This is the mail archive of the
libstdc++@gcc.gnu.org
mailing list for the libstdc++ project.
Re: 27_io/istream_extractor_arith.cc fixup
- From: Paolo Carlini <pcarlini at unitus dot it>
- To: Andreas Schwab <schwab at suse dot de>
- Cc: libstdc++ at gcc dot gnu dot org
- Date: Wed, 30 Jan 2002 10:39:08 +0100
- Subject: Re: 27_io/istream_extractor_arith.cc fixup
- References: <200201291947.g0TJlR302024@fillmore.constant.com> <3C5733E5.5A2A6BCD@unitus.it> <jeit9ki3p3.fsf@sykes.suse.de>
Andreas Schwab wrote:
>Paolo Carlini <pcarlini@unitus.it> writes:
>
>|> What *really* puzzles me is this:
>|>
>|> http://gcc.gnu.org/ml/gcc-testresults/2002-01/msg00582.html
>|>
>|> How possibly, ostream_inserter_arith.cc can fail now on ia64? It did'nt fail
>|> before enabling test03, then Andreas tested positively the patch above, and then
>|> it fails now?!?
>
>There were some conflicts with the change I used locally, due to some
>stupid mailer that has mangled whitespace.
>
I see.
Therefore, if you can confirm that indeed the testcase is failing now
with a freshly updated source from CVS, could you please post some
details? It would be important to see the relevant section of
libstdc++-v3.log in order to debug it.
Perhaps the improvement suggested by Richard Henderson is *necessary* in
order to fix the test on ia64 but we cannot be sure right now.
Thanks for your help,
Paolo.