From: Andy Lutomirski <luto@kernel.org>
To: libc-alpha <libc-alpha@sourceware.org>,
"H. J. Lu" <hjl.tools@gmail.com>, X86 ML <x86@kernel.org>,
LKML <linux-kernel@vger.kernel.org>,
"Bae, Chang Seok" <chang.seok.bae@intel.com>,
Florian Weimer <fweimer@redhat.com>,
"Carlos O'Donell" <carlos@redhat.com>,
Rich Felker <dalias@libc.org>
Subject: Why does glibc use AVX-512?
Date: Thu, 25 Mar 2021 21:38:24 -0700 [thread overview]
Message-ID: <CALCETrURmk4ZijJVUtJwouj=_0NPiUvUFr9XMvdniRRFqeU+fg@mail.gmail.com> (raw)
Hi all-
glibc appears to use AVX512F for memcpy by default. (Unless
Prefer_ERMS is default-on, but I genuinely can't tell if this is the
case. I did some searching.) The commit adding it refers to a 2016
email saying that it's 30% on KNL. Unfortunately, AVX-512 is now
available in normal hardware, and the overhead from switching between
normal and AVX-512 code appears to vary from bad to genuinely
horrible. And, once anything has used the high parts of YMM and/or
ZMM, those states tend to get stuck with XINUSE=1.
I'm wondering whether glibc should stop using AVX-512 by default.
Meanwhile, some of you may have noticed a little ABI break we have.
On AVX-512 hardware, the size of a signal frame is unreasonably large,
and this is causing problems even for existing software that doesn't
use AVX-512. Do any of you have any clever ideas for how to fix it?
We have some kernel patches around to try to fail more cleanly, but we
still fail.
I think we should seriously consider solutions in which, for new
tasks, XCR0 has new giant features (e.g. AMX) and possibly even
AVX-512 cleared, and programs need to explicitly request enablement.
This would allow programs to opt into not saving/restoring across
signals or to save/restore in buffers supplied when the feature is
enabled. This has all kinds of pros and cons, and I'm not sure it's a
great idea. But, in the absence of some change to the ABI, the
default outcome is that, on AMX-enabled kernels on AMX-enabled
hardware, the signal frame will be more than 8kB, and this will affect
*every* signal regardless of whether AMX is in use.
--Andy
next reply other threads:[~2021-03-26 4:38 UTC|newest]
Thread overview: 14+ messages / expand[flat|nested] mbox.gz Atom feed top
2021-03-26 4:38 Andy Lutomirski [this message]
2021-03-26 10:06 ` Borislav Petkov
2021-03-26 18:17 ` Andy Lutomirski
2021-03-26 12:12 ` Florian Weimer
2021-03-26 18:14 ` Andy Lutomirski
2021-03-26 19:34 ` Florian Weimer
2021-03-26 19:47 ` Andy Lutomirski
2021-03-26 20:06 ` Andrew Cooper
2021-03-26 20:35 ` Florian Weimer
2021-03-26 20:43 ` H.J. Lu
2021-03-26 20:48 ` Andy Lutomirski
2021-03-26 21:11 ` Florian Weimer
2021-03-26 21:21 ` Andy Lutomirski
2021-03-26 13:32 ` David Laight
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to='CALCETrURmk4ZijJVUtJwouj=_0NPiUvUFr9XMvdniRRFqeU+fg@mail.gmail.com' \
--to=luto@kernel.org \
--cc=carlos@redhat.com \
--cc=chang.seok.bae@intel.com \
--cc=dalias@libc.org \
--cc=fweimer@redhat.com \
--cc=hjl.tools@gmail.com \
--cc=libc-alpha@sourceware.org \
--cc=linux-kernel@vger.kernel.org \
--cc=x86@kernel.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).