This is the mail archive of the gcc-patches@gcc.gnu.org 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: rewrite x86-64 libffi support


Richard Henderson <rth@redhat.com> writes:

> The previous implementation still had a bunch of FIXMEs and
> suchlike for just about everything connected with structures.
>
> The following is essentially a complete rewrite, based off
> the rather more efficient split between C and assembly that
> I used for the Alpha port.  Rather than doing C->asm->C, we
> do C->asm and play tricks with alloca such that the stack
> frame for the assembly is created while still in C.
>
> Anyway, there are no longer any failures at all in libffi.

Excellent.

But what strikes me are the java failures.

This is from:
http://gcc.gnu.org/ml/gcc-testresults/2004-12/msg01129.html
which was before your change:

	=== libjava Summary for unix ===

# of expected passes		3706
# of unexpected successes	2
# of expected failures		12
# of untested testcases		12



But now:
http://gcc.gnu.org/ml/gcc-testresults/2004-12/msg01154.html
	=== libjava Summary for unix ===

# of expected passes		3575
# of unexpected failures	65
# of unexpected successes	2
# of expected failures		12
# of untested testcases		77


But this might also be an unrelated change,

Andreas
-- 
 Andreas Jaeger, aj@suse.de, http://www.suse.de/~aj
  SUSE Linux Products GmbH, Maxfeldstr. 5, 90409 NÃrnberg, Germany
   GPG fingerprint = 93A3 365E CE47 B889 DF7F  FED1 389A 563C C272 A126

Attachment: pgp00000.pgp
Description: PGP signature


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