2000 09:39:22 -0700 (PDT) mail_from (rat@cchkms.engr.sgi.com) Received: (from rat@localhost) by cchkms.engr.sgi.com (980427.SGI.8.8.8/970903.SGI.AUTOCF) id JAA86484; Fri, 8 Sep 2000 09:44:05 -0700 (PDT) From: "Ross A. Towle" Message-Id: <10009080944.ZM86645@cchkms.engr.sgi.com> Date: Fri, 8 Sep 2000 09:44:04 -0700 In-Reply-To: SCHAN@uk.ibm.com "Building cross compiler" (Sep 8, 1:10pm) References: <80256954.0042ED01.00@d06mta07.portsmouth.uk.ibm.com> X-Mailer: Z-Mail (3.2.3 08feb96 MediaMail) To: SCHAN@uk.ibm.com, pro64-support@oss.sgi.com Subject: Re: Building cross compiler Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Sender: owner-pro64-support@oss.sgi.com Precedence: bulk Return-Path: X-Orcpt: rfc822;pro64-support-outgoing So what do you mean as a cross compiler? !. Host on Linux/IA32 & IA64 as an IA32-executable Target for Linux/IA64 Just goto the targia32_ia64 directory and type gmake to build executables and goto the targia64 directory and type gmake to build libraries You will need a Linux/IA32 Fortran90 to build the Fortran90 compiler. If you don't want a Fortran compiler then you can skip this. Don't mount or add any Linux/IA64 headers. The setup works correctly as is. Remember that we are build IA-32 executables and will take advantage of the IA-32 compatability mode on Itanium. 2. Host on Linux/IA64 as an IA64 executable Target for Linux/IA64 Will be available soon -Ross From owner-pro64-support@oss.sgi.com Fri Sep 8 09:51:43 2000 Received: by oss.sgi.com id ; Fri, 8 Sep 2000 09:51:24 -0700 Received: from pneumatic-tube.sgi.com ([204.94.214.22]:24408 "EHLO pneumatic-tube.sgi.com") by oss.sgi.com with ESMTP id ; Fri, 8 Sep 2000 09:51:09 -0700 Received: from cchkms.engr.sgi.com (cchkms.engr.sgi.com [130.62.180.48]) by pneumatic-tube.sgi.com (980327.SGI.8.8.8-aspam/980310.SGI-aspam) via ESMTP id JAA03389 for ; Fri, 8 Sep 2000 09:57:44 -0700 (PDT) mail_from (rat@cchkms.engr.sgi.com) Received: (from rat@localhost) by cchkms.engr.sgi.com (980427.SGI.8.8.8/970903.SGI.AUTOCF) id JAA48752; Fri, 8 Sep 2000 09:49:35 -0700 (PDT) From: "Ross A. Towle" Message-Id: <10009080949.ZM85505@cchkms.engr.sgi.com> Date: Fri, 8 Sep 2000 09:49:34 -0700 In-Reply-To: Bob Deblier "SuSE IA-64 compile farm installed Pro64 compilers" (Sep 8, 4:16pm) References: <39B8F4A1.A00FE2D2@virtual-unlimited.com> X-Mailer: Z-Mail (3.2.3 08feb96 MediaMail) To: Bob Deblier , pro64-support@oss.sgi.com Subject: Re: SuSE IA-64 compile farm installed Pro64 compilers Cc: schwab@suse.de Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Sender: owner-pro64-support@oss.sgi.com Precedence: bulk Return-Path: X-Orcpt: rfc822;pro64-support-outgoing Please don't use -64. We are removing that option. It is a way for us to generate MIPS4 ISA with a 64bit ABI. Just use "sgicc -o test test.c". That is what I do. If you find that you are getting many segfaults/crashes with the compiler, please be aware that the compiler is fine. The compiler is stressing the IA32 compatability mode of Itanium and Linux. The binaries work fine on an IA32 box. See some of the threads in the ia64-tools and ia64-kernel mailing lists for more details. -Ross From owner-pro64-support@oss.sgi.com Fri Sep 8 10:04:54 2000 Received: by oss.sgi.com id ; Fri, 8 Sep 2000 10:04:44 -0700 Received: from d06lmsgate.uk.ibm.com ([195.212.29.1]:30607 "EHLO d06lmsgate.uk.ibm.COM") by oss.sgi.com with ESMTP id ; Fri, 8 Sep 2000 10:04:26 -0700 Received: from d06relay01.portsmouth.uk.ibm.com (d06relay01.portsmouth.uk.ibm.com [9.166.84.147]) by d06lmsgate.uk.ibm.COM (1.0.0) with ESMTP id RAA127920; Fri, 8 Sep 2000 17:51:27 +0100 From: SCHAN@uk.ibm.com Received: from d06mta07.portsmouth.uk.ibm.com (d06mta07_cs0 [9.180.35.5]) by d06relay01.portsmouth.uk.ibm.com (8.8.8m3/NCO v4.93) with SMTP id SAA36456; Fri, 8 Sep 2000 18:04:07 +0100 Received: by d06mta07.portsmouth.uk.ibm.com(Lotus SMTP MTA v4.6.5 (863.2 5-20-1999)) id 80256954.005DC19E ; Fri, 8 Sep 2000 18:04:04 +0100 X-Lotus-FromDomain: IBMGB To: "Ross A. Towle" , pro64-support@oss.sgi.com Message-ID: <80256954.005DC01C.00@d06mta07.portsmouth.uk.ibm.com> Date: Fri, 8 Sep 2000 18:03:57 +0100 Subject: Re: Building cross compiler Mime-Version: 1.0 Content-type: text/plain; charset=us-ascii Content-Disposition: inline Sender: owner-pro64-support@oss.sgi.com Precedence: bulk Return-Path: X-Orcpt: rfc822;pro64-support-outgoing Hello Ross, Well, when I say cross compiler I mean option 1. The impression I have got from README.src is that if you need to build the entire compiler in IA32 you will need the header files (from nue)- I tried to do make -f Make.cross it will just die and say it can't find the fortran compiler. I then follow the instruction to build each compoent by going to each directory under taria32_ia64_nodebug/ - does the g++ compiler from Redhat 6.2 (egcs-2.91.66) works for compiling? seems like some of the C++ function is missing when I tried to compile gccfe Does the binary rpm files you provided on the website compiles WITHOUT nue? I have a real IA64 box to test the binary the compiler produce. Thanks for your help! -x- Sunny Chan Java Technology Center, Mail Point 146, IBM UK, Hursley Park, Winchester, England SO21 2JN Notes ID: Sunny Chan/UK/IBM Email: schan@uk.ibm.com Telephones: Internal: 246762 External: 01962 816762 Mobile: 07887-704088 "Ross A. Towle" on 08/09/2000 17:44:04 Please respond to "Ross A. Towle" To: Sunny Chan/UK/IBM@IBMGB, pro64-support@oss.sgi.com cc: Subject: Re: Building cross compiler So what do you mean as a cross compiler? !. Host on Linux/IA32 & IA64 as an IA32-executable Target for Linux/IA64 Just goto the targia32_ia64 directory and type gmake to build executables and goto the targia64 directory and type gmake to build libraries You will need a Linux/IA32 Fortran90 to build the Fortran90 compiler. If you don't want a Fortran compiler then you can skip this. Don't mount or add any Linux/IA64 headers. The setup works correctly as is. Remember that we are build IA-32 executables and will take advantage of the IA-32 compatability mode on Itanium. 2. Host on Linux/IA64 as an IA64 executable Target for Linux/IA64 Will be available soon -Ross From owner-pro64-support@oss.sgi.com Fri Sep 8 10:12:44 2000 Received: by oss.sgi.com id ; Fri, 8 Sep 2000 10:12:24 -0700 Received: from [209.220.25.99] ([209.220.25.99]:476 "EHLO chamfs.chameleonsystems.com") by oss.sgi.com with ESMTP id ; Fri, 8 Sep 2000 10:12:15 -0700 Received: from ocelot.chameleonsystems.com (ocelot.chameleonsystems.com [192.168.1.125]) by chamfs.chameleonsystems.com (8.9.3/8.9.3) with ESMTP id KAA04524; Fri, 8 Sep 2000 10:11:01 -0700 (PDT) Date: Fri, 8 Sep 2000 10:11:00 -0700 (PDT) From: Xinan Tang To: SCHAN@uk.ibm.com cc: "Ross A. Towle" , pro64-support@oss.sgi.com Subject: Re: Building cross compiler In-Reply-To: <80256954.005DC01C.00@d06mta07.portsmouth.uk.ibm.com> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-pro64-support@oss.sgi.com Precedence: bulk Return-Path: X-Orcpt: rfc822;pro64-support-outgoing Hi On Fri, 8 Sep 2000 SCHAN@uk.ibm.com wrote: > make -f Make.cross it will just die and say it can't find the fortran > compiler. I then follow the instruction to build each compoent by going to That is easy to fix. Simply comment out the line appeared in CROSS_COMPONENTS = \ ... \ # $(CROSS_BUILD_DIR)/cray90/sgi/mfee90 Then, the fortran part won't be touched. --Tang From owner-pro64-support@oss.sgi.com Fri Sep 8 10:16:44 2000 Received: by oss.sgi.com id ; Fri, 8 Sep 2000 10:16:24 -0700 Received: from cosrel1.hp.com ([156.153.255.170]:27155 "HELO cosrel1.hp.com") by oss.sgi.com with SMTP id ; Fri, 8 Sep 2000 10:16:20 -0700 Received: from hpfctwc.fc.hp.com (hpfctwc.fc.hp.com [15.6.247.128]) by cosrel1.hp.com (Postfix) with ESMTP id 1042E57B; Fri, 8 Sep 2000 11:16:28 -0600 (MDT) Received: from fc.hp.com (athena.nsr.hp.com [15.116.178.38]) by hpfctwc.fc.hp.com with ESMTP (8.8.6 (PHNE_14041)/8.8.6 SMKit7.02) id LAA02584; Fri, 8 Sep 2000 11:22:56 -0600 (MDT) Message-ID: <39B91F26.A4FE845B@fc.hp.com> Date: Fri, 08 Sep 2000 11:17:26 -0600 From: Christopher Worley Organization: Hewlett-Packard Laboratories X-Mailer: Mozilla 4.72 [en] (Windows NT 5.0; U) X-Accept-Language: en MIME-Version: 