From: Florian Weimer <fw@deneb.enyo.de>
To: "H.J. Lu via Libc-alpha" <libc-alpha@sourceware.org>
Cc: Florian Weimer <fweimer@redhat.com>,
"H.J. Lu" <hjl.tools@gmail.com>,
Paul Zimmermann <Paul.Zimmermann@inria.fr>
Subject: Re: [PATCH] x86_64: Add sinh with FMA
Date: Sun, 20 Aug 2023 16:28:31 +0200 [thread overview]
Message-ID: <87y1i5lre8.fsf@mid.deneb.enyo.de> (raw)
In-Reply-To: <CAMe9rOrfesG-bNTe4BdySCjHXpRtW9misb2rv3dhHS25j_0O9A@mail.gmail.com> (H. J. Lu via Libc-alpha's message of "Thu, 17 Aug 2023 12:21:50 -0700")
* H. J. Lu via Libc-alpha:
> On Thu, Aug 17, 2023 at 12:15 PM Florian Weimer <fweimer@redhat.com> wrote:
>>
>> * H. J. Lu via Libc-alpha:
>> > There is only one FMA insn in __ieee754_sinh_fma:
>> >
>> > 114: c4 e2 e9 9b 05 00 00 00 00 vfmsub132sd 0x0(%rip),%xmm2,%xmm0
>> >
>> > Since it calls __ieee754_exp and __expm1, with
>> >
>> > commit 1b214630ce6f7e0099b8b6f87246246739b079cf (master)
>> > Author: H.J. Lu <hjl.tools@gmail.com>
>> > Date: Fri Aug 11 08:04:08 2023 -0700
>> >
>> > x86_64: Add expm1 with FMA
>> >
>> > On Skylake, it improves expm1 bench performance by:
>> >
>> > Before After Improvement
>> > max 70.204 68.054 3%
>> > min 20.709 16.2 22%
>> > mean 22.1221 16.7367 24%
>> >
>> > There is very little improvement with a single FMA insn.
>>
>> If you specialize the implementation for FMA, could you avoid the
>> indirect call to __ieee754_exp and __expm1?
>>
>
> That is true. Performance differences are
>
> 1. Non-FMA
>
> "sinh": {
> "": {
> "duration": 1.60856e+09,
> "iterations": 1.248e+08,
> "max": 173.747,
> "min": 10.706,
> "mean": 12.8891
> }
> }%
>
> 2. FMA
>
> "sinh": {
> "": {
> "duration": 1.61017e+09,
> "iterations": 1.275e+08,
> "max": 100.218,
> "min": 9.678,
> "mean": 12.6288
> }
> }%
>
> FMA is a little faster. Should I repost it with updated performance
> numbers?
So the baseline is the before column from the original patch?
> On Skylake, it improves sinh bench performance by:
>
> Before After Improvement
> max 64.06 60.416 7%
> min 10.874 9.674 11%
> mean 14.0546 12.2018 13%
So we go from 14.0546 to 12.6288 for the mean case?
next prev parent reply other threads:[~2023-08-20 14:28 UTC|newest]
Thread overview: 10+ messages / expand[flat|nested] mbox.gz Atom feed top
2023-08-14 15:16 H.J. Lu
2023-08-14 15:50 ` Andreas Schwab
2023-08-14 15:57 ` H.J. Lu
2023-08-14 16:06 ` Andreas Schwab
2023-08-16 14:45 ` Paul Zimmermann
2023-08-16 17:23 ` H.J. Lu
2023-08-17 19:15 ` Florian Weimer
2023-08-17 19:21 ` H.J. Lu
2023-08-20 14:28 ` Florian Weimer [this message]
2023-08-20 14:29 ` H.J. Lu
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=87y1i5lre8.fsf@mid.deneb.enyo.de \
--to=fw@deneb.enyo.de \
--cc=Paul.Zimmermann@inria.fr \
--cc=fweimer@redhat.com \
--cc=hjl.tools@gmail.com \
--cc=libc-alpha@sourceware.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).