On Fri, Dec 17, 2004 at 11:05:57AM -0800, Brian D. McGrew wrote:
> I get the following:
>
> /usr/bin/g++ -c -fPIC -g -O0 -DALGODEBUG -D__EXTENSIONS__
> -I/usr/X11R6/include -DHAS_SYS_IOCTL_H -UHAS_SYS_FILIO_H -DHAS_STDINT_H
> -DHAS_STRING_H -DHAS_ASM_IOCTL_H -DHAS_SYS_STROPTS_H -DHAS_TIME_H
> -DHAS_SYS_WAIT_H -DUNIXHOST -DTHREAD -D_PTHREADS -D_REENTRANT
> -D_POSIX_C_SOURCE=199506L -pipe -D_G_NO_EXTERN_TEMPLATES -funroll-loops
> -finline-limit=2750 -pipe -ansi -pedantic -Wimplicit -Wconversion
> -Wsign-compare -Wno-long-long -Wreorder -Woverloaded-virtual
> -Wsign-promo -DIN_IS_LIBRARY -DDPC_DEBUG -DINTERNAL_UNDEFINED_RELEASE
> -DX86 -DLINUX -D_BSD_SOURCE -D_XOPEN_SOURCE -I../.././include
> -I../.././include/libinc -I../.././include/localinc
> barcode_alert_resources.c -o barcode_alert_resources.sh
> barcode_alert_resources.c:225: internal compiler error: Segmentation
> fault
Verified, ICE on invalid, simplified testcase:
static struct { const char *a, *b; } s[] = { { "foo",
G++ 3.2.3-RH doesn't ICE on it and neither does G++ CVS HEAD.
But, please, next time:
1) such bugreports belong to bugzilla, not on the mailing list
2) even if you for some strange reason insist on posting it,
send the file as attachment, your mailer horribly wrapped the whole file
3) and given that it is big, bzip2ing or gzipping it first is useful too
Jakub
