Bug#627672: PR 28322 patch is still wrong
Package: gcc-4.3
Version: 4.3.2-1.1
The changelog for gcc-4.3 says:
gcc-4.3 (4.3.1-5) unstable; urgency=low
* Update to SVN 20080705 from the gcc-4_3-branch.
- Fix PR target/36634, wrong-code on powerpc with -msecure-plt.
* Fix PR target/35965, PIC + -fstack-protector on arm/armel. Closes:
#469517.
* Don't run the libjava testsuite with -mabi=n32.
* Update patch for PR other/28322, that unknown -Wno-* options do not
cause errors, but warnings instead.
* On m68k, add -fgnu89-inline when in gnu99 mode (requested by Michael
Casadeval for the m68k port). Closes: #489234.
-- Matthias Klose Sun, 06 Jul 2008 01:39:30 +0200
However, on released lenny:
mariner:~/junk> gcc -DHACK -Wno-sponge t.c
t.c: In function 'main':
t.c:7: warning: initialization makes pointer from integer without a cast
At top level:
cc1: error: unrecognized command line option "-Wno-sponge"
mariner:~/junk> cat t.c
#include
int main(void)
{
#ifdef HACK
void *p = 0x1234;
#else
void *p = NULL;
#endif
printf("P is %p\n", p);
}
mariner:~/junk>
This means that this gcc has a broken version of the fix for
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=28322
This is particularly annoying as it means that programs' whose
configure scripts attempt to discover whether a particular -Wno-foo
option is supported will fail to spot that it isn't, pass it anyway,
and then fall over if there are any other warnings. (Only programs
not using -Werror are affected.)
sqeeze is fine. Could we consider an update to improve this for
lenny ?
Ian.
--
To UNSUBSCRIBE, email to debian-gcc-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive:
http://lists.debian.org/19930.28667.746285.390...@chiark.greenend.org.uk
Bug#624743: cannot reproduce
found 624743 4.6.0-8 thanks Hello, hurd-i386 is also hit by the bug. Andreas Metzler, le Fri 06 May 2011 20:16:49 +0200, a écrit : > On 2011-05-05 Kees Cook wrote: > > Hi! Thanks for this report. I can't reproduce this segfault. I tried the > > builds both amd64 and i386, and both build fine with 4.6.0-6 for me. Do you > > have any minimal reproducers that might show this more specifically? > > Sadly I cannot reproduce this anymore either. Some of the since > upgraded build-deps must have changed. I can reproduce it with gcc-4.6 4.6.0-8, but only by using valgrind: in debian/minimaltest, prepend valgrind to the line $2 -C "$top/exim4.conf" -bV valgrind will then warn: ==29180== Source and destination overlap in memcpy(0x6fad4e8, 0x6fad4e0, 88) ==29180==at 0x4C25F6A: memcpy (mc_replace_strmem.c:497) ==29180==by 0x13B5D9: init_lookup_list (string3.h:59) ==29180==by 0x11FF6E: main (exim.c:3615) which is the same backtrace reported here. Disassembling init_lookup_list shows this: 0x000335cb <+459>: cltq 0x000335cd <+461>: lea0x0(,%rax,8),%rdx 0x000335d5 <+469>: callq 0x153a8 0x000335da <+474>: mov%rbx,%rcx 0x000335dd <+477>: add0x2b6fac(%rip),%rcx# 0x2ea590 0x000335e4 <+484>: mov0x18(%rsp),%rdx i.e. a memcpy call. l * 0x000335d5 points at 59return __builtin___memmove_chk (__dest, __src, __len, __bos0 (__dest)); which is actually a memmove call, not a memcpy call! The memmove call comes from the add_lookup_to_list() inline. By replacing it with __builtin_memmove() to avoid the _chk version, I don't get any valgrind issue any more. It looks like gcc-4.6 is here erroneously optimizing __builtin___memmove_chk into a memcpy call! Samuel -- To UNSUBSCRIBE, email to debian-gcc-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/20110523215927.ga3...@const.famille.thibault.fr
Processed: Re: Bug#624743: cannot reproduce
Processing commands for cont...@bugs.debian.org: > found 624743 4.6.0-8 Bug #624743 [gcc-4.6] gcc-4.6 -D_FORTIFY_SOURCE=2 miscompiles exim4 Bug Marked as found in versions gcc-4.6/4.6.0-8. > thanks Stopping processing here. Please contact me if you need assistance. -- 624743: http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=624743 Debian Bug Tracking System Contact ow...@bugs.debian.org with problems -- To UNSUBSCRIBE, email to debian-gcc-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/handler.s.c.13061879731935.transcr...@bugs.debian.org
Bug#626719: g++-4.6 intenal error
G++ fail to compile anything (void file, main stub, helo world, other app) > g++: internal compiler error: Illegal instruction (program cc1plus) > Please submit a full bug report, > with preprocessed source if appropriate. > See for instructions. uname -a > Linux SQLek 2.6.38-2-486 #1 Sun May 8 14:04:15 UTC 2011 i586 GNU/Linux g++ --version > g++ (Debian 4.6.0-8) 4.6.1 20110521 (prerelease) > Copyright (C) 2011 Free Software Foundation, Inc. > This is free software; see the source for copying conditions. There is NO > warranty; not even for MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. What more can i append to bugreport ? -- To UNSUBSCRIBE, email to debian-gcc-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/20110524030001.GA12299@SQLek.DOMek
