Re: c++/8889: g++ 3.2.1 (also 3.2 it) miscompile Qt

Re: c++/8889: g++ 3.2.1 (also 3.2 it) miscompile Qt

[reichelt]

Synopsis: g++ 3.2.1 (also 3.2 it) miscompile Qt

State-Changed-From-To: feedback->open
State-Changed-By: reichelt
State-Changed-When: Thu Dec 12 08:12:04 2002
State-Changed-Why:
    Might be still an ABI issue, given the fact that
    the problems arose between gcc 3.1.1 and 3.2.
    Maybe Qt makes some assumptions about the ABI
    which do not hold any more, but that's just a wild guess.

http://gcc.gnu.org/cgi-bin/gnatsweb.pl?cmd=view%20audit-trail&database=gcc&pr=8889

Re: c++/8889: g++ 3.2.1 (also 3.2 it) miscompile Qt

[Zack Weinberg]

The following reply was made to PR c++/8889; it has been noted by GNATS.

From: Zack Weinberg <zack@codesourcery.com>
To: Felix von Leitner <felix-gcc@fefe.de>
Cc: bangerth@dealii.org, gcc-bugs@gcc.gnu.org, gcc-gnats@gcc.gnu.org
Subject: Re: c++/8889: g++ 3.2.1 (also 3.2 it) miscompile Qt
Date: Mon, 16 Dec 2002 17:25:09 -0800

 Felix von Leitner <felix-gcc@fefe.de> writes:
 
 > I made a bug report and you refuse to even make a feeble attempt to
 > reproduce it.  Instead you first claim that it's an ABI problem (which
 > is not and I wrote that), then you claim you need more information (you
 > don't, I included everything that is necessary to reproduce the problem)
 > and then you claim that I withhold necessary information.
 
 As a matter of project policy, the GCC team expects you to do more of
 the work of isolating the bug.  I would like to draw your attention to
 http://gcc.gnu.org/bugs.html, which clearly states that you are
 expected to provide a self-contained, already preprocessed, as small
 as feasible, source file which is being miscompiled, along with an
 explanation of what it is supposed to do and what it does instead.
 
 We make this requirement because our experience is that, in a
 circumstance such as yours, we would be unable to reproduce the bug by
 downloading the software packages you mentioned and compiling them
 ourselves.  Tiny differences in system headers, for instance, can
 perturb the bug out of manifesting.  We would then have wasted a great
 deal of effort.
 
 zw

Re: c++/8889: g++ 3.2.1 (also 3.2 it) miscompile Qt

[Felix von Leitner]

The following reply was made to PR c++/8889; it has been noted by GNATS.

From: Felix von Leitner <felix-gcc@fefe.de>
To: bangerth@dealii.org, gcc-bugs@gcc.gnu.org, gcc-prs@gcc.gnu.org,
  nobody@gcc.gnu.org, gcc-gnats@gcc.gnu.org
Cc:  
Subject: Re: c++/8889: g++ 3.2.1 (also 3.2 it) miscompile Qt
Date: Tue, 17 Dec 2002 01:57:15 +0100

 Thus spake bangerth@dealii.org (bangerth@dealii.org):
 > State-Changed-From-To: feedback->closed
 > State-Changed-By: bangerth
 > State-Changed-When: Mon Dec 16 16:08:37 2002
 > State-Changed-Why:
 >     I refuse to work on PRs where I am discriminated against
 >     for my nationality (which the reporter mistakenly assumed
 >     to be different from his own), and I think that in general
 >     gcc is not the place for insult.
 
 I made a bug report and you refuse to even make a feeble attempt to
 reproduce it.  Instead you first claim that it's an ABI problem (which
 is not and I wrote that), then you claim you need more information (you
 don't, I included everything that is necessary to reproduce the problem)
 and then you claim that I withhold necessary information.
 
 As if this travesty is not enough, now you feel discriminated against?
 
 Is nobody here interested in whether the latest gcc release actually
 miscompiles the most widely used C++ toolkit (Qt) on the most widely
 used gcc platform (x86-linux) using the recommended flags?
 
 What is the bug report form good for, then?
 
