[PATCH x86] Update options for avx512 testsuite.
H.J. Lu
hjl.tools@gmail.com
Mon Dec 1 14:54:00 GMT 2014
On Mon, Dec 1, 2014 at 6:37 AM, H.J. Lu <hjl.tools@gmail.com> wrote:
> On Fri, Nov 28, 2014 at 7:15 AM, Uros Bizjak <ubizjak@gmail.com> wrote:
>> Hello!
>>
>>> As discussed for AVX512F submission, using -D.... in dg-options,
>>> should be replaced with parts of the testcase itself. Patch below
>>> does that.
>>> Ok for trunk?
>>>
>>> P. S.
>>> As patch itself is hard to review, this was done by sed like this:
>>>
>>> sed -i 's/ -DAVX512IFMA//g' ../gcc/testsuite/gcc.target/i386/avx512*
>>> ...
>>> sed -i 's/#define AVX512F_LEN 256/#define AVX512VL\n#define AVX512F_LEN256/g' ../gcc/testsuite/gcc.target/i386/avx512vl-*
>>> ...
>>
>> Looking into a couple of cases, it looks OK.
>>
>
> I saw
>
> FAIL: gcc.dg/vect/costmodel/i386/costmodel-fast-math-vect-pr29925.c
> scan-tree-dump-times vect "vectorized 1 loops" 1
> FAIL: gcc.dg/vect/costmodel/x86_64/costmodel-fast-math-vect-pr29925.c
> scan-tree-dump-times vect "vectorized 1 loops" 1
> FAIL: gcc.target/i386/avx256-unaligned-store-2.c scan-assembler
> vmovups.*\\\\*movv16qi_internal/3
> FAIL: gcc.target/i386/avx512ifma-vpmaddhuq-2.c (test for excess errors)
> UNRESOLVED: gcc.target/i386/avx512ifma-vpmaddhuq-2.c compilation
> failed to produce executable
> FAIL: gcc.target/i386/avx512ifma-vpmaddluq-2.c (test for excess errors)
> UNRESOLVED: gcc.target/i386/avx512ifma-vpmaddluq-2.c compilation
> failed to produce executable
> FAIL: gcc.target/i386/avx512vbmi-vpermb-2.c (test for excess errors)
> UNRESOLVED: gcc.target/i386/avx512vbmi-vpermb-2.c compilation failed
> to produce executable
> FAIL: gcc.target/i386/avx512vbmi-vpermi2b-2.c (test for excess errors)
> UNRESOLVED: gcc.target/i386/avx512vbmi-vpermi2b-2.c compilation failed
> to produce executable
> FAIL: gcc.target/i386/avx512vbmi-vpermt2b-2.c (test for excess errors)
> UNRESOLVED: gcc.target/i386/avx512vbmi-vpermt2b-2.c compilation failed
> to produce executable
> FAIL: gcc.target/i386/avx512vbmi-vpmultishiftqb-2.c (test for excess errors)
> UNRESOLVED: gcc.target/i386/avx512vbmi-vpmultishiftqb-2.c compilation
> failed to produce executable
>
> on x86:
>
> https://gcc.gnu.org/ml/gcc-testresults/2014-12/msg00030.html
>
>
I took a look at one of them:
diff --git a/gcc/testsuite/gcc.target/i386/avx512vbmi-vpmultishiftqb-2.c
b/gcc/testsuite/gcc.target/i386/avx512vbmi-vpmultishiftqb-2.c
index 936d938..861dce2 100644
--- a/gcc/testsuite/gcc.target/i386/avx512vbmi-vpmultishiftqb-2.c
+++ b/gcc/testsuite/gcc.target/i386/avx512vbmi-vpmultishiftqb-2.c
@@ -1,5 +1,5 @@
/* { dg-do run } */
-/* { dg-options "-O2 -mavx512vbmi -DAVX512VBMI" } */
+/* { dg-options "-O2 -mavx512vbmi" } */
/* { dg-require-effective-target avx512vbmi } */
#include "avx512f-helper.h"
There is no "#define AVX512VBM" added.
--
H.J.
More information about the Gcc-patches
mailing list