This is the mail archive of the
mailing list for the GCC project.
Re: PATCH: x86/x86-64: Support Intel SSSE3/MNI intrinsics
- From: "H. J. Lu" <hjl at lucon dot org>
- To: Jan Hubicka <hubicka at ucw dot cz>
- Cc: Mike Stump <mrs at apple dot com>, Mark Mitchell <mark at codesourcery dot com>, Andrew Pinski <pinskia at physics dot uc dot edu>, Ian Lance Taylor <iant at google dot com>, gcc-patches at gcc dot gnu dot org
- Date: Sat, 21 Oct 2006 08:51:15 -0700
- Subject: Re: PATCH: x86/x86-64: Support Intel SSSE3/MNI intrinsics
- References: <20060705183055.GA9428@lucon.org> <20060710223553.GA10395@lucon.org> <firstname.lastname@example.org> <44B6FFDF.email@example.com> <20060928134749.GA19128@lucon.org> <firstname.lastname@example.org> <20060928164944.GA20298@lucon.org> <451C2FB0.email@example.com> <8E373DBA-1E2A-4A6A-90B5-244B2D008A29@apple.com> <20061009222218.GB5896@atrey.karlin.mff.cuni.cz>
On Tue, Oct 10, 2006 at 12:22:18AM +0200, Jan Hubicka wrote:
> > On Sep 28, 2006, at 1:25 PM, Mark Mitchell wrote:
> > >I have no comments on the substance of the patch itself. If the
> > >IA32 maintainers want to put this in for GCC 4.2, that's OK with me;
> > >if they want to hold off until 4.3, that's fine too.
> > I'd like it for 4.2, I already have it in my 4.0.
> I am not x86 maintainer but as x86-64 maintainer I would vote for
> including it ;) The patch is almost obviously harmless for non SSSE
> users and users tends to be locked to one GCC version for a while.
> However it would be nice to have some testsuite for the intrincisc. In
> the past there was many bugs in the original SSE/SSE(2) implementation
> and we never gained any sensible public testsuite, sadly. We however
> should not keep this habit and I would say we sould adopt the need to
> have at least basic testcases for every added target builtin.
I am checking SSSE3 intrinsic into gcc 4.3. I will provide run-time
SSE3 and SSSE3 intrinsic tests later.