1.0 To: SCHAN@uk.ibm.com Cc: pro64-support@oss.sgi.com Subject: Re: Building cross compiler References: <80256954.005DC01C.00@d06mta07.portsmouth.uk.ibm.com> Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Sender: owner-pro64-support@oss.sgi.com Precedence: bulk Return-Path: X-Orcpt: rfc822;pro64-support-outgoing Sunny; You need to have g77 installed and build the cross comiler without being in the nue shell. I was able to do it with no problems. Christopher -- Christopher Worley Software Design Engineer Hewlett-Packard Laboratories E-Mail: cworley@fc.hp.com Phone: (720) 528-9500 Telnet: (970) 898-0617 FAX: (720) 528-9499 From owner-pro64-support@oss.sgi.com Fri Sep 8 11:46:35 2000 Received: by oss.sgi.com id ; Fri, 8 Sep 2000 11:46:16 -0700 Received: from pneumatic-tube.sgi.com ([204.94.214.22]:36197 "EHLO pneumatic-tube.sgi.com") by oss.sgi.com with ESMTP id ; Fri, 8 Sep 2000 11:45:59 -0700 Received: from cchkms.engr.sgi.com (cchkms.engr.sgi.com [130.62.180.48]) by pneumatic-tube.sgi.com (980327.SGI.8.8.8-aspam/980310.SGI-aspam) via ESMTP id LAA09365 for ; Fri, 8 Sep 2000 11:52:33 -0700 (PDT) mail_from (rat@cchkms.engr.sgi.com) Received: (from rat@localhost) by cchkms.engr.sgi.com (980427.SGI.8.8.8/970903.SGI.AUTOCF) id LAA01261; Fri, 8 Sep 2000 11:44:24 -0700 (PDT) From: "Ross A. Towle" Message-Id: <10009081144.ZM1236@cchkms.engr.sgi.com> Date: Fri, 8 Sep 2000 11:44:22 -0700 In-Reply-To: SCHAN@uk.ibm.com "Re: Building cross compiler" (Sep 8, 6:03pm) References: <80256954.005DC01C.00@d06mta07.portsmouth.uk.ibm.com> X-Mailer: Z-Mail (3.2.3 08feb96 MediaMail) To: SCHAN@uk.ibm.com, pro64-support@oss.sgi.com Subject: Re: Building cross compiler Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Sender: owner-pro64-support@oss.sgi.com Precedence: bulk Return-Path: X-Orcpt: rfc822;pro64-support-outgoing It is not building the compilers that needs the IA64 headers but rather when you use the compilers. That is when you need the headers. The binary rpms can be installed 1. under NUE which provides the IA64 headers and libraries also as, ld, nm, ar or 2. on IA64 box running a Linux distribution which provides the IA64 headers, libraries, as, ld, nm, ar. If you go this route be aware of various problems with IA32 compatability mode that has been discussed on the ia64-kernel and ia64-tools mailing list. When we build we use gcc version 2.96 19991114 (see http://oss.sgi.com/projects/Pro64/READMEsrc.html) on IA32 Linux boxes running RedHat 6.1. We have not tried to build with other versions of gcc. -Ross From owner-pro64-support@oss.sgi.com Fri Sep 8 19:24:37 2000 Received: by oss.sgi.com id ; Fri, 8 Sep 2000 19:24:26 -0700 Received: from pneumatic-tube.sgi.com ([204.94.214.22]:12305 "EHLO pneumatic-tube.sgi.com") by oss.sgi.com with ESMTP id ; Fri, 8 Sep 2000 19:24:16 -0700 Received: from cchkms.engr.sgi.com (cchkms.engr.sgi.com [130.62.180.48]) by pneumatic-tube.sgi.com (980327.SGI.8.8.8-aspam/980310.SGI-aspam) via ESMTP id TAA02067 for ; Fri, 8 Sep 2000 19:30:50 -0700 (PDT) mail_from (rat@cchkms.engr.sgi.com) Received: (from rat@localhost) by cchkms.engr.sgi.com (980427.SGI.8.8.8/970903.SGI.AUTOCF) id TAA04901 for pro64-support@oss.sgi.com; Fri, 8 Sep 2000 19:22:32 -0700 (PDT) From: "Ross A. Towle" Message-Id: <10009081922.ZM4965@cchkms.engr.sgi.com> Date: Fri, 8 Sep 2000 19:22:31 -0700 X-Mailer: Z-Mail (3.2.3 08feb96 MediaMail) To: pro64-support@oss.sgi.com Subject: SGI Pro64 compilers causing system to need reboot:a fix Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Sender: owner-pro64-support@oss.sgi.com Precedence: bulk Return-Path: X-Orcpt: rfc822;pro64-support-outgoing I would recommend that you obtain workaround-1.1.efi which was part of a message by Rohit Seth on September 1 to the ia64-tools mailing list. When we applied it to a Lion box with A3 parts, we found the need to reboot due to compilations hanging the system went away. -Ross From owner-pro64-support@oss.sgi.com Tue Sep 12 11:09:42 2000 Received: by oss.sgi.com id ; Tue, 12 Sep 2000 11:09:33 -0700 Received: from louie.udel.edu ([128.175.7.39]:19392 "HELO mail.eecis.udel.edu") by oss.sgi.com with SMTP id ; Tue, 12 Sep 2000 11:09:12 -0700 Received: from ren.eecis.udel.edu by mail.eecis.udel.edu id aa11575; 12 Sep 2000 14:07 EDT Date: Tue, 12 Sep 2000 14:07:05 -0400 (EDT) From: Ziang Hu To: pro64-support@oss.sgi.com Subject: about LNO Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-pro64-support@oss.sgi.com Precedence: bulk Return-Path: X-Orcpt: rfc822;pro64-support-outgoing Hi, I tried to test the effect of LNO with Pro64. The code is as follows: ... for (i=0; i<1000; i++) for (j=0; j<1000; j++) a[i][j] = a[i][j] + a[i-1][j]; ... With -O3, no loop interchange happened. How about LNO with C programs ? Thanks Ziang From owner-pro64-support@oss.sgi.com Tue Sep 12 11:14:02 2000 Received: by oss.sgi.com id ; Tue, 12 Sep 2000 11:13:52 -0700 Received: from deliverator.sgi.com ([204.94.214.10]:7020 "EHLO deliverator.sgi.com") by oss.sgi.com with ESMTP id ; Tue, 12 Sep 2000 11:13:46 -0700 Received: from cchkms.engr.sgi.com (cchkms.engr.sgi.com [130.62.180.48]) by deliverator.sgi.com (980309.SGI.8.8.8-aspam-6.2/980310.SGI-aspam) via ESMTP id LAA03756 for ; Tue, 12 Sep 2000 11:06:05 -0700 (PDT) mail_from (rat@cchkms.engr.sgi.com) Received: (from rat@localhost) by cchkms.engr.sgi.com (980427.SGI.8.8.8/970903.SGI.AUTOCF) id LAA40179; Tue, 12 Sep 2000 11:11:58 -0700 (PDT) From: "Ross A. Towle" Message-Id: <10009121111.ZM40419@cchkms.engr.sgi.com> Date: Tue, 12 Sep 2000 11:11:57 -0700 In-Reply-To: Ziang Hu "about LNO" (Sep 12, 2:07pm) References: X-Mailer: Z-Mail (3.2.3 08feb96 MediaMail) To: Ziang Hu , pro64-support@oss.sgi.com Subject: Re: about LNO Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Sender: owner-pro64-support@oss.sgi.com Precedence: bulk Return-Path: X-Orcpt: rfc822;pro64-support-outgoing There will be a switch to a new c front end later this year. At that time there will be more LNO optimizations performed on c and c++ programs. -Ross From owner-pro64-support@oss.sgi.com Tue Sep 12 16:30:44 2000 Received: by oss.sgi.com id ; Tue, 12 Sep 2000 16:30:24 -0700 Received: from [38.170.141.29] ([38.170.141.29]:25328 "EHLO mail-in.hq.tensilica.com") by oss.sgi.com with ESMTP id ; Tue, 12 Sep 2000 16:29:58 -0700 Received: from gobi-pc (gobi-pc.hq.tensilica.com [192.168.10.121]) by mail-in.hq.tensilica.com (8.9.3/8.9.3) with SMTP id QAA05487; Tue, 12 Sep 2000 16:29:51 -0700 Received: by localhost with Microsoft MAPI; Tue, 12 Sep 2000 16:29:38 -0700 Message-ID: <01C01CD6.A4E64F70.tu@tensilica.com> From: Peng Tu Reply-To: "tu@tensilica.com" To: "'Ziang Hu'" , "pro64-support@oss.sgi.com" Subject: RE: about LNO Date: Tue, 12 Sep 2000 16:29:36 -0700 Organization: Tensilica Inc X-Mailer: Microsoft Internet E-mail/MAPI - 8.0.0.4211 MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Sender: owner-pro64-support@oss.sgi.com Precedence: bulk Return-Path: X-Orcpt: rfc822;pro64-support-outgoing For this C program, I don't see why it is beneficial to interchange the loop because the inner loop is already stride-1 (C is row-major). It does need an unroll-and-jam transformation to capture the reuse of the outer loop though. Maybe the LNO machine model is not up-to-date with the IA64. A useful flag to try for C program is -OPT:alias=restrict if the array is an input parameter. Peng. On Tuesday, September 12, 2000 11:07 AM, Ziang Hu [SMTP:hu@mail.eecis.udel.edu] wrote: > > Hi, > > I tried to test the effect of LNO with Pro64. The code is as follows: > > ... > for (i=0; i<1000; i++) > for (j=0; j<1000; j++) > a[i][j] = a[i][j] + a[i-1][j]; > ... > > With -O3, no loop interchange happened. > > How about LNO with C programs ? > > Thanks > > Ziang From owner-pro64-support@oss.sgi.com Tue Sep 12 17:44:15 2000 Received: by oss.sgi.com id ; Tue, 12 Sep 2000 17:44:06 -0700 Received: from louie.udel.edu ([128.175.7.39]:29673 "HELO mail.eecis.udel.edu") by oss.sgi.com with SMTP id ; Tue, 12 Sep 2000 17:43:47 -0700 Received: from louie.udel.edu by mail.eecis.udel.edu id aa16667; 12 Sep 2000 20:42 EDT Date: Tue, 12 Sep 2000 20:42:50 -0400 (EDT) From: Ziang Hu To: Peng Tu cc: 'Ziang Hu' , "pro64-support@oss.sgi.com" Subject: