From Mon May 19 10:42:15 2008 Date: Mon, 19 May 2008 10:42:15 -0700 (PDT) From: Postmaster Subject: Message from mail server Content-Length: 94 Mime-Version: 1.0 Status: RO X-IMAP: 1211218934 28 Delete. This is a system message. --END+PSEUDO-- From sacadmin Sun May 11 11:57:31 2008 Received: from sunraf.sfbay.sun.com (sunraf.SFBay.Sun.COM [129.146.177.11]) by sac.sfbay.sun.com (8.13.8+Sun/8.13.8) with ESMTP id m4BIvVY0014688; Sun, 11 May 2008 11:57:31 -0700 (PDT) Received: from sunraf.sfbay.sun.com (localhost [127.0.0.1]) by sunraf.sfbay.sun.com (8.14.3+Sun/8.14.3) with ESMTP id m4BIvUsf101239; Sun, 11 May 2008 11:57:30 -0700 (PDT) Received: (from raf@localhost) by sunraf.sfbay.sun.com (8.14.3+Sun/8.14.3/Submit) id m4BIvTIo101235; Sun, 11 May 2008 11:57:30 -0700 (PDT) Date: Sun, 11 May 2008 11:57:30 -0700 (PDT) From: "Roger A. Faulkner" Message-Id: <200805111857.m4BIvTIo101235@sunraf.sfbay.sun.com> To: PSARC-record@sac.sfbay.sun.com Subject: expunge synonyms.h [PSARC/2008/309 FastTrack timeout 05/19/2008] Content-Length: 556 Status: RO X-Status: $$$$ X-UID: 0000000001 Template Version: @(#)sac_nextcase 1.66 04/17/08 SMI This information is Copyright 2008 Sun Microsystems 1. Introduction 1.1. Project/Component Working Name: expunge synonyms.h 1.2. Name of Document Author/Supplier: Author: Roger Faulkner 1.3 Date of This Document: 11 May, 2008 4. Technical Description See the case directory for more detail 6. Resources and Schedule 6.4. Steering Committee requested information 6.4.1. Consolidation C-team Name: OS/NET 6.5. ARC review type: FastTrack 6.6. ARC Exposure: open From Roger.Faulkner@sun.com Sun May 11 12:14:21 2008 Received: from jurassic-x4600.sfbay.sun.com (cretaceous.SFBay.Sun.COM [129.146.17.59]) by sac.sfbay.sun.com (8.13.8+Sun/8.13.8) with ESMTP id m4BJELVp014844 for ; Sun, 11 May 2008 12:14:21 -0700 (PDT) Received: from jurassic-x4600.sfbay.sun.com (jurassic-x4600.sfbay.sun.com [129.146.17.59]) by jurassic-x4600.sfbay.sun.com (8.14.2+Sun/8.14.2) with SMTP id m4BJELn2338278; Sun, 11 May 2008 12:14:21 -0700 (PDT) Message-Id: <200805111914.m4BJELn2338278@jurassic-x4600.sfbay.sun.com> Date: Sun, 11 May 2008 12:14:21 -0700 (PDT) From: "Roger A. Faulkner" Reply-To: "Roger A. Faulkner" Subject: PSARC/2008/309 - expunge synonyms.h To: psarc-ext@sac.sfbay.sun.com Cc: darrin.johnson@sun.com, rod.evans@sun.com, bart.smaalders@sun.com MIME-Version: 1.0 Content-Type: TEXT/plain; charset=us-ascii Content-MD5: xu7TNr1LCSu5f6dgw2kvlg== X-Mailer: dtmail 1.3.0 @(#)CDE Version 1.7_88 SunOS 5.11 i86pc i386 Content-Length: 6974 Status: RO X-Status: $$$$ X-UID: 0000000002 I am sponsoring this fast-track case for myself. This proposal will time out 05/19/2008 It's not clear that this actually needs PSARC review since no official interfaces are affected, but it is a change to the system for old stuff that has been pervasive in Solaris since Solaris 2.0 and I'd like to get your opinions. There are no exported interfaces involved, so there is no stability level issue. The proposed release binding (if required) is "minor release" since no one, certainly not I, wants this to be back-ported to Solaris 10. ============================================================== History Since the dawn of time (that is, since SVR4 came from New Jersey and became Solaris 2.0), there have been synonym symbols in the system-supplied libraries, most prevalently in libc. Synonym symbols come in pairs like this: opendir _opendir These two functions are identical. Not identical contents but exactly the same location in memory. This is most commonly achieved with #pragma weak directives. (Back in the days of static system libraries, the notion of 'strong' and 'weak' symbols had some meaning, but with dynamically linked system libraries the distinction is lost and the two symbols are just alternate names for the same function.) The motivation for synonym symbols is to allow applications and system libraries to define and use the same symbol names without hurting with each other. Standards-conforming applications are constrained not to define or call functions with leading underscores in their names. The system libraries are not so constrained. So, for example, libc defines opendir() and it calls opendir() internally in its implementation of ttyname(). If an application defines its own version of opendir() (which does something entirely different from opendir() in libc) and the application also calls ttyname() in libc, which instance of opendir() will be called by libc's ttyname()? The solution to this problem has historically (since the dawn of time) been synonym symbols. Internally, libc calls the _opendir() symbol. This way, the application can define and use its opendir() function for whatever it chooses and there will be no problem when the application calls ttyname(). The implementation of this scheme for system libraries involves special header files, usr/src/lib/libc/inc/synonyms.h for libc and usr/src/lib/common/inc/c_synonyms.h for all other system libraries. These header files contain definitions like these: #define openat _openat #define opendir _opendir #define openlog _openlog #define open _open There are over one thousand definitions in each file. For the scheme to work, every source file in libc must do this: #include "synonyms.h" before including any other header file. Neither synonyms.h nor c_synonyms.h are delivered as header files for Solaris. They are OS/NET consolidation-private files. The source code for ttyname() in libc shows a call to opendir(), but by virtue of including "synonyms.h", the compiled code actually calls _opendir(). For the scheme to work for all system libraries delivered from the OS/NET consolidation, all other libraries are required to include "c_synonyms.h" before any other header file in every one of their source files. Some of them do (libnsl, libresolve, libgen plus a few others) but the vast majority of them do not. So the synonyms scheme is a failure for most libraries, due to historical omission. ============================================================== The better solution So much for history. Now for today's situation. The change for this bug report was recently integrated into /ws/onnv-gate: 4947191 OSNet should use direct bindings This causes all OS/NET libraries, including libc, to be built using direct binding (-B direct). As a result, all symbols referenced by a library are bound either to the definition within itself or to the definition within its direct dependencies. This solves the problem for which synonyms were invented in an elegant and global way. Now all libraries in OS/NET have the benefit of not damaging or being damaged by an application's redefinition of a symbol that the library calls. The application calls its instance and the library calls its instance, even though both instances have the same name. (Excepting, of course, symbols in the malloc() family and other symbols explicitly defined to be interposable symbols by default. These have all been marked NODIRECT in the appropriate mapfiles. [The devil is in the details.]) This renders the whole synonyms scheme obsolete. The final stage of this evolution is to expunge the various synonyms.h files: usr/src/cmd/sgs/libelf/common/syn.h usr/src/cmd/sgs/rtld/common/_synonyms.h usr/src/lib/common/inc/c_synonyms.h usr/src/lib/libc/inc/synonyms.h usr/src/lib/libcrypt/inc/des_synonyms.h usr/src/lib/libgen/inc/gen_synonyms.h usr/src/lib/libtsnet/common/synonyms.h and to fix all of the source files that include them. Also, most of the leading-underscore symbols involved in the synonyms scheme are versioned SUNWprivate_1.1 (over 600 of them). Many others are versioned public, some because they are defined in the SVID, hence the ABI, but others because of mistakes made along the way wherein developers mistakenly thought that the leading-underscore symbols should be versioned the same as the primary symbols. In an ideal world, the SUNWprivate symbols could just be eliminated since nothing outside of OS/NET can legitimately call them, and they are not part of the ABI. This would reduce the number of symbols exported from libc by 25%. The non-private symbols cannot be eliminated (although I wish they could). However, a search of all of the executables and libraries not delivered from OS/NET and also all the executables and libraries in /usr/dist reveals that /lib/libm.so.2, some older versions of the Studio compiler/debugger components, and some ancient programs in /usr/dist call 55 of the SUNWprivate symbols. So these 55 must be retained, hopefully to be removed in some future update of Solaris. ============================================================== The proposal This brings us to the change being proposed by this PSARC case: 1. Delete these header files from the OS/NET consolidation: usr/src/cmd/sgs/libelf/common/syn.h usr/src/cmd/sgs/rtld/common/_synonyms.h usr/src/lib/common/inc/c_synonyms.h usr/src/lib/libc/inc/synonyms.h usr/src/lib/libcrypt/inc/des_synonyms.h usr/src/lib/libgen/inc/gen_synonyms.h usr/src/lib/libtsnet/common/synonyms.h and fix all of the source files that include them. 2. Eliminate all of the SUNWprivate leading-underscore symbols involved in the synonyms scheme except for the 55 that were discovered to be used outside of OS/NET. Roger Faulkner From Andrew.Gabriel@Sun.COM Sun May 11 16:56:53 2008 Received: from dm-sfbay-02.sfbay.sun.com (dm-sfbay-02.SFBay.Sun.COM [129.146.11.31]) by sac.sfbay.sun.com (8.13.8+Sun/8.13.8) with ESMTP id m4BNurRn018974 for ; Sun, 11 May 2008 16:56:53 -0700 (PDT) Received: from gmp-eb-inf-2.sun.com (gmp-eb-inf-2.EU.Sun.COM [192.18.6.24]) by dm-sfbay-02.sfbay.sun.com (8.13.8+Sun/8.13.8/ENSMAIL,v2.2) with ESMTP id m4BNuqWI013919 for ; Sun, 11 May 2008 16:56:52 -0700 (PDT) Received: from fe-emea-10.sun.com (gmp-eb-lb-2-fe1.eu.sun.com [192.18.6.10]) by gmp-eb-inf-2.sun.com (8.13.7+Sun/8.12.9) with ESMTP id m4BNukIF000236 for ; Sun, 11 May 2008 23:56:47 GMT Received: from conversion-daemon.fe-emea-10.sun.com by fe-emea-10.sun.com (Sun