Messages in this thread | | | From | Eric Dumazet <> | Date | Fri, 26 Nov 2021 22:51:05 -0800 | Subject | Re: [PATCH v3] x86/lib: Optimize 8x loop and memory clobbers in csum_partial.c |
| |
On Fri, Nov 26, 2021 at 10:39 PM Noah Goldstein <goldstein.w.n@gmail.com> wrote: > > Modify the 8x loop to that it uses two independent > accumulators. Despite adding more instructions the latency and > throughput of the loop is improved because the `adc` chains can now > take advantage of multiple execution units.
Oh well, there was really no need to resend this, especially if you do not add my ack.
Reviewed-by: Eric Dumazet <edumazet@google.com>
| |