RE: about LNO In-Reply-To: <01C01CD6.A4E64F70.tu@tensilica.com> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-pro64-support@oss.sgi.com Precedence: bulk Return-Path: X-Orcpt: rfc822;pro64-support-outgoing I mean loop interchange. Generally, MipsPro will do loop interchange and tiling with this code. Thanks Hu On Tue, 12 Sep 2000, Peng Tu wrote: > For this C program, I don't see why it is beneficial to interchange > the loop because the inner loop is already stride-1 (C is row-major). > > It does need an unroll-and-jam transformation to capture the reuse > of the outer loop though. Maybe the LNO machine model is not up-to-date > with the IA64. > > A useful flag to try for C program is -OPT:alias=restrict if the array is > an input parameter. > > Peng. > > On Tuesday, September 12, 2000 11:07 AM, Ziang Hu [SMTP:hu@mail.eecis.udel.edu] wrote: > > > > Hi, > > > > I tried to test the effect of LNO with Pro64. The code is as follows: > > > > ... > > for (i=0; i<1000; i++) > > for (j=0; j<1000; j++) > > a[i][j] = a[i][j] + a[i-1][j]; > > ... > > > > With -O3, no loop interchange happened. > > > > How about LNO with C programs ? > > > > Thanks > > > > Ziang > From owner-pro64-support@oss.sgi.com Tue Sep 12 18:33:06 2000 Received: by oss.sgi.com id ; Tue, 12 Sep 2000 18:32:45 -0700 Received: from louie.udel.edu ([128.175.7.39]:54510 "HELO mail.eecis.udel.edu") by oss.sgi.com with SMTP id ; Tue, 12 Sep 2000 18:32:14 -0700 Received: from louie.udel.edu by mail.eecis.udel.edu id aa20528; 12 Sep 2000 21:31 EDT Date: Tue, 12 Sep 2000 21:31:51 -0400 (EDT) From: Ziang Hu To: Peng Tu cc: 'Ziang Hu' , "pro64-support@oss.sgi.com" Subject: RE: about LNO In-Reply-To: <01C01CD6.A4E64F70.tu@tensilica.com> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-pro64-support@oss.sgi.com Precedence: bulk Return-Path: X-Orcpt: rfc822;pro64-support-outgoing Sorry. I think my question should be as follows: 1. With MisPro on SGI machine, LNO does tiling and interchanged the loop (i and j) of the code, even it doesn't need to do loop interchange. 2. With Pro64, you do loop fission, instead of tiling. Can you explain the reason? Thanks Ziang Hu ECE Department, University of Delaware On Tue, 12 Sep 2000, Peng Tu wrote: > For this C program, I don't see why it is beneficial to interchange > the loop because the inner loop is already stride-1 (C is row-major). > > It does need an unroll-and-jam transformation to capture the reuse > of the outer loop though. Maybe the LNO machine model is not up-to-date > with the IA64. > > A useful flag to try for C program is -OPT:alias=restrict if the array is > an input parameter. > > Peng. > > On Tuesday, September 12, 2000 11:07 AM, Ziang Hu [SMTP:hu@mail.eecis.udel.edu] wrote: > > > > Hi, > > > > I tried to test the effect of LNO with Pro64. The code is as follows: > > > > ... > > for (i=0; i<1000; i++) > > for (j=0; j<1000; j++) > > a[i][j] = a[i][j] + a[i-1][j]; > > ... > > > > With -O3, no loop interchange happened. > > > > How about LNO with C programs ? > > > > Thanks > > > > Ziang > From owner-pro64-support@oss.sgi.com Tue Sep 12 18:42:35 2000 Received: by oss.sgi.com id ; Tue, 12 Sep 2000 18:42:25 -0700 Received: from pneumatic-tube.sgi.com ([204.94.214.22]:2624 "EHLO pneumatic-tube.sgi.com") by oss.sgi.com with ESMTP id ; Tue, 12 Sep 2000 18:42:10 -0700 Received: from cchkms.engr.sgi.com (cchkms.engr.sgi.com [130.62.180.48]) by pneumatic-tube.sgi.com (980327.SGI.8.8.8-aspam/980310.SGI-aspam) via ESMTP id SAA04438 for ; Tue, 12 Sep 2000 18:48:49 -0700 (PDT) mail_from (rat@cchkms.engr.sgi.com) Received: (from rat@localhost) by cchkms.engr.sgi.com (980427.SGI.8.8.8/970903.SGI.AUTOCF) id SAA41918; Tue, 12 Sep 2000 18:40:23 -0700 (PDT) From: "Ross A. Towle" Message-Id: <10009121840.ZM42676@cchkms.engr.sgi.com> Date: Tue, 12 Sep 2000 18:40:22 -0700 In-Reply-To: Ziang Hu "RE: about LNO" (Sep 12, 9:31pm) References: X-Mailer: Z-Mail (3.2.3 08feb96 MediaMail) To: Ziang Hu , Peng Tu Subject: Re: about LNO Cc: "pro64-support@oss.sgi.com" Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Sender: owner-pro64-support@oss.sgi.com Precedence: bulk Return-Path: X-Orcpt: rfc822;pro64-support-outgoing Please see my previous response. It will take front end work. -Ross From owner-pro64-support@oss.sgi.com Tue Sep 12 18:47:36 2000 Received: by oss.sgi.com id ; Tue, 12 Sep 2000 18:47:16 -0700 Received: from deliverator.sgi.com ([204.94.214.10]:25704 "EHLO deliverator.sgi.com") by oss.sgi.com with ESMTP id ; Tue, 12 Sep 2000 18:47:12 -0700 Received: from gaea.engr.sgi.com (gaea.engr.sgi.com [130.62.180.97]) by deliverator.sgi.com (980309.SGI.8.8.8-aspam-6.2/980310.SGI-aspam) via ESMTP id SAA03511 for ; Tue, 12 Sep 2000 18:39:30 -0700 (PDT) mail_from (murthy@sgi.com) Received: from sgi.com (localhost [127.0.0.1]) by gaea.engr.sgi.com (980427.SGI.8.8.8/970903.SGI.AUTOCF) via ESMTP id SAA98281; Tue, 12 Sep 2000 18:45:32 -0700 (PDT) Message-ID: <39BEDC3B.D8D0CF5A@sgi.com> Date: Tue, 12 Sep 2000 18:45:31 -0700 From: Chandrasekhar Murthy X-Mailer: Mozilla 4.51C-SGI [en] (X11; I; IRIX 6.5 IP32) X-Accept-Language: en MIME-Version: 1.0 To: Ziang Hu CC: Peng Tu , "pro64-support@oss.sgi.com" Subject: Re: about LNO References: Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Sender: owner-pro64-support@oss.sgi.com Precedence: bulk Return-Path: X-Orcpt: rfc822;pro64-support-outgoing Ziang Hu wrote: > > Sorry. I think my question should be as follows: > > 1. With MisPro on SGI machine, LNO does tiling and interchanged the loop > (i and j) of the code, even it doesn't need to do loop interchange. > > 2. With Pro64, you do loop fission, instead of tiling. > > Can you explain the reason? > > Thanks > > Ziang Hu > ECE Department, > University of Delaware > > On Tue, 12 Sep 2000, Peng Tu wrote: > > > For this C program, I don't see why it is beneficial to interchange > > the loop because the inner loop is already stride-1 (C is row-major). > > > > It does need an unroll-and-jam transformation to capture the reuse > > of the outer loop though. Maybe the LNO machine model is not up-to-date > > with the IA64. > > > > A useful flag to try for C program is -OPT:alias=restrict if the array is > > an input parameter. > > > > Peng. > > > > On Tuesday, September 12, 2000 11:07 AM, Ziang Hu [SMTP:hu@mail.eecis.udel.edu] wrote: > > > > > > Hi, > > > > > > I tried to test the effect of LNO with Pro64. The code is as follows: > > > > > > ... > > > for (i=0; i<1000; i++) > > > for (j=0; j<1000; j++) > > > a[i][j] = a[i][j] + a[i-1][j]; > > > ... > > > > > > With -O3, no loop interchange happened. > > > > > > How about LNO with C programs ? > > > > > > Thanks > > > > > > Ziang > > Some of the transformations which were done in the MipsPro C/C++ frontends have not yet been implemented in the Pro64 C/C++ frontends. As Ross mentioned in an earlier mail, these will be implemented in the release later this year. Murthy From owner-pro64-support@oss.sgi.com Wed Sep 13 08:12:40 2000 Received: by oss.sgi.com id ; Wed, 13 Sep 2000 08:12:20 -0700 Received: from deliverator.sgi.com ([204.94.214.10]:57195 "EHLO deliverator.sgi.com") by oss.sgi.com with ESMTP id ; Wed, 13 Sep 2000 08:11:51 -0700 Received: from sgihud.hudson.sgi.com (sgihud.hudson.sgi.com [169.238.41.4]) by deliverator.sgi.com (980309.SGI.8.8.8-aspam-6.2/980310.SGI-aspam) via ESMTP id IAA10455 for ; Wed, 13 Sep 2000 08:04:09 -0700 (PDT) mail_from (dlstephe@sgi.com) Received: from zippy.hudson.sgi.com (zippy.hudson.sgi.com [169.238.41.37]) by sgihud.hudson.sgi.com (980427.SGI.8.8.8/980728.SGI.AUTOCF) via ESMTP id LAA48112; Wed, 13 Sep 2000 11:09:11 -0400 (EDT) Received: from sgi.com (localhost [127.0.0.1]) by zippy.hudson.sgi.com (980427.SGI.8.8.8/980728.SGI.AUTOCF) via ESMTP id LAA36964; Wed, 13 Sep 2000 11:09:00 -0400 (EDT) Message-ID: <39BF988B.4EC84454@sgi.com> Date: Wed, 13 Sep 2000 11:08:59 -0400 From: David L Stephenson Organization: Strategic Software Organization -- Compilers X-Mailer: Mozilla 4.51C-SGI [en] (X11; I; IRIX 6.5 IP22) X-Accept-Language: en MIME-Version: 1.0 To: pro64-support@oss.sgi.com Subject: Re: about LNO References: <01C01CD6.A4E64F70.tu@tensilica.com> Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Sender: owner-pro64-support@oss.sgi.com Precedence: bulk Return-Path: X-Orcpt: rfc822;pro64-support-outgoing Peng Tu wrote: > For this C program, I don't see why it is beneficial to interchange > the loop because the inner loop is already stride-1 (C is row-major). Interchanging the loops allows one of the array loads to be removed: for (i=1; i<1000; i++) for (j=1; j<1000; j++) a[i][j] = a[i][j] + a[i-1][j]; becomes for (j=1; j<1000; j++) { t = a[i][0]; for (i=1; i<1000; i++) { a[i][j] = t = a[i][j] + t; } But see Ross Towle's response. -- David Stephenson http://reality.sgi.com/dlstephe_engr/ From owner-pro64-support@oss.sgi.com Wed Sep 13 09:49:30 2000 Received: by oss.sgi.com id ; Wed, 13 Sep 2000 09:49:20 -0700 Received: from [38.170.141.29] ([38.170.141.29]:39670 "EHLO mail-in.hq.tensilica.com") by oss.sgi.com with ESMTP id ; Wed, 13 Sep 2000 09:49:07 -0700 Received: from gobi-pc (gobi-pc.hq.tensilica.com [192.168.10.121]) by mail-in.hq.tensilica.com (8.9.3/8.9.3) with SMTP id JAA08364; Wed, 13 Sep 2000 09:47:59 -0700 Received: by localhost with Microsoft MAPI; Wed, 13 Sep 2000 09:47:50 -0700 Message-ID: <01C01D67.AE461780.tu@tensilica.com> From: Peng Tu Reply-To: "tu@tensilica.com" To: "'David L Stephenson'" , "pro64-support@oss.sgi.com" Subject: RE: about LNO Date: Wed, 13 Sep 2000 09:47:49 -0700 Organization: Tensilica Inc X-Mailer: Microsoft Internet E-mail/MAPI - 8.0.0.4211 MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Sender: owner-pro64-support@oss.sgi.com Precedence: bulk Return-Path: X-Orcpt: rfc822;pro64-support-outgoing My argument is: it is better to do unroll-and-jam (a.k.a. outer unroll in LNO), instead of interchange: for (i = 0; i < 1000; i = i + 4) { for (j = 1; j < 1000; j++) { a[i][j] = a[i][j] + a[i-1][j]; a[i+1][j] = a[i+1][j] + a[i][j]; a[i+2][j] = a[i+2][j] + a[i+1][j] a[i+3][j] = a[i+3][j] + a[i+2][j] } } This way, you get the benefit of reuse on the i-loop without losing the stride-1 access on the j-loop. Anyway, I don't see the whole program. Ross is probably right about the frontend issue. Peng. On Wednesday, September 13, 2000 8:09 AM, David L Stephenson [SMTP:dlstephe@sgi.com] wrote: > Peng Tu wrote: > > > For this C program, I don't see why it is beneficial to interchange > > the loop because the inner loop is already stride-1 (C is row-major). > > Interchanging the loops allows one of the array loads to be removed: > > for (i=1; i<1000; i++) > for (j=1; j<1000; j++) > a[i][j] = a[i][j] + a[i-1][j]; > > becomes > > for (j=1; j<1000; j++) { > t = a[i][0]; > for (i=1; i<1000; i++) { > a[i][j] = t = a[i][j] + t; > } > > But see Ross Towle's response. > > -- > David Stephenson http://reality.sgi.com/dlstephe_engr/ From owner-pro64-support@oss.sgi.com Wed Sep 13 16:46:34 2000 Received: by oss.sgi.com id ; Wed, 13 Sep 2000 16:46:24 -0700 Received: from thalia.fm.intel.com ([132.233.247.11]:40717 "EHLO thalia.fm.intel.com") by oss.sgi.com with ESMTP id ; Wed, 13 Sep 2000 16:46:15 -0700 Received: from SMTP (fmsmsxvs01-1.fm.intel.com [132.233.42.201]) by thalia.fm.intel.com (8.9.1a+p1/8.9.1/d: relay.m4,v 1.31 2000/08/22 00:15:13 dmccart Exp $) with SMTP id XAA09668 for ; Wed, 13 Sep 2000 23:47:05 GMT Received: from fmsmsx17.intel.com ([132.233.48.17]) by 132.233.48.201 (Norton AntiVirus for Internet Email Gateways 1.0) ; Wed, 13 Sep 2000 23:46:03 0000 (GMT) Received: by fmsmsx17.fm.intel.com with Internet Mail Service (5.5.2650.21) id ; Wed, 13 Sep 2000 16:46:01 -0700 Message-ID: <3677E033A5F3D211AC4000A0C96B53EB02E7008A@FMSMSX94> From: "Murty, Ravi" To: "'pro64-support@oss.sgi.com'" Subject: Problem invoking sgicc and sgi++ Date: Wed, 13 Sep 2000 16:45:53 -0700 MIME-Version: 1.0 X-Mailer: Internet Mail Service (5.5.2650.21) Content-Type: text/plain; charset="iso-8859-1" Sender: owner-pro64-support@oss.sgi.com Precedence: bulk Return-Path: X-Orcpt: rfc822;pro64-support-outgoing Hi, I used your rpm to install the latest release of the C++ compiler. However when I run - $ sgicc hello.c I get bash: /usr/bin/sgi : no such file or directory. I can see that sgicc is a link to a file called driver in the /lib directory. which sgicc points me to /usr/bin/sgicc.. Am I missing some configuration file or do I have to set some environment variables or something. Thanks, Ravi ------------------------------------------------------------------- Ravi Murty (503) 677 4233 Enterprise Server Group Ravi.Murty@intel.com Intel Corporation Portland, Oregon From owner-pro64-support@oss.sgi.com Wed Sep 13 17:01:34 2000 Received: by oss.sgi.com id ; Wed, 13 Sep 2000 17:01:24 -0700 Received: from deliverator.sgi.com ([204.94.214.10]:43 "EHLO deliverator.sgi.com") by oss.sgi.com with ESMTP id ; Wed, 13 Sep 2000 17:01:07 -0700 Received: from cchkms.engr.sgi.com (cchkms.engr.sgi.com [130.62.180.48]) by deliverator.sgi.com (980309.SGI.8.8.8-aspam-6.2/980310.SGI-aspam) via ESMTP id QAA10289 for ; Wed, 13 Sep 2000 16:53:26 -0700 (PDT) mail_from (rat@cchkms.engr.sgi.com) Received: (from rat@localhost) by cchkms.engr.sgi.com (980427.SGI.8.8.8/970903.SGI.AUTOCF) id QAA01454; Wed, 13 Sep 2000 16:58:05 -0700 (PDT) From: "Ross A. Towle" Message-Id: <10009131658.ZM1460@cchkms.engr.sgi.com> Date: Wed, 13 Sep 2000 16:58:05 -0700 In-Reply-To: "Murty, Ravi" "Problem invoking sgicc and sgi++" (Sep 13, 4:45pm) References: <3677E033A5F3D211AC4000A0C96B53EB02E7008A@FMSMSX94> X-Mailer: Z-Mail (3.2.3 08feb96 MediaMail) To: "Murty, Ravi" , "'pro64-support@oss.sgi.com '" Subject: Re: Problem invoking sgicc and sgi++ Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Sender: owner-pro64-support@oss.sgi.com Precedence: bulk Return-Path: X-Orcpt: rfc822;pro64-support-outgoing What kind of a system did you install it on? And what is the OS? I suspect that you are installing on an IA64 box with a very old Linux that does not support IA32-compatability mode. In that case you will have to get a newer Linux. -Ross From owner-pro64-support@oss.sgi.com Thu Sep 14 12:19:00 2000 Received: by oss.sgi.com id ; Thu, 14 Sep 2000 12:18:51 -0700 Received: from ganymede.or.intel.com ([134.134.248.3]:32010 "EHLO ganymede.or.intel.com") by oss.sgi.com with ESMTP id ; Thu, 14 Sep 2000 12:18:41 -0700 Received: from babel.co.intel.com (babel.co.intel.com [10.8.3.28]) by ganymede.or.intel.com (8.9.1a+p1/8.9.1/d: relay.m4,v 1.31 2000/08/22 00:15:13 dmccart Exp $) with ESMTP id MAA01770 for ; Thu, 14 Sep 2000 12:18:40 -0700 (PDT) Received: from co.intel.com (IDENT:cfleck@dude1.co.intel.com [10.8.90.37]) by babel.co.intel.com (8.9.1a/8.9.1/d: client.m4,v 1.3 1998/09/29 16:36:11 sedayao Exp sedayao $) with ESMTP id MAA18383 for ; Thu, 14 Sep 2000 12:18:32 -0700 (PDT) Message-ID: <39C0C4B4.832ED4F6@co.intel.com> Date: Thu, 14 Sep 2000 12:29:40 +0000 From: Chuck Fleckenstein Reply-To: cfleck@co.intel.com Organization: Intel X-Mailer: Mozilla 4.61 [en] (X11; U; Linux 2.2.13 i686) X-Accept-Language: en MIME-Version: 1.0 To: pro64-support@oss.sgi.com Subject: OpenMP support... Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Sender: owner-pro64-support@oss.sgi.com Precedence: bulk Return-Path: X-Orcpt: rfc822;pro64-support-outgoing Does the current IA-64 sgif90 support OpenMP ? If so, how do you enable it ? thanks, Chuck Fleckenstein cfleck@co.intel.com From owner-pro64-support@oss.sgi.com Thu Sep 14 13:30:11 2000 Received: by oss.sgi.com id ; Thu, 14 Sep 2000 13:29:51 -0700 Received: from deliverator.sgi.com ([204.94.214.10]:6194 "EHLO deliverator.sgi.com") by oss.sgi.com with ESMTP id ; Thu, 14 Sep 2000 13:29:41 -0700 Received: from cchkms.engr.sgi.com (cchkms.engr.sgi.com [130.62.180.48]) by deliverator.sgi.com (980309.SGI.8.8.8-aspam-6.2/980310.SGI-aspam) via ESMTP id NAA09773 for ; Thu, 14 Sep 2000 13:22:00 -0700 (PDT) mail_from (rat@cchkms.engr.sgi.com) Received: (from rat@localhost) by cchkms.engr.sgi.com (980427.SGI.8.8.8/970903.SGI.AUTOCF) id NAA03083; Thu, 14 Sep 2000 13:28:00 -0700 (PDT) From: "Ross A. Towle" Message-Id: <10009141328.ZM3091@cchkms.engr.sgi.com> Date: Thu, 14 Sep 2000 13:27:59 -0700 In-Reply-To: Chuck Fleckenstein "OpenMP support..." (Sep 14, 12:29pm) References: <39C0C4B4.832ED4F6@co.intel.com> X-Mailer: Z-Mail (3.2.3 08feb96 MediaMail) To: cfleck@co.intel.com, pro64-support@oss.sgi.com Subject: Re: OpenMP support... Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Sender: owner-pro64-support@oss.sgi.com Precedence: bulk Return-Path: X-Orcpt: rfc822;pro64-support-outgoing It is not available today. There is still more implementation work to be done. -Ross From owner-pro64-support@oss.sgi.com Fri Sep 15 07:45:25 2000 Received: by oss.sgi.com id ; Fri, 15 Sep 2000 07:45:15 -0700 Received: from e24.nc.us.ibm.com ([32.97.136.230]:47537 "EHLO e24.nc.us.ibm.com") by oss.sgi.com with ESMTP id ; Fri, 15 Sep 2000 07:44:52 -0700 Received: from southrelay02.raleigh.ibm.com (southrelay02.raleigh.ibm.com [9.37.3.209]) by e24.nc.us.ibm.com (8.9.3/8.9.3) with ESMTP id KAA10774 for ; Fri, 15 Sep 2000 10:46:55 -0500 From: gjertsen@us.ibm.com Received: from d54mta01.raleigh.ibm.com (d54mta01.raleigh.ibm.com [9.67.228.33]) by southrelay02.raleigh.ibm.com (8.8.8m3/NCO v4.93) with SMTP id KAA30962 for ; Fri, 15 Sep 2000 10:44:50 -0400 Received: by d54mta01.raleigh.ibm.com(Lotus SMTP MTA v4.6.5 (863.2 5-20-1999)) id 8525695B.0050FFC7 ; Fri, 15 Sep 2000 10:44:43 -0400 X-Lotus-FromDomain: IBMUS To: pro64-support@oss.sgi.com Message-ID: <8525695B.0050FE73.00@d54mta01.raleigh.ibm.com> Date: Fri, 15 Sep 2000 10:43:39 -0400 Subject: C++ variable argument problem with sgiCC Mime-Version: 1.0 Content-type: text/plain; charset=us-ascii Content-Disposition: inline Sender: owner-pro64-support@oss.sgi.com Precedence: bulk Return-Path: X-Orcpt: rfc822;pro64-support-outgoing I am getting an internal compiler error when using variable arguments in C++ (stuff in stdarg.h) with sgiCC. I've tried to generate the cpp output to get a better idea of what is going on, but it doesn't look like /usr/lib/gcc-lib/ia64-sgi-linux/sgicc-1.0/cpp is provided. Note that the same fragment of code is OK when I compile it with sgicc (and this also works OK with g++). The full details follow below. If there is a simple workaround I can do, then let me know. In addition, is there an archive web site that you keep with all of the postings to this list. I would actually prefer that to email and it would allow people to check and see if their problem was already solved. Thanks, --Rob ***** MACHINE SETUP ***** [gjertsen@clue5 dirlinux]$ sgiCC -v SGIcc Compilers: Version 0.01.0-10 [gjertsen@clue5 dirlinux]$ uname -a Linux clue5 2.4.0test7-000823-43 #1 Wed Sep 13 16:48:51 CDT 2000 ia64 unknown ***** SOURCE CODE ***** [gjertsen@clue5 dirlinux]$ more bobo.C #include #include int ktrace(const char *fmt, ...) { va_list vargs; va_start(vargs, fmt); return 1; } ***** INTERNAL COMPILER ERROR ***** [gjertsen@clue5 dirlinux]$ sgiCC -c bobo.C bobo.C: In function `int ktrace (const char *, ...)': bobo.C:8: could not convert `&vargs' to `void *&' :8: in passing argument 1 of `__builtin_stdarg_start (void *&, ...)' From owner-pro64-support@oss.sgi.com Fri Sep 15 11:38:20 2000 Received: by oss.sgi.com id ; Fri, 15 Sep 2000 11:38:10 -0700 Received: from pneumatic-tube.sgi.com ([204.94.214.22]:53860 "EHLO pneumatic-tube.sgi.com") by oss.sgi.com with ESMTP id ; Fri, 15 Sep 2000 11:37:51 -0700 Received: from cchkms.engr.sgi.com (cchkms.engr.sgi.com [130.62.180.48]) by pneumatic-tube.sgi.com (980327.SGI.8.8.8-aspam/980310.SGI-aspam) via ESMTP id LAA04782 for ; Fri, 15 Sep 2000 11:44:33 -0700 (PDT) mail_from (rat@cchkms.engr.sgi.com) Received: (from rat@localhost) by cchkms.engr.sgi.com (980427.SGI.8.8.8/970903.SGI.AUTOCF) id LAA04683; Fri, 15 Sep 2000 11:32:33 -0700 (PDT) From: "Ross A. Towle" Message-Id: <10009151132.ZM4711@cchkms.engr.sgi.com> Date: Fri, 15 Sep 2000 11:32:32 -0700 In-Reply-To: gjertsen@us.ibm.com "C++ variable argument problem with sgiCC" (Sep 15, 10:43am) References: <8525695B.0050FE73.00@d54mta01.raleigh.ibm.com> X-Mailer: Z-Mail (3.2.3 08feb96 MediaMail) To: gjertsen@us.ibm.com, pro64-support@oss.sgi.com Subject: Re: C++ variable argument problem with sgiCC Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Sender: owner-pro64-support@oss.sgi.com Precedence: bulk Return-Path: X-Orcpt: rfc822;pro64-support-outgoing You need to pickup update.0828.tar.gz from http://oss.sgi.com/projects/Pro64/download which will fix your vararg problem -Ross From owner-pro64-support@oss.sgi.com Tue Sep 19 16:07:55 2000 Received: by oss.sgi.com id ; Tue, 19 Sep 2000 16:07:45 -0700 Received: from [38.170.141.29] ([38.170.141.29]:54772 "EHLO heart.hq.tensilica.com") by oss.sgi.com with ESMTP id ; Tue, 19 Sep 2000 16:07:37 -0700 Received: (from goodwin@localhost) by heart.hq.tensilica.com (8.9.3/8.9.3) id QAA30560; Tue, 19 Sep 2000 16:07:37 -0700 X-Authentication-Warning: heart.hq.tensilica.com: goodwin set sender to goodwin@tensilica.com using -f From: David Goodwin MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Message-ID: <14791.61881.466122.621594@heart.hq.tensilica.com> Date: Tue, 19 Sep 2000 16:07:37 -0700 (PDT) To: pro64-support@oss.sgi.com Cc: goodwin@tensilica.com Subject: Bad fe whirl generated for struct initialization X-Mailer: VM 6.75 under Emacs 20.5.1 Reply-to: goodwin@tensilica.com Sender: owner-pro64-support@oss.sgi.com Precedence: bulk Return-Path: X-Orcpt: rfc822;pro64-support-outgoing The following function should set s1.m3 to 0 since there is not an initializer for it. >From K&R 2e, p. 219, middle of the page: "The initializer for a structure is either an expression of the same type, or a brace-enclosed list of initializers for its members in order. Unnamed bif-field members are ignored, and are not initialized. If there are fewer initializers in the list than members of the structure, the trailing members are initialized with 0. ..." void finit (void) { struct s1 { unsigned m1:2; unsigned :2; unsigned m2:2; unsigned m3:2; } s1 = { 1, 2 }; if (s1.m3 != 0) printf("ERROR: s1.m3 = %d\n", s1.m3); } But looking at the whirl comming out of gccfe, we see that only the m1 and m2 fields are initialized. m3 has an unitialized value (equal to whatever garbage is on the stack at that location). LOC 0 0 source files: 1 "" LOC 1 2 main () LOC 1 3 { FUNC_ENTRY <1,20,main> BODY BLOCK END_BLOCK BLOCK END_BLOCK BLOCK PRAGMA 0 120 0 (0x0) # PREAMBLE_END LOC 1 6 unsigned m1:2; LOC 1 7 unsigned :2; LOC 1 8 unsigned m2:2; LOC 1 9 unsigned m3:2; LOC 1 10 } s1 = { 1, 2 }; U4INTCONST 1 (0x1) BSSTID 0 <2,1,s1> T<28,s1,4> U4INTCONST 2 (0x2) BSSTID 0 <2,1,s1> T<28,s1,4> U4INTCONST 0 (0x0) U8LDA 0 <2,1,s1> T<29,anon_ptr.,8> U4INTCONST 3 (0x3) MSTORE 1 T<30,anon_ptr.,8> LOC 1 11 LOC 1 12 if (s1.m3 != 0) IF U4BSLDID 0 <2,1,s1> T<28,s1,4> I4U4CVT I4INTCONST 0 (0x0) I4I4NE THEN BLOCK LOC 1 13 printf("ERROR: s1.m3 = %d\n", s1.m3); U8LDA 0 <1,22,(19_bytes)_"ERROR:_s1.m3_=_%d\n\000"> T<34,anon_ptr.,8> U8PARM 2 T<31,anon_ptr.,8> # by_value U4BSLDID 0 <2,1,s1> T<28,s1,4> I4U4CVT I4PARM 2 T<4,.predef_I4,4> # by_value VCALL 126 <1,21,printf> # flags 0x7e END_BLOCK ELSE LOC 1 12 BLOCK END_BLOCK END_IF RETURN END_BLOCK From owner-pro64-support@oss.sgi.com Tue Sep 19 22:46:15 2000 Received: by oss.sgi.com id ; Tue, 19 Sep 2000 22:45:55 -0700 Received: from [159.226.39.1] ([159.226.39.1]:63248 "HELO ns.ict.ac.cn") by oss.sgi.com with SMTP id ; Tue, 19 Sep 2000 22:45:42 -0700 Received: (qmail 22079 invoked from network); 20 Sep 2000 05:42:09 -0000 Received: from unknown (HELO ACTNOTEBOOK) (159.226.40.182) by ns.ict.ac.cn with SMTP; 20 Sep 2000 05:42:09 -0000 Message-ID: <00fe01c022c6$6310faa0$b628e29f@ACTMISC> From: "Wu Chengyong" To: Cc: "Zhang Zhaoqing" , "Ju, Roy" , "Wu Chengyong" , "Liu Yang" , "Lian Ruiqi" , "Gan Ge" , "Fu Chen" , "Feng Xiaobing" , "Dong Xiqian" , "Gao Lin" Subject: The long loading time of .so of Pro64 under GDB Date: Wed, 20 Sep 2000 13:48:22 +0800 MIME-Version: 1.0 Content-Type: multipart/alternative; boundary="----=_NextPart_000_00FB_01C02309.7112A8E0" 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-pro64-support@oss.sgi.com Precedence: bulk Return-Path: X-Orcpt: rfc822;pro64-support-outgoing This is a multi-part message in MIME format. ------=_NextPart_000_00FB_01C02309.7112A8E0 Content-Type: text/plain; charset="gb2312" Content-Transfer-Encoding: base64 SGksIGV2ZXJ5b25lLA0KDQpXZSBhcmUgdHJhY2luZyBhbmQgYW5hbHl6aW5nIFBybzY0IHVzaW5n