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: [PATCH] do not lower a/b to a*(1/b)


On Mon, May 16, 2005 at 11:13:27AM -0400, David Edelsohn wrote:
> 	Janis Johnson might be able to test with SPEC on PowerPC.  SUSE
> and Red Hat might be able to test the patch on some of their machines.

I tried Paolo's patch against mainline sources from last night on
powerpc64-linux by building the SPEC CPU2000 tests with the following
options and running with the small "test" input:

  -m32 -O2 -ffast-math
  -m64 -O2 -ffast-math
  -m32 -O2 -funsafe-math-optimizations
  -m64 -O2 -funsafe-math-optimizations

eon failed to build, but fails with -O2 as well.  Everything else went
fine.

Janis


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