[PATCH, PR 42855] Skip test for PR 42585 on ppc and arm

[Richard Earnshaw]

On Mon, 2010-03-08 at 16:51 +0100, Martin Jambor wrote:
> Hi,
> 
> On Fri, Mar 05, 2010 at 11:20:38AM -0800, Janis Johnson wrote:
> > On Thu, 2010-03-04 at 12:23 +0100, Martin Jambor wrote:
> > > test "gcc.dg/tree-ssa/pr42585.c scan-tree-dump-times optimized"
> > > reportedly fails on powerpc and arm, probably because MOVE_RATIO is
> > > not defined for them and so the value of 2 is assumed and therefore
> > > total scalarization is not performed.  I have therefore changed it to
> > > be skipped on the two platforms.
> > 
> > Is it a bug that those targets don't define MOVE_RATIO?  If so then the
> > test could XFAIL those steps just before the release, but in the
> > meantime the relevant maintainers should be made aware of the reason
> > for these failures so they can fix them.
> > 
> 
> Even the old SRA used MOVE_RATIO, in fact in very similar fashion.  I
> therefore assumed that the behavior for those platforms was expected.
> OTOH, since it often leads to code size improvements, it is peculiar
> that at least arm does not (reportedly) define it.
> 

There should be no need to define it:

in tm.texi, it says: "If you don't define this, a reasonable default is
used."

So there's always an implicit definition, and if some code insists on
there being an actual definition, then it's broken.


> In the end, the respective maintainers should decide, so I included
> them in CC.
> 
> > > I have built a c-compiler on ppc and verified it is skipped there and
> > > also made sure it is still executed on x86_64.  OK for trunk and 4.4?
> > 
> > If it's not a bug, but missing functionality, then it's OK to skip the
> > check for those targets.
> 
> I don't really care much, I just want to get rid off this issue (that
> I sort-of caused).
> 
> Thanks,
> 
> Martin
> 
> > > 
> > > 
> > > 2010-03-03  Martin Jambor  <mjambor@suse.cz>
> > > 
> > > 	PR testsuite/42855
> > > 	* testsuite/gcc.dg/tree-ssa/pr42585.c: Skip dump scan on powerpc
> > > 	and arm.
> > > 
> > > Index: mine/gcc/testsuite/gcc.dg/tree-ssa/pr42585.c
> > > ===================================================================
> > > --- mine.orig/gcc/testsuite/gcc.dg/tree-ssa/pr42585.c
> > > +++ mine/gcc/testsuite/gcc.dg/tree-ssa/pr42585.c
> > > @@ -32,6 +32,6 @@ Cyc_string_ungetc (int ignore, struct _f
> > >  }
> > > 
> > >  /* The local aggregates . */
> > > -/* { dg-final { scan-tree-dump-times "struct _fat_ptr _ans" 0 "optimized"} } */
> > > -/* { dg-final { scan-tree-dump-times "struct _fat_ptr _T2" 0 "optimized"} } */
> > > +/* { dg-final { scan-tree-dump-times "struct _fat_ptr _ans" 0 "optimized" { target { { ! powerpc*-*-* } && { ! arm-*-* } } } } } */
> > > +/* { dg-final { scan-tree-dump-times "struct _fat_ptr _T2" 0 "optimized" { target { { ! powerpc*-*-* } && { ! arm-*-* } } } } } */
> > >  /* { dg-final { cleanup-tree-dump "optimized" } } */
> > 
> >