 Is this a technical bug tracking and removal instrument or the personal
 playground of some "I don't have the time to try to reproduce your bug,
 but maybe you could try running ldd again" weenies?
 
 Sheesh.  I could care less whether Mr bangerth here has the time to try
 and reproduce my bug report.  But I do care whether he sabotages gcc by
 closing open bugs he did not even try to reproduce!
 
 Felix

Re: c++/8889: g++ 3.2.1 (also 3.2 it) miscompile Qt

[bangerth]

Synopsis: g++ 3.2.1 (also 3.2 it) miscompile Qt

State-Changed-From-To: feedback->closed
State-Changed-By: bangerth
State-Changed-When: Mon Dec 16 16:08:37 2002
State-Changed-Why:
    I refuse to work on PRs where I am discriminated against
    for my nationality (which the reporter mistakenly assumed
    to be different from his own), and I think that in general
    gcc is not the place for insult.
    
    I thus take the liberty to close this PR. If someone should
    have the strong urge to work on this anyway, please feel
    free to reopen it.

http://gcc.gnu.org/cgi-bin/gnatsweb.pl?cmd=view%20audit-trail&database=gcc&pr=8889

Re: c++/8889: g++ 3.2.1 (also 3.2 it) miscompile Qt

[Wolfgang Bangerth]

[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #1: Type: text/plain, Size: 1329 bytes --]

The following reply was made to PR c++/8889; it has been noted by GNATS.

From: Wolfgang Bangerth <bangerth@ticam.utexas.edu>
To: gcc-gnats@gcc.gnu.org
Cc:  
Subject: Re: c++/8889: g++ 3.2.1 (also 3.2 it) miscompile Qt
Date: Mon, 16 Dec 2002 17:58:47 -0600 (CST)

 ---------- Forwarded message ----------
 Date: Tue, 17 Dec 2002 00:21:05 +0100
 From: Felix von Leitner <felix-gcc@fefe.de>
 To: Wolfgang Bangerth <bangerth@ticam.utexas.edu>
 Subject: Re: c++/8889: g++ 3.2.1 (also 3.2 it) miscompile Qt
 
 Thus spake Wolfgang Bangerth (bangerth@ticam.utexas.edu):
 > Falls Dir das nicht aufgefallen sein sollte: ich komme aus dem selben Land 
 > wie Du, spreche die selbe Sprache (allerdings wesentlich hoeflicher) und 
 > habe auch ansonsten noch anderes zu tun (zum Beispiel mein eigenes Open 
 > Source Projekt).
 
 Woher soll mir das auffallen, utexas.edu sieht nicht danach aus.
 
 > Ansonsten ist wohl alles gesagt.
 
 In der Tat.  Selten hat mich jemand derartig verarscht bei einem Bug
 Report für ein open source Projekt.  Insbesondere vom GNU-Projekt ist
 man gewohnt, daß der Autor zumindest das Problem reproduziert.
 
 Hast du überhaupt schon mal jemandem helfen können oder machst du immer
 nur irgendwelche nebulösen Andeutungen auf Basis von ignoriertem Inhalt
 eines Bug Reports?
 
 Unglaublich.
 
 Felix
 

Re: c++/8889: g++ 3.2.1 (also 3.2 it) miscompile Qt

[Wolfgang Bangerth]

[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #1: Type: text/plain, Size: 1691 bytes --]

The following reply was made to PR c++/8889; it has been noted by GNATS.

From: Wolfgang Bangerth <bangerth@ticam.utexas.edu>
To: gcc-gnats@gcc.gnu.org
Cc:  
Subject: Re: c++/8889: g++ 3.2.1 (also 3.2 it) miscompile Qt
Date: Mon, 16 Dec 2002 18:00:36 -0600 (CST)

 ---------- Forwarded message ----------
 Date: Tue, 17 Dec 2002 00:40:11 +0100
 From: Felix von Leitner <felix-dietlibc@fefe.de>
 To: Wolfgang Bangerth <bangerth@ticam.utexas.edu>
 Subject: Re: c++/8889: g++ 3.2.1 (also 3.2 it) miscompile Qt
 
 Thus spake Wolfgang Bangerth (bangerth@ticam.utexas.edu):
 > > Woher soll mir das auffallen, utexas.edu sieht nicht danach aus.
 > Dafuer habe ich ja meine Homepage in der Signatur.
 
