This is the mail archive of the
libstdc++@gcc.gnu.org
mailing list for the libstdc++ project.
Re: 3.1.1 PATCH: Fix libstdc++-v3 isprint for Solaris 2.5.1
- From: Gerald Pfeifer <pfeifer at dbai dot tuwien dot ac dot at>
- To: Rainer Orth <ro at TechFak dot Uni-Bielefeld dot DE>
- Cc: gcc-patches at gcc dot gnu dot org, <libstdc++ at gcc dot gnu dot org>
- Date: Sat, 13 Jul 2002 18:04:12 +0200 (CEST)
- Subject: Re: 3.1.1 PATCH: Fix libstdc++-v3 isprint for Solaris 2.5.1
On Sat, 13 Jul 2002, Rainer Orth wrote:
> Comparing Solaris 2.5.1/SPARC testsuite results between 3.1 and the current
> branch, I noticed a new failure:
>
> +FAIL: 22_locale/ctype_scan_char.cc execution test
>
> Assertion failed: gctype.scan_is((std::ctype_base::print), (cs), (cs) + traits_type::length(cs)) == (cs), file /vol/gnu/src/gcc/gcc-3.1-branch-dist/libstdc++-v3/testsuite/22_locale/ctype_scan_char.cc, line 226
>
> This is not a regression since the test was only introduced after the 3.1
> release, but it's a stupid bug and trivial to fix:
Just to make sure: A regression is when a current version of GCC fails to
correctly process a program that an earlier version of GCC processed fine.
"Program" here is not limited to programs that were already part of the
testsuite of that earlier version of GCC, it can also be a completely new
testcase.
Gerald