Hi
Well, I 've just tried to compile the new glibc-2.3.5 with the latest gcc-3.4-20050422 and ran into the following internal compiler error during the glibc math checks:
make[2]: Entering directory `/share/home/clemens/newbuild/glibc-2.3.5/math'
gcc test-idouble.c -c -std=gnu99 -O2 -Wall -Winline -Wstrict-prototypes -Wwrite-strings -g -msoft-float -mnew-mnemonics -Wno-uninitialized -D__NO_MATH_INLINES -D__LIBC_INTERNAL_MATH_INLINES -DNO_LONG_DOUBLE -D_Mlong_double_=double -I../include -I. -I/home/clemens/newbuild/glibc-2.3.5-build/math -I.. -I../libio -I../nptl -I/home/clemens/newbuild/glibc-2.3.5-build -I../sysdeps/powerpc/powerpc32/elf -I../sysdeps/powerpc/elf -I../libidn/sysdeps/unix -I../nptl/sysdeps/unix/sysv/linux/powerpc/powerpc32 -I../nptl/sysdeps/unix/sysv/linux/powerpc -I../nptl/sysdeps/unix/sysv/linux -I../nptl/sysdeps/pthread -I../sysdeps/pthread -I../nptl/sysdeps/unix/sysv -I../nptl/sysdeps/unix -I../nptl/sysdeps/powerpc -I../sysdeps/unix/sysv/linux/powerpc/powerpc32 -I../sysdeps/unix/sysv/linux/powerpc -I../sysdeps/unix/sysv/linux -I../sysdeps/gnu -I../sysdeps/unix/common -I../sysdeps/unix/mman -I../sysdeps/unix/inet -I../sysdeps/unix/sysv -I../sysdeps/unix/powerpc -I../sysdeps/unix -I../sysdeps/po
six -I../sysdeps/powerpc/powerpc32 -I../sysdeps/wordsize-32 -I../sysdeps/powerpc/soft-fp -I../sysdeps/powerpc/nofpu -I../sysdeps/powerpc -I../sysdeps/ieee754/flt-32 -I../sysdeps/ieee754/dbl-64 -I../sysdeps/ieee754 -I../sysdeps/generic/elf -I../sysdeps/generic -I /lib/modules/2.6.11.6/build/include -D_LIBC_REENTRANT -D_LIBC_REENTRANT -include ../include/libc-symbols.h -DNOT_IN_libc=1 -U__LIBC_INTERNAL_MATH_INLINES -D__FAST_MATH__ -DTEST_FAST_MATH -fno-builtin -I../soft-fp/ -o /home/clemens/newbuild/glibc-2.3.5-build/math/test-idouble.o -MD -MP -MF /home/clemens/newbuild/glibc-2.3.5-build/math/test-idouble.o.dt -MT /home/clemens/newbuild/glibc-2.3.5-build/math/test-idouble.o
/home/clemens/newbuild/glibc-2.3.5-build/math/libm-test.c: In function `ctanh_test':
/home/clemens/newbuild/glibc-2.3.5-build/math/libm-test.c:2333: internal compiler error: Segmentation fault
Please submit a full bug report,
with preprocessed source if appropriate.
See <URL:http://gcc.gnu.org/bugs.html> for instructions.
make[2]: *** [/home/clemens/newbuild/glibc-2.3.5-build/math/test-idouble.o] Error 1
make[2]: Leaving directory `/share/home/clemens/newbuild/glibc-2.3.5/math'
make[1]: *** [math/tests] Error 2
make[1]: Leaving directory `/share/home/clemens/newbuild/glibc-2.3.5'
make: *** [check] Error 2
gcc was built with: Reading specs from /usr/local/lib/gcc/powerpc-unknown-linux-gnu/3.4.4/specs Configured with: ../gcc-3.4-20050422/configure --with-float=soft --enable-shared --enable-threads=posix --enable-__cxa_atexit --enable-languages=c,c++,objc --enable-nls=yes --enable-clocale=gnu Thread model: posix gcc version 3.4.4 20050422 (prerelease)
all that is on a: $ cat /proc/cpuinfo processor : 0 cpu : e500 revision : 2.0 (pvr 8020 0020) bogomips : 823.29 Vendor : MicroSys Electronics GmbH Machine : PM854 clock : 825MHz PVR : 0x80200020 SVR : 0x80300020 PLL setting : 0x5 Memory : 256 MB
which is a PowerPC e500 MPC8540 which has no FPU system running: $ uname -a Linux ecam.anagramm.de 2.6.11.6 #1 Mon Apr 4 15:31:11 CEST 2005 ppc ppc ppc GNU/Linux $ ld --version GNU ld version 2.16.90.0.1 20050408 ...
I know that there are precision issues with the test-float and test-ifloat tests with the FPU emulation code, but this crash is new to me. I haven't seen this problem with gcc-3.4.3 and glibc-2.3.4. Well... is this already a known problem, a mistake I made, or do you need more debugging info?
Best greets,
Clemens Koller _______________________________ R&D Imaging Devices Anagramm GmbH Rupert-Mayer-Str. 45/1 81379 Muenchen Germany
http://www.anagramm.de Phone: +49-89-741518-50 Fax: +49-89-741518-19
