This is the mail archive of the
mailing list for the glibc project.
Re: [PATCH 1/2] MIPS: Add wrappers to get/setrlimit64 to fix RLIM64_INFINITY constant
- From: Aurelien Jarno <aurelien at aurel32 dot net>
- To: "Joseph S. Myers" <joseph at codesourcery dot com>
- Cc: libc-alpha at sourceware dot org
- Date: Tue, 26 Nov 2013 21:37:45 +0100
- Subject: Re: [PATCH 1/2] MIPS: Add wrappers to get/setrlimit64 to fix RLIM64_INFINITY constant
- Authentication-results: sourceware.org; auth=none
- References: <1384921106-1144-1-git-send-email-aurelien at aurel32 dot net> <1384921106-1144-2-git-send-email-aurelien at aurel32 dot net> <Pine dot LNX dot 4 dot 64 dot 1311201554490 dot 28804 at digraph dot polyomino dot org dot uk> <20131123151646 dot GA1840 at ohm dot rr44 dot fr> <Pine dot LNX dot 4 dot 64 dot 1311251805010 dot 12149 at digraph dot polyomino dot org dot uk>
On Mon, Nov 25, 2013 at 06:06:45PM +0000, Joseph S. Myers wrote:
> On Sat, 23 Nov 2013, Aurelien Jarno wrote:
> > On Wed, Nov 20, 2013 at 03:55:09PM +0000, Joseph S. Myers wrote:
> > > Has this patch been tested on its own with the glibc testsuite for both
> > > affected ABIs, including passing the ABI tests?
> > I have run the glibc testsuite, but not the ABI tests.
> The ABI tests have been part of a default "make check" run since:
> commit ecf0ebfb04f6e31e3ba709521b6955d5f86ff724
> Author: Andreas Jaeger <email@example.com>
> Date: Sun Apr 29 06:31:26 2012 +0200
> Always run check-abi
> Remove the configure options for running of check-abi and always
> run it as part of the testsuite
> Please make sure you aren't testing with any local patches that disable
> this or other parts of the testsuite.
Indeed I confirm check-abi is run as part of the testsuite and that it
in fact detects the issue.
My problem is that given the testsuite is known to fail for some tests,
I ran "make -k check" and redirected the output to a file. I then used
convertlog.sh from the Debian package  to extract the failures and to
compare them to the build without the patch. It happened that the regular
expression from the script doesn't catch the check-abi failure.
Aurelien Jarno GPG: 1024D/F1BCDB73