public inbox for gcc-patches@gcc.gnu.org
 help / color / mirror / Atom feed
* PATCH: Update  gcc.dg/vect/costmodel/x86_64/costmodel-vect-reduc-1char.c
@ 2007-07-22  1:40 H.J. Lu
  2007-07-24  9:29 ` Dorit Nuzman
  0 siblings, 1 reply; 5+ messages in thread
From: H.J. Lu @ 2007-07-22  1:40 UTC (permalink / raw)
  To: gcc-patches; +Cc: harsha.jagasia

This patch:

http://gcc.gnu.org/ml/gcc-patches/2007-07/msg00496.html

causes

gcc.dg/vect/costmodel/x86_64/costmodel-vect-reduc-1char.c to fail
since gcc now can vectorize 3 loops intead of 1. How should we
update gcc.dg/vect/costmodel/x86_64/costmodel-vect-reduc-1char.c?


H.J.
----
2007-07-21  H.J. Lu  <hongjiu.lu@intel.com>

	* gcc.dg/vect/costmodel/x86_64/costmodel-vect-reduc-1char.c: Updated.

--- gcc/testsuite/gcc.dg/vect/costmodel/x86_64/costmodel-vect-reduc-1char.c.x86	2007-06-08 13:49:10.000000000 -0700
+++ gcc/testsuite/gcc.dg/vect/costmodel/x86_64/costmodel-vect-reduc-1char.c	2007-07-21 16:29:53.000000000 -0700
@@ -46,6 +46,5 @@ int main (void)
   return 0;
 }
 
-/* { dg-final { scan-tree-dump-times "vectorized 1 loops" 1 "vect" { xfail vect_no_int_max } } } */
-/* { dg-final { scan-tree-dump-times "vectorization not profitable" 2 "vect" { xfail vect_no_int_max } } } */
+/* { dg-final { scan-tree-dump-times "vectorized 3 loops" 1 "vect" { xfail vect_no_int_max } } } */
 /* { dg-final { cleanup-tree-dump "vect" } } */

^ permalink raw reply	[flat|nested] 5+ messages in thread

* Re: PATCH: Update  gcc.dg/vect/costmodel/x86_64/costmodel-vect-reduc-1char.c
  2007-07-22  1:40 PATCH: Update gcc.dg/vect/costmodel/x86_64/costmodel-vect-reduc-1char.c H.J. Lu
@ 2007-07-24  9:29 ` Dorit Nuzman
  2007-07-24 18:20   ` H.J. Lu
  0 siblings, 1 reply; 5+ messages in thread
From: Dorit Nuzman @ 2007-07-24  9:29 UTC (permalink / raw)
  To: H.J. Lu; +Cc: gcc-patches, harsha.jagasia

> This patch:
>
> http://gcc.gnu.org/ml/gcc-patches/2007-07/msg00496.html
>
> causes
>
> gcc.dg/vect/costmodel/x86_64/costmodel-vect-reduc-1char.c to fail
> since gcc now can vectorize 3 loops intead of 1. How should we
> update gcc.dg/vect/costmodel/x86_64/costmodel-vect-reduc-1char.c?
>

I guess we can just apply the same change I applied to
gcc.dg/vect/costmodel/i386/costmodel-vect-reduc-1char.c (from the link
above):

Index: testsuite/gcc.dg/vect/costmodel/i386/costmodel-vect-reduc-1char.c
===================================================================
*** testsuite/gcc.dg/vect/costmodel/i386/costmodel-vect-reduc-1char.c
 (revision 126381)
--- testsuite/gcc.dg/vect/costmodel/i386/costmodel-vect-reduc-1char.c
 (working copy)
*************** int main (void)
*** 46,51 ****
    return 0;
  }

! /* { dg-final { scan-tree-dump-times "vectorized 1 loops" 1 "vect" {
xfail vect_no_int_max } } } */
! /* { dg-final { scan-tree-dump-times "vectorization not profitable" 2
"vect" { xfail vect_no_int_max } } } */
  /* { dg-final { cleanup-tree-dump "vect" } } */
--- 46,51 ----
    return 0;
  }