 ... auch in Englisch.  Du hast in Heidelberg studiert, das heißt auch
 noch nicht viel.  An der FU Berlin gibt es jedenfalls Studenten, die
 kein Deutsch sprechen.
 
 > > Unglaublich.
 > Und ich denke dafuer waere ein Blick auf die Mailinglisten von gcc-bugs
 > ganz gut.
 
 Ach weißt du, so wichtig ist mir das nicht.
 Aber wenn schon die drei Stützpfeiler des GNU-Projekts (abgesehen von
 Hurd, da wundert sich ja eh niemand mehr) nicht miteinander wollen, dann
 kann man m.E. gleich das ganze Projekt einstellen.
 
 Im Übrigen ist gerade gcc gerade seit egcs-Übernahme das Teilprojekt mit
 der größten Test-Suite.  Da muß alles mögliche sauber kompilieren, damit
 der gcc überhaupt released wird, und da gehört eben dann auch Qt rein,
 wenn es nicht schon drin ist.  Ich finde die Seite gerade nicht.  Wenn
 Qt schon drin ist, dann reicht es eben nicht, wenn es kompiliert,
 sondern man muß es auch ein bißchen testen.  Keine Ahnung wie, ist auch
 nicht mein Problem.
 
 Felix
 

Re: c++/8889: g++ 3.2.1 (also 3.2 it) miscompile Qt

[Wolfgang Bangerth]

The following reply was made to PR c++/8889; it has been noted by GNATS.

From: Wolfgang Bangerth <bangerth@ticam.utexas.edu>
To: gcc-gnats@gcc.gnu.org
Cc:  
Subject: Re: c++/8889: g++ 3.2.1 (also 3.2 it) miscompile Qt
Date: Mon, 16 Dec 2002 17:16:12 -0600 (CST)

 ---------- Forwarded message ----------
 Date: Tue, 17 Dec 2002 00:11:38 +0100
 From: Felix von Leitner <felix-gcc@fefe.de>
 To: Wolfgang Bangerth <bangerth@ticam.utexas.edu>
 Subject: Re: c++/8889: g++ 3.2.1 (also 3.2 it) miscompile Qt
 
 Wolfgang,
 
 I find it highly insulting that you are now trying ot threaten me in
 this childish way.  If you think it is too much to ask of THE GNU
 COMPILER to expect it to properly compile THE GNU LIBC using THE GNU
 BINUTILS, then it's good to hear now and I will switch to Windows.  At
 least they don't try to bullshit me like this.
 
 In case you really didn't notice: I gave you _all_ of the information I
 have.  I told you the binutils version, the gcc version, the glibc
 version, and the qt version, and I gave the the CFLAGS I used.
 
 And now you tell me it's too much to ask that you at least try to
 reproduce the problem?  Instead you give me vague conspiracy theories
 about potential third party C++ libraries I might use (which, by the
 way, is not the case, as I also wrote in the email).  So if you feel I
 am withholding information, I can't help you.
 
 I manage quite a few open source projects myself.  One of them is a
 libc.  I wouldn't be bothering you if I hadn't exhausted the other
 options.  If trying to reproduce a reported is too much to ask, then
 please get out of the way and let people handle this who are willing to
 invest the necessary time.
 
 > I'm not telling you anything except for this: gcc is in most part a 
 > volunteer project.
 
 No, really?
 I didn't know!
 Now _that_ changes everything!
 Of course then you don't have to try to reproduce problems... ?!  What
 the hell are you smoking there?
 
 > Nobody's paying me for the work, and you can't force me to do
 > something.
 
 Why do you waste my time with this?
 Stop whining and do something useful.
 If it ain't for gcc, then do something else useful.
 
 > If you don't want to cooperate and try to provide more information,
 > that's fine with me, but don't insult me.
 
 ROTFL.  So I insulted you.  I think I start to understand why so many
 Americans need therapy.
 
 > It's not reasonable to ask me to build Qt+KDE with a new compiler just to 
 > experience "that it does not work".
 
