From: Noah Goldstein <goldstein.w.n@gmail.com>
To: David Laight <David.Laight@aculab.com>
Cc: "tglx@linutronix.de" <tglx@linutronix.de>,
"mingo@redhat.com" <mingo@redhat.com>,
"bp@alien8.de" <bp@alien8.de>, "x86@kernel.org" <x86@kernel.org>,
"hpa@zytor.com" <hpa@zytor.com>,
"luto@kernel.org" <luto@kernel.org>,
"linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>
Subject: Re: [PATCH v4] arch/x86: Improve 'rep movs{b|q}' usage in memmove_64.S
Date: Fri, 19 Nov 2021 18:05:56 -0600 [thread overview]
Message-ID: <CAFUsyfLUQLj5py1AQ+4NptM6htWxV5i0qxkeXDUdFPfAnqRY2w@mail.gmail.com> (raw)
In-Reply-To: <e41913010e61429bbd2980a5c5ccfb39@AcuMS.aculab.com>
On Fri, Nov 19, 2021 at 4:31 PM David Laight <David.Laight@aculab.com> wrote:
>
> From: Noah Goldstein
> > Sent: 17 November 2021 22:45
> >
> > On Wed, Nov 17, 2021 at 4:31 PM David Laight <David.Laight@aculab.com> wrote:
> > >
> > > From: Noah Goldstein
> > > > Sent: 17 November 2021 21:03
> > > >
> > > > Add check for "short distance movsb" for forwards FSRM usage and
> > > > entirely remove backwards 'rep movsq'. Both of these usages hit "slow
> > > > modes" that are an order of magnitude slower than usual.
> > > >
> > > > 'rep movsb' has some noticeable VERY slow modes that the current
> > > > implementation is either 1) not checking for or 2) intentionally
> > > > using.
> > >
> > > How does this relate to the decision that glibc made a few years
> > > ago to use backwards 'rep movs' for non-overlapping copies?
> >
> > GLIBC doesn't use backwards `rep movs`. Since the regions are
> > non-overlapping it just uses forward copy. Backwards `rep movs` is
> > from setting the direction flag (`std`) and is a very slow byte
> > copy. For overlapping regions where backwards copy is necessary GLIBC
> > uses 4x vec copy loop.
>
> Try to find this commit 6fb8cbcb58a29fff73eb2101b34caa19a7f88eba
>
> Or follow links from https://www.win.tue.nl/~aeb/linux/misc/gcc-semibug.html
> But I can't find the actual patch.
>
> The claims were a massive performance increase for the reverse copy.
>
I don't think that's referring to optimizations around `rep movs`. It
appears to be referring to fallout from this patch:
https://sourceware.org/git/?p=glibc.git;a=commit;h=6fb8cbcb58a29fff73eb2101b34caa19a7f88eba
which broken programs misusing `memcpy` with overlapping regions
resulting in this fix:
https://sourceware.org/git/?p=glibc.git;a=commit;h=0354e355014b7bfda32622e0255399d859862fcd
AFAICT support for ERMS was only added around:
https://sourceware.org/git/?p=glibc.git;a=commit;h=13efa86ece61bf84daca50cab30db1b0902fe2db
Either way GLIBC memcpy/memmove moment most certainly does not
use backwards `rep movs`:
https://sourceware.org/git/?p=glibc.git;a=blob;f=sysdeps/x86_64/multiarch/memmove-vec-unaligned-erms.S;hb=HEAD#l655
as it is very slow.
> The pdf from www.agner.org/optimize may well indicate why some
> copies are unexpectedly slow due to cache access aliasing.
Even in the `4k` aliasing case `rep movsb` seems to stay within a
factor of 2 of optimal whereas the `std` backwards `rep movs` loses
by a factor of 10.
Either way, `4k` aliasing detection is mostly a concern of `memcpy` as
the direction of copy for `memmove` is a correctness question, not
an optimization.
>
> I'm pretty sure that Intel cpu (possibly from Ivy bridge onwards)
> can be persuaded to copy 8 bytes/clock for in-cache data with
> a fairly simple loop that contains 2 reads (maybe misaligned)
> and two writes (so 16 bytes per iteration).
> Extra unrolling just adds extra code top and bottom.
>
> You might want a loop like:
> 1: mov 0(%rsi, %rcx),%rax
> mov 8(%rsi, %rcx),%rdx
> mov %rax, 0(%rdi, %rcx)
> mov %rdx, 8(%rdi, %rcx)
> add $16, %rcx
> jnz 1b
>
> David
The backwards loop already has 4x unrolled `movq` loop.
>
> -
> Registered Address Lakeside, Bramley Road, Mount Farm, Milton Keynes, MK1 1PT, UK
> Registration No: 1397386 (Wales)
next prev parent reply other threads:[~2021-11-20 0:06 UTC|newest]
Thread overview: 13+ messages / expand[flat|nested] mbox.gz Atom feed top
2021-11-01 4:49 [PATCH v1] arch/x86: Improve 'rep movs{b|q}' usage in memmove_64.S Noah Goldstein
2021-11-01 18:59 ` [PATCH v2] " Noah Goldstein
2021-11-02 23:15 ` [PATCH v3] " Noah Goldstein
2021-11-17 21:02 ` [PATCH v4] " Noah Goldstein
2021-11-17 22:31 ` David Laight
2021-11-17 22:45 ` Noah Goldstein
2021-11-19 22:31 ` David Laight
2021-11-20 0:05 ` Noah Goldstein [this message]
2021-12-10 18:35 ` Noah Goldstein
2022-01-12 3:13 ` Noah Goldstein
2022-02-10 9:08 ` Noah Goldstein
2022-03-16 14:15 ` Noah Goldstein
2022-03-16 16:23 ` 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=CAFUsyfLUQLj5py1AQ+4NptM6htWxV5i0qxkeXDUdFPfAnqRY2w@mail.gmail.com \
--to=goldstein.w.n@gmail.com \
--cc=David.Laight@aculab.com \
--cc=bp@alien8.de \
--cc=hpa@zytor.com \
--cc=linux-kernel@vger.kernel.org \
--cc=luto@kernel.org \
--cc=mingo@redhat.com \
--cc=tglx@linutronix.de \
--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 NNTP newsgroup(s).