This is the mail archive of the
libc-alpha@sourceware.org
mailing list for the glibc project.
Re: Don't include expected results in libm-test test names
- From: Andreas Jaeger <aj at suse dot com>
- To: libc-alpha at sourceware dot org
- Date: Wed, 22 May 2013 08:33:05 +0200
- Subject: Re: Don't include expected results in libm-test test names
- References: <Pine dot LNX dot 4 dot 64 dot 1305212049130 dot 7003 at digraph dot polyomino dot org dot uk>
On 05/21/2013 10:53 PM, Joseph S. Myers wrote:
This patch stops libm test names, as used in libm-test-ulps files,
from including the expected result of the test. This is with a view
to most expected results being generated automatically by MPFR/MPC in
future separately for each floating-point format: then the text of the
expected result for a given input will differ slightly between
floating-point formats, but if the input is exactly the same it's
still good for them to be considered the same test for libm-test-ulps
files.
Tested x86_64 and x86. I updated all the libm-test-ulps files using
sed (and running gen-libm-test.pl -n on them in case any consequent
reordering was needed); full diff attached gzipped, since it's nearly
4MB uncompressed. Note that for the actual move to MPFR-based
generation of expected values (probably not for 2.18), it will be
necessary to regenerate all the libm-test-ulps files from scratch, as
all test inputs for MPFR-based tests will I expect end up in hex float
format.
Ok, thanks,
Andreas
--
Andreas Jaeger aj@{suse.com,opensuse.org} Twitter/Identica: jaegerandi
SUSE LINUX Products GmbH, Maxfeldstr. 5, 90409 Nürnberg, Germany
GF: Jeff Hawn,Jennifer Guild,Felix Imendörffer,HRB16746 (AG Nürnberg)
GPG fingerprint = 93A3 365E CE47 B889 DF7F FED1 389A 563C C272 A126