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

[Jingyue Wu]

Did you try pinning the benchmark on one CPU? Scheduling can significantly
disturb the benchmark results. Also, if your CPU has any boosting feature
(similar to Intel's turbo boost), try disabling that. One of our folks,
Mark, found that can make benchmarking sensitive too.

Jingyue

On Tue Nov 18 2014 at 10:08:50 PM Hao Liu <Hao.Liu at arm.com> wrote:

> >>! In D5864#45, @Gerolf wrote:
> > Hi Hao,
> >
> > could you share your current performance data + options you use? With
> your patch I see small gains on astar + xalancbmk, but a ~4% loss on
> libquantum.  The data is based on the 11/07 trunk on top of r221549, O3 +
> LTO, cyclone, ref input.
> > In addition to cortex, could you run on x86 also?
> Hi Gerolf,
>
> I ran spec2006 4 times on my x86 machine, but the results are always
> fluctuating. So I can not see whether it is better or not by enabling this
> pass. I think it is because I don't have much experience on testing x86.
>
> For the AArch64 backend, recently I ran twice on r222083 at trunk and
> r222017 at trunk on Cortex-A57. The results are similar to the previous
> results. But it has about 1%-2% regression on 471.omnetpp. That benchmark
> is always fluctuating, so I can't measure it accurately. Anyway, we have
> much improvements than regressions.
>
> Thanks,
> -Hao
>
> http://reviews.llvm.org/D5864
>
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20141119/b7941284/attachment.html>