Bug in cpplibs usage of hash tables

Andreas Jaeger aj@suse.de
Mon Mar 13 10:03:00 GMT 2000


>>>>> Zack Weinberg writes:

Zack> On Mon, Mar 13, 2000 at 06:32:52PM +0100, Andreas Jaeger wrote:
>> 
>> Running the testsuite on i686 with the current CVS version (which
>> includes Zack's fix for cpplib), I get now an ICE in cpp:
>> 
>> make[3]: Entering directory `/usr/src/aj/build-egcs-i686/i686-pc-linux-gnu/libio/dbz'
>> /usr/src/aj/build-egcs-i686/gcc/xgcc -B/usr/src/aj/build-egcs-i686/gcc/ -B/opt/aj-mips//i686-pc-linux-gnu/bin/ -g -O2 -I../../../../egcs/libio/dbz -I.. -I../../../../egcs/libio/dbz/.. -DDBZ_FINISH='_IO_flush_all()' -c ../../../../egcs/libio/dbz/fake.c
>> xgcc: Internal compiler error: program cpp got fatal signal 11
Zack> ...
>> This should be easy to reproduce - if not I can send preprocessed
>> sources.

Zack> Please do, it doesn't happen for me despite having a nearly identical
Zack> machine.

Upps.  I can't send preprocessed sources since the preprocessor failes:-(

What do you need?  I'll try to debug it myself now.

Andreas
-- 
 Andreas Jaeger
  SuSE Labs aj@suse.de
   private aj@arthur.rhein-neckar.de


More information about the Gcc-bugs mailing list