Java System Messaging Server 6.2-8.04 (built Feb 28 2007)) id <0K0Q00D01A7HSZ00@fe-emea-10.sun.com> (original mail from Andrew.Gabriel@Sun.COM) for psarc-ext@sac.sfbay.sun.com; Mon, 12 May 2008 00:56:46 +0100 (BST) Received: from [192.168.2.137] ([62.31.246.243]) by fe-emea-10.sun.com (Sun Java System Messaging Server 6.2-8.04 (built Feb 28 2007)) with ESMTPSA id <0K0Q00K5WAI9DB70@fe-emea-10.sun.com>; Mon, 12 May 2008 00:56:44 +0100 (BST) Date: Mon, 12 May 2008 00:56:31 +0100 From: Andrew Gabriel Subject: Re: PSARC/2008/309 - expunge synonyms.h In-reply-to: <200805111914.m4BJELn2338278@jurassic-x4600.sfbay.sun.com> Sender: Andrew.Gabriel@Sun.COM To: "Roger A. Faulkner" Cc: psarc-ext@sac.sfbay.sun.com, Darrin.Johnson@Sun.COM, Rod.Evans@Sun.COM, Bart.Smaalders@Sun.COM Message-id: <482787AF.6050809@sun.com> MIME-version: 1.0 Content-type: text/plain; format=flowed; charset=ISO-8859-1 Content-transfer-encoding: 7BIT References: <200805111914.m4BJELn2338278@jurassic-x4600.sfbay.sun.com> User-Agent: Thunderbird 2.0.0.12 (X11/20080310) Content-Length: 1275 Status: RO X-Status: $$$$ X-UID: 0000000003 Roger A. Faulkner wrote: > However, a search of all of the executables and libraries > not delivered from OS/NET and also all the executables and > libraries in /usr/dist reveals that /lib/libm.so.2, some > older versions of the Studio compiler/debugger components, > and some ancient programs in /usr/dist call 55 of the > SUNWprivate symbols. So these 55 must be retained, > hopefully to be removed in some future update of Solaris. Few thoughts... 1) Feels kind of selfish to leave symbols there because it would be painful for us to remove them, without applying similar considerations to others' code. 2) Could we supply a sticky plaster, possibly in the form of an LD_PRELOAD library, which would reinstate the symbols for when some unsuspecting customer trips over some application which relied on them? Perhaps this could even be a workaround which avoids retaining the 55 symbols? 3) When appcert(1) is run, it (optionally) mails off the results to Sun, apparently so we can check for things like this before removing symbols. I've no idea where that data ends up in Sun, but seems like it might be a good idea to scan it to see if we can identify a worse problem than currently anticipated. -- Regards Andrew From gdamore@sun.com Sun May 11 20:18:36 2008 Received: from dm-sfbay-01.sfbay.sun.com (dm-sfbay-01.SFBay.Sun.COM [129.145.155.118]) by sac.sfbay.sun.com (8.13.8+Sun/8.13.8) with ESMTP id m4C3IaML022512 for ; Sun, 11 May 2008 20:18:36 -0700 (PDT) Received: from sca-es-mail-1.sun.com (sca-es-mail-1.Sun.COM [192.18.43.132]) by dm-sfbay-01.sfbay.sun.com (8.13.8+Sun/8.13.8/ENSMAIL,v2.2) with ESMTP id m4C3IZf1040961 for ; Sun, 11 May 2008 20:18:35 -0700 (PDT) Received: from fe-sfbay-09.sun.com ([192.18.43.129]) by sca-es-mail-1.sun.com (8.13.7+Sun/8.12.9) with ESMTP id m4C3IU1g026855 for ; Sun, 11 May 2008 20:18:30 -0700 (PDT) Received: from conversion-daemon.fe-sfbay-09.sun.com by fe-sfbay-09.sun.com (Sun Java System Messaging Server 6.2-8.04 (built Feb 28 2007)) id <0K0Q00E01JUAHS00@fe-sfbay-09.sun.com> (original mail from gdamore@sun.com) for psarc-ext@sac.sfbay.sun.com; Sun, 11 May 2008 20:18:30 -0700 (PDT) Received: from [192.168.251.106] ([76.174.83.55]) by fe-sfbay-09.sun.com (Sun Java System Messaging Server 6.2-8.04 (built Feb 28 2007)) with ESMTPSA id <0K0Q00BXSJUTYK00@fe-sfbay-09.sun.com>; Sun, 11 May 2008 20:18:30 -0700 (PDT) Date: Sun, 11 May 2008 20:12:30 -0700 From: "Garrett D'Amore" Subject: Re: PSARC/2008/309 - expunge synonyms.h In-reply-to: <200805111914.m4BJELn2338278@jurassic-x4600.sfbay.sun.com> Sender: Garrett.Damore@sun.com To: "Roger A. Faulkner" Cc: psarc-ext@sac.sfbay.sun.com, Darrin.Johnson@sun.com, Rod.Evans@sun.com, Bart.Smaalders@sun.com Message-id: <4827B59E.6070005@sun.com> MIME-version: 1.0 Content-type: text/plain; format=flowed; charset=ISO-8859-1 Content-transfer-encoding: 7BIT References: <200805111914.m4BJELn2338278@jurassic-x4600.sfbay.sun.com> User-Agent: Thunderbird 2.0.0.6 (X11/20071023) Content-Length: 8379 Status: RO X-Status: $$$$ X-UID: 0000000004 A probably naive question, but I'm going to ask it anyway. What impact does reducing the number of symbols exported by libc by 25% have on the system? Does it save memory significantly, does it improve runtime somehow? The reason for the question is one of simply understanding the benefit to risk assessment. If there are applications which rely on SUNWprivate symbols which we don't know about, it seems like it could be painful for some folks. If there's not a reasonable benefit to cleaning up the cruft (and normally I'm hugely in favor of cleaning cruft), then maybe its better to avoid the risk? (Conversely, if there is likely to be noticeable improvement for a significant number of applications, or even in the time it takes to compile things like ON, then the benefits may well be worth the risk.) (Certainly it seems that the risk is non-zero, since even within Sun we've identified applications which have erroneously relied on private APIs.) Now removing the synonyms *headers* seems like a great idea, unconditionally. -- Garrett Roger A. Faulkner wrote: > I am sponsoring this fast-track case for myself. > This proposal will time out 05/19/2008 > > It's not clear that this actually needs PSARC review > since no official interfaces are affected, but it is > a change to the system for old stuff that has been > pervasive in Solaris since Solaris 2.0 and I'd like > to get your opinions. > > There are no exported interfaces involved, so there is > no stability level issue. > > The proposed release binding (if required) is "minor release" > since no one, certainly not I, wants this to be back-ported > to Solaris 10. > > ============================================================== > History > > Since the dawn of time (that is, since SVR4 came from New Jersey > and became Solaris 2.0), there have been synonym symbols in > the system-supplied libraries, most prevalently in libc. > > Synonym symbols come in pairs like this: > opendir > _opendir > These two functions are identical. Not identical contents but > exactly the same location in memory. This is most commonly > achieved with #pragma weak directives. (Back in the days of > static system libraries, the notion of 'strong' and 'weak' > symbols had some meaning, but with dynamically linked system > libraries the distinction is lost and the two symbols are just > alternate names for the same function.) > > The motivation for synonym symbols is to allow applications > and system libraries to define and use the same symbol names > without hurting with each other. Standards-conforming applications > are constrained not to define or call functions with leading > underscores in their names. The system libraries are not so > constrained. > > So, for example, libc defines opendir() and it calls opendir() > internally in its implementation of ttyname(). If an application > defines its own version of opendir() (which does something entirely > different from opendir() in libc) and the application also calls > ttyname() in libc, which instance of opendir() will be called by > libc's ttyname()? > > The solution to this problem has historically (since the dawn of time) > been synonym symbols. Internally, libc calls the _opendir() symbol. > This way, the application can define and use its opendir() function > for whatever it chooses and there will be no problem when the > application calls ttyname(). > > The implementation of this scheme for system libraries involves > special header files, usr/src/lib/libc/inc/synonyms.h for libc and > usr/src/lib/common/inc/c_synonyms.h for all other system libraries. > These header files contain definitions like these: > #define openat _openat > #define opendir _opendir > #define openlog _openlog > #define open _open > There are over one thousand definitions in each file. > For the scheme to work, every source file in libc must do this: > #include "synonyms.h" > before including any other header file. > > Neither synonyms.h nor c_synonyms.h are delivered as header files > for Solaris. They are OS/NET consolidation-private files. > > The source code for ttyname() in libc shows a call to opendir(), > but by virtue of including "synonyms.h", the compiled code > actually calls _opendir(). > > For the scheme to work for all system libraries delivered from > the OS/NET consolidation, all other libraries are required to > include "c_synonyms.h" before any other header file in every > one of their source files. Some of them do (libnsl, libresolve, > libgen plus a few others) but the vast majority of them do not. > > So the synonyms scheme is a failure for most libraries, due to > historical omission. > > ============================================================== > The better solution > > So much for history. Now for today's situation. > > The change for this bug report was recently integrated > into /ws/onnv-gate: > > 4947191 OSNet should use direct bindings > > This causes all OS/NET libraries, including libc, to be built > using direct binding (-B direct). As a result, all symbols > referenced by a library are bound either to the definition > within itself or to the definition within its direct dependencies. > > This solves the problem for which synonyms were invented > in an elegant and global way. Now all libraries in OS/NET > have the benefit of not damaging or being damaged by an > application's redefinition of a symbol that the library calls. > The application calls its instance and the library calls its > instance, even though both instances have the same name. > (Excepting, of course, symbols in the malloc() family and > other symbols explicitly defined to be interposable symbols > by default. These have all been marked NODIRECT in the > appropriate mapfiles. [The devil is in the details.]) > > This renders the whole synonyms scheme obsolete. > > The final stage of this evolution is to expunge the > various synonyms.h files: > usr/src/cmd/sgs/libelf/common/syn.h > usr/src/cmd/sgs/rtld/common/_synonyms.h > usr/src/lib/common/inc/c_synonyms.h > usr/src/lib/libc/inc/synonyms.h > usr/src/lib/libcrypt/inc/des_synonyms.h > usr/src/lib/libgen/inc/gen_synonyms.h > usr/src/lib/libtsnet/common/synonyms.h > and to fix all of the source files that include them. > > Also, most of the leading-underscore symbols involved > in the synonyms scheme are versioned SUNWprivate_1.1 > (over 600 of them). Many others are versioned public, > some because they are defined in the SVID, hence the ABI, > but others because of mistakes made along the way wherein > developers mistakenly thought that the leading-underscore > symbols should be versioned the same as the primary symbols. > > In an ideal world, the SUNWprivate symbols could just be > eliminated since nothing outside of OS/NET can legitimately > call them, and they are not part of the ABI. This would > reduce the number of symbols exported from libc by 25%. > The non-private symbols cannot be eliminated (although > I wish they could). > > However, a search of all of the executables and libraries > not delivered from OS/NET and also all the executables and > libraries in /usr/dist reveals that /lib/libm.so.2, some > older versions of the Studio compiler/debugger components, > and some ancient programs in /usr/dist call 55 of the > SUNWprivate symbols. So these 55 must be retained, > hopefully to be removed in some future update of Solaris. > > ============================================================== > The proposal > > This brings us to the change being proposed by this PSARC case: > > 1. Delete these header files from the OS/NET consolidation: > usr/src/cmd/sgs/libelf/common/syn.h > usr/src/cmd/sgs/rtld/common/_synonyms.h > usr/src/lib/common/inc/c_synonyms.h > usr/src/lib/libc/inc/synonyms.h > usr/src/lib/libcrypt/inc/des_synonyms.h > usr/src/lib/libgen/inc/gen_synonyms.h > usr/src/lib/libtsnet/common/synonyms.h > and fix all of the source files that include them. > > 2. Eliminate all of the SUNWprivate leading-underscore symbols > involved in the synonyms scheme except for the 55 that were > discovered to be used outside of OS/NET. > > Roger Faulkner > > From scott.rotondo@sun.com Sun May 11 23:10:34 2008 Received: from jurassic-x4600.sfbay.sun.com (cretaceous.SFBay.Sun.COM [129.146.17.63]) by sac.sfbay.sun.com (8.13.8+Sun/8.13.8) with ESMTP id m4C6AYU0025436 for ; Sun, 11 May 2008 23:10:34 -0700 (PDT) Received: from viaggio.local (punchin-rotondo.SFBay.Sun.COM [10.7.251.213]) by jurassic-x4600.sfbay.sun.com (8.14.2+Sun/8.14.2) with ESMTP id m4C6AUoR364667; Sun, 11 May 2008 23:10:30 -0700 (PDT) Message-ID: <4827DF91.4060300@sun.com> Date: Sun, 11 May 2008 23:11:29 -0700 From: Scott Rotondo User-Agent: Thunderbird 2.0.0.14 (Macintosh/20080421) MIME-Version: 1.0 To: "Roger A. Faulkner" CC: psarc-ext@sac.sfbay.sun.com, darrin.johnson@sun.com, rod.evans@sun.com, bart.smaalders@sun.com Subject: Re: PSARC/2008/309 - expunge synonyms.h References: <200805111914.m4BJELn2338278@jurassic-x4600.sfbay.sun.com> In-Reply-To: <200805111914.m4BJELn2338278@jurassic-x4600.sfbay.sun.com> Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Content-Length: 532 Status: RO X-Status: $$$$ X-UID: 0000000005 Roger A. Faulkner wrote: > > However, a search of all of the executables and libraries > not delivered from OS/NET and also all the executables and > libraries in /usr/dist reveals that /lib/libm.so.2, some > older versions of the Studio compiler/debugger components, > and some ancient programs in /usr/dist call 55 of the > SUNWprivate symbols. So these 55 must be retained, > hopefully to be removed in some future update of Solaris. If only to satisfy voyeuristic curiosity, could you identify the 55 symbols? Thanks, Scott From jek3@sun.com Mon May 12 00:26:20 2008 Received: from jurassic-x4600.sfbay.sun.com (cretaceous.SFBay.Sun.COM [129.146.17.63]) by sac.sfbay.sun.com (8.13.8+Sun/8.13.8) with ESMTP id m4C7QJ9M027227 for ; Mon, 12 May 2008 00:26:19 -0700 (PDT) Received: from [129.150.13.200] (vpn-129-150-13-200.SFBay.Sun.COM [129.150.13.200]) by jurassic-x4600.sfbay.sun.com (8.14.2+Sun/8.14.2) with ESMTP id m4C7QIDV368711; Mon, 12 May 2008 00:26:19 -0700 (PDT) Message-ID: <4827F172.7090001@sun.com> Date: Sun, 11 May 2008 21:27:46 -1000 From: Joseph Kowalski User-Agent: Thunderbird 2.0.0.9 (X11/20080225) MIME-Version: 1.0 To: "Roger A. Faulkner" CC: psarc-ext@sac.sfbay.sun.com, darrin.johnson@sun.com, rod.evans@sun.com, bart.smaalders@sun.com Subject: Re: PSARC/2008/309 - expunge synonyms.h References: <200805111914.m4BJELn2338278@jurassic-x4600.sfbay.sun.com> In-Reply-To: <200805111914.m4BJELn2338278@jurassic-x4600.sfbay.sun.com> Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Content-Length: 704 Status: RO X-Status: $$$$ X-UID: 0000000006 Roger A. Faulkner wrote: > 2. Eliminate all of the SUNWprivate leading-underscore symbols > involved in the synonyms scheme except for the 55 that were > discovered to be used outside of OS/NET. > More correctly, those were the 55 that were discovered by what ever happened to be in /usr/dist. I don't think that is sufficient coverage. However, I also know we can't possibly get enough coverage to be 100% sure. Which leaves me to the initial view of "Yea, I know its ugly, but is this worth taking chances for"? Is the benefit enough? I need to think about this. I hope other express their views... in few words. I think all of us are well aware of the history (and the bugs). - jek3 From jek3@sun.com Mon May 12 00:29:46 2008 Received: from jurassic-x4600.sfbay.sun.com (cretaceous.SFBay.Sun.COM [129.146.17.59]) by sac.sfbay.sun.com (8.13.8+Sun/8.13.8) with ESMTP id m4C7TkRN027241 for ; Mon, 12 May 2008 00:29:46 -0700 (PDT) Received: from [129.150.13.200] (vpn-129-150-13-200.SFBay.Sun.COM [129.150.13.200]) by jurassic-x4600.sfbay.sun.com (8.14.2+Sun/8.14.2) with ESMTP id m4C7Tjpt368981; Mon, 12 May 2008 00:29:46 -0700 (PDT) Message-ID: <4827F241.40504@sun.com> Date: Sun, 11 May 2008 21:31:13 -1000 From: Joseph Kowalski User-Agent: Thunderbird 2.0.0.9 (X11/20080225) MIME-Version: 1.0 To: "Garrett D'Amore" CC: "Roger A. Faulkner" , psarc-ext@sac.sfbay.sun.com, Darrin.Johnson@sun.com, Rod.Evans@sun.com, Bart.Smaalders@sun.com Subject: Re: PSARC/2008/309 - expunge synonyms.h References: <200805111914.m4BJELn2338278@jurassic-x4600.sfbay.sun.com> <4827B59E.6070005@sun.com> In-Reply-To: <4827B59E.6070005@sun.com> Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Content-Length: 297 Status: RO X-Status: $$$$ X-UID: 0000000007 Garrett D'Amore wrote: > Now removing the synonyms *headers* seems like a great idea, > unconditionally. Those headers are only in the source tree. They have never been in /usr/include. I seem to recall squashing a request to do that many years ago (with great relish). Maybe twice. - jek3 From jek3@sun.com Mon May 12 00:35:24 2008 Received: from jurassic-x4600.sfbay.sun.com (cretaceous.SFBay.Sun.COM [129.146.17.59]) by sac.sfbay.sun.com (8.13.8+Sun/8.13.8) with ESMTP id m4C7ZOTQ027448 for ; Mon, 12 May 2008 00:35:24 -0700 (PDT) Received: from [129.150.13.200] (vpn-129-150-13-200.SFBay.Sun.COM [129.150.13.200]) by jurassic-x4600.sfbay.sun.com (8.14.2+Sun/8.14.2) with ESMTP id m4C7ZIfG369430; Mon, 12 May 2008 00:35:23 -0700 (PDT) Message-ID: <4827F38E.1020106@sun.com> Date: Sun, 11 May 2008 21:36:46 -1000 From: Joseph Kowalski User-Agent: Thunderbird 2.0.0.9 (X11/20080225) MIME-Version: 1.0 To: "Garrett D'Amore" CC: "Roger A. Faulkner" , psarc-ext@sac.sfbay.sun.com, Darrin.Johnson@sun.com, Rod.Evans@sun.com, Bart.Smaalders@sun.com Subject: Re: PSARC/2008/309 - expunge synonyms.h References: <200805111914.m4BJELn2338278@jurassic-x4600.sfbay.sun.com> <4827B59E.6070005@sun.com> <4827F241.40504@sun.com> In-Reply-To: <4827F241.40504@sun.com> Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Content-Length: 466 Status: RO X-Status: $$$$ X-UID: 0000000008 Joseph Kowalski wrote: > I seem to recall squashing a request to do that many years ago (with > great relish). > Maybe twice. Speaking of "squashing", there have been multiple attempts to publicly expose these symbols. The rationale is that third party libraries could use the same trick - by binding to _read() rather than read(). That "trick" has merit, but I don't think it ever happened. (Bart, I think you were a proponent of this. Comments?) - jek3 From Roger.Faulkner@sun.com Mon May 12 04:43:26 2008 Received: from jurassic-x4600.sfbay.sun.com (cretaceous.SFBay.Sun.COM [129.146.17.59]) by sac.sfbay.sun.com (8.13.8+Sun/8.13.8) with ESMTP id m4CBhQ2w002348 for ; Mon, 12 May 2008 04:43:26 -0700 (PDT) Received: from jurassic-x4600.sfbay.sun.com (jurassic-x4600.sfbay.sun.com [129.146.17.63]) by jurassic-x4600.sfbay.sun.com (8.14.2+Sun/8.14.2) with SMTP id m4CBhQpq386098; Mon, 12 May 2008 04:43:26 -0700 (PDT) Message-Id: <200805121143.m4CBhQpq386098@jurassic-x4600.sfbay.sun.com> Date: Mon, 12 May 2008 04:43:26 -0700 (PDT) From: "Roger A. Faulkner" Reply-To: "Roger A. Faulkner" Subject: Re: PSARC/2008/309 - expunge synonyms.h To: scott.rotondo@sun.com Cc: psarc-ext@sac.sfbay.sun.com, darrin.johnson@sun.com, rod.evans@sun.com, bart.smaalders@sun.com MIME-Version: 1.0 Content-Type: TEXT/plain; charset=us-ascii Content-MD5: 8X9XIyin0I1OtVDs0WbzcQ== X-Mailer: dtmail 1.3.0 @(#)CDE Version 1.7_88 SunOS 5.11 i86pc i386 Content-Length: 4086 Status: RO X-Status: $$$$ X-UID: 0000000009 > Date: Sun, 11 May 2008 23:11:29 -0700 > From: Scott Rotondo > Subject: Re: PSARC/2008/309 - expunge synonyms.h > To: "Roger A. Faulkner" > Cc: psarc-ext@sac.sfbay.sun.com, darrin.johnson@sun.com, rod.evans@sun.com, bart.smaalders@sun.com > > Roger A. Faulkner wrote: > > > > However, a search of all of the executables and libraries > > not delivered from OS/NET and also all the executables and > > libraries in /usr/dist reveals that /lib/libm.so.2, some > > older versions of the Studio compiler/debugger components, > > and some ancient programs in /usr/dist call 55 of the > > SUNWprivate symbols. So these 