 That's why I only asked you to build Qt and xca, which is about 1/4 the
 efford.  You didn't even build gcc and Qt yourself.  What makes you
 think you are even qualified to debug this?
 
 > We need more information to figure out what goes wrong.
 
 No, you don't.
 Just build Qt and xca, start it, and see if it segfaults within the
 first second for no apparent reason.
 
 > I leave this report in feedback state.
 
 Good.
 
 Felix
 

Re: c++/8889: g++ 3.2.1 (also 3.2 it) miscompile Qt

[Wolfgang Bangerth]

The following reply was made to PR c++/8889; it has been noted by GNATS.

From: Wolfgang Bangerth <bangerth@ticam.utexas.edu>
To: Felix von Leitner <felix-gcc@fefe.de>
Cc: bangerth@dealii.org, <gcc-bugs@gcc.gnu.org>, <gcc-gnats@gcc.gnu.org>
Subject: Re: c++/8889: g++ 3.2.1 (also 3.2 it) miscompile Qt
Date: Mon, 16 Dec 2002 15:01:29 -0600 (CST)

 I'm giving up.
 
 > It should be a trivial exercise to try to reproduce my problem.
 > [...]
 > 
 > >     On my system, these libraries all contain no C++ symbols,
 > >     except libstdc++, of course, but you should repeat this
 > >     with the executables/libraries that fail for you.
 > 
 > No, I shouldn't.
 > These are no obscure projects from some Joe Schmoe.  I used standard
 > packages and default flags.  It ought to be a matter of seconds to find
 > someone in the gcc team who compiled these packages with the current
 > gcc.
 > [...] 
 >
 > And before someone asks: no, I am not overclocking my machine.
 > I am not using unstable or CVS versions of anything.
 > I just want to run a system using the latest version of each of the
 > major packages, including binutils, gcc, glibc and Qt.  That ought to be
 > possible, right?  I mean, come on, Qt is the most important C++ project!
 > For many people it is the only reason to install C++ support at all!
 > 
 > Are you telling me that noone has compiled Qt with the current gcc and
 > binutils yet?
 
 I'm not telling you anything except for this: gcc is in most part a 
 volunteer project. Nobody's paying me for the work, and you can't force me 
 to do something. If you don't want to cooperate and try to provide more 
 information, that's fine with me, but don't insult me.
 
 It's not reasonable to ask me to build Qt+KDE with a new compiler just to 
 experience "that it does not work". We need more information to figure out 
 what goes wrong. If you don't want to provide it -- fine.
 
 I leave this report in feedback state. This way either someone else will 
 look at it, or someone (possibly me) will close it in 3 months time since 
 no reasonable feedback is forthcoming.
 
 Regards
   Wolfgang
 
 -------------------------------------------------------------------------
 Wolfgang Bangerth              email:           bangerth@ticam.utexas.edu
                                www: http://www.ticam.utexas.edu/~bangerth
 
 

Re: c++/8889: g++ 3.2.1 (also 3.2 it) miscompile Qt

[Felix von Leitner]

The following reply was made to PR c++/8889; it has been noted by GNATS.

From: Felix von Leitner <felix-gcc@fefe.de>
To: bangerth@dealii.org, gcc-bugs@gcc.gnu.org, gcc-prs@gcc.gnu.org,
  nobody@gcc.gnu.org, gcc-gnats@gcc.gnu.org
Cc:  
Subject: Re: c++/8889: g++ 3.2.1 (also 3.2 it) miscompile Qt
Date: Mon, 16 Dec 2002 21:53:39 +0100

