From owner-pro64-support@oss.sgi.com Mon Jan 1 22:14:12 2001 Received: by oss.sgi.com id ; Mon, 1 Jan 2001 22:14:02 -0800 Received: from www.pspl.co.in ([202.54.11.65]:11018 "EHLO www.pspl.co.in") by oss.sgi.com with ESMTP id ; Mon, 1 Jan 2001 22:13:41 -0800 Received: from kampala (kampala.intranet.pspl.co.in [192.168.1.57]) by www.pspl.co.in (8.11.0/8.11.0) with SMTP id f026KrY29430; Tue, 2 Jan 2001 11:50:54 +0530 From: "Rajalakshmi Iyer" To: Cc: , , Subject: Column no. info required Date: Tue, 2 Jan 2001 11:45:21 +0530 Message-ID: <000001c07483$63c38670$3901a8c0@intranet.pspl.co.in> MIME-Version: 1.0 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: 7bit X-Priority: 3 (Normal) X-MSMail-Priority: Normal X-Mailer: Microsoft Outlook CWS, Build 9.0.2416 (9.0.2910.0) Importance: Normal X-MimeOLE: Produced By Microsoft MimeOLE V5.00.2314.1300 Sender: owner-pro64-support@oss.sgi.com Precedence: bulk Return-Path: X-Orcpt: rfc822;pro64-support-outgoing We are using the SGI Pro64 compiler version 0.01.0-12 and are mainly interested in the intermediate representation produced by the compiler. We need to make certain changes in a sample source code based on the information provided by the intermediate representation. We are not able to get the proper source position information of a statement node. We have used the APIs provided in the file srcpos.h. Although we are getting proper line number information, the column number information is incorrect (it is always 0). We also tried having 2 statements on the same line. Any help in this regard would be greatly appreciated. Thanks in advance Rajalakshmi Iyer Member of Technical Staff Persistent Systems. From owner-pro64-support@oss.sgi.com Mon Jan 1 23:42:22 2001 Received: by oss.sgi.com id ; Mon, 1 Jan 2001 23:42:12 -0800 Received: from Cantor.suse.de ([194.112.123.193]:2567 "HELO Cantor.suse.de") by oss.sgi.com with SMTP id ; Mon, 1 Jan 2001 23:41:54 -0800 Received: from Hermes.suse.de (Hermes.suse.de [194.112.123.136]) by Cantor.suse.de (Postfix) with ESMTP id 2D06A1E20C; Tue, 2 Jan 2001 08:41:52 +0100 (MET) Received: from gruyere.muc.suse.de (unknown [10.23.1.2]) by Hermes.suse.de (Postfix) with ESMTP id 870FE3E44F; Tue, 2 Jan 2001 08:41:51 +0100 (MET) Received: by gruyere.muc.suse.de (Postfix, from userid 14446) id 80C0C2F300; Tue, 2 Jan 2001 08:41:49 +0100 (MET) Date: Tue, 2 Jan 2001 08:41:49 +0100 From: Andi Kleen To: Rajalakshmi Iyer Cc: pro64-support@oss.sgi.com, hammigi@pspl.co.in, svivek@pspl.co.in Subject: Re: Column no. info required Message-ID: <20010102084149.A4546@gruyere.muc.suse.de> References: <000001c07483$63c38670$3901a8c0@intranet.pspl.co.in> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline User-Agent: Mutt/1.2.5i In-Reply-To: <000001c07483$63c38670$3901a8c0@intranet.pspl.co.in>; from rajiyer@pspl.co.in on Tue, Jan 02, 2001 at 11:45:21AM +0530 Sender: owner-pro64-support@oss.sgi.com Precedence: bulk Return-Path: X-Orcpt: rfc822;pro64-support-outgoing On Tue, Jan 02, 2001 at 11:45:21AM +0530, Rajalakshmi Iyer wrote: > > We are using the SGI Pro64 compiler version 0.01.0-12 and are mainly > interested in the intermediate representation produced by the compiler. > > We need to make certain changes in a sample source code based on the > information provided by the intermediate representation. > > We are not able to get the proper source position information of a statement > node. We have used the APIs provided in the file srcpos.h. Although we are > getting proper line number information, the column number information is > incorrect (it is always 0). We also tried having 2 statements on the same > line. Normal gcc frontend does not supply column information, because that requires a builtin preprocessor to be meaningfull. Latest gcc 2.97 optionally supports a preprocessor now, but I guess the gcc snapshot used in pro64 is too old to have that already. You would probably need to port a newer snapshot. -Andi From owner-pro64-support@oss.sgi.com Tue Jan 2 09:29:35 2001 Received: by oss.sgi.com id ; Tue, 2 Jan 2001 09:29:25 -0800 Received: from deliverator.sgi.com ([204.94.214.10]:1548 "EHLO deliverator.sgi.com") by oss.sgi.com with ESMTP id ; Tue, 2 Jan 2001 09:29:02 -0800 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 JAA28293 for ; Tue, 2 Jan 2001 09:28:11 -0800 (PST) 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 JAA15430; Tue, 2 Jan 2001 09:21:35 -0800 (PST) From: "Ross A. Towle" Message-Id: <10101020921.ZM15390@cchkms.engr.sgi.com> Date: Tue, 2 Jan 2001 09:21:34 -0800 In-Reply-To: "Rajalakshmi Iyer" "Column no. info required" (Jan 2, 11:45am) References: <000001c07483$63c38670$3901a8c0@intranet.pspl.co.in> X-Mailer: Z-Mail (3.2.3 08feb96 MediaMail) To: Subject: Re: Column no. info required Cc: , 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 The column # is not filled in. That is why you are seeing 0. -Ross From owner-pro64-support@oss.sgi.com Fri Jan 5 09:38:13 2001 Received: by oss.sgi.com id ; Fri, 5 Jan 2001 09:37:53 -0800 Received: from [38.170.141.29] ([38.170.141.29]:52210 "EHLO heart.hq.tensilica.com") by oss.sgi.com with ESMTP id ; Fri, 5 Jan 2001 09:37:38 -0800 Received: (from goodwin@localhost) by heart.hq.tensilica.com (8.9.3/8.9.3) id JAA10829; Fri, 5 Jan 2001 09:37:38 -0800 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: <14934.1634.87891.229902@heart.hq.tensilica.com> Date: Fri, 5 Jan 2001 09:37:38 -0800 (PST) To: pro64-support@oss.sgi.com Cc: goodwin@tensilica.com Subject: gcc extension to control variable names works with sgiCC, but not sgicc. 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 gcc allows this use of "asm" to set the name used for a variable in the assembly file to be different than the variable name in the source: char myarray[10] asm ("asmname_myarray"); char * proc (void) { return myarray; } > sgicc -v SGIcc Compilers: Version 0.01.0-12 > sgiCC -v SGIcc Compilers: Version 0.01.0-12 Compiling with sgiCC gives the correct result, i.e. the array is referred to as "asmname_myarray". Using sgicc, the array is incorrectly referred to as "myarray". From owner-pro64-support@oss.sgi.com Fri Jan 5 10:10:14 2001 Received: by oss.sgi.com id ; Fri, 5 Jan 2001 10:10:04 -0800 Received: from deliverator.sgi.com ([204.94.214.10]:62825 "EHLO deliverator.sgi.com") by oss.sgi.com with ESMTP id ; Fri, 5 Jan 2001 10:09:47 -0800 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 KAA02734 for ; Fri, 5 Jan 2001 10:08:55 -0800 (PST) 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 KAA41087; Fri, 5 Jan 2001 10:01:30 -0800 (PST) Message-ID: <3A560BFA.5DF2AFFD@sgi.com> Date: Fri, 05 Jan 2001 10:01:30 -0800 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: goodwin@tensilica.com CC: pro64-support@oss.sgi.com Subject: Re: gcc extension to control variable names works with sgiCC, but not sgicc. References: <14934.1634.87891.229902@heart.hq.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 David Goodwin wrote: > > gcc allows this use of "asm" to set the name used for a variable in the > assembly file to be different than the variable name in the source: > > char myarray[10] asm ("asmname_myarray"); > char * > proc (void) > { > return myarray; > } > > > sgicc -v > SGIcc Compilers: Version 0.01.0-12 > > sgiCC -v > SGIcc Compilers: Version 0.01.0-12 > > Compiling with sgiCC gives the correct result, i.e. the array is referred > to as "asmname_myarray". Using sgicc, the array is incorrectly referred to > as "myarray". sgicc is using DECL_NAME instead of DECL_ASSEMBLER_NAME in tree_symtab.cxx. I think this should fix the problem. Murthy From owner-pro64-support@oss.sgi.com Fri Jan 5 10:53:34 2001 Received: by oss.sgi.com id ; Fri, 5 Jan 2001 10:53:14 -0800 Received: from [38.170.141.29] ([38.170.141.29]:27895 "EHLO heart.hq.tensilica.com") by oss.sgi.com with ESMTP id ; Fri, 5 Jan 2001 10:53:01 -0800 Received: (from goodwin@localhost) by heart.hq.tensilica.com (8.9.3/8.9.3) id KAA05294; Fri, 5 Jan 2001 10:52:57 -0800 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: <14934.6152.936589.738478@heart.hq.tensilica.com> Date: Fri, 5 Jan 2001 10:52:56 -0800 (PST) To: Chandrasekhar Murthy Cc: goodwin@tensilica.com, pro64-support@oss.sgi.com Subject: Re: gcc extension to control variable names works with sgiCC, but not sgicc. In-Reply-To: <3A560BFA.5DF2AFFD@sgi.com> References: <14934.1634.87891.229902@heart.hq.tensilica.com> <3A560BFA.5DF2AFFD@sgi.com> 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 Yes, that seems to work. Here's the patch: ==== gccfe/tree_symtab.cxx ==== *************** *** 554,561 **** if (TREE_CODE(decl_node) == ERROR_MARK) exit (RC_USER_ERROR); ! if (DECL_NAME (decl_node)) ! name = IDENTIFIER_POINTER (DECL_NAME (decl_node)); else { DevWarn ("no name for DECL_NODE"); name = "__unknown__"; --- 554,561 ---- if (TREE_CODE(decl_node) == ERROR_MARK) exit (RC_USER_ERROR); ! if (DECL_ASSEMBLER_NAME (decl_node)) ! name = IDENTIFIER_POINTER (DECL_ASSEMBLER_NAME (decl_node)); else { DevWarn ("no name for DECL_NODE"); name = "__unknown__"; Chandrasekhar Murthy writes: > David Goodwin wrote: > > > > gcc allows this use of "asm" to set the name used for a variable in the > > assembly file to be different than the variable name in the source: > > > > char myarray[10] asm ("asmname_myarray"); > > char * > > proc (void) > > { > > return myarray; > > } > > > > > sgicc -v > > SGIcc Compilers: Version 0.01.0-12 > > > sgiCC -v > > SGIcc Compilers: Version 0.01.0-12 > > > > Compiling with sgiCC gives the correct result, i.e. the array is referred > > to as "asmname_myarray". Using sgicc, the array is incorrectly referred to > > as "myarray". > > sgicc is using DECL_NAME instead of DECL_ASSEMBLER_NAME in > tree_symtab.cxx. > > I think this should fix the problem. > > Murthy From owner-pro64-support@oss.sgi.com Mon Jan 8 08:14:47 2001 Received: by oss.sgi.com id ; Mon, 8 Jan 2001 08:14:37 -0800 Received: from mailgate.rz.uni-karlsruhe.de ([129.13.64.97]:65291 "EHLO mailgate.rz.uni-karlsruhe.de") by oss.sgi.com with ESMTP id ; Mon, 8 Jan 2001 08:14:16 -0800 Received: from wbkpc107 (wbkpc107.mach.uni-karlsruhe.de [129.13.175.107]) by mailgate.rz.uni-karlsruhe.de with smtp (Exim 3.16 #1) id 14Few0-0002fb-00; Mon, 08 Jan 2001 17:14:12 +0100 Received: by localhost with Microsoft MAPI; Mon, 8 Jan 2001 17:13:13 +0100 Message-ID: <0FE3AC51A09E584DB0D40763A659269E0D3424@wbkex.wbk.uni-karlsruhe.de> From: Dambacher Reply-To: "ulf.dambacher@mach.uni-karlsruhe.de" To: "'pro64-support@oss.sgi.com'" Subject: -apo to be licesed? Date: Mon, 8 Jan 2001 17:17:09 +0100 Organization: wbk 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 Hi at sgi I' m just testing the sgicc package for linux-itanium and wanted to compile with -apo, but the linker can't find the library libmp. $ sgicc -o compilertest_sgicc_o3_apo -O3 -apo compilertest.c /usr/lib/gcc-lib/ia64-redhat-linux/2.96-ia64-000717/../../../../ia64-redhat-linux/bin/ld: cannot find -lmp collect2: ld returned 1 exit status The manual sais something about 'if you are licensed to use -apo'... Can you give me more information? Thanks! Ulf Dambacher ----- Dipl.