From: Ingo Molnar <mingo@kernel.org>
To: Andy Lutomirski <luto@amacapital.net>
Cc: Denys Vlasenko <vda.linux@googlemail.com>,
Oleg Nesterov <oleg@redhat.com>, Rik van Riel <riel@redhat.com>,
X86 ML <x86@kernel.org>,
Linux Kernel Mailing List <linux-kernel@vger.kernel.org>,
Borislav Petkov <bp@alien8.de>,
Linus Torvalds <torvalds@linux-foundation.org>,
Andrew Morton <akpm@linux-foundation.org>,
"H. Peter Anvin" <hpa@zytor.com>,
Thomas Gleixner <tglx@linutronix.de>
Subject: Re: [RFC PATCH] x86, fpu: Use eagerfpu by default on all CPUs
Date: Wed, 25 Feb 2015 11:45:45 +0100 [thread overview]
Message-ID: <20150225104544.GD554@gmail.com> (raw)
In-Reply-To: <CALCETrXY_=xas8NP=K0JJzFYZ1mSFdV5hs28Xseq8f_SJ2D+Sg@mail.gmail.com>
* Andy Lutomirski <luto@amacapital.net> wrote:
> > I'm a big fan of simplifying things, but.
> >
> > SIMD registers were growing in x86, and they are going
> > to grow again, this time four-fold in Intel MIC: from
> > sixteen 256-bit registers to thirty two 512-bit
> > registers.
> >
> > That's 2 kbytes of data. Just moving this data out
> > to/from memory will take some time.
> >
> > And some people talk about 1024-bit registers
> > already...
> >
> > Let's not completely remove lazy FPU saving code just
> > yet. Maybe we'll be forced to reinstate it.
>
> I'd prefer a different partial solution: encourage
> everyone to clear the xstate before making syscalls
> (using e.g. vzeroall). In fact, maybe user code should
> aggressively clear newly-unused xstate.
Also, xstate has various compaction features and could grow
new ones in the future as well, should the xsave area
become overly sparse: see xstate_comp_*[] et al in
arch/x86/kernel/xsave.c.
This is the better, hardware driven, synchronous
alternative to lazy, async register state save/restore, as
it gets us similar benefits of not saving/restoring unused
space, but avoids any async trap overhead.
Also ... with more and wider vector CPU registers context
switches between different sets of registers are going to
be inevitably more expensive, no matter what.
Thanks,
Ingo
next prev parent reply other threads:[~2015-02-25 10:45 UTC|newest]
Thread overview: 45+ messages / expand[flat|nested] mbox.gz Atom feed top
2015-02-20 18:58 [RFC PATCH] x86, fpu: Use eagerfpu by default on all CPUs Andy Lutomirski
2015-02-20 19:05 ` Borislav Petkov
2015-02-21 9:31 ` Ingo Molnar
2015-02-21 16:38 ` Borislav Petkov
2015-02-21 17:29 ` Borislav Petkov
2015-02-21 18:39 ` Ingo Molnar
2015-02-21 19:15 ` Borislav Petkov
2015-02-21 19:23 ` Ingo Molnar
2015-02-21 21:36 ` Borislav Petkov
2015-02-22 8:18 ` Ingo Molnar
2015-02-22 8:22 ` Ingo Molnar
2015-02-22 10:48 ` Borislav Petkov
2015-02-22 12:50 ` Borislav Petkov
2015-02-22 12:57 ` Ingo Molnar
2015-02-22 13:21 ` Borislav Petkov
2015-02-22 0:34 ` Maciej W. Rozycki
2015-02-22 2:18 ` Andy Lutomirski
2015-02-22 11:06 ` Borislav Petkov
2015-02-23 1:45 ` Rik van Riel
2015-02-23 5:22 ` Andy Lutomirski
2015-02-23 12:51 ` Rik van Riel
2015-02-23 15:03 ` Borislav Petkov
2015-02-23 15:51 ` Rik van Riel
2015-02-23 18:06 ` Borislav Petkov
2015-02-23 21:17 ` Maciej W. Rozycki
2015-02-23 21:21 ` Rik van Riel
2015-02-23 22:14 ` Linus Torvalds
2015-02-24 0:56 ` Maciej W. Rozycki
2015-02-24 0:59 ` Andy Lutomirski
2015-02-23 22:27 ` Maciej W. Rozycki
2015-02-23 23:44 ` Andy Lutomirski
2015-02-24 2:14 ` Maciej W. Rozycki
2015-02-24 2:31 ` Andy Lutomirski
2015-02-24 14:43 ` Rik van Riel
2015-02-21 18:34 ` Ingo Molnar
2015-02-23 14:59 ` Oleg Nesterov
2015-02-23 15:11 ` Borislav Petkov
2015-02-23 15:53 ` Rik van Riel
2015-02-23 18:40 ` Oleg Nesterov
2015-02-24 19:15 ` Denys Vlasenko
2015-02-25 0:07 ` Andy Lutomirski
2015-02-25 10:37 ` Borislav Petkov
2015-02-25 10:50 ` Ingo Molnar
2015-02-25 10:45 ` Ingo Molnar [this message]
2015-02-25 17:12 ` Some results (was: Re: [RFC PATCH] x86, fpu: Use eagerfpu by default on all CPUs) Borislav Petkov
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=20150225104544.GD554@gmail.com \
--to=mingo@kernel.org \
--cc=akpm@linux-foundation.org \
--cc=bp@alien8.de \
--cc=hpa@zytor.com \
--cc=linux-kernel@vger.kernel.org \
--cc=luto@amacapital.net \
--cc=oleg@redhat.com \
--cc=riel@redhat.com \
--cc=tglx@linutronix.de \
--cc=torvalds@linux-foundation.org \
--cc=vda.linux@googlemail.com \
--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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.