This is the mail archive of the
libc-alpha@sourceware.org
mailing list for the glibc project.
Re: XFAIL libm-test.inc tests as needed for ibm128 [committed]
- From: "Tulio Magno Quites Machado Filho" <tuliom at linux dot vnet dot ibm dot com>
- To: Joseph Myers <joseph at codesourcery dot com>, libc-alpha at sourceware dot org
- Cc:
- Date: Wed, 11 Jan 2017 13:59:29 -0200
- Subject: Re: XFAIL libm-test.inc tests as needed for ibm128 [committed]
- Authentication-results: sourceware.org; auth=none
- References: <alpine.DEB.2.20.1701100014330.8116@digraph.polyomino.org.uk>
Joseph Myers <joseph@codesourcery.com> writes:
> This patch arranges for various libm-test.inc tests to be XFAILed for
> ibm128-libgcc in non-default rounding modes. The tests are marked
> with XFAIL_ROUNDING_IBM128_LIBGCC and gen-libm-test.pl is made to
> transform that to XFAIL_IBM128_LIBGCC or 0 depending on the rounding
> mode.
>
> This should allow test-ldouble, test-ildouble and test-ldouble-finite
> to pass with unmodified libgcc, given an ulps regeneration.
Just for completeness: building without --with-cpu.
I'm still seeing errors on powerpc using --with-cpu=power8, powerpc64
and powerpc64le.
> (The case of patched libgcc was already clean up to ulps
Which patch are you referring to?
Thanks!
--
Tulio Magno