-Ing Ulf Dambacher Institut fur Werkzeugmaschinen und Betriebstechnik - Universitat Karlsruhe Tel.: +49 721 608 6022 Fax: +49 721 69 91 53 ulf.dambacher@mach.uni-karlsruhe.de http://www-wbk.mach.uni-karlsruhe.de From owner-pro64-support@oss.sgi.com Mon Jan 8 22:24:51 2001 Received: by oss.sgi.com id ; Mon, 8 Jan 2001 22:24:42 -0800 Received: from www.pspl.co.in ([202.54.11.65]:1542 "EHLO www.pspl.co.in") by oss.sgi.com with ESMTP id ; Mon, 8 Jan 2001 22:24:24 -0800 Received: from kampala (kampala.intranet.pspl.co.in [192.168.1.57]) by www.pspl.co.in (8.11.0/8.11.0) with SMTP id f096VCY05477 for ; Tue, 9 Jan 2001 12:01:12 +0530 From: "Rajalakshmi Iyer" To: Subject: Clarifications required Date: Tue, 9 Jan 2001 11:56:12 +0530 Message-ID: <000001c07a05$10660340$3901a8c0@intranet.pspl.co.in> MIME-Version: 1.0 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: 7bit X-Priority: 3 (Normal) X-MSMail-Priority: Normal X-Mailer: Microsoft Outlook CWS, Build 9.0.2416 (9.0.2910.0) Importance: Normal X-MimeOLE: Produced By Microsoft MimeOLE V5.00.2314.1300 Sender: owner-pro64-support@oss.sgi.com Precedence: bulk Return-Path: X-Orcpt: rfc822;pro64-support-outgoing We are using the SGI Pro64 compiler version 0.01.0-12 and are mainly interested in the intermediate representation produced by the compiler. We need certain clarifications : a) There are certain function calls eg. strlen which have been made intrinsic and the compiler automatically typecasts them from size_t to int. Is there any option to make sure that the return value of such functions is not automatically type-casted. b) We need access to column number information. We are not able to get column number information. The column number field always returns 0. Is there any specific reason why the column number info is not provided for nodes other than statement nodes ? Any help in this regard would be appreciated. Thanks in advance Rajalakshmi. Iyer Member of Technical Staff Persistent Systems. From owner-pro64-support@oss.sgi.com Tue Jan 9 09:59:46 2001 Received: by oss.sgi.com id ; Tue, 9 Jan 2001 09:59:26 -0800 Received: from deliverator.sgi.com ([204.94.214.10]:64567 "EHLO deliverator.sgi.com") by oss.sgi.com with ESMTP id ; Tue, 9 Jan 2001 09:59:19 -0800 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 JAA29386 for ; Tue, 9 Jan 2001 09:58:26 -0800 (PST) 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 JAA61420; Tue, 9 Jan 2001 09:49:40 -0800 (PST) Message-ID: <3A5B4F34.DE716849@sgi.com> Date: Tue, 09 Jan 2001 09:49:40 -0800 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: Rajalakshmi Iyer CC: pro64-support@oss.sgi.com Subject: Re: Clarifications required References: <000001c07a05$10660340$3901a8c0@intranet.pspl.co.in> 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 Rajalakshmi Iyer wrote: > > We are using the SGI Pro64 compiler version 0.01.0-12 and are mainly > interested in the intermediate representation produced by the compiler. > > We need certain clarifications : > a) There are certain function calls eg. strlen which have been made > intrinsic and the compiler automatically typecasts them from size_t to int. > Is there any option to make sure that the return value of such functions is > not automatically type-casted. You can specify -fno-builtin on the commandline to suppress handling of these functions as builtins. > b) We need access to column number information. We are not able to get > column number information. The column number field always returns 0. Is > there any specific reason why the column number info is not provided for > nodes other than statement nodes ? > Could you provide a small example of this. > Any help in this regard would be appreciated. > Thanks in advance > Rajalakshmi. Iyer > Member of Technical Staff > Persistent Systems. Murthy From owner-pro64-support@oss.sgi.com Tue Jan 9 13:30:17 2001 Received: by oss.sgi.com id ; Tue, 9 Jan 2001 13:29:57 -0800 Received: from deliverator.sgi.com ([204.94.214.10]:34395 "EHLO deliverator.sgi.com") by oss.sgi.com with ESMTP id ; Tue, 9 Jan 2001 13:29:37 -0800 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 NAA27271 for ; Tue, 9 Jan 2001 13:28:45 -0800 (PST) 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 NAA63579; Tue, 9 Jan 2001 13:28:08 -0800 (PST) Date: Tue, 9 Jan 2001 13:28:08 -0800 (PST) From: mpm@rohi.engr.sgi.com (Michael Murphy) Message-Id: <200101092128.NAA63579@rohi.engr.sgi.com> To: , "Rajalakshmi Iyer" Subject: Re: Clarifications required Sender: owner-pro64-support@oss.sgi.com Precedence: bulk Return-Path: X-Orcpt: rfc822;pro64-support-outgoing From: "Rajalakshmi Iyer" b) We need access to column number information. We are not able to get column number information. The column number field always returns 0. Is there any specific reason why the column number info is not provided for nodes other than statement nodes ? I don't have the email handy, but this question was previously answered on this list. The root of the problem is that gcc doesn't give column numbers, so we can't. The newest gcc, which integrates cpp into the front end, may be able to do this, but we haven't released with that yet. -- Mike Murphy -- mpm@sgi.com -- quote of the day: -- "The mark of immaturity is a man who insists that -- 'I am the master of my fate, I am the captain of my soul'". (John Sanford) From owner-pro64-support@oss.sgi.com Wed Jan 10 10:32:54 2001 Received: by oss.sgi.com id ; Wed, 10 Jan 2001 10:32:45 -0800 Received: from paco.llnl.gov ([134.9.17.119]:50958 "EHLO paco.llnl.gov") by oss.sgi.com with ESMTP id ; Wed, 10 Jan 2001 10:32:32 -0800 Received: from paco.llnl.gov (IDENT:waynev@localhost [127.0.0.1]) by paco.llnl.gov (SGI-8.9.3/8.9.3) with ESMTP id KAA15614 for ; Wed, 10 Jan 2001 10:32:17 -0800 (PST) Message-ID: <3A5CAAB1.F6E15DA8@paco.llnl.gov> Date: Wed, 10 Jan 2001 10:32:17 -0800 From: Wayne Vieira Organization: SGI X-Mailer: Mozilla 4.75C-SGI [en] (X11; I; IRIX64 6.5 IP30) X-Accept-Language: en MIME-Version: 1.0 To: pro64-support@oss.sgi.com Subject: Optimizations 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 When I compile codes with -Ofast, I get the following error at link time: % 13% sgicc -Ofast hello.c sgicc ERROR: cannot exec /usr/lib/gcc-lib/ia64-sgi-linux/sgicc-1.0/ipa_link The file referenced isn't there. Is it supposed to be? Is this option just not available yet, or was my RPM just missing this file? Wayne From owner-pro64-support@oss.sgi.com Wed Jan 10 10:40:45 2001 Received: by oss.sgi.com id ; Wed, 10 Jan 2001 10:40:25 -0800 Received: from thalia.fm.intel.com ([132.233.247.11]:45830 "EHLO thalia.fm.intel.com") by oss.sgi.com with ESMTP id ; Wed, 10 Jan 2001 10:40:10 -0800 Received: from SMTP (fmsmsxvs05-1.fm.intel.com [132.233.42.205]) by thalia.fm.intel.com (8.9.1a+p1/8.9.1/d: relay.m4,v 1.33 2000/11/21 19:27:27 smothers Exp $) with SMTP id SAA18730; Wed, 10 Jan 2001 18:41:26 GMT Received: from fmsmsx28.fm.intel.com ([132.233.48.28]) by 132.233.48.205 (Norton AntiVirus for Internet Email Gateways 1.0) ; Wed, 10 Jan 2001 18:40:01 0000 (GMT) Received: by fmsmsx28.fm.intel.com with Internet Mail Service (5.5.2650.21) id ; Wed, 10 Jan 2001 10:40:00 -0800 Message-ID: <9287DC1579B0D411AA2F009027F44C3F3B0AF9@FMSMSX41> From: "Chan, Sun C" To: "'Wayne Vieira'" , pro64-support@oss.sgi.com Subject: RE: Optimizations Date: Wed, 10 Jan 2001 10:39:57 -0800 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 -Ofast invokes IPA (interprocedural optimizer) which is not supported at this point. You should be using -O (same as -O2), -O1, -O0 and -O3 -O3 invokes lno (loop nest optimizer). -O2 invokes wopt (global optimizer). Perhaps SGI people can consider releasing the changes needed to make gld to work with IPA so people can use -Ofast at their own risk. There's still might be a problem. I don't remember exactly, -Ofast might also involves feedback which is not released yet. Sun Chan > -----Original Message----- > From: Wayne Vieira [mailto:waynev@paco.llnl.gov] > Sent: Wednesday, January 10, 2001 10:32 AM > To: pro64-support@oss.sgi.com > Subject: Optimizations > > > When I compile codes with -Ofast, I get the following error at link > time: > > % 13% sgicc -Ofast hello.c > sgicc ERROR: cannot exec > /usr/lib/gcc-lib/ia64-sgi-linux/sgicc-1.0/ipa_link > > The file referenced isn't there. Is it supposed to be? Is > this option > just not available yet, or was my RPM just missing this file? > > Wayne > From owner-pro64-support@oss.sgi.com Wed Jan 10 10:41:24 2001 Received: by oss.sgi.com id ; Wed, 10 Jan 2001 10:41:14 -0800 Received: from pneumatic-tube.sgi.com ([204.94.214.22]:14177 "EHLO pneumatic-tube.sgi.com") by oss.sgi.com with ESMTP id ; Wed, 10 Jan 2001 10:41:04 -0800 Received: from sgihud.hudson.sgi.com (sgihud.hudson.sgi.com [169.238.41.4]) by pneumatic-tube.sgi.com (980327.SGI.8.8.8-aspam/980310.SGI-aspam) via ESMTP id KAA02759 for ; Wed, 10 Jan 2001 10:49:49 -0800 (PST) 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 NAA13800; Wed, 10 Jan 2001 13:36:07 -0500 (EST) Date: Wed, 10 Jan 2001 13:36:07 -0500 (EST) From: lesniak@sgihud.hudson.sgi.com (Ken Lesniak) Message-Id: <200101101836.NAA13800@sgihud.hudson.sgi.com> To: pro64-support@oss.sgi.com, Wayne Vieira Subject: Re: Optimizations Reply-To: lesniak@sgihud.hudson.sgi.com References: Sender: owner-pro64-support@oss.sgi.com Precedence: bulk Return-Path: X-Orcpt: rfc822;pro64-support-outgoing >When I compile codes with -Ofast, I get the following error at link >time: > >% 13% sgicc -Ofast hello.c >sgicc ERROR: cannot exec >/usr/lib/gcc-lib/ia64-sgi-linux/sgicc-1.0/ipa_link > >The file referenced isn't there. Is it supposed to be? Is this option >just not available yet, or was my RPM just missing this file? > >Wayne Nothing missing, it's just not supported yet. Ken From owner-pro64-support@oss.sgi.com Wed Jan 10 14:34:25 2001 Received: by oss.sgi.com id ; Wed, 10 Jan 2001 14:34:06 -0800 Received: from louie.udel.edu ([128.175.2.33]:37342 "HELO mail.eecis.udel.edu") by oss.sgi.com with SMTP id ; Wed, 10 Jan 2001 14:33:41 -0800 Received: from ren.eecis.udel.edu by mail.eecis.udel.edu id aa14543; 10 Jan 2001 17:33 EST Date: Wed, 10 Jan 2001 17:33:23 -0500 (EST) From: Ziang Hu To: pro64-support@oss.sgi.com MMDF-Warning: Parse error in original version of preceding line at mail.eecis.udel.edu Subject: any option can force "inline" all the function calls ? 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 just want to know whether Pro64 has options that forces all function calls inlined or not. Thanks Ziang Hu ECE Department, University of Delaware From owner-pro64-support@oss.sgi.com Wed Jan 10 17:29:06 2001 Received: by oss.sgi.com id ; Wed, 10 Jan 2001 17:28:56 -0800 Received: from thalia.fm.intel.com ([132.233.247.11]:9477 "EHLO thalia.fm.intel.com") by oss.sgi.com with ESMTP id ; Wed, 10 Jan 2001 17:28:36 -0800 Received: from SMTP (fmsmsxvs02-1.fm.intel.com [132.233.42.202]) by thalia.fm.intel.com (8.9.1a+p1/8.9.1/d: relay.m4,v 1.33 2000/11/21 19:27:27 smothers Exp $) with SMTP id BAA19236; Thu, 11 Jan 2001 01:29:55 GMT Received: from fmsmsx18.intel.com ([132.233.48.18]) by 132.233.48.202 (Norton AntiVirus for Internet Email Gateways 1.0) ; Thu, 11 Jan 2001 01:28:31 0000 (GMT) Received: by fmsmsx18.fm.intel.com with Internet Mail Service (5.5.2650.21) id ; Wed, 10 Jan 2001 17:28:30 -0800 Message-ID: <9287DC1579B0D411AA2F009027F44C3F3B0B00@FMSMSX41> From: "Chan, Sun C" To: "'Ziang Hu'" , pro64-support@oss.sgi.com Subject: RE: any option can force "inline" all the function calls ? Date: Wed, 10 Jan 2001 17:28:26 -0800 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 You can find out all the flags in common/com/config_xxx.{h,cxx} where xxx are the different components such as lno, wopt etc. CG options can be found in be/cg/cgdriver.cxx though. Sun Chan > -----Original Message----- > From: Ziang Hu [mailto:hu@mail.eecis.udel.edu] > Sent: Wednesday, January 10, 2001 2:33 PM > To: pro64-support@oss.sgi.com > Subject: any option can force "inline" all the function calls ? > > > > I just want to know whether Pro64 has options that forces all function > calls inlined or not. > > Thanks > > Ziang Hu > ECE Department, > University of Delaware > > From owner-pro64-support@oss.sgi.com Wed Jan 10 18:31:37 2001 Received: by oss.sgi.com id ; Wed, 10 Jan 2001 18:31:27 -0800 Received: from pneumatic-tube.sgi.com ([204.94.214.22]:34071 "EHLO pneumatic-tube.sgi.com") by oss.sgi.com with ESMTP id ; Wed, 10 Jan 2001 18:31:12 -0800 Received: from chuchai.engr.sgi.com (chuchai.engr.sgi.com [130.62.180.76]) by pneumatic-tube.sgi.com (980327.SGI.8.8.8-aspam/980310.SGI-aspam) via ESMTP id SAA08466 for ; Wed, 10 Jan 2001 18:39:58 -0800 (PST) mail_from (lilian@chuchai.engr.sgi.com) Received: (from lilian@localhost) by chuchai.engr.sgi.com (980427.SGI.8.8.8/970903.SGI.AUTOCF) id SAA74869; Wed, 10 Jan 2001 18:28:43 -0800 (PST) From: lilian@chuchai.engr.sgi.com (Lilian Leung) Message-Id: <200101110228.SAA74869@chuchai.engr.sgi.com> Subject: Fwd: any option can force "inline" all the function calls? To: hu@mail.eecis.udel.edu Date: Wed, 10 Jan 2001 18:28:43 -0800 (PST) Cc: pro64-support@oss.sgi.com, shinmingliu@yahoo.com X-Mailer: ELM [version 2.5 PL2] 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 > Date: Wed, 10 Jan 2001 17:33:23 -0500 (EST) > From: Ziang Hu > > > I just want to know whether Pro64 has options that forces all function > calls inlined or not. > > Thanks > > Ziang Hu > ECE Department, > University of Delaware > I am not exactly sure what you want but you can try -INLINE:all:preempt:list to see if that works for you. -- Lilian Leung Silicon Graphics Computer Systems From owner-pro64-support@oss.sgi.com Thu Jan 11 12:11:03 2001 Received: by oss.sgi.com id ; Thu, 11 Jan 2001 12:10:44 -0800 Received: from [38.170.141.29] ([38.170.141.29]:42483 "EHLO mail-in.hq.tensilica.com") by oss.sgi.com with ESMTP id ; Thu, 11 Jan 2001 12:10:41 -0800 Received: from tensilica.com (IDENT:jonhsu@leo-home.hq.tensilica.com [192.168.10.103]) by mail-in.hq.tensilica.com (8.9.3/8.9.3) with ESMTP id MAA27585 for ; Thu, 11 Jan 2001 12:10:40 -0800 Message-ID: <3A5E134C.88060A9F@tensilica.com> Date: Thu, 11 Jan 2001 12:10:52 -0800 From: John Hsu X-Mailer: Mozilla 4.76 [en] (X11; U; Linux 2.2.16-3 i686) X-Accept-Language: en MIME-Version: 1.0 To: pro64-support@oss.sgi.com Subject: Header files for building libgcc 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 Hello, I was wondering what standard header files are used when building libgcc for the Pro64 compiler. In particular, I'm trying to build tinfo.cc and I get conflicts with the typeinfo header file, which does not appear to be in the Pro64 source distribution. I think perhaps the standard C++ header files I have are outdated, but I don't know what version is needed. Jon Hsu From owner-pro64-support@oss.sgi.com Thu Jan 11 13:37:53 2001 Received: by oss.sgi.com id ; Thu, 11 Jan 2001 13:37:33 -0800 Received: from sgi.SGI.COM ([192.48.153.1]:56635 "EHLO sgi.com") by oss.sgi.com with ESMTP id ; Thu, 11 Jan 2001 13:37:24 -0800 Received: from cchkms.engr.sgi.com ([130.62.180.48]) by sgi.com (980327.SGI.8.8.8-aspam/980304.SGI-aspam: SGI does not authorize the use of its proprietary systems or networks for unsolicited or bulk email from the Internet.) via ESMTP id NAA03521 for ; Thu, 11 Jan 2001 13:37:21 -0800 (PST) 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 NAA44483; Thu, 11 Jan 2001 13:27:23 -0800 (PST) From: "Ross A. Towle" Message-Id: <10101111327.ZM43788@cchkms.engr.sgi.com> Date: Thu, 11 Jan 2001 13:27:22 -0800 In-Reply-To: John Hsu "Header files for building libgcc" (Jan 11, 12:10pm) References: <3A5E134C.88060A9F@tensilica.com> X-Mailer: Z-Mail (3.2.3 08feb96 MediaMail) To: John Hsu , pro64-support@oss.sgi.com Subject: Re: Header files for building libgcc 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 Header files that are "missing" actually should come from a standard Linux distribution. We use TurboLinux. -Ross From owner-pro64-support@oss.sgi.com Thu Jan 11 13:57:54 2001 Received: by oss.sgi.com id ; Thu, 11 Jan 2001 13:57:34 -0800 Received: from [38.170.141.29] ([38.170.141.29]:60920 "EHLO mail-in.hq.tensilica.com") by oss.sgi.com with ESMTP id ; Thu, 11 Jan 2001 13:57:20 -0800 Received: from tensilica.com (IDENT:jonhsu@leo-home.hq.tensilica.com [192.168.10.103]) by mail-in.hq.tensilica.com (8.9.3/8.9.3) with ESMTP id NAA18708; Thu, 11 Jan 2001 13:57:16 -0800 Message-ID: <3A5E2C45.240B20FB@tensilica.com> Date: Thu, 11 Jan 2001 13:57:27 -0800 From: John Hsu X-Mailer: Mozilla 4.76 [en] (X11; U; Linux 2.2.16-3 i686) X-Accept-Language: en MIME-Version: 1.0 To: "Ross A. Towle" , pro64-support@oss.sgi.com Subject: Re: Header files for building libgcc References: <3A5E134C.88060A9F@tensilica.com> <10101111327.ZM43788@cchkms.engr.sgi.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 "Ross A. Towle" wrote: > Header files that are "missing" actually should come from a standard > Linux distribution. We use TurboLinux. > > -Ross What version of gcc do you have on your system? I think we're using egcs-2.90.29. From owner-pro64-support@oss.sgi.com Fri Jan 12 12:49:30 2001 Received: by oss.sgi.com id ; Fri, 12 Jan 2001 12:49:21 -0800 Received: from cnxtsmtp2.conexant.com ([198.62.9.253]:7671 "EHLO cnxtsmtp2.conexant.com") by oss.sgi.com with ESMTP id ; Fri, 12 Jan 2001 12:49:09 -0800 Received: from npbsmtp1.conexant.com (npbsmtp1.nb.conexant.com [157.152.161.153]) by cnxtsmtp2.conexant.com (8.9.3/8.9.3) with ESMTP id MAA24701 for ; Fri, 12 Jan 2001 12:47:53 -0800 (PST) From: david.kolson@conexant.com Subject: Configure To: pro64-support@oss.sgi.com X-Mailer: Lotus Notes Release 5.0.4 June 8, 2000 Message-ID: Date: Fri, 12 Jan 2001 12:47:46 -0800 X-MIMETrack: Serialize by Router on NPBSMTP1/Server/Conexant(Release 5.0.5 |September 22, 2000) at 01/12/2001 12:49:05 PM 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 Hello, Are there any plans to organize the build structure of the compiler to utilize the Gnu autotools, so that a single 'configure ; gmake' will build the compilers and libraries? Thanks, David Kolson From owner-pro64-support@oss.sgi.com Fri Jan 12 13:15:40 2001 Received: by oss.sgi.com id ; Fri, 12 Jan 2001 13:15:30 -0800 Received: from deliverator.sgi.com ([204.94.214.10]:35430 "EHLO deliverator.sgi.com") by oss.sgi.com with ESMTP id ; Fri, 12 Jan 2001 13:15:16 -0800 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 NAA25269 for ; Fri, 12 Jan 2001 13:14:23 -0800 (PST) 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 NAA44398 for pro64-support@oss.sgi.com; Fri, 12 Jan 2001 13:10:33 -0800 (PST) From: "Ross A. Towle" Message-Id: <10101121310.ZM46924@cchkms.engr.sgi.com> Date: Fri, 12 Jan 2001 13:10:32 -0800 In-Reply-To: david.kolson@conexant.com "Configure" (Jan 12, 12:47pm) References: X-Mailer: Z-Mail (3.2.3 08feb96 MediaMail) To: pro64-support@oss.sgi.com Subject: Re: Configure 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 The answer is no, we are not switching to Configure. Right now, it is just as simple. Just gmake in the targ* directories (depending upon your target). We need the capability of building multiple targets from the same sources without one build messing up another build (which is a side effect of Configure). -Ross From owner-pro64-support@oss.sgi.com Fri Jan 12 18:56:52 2001 Received: by oss.sgi.com id ; Fri, 12 Jan 2001 18:56:42 -0800 Received: from deliverator.sgi.com ([204.94.214.10]:44143 "EHLO deliverator.sgi.com") by oss.sgi.com with ESMTP id ; Fri, 12 Jan 2001 18:56:20 -0800 Received: from harpoon.engr.sgi.com (harpoon.engr.sgi.com [130.62.41.49]) by deliverator.sgi.com (980309.SGI.8.8.8-aspam-6.2/980310.SGI-aspam) via ESMTP id SAA17809 for ; Fri, 12 Jan 2001 18:55:26 -0800 (PST) mail_from (jkingdon@embryo.engr.sgi.com) Received: from embryo.engr.sgi.com (kingdon.engr.sgi.com [130.62.180.80]) by harpoon.engr.sgi.com (980427.SGI.8.8.8/970903.SGI.AUTOCF) via ESMTP id SAA89181; Fri, 12 Jan 2001 18:55:02 -0800 (PST) Message-ID: <3A5FC385.AD71F4AF@embryo.engr.sgi.com> Date: Fri, 12 Jan 2001 18:55:01 -0800 From: James Kingdon X-Mailer: Mozilla 4.7C-SGI [en] (X11; I; IRIX 6.5 IP32) X-Accept-Language: en MIME-Version: 1.0 To: pro64-support@oss.sgi.com CC: jkingdon@cthulhu.engr.sgi.com Subject: Re: Configure References: <10101121310.ZM46924@cchkms.engr.sgi.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 "Ross A. Towle" wrote: > We need the > capability of building multiple targets from the same > sources without one build messing up another build (which is > a side effect of Configure). GNU configure handles this with the --srcdir argument to configure (which is the default if you specify a configure script in another directory other than where you are running configure from). For example, here's how you can build both a ia32 native GCC compiler and a ia32 x ia64 compiler from the same source tree (I think I even did it with the source tree being read-only although there are a few minor annoyances related to yacc and makeinfo and such, which generate non-platform-specific generated files): On an ia32 system, go to OBJDIR32 and issue the following commands: SRC/gcc/configure --enable-languages=c++ --disable-nls make bootstrap On an ia32 system, go to OBJDIR64 and issue the following commands: SRC/gcc/configure --enable-languages=c++ --target=ia64-unknown-linux --disable-nls make (for those within SGI, this is an excerpt from http://sahara/ia64/merge.html although it could go on a public site if anyone actually cares). Having said all that, I'm not sure I see a lot of reasons to autoconfiscate Pro64. But it does have this particular feature. (There's also the "hmm, interesting idea, you feel like working up some patches to see what would be involved in autoconfiscating it?" response.... :-)). From owner-pro64-support@oss.sgi.com Fri Jan 12 19:04:52 2001 Received: by oss.sgi.com id ; Fri, 12 Jan 2001 19:04:42 -0800 Received: from pneumatic-tube.sgi.com ([204.94.214.22]:60439 "EHLO pneumatic-tube.sgi.com") by oss.sgi.com with ESMTP id ; Fri, 12 Jan 2001 19:04:33 -0800 Received: from harpoon.engr.sgi.com (harpoon.engr.sgi.com [130.62.41.49]) by pneumatic-tube.sgi.com (980327.SGI.8.8.8-aspam/980310.SGI-aspam) via ESMTP id TAA02948 for ; Fri, 12 Jan 2001 19:13:21 -0800 (PST) mail_from (jkingdon@embryo.engr.sgi.com) Received: from embryo.engr.sgi.com (kingdon.engr.sgi.com [130.62.180.80]) by harpoon.engr.sgi.com (980427.SGI.8.8.8/970903.SGI.AUTOCF) via ESMTP id TAA89251; Fri, 12 Jan 2001 19:03:15 -0800 (PST) Message-ID: <3A5FC572.7E15221C@embryo.engr.sgi.com> Date: Fri, 12 Jan 2001 19:03:14 -0800 From: James Kingdon X-Mailer: Mozilla 4.7C-SGI [en] (X11; I; IRIX 6.5 IP32) X-Accept-Language: en MIME-Version: 1.0 To: John Hsu CC: pro64-support@oss.sgi.com Subject: Re: Header files for building libgcc References: <3A5E134C.88060A9F@tensilica.com> <10101111327.ZM43788@cchkms.engr.sgi.com> <3A5E2C45.240B20FB@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 John Hsu wrote: > What version of gcc do you have on your system? I think we're using > egcs-2.90.29. On IA32, bash$ gcc -v Reading specs from /usr/lib/gcc-lib/i686-pc-linux-gnu/2.96/specs gcc version 2.96 19991114 (experimental) bash$ I believe that the Red Hat 6.x compiler (egcs 1.1.2 based) is too old, GCC 2.95 might be OK (although don't quote me on that); I'm not sure of the exact cutoff. If the egcs vs. gcc numbers are confusing, there's a table somewhere on http://gcc.gnu.org/ which gives the correspondance between the egcs 1.1.* numbers and the 2.90.* style numbers. From owner-pro64-support@oss.sgi.com Sat Jan 13 07:57:35 2001 Received: by oss.sgi.com id ; Sat, 13 Jan 2001 07:57:25 -0800 Received: from Cantor.suse.de ([194.112.123.193]:61451 "HELO Cantor.suse.de") by oss.sgi.com with SMTP id ; Sat, 13 Jan 2001 07:57:05 -0800 Received: from Hermes.suse.de (Hermes.suse.de [194.112.123.136]) by Cantor.suse.de (Postfix) with ESMTP id 582021E269; Sat, 13 Jan 2001 16:57:00 +0100 (MET) Date: Sat, 13 Jan 2001 16:56:57 +0100 From: Andi Kleen To: "Ross A. Towle" Cc: pro64-support@oss.sgi.com Subject: Re: Configure Message-ID: <20010113165657.A14072@gruyere.muc.suse.de> References: <10101121310.ZM46924@cchkms.engr.sgi.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline User-Agent: Mutt/1.2.5i In-Reply-To: <10101121310.ZM46924@cchkms.engr.sgi.com>; from rat@cchkms.engr.sgi.com on Fri, Jan 12, 2001 at 01:10:32PM -0800 Sender: owner-pro64-support@oss.sgi.com Precedence: bulk Return-Path: X-Orcpt: rfc822;pro64-support-outgoing On Fri, Jan 12, 2001 at 01:10:32PM -0800, Ross A. Towle wrote: > The answer is no, we are not switching to Configure. > Right now, it is just as simple. Just gmake in the targ* > directories (depending upon your target). We need the > capability of building multiple targets from the same > sources without one build messing up another build (which is > a side effect of Configure). GNU autoconf supports building in separate directories without touching the original source tree. I do this all the time when building gcc. -Andi From owner-pro64-support@oss.sgi.com Mon Jan 15 08:45:14 2001 Received: by oss.sgi.com id ; Mon, 15 Jan 2001 08:45:04 -0800 Received: from [38.170.141.29] ([38.170.141.29]:19190 "EHLO heart.hq.tensilica.com") by oss.sgi.com with ESMTP id ; Mon, 15 Jan 2001 08:44:42 -0800 Received: (from goodwin@localhost) by heart.hq.tensilica.com (8.9.3/8.9.3) id IAA27693; Mon, 15 Jan 2001 08:44:38 -0800 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: <14947.10486.211698.511712@heart.hq.tensilica.com> Date: Mon, 15 Jan 2001 08:44:38 -0800 (PST) To: pro64-support@oss.sgi.com Subject: followup on inliner bug 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 In October I sent in the following bug. It was not fixed in pro64-0.01.0-12. Were you able to reproduce it? What is the status on a fix? Thanks, David Compiling the attached example at -O3 causes the inliner to do a couple of things that appear wrong. sgicc -O3 -S alias.c static int x; static int y; static int *get_x(){ return &x;} static int *get_y(){ int *p = &y; return p;} int foo(int); int test() { (*(get_x())) = 8; foo(x); (*(get_y())) = 9; return x + y; } 1. Correctness problem: The inliner changes some symbol table flags for x and y that cause alias problems in be. Diffing the whirl after gccfe and inline, it seems the only thing the inliner does is change 'x's "addr_saved" to "addr_passed", and add "addr_passed" to 'y'. Removing 'x's "addr_saved" causes be to think that the load of 'x' for the call to "foo" can't alias with the store of 8 into get_x's return value. Of course that is wrong. 2. Performance problem: Neither get_x or get_y are inlined!! From owner-pro64-support@oss.sgi.com Thu Jan 18 00:20:47 2001 Received: by oss.sgi.com id ; Thu, 18 Jan 2001 00:20:28 -0800 Received: from narkis.wisdom.weizmann.ac.il ([132.76.80.32]:15593 "EHLO narkis.wisdom.weizmann.ac.il") by oss.sgi.com with ESMTP id ; Thu, 18 Jan 2001 00:19:54 -0800 Received: from wisdom.weizmann.ac.il (amir8-pc.wisdom.weizmann.ac.il [132.76.81.32]) by narkis.wisdom.weizmann.ac.il (8.9.3/8.9.3) with ESMTP id KAA00603 for ; Thu, 18 Jan 2001 10:19:19 +0200 (IST) Message-ID: <3A66A678.C60CEF33@wisdom.weizmann.ac.il> Date: Thu, 18 Jan 2001 10:16:56 +0200 From: raya Organization: Weizmann Institute of Science X-Mailer: Mozilla 4.75 [en] (WinNT; U) X-Accept-Language: il,en MIME-Version: 1.0 To: pro64 Subject: Examples 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 Hello, I'm looking for a data base of examples that demonstrate the compiler optimizations, so I know ALL optimizer modules are used, and on which part of the code. Regards, -- Raya Leviathan Tel. 972-8-9344208 (office) Tel. 972-3-6358481 (home) Email: raya@wisdom.weizmann.ac.il From owner-pro64-support@oss.sgi.com Thu Jan 18 16:24:51 2001 Received: by oss.sgi.com id ; Thu, 18 Jan 2001 16:24:42 -0800 Received: from mta5.snfc21.pbi.net ([206.13.28.241]:61635 "EHLO mta5.snfc21.pbi.net") by oss.sgi.com with ESMTP id ; Thu, 18 Jan 2001 16:24:31 -0800 Received: from ted ([63.206.179.222]) by mta5.snfc21.pbi.net (Sun Internet Mail Server sims.3.5.2000.01.05.12.18.p9) with SMTP id <0G7D00MNIVI7KJ@mta5.snfc21.pbi.net> for pro64-support@oss.sgi.com; Thu, 18 Jan 2001 16:18:07 -0800 (PST) Date: Thu, 18 Jan 2001 16:28:56 -0800 From: Wayne Vieira Subject: sgif90 -cpp errors In-reply-to: <20010113165657.A14072@gruyere.muc.suse.de> To: pro64-support@oss.sgi.com Message-id: MIME-version: 1.0 X-Mailer: Microsoft Outlook IMO, Build 9.0.2416 (9.0.2911.0) Content-type: text/plain; charset="us-ascii" Content-transfer-encoding: 