Re: [PATCH v2] xfail fetestexcept test - ppc always uses fcmpu

[Alexandre Oliva <oliva@adacore.com>]

On Apr 23, 2024, "Kewen.Lin" <linkw@linux.ibm.com> wrote:

>> --- a/gcc/testsuite/gcc.dg/torture/pr91323.c
>> +++ b/gcc/testsuite/gcc.dg/torture/pr91323.c
>> @@ -1,4 +1,5 @@
>> -/* { dg-do run } */
>> +/* { dg-do run { xfail powerpc*-*-* } } */
>> +/* The ppc xfail is because of PR target/58684.  */

> OK, though the proposed comment is slightly different from what's in
> the related commit r8-6445-g86145a19abf39f. :)  Thanks!

Oh, thanks for the pointer, that was easy to fix.  Here's what I'm
pushing momentarily...


xfail fetestexcept test - ppc always uses fcmpu

gcc.dg/torture/pr91323.c tests that a compare with NaNf doesn't set an
exception using builtin compare intrinsics, and that it does when
using regular compare operators.

That doesn't seem to be expected to work on powerpc targets.  It fails
on GNU/Linux, it's marked to be skipped on AIX, and a similar test,
gcc.dg/torture/pr93133.c, has the execution test xfailed for all of
powerpc*-*-*.

In this test, the functions that use intrinsics for the compare end up
with the same code as the one that uses compare operators, using
fcmpu, a floating compare that, unlike fcmpo, does not set the invalid
operand exception for quiet NaN.  I couldn't find any evidence that
the rs6000 backend ever outputs fcmpo.  Therefore, I'm adding the same
execution xfail marker to this test.


for  gcc/testsuite/ChangeLog

	PR target/58684
	* gcc.dg/torture/pr91323.c: Expect execution fail on
	powerpc*-*-*.
---
 gcc/testsuite/gcc.dg/torture/pr91323.c |    3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/gcc/testsuite/gcc.dg/torture/pr91323.c b/gcc/testsuite/gcc.dg/torture/pr91323.c
index 1411fcaa3966c..4574342e728db 100644
--- a/gcc/testsuite/gcc.dg/torture/pr91323.c
+++ b/gcc/testsuite/gcc.dg/torture/pr91323.c
@@ -1,4 +1,5 @@
-/* { dg-do run } */
+/* { dg-do run { xfail powerpc*-*-* } } */
+/* remove the xfail for powerpc when pr58684 is fixed */
 /* { dg-add-options ieee } */
 /* { dg-require-effective-target fenv_exceptions } */
 /* { dg-skip-if "fenv" { powerpc-ibm-aix* } } */


-- 
Alexandre Oliva, happy hacker            https://FSFLA.org/blogs/lxo/
   Free Software Activist                   GNU Toolchain Engineer
More tolerance and less prejudice are key for inclusion and diversity
Excluding neuro-others for not behaving ""normal"" is *not* inclusive