This is the mail archive of the
gcc-patches@gcc.gnu.org
mailing list for the GCC project.
Re: [RFC][AARCH64][PATCH 2/5]: Add number of hw prefetchers available to cpu_prefetch_tune
- From: Kugan Vivekanandarajah <kugan dot vivekanandarajah at linaro dot org>
- To: Andrew Pinski <pinskia at gmail dot com>
- Cc: "gcc-patches at gcc dot gnu dot org" <gcc-patches at gcc dot gnu dot org>
- Date: Sun, 17 Sep 2017 08:51:07 +1000
- Subject: Re: [RFC][AARCH64][PATCH 2/5]: Add number of hw prefetchers available to cpu_prefetch_tune
- Authentication-results: sourceware.org; auth=none
- References: <CAELXzTORASUv0UGqNwWByWNNDY5xiiZA3Pt-TJfd+S1MWbgdNw@mail.gmail.com> <CA+=Sn1ntSpFFppQXkQOpESJ=KgRRLOUs034HvOASD46EZ1svUw@mail.gmail.com>
Hi Andrew,
On 15 September 2017 at 13:20, Andrew Pinski <pinskia@gmail.com> wrote:
> On Thu, Sep 14, 2017 at 6:28 PM, Kugan Vivekanandarajah
> <kugan.vivekanandarajah@linaro.org> wrote:
>> This patch adds number of hw prefetchers available to
>> cpu_prefetch_tune so it can be used in loop unrolling decisions.
>
> Can you explain the difference between this and num_slots
> (PARAM_SIMULTANEOUS_PREFETCHES)? Because it seems like they should be
> the same here.
>
I kept it different for two reason.
1. I am not sure if this would have the same effect on all the
micro-architecture. Keeping it separate allows each microarchitecture
to enable prefetch loop arrays and aiding hw prefetcher (my goal here)
by limiting prefetch streams.
2. The params used for ARAM_SIMULTANEOUS_PREFETCHES seems to be a
value determined by experimentation rather than based on functional
units in hardware. This also allows tuning them speretaterly.
Thanks,
Kugan