7bit Importance: Normal X-MSMail-Priority: Normal X-MimeOLE: Produced By Microsoft MimeOLE V5.50.4133.2400 X-Priority: 3 (Normal) Sender: owner-pro64-support@oss.sgi.com Precedence: bulk Return-Path: X-Orcpt: rfc822;pro64-support-outgoing The man page for sgif90 says that the -cpp option calls the Bcpp compiler (?) which locate could not find on my Big Sur system. (/lib/cpp was there) Using this command option generates an error. sgif90 -cpp test.f I don't have access to my system at the moment %$#@ Rolling Calif blackouts!! The error is similar to: don't recognize '-traditional-cpp' As in the example above, I am not using that option, so I must guess that it is being passed onto the /lib/cpp. My questiosn are: 1. Is the man page reference to Bcpp an error? If not where does one look for Bcpp? 2. using -traditional-cpp on gcc will produce a similar error. But no error appears on sgicc. I have to guess that sgif90 is calling gcc as it's pre-processor. Is that correct? Why isn't sgif90 using sgicc?? Thanks, Wayne Wayne Vieira, SGI Federal Systems Engineer, RHCE, Linux/IRIX http://www.sgi.com/linux http://hose3.corp.sgi.com/linuxsgi From owner-pro64-support@oss.sgi.com Thu Jan 18 16:38:42 2001 Received: by oss.sgi.com id ; Thu, 18 Jan 2001 16:38:33 -0800 Received: from deliverator.sgi.com ([204.94.214.10]:40047 "EHLO deliverator.sgi.com") by oss.sgi.com with ESMTP id ; Thu, 18 Jan 2001 16:38:04 -0800 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 QAA07022 for ; Thu, 18 Jan 2001 16:37:07 -0800 (PST) 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 QAA01360; Thu, 18 Jan 2001 16:36:06 -0800 (PST) Date: Thu, 18 Jan 2001 16:36:06 -0800 (PST) From: mpm@rohi.engr.sgi.com (Michael Murphy) Message-Id: <200101190036.QAA01360@rohi.engr.sgi.com> To: pro64-support@oss.sgi.com, Wayne Vieira Subject: Re: sgif90 -cpp errors Sender: owner-pro64-support@oss.sgi.com Precedence: bulk Return-Path: X-Orcpt: rfc822;pro64-support-outgoing From: Wayne Vieira The man page for sgif90 says that the -cpp option calls the Bcpp compiler (?) which locate could not find on my Big Sur system. (/lib/cpp was there) I think this is a typo. Should be that it calls "cpp". The B is probably from the nroff font stuff. Using this command option generates an error. sgif90 -cpp test.f I don't have access to my system at the moment %$#@ Rolling Calif blackouts!! The error is similar to: don't recognize '-traditional-cpp' This was talked about before. It is due to the recent upgrade of gcc, which no longer accepts that option. The newest, unreleased sgif90 passes -cpp rather than -traditional-cpp to gcc. As in the example above, I am not using that option, so I must guess that it is being passed onto the /lib/cpp. My questiosn are: 1. Is the man page reference to Bcpp an error? If not where does one look for Bcpp? Yes, see above. 2. using -traditional-cpp on gcc will produce a similar error. But no error appears on sgicc. I have to guess that sgif90 is calling gcc as it's pre-processor. Is that correct? Yes. Why isn't sgif90 using sgicc?? Because sgicc also uses gcc as its preprocessor :-) -- Mike Murphy -- mpm@sgi.com -- quote of the day: -- "Power is like a narcotic taken to diminish the pain of an empty life" -- (David Petrescue) From owner-pro64-support@oss.sgi.com Mon Jan 22 09:04:02 2001 Received: by oss.sgi.com id ; Mon, 22 Jan 2001 09:03:52 -0800 Received: from lslsun.epfl.ch ([128.178.150.20]:30341 "EHLO lslsun.epfl.ch") by oss.sgi.com with ESMTP id ; Mon, 22 Jan 2001 09:03:43 -0800 Received: from epfl.ch (johaenni@lslsun8 [128.178.150.31]) by lslsun.epfl.ch (8.8.X/EPFL-8.1a) with ESMTP id SAA24623 for ; Mon, 22 Jan 2001 18:03:40 +0100 (MET) Message-ID: <3A6C67EC.6770D697@epfl.ch> Date: Mon, 22 Jan 2001 18:03:40 +0100 From: Jacques-Olivier Haenni Organization: Swiss Federal Institute of Technology, Lausanne (EPFL) X-Mailer: Mozilla 4.75 [en] (X11; U; SunOS 5.6 sun4u) X-Accept-Language: en, fr-CH, fr MIME-Version: 1.0 To: pro64-support@oss.sgi.com Subject: Problem building Pro64 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 Hi, I'm desesperatly trying to build SGI Pro64 compiler from the sources on my IA-32 Linux box. I've been able to solve many problems, but I now ask for help for the following one: When compiling the ipa, I get the following error : g++ -D_SGI_SOURCE -D_LANGUAGE_C_PLUS_PLUS -funsigned-char -D__GNU_BUG_WORKAROUND -D_NOTHREADS -c -D__MIPS_AND_IA64_ELF_H -DTARG_IA64 -D_NEW_SYMTAB -D_MERGE_SUMMARY_ST_IDX_=1 -DBACK_END -DSTD_MONGOOSE_LOC='"/usr/lib32/cmplrs"' -D_SUPPORT_IPA -D_MULTIGOT -D_NEW_SYMTAB -D_MERGE_SUMMARY_ST_IDX_=1 -DBACK_END -DSTD_MONGOOSE_LOC='"/usr/lib32/cmplrs"' -D_SUPPORT_IPA -DMONGOOSE_BE -I../../ipa/common -I../../ipa/local -I../../ipa/main/optimize -I../../ipa/main/analyze -I../../be/lno -I../../be/com -I../../common/com -I../../be/region -I../../be/opt -I../../common/com/ia64 -I../../common/util -I../../fake_ld/common -I../../fake_ld -I../be -I../include/libelf -I../include -O0 -D_MIPSEL -D_LONGLONG -D_MIPS_SZINT=32 -D_MIPS_SZPTR=32 -D_MIPS_SZLONG=32 -MD ../../ipa/common/ipc_dst_merge.cxx -o ipc_dst_merge.o In file included from ../../be/com/fb_whirl.h:67, from ../../ipa/local/ipl_summary.h:76, from ../../ipa/main/analyze/ipa_cg.h:65, from ../../ipa/common/ipc_dst_merge.cxx:51: ../../common/com/fb_info.h:142:2: warning: multi-line comment ../../common/com/fb_info.h:145:2: warning: multi-line comment /tmp/ccBVmgBK.s: Assembler messages: /tmp/ccBVmgBK.s:165: Error: invalid character '=' in operand 1 /tmp/ccBVmgBK.s:446: Error: invalid character '=' in operand 1 /tmp/ccBVmgBK.s:518: Error: invalid character '=' in operand 1 [about 20 similar errors have been cut off] I've tried to debug this and I've noticed that the generated assembly file contains calls such as 'call operator!=' which cause this problem. My question is now : How can I get rid off this error ? Do I have to upgrade or install another version (which one ?) of binutils or gcc ? Could you tell me which versions of gnu tools are able to correctly compile Pro64 ? Is somebody else experiencing the same problem ? My current configuration is RedHat 7.0 with - gcc version 2.96 20000731 (Red Hat Linux 7.0) - GNU assembler version 2.10.91 (i386-redhat-linux) using BFD version 2.10.1.0.2 Does somebody has any clue for me to be able to compile Pro64 ? Thanks a lot for your help. Jacques-Olivier -- Jacques-Olivier Haenni http://lslwww.epfl.ch/~johaenni/ Jacques-Olivier.Haenni@epfl.ch Logic Systems Laboratory Swiss Federal Institute of Technology (EPFL) | Tel: (+41 21) 693 66 30 1015 Lausanne - Switzerland | Fax: (+41 21) 693 37 05 From owner-pro64-support@oss.sgi.com Mon Jan 22 12:07:55 2001 Received: by oss.sgi.com id ; Mon, 22 Jan 2001 12:07:36 -0800 Received: from deliverator.sgi.com ([204.94.214.10]:25694 "EHLO deliverator.sgi.com") by oss.sgi.com with ESMTP id ; Mon, 22 Jan 2001 12:07:18 -0800 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 MAA29269 for ; Mon, 22 Jan 2001 12:06:20 -0800 (PST) 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 MAA12974; Mon, 22 Jan 2001 12:05:38 -0800 (PST) Date: Mon, 22 Jan 2001 12:05:38 -0800 (PST) From: mpm@rohi.engr.sgi.com (Michael Murphy) Message-Id: <200101222005.MAA12974@rohi.engr.sgi.com> To: pro64-support@oss.sgi.com, Jacques-Olivier Haenni Subject: Re: Problem building Pro64 Sender: owner-pro64-support@oss.sgi.com Precedence: bulk Return-Path: X-Orcpt: rfc822;pro64-support-outgoing From: Jacques-Olivier Haenni /tmp/ccBVmgBK.s: Assembler messages: /tmp/ccBVmgBK.s:165: Error: invalid character '=' in operand 1 /tmp/ccBVmgBK.s:446: Error: invalid character '=' in operand 1 /tmp/ccBVmgBK.s:518: Error: invalid character '=' in operand 1 [about 20 similar errors have been cut off] I've tried to debug this and I've noticed that the generated assembly file contains calls such as 'call operator!=' which cause this problem. I tried compiling the same file with your same command line, and did not see this problem in the .s file. Must be a problem in gcc rather than the assembler. Actually looks like perhaps a problem in the name mangling. I use gcc version 2.96 19991114 (experimental) which is just slightly older than what you are using, but I would hope that gcc hasn't regressed like this. Since this is the ia32 compiler, you should try to create a small test case and send it to cygnus. -- Mike Murphy -- mpm@sgi.com -- quote of the day: -- "What does the LORD require of you? -- To act justly and to love mercy and to walk humbly with your God." -- (Micah 6:8, The Bible) From owner-pro64-support@oss.sgi.com Mon Jan 22 14:11:47 2001 Received: by oss.sgi.com id ; Mon, 22 Jan 2001 14:11:28 -0800 Received: from deliverator.sgi.com ([204.94.214.10]:59680 "EHLO deliverator.sgi.com") by oss.sgi.com with ESMTP id ; Mon, 22 Jan 2001 14:11:22 -0800 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 OAA29024 for ; Mon, 22 Jan 2001 14:10:24 -0800 (PST) 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 OAA20041; Mon, 22 Jan 2001 14:05:30 -0800 (PST) Message-ID: <3A6CAEAA.84CC181E@sgi.com> Date: Mon, 22 Jan 2001 14:05:30 -0800 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: Jacques-Olivier Haenni CC: pro64-support@oss.sgi.com Subject: Re: Problem building Pro64 References: <3A6C67EC.6770D697@epfl.ch> 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 Jacques-Olivier Haenni wrote: > > Hi, > > I'm desesperatly trying to build SGI Pro64 compiler from the sources on my > IA-32 Linux box. I've been able to solve many problems, but I now ask for help > for the following one: > > When compiling the ipa, I get the following error : > > g++ -D_SGI_SOURCE -D_LANGUAGE_C_PLUS_PLUS -funsigned-char > -D__GNU_BUG_WORKAROUND -D_NOTHREADS -c -D__MIPS_AND_IA64_ELF_H > -DTARG_IA64 -D_NEW_SYMTAB -D_MERGE_SUMMARY_ST_IDX_=1 -DBACK_END > -DSTD_MONGOOSE_LOC='"/usr/lib32/cmplrs"' -D_SUPPORT_IPA -D_MULTIGOT > -D_NEW_SYMTAB -D_MERGE_SUMMARY_ST_IDX_=1 -DBACK_END > -DSTD_MONGOOSE_LOC='"/usr/lib32/cmplrs"' -D_SUPPORT_IPA -DMONGOOSE_BE > -I../../ipa/common -I../../ipa/local -I../../ipa/main/optimize > -I../../ipa/main/analyze -I../../be/lno -I../../be/com -I../../common/com > -I../../be/region -I../../be/opt -I../../common/com/ia64 -I../../common/util > -I../../fake_ld/common -I../../fake_ld -I../be -I../include/libelf > -I../include -O0 -D_MIPSEL -D_LONGLONG -D_MIPS_SZINT=32 -D_MIPS_SZPTR=32 > -D_MIPS_SZLONG=32 -MD ../../ipa/common/ipc_dst_merge.cxx -o ipc_dst_merge.o > In file included from ../../be/com/fb_whirl.h:67, > from ../../ipa/local/ipl_summary.h:76, > from ../../ipa/main/analyze/ipa_cg.h:65, > from ../../ipa/common/ipc_dst_merge.cxx:51: > ../../common/com/fb_info.h:142:2: warning: multi-line comment > ../../common/com/fb_info.h:145:2: warning: multi-line comment > /tmp/ccBVmgBK.s: Assembler messages: > /tmp/ccBVmgBK.s:165: Error: invalid character '=' in operand 1 > /tmp/ccBVmgBK.s:446: Error: invalid character '=' in operand 1 > /tmp/ccBVmgBK.s:518: Error: invalid character '=' in operand 1 > [about 20 similar errors have been cut off] > > I've tried to debug this and I've noticed that the generated assembly file > contains calls such as 'call operator!