This is the mail archive of the
mailing list for the GCC project.
Re: [Patch FR30] Use existing predicates for RTL objects
- From: Dave Korn <dave dot korn dot cygwin at googlemail dot com>
- To: Shujing Zhao <pearly dot zhao at oracle dot com>
- Cc: gcc-patches at gcc dot gnu dot org, Nick Clifton <nickc at redhat dot com>, Paolo Carlini <paolo dot carlini at oracle dot com>
- Date: Tue, 02 Jun 2009 12:18:59 +0100
- Subject: Re: [Patch FR30] Use existing predicates for RTL objects
- References: <4A24F262.email@example.com>
Shujing Zhao wrote:
> There are many unexpected failures when test gcc on fr30-elf target.
> Please to see the following gcc summary. Is it normal?
Well, it is not uncommon for some of the less-common targets to accumulate
some bitrot in the testsuite. But 8896 FAILs seems like rather a lot. It's
possible that there's some systematic problem between the testsuite and the
simulator; you'd have to rummage through the logs and see if there's a common
factor to the failures.
> === gcc Summary ===
> # of expected passes 40392
> # of unexpected failures 8896
> # of unexpected successes 4
> # of expected failures 71
> # of unresolved testcases 104
> # of unsupported tests 613
> /home/pzhao/cross-compile/fr30-build/gcc/xgcc version 4.5.0 20090601
> (experimental) [trunk revision 148068] (GCC)
You need to run the tests both before and after applying your patch to the
sources, and check if any differences show up.