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>
 <CANn89iLjw7YeWNGNtVNi690adJfoSVwgLScrtx-zSyh=COZRsQ@mail.gmail.com>
 <CAFUsyfK5r+P6aQLBpYZoi0FSgvLwfawBUYTHtQL0TJRozujp3g@mail.gmail.com> <CANn89iKkK0eMQkOTSMGbvdtf+G=eRfXJG0Semju5h-C63QytaQ@mail.gmail.com>
In-Reply-To: <CANn89iKkK0eMQkOTSMGbvdtf+G=eRfXJG0Semju5h-C63QytaQ@mail.gmail.com>
From:   Noah Goldstein <goldstein.w.n@gmail.com>
Date:   Fri, 26 Nov 2021 14:33:31 -0600
Message-ID: <CAFUsyf+9F9b2V-YDm8UdbnANrQ0_fYV6cGfRgPciASfvp9p=7w@mail.gmail.com>
Subject: Re: [PATCH v1] x86/lib: Optimize 8x loop and memory clobbers in csum_partial.c
To:     Eric Dumazet <edumazet@google.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 2:07 PM Eric Dumazet <edumazet@google.com> wrote:
>
> On Fri, Nov 26, 2021 at 11:50 AM Noah Goldstein <goldstein.w.n@gmail.com> wrote:
> >
> > Bright :) but it will need a BMI support check.
>
> Yes, probably not worth the pain.

Making a V2 for my patch with your optimization for the loop case. Do you think
1 or 2 accum for the 32 byte case?
>
> >
> > I actually get better performance in hyperthread benchmarks with 2 accum:
> >
> > Used:
> >
> >         u64 res;
> >         temp64 = (__force uint64_t)sum;
> >         asm("movq 0*8(%[src]),%[res]\n\t"
> >             "addq 1*8(%[src]),%[res]\n\t"
> >             "adcq 2*8(%[src]),%[res]\n\t"
> >             "adcq   $0, %[res]\n"
> >             "addq 3*8(%[src]),%[temp64]\n\t"
> >             "adcq 4*8(%[src]),%[temp64]\n\t"
> >             "adcq   %[res], %[temp64]\n\t"
> >             "mov  %k[temp64],%k[res]\n\t"
> >             "rorx $32,%[temp64],%[temp64]\n\t"
> >             "adcl %k[temp64],%k[res]\n\t"
> >             "adcl $0,%k[res]"
> >             : [temp64] "+r"(temp64), [res] "=&r"(res)
> >             : [src] "r"(buff)
> >             : "memory");
> >         return (__force __wsum)res;
> >
> > w/ hyperthread:
> > size,    2acc lat,    1acc lat,   2acc tput,   1acc tput
> >   40,       6.511,       7.863,       6.177,       6.157
> >
> > w/o hyperthread:
> > size,    2acc lat,    1acc lat,   2acc tput,   1acc tput
> >   40,       5.577,       6.764,       3.150,       3.210