IEdEQi4NCkJ1dCB3ZSBleHBlcmllbmNlZCBhIHZlcnkgbG9uZyBsb2FkaW5nIHRpbWUgd2hlbiBH REIgdHJ5IHRvIGxvYWQgc29tZSAuc28gKGUuZy4gYmUuc28sIGNnLnNvLCAuLi4pLg0KSW4gb3Vy IGVudmlyb25tZW50IGl0IG5lZWRzIGFib3V0IDQwIG1pbnV0ZXMgLSAxIGhvdXIgdG8gbG9hZCB0 aGUgc3ltYm9sIHRhYmxlcyBvZiB0aGVzZSB0d28gLnNvLg0KV2UgdXNlIEdDQyAyLjk1LjIgdG8g Y29tcGlsZSBQcm82NCAodmVyc2lvbiAwLjAxLjAtMTApIGFuZCB3ZSBtb2RpZmllZCAuLi9vc3By ZXkxLjAvTWFrZWZpbGUuZ3NldHVwDQp0byBhZGQgZGVidWdnaW5nIGluZm9ybWF0aW9uIGludG8g dGhvc2UgLnNvLg0KDQpQZW50aXVtIElJSSAxRyBNSHoNCjUxMiBNQiBNZW1vcnkNCjMwIEdCIEhh cmQgZGlzaw0KUmVkSGF0IExpbnV4IDYuMg0KR251IGdjYyAyLjk1LjINCkdudSBnZGIgMTk5OTEw MDQNCg0KSGFzIGFueWJvZHkgZXhwZXJpZW5jZWQgdGhpcyBwcm9ibGVtIG9yIGtub3cgaG93IHRv IHJlc29sdmUgaXQsIHBsZWFzZSBsZXQgbWUga25vdy4NCg0KQW55IGluZm9ybWF0aW9uIHdpbGwg YmUgYXBwcmVjaWF0ZWQuDQoNCkNoZW5neW9uZw0K ------=_NextPart_000_00FB_01C02309.7112A8E0 Content-Type: text/html; charset="gb2312" Content-Transfer-Encoding: base64 PCFET0NUWVBFIEhUTUwgUFVCTElDICItLy9XM0MvL0RURCBIVE1MIDQuMCBUcmFuc2l0aW9uYWwv L0VOIj4NCjxIVE1MPjxIRUFEPg0KPE1FVEEgY29udGVudD0idGV4dC9odG1sOyBjaGFyc2V0PWdi MjMxMiIgaHR0cC1lcXVpdj1Db250ZW50LVR5cGU+DQo8TUVUQSBjb250ZW50PSJNU0hUTUwgNS4w MC4yNjE0LjM1MDAiIG5hbWU9R0VORVJBVE9SPg0KPFNUWUxFPjwvU1RZTEU+DQo8L0hFQUQ+DQo8 Qk9EWT4NCjxESVY+PEZPTlQgc2l6ZT0yPkhpLCBldmVyeW9uZSw8L0ZPTlQ+PC9ESVY+DQo8RElW PjxGT05UIHNpemU9Mj48L0ZPTlQ+Jm5ic3A7PC9ESVY+DQo8RElWPjxGT05UIHNpemU9Mj5XZSBh cmUgdHJhY2luZyBhbmQgYW5hbHl6aW5nIFBybzY0IHVzaW5nIEdEQi48L0ZPTlQ+PC9ESVY+DQo8 RElWPjxGT05UIHNpemU9Mj5CdXQgd2UgZXhwZXJpZW5jZWQgYSB2ZXJ5IGxvbmcgbG9hZGluZyB0 aW1lIHdoZW4gR0RCIHRyeSB0byANCmxvYWQgc29tZSAuc28gKGUuZy4gYmUuc28sIGNnLnNvLCAu Li4pLjwvRk9OVD48L0RJVj4NCjxESVY+PEZPTlQgc2l6ZT0yPkluIG91ciBlbnZpcm9ubWVudCZu YnNwO2l0IG5lZWRzIGFib3V0IDQwIG1pbnV0ZXMgLSAxIGhvdXIgdG8gDQpsb2FkIHRoZSBzeW1i b2wgdGFibGVzIG9mIHRoZXNlIHR3byAuc28uPC9GT05UPjwvRElWPg0KPERJVj48Rk9OVCBzaXpl PTI+V2UgdXNlIDxFTT48Rk9OVCBjb2xvcj0jMDAwMGZmPkdDQyAyLjk1LjIgPC9GT05UPjwvRU0+ dG8gDQpjb21waWxlIFBybzY0ICh2ZXJzaW9uIDxFTT48Rk9OVCBjb2xvcj0jMDAwMGZmPjAuMDEu MC0xMDwvRk9OVD48L0VNPikgYW5kIHdlIA0KbW9kaWZpZWQgPEZPTlQgDQpjb2xvcj0jMDAwMGZm PjxFTT4uLi9vc3ByZXkxLjAvTWFrZWZpbGUuZ3NldHVwPC9FTT48L0ZPTlQ+PC9GT05UPjwvRElW Pg0KPERJVj48Rk9OVCBzaXplPTI+dG8gYWRkIGRlYnVnZ2luZyBpbmZvcm1hdGlvbiBpbnRvIHRo b3NlIC5zby48L0ZPTlQ+PC9ESVY+DQo8RElWPiZuYnNwOzwvRElWPg0KPERJVj48Rk9OVCBzaXpl PTI+UGVudGl1bSBJSUkgMUcgTUh6PC9GT05UPjwvRElWPg0KPERJVj48Rk9OVCBzaXplPTI+NTEy IE1CIE1lbW9yeTwvRk9OVD48L0RJVj4NCjxESVY+PEZPTlQgc2l6ZT0yPjMwIEdCIEhhcmQgZGlz azwvRk9OVD48L0RJVj4NCjxESVY+PEZPTlQgc2l6ZT0yPlJlZEhhdCBMaW51eCA2LjI8L0ZPTlQ+ PC9ESVY+DQo8RElWPjxGT05UIHNpemU9Mj5HbnUgZ2NjIDIuOTUuMjwvRk9OVD48L0RJVj4NCjxE SVY+PEZPTlQgc2l6ZT0yPkdudSBnZGIgMTk5OTEwMDQ8L0ZPTlQ+PC9ESVY+DQo8RElWPiZuYnNw OzwvRElWPg0KPERJVj48Rk9OVCBzaXplPTI+SGFzIGFueWJvZHkgZXhwZXJpZW5jZWQgdGhpcyBw cm9ibGVtIG9yIGtub3cgaG93IHRvIHJlc29sdmUgDQppdCwgcGxlYXNlIGxldCBtZSBrbm93Ljwv Rk9OVD48L0RJVj4NCjxESVY+PEZPTlQgc2l6ZT0yPjwvRk9OVD4mbmJzcDs8L0RJVj4NCjxESVY+ PEZPTlQgc2l6ZT0yPkFueSBpbmZvcm1hdGlvbiB3aWxsIGJlIGFwcHJlY2lhdGVkLjwvRk9OVD48 L0RJVj4NCjxESVY+Jm5ic3A7PC9ESVY+DQo8RElWPjxGT05UIHNpemU9Mj5DaGVuZ3lvbmc8L0ZP TlQ+PC9ESVY+PC9CT0RZPjwvSFRNTD4NCg== ------=_NextPart_000_00FB_01C02309.7112A8E0-- From owner-pro64-support@oss.sgi.com Wed Sep 20 02:38:18 2000 Received: by oss.sgi.com id ; Wed, 20 Sep 2000 02:38:08 -0700 Received: from Cantor.suse.de ([194.112.123.193]:43280 "HELO Cantor.suse.de") by oss.sgi.com with SMTP id ; Wed, 20 Sep 2000 02:38:00 -0700 Received: from Hermes.suse.de (Hermes.suse.de [194.112.123.136]) by Cantor.suse.de (Postfix) with ESMTP id 345811E12B; Wed, 20 Sep 2000 11:37:58 +0200 (MEST) Received: from gruyere.muc.suse.de (unknown [10.23.1.2]) by Hermes.suse.de (Postfix) with ESMTP id 83B2910A02D; Wed, 20 Sep 2000 11:37:57 +0200 (MEST) Received: by gruyere.muc.suse.de (Postfix, from userid 14446) id C68332F300; Wed, 20 Sep 2000 11:37:48 +0200 (MEST) Date: Wed, 20 Sep 2000 11:37:48 +0200 From: "Andi Kleen" To: "Wu Chengyong" Cc: , "Zhang Zhaoqing" , "Ju, Roy" , "Liu Yang" , "Lian Ruiqi" , "Gan Ge" , "Fu Chen" , "Feng Xiaobing" , "Dong Xiqian" , "Gao Lin" Subject: Re: The long loading time of .so of Pro64 under GDB Message-ID: <20000920113748.A4521@gruyere.muc.suse.de> References: <00fe01c022c6$6310faa0$b628e29f@ACTMISC> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Mailer: Mutt 1.0i In-Reply-To: <00fe01c022c6$6310faa0$b628e29f@ACTMISC>; from cwu@ict.ac.cn on Wed, Sep 20, 2000 at 01:48:22PM +0800 Sender: owner-pro64-support@oss.sgi.com Precedence: bulk Return-Path: X-Orcpt: rfc822;pro64-support-outgoing On Wed, Sep 20, 2000 at 01:48:22PM +0800, Wu Chengyong wrote: > Hi, everyone, > > We are tracing and analyzing Pro64 using GDB. > But we experienced a very long loading time when GDB try to load some .so (e.g. be.so, cg.so, ...). > In our environment it needs about 40 minutes - 1 hour to load the symbol tables of these two .so. > We use GCC 2.95.2 to compile Pro64 (version 0.01.0-10) and we modified ../osprey1.0/Makefile.gsetup > to add debugging information into those .so. Your machine is probably heavily swapping/trashing because gdb needs so much memory. You can check its memory use with running top during the load. gdb also deals very poorly with very big symbol tables (and worse with big C++ programs which have very long symbol names too). An useful resource is http://www.mozilla.org/unix/debugging-faq.html , it explains some techniques that may help. -Andi From owner-pro64-support@oss.sgi.com Wed Sep 20 06:32:59 2000 Received: by oss.sgi.com id ; Wed, 20 Sep 2000 06:32:49 -0700 Received: from runyon.cygnus.com ([205.180.230.5]:40884 "EHLO cygnus.com") by oss.sgi.com with ESMTP id ; Wed, 20 Sep 2000 06:32:31 -0700 Received: from dan2.cygnus.com (cse.cygnus.com [205.180.230.236]) by runyon.cygnus.com (8.8.7-cygnus/8.8.7) with SMTP id GAA03113; Wed, 20 Sep 2000 06:31:15 -0700 (PDT) To: "Andi Kleen" Cc: "Wu Chengyong" , , "Zhang Zhaoqing" , "Ju, Roy" , "Liu Yang" , "Lian Ruiqi" , "Gan Ge" , "Fu Chen" , "Feng Xiaobing" , "Dong Xiqian" , "Gao Lin" Subject: Re: The long loading time of .so of Pro64 under GDB References: <00fe01c022c6$6310faa0$b628e29f@ACTMISC> <20000920113748.A4521@gruyere.muc.suse.de> From: Daniel Berlin Date: 20 Sep 2000 09:31:14 -0400 In-Reply-To: "Andi Kleen"'s message of "Wed, 20 Sep 2000 11:37:48 +0200" Message-ID: Lines: 33 User-Agent: Gnus/5.0807 (Gnus v5.8.7) XEmacs/21.1 (GTK) MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Sender: owner-pro64-support@oss.sgi.com Precedence: bulk Return-Path: X-Orcpt: rfc822;pro64-support-outgoing "Andi Kleen" writes: > > Your machine is probably heavily swapping/trashing because gdb needs so much > memory. You can check its memory use with running top during the load. > gdb also deals very poorly with very big symbol tables (and worse with big > C++ programs which have very long symbol names too). An useful resource is > http://www.mozilla.org/unix/debugging-faq.html , it explains some techniques > that may help. Err, not anymore. I've fixed the memory usage as of a few weeks ago, at least for DWARF2. The patches haven't been commited yet (I'll do it in about a week), but they are on gdb-patches C++ symbol searches also used to be linear, and are now binary searches (It was a mangling issue). You want the most recent dwarf2 patch i submitted, and the C++ patch from a few weeks ago. This should bring your load time down to something much more reasonable, as well as your memory usage. I also fixed name completion so quoting is not necessary anymore. It's cute when people don't submit bug reports on this stuff, so that I can't make a case to anyone to schedule me to fix it. I've also only been working on fixing C++ in GDB since february, and it was amazingly broken, so give me time. The mozilla guys already love me. --Dan C++ support maintainer - GDB. From owner-pro64-support@oss.sgi.com Wed Sep 20 06:34:49 2000 Received: by oss.sgi.com id ; Wed, 20 Sep 2000 06:34:39 -0700 Received: from [159.226.39.1] ([159.226.39.1]:47880 "HELO ns.ict.ac.cn") by oss.sgi.com with SMTP id ; Wed, 20 Sep 2000 06:34:38 -0700 Received: (qmail 11474 invoked from network); 20 Sep 2000 13:30:37 -0000 Received: from unknown (HELO ACTNOTEBOOK) (159.226.40.182) by ns.ict.ac.cn with SMTP; 20 Sep 2000 