From: Paul Zimmermann <Paul.Zimmermann@inria.fr>
To: Alexander Monakov <amonakov@ispras.ru>
Cc: gcc-help@gcc.gnu.org, stephane.glondu@inria.fr,
marc.glisse@inria.fr, sibid@uvic.ca
Subject: Re: slowdown with -std=gnu18 with respect to -std=c99
Date: Sat, 07 May 2022 08:11:36 +0200 [thread overview]
Message-ID: <mw4k2150tj.fsf@tomate.loria.fr> (raw)
In-Reply-To: <2b4e81-fee-e79-5ea0-bf658f20b4c2@ispras.ru> (message from Alexander Monakov on Fri, 6 May 2022 12:27:39 +0300 (MSK))
thank you very much Alexander for your analysis and the bugzilla report!
Paul
> Date: Fri, 6 May 2022 12:27:39 +0300 (MSK)
> From: Alexander Monakov <amonakov@ispras.ru>
>
> On Fri, 6 May 2022, Paul Zimmermann via Gcc-help wrote:
>
> > here are latency metrics (still on i5-4590):
> >
> > | gcc-9 | gcc-10 | gcc-11 |
> > ------------|-------|--------|--------|
> > -std=c99 | 70.8 | 70.3 | 70.2 |
> > -std=gnu18 | 59.5 | 59.5 | 59.5 |
> >
> > It thus seems the issue only appears for the reciprocal throughput.
>
> Thanks.
>
> The primary issue here is false dependency on vcvtss2sd instruction. In the
> snippet shown in Stéphane's email, the slower variant begins with
>
> vcvtss2sd -0x4(%rsp),%xmm1,%xmm1
>
> The cvtss2sd instruction is specified to take the upper bits of SSE register
> unmodified, so here it merges high bits of xmm1 with results of float->double
> conversion (in low bits) into new xmm1. Unless the CPU can track dependencies
> separately for vector register components, it has to delay this instruction
> until the previous computation that modified xmm1 has completed (AMD Zen2 is
> an example of a microarchitecture that apparently can).
>
> This limits the degree to which separate cr_log10f can overlap, affecting
> throughput. In latency measurements, the calls are already serialized by
> dependency over xmm0, so the additional false dependency does not matter.
>
> (so fma is a "red herring", it's just that depending on compiler version and
> flags, register allocation will place last assignment into xmm1 differently)
>
> If you want to experiment, you can hand-edit assembly to replace the problematic
> instruction with variants that avoid the false dependency, such as
>
> vcvtss2sd %xmm0, %xmm0, %xmm1
>
> or
>
> vpxor %xmm1, %xmm1, %xmm1
> vcvtss2sd -0x4(%rsp),%xmm1,%xmm1
>
> GCC has code to do this automatically, but for some reason it doesn't work for
> your function. I have reported in to the Bugzilla:
> https://gcc.gnu.org/bugzilla/show_bug.cgi?id=105504
>
> Alexander
next prev parent reply other threads:[~2022-05-07 6:11 UTC|newest]
Thread overview: 13+ messages / expand[flat|nested] mbox.gz Atom feed top
2022-05-03 8:28 Paul Zimmermann
2022-05-03 9:09 ` Alexander Monakov
2022-05-03 11:45 ` Paul Zimmermann
2022-05-03 12:12 ` Alexander Monakov
2022-05-05 8:57 ` Stéphane Glondu
2022-05-05 14:31 ` Stéphane Glondu
2022-05-05 14:41 ` Marc Glisse
2022-05-05 14:56 ` Alexander Monakov
2022-05-06 7:46 ` Paul Zimmermann
2022-05-06 9:27 ` Alexander Monakov
2022-05-07 6:11 ` Paul Zimmermann [this message]
2022-05-11 13:26 ` Alexander Monakov
2022-05-05 17:50 ` Paul Zimmermann
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=mw4k2150tj.fsf@tomate.loria.fr \
--to=paul.zimmermann@inria.fr \
--cc=amonakov@ispras.ru \
--cc=gcc-help@gcc.gnu.org \
--cc=marc.glisse@inria.fr \
--cc=sibid@uvic.ca \
--cc=stephane.glondu@inria.fr \
/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).