55 must be retained, > > hopefully to be removed in some future update of Solaris. > > If only to satisfy voyeuristic curiosity, could you identify the 55 symbols? > > Thanks, > Scott These are the ones, from the SUNWprivate_1.1 section of usr/src/lib/libc/port/mapfile-vers : # The following functions should not be exported from libc, # but /lib/libm.so.2, some older versions of the Studio # compiler/debugger components, and some ancient programs # found in /usr/dist reference them. When we no longer # care about these old and broken binary objects, these # symbols should be deleted. _brk = NODYNSORT; _cond_broadcast = NODYNSORT; _cond_init = NODYNSORT; _cond_signal = NODYNSORT; _cond_wait = NODYNSORT; _ecvt = NODYNSORT; _fcvt = NODYNSORT; _getc_unlocked = NODYNSORT; _llseek = NODYNSORT; _pthread_attr_getdetachstate = NODYNSORT; _pthread_attr_getinheritsched = NODYNSORT; _pthread_attr_getschedparam = NODYNSORT; _pthread_attr_getschedpolicy = NODYNSORT; _pthread_attr_getscope = NODYNSORT; _pthread_attr_getstackaddr = NODYNSORT; _pthread_attr_getstacksize = NODYNSORT; _pthread_attr_init = NODYNSORT; _pthread_condattr_getpshared = NODYNSORT; _pthread_condattr_init = NODYNSORT; _pthread_cond_init = NODYNSORT; _pthread_create = NODYNSORT; _pthread_getschedparam = NODYNSORT; _pthread_join = NODYNSORT; _pthread_key_create = NODYNSORT; _pthread_mutexattr_getprioceiling = NODYNSORT; _pthread_mutexattr_getprotocol = NODYNSORT; _pthread_mutexattr_getpshared = NODYNSORT; _pthread_mutexattr_init = NODYNSORT; _pthread_mutex_getprioceiling = NODYNSORT; _pthread_mutex_init = NODYNSORT; _pthread_sigmask = NODYNSORT; _rwlock_init = NODYNSORT; _rw_rdlock = NODYNSORT; _rw_unlock = NODYNSORT; _rw_wrlock = NODYNSORT; _sbrk_unlocked = NODYNSORT; _select = NODYNSORT; _sema_init = NODYNSORT; _sema_post = NODYNSORT; _sema_trywait = NODYNSORT; _sema_wait = NODYNSORT; _sysfs = NODYNSORT; _thr_create = NODYNSORT; _thr_exit = NODYNSORT; _thr_getprio = NODYNSORT; _thr_getspecific = NODYNSORT; _thr_join = NODYNSORT; _thr_keycreate = NODYNSORT; _thr_kill = NODYNSORT; _thr_main = NODYNSORT; _thr_self = NODYNSORT; _thr_setspecific = NODYNSORT; _thr_sigsetmask = NODYNSORT; _thr_stksegment = NODYNSORT; _ungetc_unlocked = NODYNSORT; From John.Plocher@Sun.COM Mon May 12 07:43:56 2008 Received: from dm-sfbay-01.sfbay.sun.com (dm-sfbay-01.SFBay.Sun.COM [129.145.155.118]) by sac.sfbay.sun.com (8.13.8+Sun/8.13.8) with ESMTP id m4CEhunQ005762 for ; Mon, 12 May 2008 07:43:56 -0700 (PDT) Received: from sca-es-mail-2.sun.com (sca-es-mail-2.Sun.COM [192.18.43.133]) by dm-sfbay-01.sfbay.sun.com (8.13.8+Sun/8.13.8/ENSMAIL,v2.2) with ESMTP id m4CEhuKh039846 for ; Mon, 12 May 2008 07:43:56 -0700 (PDT) Received: from fe-sfbay-10.sun.com ([192.18.43.129]) by sca-es-mail-2.sun.com (8.13.7+Sun/8.12.9) with ESMTP id m4CEhpc7012836 for ; Mon, 12 May 2008 07:43:51 -0700 (PDT) Received: from conversion-daemon.fe-sfbay-10.sun.com by fe-sfbay-10.sun.com (Sun Java System Messaging Server 6.2-8.04 (built Feb 28 2007)) id <0K0R00D01FA5SA00@fe-sfbay-10.sun.com> (original mail from John.Plocher@Sun.COM) for psarc-ext@sac.sfbay.sun.com; Mon, 12 May 2008 07:43:51 -0700 (PDT) Received: from wp668.local ([208.74.177.212]) by fe-sfbay-10.sun.com (Sun Java System Messaging Server 6.2-8.04 (built Feb 28 2007)) with ESMTPSA id <0K0R0092MFL0RK50@fe-sfbay-10.sun.com>; Mon, 12 May 2008 07:43:48 -0700 (PDT) Date: Mon, 12 May 2008 07:43:48 -0700 From: John Plocher Subject: Re: PSARC/2008/309 - expunge synonyms.h In-reply-to: <200805111914.m4BJELn2338278@jurassic-x4600.sfbay.sun.com> Sender: John.Plocher@Sun.COM To: "Roger A. Faulkner" Cc: psarc-ext@sac.sfbay.sun.com, Bart.Smaalders@Sun.COM, Rod.Evans@Sun.COM, Darrin.Johnson@Sun.COM Message-id: <482857A4.6060707@Sun.Com> MIME-version: 1.0 Content-type: text/plain; format=flowed; charset=ISO-8859-1 Content-transfer-encoding: 7BIT References: <200805111914.m4BJELn2338278@jurassic-x4600.sfbay.sun.com> User-Agent: Thunderbird 2.0.0.14 (Macintosh/20080421) Content-Length: 1363 Status: RO X-Status: $$$$ X-UID: 0000000010 Roger A. Faulkner wrote: > The proposal > > This brings us to the change being proposed by this PSARC case: > > 1. Delete these header files from the OS/NET consolidation: > usr/src/cmd/sgs/libelf/common/syn.h > usr/src/cmd/sgs/rtld/common/_synonyms.h > usr/src/lib/common/inc/c_synonyms.h > usr/src/lib/libc/inc/synonyms.h > usr/src/lib/libcrypt/inc/des_synonyms.h > usr/src/lib/libgen/inc/gen_synonyms.h > usr/src/lib/libtsnet/common/synonyms.h > and fix all of the source files that include them. > > 2. Eliminate all of the SUNWprivate leading-underscore symbols > involved in the synonyms scheme except for the 55 that were > discovered to be used outside of OS/NET. I'd go out on a limb and suggest that, since this is an obvious case of not following the instructions, and since we are quickly moving towards a major release classification for Nevada/ON, which gives one (and probably only one) honeymoon opportunity to clean up such things, we should simply: Delete the header files as above, Create a LD_PRELOAD shim that restores these symbols for the things that need it, and Eliminate all of the SUNWprivate leading-underscore symbols involved in the synonyms scheme. Period. If something breaks, let it use the preload shim. End of story. -John From sacadmin Mon May 12 09:09:59 2008 Received: from dm-sfbay-01.sfbay.sun.com (dm-sfbay-01.SFBay.Sun.COM [129.145.155.118]) by sac.sfbay.sun.com (8.13.8+Sun/8.13.8) with ESMTP id m4CG9x6v008542 for ; Mon, 12 May 2008 09:09:59 -0700 (PDT) Received: from sca-es-mail-1.sun.com (sca-es-mail-1.Sun.COM [192.18.43.132]) by dm-sfbay-01.sfbay.sun.com (8.13.8+Sun/8.13.8/ENSMAIL,v2.2) with ESMTP id m4CG9xlh036116 for ; Mon, 12 May 2008 09:09:59 -0700 (PDT) Received: from fe-sfbay-09.sun.com ([192.18.43.129]) by sca-es-mail-1.sun.com (8.13.7+Sun/8.12.9) with ESMTP id m4CG9sJS025532 for ; Mon, 12 May 2008 09:09:54 -0700 (PDT) Received: from conversion-daemon.fe-sfbay-09.sun.com by fe-sfbay-09.sun.com (Sun Java System Messaging Server 6.2-8.04 (built Feb 28 2007)) id <0K0R00701J3VAC00@fe-sfbay-09.sun.com> (original mail from John.Plocher@Sun.COM) for psarc-record@sac.sfbay.sun.com; Mon, 12 May 2008 09:09:54 -0700 (PDT) Received: from wp668.local ([208.74.177.212]) by fe-sfbay-09.sun.com (Sun Java System Messaging Server 6.2-8.04 (built Feb 28 2007)) with ESMTPSA id <0K0R00FZXJKE0C20@fe-sfbay-09.sun.com> for psarc-record@sac.sfbay.sun.com; Mon, 12 May 2008 09:09:51 -0700 (PDT) Date: Mon, 12 May 2008 09:09:50 -0700 From: John Plocher Subject: [Fwd: Re: [arc-discuss] PSARC/2008/309 - expunge synonyms.h] Sender: John.Plocher@Sun.COM To: psarc-record@sac.sfbay.sun.com Message-id: <48286BCE.9080307@Sun.Com> MIME-version: 1.0 Content-type: text/plain; format=flowed; charset=ISO-8859-1 Content-transfer-encoding: 7BIT User-Agent: Thunderbird 2.0.0.14 (Macintosh/20080421) Content-Length: 1749 Status: RO X-Status: $$$$ X-UID: 0000000011 -------- Original Message -------- Subject: Re: [arc-discuss] PSARC/2008/309 - expunge synonyms.h Date: Mon, 12 May 2008 08:43:52 -0700 From: John Plocher To: Cyril Plisko CC: arc-discuss@opensolaris.org References: <200805111914.m4BJELn2338278@jurassic-x4600.sfbay.sun.com> <482857A4.6060707@Sun.Com> Cyril Plisko wrote: >> and since we are quickly moving towards >> a major release classification for Nevada/ON, which gives one (and probably > > Wow, that's interesting ! s/we are/I believe we are/ See my earlier email exploring this topic, starting last April and May with: http://archive.netbsd.se/?ml=opensolaris-discuss&a=2007-03&t=3397845 http://www.opensolaris.org/jive/thread.jspa?messageID=118803 http://opensolaris.org/jive/thread.jspa?threadID=58528&tstart=60 http://mail.opensolaris.org/pipermail/arc-discuss/2008-April/000817.html and others where we have discussed relaxing our heavy handed "minor release means absolutely no incompatible changes in /any/ part of nevada" mindset to allow us to at least consider changing the default for df, $SHELL and (gasp) maybe even $PATH. Recently (at the summit) we talked more about this in terms of repositories and how that might change how we build things. Nothing is set in stone, except maybe for my belief that trying to do both "Solaris10.next" and "OpenSolaris.modernization" at the same time and in the same minor release bound source tree is not the right answer. -John _______________________________________________ arc-discuss mailing list arc-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/arc-discuss From bart.smaalders@Sun.COM Mon May 12 11:07:27 2008 Received: from jurassic-x4600.sfbay.sun.com (cretaceous.SFBay.Sun.COM [129.146.17.59]) by sac.sfbay.sun.com (8.13.8+Sun/8.13.8) with ESMTP id m4CI7RiY014131 for ; Mon, 12 May 2008 11:07:27 -0700 (PDT) Received: from [129.146.228.109] (cyber.SFBay.Sun.COM [129.146.228.109]) by jurassic-x4600.sfbay.sun.com (8.14.2+Sun/8.14.2) with ESMTP id m4CI7Qrj442497; Mon, 12 May 2008 11:07:27 -0700 (PDT) Message-ID: <4828875E.1030800@Sun.COM> Date: Mon, 12 May 2008 11:07:26 -0700 From: Bart Smaalders Organization: Sun Microsystems User-Agent: Thunderbird 2.0.0.12 (X11/20080310) MIME-Version: 1.0 To: Joseph Kowalski CC: "Garrett D'Amore" , "Roger A. Faulkner" , psarc-ext@sac.sfbay.sun.com, Darrin.Johnson@Sun.COM, Rod.Evans@Sun.COM Subject: Re: PSARC/2008/309 - expunge synonyms.h References: <200805111914.m4BJELn2338278@jurassic-x4600.sfbay.sun.com> <4827B59E.6070005@sun.com> <4827F241.40504@sun.com> <4827F38E.1020106@sun.com> In-Reply-To: <4827F38E.1020106@sun.com> Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Content-Length: 1428 Status: RO X-Status: $$$$ X-UID: 0000000012 Joseph Kowalski wrote: > Joseph Kowalski wrote: >> I seem to recall squashing a request to do that many years ago >> (with great relish). Maybe twice. > Speaking of "squashing", there have been multiple attempts to > publicly expose these symbols. The rationale is that third party > libraries could use the same trick - by binding to _read() rather > than read(). That "trick" has merit, but I don't think it ever > happened. > > (Bart, I think you were a proponent of this. Comments?) > > - jek3 > I was and am a proponent of fixing the "surprise" factor that occurs when application/library function (and data, shudder) names intersect unexpectedly. Originally, we did not have the linker technology to do this w/o synonyms; the enhancements made to the linker(s) in the intervening years (and the massive cleanups of ON by Roger, Rod, and others) now allows us to use direct binding to address this problem. This is critical for long-term stability of dynamic library interfaces; as we move to a more asynchronous software delivery model & mechanism, removing the chance of accidental interposition after the fact is even more critical. I'm wondering if we should be changing some default linking behaviors when building shared libraries... - Bart -- Bart Smaalders Solaris Kernel Performance barts@cyber.eng.sun.com http://blogs.sun.com/barts "You will contribute more with mercurial than with thunderbird." From jek3@sun.com Mon May 12 12:57:41 2008 Received: from jurassic-x4600.sfbay.sun.com (cretaceous.SFBay.Sun.COM [129.146.17.63]) by sac.sfbay.sun.com (8.13.8+Sun/8.13.8) with ESMTP id m4CJvfe3017589 for ; Mon, 12 May 2008 12:57:41 -0700 (PDT) Received: from [129.150.13.200] (vpn-129-150-13-200.SFBay.Sun.COM [129.150.13.200]) by jurassic-x4600.sfbay.sun.com (8.14.2+Sun/8.14.2) with ESMTP id m4CJve5A460596; Mon, 12 May 2008 12:57:40 -0700 (PDT) Message-ID: <4828A18E.2070509@sun.com> Date: Mon, 12 May 2008 09:59:10 -1000 From: Joseph Kowalski User-Agent: Thunderbird 2.0.0.9 (X11/20080225) MIME-Version: 1.0 To: John Plocher CC: "Roger A. Faulkner" , psarc-ext@sac.sfbay.sun.com, Bart.Smaalders@sun.com, Rod.Evans@sun.com, Darrin.Johnson@sun.com Subject: Re: PSARC/2008/309 - expunge synonyms.h References: <200805111914.m4BJELn2338278@jurassic-x4600.sfbay.sun.com> <482857A4.6060707@Sun.Com> In-Reply-To: <482857A4.6060707@Sun.Com> Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Content-Length: 1460 Status: RO X-Status: $$$$ X-UID: 0000000013 John Plocher wrote: > > I'd go out on a limb and suggest that, since this is an obvious case of > not following the instructions, and since we are quickly moving towards > a major release classification for Nevada/ON, which gives one (and > probably > only one) honeymoon opportunity to clean up such things, we should simply: I clearly heard that the same compatible rules we to be inplace for the library interfaces. The semi-major release classification was for utilities (and perhaps administration - not clear to me). These are library interfaces, hence I think we need to think in Minor release terms. Maybe this will be clearer on Wednesday. > Delete the header files as above, They are in source. Who cares? They are implementation. Certainly not architecture. > Create a LD_PRELOAD shim that restores these symbols for the > things that need it, and I need to think about this more, but I don't think this works reliably if third-party libraries are in the mix. At this point, its just a nit. > Eliminate all of the SUNWprivate leading-underscore symbols > involved in the synonyms scheme. Period. Comma or semi-colon perhaps? (See: how Major is Major) > If something breaks, let it use the preload shim. End of story. The ARC shouldn't be designing this. *If* there is a need for a shim it should be part of the proposal, including any caveats. Like I I said, I'm not sure we need a shim (TBD). > > -John - jek3 From Joerg.Schilling@fokus.fraunhofer.de Wed May 14 03:46:36 2008 Received: from dm-sfbay-02.sfbay.sun.com (dm-sfbay-02.SFBay.Sun.COM [129.146.11.31]) by sac.sfbay.sun.com (8.13.8+Sun/8.13.8) with ESMTP id m4EAkaCm025967 for ; Wed, 14 May 2008 03:46:36 -0700 (PDT) Received: from sca-ea-mail-1.sun.com (sca-ea-mail-1.Sun.COM [192.18.43.24]) by dm-sfbay-02.sfbay.sun.com (8.13.8+Sun/8.13.8/ENSMAIL,v2.2) with ESMTP id m4EAkZf1052118 for ; Wed, 14 May 2008 03:46:35 -0700 (PDT) Received: from relay21.sun.com (relay21.sun.com [192.12.251.24] (may be forged)) by sca-ea-mail-1.sun.com (8.13.7+Sun/8.12.9) with ESMTP id m4EAgmYe022927 for ; Wed, 14 May 2008 10:46:30 GMT Received: from mms22es.sun.com ([150.143.232.34] [150.143.232.34]) by relay21i.sun.com with ESMTP id BT-MMP-113297 for psarc-ext@sac.sfbay.sun.com; Wed, 14 May 2008 10:46:30 Z Received: from relay22.sun.com (relay22.sun.com [192.12.251.34]) by mms22es.sun.com with ESMTP id BT-MMP-49309558 for psarc-ext@sac.sfbay.sun.com; Wed, 14 May 2008 10:46:30 Z Received: from mailgw1.fraunhofer.de ([153.96.1.17] [153.96.1.17]) by relay22i.sun.com with ESMTP id BT-MMP-4128544 for psarc-ext@sac.sfbay.sun.com; Wed, 14 May 2008 10:46:29 Z Received: from mailgw1.fraunhofer.de (localhost [127.0.0.1]) by mailgw1.fraunhofer.de[host mailgw22] (8.14.2+/8.14.2) with ESMTP id m4EAh8c3009703; Wed, 14 May 2008 12:43:09 +0200 (CEST) Received: from pluto.fokus.fraunhofer.de (pluto.fokus.fraunhofer.de [195.37.77.164]) by mailgw1.fraunhofer.de (8.14.2+/8.14.2) with ESMTP id m4EAh8eg009694 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=OK); Wed, 14 May 2008 12:43:08 +0200 (CEST) Received: from EXCHSRV.fokus.fraunhofer.de (bohr [10.147.9.231]) by pluto.fokus.fraunhofer.de (8.13.7/8.13.7) with SMTP id m4EAh7CZ008128; Wed, 14 May 2008 12:43:08 +0200 (MEST) Received: from rigel ([10.147.65.195]) by EXCHSRV.fokus.fraunhofer.de with Microsoft SMTPSVC(6.0.3790.3959); Wed, 14 May 2008 12:43:08 +0200 Date: Wed, 14 May 2008 12:43:07 +0200 From: Joerg.Schilling@fokus.fraunhofer.de (Joerg Schilling) To: Roger.Faulkner@sun.com, psarc-ext@sac.sfbay.sun.com Cc: rod.evans@sun.com, darrin.johnson@sun.com, bart.smaalders@sun.com Subject: Re: PSARC/2008/309 - expunge synonyms.h Message-ID: <482ac23b.hwNBXU5mto4SQ5N/%Joerg.Schilling@fokus.fraunhofer.de> References: <200805111914.m4BJELn2338278@jurassic-x4600.sfbay.sun.com> In-Reply-To: <200805111914.m4BJELn2338278@jurassic-x4600.sfbay.sun.com> User-Agent: nail 11.22 3/20/05 X-Brightmail-Tracker: AAAAAA== X-OriginalArrivalTime: 14 May 2008 10:43:08.0042 (UTC) FILETIME=[4C152EA0:01C8B5AF] X-Fraunhofer-Email-Policy: accepted X-Antispam: No, score=0.0/5.0, scanned in 0.245sec at (localhost [127.0.0.1]) by smf-spamd v1.3.1 - http://smfs.sf.net/ MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit Content-Length: 583 Status: RO X-Status: $$$$ X-UID: 0000000014 "Roger A. Faulkner" wrote: .... > ============================================================== > The better solution > > So much for history. Now for today's situation. > > The change for this bug report was recently integrated > into /ws/onnv-gate: > > 4947191 OSNet should use direct bindings > > This causes all OS/NET libraries, including libc, to be built > using direct binding (-B direct). As a result, all symbols > referenced by a library are bound either to the definition > within itself or to the definition within its direct dependencies. From my understanding, this would prevent applications to replace "malloc" Status: RO X-Status: $$$$ X-UID: 0000000015 that are based on brk(2)/sbrk(2). libc currently binds to the malloc()/.... functions from the application. Binding to libc internal functions is currently only done for those functions that need to be the way libc defines them. Using a "direct" binding to all libc functions looks to me like ignoring linking rules. if you are sure that this change would not cause any problems please explain in more details. Jörg -- EMail:joerg@schily.isdn.cs.tu-berlin.de (home) Jörg Schilling D-13353 Berlin js@cs.tu-berlin.de (uni) schilling@fokus.fraunhofer.de (work) Blog: http://schily.blogspot.com/ URL: http://cdrecord.berlios.de/old/private/ ftp://ftp.berlios.de/pub/schily From Roger.Faulkner@sun.com Wed May 14 04:54:25 2008 Received: from jurassic-x4600.sfbay.sun.com (cretaceous.SFBay.Sun.COM [129.146.17.63]) by sac.sfbay.sun.com (8.13.8+Sun/8.13.8) with ESMTP id m4EBsPlH026986 for ; Wed, 14 May 2008 04:54:25 -0700 (PDT) Received: from jurassic-x4600.sfbay.sun.com (jurassic-x4600.sfbay.sun.com [129.146.17.59]) by jurassic-x4600.sfbay.sun.com (8.14.2+Sun/8.14.2) with SMTP id m4EBsPAZ788105; Wed, 14 May 2008 04:54:25 -0700 (PDT) Message-Id: <200805141154.m4EBsPAZ788105@jurassic-x4600.sfbay.sun.com> Date: Wed, 14 May 2008 04:54:25 -0700 (PDT) From: "Roger A. Faulkner" Reply-To: "Roger A. Faulkner" Subject: Re: PSARC/2008/309 - expunge synonyms.h To: Joerg.Schilling@fokus.fraunhofer.de Cc: rod.evans@sun.com, darrin.johnson@sun.com, bart.smaalders@sun.com, psarc-ext@sac.sfbay.sun.com MIME-Version: 1.0 Content-Type: TEXT/plain; charset=ISO-8859-1 Content-MD5: YcBPBAlFL4Tqdugqey11bQ== X-Mailer: dtmail 1.3.0 @(#)CDE Version 1.7_88 SunOS 5.11 i86pc i386 Content-Transfer-Encoding: 8bit X-MIME-Autoconverted: from QUOTED-PRINTABLE to 8bit by sac.sfbay.sun.com id m4EBsPlH026986 Content-Length: 2227 Status: RO X-Status: $$$$ X-UID: 0000000016 > Date: Wed, 14 May 2008 12:43:07 +0200 > From: Joerg.Schilling@fokus.fraunhofer.de (Joerg Schilling) > Subject: Re: PSARC/2008/309 - expunge synonyms.h > To: Roger.Faulkner@sun.com, psarc-ext@sac.sfbay.sun.com > Cc: rod.evans@sun.com, darrin.johnson@sun.com, bart.smaalders@sun.com > > "Roger A. Faulkner" wrote: > > .... > > > ============================================================== > > The better solution > > > > So much for history. Now for today's situation. > > > > The change for this bug report was recently integrated > > into /ws/onnv-gate: > > > > 4947191 OSNet should use direct bindings > > > > This causes all OS/NET libraries, including libc, to be built > > using direct binding (-B direct). As a result, all symbols > > referenced by a library are bound either to the definition > > within itself or to the definition within its direct dependencies. > > From my understanding, this would prevent applications to replace "malloc" > that are based on brk(2)/sbrk(2). > > libc currently binds to the malloc()/.... functions from the application. > Binding to libc internal functions is currently only done for those functions > that need to be the way libc defines them. > > Using a "direct" binding to all libc functions looks to me like ignoring > linking rules. > > if you are sure that this change would not cause any problems please explain in > more details. > > Jörg The '-B direct' change for all system libraries has already been done. It is independent of (but required by) my proposed synonyms.h elimination. As my mail went on to say: : The application calls its instance and the library calls its : instance, even though both instances have the same name. : (Excepting, of course, symbols in the malloc() family and : other symbols explicitly defined to be interposable symbols : by default. These have all been marked NODIRECT in the : appropriate mapfiles. [The devil is in the details.]) The malloc family is marked NODIRECT in libc's mapfiles: calloc = NODIRECT; cfree = NODIRECT; free = NODIRECT; malloc = NODIRECT; memalign = NODIRECT; realloc = NODIRECT; valloc = NODIRECT; Roger From Roger.Faulkner@sun.com Thu May 15 11:28:24 2008 Received: from jurassic-x4600.sfbay.sun.com (cretaceous.SFBay.Sun.COM [129.146.17.63]) by sac.sfbay.sun.com (8.13.8+Sun/8.13.8) with ESMTP id m4FISOA7021544 for ; Thu, 15 May 2008 11:28:24 -0700 (PDT) Received: from jurassic-x4600.sfbay.sun.com (jurassic-x4600.sfbay.sun.com [129.146.17.59]) by jurassic-x4600.sfbay.sun.com (8.14.2+Sun/8.14.2) with SMTP id m4FISOTh109301; Thu, 15 May 2008 11:28:24 -0700 (PDT) Message-Id: <200805151828.m4FISOTh109301@jurassic-x4600.sfbay.sun.com> Date: Thu, 15 May 2008 11:28:24 -0700 (PDT) From: "Roger A. Faulkner" Reply-To: "Roger A. Faulkner" Subject: Re: PSARC/2008/309 - expunge synonyms.h To: John.Plocher@sun.com, jek3@sun.com Cc: psarc-ext@sac.sfbay.sun.com, Bart.Smaalders@sun.com, Rod.Evans@sun.com, Darrin.Johnson@sun.com MIME-Version: 1.0 Content-Type: TEXT/plain; charset=us-ascii Content-MD5: iRz5YeVfouQ7F5uAnJsMKg== X-Mailer: dtmail 1.3.0 @(#)CDE Version 1.7_88 SunOS 5.11 i86pc i386 Content-Length: 2305 Status: RO X-Status: $$$$ X-UID: 0000000017 > Date: Mon, 12 May 2008 09:59:10 -1000 > From: Joseph Kowalski > Subject: Re: PSARC/2008/309 - expunge synonyms.h > To: John Plocher > Cc: "Roger A. Faulkner" , psarc-ext@sac.sfbay.sun.com, Bart.Smaalders@sun.com, Rod.Evans@sun.com, Darrin.Johnson@sun.com > > John Plocher wrote: > > > > I'd go out on a limb and suggest that, since this is an obvious case of > > not following the instructions, and since we are quickly moving towards > > a major release classification for Nevada/ON, which gives one (and > > probably > > only one) honeymoon opportunity to clean up such things, we should simply: > I clearly heard that the same compatible rules we to be inplace for the > library > interfaces. The semi-major release classification was for utilities > (and perhaps > administration - not clear to me). These are library interfaces, hence > I think > we need to think in Minor release terms. Maybe this will be clearer on > Wednesday. > > Delete the header files as above, > They are in source. Who cares? They are implementation. Certainly not > architecture. > > Create a LD_PRELOAD shim that restores these symbols for the > > things that need it, and > I need to think about this more, but I don't think this works reliably if > third-party libraries are in the mix. At this point, its just a nit. > > Eliminate all of the SUNWprivate leading-underscore symbols > > involved in the synonyms scheme. Period. > Comma or semi-colon perhaps? (See: how Major is Major) > > If something breaks, let it use the preload shim. End of story. > The ARC shouldn't be designing this. *If* there is a need for a > shim it should be part of the proposal, including any caveats. > Like I I said, I'm not sure we need a shim (TBD). > > > > -John > - jek3 Regardless of the merits of the Draconian approach (which I like!), there is merit to the idea of providing a preloadable shim library that provides the old leading-underscore symbols and jumps directly (via the PLT and GOT) to the corresponding non-underscore symbols. I created it yesterday and named it: /lib/libshim.so.1 /lib/64/libshim.so.1 So I'm adding this item to the official proposal. I welcome suggestions for a better name. Roger From johnsonnenschein@gmail.com Thu May 15 11:31:34 2008 Received: from dm-sfbay-02.sfbay.sun.com (dm-sfbay-02.SFBay.Sun.COM [129.146.11.31]) by sac.sfbay.sun.com (8.13.8+Sun/8.13.8) with ESMTP id m4FIVYk1021579 for ; Thu, 15 May 2008 11:31:34 -0700 (PDT) Received: from brmea-mail-2.sun.com (brmea-mail-2.Sun.COM [192.18.98.43]) by dm-sfbay-02.sfbay.sun.com (8.13.8+Sun/8.13.8/ENSMAIL,v2.2) with ESMTP id m4FIVYSm061080 for ; Thu, 15 May 2008 11:31:34 -0700 (PDT) Received: from relay15i.sun.com (ip125.net129179-4.block1.us.syntegra.com [129.179.4.125]) by brmea-mail-2.sun.com (8.13.6+Sun/8.12.9) with ESMTP id m4FIML0k001925 for ; Thu, 15 May 2008 18:31:33 GMT Received: from mmp14es.sun.com ([160.41.209.24] [160.41.209.24]) by relay15i.sun.com with ESMTP id BT-MMP-149609 for psarc-ext@sac.sfbay.sun.com; Thu, 15 May 2008 18:31:33 Z Received: from relay12i.sun.com (relay12i.sun.com [129.179.4.122]) by mmp14es.sun.com with ESMTP id BT-MMP-348086 for psarc-ext@sac.sfbay.sun.com; Thu, 15 May 2008 18:31:33 Z Received: from wr-out-0506.google.com ([64.233.184.224] [64.233.184.224]) by relay1ib.sun.com with ESMTP id BT-MMP-1748751 for psarc-ext@sac.sfbay.sun.com; Thu, 15 May 2008 18:31:33 Z Received: by wr-out-0506.google.com with SMTP id c57so267169wra.9 for ; Thu, 15 May 2008 11:31:15 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=gamma; h=domainkey-signature:received:received:message-id:date:from:to:subject:cc:in-reply-to:mime-version:content-type:content-transfer-encoding:content-disposition:references; bh=rJeJWKaHx3f4d0G1wHEz4Pzbj7huDWcDo+wbdsW3JHY=; b=EBqYupTcK3dKoryWPlnQ1Rj7677OSG8WSjCUDJaRMSyms5OBXb2bCLyVVPdvGSikgLqdvT2mqLb3+l+Q9VkC5yJRz9KlJTKIrnobZm7Qh/neGhRpR7fx94B4ILr5dYBRYqRJAdJ8upvnfHJMIm+hMJIKZ1Hj7RQDdhW18My+3DU= DomainKey-Signature: a=rsa-sha1; c=nofws; d=gmail.com; s=gamma; h=message-id:date:from:to:subject:cc:in-reply-to:mime-version:content-type:content-transfer-encoding:content-disposition:references; b=dEpPzZNYixJUlj1htDBA3gi3pRM/JFQw1JJKIyK8jg+PmMgb7Wd2aqSex9x9XO/5ZgBHPpO0wpXDR4c5gln4lqvcgJ0EIwOPiYoJ/QSxJ3WH59hMT+Z6pLld8Kkur+vE8lC8HU3n8Ze6iMlggNb3xqcyrf45Mhb9m5XQwS+BIOg= Received: by 10.114.66.2 with SMTP id o2mr2836090waa.102.1210876275046; Thu, 15 May 2008 11:31:15 -0700 (PDT) Received: by 10.114.150.11 with HTTP; Thu, 15 May 2008 11:31:14 -0700 (PDT) Message-ID: <241540330805151131x239fe5c9s588c6a791c8d6432@mail.gmail.com> Date: Thu, 15 May 2008 11:31:14 -0700 From: "John Sonnenschein" To: "Roger A. Faulkner" Subject: Re: PSARC/2008/309 - expunge synonyms.h Cc: John.Plocher@sun.com, jek3@sun.com, Rod.Evans@sun.com, Bart.Smaalders@sun.com, psarc-ext@sac.sfbay.sun.com, Darrin.Johnson@sun.com In-Reply-To: <200805151828.m4FISOTh109301@jurassic-x4600.sfbay.sun.com> References: <200805151828.m4FISOTh109301@jurassic-x4600.sfbay.sun.com> X-Brightmail-Tracker: AAAAAA== X-Antispam: No, score=0.0/5.0, scanned in 0.083sec at (localhost [127.0.0.1]) by smf-spamd v1.3.1 - http://smfs.sf.net/ MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Disposition: inline Content-Length: 2664 Status: RO X-Status: $$$$ X-UID: 0000000018 libdeprecated ? libfixbadcoding ? On Thu, May 15, 2008 at 11:28 AM, Roger A. Faulkner wrote: > >> Date: Mon, 12 May 2008 09:59:10 -1000 >> From: Joseph Kowalski >> Subject: Re: PSARC/2008/309 - expunge synonyms.h >> To: John Plocher >> Cc: "Roger A. Faulkner" , psarc-ext@sac.sfbay.sun.com, > Bart.Smaalders@sun.com, Rod.Evans@sun.com, Darrin.Johnson@sun.com >> >> John Plocher wrote: >> > >> > I'd go out on a limb and suggest that, since this is an obvious case of >> > not following the instructions, and since we are quickly moving towards >> > a major release classification for Nevada/ON, which gives one (and >> > probably >> > only one) honeymoon opportunity to clean up such things, we should simply: >> I clearly heard that the same compatible rules we to be inplace for the >> library >> interfaces. The semi-major release classification was for utilities >> (and perhaps >> administration - not clear to me). These are library interfaces, hence >> I think >> we need to think in Minor release terms. Maybe this will be clearer on >> Wednesday. >> > Delete the header files as above, >> They are in source. Who cares? They are implementation. Certainly not >> architecture. >> > Create a LD_PRELOAD shim that restores these symbols for the >> > things that need it, and >> I need to think about this more, but I don't think this works reliably if >> third-party libraries are in the mix. At this point, its just a nit. >> > Eliminate all of the SUNWprivate leading-underscore symbols >> > involved in the synonyms scheme. Period. >> Comma or semi-colon perhaps? (See: how Major is Major) >> > If something breaks, let it use the preload shim. End of story. >> The ARC shouldn't be designing this. *If* there is a need for a >> shim it should be part of the proposal, including any caveats. >> Like I I said, I'm not sure we need a shim (TBD). >> > >> > -John >> - jek3 > > Regardless of the merits of the Draconian approach (which I like!), > there is merit to the idea of providing a preloadable shim library > that provides the old leading-underscore symbols and jumps directly > (via the PLT and GOT) to the corresponding non-underscore symbols. > > I created it yesterday and named it: > > /lib/libshim.so.1 > /lib/64/libshim.so.1 > > So I'm adding this item to the official proposal. > > I welcome suggestions for a better name. > > Roger > > _______________________________________________ > opensolaris-arc mailing list > opensolaris-arc@opensolaris.org > -- PGP Public Key 0x437AF1A1 Available on hkp://pgp.mit.edu From Darren.Moffat@Sun.COM Thu May 15 11:34:24 2008 Received: from dm-sfbay-01.sfbay.sun.com (dm-sfbay-01.SFBay.Sun.COM [129.145.155.118]) by sac.sfbay.sun.com (8.13.8+Sun/8.13.8) with ESMTP id m4FIYO35021615 for ; Thu, 15 May 2008 11:34:24 -0700 (PDT) Received: from gmp-eb-inf-2.sun.com (gmp-eb-inf-2.EU.Sun.COM [192.18.6.24]) by dm-sfbay-01.sfbay.sun.com (8.13.8+Sun/8.13.8/ENSMAIL,v2.2) with ESMTP id m4FIYN0p020382 for ; Thu, 15 May 2008 11:34:24 -0700 (PDT) Received: from fe-emea-09.sun.com (gmp-eb-lb-2-fe1.eu.sun.com [192.18.6.10]) by gmp-eb-inf-2.sun.com (8.13.7+Sun/8.12.9) with ESMTP id m4FIYIpd027714 for ; Thu, 15 May 2008 18:34:18 GMT Received: from conversion-daemon.fe-emea-09.sun.com by fe-emea-09.sun.com (Sun Java System Messaging Server 6.2-8.04 (built Feb 28 2007)) id <0K0X00L01A5Y2B00@fe-emea-09.sun.com> (original mail from Darren.Moffat@Sun.COM) for psarc-ext@sac.sfbay.sun.com; Thu, 15 May 2008 19:34:16 +0100 (BST) Received: from [129.156.173.21] by fe-emea-09.sun.com (Sun Java System Messaging Server 6.2-8.04 (built Feb 28 2007)) with ESMTPSA id <0K0X00MP2A921W20@fe-emea-09.sun.com>; Thu, 15 May 2008 19:34:16 +0100 (BST) Date: Thu, 15 May 2008 19:34:14 +0100 From: Darren J Moffat Subject: Re: PSARC/2008/309 - expunge synonyms.h In-reply-to: <200805151828.m4FISOTh109301@jurassic-x4600.sfbay.sun.com> Sender: Darren.Moffat@Sun.COM To: "Roger A. Faulkner" Cc: John.Plocher@Sun.COM, jek3@Sun.COM, psarc-ext@sac.sfbay.sun.com, Bart.Smaalders@Sun.COM, Rod.Evans@Sun.COM, Darrin.Johnson@Sun.COM Message-id: <482C8226.2070502@Sun.COM> MIME-version: 1.0 Content-type: text/plain; format=flowed; charset=ISO-8859-1 Content-transfer-encoding: 7BIT References: <200805151828.m4FISOTh109301@jurassic-x4600.sfbay.sun.com> User-Agent: Thunderbird 2.0.0.12 (X11/20080326) Content-Length: 508 Status: RO X-Status: $$$$ X-UID: 0000000019 Roger A. Faulkner wrote: > I created it yesterday and named it: > > /lib/libshim.so.1 > /lib/64/libshim.so.1 > > So I'm adding this item to the official proposal. With that name ? Isn't this more like 0@0.so.1, extendedFILE.so.1, madv.so.1 etc ? In other words I don't think it should be specified on a link line so does it need to begin with lib. Also isn't there a better name than the highly generic "shim" ? My suggestion would be: leading_underscore_synonyms.so.1 -- Darren J Moffat From scott.rotondo@sun.com Thu May 15 11:35:50 2008 Received: from jurassic-x4600.sfbay.sun.com (cretaceous.SFBay.Sun.COM [129.146.17.63]) by sac.sfbay.sun.com (8.13.8+Sun/8.13.8) with ESMTP id m4FIZot0021632 for ; Thu, 15 May 2008 11:35:50 -0700 (PDT) Received: from [129.146.108.62] (vinifera.SFBay.Sun.COM [129.146.108.62]) by jurassic-x4600.sfbay.sun.com (8.14.2+Sun/8.14.2) with ESMTP id m4FIZolW111331; Thu, 15 May 2008 11:35:50 -0700 (PDT) Message-ID: <482C8286.3000902@sun.com> Date: Thu, 15 May 2008 11:35:50 -0700 From: Scott Rotondo User-Agent: Thunderbird 2.0.0.6 (X11/20071119) MIME-Version: 1.0 To: "Roger A. Faulkner" CC: psarc-ext@sac.sfbay.sun.com Subject: Re: PSARC/2008/309 - expunge synonyms.h References: <200805151828.m4FISOTh109301@jurassic-x4600.sfbay.sun.com> <241540330805151131x239fe5c9s588c6a791c8d6432@mail.gmail.com> In-Reply-To: <241540330805151131x239fe5c9s588c6a791c8d6432@mail.gmail.com> Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Content-Length: 112 Status: RO X-Status: $$$$ X-UID: 0000000020 John Sonnenschein wrote: > libdeprecated ? libfixbadcoding ? Or less judgmentally, perhaps libsynonym? Scott From jek3@sun.com Thu May 15 11:46:25 2008 Received: from jurassic-x4600.sfbay.sun.com (cretaceous.SFBay.Sun.COM [129.146.17.63]) by sac.sfbay.sun.com (8.13.8+Sun/8.13.8) with ESMTP id m4FIkPnV021808 for ; Thu, 15 May 2008 11:46:25 -0700 (PDT) Received: from [129.150.13.200] (vpn-129-150-13-200.SFBay.Sun.COM [129.150.13.200]) by jurassic-x4600.sfbay.sun.com (8.14.2+Sun/8.14.2) with ESMTP id m4FIkO3K112880; Thu, 15 May 2008 11:46:24 -0700 (PDT) Message-ID: <482C8562.8010700@sun.com> Date: Thu, 15 May 2008 08:48:02 -1000 From: Joseph Kowalski User-Agent: Thunderbird 2.0.0.9 (X11/20080225) MIME-Version: 1.0 To: "Roger A. Faulkner" CC: John.Plocher@sun.com, psarc-ext@sac.sfbay.sun.com, Bart.Smaalders@sun.com, Rod.Evans@sun.com, Darrin.Johnson@sun.com Subject: Re: PSARC/2008/309 - expunge synonyms.h References: <200805151828.m4FISOTh109301@jurassic-x4600.sfbay.sun.com> In-Reply-To: <200805151828.m4FISOTh109301@jurassic-x4600.sfbay.sun.com> Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Content-Length: 1046 Status: RO X-Status: $$$$ X-UID: 0000000021 Roger A. Faulkner wrote: > Regardless of the merits of the Draconian approach (which I like!), > there is merit to the idea of providing a preloadable shim library > that provides the old leading-underscore symbols and jumps directly > (via the PLT and GOT) to the corresponding non-underscore symbols. > > I created it yesterday and named it: > > /lib/libshim.so.1 > /lib/64/libshim.so.1 > > So I'm adding this item to the official proposal. > > I welcome suggestions for a better name. > > Roger > lib-toxic-waste-from-NJ.so.1 ??? :-) (Actually, that's not fair given 1990's linker technology) Unfortunately, I don't have a better suggestion, but I hope somebody does. At least some refernance to "sym". ...... What is the commitment level to "libshim"? More relevantly, does this get a man page? Does it get no man page, but some other documentation (linker guide?)? Its kinda strange, because I can understand lots of reasons to document it in unique ways, but I suspect it should be documented somewhere. - jek3 From jek3@sun.com Thu May 15 11:51:08 2008 Received: from jurassic-x4600.sfbay.sun.com (cretaceous.SFBay.Sun.COM [129.146.17.59]) by sac.sfbay.sun.com (8.13.8+Sun/8.13.8) with ESMTP id m4FIp8b4021877 for ; Thu, 15 May 2008 11:51:08 -0700 (PDT) Received: from [129.150.13.200] (vpn-129-150-13-200.SFBay.Sun.COM [129.150.13.200]) by jurassic-x4600.sfbay.sun.com (8.14.2+Sun/8.14.2) with ESMTP id m4FIp6vt113831; Thu, 15 May 2008 11:51:07 -0700 (PDT) Message-ID: <482C866E.4060909@sun.com> Date: Thu, 15 May 2008 08:52:30 -1000 From: Joseph Kowalski User-Agent: Thunderbird 2.0.0.9 (X11/20080225) MIME-Version: 1.0 To: Darren J Moffat CC: "Roger A. Faulkner" , John.Plocher@sun.com, psarc-ext@sac.sfbay.sun.com, Bart.Smaalders@sun.com, Rod.Evans@sun.com, Darrin.Johnson@sun.com Subject: Re: PSARC/2008/309 - expunge synonyms.h References: <200805151828.m4FISOTh109301@jurassic-x4600.sfbay.sun.com> <482C8226.2070502@Sun.COM> In-Reply-To: <482C8226.2070502@Sun.COM> Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Content-Length: 740 Status: RO X-Status: $$$$ X-UID: 0000000022 Darren J Moffat wrote: > Roger A. Faulkner wrote: >> I created it yesterday and named it: >> >> /lib/libshim.so.1 >> /lib/64/libshim.so.1 >> >> So I'm adding this item to the official proposal. > > With that name ? > > Isn't this more like 0@0.so.1, extendedFILE.so.1, madv.so.1 etc ? > > In other words I don't think it should be specified on a link line so > does it need to begin with lib. +1 (nice catch) > Also isn't there a better name than the highly generic "shim" ? > > My suggestion would be: > > leading_underscore_synonyms.so.1 > Would be "underscore_syn.so.1" be enough? (I know we are supposed to be "linux familiar", but does that include "linix-prefers-really-long-and-awkward-names"? :-) :-) ;-) ) - jek3 From Rod.Evans@sun.com Thu May 15 11:59:22 2008 Received: from jurassic-x4600.sfbay.sun.com (cretaceous.SFBay.Sun.COM [129.146.17.63]) by sac.sfbay.sun.com (8.13.8+Sun/8.13.8) with ESMTP id m4FIxM9D022622 for ; Thu, 15 May 2008 11:59:22 -0700 (PDT) Received: from [129.146.228.120] (chaz.SFBay.Sun.COM [129.146.228.120]) by jurassic-x4600.sfbay.sun.com (8.14.2+Sun/8.14.2) with ESMTP id m4FIxL39115155; Thu, 15 May 2008 11:59:21 -0700 (PDT) Message-ID: <482C879A.6000203@sun.com> Date: Thu, 15 May 2008 11:57:30 -0700 From: Rod Evans Reply-To: Rod.Evans@sun.com Organization: Sun Microsystems Inc. User-Agent: Thunderbird 2.0.0.6 (X11/20071119) MIME-Version: 1.0 To: Joseph Kowalski CC: "Roger A. Faulkner" , John.Plocher@sun.com, psarc-ext@sac.sfbay.sun.com, Bart.Smaalders@sun.com, Darrin.Johnson@sun.com Subject: Re: PSARC/2008/309 - expunge synonyms.h References: <200805151828.m4FISOTh109301@jurassic-x4600.sfbay.sun.com> <482C8562.8010700@sun.com> In-Reply-To: <482C8562.8010700@sun.com> Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Content-Length: 1230 Status: RO X-Status: $$$$ X-UID: 0000000023 Joseph Kowalski wrote: > What is the commitment level to "libshim"? More relevantly, does this > get a man page? Does it get no man page, but some other documentation > (linker guide?)? Its kinda strange, because I can understand lots of > reasons to document it in unique ways, but I suspect it should be > documented somewhere. 