This is the mail archive of the
gcc-patches@gcc.gnu.org
mailing list for the GCC project.
Re: [PATCH v2, rs6000] Fix implementation of vec_pack (vector double, vector double) built-in function
- From: Segher Boessenkool <segher at kernel dot crashing dot org>
- To: Kelvin Nilsen <kdnilsen at linux dot ibm dot com>
- Cc: gcc-patches at gcc dot gnu dot org
- Date: Tue, 19 Jun 2018 17:37:21 -0500
- Subject: Re: [PATCH v2, rs6000] Fix implementation of vec_pack (vector double, vector double) built-in function
- References: <46349fd1-088a-c404-f300-19fd6fda50af@linux.ibm.com>
Hi!
On Tue, Jun 19, 2018 at 01:37:51PM -0500, Kelvin Nilsen wrote:
> --- gcc/testsuite/gcc.target/powerpc/builtins-9.c (nonexistent)
> +++ gcc/testsuite/gcc.target/powerpc/builtins-9.c (working copy)
> @@ -0,0 +1,21 @@
> +/* { dg-do compile } */
> +/* { dg-require-effective-target powerpc_p8vector_ok } */
> +/* Expect same instruction selecton on p8 and above. Fix if future
> + targets behave differently. */
> +/* { dg-options "-O3 -maltivec" } */
But this doesn't use -mcpu=power8 or similar. Does it need it anyway?
Both xxpermdi and xvcvdpsp are Power7 (ISA 2.06) and the rest is AltiVec?
So maybe just powerpc_vsx_ok?
> +/* { dg-skip-if "do not override -mcpu" { powerpc*-*-* } { "-mcpu=*" } } */
You do not use -mcpu= so you don't need this.
Same issues in the next test. Rest looks good though :-)
Segher