On Thu, Jan 29, 2015 at 6:56 AM, Jakub Jelinek <ja...@redhat.com> wrote:
> On Thu, Jan 29, 2015 at 03:54:34PM +0100, Uros Bizjak wrote:
>> On Thu, Jan 29, 2015 at 3:47 PM, Jakub Jelinek <ja...@redhat.com> wrote:
>>
>> >> > So here's the updated patch which handles all 4 testcases from the PR
>> >> > as well as a couple of my own.
>> >>
>> >> @@ -0,0 +1,33 @@
>> >> +/* PR 15184 first two tests, plus two addition ones. */
>> >> +/* { dg-do compile } */
>> >> +/* { dg-options "-O2 -m32 -march=pentiumpro" } */
>> >>
>> >> No, we don't want -m32 in dg-options. Please write this part as:
>> >>
>> >> /* { dg-do compile { target ia32 } } */
>> >> /* { dg-options "-O2 -march=pentiumpro" } */
>> >
>> > Or
>> > /* { dg-do compile } */
>> > /* { dg-options "-O2" } */
>> > /* { dg-additional-options "-march=pentiumpro" { target ia32 } } */
>> > if there is nothing ia32 specific on the testcase, just that you want
>> > to use that -march option if you are compiling it as 32-bit.
>>
>> In this case, you will need to add -mregparm=3 to
>> dg-additional-options to avoid "attribute ... ignored" awarning. There
>> are some examples in dg.target/i386/ testsuite directory.
>
> Ah, then better limit that to just ia32 in dg-do.
>
Yea, please. I have seen:
New failures:
FAIL: gcc.target/i386/pr15184-1.c (test for excess errors)
FAIL: gcc.target/i386/pr15184-2.c (test for excess errors)
for -mx32.
--
H.J.