! /* { dg-final { scan-tree-dump-times "vectorized 3 loops" 1 "vect" {
xfail vect_no_int_max } } } */
! /* { dg-final { scan-tree-dump-times "vectorization not profitable" 0
"vect" } } */
  /* { dg-final { cleanup-tree-dump "vect" } } */

dorit

>
> H.J.
> ----
> 2007-07-21  H.J. Lu  <hongjiu.lu@intel.com>
>
>    * gcc.dg/vect/costmodel/x86_64/costmodel-vect-reduc-1char.c: Updated.
>
> --- gcc/testsuite/gcc.dg/vect/costmodel/x86_64/costmodel-vect-
> reduc-1char.c.x86   2007-06-08 13:49:10.000000000 -0700
> +++ gcc/testsuite/gcc.dg/vect/costmodel/x86_64/costmodel-vect-
> reduc-1char.c   2007-07-21 16:29:53.000000000 -0700
> @@ -46,6 +46,5 @@ int main (void)
>    return 0;
>  }
>
> -/* { dg-final { scan-tree-dump-times "vectorized 1 loops" 1 "vect"
> { xfail vect_no_int_max } } } */
> -/* { dg-final { scan-tree-dump-times "vectorization not profitable"
> 2 "vect" { xfail vect_no_int_max } } } */
> +/* { dg-final { scan-tree-dump-times "vectorized 3 loops" 1 "vect"
> { xfail vect_no_int_max } } } */
>  /* { dg-final { cleanup-tree-dump "vect" } } */

^ permalink raw reply	[flat|nested] 5+ messages in thread

* Re: PATCH: Update  gcc.dg/vect/costmodel/x86_64/costmodel-vect-reduc-1char.c
  2007-07-24  9:29 ` Dorit Nuzman
@ 2007-07-24 18:20   ` H.J. Lu
  2007-07-25 11:20     ` Dorit Nuzman
  0 siblings, 1 reply; 5+ messages in thread
From: H.J. Lu @ 2007-07-24 18:20 UTC (permalink / raw)
  To: Dorit Nuzman, janis187; +Cc: gcc-patches, harsha.jagasia

On Tue, Jul 24, 2007 at 11:17:41AM +0300, Dorit Nuzman wrote:
> > This patch:
> >
> > http://gcc.gnu.org/ml/gcc-patches/2007-07/msg00496.html
> >
> > causes
> >
> > gcc.dg/vect/costmodel/x86_64/costmodel-vect-reduc-1char.c to fail
> > since gcc now can vectorize 3 loops intead of 1. How should we
> > update gcc.dg/vect/costmodel/x86_64/costmodel-vect-reduc-1char.c?
> >
> 
> I guess we can just apply the same change I applied to
> gcc.dg/vect/costmodel/i386/costmodel-vect-reduc-1char.c (from the link
> above):
> 

This updated patch works for me. OK to install?


H.J.
----
2007-07-24  H.J. Lu  <hongjiu.lu@intel.com>

	* gcc.dg/vect/costmodel/x86_64/costmodel-vect-reduc-1char.c: Updated.

--- gcc/testsuite/gcc.dg/vect/costmodel/x86_64/costmodel-vect-reduc-1char.c.x86	2007-06-08 13:49:10.000000000 -0700
+++ gcc/testsuite/gcc.dg/vect/costmodel/x86_64/costmodel-vect-reduc-1char.c	2007-07-21 16:29:53.000000000 -0700
@@ -46,6 +46,6 @@ int main (void)
   return 0;
 }
 
-/* { dg-final { scan-tree-dump-times "vectorized 1 loops" 1 "vect" { xfail vect_no_int_max } } } */
-/* { dg-final { scan-tree-dump-times "vectorization not profitable" 2 "vect" { xfail vect_no_int_max } } } */
+/* { dg-final { scan-tree-dump-times "vectorized 3 loops" 1 "vect" { xfail vect_no_int_max } } } */
+/* { dg-final { scan-tree-dump-times "vectorization not profitable" 0 "vect" { xfail vect_no_int_max } } } */
 /* { dg-final { cleanup-tree-dump "vect" } } */

