[PATCH] [AArch64] Improve and enable the SeparateConstOffsetFromGEP for AArch64 backend.

Hao Liu Hao.Liu at arm.com
Wed Nov 19 17:30:07 PST 2014


>>! In D5864#61, @jingyue wrote:
> SPEC2006 benchmarks are single-threaded, but this single thread can be
> scheduled by the operating system on different CPUs (cores, hyperthreads,
> etc.) in different time slices. I used to encounter fluctuation too and
> fixed that using the taskset command. Not sure if that can help you too.
> 
> Also, I don't think x86 matters because this pass is only enabled in
> AArch64 and NVPTX for now. I do feel a little worried on the 1%-2%
> regression you observed on AArch64. It would be great if you can get a more
> stable result.
> 
> Jingyue

Hi Jingyue,

When I run benchmarks on AArch64 machine, I use taskset because there are two kinds of cores (A53 and A57). I thought all x86 cores on my machine are the same, so I didn't use taskset. Thanks for your share. It is helpful for our future perf tests.

For the regressions, I'll keep on watching the performance results.

Thanks,
-Hao

http://reviews.llvm.org/D5864






More information about the llvm-commits mailing list