Received-SPF: pass (google.com: domain of linux-crypto-owner@vger.kernel.org designates 23.128.96.18 as permitted sender) client-ip=23.128.96.18;
MIME-Version: 1.0
References: <1607686144-2604-1-git-send-email-TonyWWang-oc@zhaoxin.com> <X9Ov3RWDpUik7gXo@sol.localdomain>
In-Reply-To: <X9Ov3RWDpUik7gXo@sol.localdomain>
From:   Ard Biesheuvel <ardb@kernel.org>
Date:   Sat, 12 Dec 2020 10:36:28 +0100
Message-ID: <CAMj1kXEDjQG_my5FWVY+b7Q43-_waW74sZyBAPCkd7EEdku+Rw@mail.gmail.com>
Subject: Re: [PATCH] crypto: x86/crc32c-intel - Don't match some Zhaoxin CPUs
To:     Eric Biggers <ebiggers@kernel.org>
Cc:     Tony W Wang-oc <TonyWWang-oc@zhaoxin.com>,
        Herbert Xu <herbert@gondor.apana.org.au>,
        "David S. Miller" <davem@davemloft.net>,
        Thomas Gleixner <tglx@linutronix.de>,
        Ingo Molnar <mingo@redhat.com>, Borislav Petkov <bp@alien8.de>,
        X86 ML <x86@kernel.org>, "H. Peter Anvin" <hpa@zytor.com>,
        Linux Crypto Mailing List <linux-crypto@vger.kernel.org>,
        Linux Kernel Mailing List <linux-kernel@vger.kernel.org>,
        TimGuo-oc@zhaoxin.com, CooperYan@zhaoxin.com,
        QiyuanWang@zhaoxin.com, HerryYang@zhaoxin.com,
        CobeChen@zhaoxin.com, SilviaZhao@zhaoxin.com
Content-Type: text/plain; charset="UTF-8"
Precedence: bulk

On Fri, 11 Dec 2020 at 20:07, Eric Biggers <ebiggers@kernel.org> wrote:
>
> On Fri, Dec 11, 2020 at 07:29:04PM +0800, Tony W Wang-oc wrote:
> > The driver crc32c-intel match CPUs supporting X86_FEATURE_XMM4_2.
> > On platforms with Zhaoxin CPUs supporting this X86 feature, When
> > crc32c-intel and crc32c-generic are both registered, system will
> > use crc32c-intel because its .cra_priority is greater than
> > crc32c-generic. This case expect to use crc32c-generic driver for
> > some Zhaoxin CPUs to get performance gain, So remove these Zhaoxin
> > CPUs support from crc32c-intel.
> >
> > Signed-off-by: Tony W Wang-oc <TonyWWang-oc@zhaoxin.com>
>
> Does this mean that the performance of the crc32c instruction on those CPUs is
> actually slower than a regular C implementation?  That's very weird.
>

This driver does not use CRC instructions, but carryless
multiplication and aggregation. So I suppose the pclmulqdq instruction
triggers some pathological performance limitation here.

That means the crct10dif driver probably needs the same treatment.