13:30:37 -0000 Message-ID: <000f01c02307$d2969100$b628e29f@ACTMISC> From: "Wu Chengyong" To: "Andi Kleen" Cc: "pro64-support" References: <00fe01c022c6$6310faa0$b628e29f@ACTMISC> <20000920113748.A4521@gruyere.muc.suse.de> Subject: Re: The long loading time of .so of Pro64 under GDB Date: Wed, 20 Sep 2000 21:36:46 +0800 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-pro64-support@oss.sgi.com Precedence: bulk Return-Path: X-Orcpt: rfc822;pro64-support-outgoing Thank you very much, Andi, I just visited the web site you recomended and according some directions there, I downloaded the yet unreleased GDB 5.0 and installed it. And all my problems disappear. Now it needs only 2 MINITES to load all symbol tables! Once more, thank you very much. Chengyong ----- Original Message ----- From: Andi Kleen To: Wu Chengyong Cc: ; Zhang Zhaoqing ; Ju, Roy ; Liu Yang ; Lian Ruiqi ; Gan Ge ; Fu Chen ; Feng Xiaobing ; Dong Xiqian ; Gao Lin Sent: Wednesday, September 20, 2000 5:37 PM Subject: Re: The long loading time of .so of Pro64 under GDB > On Wed, Sep 20, 2000 at 01:48:22PM +0800, Wu Chengyong wrote: > > Hi, everyone, > > > > We are tracing and analyzing Pro64 using GDB. > > But we experienced a very long loading time when GDB try to load some .so (e.g. be.so, cg.so, ...). > > In our environment it needs about 40 minutes - 1 hour to load the symbol tables of these two .so. > > We use GCC 2.95.2 to compile Pro64 (version 0.01.0-10) and we modified ../osprey1.0/Makefile.gsetup > > to add debugging information into those .so. > > Your machine is probably heavily swapping/trashing because gdb needs so much > memory. You can check its memory use with running top during the load. > gdb also deals very poorly with very big symbol tables (and worse with big > C++ programs which have very long symbol names too). An useful resource is > http://www.mozilla.org/unix/debugging-faq.html , it explains some techniques > that may help. > > > > -Andi > From owner-pro64-support@oss.sgi.com Wed Sep 20 06:44:19 2000 Received: by oss.sgi.com id ; Wed, 20 Sep 2000 06:44:09 -0700 Received: from runyon.cygnus.com ([205.180.230.5]:57524 "EHLO cygnus.com") by oss.sgi.com with ESMTP id ; Wed, 20 Sep 2000 06:43:43 -0700 Received: from dan2.cygnus.com (cse.cygnus.com [205.180.230.236]) by runyon.cygnus.com (8.8.7-cygnus/8.8.7) with SMTP id GAA03339; Wed, 20 Sep 2000 06:41:29 -0700 (PDT) To: "Andi Kleen" Cc: "Wu Chengyong" , , "Zhang Zhaoqing" , "Ju, Roy" , "Liu Yang" , "Lian Ruiqi" , "Gan Ge" , "Fu Chen" , "Feng Xiaobing" , "Dong Xiqian" , "Gao Lin" Subject: Re: The long loading time of .so of Pro64 under GDB References: <00fe01c022c6$6310faa0$b628e29f@ACTMISC> <20000920113748.A4521@gruyere.muc.suse.de> From: Daniel Berlin Date: 20 Sep 2000 09:41:28 -0400 In-Reply-To: "Andi Kleen"'s message of "Wed, 20 Sep 2000 11:37:48 +0200" Message-ID: Lines: 31 User-Agent: Gnus/5.0807 (Gnus v5.8.7) XEmacs/21.1 (GTK) MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Sender: owner-pro64-support@oss.sgi.com Precedence: bulk Return-Path: X-Orcpt: rfc822;pro64-support-outgoing "Andi Kleen" writes: > memory. You can check its memory use with running top during the load. > gdb also deals very poorly with very big symbol tables (and worse with big > C++ programs which have very long symbol names too). An useful resource is > http://www.mozilla.org/unix/debugging-faq.html , it explains some techniques > that may help. Oh, and for those who are into stats, before my dwarf2 patch, debugging a 100 meg executable with dwarf2 info took ~140 meg of memory, just to load the thing in on start. You ended up taking a lot more as time progresses. Now it takes ~80 meg to load the thing in (not too shabby for a 100 meg executable), and you don't end up taking all that much more as time progresses. Also, most of the 80 meg is shared memory, while none of the 140 is shared. I could do a little better, probably shaving it down to 70 or 60 meg, but i dunno if it'd be worth my trouble. The main memory user becomes the type structures, which i'm in the middle of redesigning. Once that is done, memory usage should go down by an order of magnitude (each C++ type structure is now 52+114 bytes minimum. If you have 500000 of these, which isn't unusual for a 100 meg executable, you end up with 57 meg of just structure space, not including the names/etc in that structure). --Dan > > -Andi From owner-pro64-support@oss.sgi.com Wed Sep 20 06:45:58 2000 Received: by oss.sgi.com id ; Wed, 20 Sep 2000 06:45:49 -0700 Received: from runyon.cygnus.com ([205.180.230.5]:60852 "EHLO cygnus.com") by oss.sgi.com with ESMTP id ; Wed, 20 Sep 2000 06:45:39 -0700 Received: from dan2.cygnus.com (cse.cygnus.com [205.180.230.236]) by runyon.cygnus.com (8.8.7-cygnus/8.8.7) with SMTP id GAA03468; Wed, 20 Sep 2000 06:45:36 -0700 (PDT) To: "Wu Chengyong" Cc: "Andi Kleen" , "pro64-support" Subject: Re: The long loading time of .so of Pro64 under GDB References: <00fe01c022c6$6310faa0$b628e29f@ACTMISC> <20000920113748.A4521@gruyere.muc.suse.de> <000f01c02307$d2969100$b628e29f@ACTMISC> From: Daniel Berlin Date: 20 Sep 2000 09:45:34 -0400 In-Reply-To: "Wu Chengyong"'s message of "Wed, 20 Sep 2000 21:36:46 +0800" Message-ID: Lines: 20 User-Agent: Gnus/5.0807 (Gnus v5.8.7) XEmacs/21.1 (GTK) MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Sender: owner-pro64-support@oss.sgi.com Precedence: bulk Return-Path: X-Orcpt: rfc822;pro64-support-outgoing "Wu Chengyong" writes: > Thank you very much, Andi, > > I just visited the web site you recomended and according some directions > there, > I downloaded the yet unreleased GDB 5.0 and installed it. GDB 5 was released a while ago, actually. Dunno what makes you think it's unreleased. > And all my problems disappear. > Now it needs only 2 MINITES to load all symbol tables! > Not surprising. It still shouldn't even be that long. > Once more, thank you very much. Your welcome, as most of that is probably my work. --Dan From owner-pro64-support@oss.sgi.com Wed Sep 20 17:30:37 2000 Received: by oss.sgi.com id ; Wed, 20 Sep 2000 17:30:27 -0700 Received: from deliverator.sgi.com ([204.94.214.10]:23076 "EHLO deliverator.sgi.com") by oss.sgi.com with ESMTP id ; Wed, 20 Sep 2000 17:30:14 -0700 Received: from cchkms.engr.sgi.com (cchkms.engr.sgi.com [130.62.180.48]) by deliverator.sgi.com (980309.SGI.8.8.8-aspam-6.2/980310.SGI-aspam) via ESMTP id RAA17961; Wed, 20 Sep 2000 17:22:32 -0700 (PDT) mail_from (rat@cchkms.engr.sgi.com) Received: (from rat@localhost) by cchkms.engr.sgi.com (980427.SGI.8.8.8/970903.SGI.AUTOCF) id RAA12131; Wed, 20 Sep 2000 17:27:34 -0700 (PDT) From: "Ross A. Towle" Message-Id: <10009201727.ZM12198@cchkms.engr.sgi.com> Date: Wed, 20 Sep 2000 17:27:33 -0700 X-Mailer: Z-Mail (3.2.3 08feb96 MediaMail) To: pro64-announce@oss.sgi.com, pro64-support@oss.sgi.com Subject: New release of Pro64 Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Sender: owner-pro64-support@oss.sgi.com Precedence: bulk Return-Path: X-Orcpt: rfc822;pro64-support-outgoing is available from http://oss.sgi.com/projects/Pro64. This fixes the bugs reported by Stephane Eranian and a performance bug from Sverre Jarp. The Fortran90 routines to "getarg" and "iargc" now work. More of the Fortran90 wrappers to get to system information are in place. A number of other correctness and performance bugs have been fixed. If you install this release on IA-64 boxes be sure to use the "workaround-1.1.efi" from Intel. It does not eliminate the spurious segfaults but it does keep the system from freezing during compilation. -Ross A. Towle Director, HPPE SGI From owner-pro64-support@oss.sgi.com Thu Sep 21 05:58:33 2000 Received: by oss.sgi.com id ; Thu, 21 Sep 2000 05:58:23 -0700 Received: from e22.nc.us.ibm.com ([32.97.136.228]:33739 "EHLO e22.nc.us.ibm.com") by oss.sgi.com with ESMTP id ; Thu, 21 Sep 2000 05:58:07 -0700 Received: from southrelay02.raleigh.ibm.com (southrelay02.raleigh.ibm.com [9.37.3.209]) by e22.nc.us.ibm.com (8.9.3/8.9.3) with ESMTP id IAA30372 for ; Thu, 21 Sep 2000 08:33:48 -0500 From: gjertsen@us.ibm.com Received: from d54mta01.raleigh.ibm.com (d54mta01.raleigh.ibm.com [9.67.228.33]) by southrelay02.raleigh.ibm.com (8.8.8m3/NCO