This is the mail archive of the
mailing list for the GCC project.
Re: RFA: patch for PR37397
H.J. Lu wrote:
On Wed, Nov 12, 2008 at 9:17 AM, Jeff Law <email@example.com> wrote:H.J., thanks for reporting this. I'll submit a patch today to fix it.
Vladimir Makarov wrote:
OK. Good find.
This patch improves SPEC benchmarks. I saw stable improvements on
x86/x86_64 and ppc. This patch implements a small trick mentioned in one
classical article by Chaitin etc (Register allocation and spilling via graph
coloring). There is no sense to spill pseudo in whose live range nothing is
dying because the spill will not make other allocnos colorable and
additional reloads for the corresponding pseudo will be generated in reload
pass for each insn it occurs.
Is it ok to commit?
This caused 30% slowdown on 454.calculix in SPEC CPU 2006
with -O2 -ffast-math on Linux/Intel64.