0@0.so.1 only exists in ld.so.1(1) as far as I know. Perhaps something similar for Rogers library. NOTES ... The user compatibility library /usr/lib/0@0.so.1 provides a mechanism that establishes a value of 0 at location 0. Some applications exist that erroneously assume a null character pointer should be treated the same as a pointer to a null string. A segmentation violation occurs in these applica- tions when a null character pointer is accessed. If this library is added to such an application at runtime using LD_PRELOAD, the library provides an environment that is sym- pathetic to this errant behavior. However, the user compati- bility library is intended neither to enable the generation of such applications, nor to endorse this particular pro- gramming practice. -- Rod. From Roger.Faulkner@sun.com Thu May 15 14:03:40 2008 Received: from jurassic-x4600.sfbay.sun.com (cretaceous.SFBay.Sun.COM [129.146.17.63]) by sac.sfbay.sun.com (8.13.8+Sun/8.13.8) with ESMTP id m4FL3exC025915 for ; Thu, 15 May 2008 14:03:40 -0700 (PDT) Received: from jurassic-x4600.sfbay.sun.com (jurassic-x4600.sfbay.sun.com [129.146.17.59]) by jurassic-x4600.sfbay.sun.com (8.14.2+Sun/8.14.2) with SMTP id m4FL3e9j137108; Thu, 15 May 2008 14:03:40 -0700 (PDT) Message-Id: <200805152103.m4FL3e9j137108@jurassic-x4600.sfbay.sun.com> Date: Thu, 15 May 2008 14:03:40 -0700 (PDT) From: "Roger A. Faulkner" Reply-To: "Roger A. Faulkner" Subject: Re: PSARC/2008/309 - expunge synonyms.h To: Darren.Moffat@sun.com, jek3@sun.com Cc: John.Plocher@sun.com, psarc-ext@sac.sfbay.sun.com, Bart.Smaalders@sun.com, Rod.Evans@sun.com, Darrin.Johnson@sun.com MIME-Version: 1.0 Content-Type: TEXT/plain; charset=us-ascii Content-MD5: mmy5ZsdCxygYMLdXqmcruQ== X-Mailer: dtmail 1.3.0 @(#)CDE Version 1.7_88 SunOS 5.11 i86pc i386 Content-Length: 3221 Status: RO X-Status: $$$$ X-UID: 0000000024 > Date: Thu, 15 May 2008 08:52:30 -1000 > From: Joseph Kowalski > Subject: Re: PSARC/2008/309 - expunge synonyms.h > To: Darren J Moffat > Cc: "Roger A. Faulkner" , John.Plocher@sun.com, psarc-ext@sac.sfbay.sun.com, Bart.Smaalders@sun.com, Rod.Evans@sun.com, Darrin.Johnson@sun.com > > Would be "underscore_syn.so.1" be enough? > > (I know we are supposed to be "linux familiar", but does that include > "linix-prefers-really-long-and-awkward-names"? :-) :-) ;-) ) > > - jek3 Ok. (Thanks to all.) This is my proposal, in the form of additional material in the libc(3LIB) manual page (a stability level of Stable would be fine with me -- it will never change): FILES /lib/c_synonyms.so.1 A compatibility library to provide access to obsolete libc synonym symbols /lib/64/c_synonyms.so.1 A 64-bit compatibility library to provide access to obsolete libc synonym symbols ... NOTES The synonyms compatibility library, c_synonyms.so.1, provides a mechanism to support old applications and libraries that were mistakenly built using now-obsolete synonym symbols from libc. Before the advent of direct binding (-B direct) libc provided many functions with two names. For example: getpwent() _getpwent() These two names referred to exactly the same function in libc. The leading-underscore symbol was intended to be used by system libraries in order to avoid conflicting with an application that might define its own version of getpwent() with completely different semantics. Standards-conforming applications may not define and use function names with leading underscores. Solaris system libraries are now built with direct binding. This means that a system library that calls getpwent() will bind directly to the instance of getpwent() in libc, even if the application to which it is linked defines a different getpwent() for its own use. The application binds to its instance of getpwent() and there is no resulting conflict. The direct binding mechanism is equally available to libraries not delivered with Solaris. As a result of this evolution, most of the leading-underscore synonym symbols have been removed from libc. This means that applications that call these now-obsolete function names will cease to work. They will typically draw the error: $ ./application ld.so.1: fatal: relocation error: symbol _getpwent: referenced symbol not found Killed All of the old leading-underscore symbols have been copied to the synonyms compatibility library. This library simply redirects the calls to the non-underscore instances of the corresponding functions in libc. Use it as a preloaded object: $ LD_PRELOAD=c_synonyms.so.1 ./application The synonyms compatibility library is intended neither to enable the generation of applications that call the obsolete leading-underscore synonym functions, nor to endorse this particular programming practice. From jek3@sun.com Thu May 15 14:20:21 2008 Received: from jurassic-x4600.sfbay.sun.com (cretaceous.SFBay.Sun.COM [129.146.17.59]) by sac.sfbay.sun.com (8.13.8+Sun/8.13.8) with ESMTP id m4FLKLcW026065 for ; Thu, 15 May 2008 14:20:21 -0700 (PDT) Received: from [129.150.13.200] (vpn-129-150-13-200.SFBay.Sun.COM [129.150.13.200]) by jurassic-x4600.sfbay.sun.com (8.14.2+Sun/8.14.2) with ESMTP id m4FLKKdZ141415; Thu, 15 May 2008 14:20:20 -0700 (PDT) Message-ID: <482CA977.2080302@sun.com> Date: Thu, 15 May 2008 11:21:59 -1000 From: Joseph Kowalski User-Agent: Thunderbird 2.0.0.9 (X11/20080225) MIME-Version: 1.0 To: "Roger A. Faulkner" CC: Darren.Moffat@sun.com, John.Plocher@sun.com, psarc-ext@sac.sfbay.sun.com, Bart.Smaalders@sun.com, Rod.Evans@sun.com, Darrin.Johnson@sun.com Subject: Re: PSARC/2008/309 - expunge synonyms.h References: <200805152103.m4FL3e9j137108@jurassic-x4600.sfbay.sun.com> In-Reply-To: <200805152103.m4FL3e9j137108@jurassic-x4600.sfbay.sun.com> Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Content-Length: 338 Status: RO X-Status: $$$$ X-UID: 0000000025 Roger A. Faulkner wrote: > Ok. (Thanks to all.) This is my proposal, in the form of additional > material in the libc(3LIB) manual page (a stability level of > Stable would be fine with me -- it will never change): > That would be Committed. 8^) > /lib/c_synonyms.so.1 > I like the choice. Thanks for doing this. - jek3 From scott.rotondo@sun.com Thu May 15 15:21:05 2008 Received: from jurassic-x4600.sfbay.sun.com (cretaceous.SFBay.Sun.COM [129.146.17.63]) by sac.sfbay.sun.com (8.13.8+Sun/8.13.8) with ESMTP id m4FML56T028596 for ; Thu, 15 May 2008 15:21:05 -0700 (PDT) Received: from [129.146.108.62] (vinifera.SFBay.Sun.COM [129.146.108.62]) by jurassic-x4600.sfbay.sun.com (8.14.2+Sun/8.14.2) with ESMTP id m4FML5EE153820; Thu, 15 May 2008 15:21:05 -0700 (PDT) Message-ID: <482CB751.7040205@sun.com> Date: Thu, 15 May 2008 15:21:05 -0700 From: Scott Rotondo User-Agent: Thunderbird 2.0.0.6 (X11/20071119) MIME-Version: 1.0 To: "Roger A. Faulkner" CC: Darren.Moffat@sun.com, jek3@sun.com, John.Plocher@sun.com, psarc-ext@sac.sfbay.sun.com, Bart.Smaalders@sun.com, Rod.Evans@sun.com, Darrin.Johnson@sun.com Subject: Re: PSARC/2008/309 - expunge synonyms.h References: <200805152103.m4FL3e9j137108@jurassic-x4600.sfbay.sun.com> In-Reply-To: <200805152103.m4FL3e9j137108@jurassic-x4600.sfbay.sun.com> Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Content-Length: 606 Status: RO X-Status: $$$$ X-UID: 0000000026 Roger A. Faulkner wrote: > Ok. (Thanks to all.) This is my proposal, in the form of additional > material in the libc(3LIB) manual page (a stability level of > Stable would be fine with me -- it will never change): > ... > > All of the old leading-underscore symbols have been copied > to the synonyms compatibility library. This library simply > redirects the calls to the non-underscore instances of the > corresponding functions in libc. Use it as a preloaded object: Just to be clear: "all" means all, not just the 55 symbols identified earlier. Is that correct? Scott From Roger.Faulkner@sun.com Thu May 15 16:30:48 2008 Received: from jurassic-x4600.sfbay.sun.com (cretaceous.SFBay.Sun.COM [129.146.17.63]) by sac.sfbay.sun.com (8.13.8+Sun/8.13.8) with ESMTP id m4FNUm70000596 for ; Thu, 15 May 2008 16:30:48 -0700 (PDT) Received: from jurassic-x4600.sfbay.sun.com (jurassic-x4600.sfbay.sun.com [129.146.17.59]) by jurassic-x4600.sfbay.sun.com (8.14.2+Sun/8.14.2) with SMTP id m4FNUmIH164257; Thu, 15 May 2008 16:30:48 -0700 (PDT) Message-Id: <200805152330.m4FNUmIH164257@jurassic-x4600.sfbay.sun.com> Date: Thu, 15 May 2008 16:30:48 -0700 (PDT) From: "Roger A. Faulkner" Reply-To: "Roger A. Faulkner" Subject: Re: PSARC/2008/309 - expunge synonyms.h To: scott.rotondo@sun.com Cc: Darren.Moffat@sun.com, jek3@sun.com, John.Plocher@sun.com, psarc-ext@sac.sfbay.sun.com, Bart.Smaalders@sun.com, Rod.Evans@sun.com, Darrin.Johnson@sun.com MIME-Version: 1.0 Content-Type: TEXT/plain; charset=us-ascii Content-MD5: yYsRKTlDsYrMAgg5m4hwXA== X-Mailer: dtmail 1.3.0 @(#)CDE Version 1.7_88 SunOS 5.11 i86pc i386 Content-Length: 1138 Status: RO X-Status: $$$$ X-UID: 0000000027 > Date: Thu, 15 May 2008 15:21:05 -0700 > From: Scott Rotondo > Subject: Re: PSARC/2008/309 - expunge synonyms.h > To: "Roger A. Faulkner" > Cc: Darren.Moffat@sun.com, jek3@sun.com, John.Plocher@sun.com, psarc-ext@sac.sfbay.sun.com, Bart.Smaalders@sun.com, Rod.Evans@sun.com, Darrin.Johnson@sun.com > > Roger A. Faulkner wrote: > > Ok. (Thanks to all.) This is my proposal, in the form of additional > > material in the libc(3LIB) manual page (a stability level of > > Stable would be fine with me -- it will never change): > > > > ... > > > > > All of the old leading-underscore symbols have been copied > > to the synonyms compatibility library. This library simply > > redirects the calls to the non-underscore instances of the > > corresponding functions in libc. Use it as a preloaded object: > > Just to be clear: "all" means all, not just the 55 symbols identified > earlier. Is that correct? > > Scott Yes. "all" means about 985 names, including symbols that aren't being removed from libc on this go-around (about 600 of them are). Roger From scott.rotondo@sun.com Thu May 15 16:35:53 2008 Received: from jurassic-x4600.sfbay.sun.com (cretaceous.SFBay.Sun.COM [129.146.17.59]) by sac.sfbay.sun.com (8.13.8+Sun/8.13.8) with ESMTP id m4FNZrH8000777 for ; Thu, 15 May 2008 16:35:53 -0700 (PDT) Received: from [129.146.108.62] (vinifera.SFBay.Sun.COM [129.146.108.62]) by jurassic-x4600.sfbay.sun.com (8.14.2+Sun/8.14.2) with ESMTP id m4FNZrAL164819; Thu, 15 May 2008 16:35:53 -0700 (PDT) Message-ID: <482CC8D9.9020900@sun.com> Date: Thu, 15 May 2008 16:35:53 -0700 From: Scott Rotondo User-Agent: Thunderbird 2.0.0.6 (X11/20071119) MIME-Version: 1.0 To: "Roger A. Faulkner" CC: psarc-ext@sac.sfbay.sun.com Subject: Re: PSARC/2008/309 - expunge synonyms.h References: <200805152330.m4FNUmIH164257@jurassic-x4600.sfbay.sun.com> In-Reply-To: <200805152330.m4FNUmIH164257@jurassic-x4600.sfbay.sun.com> Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Content-Length: 629 Status: RO X-Status: $$$$ X-UID: 0000000028 Roger A. Faulkner wrote: >>> All of the old leading-underscore symbols have been copied >>> to the synonyms compatibility library. This library simply >>> redirects the calls to the non-underscore instances of the >>> corresponding functions in libc. Use it as a preloaded object: >> Just to be clear: "all" means all, not just the 55 symbols identified >> earlier. Is that correct? >> >> Scott > > Yes. "all" means about 985 names, including symbols that > aren't being removed from libc on this go-around (about 600 > of them are). > > Roger Great. I think this is a really good solution. Scott