=' which cause this problem. > > My question is now : How can I get rid off this error ? Do I have to upgrade > or install another version (which one ?) of binutils or gcc ? Could you tell > me which versions of gnu tools are able to correctly compile Pro64 ? Is > somebody else experiencing the same problem ? > > My current configuration is RedHat 7.0 with > - gcc version 2.96 20000731 (Red Hat Linux 7.0) > - GNU assembler version 2.10.91 (i386-redhat-linux) > using BFD version 2.10.1.0.2 > > Does somebody has any clue for me to be able to compile Pro64 ? > > Thanks a lot for your help. > > Jacques-Olivier > It is a bug in g++. You might want to preprocess the file and send it to Cygnus. I am not sure why you are trying to build ipa. All components necessary for using ipa have not yet been released. In case you still want to build ipa, you could do the following. Modify the g++ commandline emitted by make to generate the assembly language file ipc_dst_merge.s Edit the assembly file replacing operator!= by __ne operator< by __lt and then generate the object file as follows as -x -o ipc_dst_merge.o ipc_dst_merge.s Murthy From owner-pro64-support@oss.sgi.com Wed Jan 24 02:18:15 2001 Received: by oss.sgi.com id ; Wed, 24 Jan 2001 02:18:06 -0800 Received: from elin.scali.no ([195.139.250.10]:15633 "EHLO elin.scali.no") by oss.sgi.com with ESMTP id ; Wed, 24 Jan 2001 02:17:48 -0800 Received: from scali.no (IDENT:eisa@pc-8.scali.no [172.16.0.108]) by elin.scali.no (8.9.3/8.8.7) with ESMTP id LAA32591 for ; Wed, 24 Jan 2001 11:17:44 +0100 Message-ID: <3A6EABC8.C0C732CA@scali.no> Date: Wed, 24 Jan 2001 11:17:44 +0100 From: Einar Saltnes Organization: Scali AS X-Mailer: Mozilla 4.72 [en] (X11; U; Linux 2.2.14-6.0.2smp i686) X-Accept-Language: en MIME-Version: 1.0 To: pro64-support@oss.sgi.com Subject: SGI pro64 on RH7.0 beta distro Content-Type: text/plain; charset=iso-8859-15 Content-Transfer-Encoding: 7bit Sender: owner-pro64-support@oss.sgi.com Precedence: bulk Return-Path: X-Orcpt: rfc822;pro64-support-outgoing Hi, I installed the pro64-0.01.0-12.ia64.rpm from your web-site, but the compiler would not run hands-off on the RH7.0 ia64 beta distro. I did follow the installation procedure from your web-site. I am not using NUE, only native Itaniums (BigSurs). : eisa@scali9-13:eisa; cat /etc/issue Red Hat Linux release 7.0 (Guinness) Kernel 2.4.0-0.43smp on a 2-processor ia64 : eisa@scali9-13:eisa; uname -a Linux scali9-13 2.4.0-0.43smp #1 SMP Mon Dec 4 18:37:06 EST 2000 ia64 unknown I am using the RH 7.0 ia64 beta distro "out of the box". Will there be an updated pro64 rpm-package compatible with the RH7.0, or is there something I am missing? We are also interested in using Fortran, besides C and C++. I believe TurboLinux is running the SGI Pro64 compiler on their disto with similar versions of glibc-2.2 and kernel-2.4.0-test11 as the RH7.0. Thank you in advance for advice to get the pro64 up and running on RH7.0, Einar Saltnes -- | Einar Saltnes | Scali AS | | mailto:eisa@scali.no | http://www.scali.com | | Phone:+47 22 62 89 50 | Scalable Linux Systems | | Fax :+47 22 62 89 51 | and Technology | From owner-pro64-support@oss.sgi.com Wed Jan 24 15:28:09 2001 Received: by oss.sgi.com id ; Wed, 24 Jan 2001 15:27:49 -0800 Received: from deliverator.sgi.com ([204.94.214.10]:51766 "EHLO deliverator.sgi.com") by oss.sgi.com with ESMTP id ; Wed, 24 Jan 2001 15:27:28 -0800 Received: from harpoon.engr.sgi.com (harpoon.engr.sgi.com [130.62.41.49]) by deliverator.sgi.com (980309.SGI.8.8.8-aspam-6.2/980310.SGI-aspam) via ESMTP id PAA15146 for ; Wed, 24 Jan 2001 15:26:29 -0800 (PST) mail_from (jkingdon@embryo.engr.sgi.com) Received: from embryo.engr.sgi.com (kingdon.engr.sgi.com [130.62.180.80]) by harpoon.engr.sgi.com (980427.SGI.8.8.8/970903.SGI.AUTOCF) via ESMTP id PAA49155; Wed, 24 Jan 2001 15:23:23 -0800 (PST) Message-ID: <3A6F63B9.A1893AF5@embryo.engr.sgi.com> Date: Wed, 24 Jan 2001 15:22:33 -0800 From: James Kingdon X-Mailer: Mozilla 4.7C-SGI [en] (X11; I; IRIX 6.5 IP32) X-Accept-Language: en MIME-Version: 1.0 To: Einar Saltnes CC: pro64-support@oss.sgi.com Subject: Re: SGI pro64 on RH7.0 beta distro References: <3A6EABC8.C0C732CA@scali.no> 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 Einar Saltnes wrote: > I installed the pro64-0.01.0-12.ia64.rpm from your web-site, but the > compiler would not run hands-off on the RH7.0 ia64 beta distro. I did > follow the installation procedure from your web-site. > I am not using NUE, only native Itaniums (BigSurs). Well, what happened when you tried? And do you know what needs to change in the RPM? > Will there be an updated pro64 rpm-package compatible with the RH7.0 If you make one, yes. If not, I don't know. I'm not working on it and I'm not sure I know of someone who is. From owner-pro64-support@oss.sgi.com Wed Jan 24 23:59:52 2001 Received: by oss.sgi.com id ; Wed, 24 Jan 2001 23:59:42 -0800 Received: from narkis.wisdom.weizmann.ac.il ([132.76.80.32]:56294 "EHLO narkis.wisdom.weizmann.ac.il") by oss.sgi.com with ESMTP id ; Wed, 24 Jan 2001 23:59:23 -0800 Received: from wisdom.weizmann.ac.il (amir8-pc.wisdom.weizmann.ac.il [132.76.81.32]) by narkis.wisdom.weizmann.ac.il (8.9.3/8.9.3) with ESMTP id JAA10650 for ; Thu, 25 Jan 2001 09:59:01 +0200 (IST) Message-ID: <3A6FDC3F.F3FBDD2@wisdom.weizmann.ac.il> Date: Thu, 25 Jan 2001 09:56:47 +0200 From: raya Organization: Weizmann Institute of Science X-Mailer: Mozilla 4.75 [en] (WinNT; U) X-Accept-Language: il,en MIME-Version: 1.0 To: pro64 Subject: whirle via RTL 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 Hello, Is there any similarity between WHIRLE and GNU RTL ? -- Raya Leviathan Tel. 972-8-9344208 (office) Tel. 972-3-6358481 (home) Email: raya@wisdom.weizmann.ac.il From owner-pro64-support@oss.sgi.com Thu Jan 25 04:51:24 2001 Received: by oss.sgi.com id ; Thu, 25 Jan 2001 04:51:04 -0800 Received: from elin.scali.no ([195.139.250.10]:40971 "EHLO elin.scali.no") by oss.sgi.com with ESMTP id ; Thu, 25 Jan 2001 04:50:55 -0800 Received: from scali.no (IDENT:eisa@pc-8.scali.no [172.16.0.108]) by elin.scali.no (8.9.3/8.8.7) with ESMTP id NAA10985; Thu, 25 Jan 2001 13:50:13 +0100 Message-ID: <3A702105.7CC76BF1@scali.no> Date: Thu, 25 Jan 2001 13:50:13 +0100 From: Einar Saltnes Organization: Scali AS X-Mailer: Mozilla 4.72 [en] (X11; U; Linux 2.2.14-6.0.2smp i686) X-Accept-Language: en MIME-Version: 1.0 To: James Kingdon CC: pro64-support@oss.sgi.com Subject: Re: SGI pro64 on RH7.0 beta distro References: <3A6EABC8.C0C732CA@scali.no> <3A6F63B9.A1893AF5@embryo.engr.sgi.com> Content-Type: text/plain; charset=iso-8859-15 Content-Transfer-Encoding: 7bit Sender: owner-pro64-support@oss.sgi.com Precedence: bulk Return-Path: X-Orcpt: rfc822;pro64-support-outgoing Hi James Kingdon, I finally found what I was missing. I had to install the RedHat/RPMS/compat-libs-7.0-1.i386.rpm from the RH7.0 distribution. After installing pro64-0.01.0-12.ia64.rpm I get : eisa@scali9-12:eisa; sgicc -v SGIcc Compilers: Version 0.01.0-12 Thank you! Best regards Einar Saltnes -- | Einar Saltnes | Scali AS | | mailto:eisa@scali.no | http://www.scali.com | | Phone:+47 22 62 89 50 | Scalable Linux Systems | | Fax :+47 22 62 89 51 | and Technology | James Kingdon wrote: > > Einar Saltnes wrote: > > I installed the pro64-0.01.0-12.ia64.rpm from your web-site, but the > > compiler would not run hands-off on the RH7.0 ia64 beta distro. I did > > follow the installation procedure from your web-site. > > I am not using NUE, only native Itaniums (BigSurs). > > Well, what happened when you tried? And do you know what > needs to change in the RPM? > sgicc would not run and there was no message > > Will there be an updated pro64 rpm-package compatible with the RH7.0 > > If you make one, yes. If not, I don't know. I'm not working on it > and I'm not sure I know of someone who is. From owner-pro64-support@oss.sgi.com Thu Jan 25 07:36:27 2001 Received: by oss.sgi.com id ; Thu, 25 Jan 2001 07:36:17 -0800 Received: from mail.ee.gatech.edu ([130.207.230.10]:21127 "EHLO mail.ee.gatech.edu") by oss.sgi.com with ESMTP id ; Thu, 25 Jan 2001 07:36:06 -0800 Received: from ece-10-21.ece-int.gatech.edu.ece.gatech.edu (IDENT:hardnett@ece-10-21.ece-int.gatech.edu [192.168.10.21]) by mail.ee.gatech.edu (8.11.0/8.11.0) with SMTP id f0PFa4R27385 for ; Thu, 25 Jan 2001 10:36:05 -0500 (EST) From: "Charles R. Hardnett" MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Message-ID: <14960.18373.364140.887236@ece-10-21.ece-int.gatech.edu> Date: Thu, 25 Jan 2001 10:35:33 -0500 (EST) To: pro64-support@oss.sgi.com Subject: Dumping .B files X-Mailer: VM 6.72 under 21.1 (patch 12) "Channel Islands" 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 Hi, I know that I saw a message about this some time ago, but I can not find that message now. BTW is there a site that is keeping a monthly archive of these messages? but the real question is how do you dump the .B files that are needed by the be component? I need to execute the be by itself to debug something, and it needs the .B file. BTW, does anyone have any useful tips on debugging the phases of the compiler using gdb? Thanks, Charles -- -------------------------------------------------------------------- Charles R. Hardnett www.spelman.edu/~hardnett From owner-pro64-support@oss.sgi.com Thu Jan 25 07:56:27 2001 Received: by oss.sgi.com id ; Thu, 25 Jan 2001 07:56:17 -0800 Received: from pneumatic-tube.sgi.com ([204.94.214.22]:43903 "EHLO pneumatic-tube.sgi.com") by oss.sgi.com with ESMTP id ; Thu, 25 Jan 2001 07:56:01 -0800 Received: from sgihud.hudson.sgi.com (sgihud.hudson.sgi.com [169.238.41.4]) by pneumatic-tube.sgi.com (980327.SGI.8.8.8-aspam/980310.SGI-aspam) via ESMTP id IAA04377 for ; Thu, 25 Jan 2001 08:05:01 -0800 (PST) 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 KAA62738; Thu, 25 Jan 2001 10:54:29 -0500 (EST) Date: Thu, 25 Jan 2001 10:54:29 -0500 (EST) From: lesniak@sgihud.hudson.sgi.com (Ken Lesniak) Message-Id: <200101251554.KAA62738@sgihud.hudson.sgi.com> To: pro64-support@oss.sgi.com, hardnett@cc.gatech.edu Subject: Re: Dumping .