 Thus spake bangerth@dealii.org (bangerth@dealii.org):
 > State-Changed-From-To: open->feedback
 > State-Changed-By: bangerth
 > State-Changed-When: Mon Dec 16 11:26:23 2002
 > State-Changed-Why:
 >     I'm still leaning with Volker's ABI theory. What one
 >     would need to check is what libraries Qt actually pulls
 >     in. For example, on my system these are:
 >     tmp/g> ldd /usr/lib/libqt.so.3
 >             libpng.so.2 => /usr/lib/libpng.so.2 (0x40618000)
 >             libz.so.1 => /lib/libz.so.1 (0x40649000)
 >             libGL.so.1 => /usr/lib/libGL.so.1 (0x40658000)
 >             libXmu.so.6 => /usr/X11R6/lib/libXmu.so.6 (0x4080f000)
 >             libICE.so.6 => /usr/X11R6/lib/libICE.so.6 (0x40826000)
 >             libSM.so.6 => /usr/X11R6/lib/libSM.so.6 (0x4083e000)
 >             libdl.so.2 => /lib/libdl.so.2 (0x40848000)
 >             libXext.so.6 => /usr/X11R6/lib/libXext.so.6 (0x4084c000)
 >             libX11.so.6 => /usr/X11R6/lib/libX11.so.6 (0x4085a000)
 >             libXrender.so.1 => /usr/X11R6/lib/libXrender.so.1 (0x4091a000)
 >             libXft.so.1 => /usr/X11R6/lib/libXft.so.1 (0x40920000)
 >             libfreetype.so.6 => /usr/lib/libfreetype.so.6 (0x4094a000)
 >             libstdc++-libc6.2-2.so.3 => /usr/lib/libstdc++-libc6.2-2.so.3 (0x4098a000)
 >             libm.so.6 => /lib/libm.so.6 (0x409d7000)
 >             libc.so.6 => /lib/libc.so.6 (0x409fa000)
 >             libXt.so.6 => /usr/X11R6/lib/libXt.so.6 (0x40b22000)
 >             libXi.so.6 => /usr/X11R6/lib/libXi.so.6 (0x40b70000)
 >             /lib/ld-linux.so.2 => /lib/ld-linux.so.2 (0x80000000)
 
 As you noted yourself, the only relevant library is libstdc++, which in
 this listing obviously is the wrong one.  In my case it's libstdc++.so.5.
 Please compile Qt yourself with gcc 3.2.1 and see if you can reproduce
 the problem.  That should be the first prudent cause of action, I think.
 
 Has anyone ever successfully compiled Qt 3.1 with g++ 3.2.1?
 I use glibc 2.2.5 because the current glibc 2.3 versions all dump core
 in the compilation process (I opened a glibc bug for this).
 I used the default Qt options and I used the cxa atexit switch that is
 recommended in the gcc web pages for 100% ABI compliance.
 
 It should be a trivial exercise to try to reproduce my problem.  I'm not
 using some obscure platform or anything.  Just default or recommended
 flags for the involved major packages.  If anyone else was able to
 successfully compile these packages, that would be enough for me.  No
 need to tell me to look for stuff I already looked at.
 
 >     On my system, these libraries all contain no C++ symbols,
 >     except libstdc++, of course, but you should repeat this
 >     with the executables/libraries that fail for you.
 
 No, I shouldn't.
 These are no obscure projects from some Joe Schmoe.  I used standard
 packages and default flags.  It ought to be a matter of seconds to find
 someone in the gcc team who compiled these packages with the current
 gcc.  I assumed these packages are important enough to be on the
 pre-release build process anyway.
 
 And before someone asks: no, I am not overclocking my machine.
 I am not using unstable or CVS versions of anything.
 I just want to run a system using the latest version of each of the
 major packages, including binutils, gcc, glibc and Qt.  That ought to be
 possible, right?  I mean, come on, Qt is the most important C++ project!
 For many people it is the only reason to install C++ support at all!
 
 Are you telling me that noone has compiled Qt with the current gcc and
 binutils yet?
 

Re: c++/8889: g++ 3.2.1 (also 3.2 it) miscompile Qt

[bangerth]