v4.93) with SMTP id IAA70224 for ; Thu, 21 Sep 2000 08:58:05 -0400 Received: by d54mta01.raleigh.ibm.com(Lotus SMTP MTA v4.6.5 (863.2 5-20-1999)) id 85256961.00473AC6 ; Thu, 21 Sep 2000 08:58:01 -0400 X-Lotus-FromDomain: IBMUS To: pro64-support@oss.sgi.com Message-ID: <85256961.0047397C.00@d54mta01.raleigh.ibm.com> Date: Thu, 21 Sep 2000 08:57:56 -0400 Subject: Re: New release of Pro64 Mime-Version: 1.0 Content-type: text/plain; charset=us-ascii Content-Disposition: inline Sender: owner-pro64-support@oss.sgi.com Precedence: bulk Return-Path: X-Orcpt: rfc822;pro64-support-outgoing Ross, Where can we find this update for "workaround-1.1.efi"? I've been rambling around the intel IA64 efi site without success. What exactly is causing the freezing of the system? I'm seeing this with sgiCC (prevoius 8/28 update), but have also encountered it with other apps. Thanks, --Rob "Ross A. Towle" @oss.sgi.com on 09/20/2000 07:27:33 PM Sent by: owner-pro64-support@oss.sgi.com To: pro64-announce@oss.sgi.com, pro64-support@oss.sgi.com cc: Subject: New release of Pro64 is available from http://oss.sgi.com/projects/Pro64. This fixes the bugs reported by Stephane Eranian and a performance bug from Sverre Jarp. The Fortran90 routines to "getarg" and "iargc" now work. More of the Fortran90 wrappers to get to system information are in place. A number of other correctness and performance bugs have been fixed. If you install this release on IA-64 boxes be sure to use the "workaround-1.1.efi" from Intel. It does not eliminate the spurious segfaults but it does keep the system from freezing during compilation. -Ross A. Towle Director, HPPE SGI From owner-pro64-support@oss.sgi.com Thu Sep 21 16:03:16 2000 Received: by oss.sgi.com id ; Thu, 21 Sep 2000 16:03:07 -0700 Received: from louie.udel.edu ([128.175.7.39]:57808 "HELO mail.eecis.udel.edu") by oss.sgi.com with SMTP id ; Thu, 21 Sep 2000 16:02:39 -0700 Received: from ren.eecis.udel.edu by mail.eecis.udel.edu id aa14539; 21 Sep 2000 14:43 EDT Date: Thu, 21 Sep 2000 14:43:47 -0400 (EDT) From: Ziang Hu To: pro64-support@oss.sgi.com cc: ggao@capsl.udel.edu Subject: need f90 compiler for Pro64 Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-pro64-support@oss.sgi.com Precedence: bulk Return-Path: X-Orcpt: rfc822;pro64-support-outgoing Dear Person, You still need a native f90 compiler to compiler Pro64 Fortran front-end (cray90/fe90), right? Can we use g77 for that ? Thanks Ziang From owner-pro64-support@oss.sgi.com Thu Sep 21 16:23:16 2000 Received: by oss.sgi.com id ; Thu, 21 Sep 2000 16:23:07 -0700 Received: from pneumatic-tube.sgi.com ([204.94.214.22]:52272 "EHLO pneumatic-tube.sgi.com") by oss.sgi.com with ESMTP id ; Thu, 21 Sep 2000 16:22:45 -0700 Received: from cchkms.engr.sgi.com (cchkms.engr.sgi.com [130.62.180.48]) by pneumatic-tube.sgi.com (980327.SGI.8.8.8-aspam/980310.SGI-aspam) via ESMTP id QAA06318 for ; Thu, 21 Sep 2000 16:29:33 -0700 (PDT) mail_from (rat@cchkms.engr.sgi.com) Received: (from rat@localhost) by cchkms.engr.sgi.com (980427.SGI.8.8.8/970903.SGI.AUTOCF) id QAA15218; Thu, 21 Sep 2000 16:21:22 -0700 (PDT) From: "Ross A. Towle" Message-Id: <10009211621.ZM14604@cchkms.engr.sgi.com> Date: Thu, 21 Sep 2000 16:21:20 -0700 In-Reply-To: Ziang Hu "need f90 compiler for Pro64" (Sep 21, 2:43pm) References: X-Mailer: Z-Mail (3.2.3 08feb96 MediaMail) To: Ziang Hu , pro64-support@oss.sgi.com Subject: Re: need f90 compiler for Pro64 Cc: ggao@capsl.udel.edu Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Sender: owner-pro64-support@oss.sgi.com Precedence: bulk Return-Path: X-Orcpt: rfc822;pro64-support-outgoing You need a f90 compiler for whatever system is the host system, ie the system where you are doing the build compiles of fe90. g77 will not work. -Ross From owner-pro64-support@oss.sgi.com Thu Sep 21 19:43:37 2000 Received: by oss.sgi.com id ; Thu, 21 Sep 2000 19:43:18 -0700 Received: from munch-it.turbolinux.com ([38.170.88.129]:65265 "EHLO mail.us.tlan") by oss.sgi.com with ESMTP id ; Thu, 21 Sep 2000 19:42:54 -0700 Received: (from nobody@localhost) by mail.us.tlan (8.9.3/8.9.3) id TAA01244 for ; Thu, 21 Sep 2000 19:42:54 -0700 Received: from ariel.dev.us.tlan(172.16.12.158), claiming to be "turbolinux.com" via SMTP by mail.us.tlan, id smtpd54wEiU; Thu Sep 21 19:42:48 2000 Message-ID: <39CAC73F.AA22DA8@turbolinux.com> Date: Thu, 21 Sep 2000 19:43:12 -0700 From: Uros Prestor Organization: Turbolinux Inc. X-Mailer: Mozilla 4.74 [en] (X11; U; Linux 2.2.13-12 i686) X-Accept-Language: sl, en MIME-Version: 1.0 CC: pro64-support@oss.sgi.com Subject: Re: New release of Pro64 References: <10009201727.ZM12198@cchkms.engr.sgi.com> Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit To: unlisted-recipients:; (no To-header on input) Sender: owner-pro64-support@oss.sgi.com Precedence: bulk Return-Path: X-Orcpt: rfc822;pro64-support-outgoing Did anybody try to compile 2.4.0-test8 kernel with this release of Pro64? I get the following error while trying to `make dep': make -C arch/ia64/tools fastdep make[2]: Entering directory `/work/BUILD/linux/arch/ia64/tools' sgicc -D__LP64__ -g -O2 -Wall -D__KERNEL__ -I/work/BUILD/linux/include print_offsets.c -o print_offsets print_offsets.o(.debug_info+0xb9d): undefined reference to `dcache_lock' print_offsets.o(.debug_info+0x6c1a): undefined reference to `tqueue_lock' print_offsets.o(.debug_info+0x7055): undefined reference to `pidhash' print_offsets.o(.debug_info+0x8694): undefined reference to `securebits' print_offsets.o(.debug_info+0x86af): undefined reference to `nr_running' print_offsets.o(.debug_info+0x86cd): undefined reference to `tasklist_lock' print_offsets.o(.debug_info+0x86e8): undefined reference to `nr_threads' collect2: ld returned 1 exit status make[2]: *** [print_offsets] Error 2 Thanks, Uros -- Uros Prestor uros@turbolinux.com From owner-pro64-support@oss.sgi.com Wed Sep 27 07:45:33 2000 Received: by oss.sgi.com id ; Wed, 27 Sep 2000 07:45:13 -0700 Received: from [159.226.39.1] ([159.226.39.1]:3852 "HELO ns.ict.ac.cn") by oss.sgi.com with SMTP id ; Wed, 27 Sep 2000 07:45:05 -0700 Received: (qmail 20708 invoked from network); 27 Sep 2000 14:42:53 -0000 Received: from unknown (HELO ACTNOTEBOOK) (159.226.40.182) by ns.ict.ac.cn with SMTP; 27 Sep 2000 14:42:53 -0000 Message-ID: <00f801c02891$d52ae120$b628e29f@ACTMISC> From: "Wu Chengyong" To: "pro64-support" Cc: "Ju, Roy" Subject: VCG Date: Wed, 27 Sep 2000 22:47:17 +0800 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-pro64-support@oss.sgi.com Precedence: bulk Return-Path: X-Orcpt: rfc822;pro64-support-outgoing Does anybody know if Pro64 can produce the input file VCG needs? There are many places in the source related to VCG, so I guess perhaps it can do it. But I don't know what options to use. Does anybody know something about it? Chengyong From owner-pro64-support@oss.sgi.com Fri Sep 29 15:08:54 2000 Received: by oss.sgi.com id ; Fri, 29 Sep 2000 15:08:44 -0700 Received: from [38.170.141.29] ([38.170.141.29]:44018 "EHLO heart.hq.tensilica.com") by oss.sgi.com with ESMTP id ; Fri, 29 Sep 2000 15:08:26 -0700 Received: (from goodwin@localhost) by heart.hq.tensilica.com (8.9.3/8.9.3) id PAA19506; Fri, 29 Sep 2000 15:08:26 -0700 X-Authentication-Warning: heart.hq.tensilica.com: goodwin set sender to goodwin@tensilica.com using -f From: David Goodwin MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Message-ID: <14805.4826.13856.60813@heart.hq.tensilica.com> Date: Fri, 29 Sep 2000 15:08:25 -0700 (PDT) To: pro64-support@oss.sgi.com Subject: compiler error with comma-separated expressions X-Mailer: VM 6.75 under Emacs 20.5.1 Reply-to: goodwin@tensilica.com Sender: owner-pro64-support@oss.sgi.com Precedence: bulk Return-Path: X-Orcpt: rfc822;pro64-support-outgoing This code fails at -O0 but passes at -O2. [test] sgicc -S paren.c ### Assertion failure at line 1012 of ../../be/com/wn_verifier.cxx: ### Compiler Error in file paren.c during Lowering phase: ### field_id and descriptor type are inconsistent sgicc INTERNAL ERROR: ia64/lib/gcc-lib/ia64-sgi-linux/sgicc-1.0/be returned non-zero status 1 typedef struct { int a; int b; int c; } astruct; int g; astruct *gs; void paren (astruct *as) { (g = 1, *gs = *as); }