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 Hi, >I know that I saw a message about this some time ago, but I can not >find that message now. BTW is there a site that is keeping a monthly >archive of these messages? No idea, but that would be nice. >but the real question is how do you dump the .B files that are needed ir_b2a will dump the .B file. I don't believe we ship the binary and that you'd have to build it (in the ir_tools build sub-directory). >by the be component? I need to execute the be by itself to debug >something, and it needs the .B file. BTW, does anyone have any useful >tips on debugging the phases of the compiler using gdb? Yes. - Run the compiler with "-keep -show" on the test case. - Capture the command line from 'be'. - Run gdb on 'be' (will probably need to point LD_LIBRARY_PATH at where cg.so, be.so, etc are). - issue the gdb 'run' command with the captured command line. Now, presumably you'd like to set breakpoints. 'be' dynamically loads the components necessary based on the command line you gave it. For example, if you do -O2 it will load wopt.so and cg.so. If you do -O3 it will load lno.so in addition. Because these dso's are dynamically linked, gdb won't know the symbols. There's a kludge in 'be' to help. Right after 'be' loads the dso's mention above, it calls the empty function be_debug(). So if you set a breakpoint on be_debug, and give the gdb 'run' command, once you hit the breakpoint you'll have access to the symbols in the dynamically loaded dso's. >Thanks, > >Charles Ken From owner-pro64-support@oss.sgi.com Thu Jan 25 08:07:18 2001 Received: by oss.sgi.com id ; Thu, 25 Jan 2001 08:07:08 -0800 Received: from mail.ee.gatech.edu ([130.207.230.10]:53648 "EHLO mail.ee.gatech.edu") by oss.sgi.com with ESMTP id ; Thu, 25 Jan 2001 08:06:52 -0800 Received: from ece-10-21.ece-int.gatech.edu.ece.gatech.edu (IDENT:hardnett@ece-10-21.ece-int.gatech.edu [192.168.10.21]) by mail.ee.gatech.edu (8.11.0/8.11.0) with SMTP id f0PG6eR00465; Thu, 25 Jan 2001 11:06:40 -0500 (EST) From: "Charles R. Hardnett" MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Message-ID: <14960.20209.412682.21320@ece-10-21.ece-int.gatech.edu> Date: Thu, 25 Jan 2001 11:06:09 -0500 (EST) To: lesniak@sgihud.hudson.sgi.com Cc: pro64-support@oss.sgi.com, hardnett@cc.gatech.edu Subject: Re: Dumping .B files In-Reply-To: <200101251554.KAA62738@sgihud.hudson.sgi.com> References: <200101251554.KAA62738@sgihud.hudson.sgi.com> X-Mailer: VM 6.72 under 21.1 (patch 12) "Channel Islands" 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 Thanks Ken, I do have ir_b2a compiled already. I must have done it a long time ago. but it requires a binary IR file to convert. So how do I get that binary IR file? Charles Ken Lesniak writes: > Hi, > > >I know that I saw a message about this some time ago, but I can not > >find that message now. BTW is there a site that is keeping a monthly > >archive of these messages? > > No idea, but that would be nice. > > >but the real question is how do you dump the .B files that are needed > > ir_b2a will dump the .B file. I don't believe we ship the binary > and that you'd have to build it (in the ir_tools build sub-directory). > > >by the be component? I need to execute the be by itself to debug > >something, and it needs the .B file. BTW, does anyone have any useful > >tips on debugging the phases of the compiler using gdb? -- -------------------------------------------------------------------- Charles R. Hardnett www.spelman.edu/~hardnett From owner-pro64-support@oss.sgi.com Thu Jan 25 08:16:18 2001 Received: by oss.sgi.com id ; Thu, 25 Jan 2001 08:15:57 -0800 Received: from pneumatic-tube.sgi.com ([204.94.214.22]:57094 "EHLO pneumatic-tube.sgi.com") by oss.sgi.com with ESMTP id ; Thu, 25 Jan 2001 08:15:48 -0800 Received: from gaea.engr.sgi.com (gaea.engr.sgi.com [130.62.180.97]) by pneumatic-tube.sgi.com (980327.SGI.8.8.8-aspam/980310.SGI-aspam) via ESMTP id IAA08835 for ; Thu, 25 Jan 2001 08:24:48 -0800 (PST) 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 IAA39768; Thu, 25 Jan 2001 08:10:43 -0800 (PST) Date: Thu, 25 Jan 2001 08:10:43 -0800 (PST) From: murthy@gaea.engr.sgi.com (Chandrasekhar Murthy) Message-Id: <200101251610.IAA39768@gaea.engr.sgi.com> To: lesniak@sgihud.hudson.sgi.com, hardnett@cc.gatech.edu Subject: Re: Dumping .B files Cc: hardnett@cc.gatech.edu, pro64-support@oss.sgi.com References: <200101251554.KAA62738@sgihud.hudson.sgi.com> Sender: owner-pro64-support@oss.sgi.com Precedence: bulk Return-Path: X-Orcpt: rfc822;pro64-support-outgoing When you compile your source program with the -keep option it will generate a file with ".B" extension. You can then use ir_b2a to convert the file. Murthy From owner-pro64-support@oss.sgi.com Thu Jan 25 08:24:28 2001 Received: by oss.sgi.com id ; Thu, 25 Jan 2001 08:24:18 -0800 Received: from pneumatic-tube.sgi.com ([204.94.214.22]:3592 "EHLO pneumatic-tube.sgi.com") by oss.sgi.com with ESMTP id ; Thu, 25 Jan 2001 08:24:07 -0800 Received: from sgihud.hudson.sgi.com (sgihud.hudson.sgi.com [169.238.41.4]) by pneumatic-tube.sgi.com (980327.SGI.8.8.8-aspam/980310.SGI-aspam) via ESMTP id IAA05754 for ; Thu, 25 Jan 2001 08:33:06 -0800 (PST) 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 LAA62708; Thu, 25 Jan 2001 11:22:33 -0500 (EST) Date: Thu, 25 Jan 2001 11:22:33 -0500 (EST) From: lesniak@sgihud.hudson.sgi.com (Ken Lesniak) Message-Id: <200101251622.LAA62708@sgihud.hudson.sgi.com> To: hardnett@cc.gatech.edu Subject: Re: Dumping .B files Cc: hardnett@cc.gatech.edu, pro64-support@oss.sgi.com Reply-To: lesniak@sgihud.hudson.sgi.com References: <200101251554.KAA62738@sgihud.hudson.sgi.com> Sender: owner-pro64-support@oss.sgi.com Precedence: bulk Return-Path: X-Orcpt: rfc822;pro64-support-outgoing >Thanks Ken, > >I do have ir_b2a compiled already. I must have done it a long time >ago. but it requires a binary IR file to convert. So how do I get that >binary IR file? > >Charles I thought since you asked how to dump a .B file that you already had one! THe .B file is the binary IR output from the front-end. As Murthy pointed out, -keep will leave the .B file behind. Ken From owner-pro64-support@oss.sgi.com Thu Jan 25 09:46:38 2001 Received: by oss.sgi.com id ; Thu, 25 Jan 2001 09:46:28 -0800 Received: from deliverator.sgi.com ([204.94.214.10]:31795 "EHLO deliverator.sgi.com") by oss.sgi.com with ESMTP id ; Thu, 25 Jan 2001 09:46:01 -0800 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 JAA06939 for ; Thu, 25 Jan 2001 09:45:03 -0800 (PST) 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 JAA18207; Thu, 25 Jan 2001 09:38:08 -0800 (PST) From: "Ross A. Towle" Message-Id: <10101250938.ZM18293@cchkms.engr.sgi.com> Date: Thu, 25 Jan 2001 09:38:07 -0800 In-Reply-To: raya "whirle via RTL" (Jan 25, 9:56am) References: <3A6FDC3F.F3FBDD2@wisdom.weizmann.ac.il> X-Mailer: Z-Mail (3.2.3 08feb96 MediaMail) To: raya , pro64 Subject: Re: whirle via RTL 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 WHIRL and GNU RTL are related only in the sense they are both intermediate representations of programs. After that they are quite different in terms of the semantic information available, levels of abstraction, annotation, ... One could probably generate GNU RTL from WHIRL but not vice versa. -Ross From owner-pro64-support@oss.sgi.com Mon Jan 29 10:02:59 2001 Received: by oss.sgi.com id ; Mon, 29 Jan 2001 10:02:50 -0800 Received: from [38.170.141.29] ([38.170.141.29]:47356 "EHLO mail-in.hq.tensilica.com") by oss.sgi.com with ESMTP id ; Mon, 29 Jan 2001 10:02:31 -0800 Received: from tensilica.com (IDENT:jonhsu@leo-home.hq.tensilica.com [192.168.10.103]) by mail-in.hq.tensilica.com (8.9.3/8.9.3) with ESMTP id KAA13631 for ; Mon, 29 Jan 2001 10:02:30 -0800 Message-ID: <3A75B017.40B9C5C1@tensilica.com> Date: Mon, 29 Jan 2001 10:02:00 -0800 From: John Hsu X-Mailer: Mozilla 4.76 [en] (X11; U; Linux 2.2.16-3 i686) X-Accept-Language: en MIME-Version: 1.0 To: pro64-support@oss.sgi.com Subject: Problems building libstdc++-v3 Content-Type: multipart/mixed; boundary="------------736B8CE9A3DAA3923DB83D74" 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. --------------736B8CE9A3DAA3923DB83D74 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Hi, I've been trying to build libstdc++-v3 with a version of the Pro64 compiler and I've been mostly successful, but I ran across the following error while building locale.cc /tmp/ccs.KlJwfL: Assembler messages: /tmp/ccs.KlJwfL:215: Error: attempt to .org backwards ignored /tmp/ccs.KlJwfL:225: Error: attempt to .org backwards ignored /tmp/ccs.KlJwfL:243: Error: attempt to .org backwards ignored /tmp/ccs.KlJwfL:196: Error: attempt to .org/.space backwards? (-4) /tmp/ccs.KlJwfL:320: Internal error! Assertion failure in write_contents at ../../src/gas/write.c line 1130. I reduced the file to a small test case, which I have attached to this e-mail. It can be compiled with: sgiCC -Wno-format -W -Wwrite-strings -Winline -fexceptions -c locale.i -Wp,-fno-implicit-templates I believe what's happening is that the front end is inserting extra data in the initializers for the static class members, which causes the .org directives to be incorrect. The extra data appears to come from template instatiations that occur while the initializer is being emitted to the .B file. In addition, if I remove the #pragma interface line from the file, the problem goes away. I may have removed too much from the test case, but I was wondering if others have seen this problem and if anyone has found a solution. --------------736B8CE9A3DAA3923DB83D74 Content-Type: text/plain; charset=us-ascii; name="locale.i" Content-Transfer-Encoding: 7bit Content-Disposition: inline; filename="locale.i" #pragma interface "typeinfo" namespace std { template struct char_traits { }; } namespace std { template > class istreambuf_iterator; } namespace std { template > class num_get; class locale { public: typedef int category; class facet; class id; class _Impl; friend class _Impl; static const category none = 0; static const category collate = 0x0100; static const category ctype = 0x0200; static const category monetary = 0x0400; static const category numeric = 0x0800; static const category time = 0x1000; static const category messages = 0x2000; static const category all = (collate | ctype | monetary | numeric | time | messages); }; class locale::_Impl { private: static const locale::id* const _S_id_collate[]; static const locale::id* const _S_id_ctype[]; static const locale::id* const _S_id_monetary[]; static const locale::id* const _S_id_numeric[]; static const locale::id* const _S_id_time[]; static const locale::id* const _S_id_messages[]; static const locale::id* const* const _S_facet_categories[]; }; class locale::id { }; } namespace std { enum _Ios_Iostate { _S_ios_iostate_end = 1<<16 }; class ios_base { public: typedef _Ios_Iostate iostate; locale _M_ios_locale; }; } namespace std { template class num_get { public: typedef _CharT char_type; typedef _InIter iter_type; typedef char_traits<_CharT> __traits_type; static locale::id id; protected: virtual iter_type do_get(iter_type, iter_type, ios_base&, ios_base::iostate&, bool&) const; }; struct money_base { enum part { none, space, symbol, sign, value }; struct pattern { char field[4]; }; static const pattern _S_default_pattern; }; template class moneypunct : public money_base { public: static locale::id id; }; } namespace std { const locale::id* const locale::_Impl::_S_id_monetary[] = { &moneypunct::id, &std::moneypunct::id, 0 }; const locale::id* const locale::_Impl::_S_id_numeric[] = { &num_get::id, 0 }; const locale::id* const* const locale::_Impl::_S_facet_categories[] = { locale::_Impl::_S_id_collate, locale::_Impl::_S_id_ctype, locale::_Impl::_S_id_monetary, locale::_Impl::_S_id_numeric, locale::_Impl::_S_id_time, locale::_Impl::_S_id_messages, 0 }; const money_base::pattern money_base::_S_default_pattern = {{symbol, sign, none, value}};; } --------------736B8CE9A3DAA3923DB83D74-- From owner-pro64-support@oss.sgi.com Mon Jan 29 11:06:00 2001 Received: by oss.sgi.com id ; Mon, 29 Jan 2001 11:05:50 -0800 Received: from deliverator.sgi.com ([204.94.214.10]:28711 "EHLO deliverator.sgi.com") by oss.sgi.com with ESMTP id ; Mon, 29 Jan 2001 11:05:33 -0800 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 LAA27301 for ; Mon, 29 Jan 2001 11:04:34 -0800 (PST) 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 LAA25155 for pro64-support@oss.sgi.com; Mon, 29 Jan 2001 11:05:10 -0800 (PST) From: "Ross A. Towle" Message-Id: <10101291105.ZM25335@cchkms.engr.sgi.com> Date: Mon, 29 Jan 2001 11:05:09 -0800 In-Reply-To: John Hsu "Problems building libstdc++-v3" (Jan 29, 10:02am) References: <3A75B017.40B9C5C1@tensilica.com> X-Mailer: Z-Mail (3.2.3 08feb96 MediaMail) To: pro64-support@oss.sgi.com Subject: Re: Problems building libstdc++-v3 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 Thanks for the report (and the small test case). The bug has been filed. -Ross From owner-pro64-support@oss.sgi.com Mon Jan 29 11:11:49 2001 Received: by oss.sgi.com id ; Mon, 29 Jan 2001 11:11:31 -0800 Received: from pneumatic-tube.sgi.com ([204.94.214.22]:38427 "EHLO pneumatic-tube.sgi.com") by oss.sgi.com with ESMTP id ; Mon, 29 Jan 2001 11:11:23 -0800 Received: from gaea.engr.sgi.com (gaea.engr.sgi.com [130.62.180.97]) by pneumatic-tube.sgi.com (980327.SGI.8.8.8-aspam/980310.SGI-aspam) via ESMTP id LAA07392 for ; Mon, 29 Jan 2001 11:20:27 -0800 (PST) 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 LAA71036; Mon, 29 Jan 2001 11:10:48 -0800 (PST) Message-ID: <3A75C038.4EC16FD0@sgi.com> Date: Mon, 29 Jan 2001 11:10:48 -0800 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: John Hsu CC: pro64-support@oss.sgi.com Subject: Re: Problems building libstdc++-v3 References: <3A75B017.40B9C5C1@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 The problem appears to be that the INITOs for _Q26locale5_Impl._S_id_monetary _10money_base._S_default_pattern land up being the same. The former should have symbolic offsets and the latter should be only the VALs. I will take a look at this. Thanks, Murthy From owner-pro64-support@oss.sgi.com Mon Jan 29 19:17:12 2001 Received: by oss.sgi.com id ; Mon, 29 Jan 2001 19:17:02 -0800 Received: from [38.170.141.29] ([38.170.141.29]:42488 "EHLO mail-in.hq.tensilica.com") by oss.sgi.com with ESMTP id ; Mon, 29 Jan 2001 19:16:53 -0800 Received: from tensilica.com (IDENT:jonhsu@leo-home.hq.tensilica.com [192.168.10.103]) by mail-in.hq.tensilica.com (8.9.3/8.9.3) with ESMTP id TAA11557 for ; Mon, 29 Jan 2001 19:16:52 -0800 Message-ID: <3A763206.D05EA80@tensilica.com> Date: Mon, 29 Jan 2001 19:16:22 -0800 From: John Hsu X-Mailer: Mozilla 4.76 [en] (X11; U; Linux 2.2.16-3 i686) X-Accept-Language: en MIME-Version: 1.0 To: pro64-support@oss.sgi.com Subject: STLport Content-Type: multipart/mixed; boundary="------------E029F187AA9EFB68888A753B" 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. --------------E029F187AA9EFB68888A753B Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Hi, I'm trying to run the test cases from STLport with a version of the Pro64 compiler and I've hit the following front-end bug: sgiCC -O -c hmap1.cpp /usr/lib/gcc-lib/ia64-hp-linux/2.9-ia64-000216/../../../../include/g++-3/ropeimpl.h: In method `void _Rope_RopeRep<_CharT, _Alloc>::_M_free_tree () [with _CharT = char, _Alloc = allocator]': /usr/lib/gcc-lib/ia64-hp-linux/2.9-ia64-000216/../../../../include/g++-3/stl_rope.h:504: instantiated from here /usr/lib/gcc-lib/ia64-hp-linux/2.9-ia64-000216/../../../../include/g++-3/ropeimpl.h:334: Internal compiler error. /usr/lib/gcc-lib/ia64-hp-linux/2.9-ia64-000216/../../../../include/g++-3/ropeimpl.h:334: Please submit a full bug report. /usr/lib/gcc-lib/ia64-hp-linux/2.9-ia64-000216/../../../../include/g++-3/ropeimpl.h:334: See for instructions. It appears to be a problem in the GNU front end, not the Pro64 compiler. The STLport website says that STLport should work with egcs 2.81 and above, although I'm not actually building STLport, just the test cases. I wasn't sure if I should send this to gnu.org directly or pass it on through you. --------------E029F187AA9EFB68888A753B Content-Type: text/plain; charset=us-ascii; name="hmap1.cpp" Content-Transfer-Encoding: 7bit Content-Disposition: inline; filename="hmap1.cpp" #ifdef MAIN #define hmap1_test main #endif // STLport regression testsuite component. // To compile as a separate example, please #define MAIN. #include #include #include #if !defined (STLPORT) || defined(__STL_USE_NAMESPACES) using namespace std; #endif int hmap1_test(int, char**) { cout<<"Results of hmap1_test:"<, equal_to > maptype; maptype m; // Store mappings between roman numerals and decimals. m['l'] = "50"; m['x'] = "20"; // Deliberate mistake. m['v'] = "5"; m['i'] = "1"; cout << "m['x'] = " << m['x'] << endl; m['x'] = "10"; // Correct mistake. cout << "m['x'] = " << m['x'] << endl; cout << "m['z'] = " << m['z'] << endl; // Note default value is added. cout << "m.count('z') = " << m.count('z') << endl; pair p = m.insert(pair('c', crope("100"))); if(p.second) cout << "First insertion successful" << endl; p = m.insert(pair('c', crope("100"))); if(p.second) cout << "Second insertion successful" << endl; else cout << "Existing pair " <<(*(p.first)).first << " -> " <<(*(p.first)).second << endl; return 0; } --------------E029F187AA9EFB68888A753B-- From owner-pro64-support@oss.sgi.com Tue Jan 30 08:35:48 2001 Received: by oss.sgi.com id ; Tue, 30 Jan 2001 08:35:39 -0800 Received: from pneumatic-tube.sgi.com ([204.94.214.22]:43535 "EHLO pneumatic-tube.sgi.com") by oss.sgi.com with ESMTP id ; Tue, 30 Jan 2001 08:35:22 -0800 Received: from gaea.engr.sgi.com (gaea.engr.sgi.com [130.62.180.97]) by pneumatic-tube.sgi.com (980327.SGI.8.8.8-aspam/980310.SGI-aspam) via ESMTP id IAA07743 for ; Tue, 30 Jan 2001 08:44:27 -0800 (PST) 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 IAA78838; Tue, 30 Jan 2001 08:34:35 -0800 (PST) Message-ID: <3A76ED1A.748CE6FC@sgi.com> Date: Tue, 30 Jan 2001 08:34:34 -0800 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: John Hsu CC: pro64-support@oss.sgi.com Subject: Re: STLport References: <3A763206.D05EA80@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 You are correct. It is a problem in the GNU frontend. Please report the problem using the link mentioned in the error report. Thanks, Murthy