Patchwork 0005-Switch-Core-2-to-new-tuning

login
register
mail settings
Submitter Jack Howarth
Date Dec. 3, 2010, 4:54 p.m.
Message ID <20101203165451.GA584@bromo.med.uc.edu>
Download mbox | patch
Permalink /patch/74169/
State New
Headers show

Comments

Jack Howarth - Dec. 3, 2010, 4:54 p.m.
On Fri, Dec 03, 2010 at 08:41:53AM -0800, H.J. Lu wrote:
> On Fri, Dec 3, 2010 at 8:32 AM, Jack Howarth <howarth@bromo.med.uc.edu> wrote:
> > On Thu, Dec 02, 2010 at 11:50:54PM +0100, Uros Bizjak wrote:
> >> On Thu, Dec 2, 2010 at 8:34 PM, Maxim Kuvyrkov <maxim@codesourcery.com> wrote:
> >> > On Dec 2, 2010, at 8:46 PM, Jack Howarth wrote:
> >> >> Maxim,
> >> >>   Any comments on the md file changes proposed by Iain Sandoe in his i386 darwin mtune=core2 patch?
> >> >>
> >> >> http://gcc.gnu.org/ml/gcc-patches/2010-08/msg01024.html
> >> >
> >> > Jack,
> >> >
> >> > I don't have any comments on the patch.  Uros or Richard should be much better judges on i386.md than I am.
> >>
> >> The patch is already approved by rth. Are there any open questions
> >> w.r.t. the patch?
> >>
> >> Uros.
> >
> > Uros,
> >   It appears that this previously approved i386 darwin mtune=core2 patch will require
> > the additional testsuite fix of...
> >
> > Index: gcc.target/i386/sse2-init-v2di-2.c
> > ===================================================================
> > --- gcc.target/i386/sse2-init-v2di-2.c  (revision 167430)
> > +++ gcc.target/i386/sse2-init-v2di-2.c  (working copy)
> > @@ -10,4 +10,4 @@
> >   return _mm_cvtsi64_si128 (b);
> >  }
> >
> > -/* { dg-final { scan-assembler "movq" } } */
> > +/* { dg-final { scan-assembler "movd" } } */
> >
> > Does that seem reasonable?
> 
> No.  You need to scan 64bit register when movd is used.

H.J.,
  Do you mean in the md file changes or the scan-assembler of the testcase itself?
I only see...


change in the assembly of the gcc.target/i386/sse2-init-v2di-2.c testcase
when the proposed darwin patch is applied.
         Jack

> 
> 
> -- 
> H.J.
H.J. Lu - Dec. 3, 2010, 5:10 p.m.
On Fri, Dec 3, 2010 at 8:54 AM, Jack Howarth <howarth@bromo.med.uc.edu> wrote:
> On Fri, Dec 03, 2010 at 08:41:53AM -0800, H.J. Lu wrote:
>> On Fri, Dec 3, 2010 at 8:32 AM, Jack Howarth <howarth@bromo.med.uc.edu> wrote:
>> > On Thu, Dec 02, 2010 at 11:50:54PM +0100, Uros Bizjak wrote:
>> >> On Thu, Dec 2, 2010 at 8:34 PM, Maxim Kuvyrkov <maxim@codesourcery.com> wrote:
>> >> > On Dec 2, 2010, at 8:46 PM, Jack Howarth wrote:
>> >> >> Maxim,
>> >> >>   Any comments on the md file changes proposed by Iain Sandoe in his i386 darwin mtune=core2 patch?
>> >> >>
>> >> >> http://gcc.gnu.org/ml/gcc-patches/2010-08/msg01024.html
>> >> >
>> >> > Jack,
>> >> >
>> >> > I don't have any comments on the patch.  Uros or Richard should be much better judges on i386.md than I am.
>> >>
>> >> The patch is already approved by rth. Are there any open questions
>> >> w.r.t. the patch?
>> >>
>> >> Uros.
>> >
>> > Uros,
>> >   It appears that this previously approved i386 darwin mtune=core2 patch will require
>> > the additional testsuite fix of...
>> >
>> > Index: gcc.target/i386/sse2-init-v2di-2.c
>> > ===================================================================
>> > --- gcc.target/i386/sse2-init-v2di-2.c  (revision 167430)
>> > +++ gcc.target/i386/sse2-init-v2di-2.c  (working copy)
>> > @@ -10,4 +10,4 @@
>> >   return _mm_cvtsi64_si128 (b);
>> >  }
>> >
>> > -/* { dg-final { scan-assembler "movq" } } */
>> > +/* { dg-final { scan-assembler "movd" } } */
>> >
>> > Does that seem reasonable?
>>
>> No.  You need to scan 64bit register when movd is used.
>
> H.J.,
>  Do you mean in the md file changes or the scan-assembler of the testcase itself?
> I only see...
>

I will update the scan-assembler of the testcase when it fails on Linux.

Patch

--- sse2-init-v2di-2.s.r167430	2010-12-03 10:51:29.000000000 -0500
+++ sse2-init-v2di-2.s.r167430_darwincore2	2010-12-03 11:19:49.000000000 -0500
@@ -3,7 +3,7 @@ 
 	.globl _test
 _test:
 LFB512:
-	movq	%rdi, %xmm0
+	movd	%rdi, %xmm0
 	ret
 LFE512:
 	.section __TEXT,__eh_frame,coalesced,no_toc+strip_static_syms+live_support