[rs6000] fold-vec testcase fix-ups

Submitted by Will Schmidt on Aug. 10, 2017, 1:27 p.m.

Details

Message ID 1502371637.6577.48.camel@brimstone.rchland.ibm.com
State New
Headers show

Commit Message

Will Schmidt Aug. 10, 2017, 1:27 p.m.
Hi, 
A testcase coverage issue and an obvious typo fix.

Mostly obvious,..  OK for trunk? 
Thanks,
-Will

[gcc/testsuite]

2017-08-10  Will Schmidt  <will_schmidt@vnet.ibm.com>

	* gcc.target/powerpc/fold-vec-msum-short.c: Fix typo.
	* gcc.target/powerpc/fold-vec/pack-longlong.c: Mark for 64-bit only.

Comments

Segher Boessenkool Aug. 10, 2017, 1:46 p.m.
On Thu, Aug 10, 2017 at 08:27:17AM -0500, Will Schmidt wrote:
> A testcase coverage issue and an obvious typo fix.
> 
> Mostly obvious,..  OK for trunk? 

Yes; one comment:

> diff --git a/gcc/testsuite/gcc.target/powerpc/fold-vec-pack-longlong.c b/gcc/testsuite/gcc.target/powerpc/fold-vec-pack-longlong.c
> index d8acc3c..73131bb 100644
> --- a/gcc/testsuite/gcc.target/powerpc/fold-vec-pack-longlong.c
> +++ b/gcc/testsuite/gcc.target/powerpc/fold-vec-pack-longlong.c
> @@ -1,9 +1,9 @@
>  /* Verify that overloaded built-ins for vec_pack with long long
>     inputs produce the right results.  */
>  
> -/* { dg-do compile } */
> +/* { dg-do compile { target { powerpc*-*-* && lp64 } } } */
>  /* { dg-require-effective-target powerpc_p8vector_ok } */
>  /* { dg-options "-mvsx -mpower8-vector -O2" } */

You can just say

/* { dg-do compile { target lp64 } } */

(make sure to test before you commit :-) )


Segher

Patch hide | download patch | download mbox

diff --git a/gcc/testsuite/gcc.target/powerpc/fold-vec-msum-short.c b/gcc/testsuite/gcc.target/powerpc/fold-vec-msum-short.c
index 61e1d35..2e590ab 100644
--- a/gcc/testsuite/gcc.target/powerpc/fold-vec-msum-short.c
+++ b/gcc/testsuite/gcc.target/powerpc/fold-vec-msum-short.c
@@ -13,11 +13,11 @@  test_msum_si (vector signed short vss2, vector signed short vss3,
 {
   return vec_msum (vss2, vss3, vsi2);
 }
 
 vector unsigned int
-test_msum)ui (vector unsigned short vus2, vector unsigned short vus3,
+test_msum_ui (vector unsigned short vus2, vector unsigned short vus3,
 	   vector unsigned int vui2)
 {
   return vec_msum (vus2, vus3, vui2);
 }
 
diff --git a/gcc/testsuite/gcc.target/powerpc/fold-vec-pack-longlong.c b/gcc/testsuite/gcc.target/powerpc/fold-vec-pack-longlong.c
index d8acc3c..73131bb 100644
--- a/gcc/testsuite/gcc.target/powerpc/fold-vec-pack-longlong.c
+++ b/gcc/testsuite/gcc.target/powerpc/fold-vec-pack-longlong.c
@@ -1,9 +1,9 @@ 
 /* Verify that overloaded built-ins for vec_pack with long long
    inputs produce the right results.  */
 
-/* { dg-do compile } */
+/* { dg-do compile { target { powerpc*-*-* && lp64 } } } */
 /* { dg-require-effective-target powerpc_p8vector_ok } */
 /* { dg-options "-mvsx -mpower8-vector -O2" } */
 
 #include <altivec.h>