public inbox for gcc-patches@gcc.gnu.org
 help / color / mirror / Atom feed
From: Will Schmidt <will_schmidt@vnet.ibm.com>
To: GCC Patches <gcc-patches@gcc.gnu.org>,
	       Segher Boessenkool <segher@kernel.crashing.org>,
	       David Edelsohn <dje.gcc@gmail.com>
Cc: Bill Schmidt <wschmidt@linux.vnet.ibm.com>
Subject: [PATCH, rs6000] fold-vec-logical-ors-longlong test update
Date: Fri, 26 May 2017 17:20:00 -0000	[thread overview]
Message-ID: <1495819205.15163.174.camel@brimstone.rchland.ibm.com> (raw)

Hi, 

This test has been flaky on both AIX and in older linux based
environments. Notably, the number of xxlor instructions
generated varies depending on the platform and the specified
bit-size (32/64), with older environments generating either 6 or 24
xxlor instructions.  The behavior appears to level out on power8 targets,
so update this test to target power8-vector and specify the
-mpower8-vector option.

[gcc/testsuite]

2017-05-26  Will Schmidt  <will_schmidt@vnet.ibm.com>

	* gcc.target/powerpc/fold-vec-logical-ors-longlong.c:
	Update the target to powerpc_p8vector_ok. Update dg-options
	value to -mpower8-vector.


diff --git a/gcc/testsuite/gcc.target/powerpc/fold-vec-logical-ors-longlong.c b/gcc/testsuite/gcc.target/powerpc/fold-vec-logical-ors-longlong.c
index 7ca23fb..10c69d3 100644
--- a/gcc/testsuite/gcc.target/powerpc/fold-vec-logical-ors-longlong.c
+++ b/gcc/testsuite/gcc.target/powerpc/fold-vec-logical-ors-longlong.c
@@ -2,8 +2,8 @@
  * long long inputs produce the right results.  */
 
 /* { dg-do compile } */
-/* { dg-require-effective-target powerpc_vsx_ok } */
-/* { dg-options "-mvsx -O2" } */
+/* { dg-require-effective-target powerpc_p8vector_ok } */
+/* { dg-options "-mpower8-vector -O2" } */
 
 #include <altivec.h>
 
@@ -151,11 +151,11 @@ test6_nor (vector unsigned long long x, vector unsigned long long y)
   return *foo;
 }
 
-// Codegen on power7 is such that the vec_or() tests generate more xxlor
-// instructions than what is seen on power8 or newer.
-// Thus, an additional target clause for the xxlor instruction check.
-/* { dg-final { scan-assembler-times {\mxxlor\M} 6 { target p8vector_hw }  } } */
-/* { dg-final { scan-assembler-times {\mxxlor\M} 24 { target { ! p8vector_hw }  }  } } */
+// The number of xxlor instructions generated varies between 6 and 24 for
+// older systems (power6,power7), as well as for 32-bit versus 64-bit targets.
+// For simplicity, this test now only targets "powerpc_p8vector_ok" environments
+// where the answer is expected to be 6.
 
+/* { dg-final { scan-assembler-times {\mxxlor\M} 6 } } */
 /* { dg-final { scan-assembler-times {\mxxlxor\M} 6 } } */
 /* { dg-final { scan-assembler-times {\mxxlnor\M} 6 } } */


             reply	other threads:[~2017-05-26 17:20 UTC|newest]

Thread overview: 2+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2017-05-26 17:20 Will Schmidt [this message]
2017-05-31 18:27 ` Segher Boessenkool

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=1495819205.15163.174.camel@brimstone.rchland.ibm.com \
    --to=will_schmidt@vnet.ibm.com \
    --cc=dje.gcc@gmail.com \
    --cc=gcc-patches@gcc.gnu.org \
    --cc=segher@kernel.crashing.org \
    --cc=wschmidt@linux.vnet.ibm.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
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).