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] Run copyrename one extra time before out-of-ssa

Steven Bosscher wrote:
- Not one binary was bigger, and a few were smaller with the patch.
 Overall the binaries are ~0.5% smaller.
- Compile time was 2s faster on 194 seconds for the unpatched compiler.
 Let's call it compile time neutral.
- For SPECint there is a 5% win for twolf, all other benchmarks are
 within 0.5%.
- For SPECfp there is a 30% improvement for mgrid (!), and 6% for mesa.
 The rest is in the noise (within 0.4%).

Very good data.

The results for i686 are more mixed, but overall still better with the
patch (peak) than without. gap failed to build and vortex miscompared,
I'm not sure why.

Do those failures occur without your patch? If so, then we can overlook them; if not, then we cannot include your patch until the problems are resolved.

Assuming that these failures do indeed occur without your patch, then you have proven that your patch is an improvement to my satisfaction, and I have no objection to including the patch.

Mark Mitchell
CodeSourcery, LLC
(916) 791-8304

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