MIME-Version: 1.0
References: <20211125193852.3617-1-goldstein.w.n@gmail.com>
 <CANn89iLnH5B11CtzZ14nMFP7b--7aOfnQqgmsER+NYNzvnVurQ@mail.gmail.com>
 <CAFUsyfK-znRWJN7FTMdJaDTd45DgtBQ9ckKGyh8qYqn0eFMMFA@mail.gmail.com>
 <CANn89iJA1JOevgLSK019VYXKkLJsMjV-u9ZHvrgZ+XUypRWwfQ@mail.gmail.com>
 <CAFUsyfLeVGW8etXHuSGvYy_RoS3RGaA1L+NLKnpc7EsSMVORBg@mail.gmail.com>
 <CANn89iJxMhGfp364rPu6p_ZLrKnM1qWF_NWrw4_oL_KG+piByg@mail.gmail.com>
 <CAFUsyfJ33cKFQdUagHQ_b4N80CfBtGQZhyA4CN_JLgEmXEX=DA@mail.gmail.com> <CAFUsyf+TuU2Xe5Guy5yiFWsV-JZSjUJxcGZv=f0BYDSmODV+_w@mail.gmail.com>
In-Reply-To: <CAFUsyf+TuU2Xe5Guy5yiFWsV-JZSjUJxcGZv=f0BYDSmODV+_w@mail.gmail.com>
From:   Eric Dumazet <edumazet@google.com>
Date:   Fri, 26 Nov 2021 11:21:34 -0800
Message-ID: <CANn89iLjw7YeWNGNtVNi690adJfoSVwgLScrtx-zSyh=COZRsQ@mail.gmail.com>
Subject: Re: [PATCH v1] x86/lib: Optimize 8x loop and memory clobbers in csum_partial.c
To:     Noah Goldstein <goldstein.w.n@gmail.com>
Cc:     tglx@linutronix.de, mingo@redhat.com,
        Borislav Petkov <bp@alien8.de>, dave.hansen@linux.intel.com,
        X86 ML <x86@kernel.org>, hpa@zytor.com, peterz@infradead.org,
        alexanderduyck@fb.com, open list <linux-kernel@vger.kernel.org>
Content-Type: text/plain; charset="UTF-8"
Precedence: bulk

On Fri, Nov 26, 2021 at 11:14 AM Noah Goldstein <goldstein.w.n@gmail.com> wrote:
>
> On Fri, Nov 26, 2021 at 12:50 PM Noah Goldstein <goldstein.w.n@gmail.com> wrote:
> >
> > On Fri, Nov 26, 2021 at 12:27 PM Eric Dumazet <edumazet@google.com> wrote:
> > >
> > > On Fri, Nov 26, 2021 at 10:17 AM Noah Goldstein <goldstein.w.n@gmail.com> wrote:
> > > >
> > >
> > > >
> > > > Makes sense. Although if you inline I think you definitely will want a more
> > > > conservative clobber than just "memory". Also I think with 40 you also will
> > > > get some value from two counters.
> > > >
> > > > Did you see the number/question I posted about two accumulators for 32
> > > > byte case?
> > > > Its a judgement call about latency vs throughput that I don't really have an
> > > > answer for.
> > > >
> > >
> > > The thing I do not know is if using more units would slow down the
> > > hyper thread ?
>
> Did some quick tests with the latency/throughput benchmarks running
> in parallel on two hyperthreads on the same processors. The 32 byte case
> latency advantage goes with 2 accum and there is still a slight regression
> in throughput. The larger cases that hit the loop still still have improvements
> both in tput and latency with 2 accum.

Great. I also played with rorx instruction, because it removes one
"adcl $0,..." step

__wsum ipv6_csum_partial(const void *buff, int len, __wsum sum)
{
u64 temp64;
u64 tmp;
u32 res32;

if (unlikely(len != 40))
return csum_partial(buff, len, sum);

temp64 = (__force u64)sum;
asm("addq 0*8(%[src]),%[temp64]\n\t"
    "adcq 1*8(%[src]),%[temp64]\n\t"
    "adcq 2*8(%[src]),%[temp64]\n\t"
    "adcq 3*8(%[src]),%[temp64]\n\t"
    "adcq 4*8(%[src]),%[temp64]\n\t"
    "mov  %k[temp64],%[res32]\n\t"
    "rorx $32,%[temp64],%[temp64]\n\t"
    "adcl %k[temp64],%[res32]\n\t"
    "adcl $0,%[res32]"
    : [temp64] "+r" (temp64), [res32] "=r" (res32)
    : [src] "r" (buff)
    : "memory");
return (__force __wsum)res32;
}