^ permalink raw reply	[flat|nested] 5+ messages in thread

* Re: PATCH: Update  gcc.dg/vect/costmodel/x86_64/costmodel-vect-reduc-1char.c
  2007-07-24 18:20   ` H.J. Lu
@ 2007-07-25 11:20     ` Dorit Nuzman
  2007-07-25 21:57       ` Janis Johnson
  0 siblings, 1 reply; 5+ messages in thread
From: Dorit Nuzman @ 2007-07-25 11:20 UTC (permalink / raw)
  To: H.J. Lu; +Cc: gcc-patches, harsha.jagasia, janis187

> On Tue, Jul 24, 2007 at 11:17:41AM +0300, Dorit Nuzman wrote:
> > > This patch:
> > >
> > > http://gcc.gnu.org/ml/gcc-patches/2007-07/msg00496.html
> > >
> > > causes
> > >
> > > gcc.dg/vect/costmodel/x86_64/costmodel-vect-reduc-1char.c to fail
> > > since gcc now can vectorize 3 loops intead of 1. How should we
> > > update gcc.dg/vect/costmodel/x86_64/costmodel-vect-reduc-1char.c?
> > >
> >
> > I guess we can just apply the same change I applied to
> > gcc.dg/vect/costmodel/i386/costmodel-vect-reduc-1char.c (from the link
> > above):
> >
>
> This updated patch works for me. OK to install?
>

I think I can approve vectorizer testcases, so, yes, ok to install.

thanks,
dorit

>
> H.J.
> ----
> 2007-07-24  H.J. Lu  <hongjiu.lu@intel.com>
>
>    * gcc.dg/vect/costmodel/x86_64/costmodel-vect-reduc-1char.c: Updated.
>
> --- gcc/testsuite/gcc.dg/vect/costmodel/x86_64/costmodel-vect-
> reduc-1char.c.x86   2007-06-08 13:49:10.000000000 -0700
> +++ gcc/testsuite/gcc.dg/vect/costmodel/x86_64/costmodel-vect-
> reduc-1char.c   2007-07-21 16:29:53.000000000 -0700
> @@ -46,6 +46,6 @@ int main (void)
>    return 0;
>  }
>
> -/* { dg-final { scan-tree-dump-times "vectorized 1 loops" 1 "vect"
> { xfail vect_no_int_max } } } */
> -/* { dg-final { scan-tree-dump-times "vectorization not profitable"
> 2 "vect" { xfail vect_no_int_max } } } */
> +/* { dg-final { scan-tree-dump-times "vectorized 3 loops" 1 "vect"
> { xfail vect_no_int_max } } } */
> +/* { dg-final { scan-tree-dump-times "vectorization not profitable"
> 0 "vect" { xfail vect_no_int_max } } } */
>  /* { dg-final { cleanup-tree-dump "vect" } } */

^ permalink raw reply	[flat|nested] 5+ messages in thread

* Re: PATCH: Update  gcc.dg/vect/costmodel/x86_64/costmodel-vect-reduc-1char.c
  2007-07-25 11:20     ` Dorit Nuzman
@ 2007-07-25 21:57       ` Janis Johnson
  0 siblings, 0 replies; 5+ messages in thread
From: Janis Johnson @ 2007-07-25 21:57 UTC (permalink / raw)
  To: Dorit Nuzman; +Cc: H.J. Lu, gcc-patches, harsha.jagasia

On Wed, 2007-07-25 at 14:17 +0300, Dorit Nuzman wrote:
> I think I can approve vectorizer testcases, so, yes, ok to install.

Yes, Dorit, you certainly can approve vectorizer tests.

Janis

^ permalink raw reply	[flat|nested] 5+ messages in thread

end of thread, other threads:[~2007-07-25 21:52 UTC | newest]

Thread overview: 5+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2007-07-22  1:40 PATCH: Update gcc.dg/vect/costmodel/x86_64/costmodel-vect-reduc-1char.c H.J. Lu
2007-07-24  9:29 ` Dorit Nuzman
2007-07-24 18:20   ` H.J. Lu
2007-07-25 11:20     ` Dorit Nuzman
2007-07-25 21:57       ` Janis Johnson

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).