[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [PATCH v2 0/4] target/arm vector improvements
From: |
Peter Maydell |
Subject: |
Re: [PATCH v2 0/4] target/arm vector improvements |
Date: |
Thu, 20 Feb 2020 14:55:49 +0000 |
On Sun, 16 Feb 2020 at 21:42, Richard Henderson
<address@hidden> wrote:
>
> Changes for v2:
>
> * Rename function arguments as requested by ajb.
> * Minor optimization in helper_gvec_pmul_b.
>
> Original blurb:
>
> The first patch has been seen before.
>
> https://patchwork.ozlabs.org/patch/1115039/
>
> It had a bug and I didn't fix it right away and then forgot.
> Fixed now; I had mixed up the operand ordering for aarch32.
>
> The next 3 are something that I noticed while doing other stuff.
>
> In particular, pmull is used heavily during https transfers.
> While cloning a repository, the old code peaks at 27% of the
> total runtime, as measured by perf top. The new code does
> not quite reach 3% repeating the same clone.
>
> In addition, the new helper functions are in the form that
> will be required for the implementation of SVE2.
Applied to target-arm.next, thanks.
-- PMM
- [PATCH v2 0/4] target/arm vector improvements, Richard Henderson, 2020/02/16
- [PATCH v2 2/4] target/arm: Convert PMUL.8 to gvec, Richard Henderson, 2020/02/16
- [PATCH v2 1/4] target/arm: Vectorize USHL and SSHL, Richard Henderson, 2020/02/16
- [PATCH v2 3/4] target/arm: Convert PMULL.64 to gvec, Richard Henderson, 2020/02/16
- [PATCH v2 4/4] target/arm: Convert PMULL.8 to gvec, Richard Henderson, 2020/02/16
- Re: [PATCH v2 0/4] target/arm vector improvements, no-reply, 2020/02/16
- Re: [PATCH v2 0/4] target/arm vector improvements,
Peter Maydell <=