On Wed, Feb 11, 2015 at 10:15:42AM +0000, Marcus Shawcroft wrote:
> On 10 February 2015 at 16:06, James Greenhalgh <james.greenha...@arm.com>
> wrote:
> >
> > Hi,
> >
> > As is already done for mips and hppa, we should XFAIL this test on
> > AArch64 as we don't currently use the store_by_pieces infrastructure.
> >
> > We may in future want to tweak this, but for GCC 5.0 the safe thing
> > to do is just to XFAIL the test.
> >
> > Tested with aarch64-none-elf.
> >
> > OK?
>
> Ok, with Mike's comment addressed... /Marcus
>
Makes sense to me.
In the end, I committed the below as r220647 after checking that that
the patch still worked for aarch64-none-elf.
Cheers,
James
---
Index: gcc/testsuite/ChangeLog
===================================================================
--- gcc/testsuite/ChangeLog (revision 220646)
+++ gcc/testsuite/ChangeLog (working copy)
@@ -1,3 +1,7 @@
+2015-02-12 James Greenhalgh <james.greenha...@arm.com>
+
+ * gfortran.dg/pr45636.f90: XFAIL for aarch64* targets.
+
2015-02-12 Jakub Jelinek <ja...@redhat.com>
PR testsuite/64930
Index: gcc/testsuite/gfortran.dg/pr45636.f90
===================================================================
--- gcc/testsuite/gfortran.dg/pr45636.f90 (revision 220646)
+++ gcc/testsuite/gfortran.dg/pr45636.f90 (working copy)
@@ -10,5 +10,7 @@
b = y
call sub(a, b)
end program main
-! { dg-final { scan-tree-dump-times "memset" 0 "forwprop2" { xfail { {
hppa*-*-* && { ! lp64 } } || { mips*-*-* && { ! nomips16 } } } } } }
+! This test will fail on targets which prefer memcpy/memset over
+! move_by_pieces/store_by_pieces.
+! { dg-final { scan-tree-dump-times "memset" 0 "forwprop2" { xfail { {
hppa*-*-* && { ! lp64 } } || { { mips*-*-* && { ! nomips16 } } || {
aarch64*-*-* } } } } } }
! { dg-final { cleanup-tree-dump "forwprop2" } }
