https://gcc.gnu.org/bugzilla/show_bug.cgi?id=82120
--- Comment #3 from Thomas Preud'homme <thopre01 at gcc dot gnu.org> ---
(In reply to Jakub Jelinek from comment #2)
> This is a total mess. I've copied a boilerplate from other tests that
> require branch cost of 2.
> I guess
> 2017-09-06 Jakub Jelinek <ja...@redhat.com>
>
> PR testsuite/82120
> * gcc.dg/tree-ssa/pr81588.c: Don't run on logical_op_short_circuit
> targets.
>
> --- gcc/testsuite/gcc.dg/tree-ssa/pr81588.c.jj 2017-08-02
> 12:10:36.000000000
> +0200
> +++ gcc/testsuite/gcc.dg/tree-ssa/pr81588.c 2017-09-06 16:09:57.373505808
> +0200
> @@ -1,5 +1,5 @@
> /* PR tree-optimization/81588 */
> -/* { dg-do compile { target { ! "m68k*-*-* mmix*-*-* bfin*-*-* v850*-*-*
> moxie*-*-* cris*-*-* m32c*-*-* fr30*-*-* mcore*-*-* powerpc*-*-* xtensa*-*-*
> hppa*-*-* nios2*-*-*" } } } */
> +/* { dg-do compile { target { ! { logical_op_short_circuit || { m68k*-*-*
> bfin*-*-* v850*-*-* moxie*-*-* m32c*-*-* fr30*-*-* mcore*-*-* xtensa*-*-*
> hppa*-*-* } } } } } */
> /* { dg-options "-O2 -fdump-tree-reassoc1-details" } */
> /* { dg-additional-options "-mbranch-cost=2" { target mips*-*-* avr-*-*
> s390*-*-* i?86-*-* x86_64-*-* } } */
>
> could fix that, but that will mean it will be untested even on mips, avr and
> s390 when it could be tested there.
>
> So perhaps better:
> 2017-09-06 Jakub Jelinek <ja...@redhat.com>
>
> PR testsuite/82120
> * gcc.dg/tree-ssa/pr81588.c: Don't run on logical_op_short_circuit
> targets
> except for those where -mbranch-cost=2 is supported.
>
> --- gcc/testsuite/gcc.dg/tree-ssa/pr81588.c.jj 2017-08-02
> 12:10:36.000000000
> +0200
> +++ gcc/testsuite/gcc.dg/tree-ssa/pr81588.c 2017-09-06 16:17:30.563118589
> +0200
> @@ -1,5 +1,5 @@
> /* PR tree-optimization/81588 */
> -/* { dg-do compile { target { ! "m68k*-*-* mmix*-*-* bfin*-*-* v850*-*-*
> moxie*-*-* cris*-*-* m32c*-*-* fr30*-*-* mcore*-*-* powerpc*-*-* xtensa*-*-*
> hppa*-*-* nios2*-*-*" } } } */
> +/* { dg-do compile { target { ! { { logical_op_short_circuit && { !
> "mips*-*-* avr-*-* s390*-*-*" } } || { m68k*-*-* bfin*-*-* v850*-*-*
> moxie*-*-* m32c*-*-* fr30*-*-* mcore*-*-* xtensa*-*-* hppa*-*-* } } } } } */
> /* { dg-options "-O2 -fdump-tree-reassoc1-details" } */
> /* { dg-additional-options "-mbranch-cost=2" { target mips*-*-* avr-*-*
> s390*-*-* i?86-*-* x86_64-*-* } } */
>
>
> Can you test it on the various arm configurations (just make check-gcc
> RUNTESTFLAGS=tree-ssa.exp=pr81588.c)?
> Wonder about the targets that aren't included in logical_op_short_circuit,
> what they actually do and why they are in the lists in the other testcases.
I've tested it for a variety of Arm Cortex-M and Cortex-A cores and it either
PASS or is marked as UNSUPPORTED so looks good to me, thanks.
Best regards.
