|
From: Christian P. <tr...@ge...> - 2005-04-02 11:02:35
|
that's strange. I just compiled valgrind-2.4.0 on a new chroot (32bit-only) environment on = my=20 amd64. I got a segfault, so I just remerged it again. But it failed. That m= ay=20 be because I first had default compiler (gcc3.3) installed and now have=20 gcc3.4. however, it complains about a const-ness problem: if i686-pc-linux-gnu-gcc -DHAVE_CONFIG_H -I. -I. -I.. -I../coregrind=20 =2DI../coregrind -I.. -I../coregrind/x86 -I../coregrind/x86=20 =2DI../coregrind/linux -I../coregrind/x86-linux -I../include -I../include=20 =2DI../include/x86 -I../include/linux -I../include/x86-linux=20 =2DDVG_LIBDIR=3D"\"/usr/lib/valgrind"\" -I./demangle -DKICKSTART_BASE=3D0xb= 0000000=20 =2DDVG_PLATFORM=3D"\"x86-linux"\" -Winline -Wall -Wshadow -O -g=20 =2Dmpreferred-stack-boundary=3D2 -DELFSZ=3D32 -fpie -nopie -O0 -ggdb3 -pipe= =20 =2Dfno-pie -Wno-long-long -MT stage2-vg_memory.o -MD -MP -MF=20 ".deps/stage2-vg_memory.Tpo" -c -o stage2-vg_memory.o `test -f 'vg_memory.c= '=20 || echo './'`vg_memory.c; \ then mv -f ".deps/stage2-vg_memory.Tpo" ".deps/stage2-vg_memory.Po"; else r= m=20 =2Df ".deps/stage2-vg_memory.Tpo"; exit 1; fi vg_memory.c: In function `vgPlain_unmap_range': vg_memory.c:161: error: initializer element is not constant vg_memory.c: In function `vgPlain_map_file_segment': vg_memory.c:327: error: initializer element is not constant vg_memory.c: In function `vgPlain_mprotect_range': vg_memory.c:482: error: initializer element is not constant vg_memory.c: In function `vgPlain_find_map_space': vg_memory.c:527: error: initializer element is not constant make[4]: *** [stage2-vg_memory.o] Error 1 battousai valgrind-2.4.0 # gcc -v Reading specs from /usr/lib/gcc/i686-pc-linux-gnu/3.4.3-20050110/specs Configured=20 with: /var/tmp/portage/gcc-3.4.3.20050110-r1/work/gcc-3.4.3/configure=20 =2D-enable-version-specific-runtime-libs --prefix=3D/usr=20 =2D-bindir=3D/usr/i686-pc-linux-gnu/gcc-bin/3.4.3-20050110=20 =2D-includedir=3D/usr/lib/gcc/i686-pc-linux-gnu/3.4.3-20050110/include=20 =2D-datadir=3D/usr/share/gcc-data/i686-pc-linux-gnu/3.4.3-20050110=20 =2D-mandir=3D/usr/share/gcc-data/i686-pc-linux-gnu/3.4.3-20050110/man=20 =2D-infodir=3D/usr/share/gcc-data/i686-pc-linux-gnu/3.4.3-20050110/info=20 =2D-with-gxx-include-dir=3D/usr/lib/gcc/i686-pc-linux-gnu/3.4.3-20050110/in= clude/g++-v3=20 =2D-host=3Di686-pc-linux-gnu --disable-altivec --enable-nls=20 =2D-without-included-gettext --with-system-zlib --disable-checking=20 =2D-disable-werror --disable-libunwind-exceptions --disable-multilib=20 =2D-disable-libgcj --enable-languages=3Dc,c++,f77 --enable-shared=20 =2D-enable-threads=3Dposix --enable-__cxa_atexit --enable-clocale=3Dgnu Thread model: posix gcc version 3.4.3-20050110 (Gentoo Linux 3.4.3.20050110-r1,=20 ssp-3.4.3.20050110-0, pie-8.7.7) Any hints? Thanks, Christian Parpart. =2D-=20 Netiquette: http://www.ietf.org/rfc/rfc1855.txt 12:57:52 up 10 days, 2:04, 0 users, load average: 0.26, 0.18, 0.07 |
|
From: Dennis L. <pla...@in...> - 2005-04-02 18:47:00
|
Am Samstag, den 02.04.2005, 13:02 +0200 schrieb Christian Parpart: > that's strange. > I just compiled valgrind-2.4.0 on a new chroot (32bit-only) environment on my > amd64. I got a segfault, so I just remerged it again. But it failed. That may > be because I first had default compiler (gcc3.3) installed and now have > gcc3.4. however, it complains about a const-ness problem: [snisnap] > > Any hints? I just tried building valgrind 2.4 from CVS with gcc 3.4.3 and 4.0CVS and it didnt complain bout anything, maybe it works with the valgrind cvs for you too... just a hint... greets Dennis |
|
From: Christian P. <tr...@ge...> - 2005-04-02 22:40:34
|
On Saturday 02 April 2005 8:45 pm, Dennis Lubert wrote: > Am Samstag, den 02.04.2005, 13:02 +0200 schrieb Christian Parpart: > > that's strange. > > I just compiled valgrind-2.4.0 on a new chroot (32bit-only) environment > > on my amd64. I got a segfault, so I just remerged it again. But it > > failed. That may be because I first had default compiler (gcc3.3) > > installed and now have gcc3.4. however, it complains about a const-ness > > problem: > > [snisnap] > > > Any hints? > > I just tried building valgrind 2.4 from CVS with gcc 3.4.3 and 4.0CVS > and it didnt complain bout anything, maybe it works with the valgrind > cvs for you too... just a hint... You don't have to believe me, but it compiles now. No clue why. I could now= even tell you where exactly gcc aborted, and how I fixed it. however, I co= mplete recompile (using emerge) of 2.4.0 on gcc3.4 worked fine now. My mach= ine's just too strange. However, the segfault problem persists (for valgrind2.4.0) but I just notic= ed, that the gentoo griffon26 (for my luck) noticed this as well, though, t= his is not my stupid box raising it alone ;) battousai libswl-0.4 # emerge -pv '=3Ddev-util/valgrind-2.4.0' These are the packages that I would merge, in order: Calculating dependencies !!! All ebuilds that could satisfy "=3Ddev-util/valgrind-2.4.0" have been m= asked. !!! One of the following masked packages is required to complete your reque= st: =2D dev-util/valgrind-2.4.0 (masked by: package.mask) # Maurice van der Pot <gri...@ge...> (02 Apr 2005) # Segfaults on everything. Masking until I get some time to # fix it, which should be in a couple of days. 2.2. works just fine in my 32bit chroot environment :D Regards, Christian Parpart. =2D-=20 Netiquette: http://www.ietf.org/rfc/rfc1855.txt 00:35:22 up 10 days, 13:41, 0 users, load average: 0.37, 0.32, 0.28 |