Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 23.128.96.18 as permitted sender) client-ip=23.128.96.18;
Content-Type: text/plain; charset=utf-8
Content-Transfer-Encoding: quoted-printable
From:   Andy Lutomirski <luto@amacapital.net>
Mime-Version: 1.0 (1.0)
Subject: Re: Candidate Linux ABI for Intel AMX and hypothetical new related features
Date:   Mon, 12 Apr 2021 08:10:15 -0700
Message-Id: <27CD4C77-E58A-4AC1-8415-3E57BE0F3663@amacapital.net>
References: <878s5nk1pk.fsf@oldenburg.str.redhat.com>
Cc:     Borislav Petkov <bp@alien8.de>, Andy Lutomirski <luto@kernel.org>,
        "Bae, Chang Seok" <chang.seok.bae@intel.com>,
        Dave Hansen <dave.hansen@intel.com>, X86 ML <x86@kernel.org>,
        LKML <linux-kernel@vger.kernel.org>, linux-abi@vger.kernel.org,
        libc-alpha@sourceware.org, Rich Felker <dalias@libc.org>,
        Kyle Huey <me@kylehuey.com>,
        Keno Fischer <keno@juliacomputing.com>
In-Reply-To: <878s5nk1pk.fsf@oldenburg.str.redhat.com>
To:     Florian Weimer <fweimer@redhat.com>
Precedence: bulk


> On Apr 12, 2021, at 7:38 AM, Florian Weimer <fweimer@redhat.com> wrote:
>=20
> =EF=BB=BF* Borislav Petkov:
>=20
>>> On Mon, Apr 12, 2021 at 04:19:29PM +0200, Florian Weimer wrote:
>>> Maybe we could have done this in 2016 when I reported this for the first=

>>> time.  Now it is too late, as more and more software is using
>>> CPUID-based detection for AVX-512.
>>=20
>> So as I said on another mail today, I don't think a library should rely
>> solely on CPUID-based detection of features especially if those features
>> need kernel support too. IOW, it should ask whether the kernel can
>> handle those too, first.
>=20
> Yes, that's why we have the XGETBV handshake.  I was imprecise.  It's
> CPUID + XGETBV of course.  Or even AT_HWCAP2 (for FSGSBASE).
>=20
>> And the CPUID-faulting thing would solve stuff like that because then
>> the kernel can *actually* get involved into answering something where it
>> has a say in, too.
>=20
> But why wouldn't we use a syscall or an entry in the auxiliary vector
> for that?  Why fault a potentially performance-critical instruction?
>=20

CPUID is horrifically slow in various virt scenarios. If user code needs to s=
erialize, use IRET or SERIALIZE.=