Synopsis: g++ 3.2.1 (also 3.2 it) miscompile Qt

State-Changed-From-To: open->feedback
State-Changed-By: bangerth
State-Changed-When: Mon Dec 16 11:26:23 2002
State-Changed-Why:
    I'm still leaning with Volker's ABI theory. What one
    would need to check is what libraries Qt actually pulls
    in. For example, on my system these are:
    tmp/g> ldd /usr/lib/libqt.so.3
            libpng.so.2 => /usr/lib/libpng.so.2 (0x40618000)
            libz.so.1 => /lib/libz.so.1 (0x40649000)
            libGL.so.1 => /usr/lib/libGL.so.1 (0x40658000)
            libXmu.so.6 => /usr/X11R6/lib/libXmu.so.6 (0x4080f000)
            libICE.so.6 => /usr/X11R6/lib/libICE.so.6 (0x40826000)
            libSM.so.6 => /usr/X11R6/lib/libSM.so.6 (0x4083e000)
            libdl.so.2 => /lib/libdl.so.2 (0x40848000)
            libXext.so.6 => /usr/X11R6/lib/libXext.so.6 (0x4084c000)
            libX11.so.6 => /usr/X11R6/lib/libX11.so.6 (0x4085a000)
            libXrender.so.1 => /usr/X11R6/lib/libXrender.so.1 (0x4091a000)
            libXft.so.1 => /usr/X11R6/lib/libXft.so.1 (0x40920000)
            libfreetype.so.6 => /usr/lib/libfreetype.so.6 (0x4094a000)
            libstdc++-libc6.2-2.so.3 => /usr/lib/libstdc++-libc6.2-2.so.3 (0x4098a000)
            libm.so.6 => /lib/libm.so.6 (0x409d7000)
            libc.so.6 => /lib/libc.so.6 (0x409fa000)
            libXt.so.6 => /usr/X11R6/lib/libXt.so.6 (0x40b22000)
            libXi.so.6 => /usr/X11R6/lib/libXi.so.6 (0x40b70000)
            /lib/ld-linux.so.2 => /lib/ld-linux.so.2 (0x80000000)
    
    
    On my system, these libraries all contain no C++ symbols,
    except libstdc++, of course, but you should repeat this
    with the executables/libraries that fail for you.
    
    Please understand that we do not want to push away
    responsibility for possible bugs to you, but that simple
    the problem you describe is so vague, and the piece of
    code in which it happens so large, that there is about
    no way for us to track this problem down. So we just
    try to eliminate the simple problems first.
    
    Regards
      Wolfgang

http://gcc.gnu.org/cgi-bin/gnatsweb.pl?cmd=view%20audit-trail&database=gcc&pr=8889

Re: c++/8889: g++ 3.2.1 (also 3.2 it) miscompile Qt

[Felix von Leitner]

The following reply was made to PR c++/8889; it has been noted by GNATS.

From: Felix von Leitner <felix-gcc@fefe.de>
To: reichelt@igpm.rwth-aachen.de, gcc-bugs@gcc.gnu.org, gcc-prs@gcc.gnu.org,
  nobody@gcc.gnu.org, gcc-gnats@gcc.gnu.org
Cc:  
Subject: Re: c++/8889: g++ 3.2.1 (also 3.2 it) miscompile Qt
Date: Thu, 12 Dec 2002 16:05:24 +0100

 > Synopsis: g++ 3.2.1 (also 3.2 it) miscompile Qt
 
 > State-Changed-From-To: open->feedback
 > State-Changed-By: reichelt
 > State-Changed-When: Thu Dec 12 02:50:53 2002
 > State-Changed-Why:
 >     Between gcc 3.1.1 and 3.2 the ABI changed.
 >     You have to recompile all the C++-libraries that you are linking.
 >     Might that be the cause for your failures?
 
 > http://gcc.gnu.org/cgi-bin/gnatsweb.pl?cmd=view%20audit-trail&database=gcc&pr=8889
 
 No.  xca does not used any C++ libraries besides libstdc++ and Qt.  I
 recompiled Qt and libstdc++ comes with gcc.
 
 Felix

Re: c++/8889: g++ 3.2.1 (also 3.2 it) miscompile Qt

[reichelt]

Synopsis: g++ 3.2.1 (also 3.2 it) miscompile Qt

State-Changed-From-To: open->feedback
State-Changed-By: reichelt
State-Changed-When: Thu Dec 12 02:50:53 2002
State-Changed-Why:
    Between gcc 3.1.1 and 3.2 the ABI changed.
    You have to recompile all the C++-libraries that you are linking.
    Might that be the cause for your failures?

http://gcc.gnu.org/cgi-bin/gnatsweb.pl?cmd=view%20audit-trail&database=gcc&pr=8889