This is the mail archive of the mailing list for the GCC project.

Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]
Other format: [Raw text]

Re: [RFC] patch: cleanup gcc.dg/vect/ (take 3)

On Tue, Nov 16, 2004 at 02:55:50PM +0200, Dorit Naishlos wrote:
> > I had hoped to
> > be able to xfail classes of targets, such as targets that don't handle
> > vector alignment, but don't yet have a way to do that so the xfail
> > lists still name specific targets.
> Thanks for trying.
> A question: how would you write a testcase that when compiled on powerpc*
> the dg-final check xfails for powerpc64* or if -m64 is used? (I want to
> xfail testcase pr18425.c for 64bit, and temporarily also xfail testcases
> vect-[77,77a,78].c for 64bit - see PR18403).

GCC for powerpc64-*-linux* can be any of the following, depending on
how it is configured:

  a) a compiler that generates only LP64 code
  b) a biarch compiler that generates ILP32 code by default
  c) a biarch compiler that generates LP64 code by default

There's currently no way to detect, in an xfail list, that a test is
being run for LP64 on a powerpc64-*-linux* target.  My nightly bootstrap
and test run uses (b) and Jon Grimm's uses (c).  SUSE and Red Hat build
GCC that is (b), but Gentoo uses (a).

I had a scheme for handling this in xfail and target lists but alas, it
depended on using TCL code within a test and that breaks a basic
requirement of the testsuite.  I'll try to come up with another
solution, but in the meantime it's more important that those tests run
on powerpc64-*-* for 32-bit code than to xfail them for 64-bit code.

A test can be skipped (reported as UNSUPPORTED) for 64-bit code on all
targets with:

  { dg-require-effective-target ilp32 }


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]