From: "Paul A. Clarke" <pc@us.ibm.com>
To: adhemerval.zanella@linaro.org
Cc: libc-alpha@sourceware.org, murphyp@linux.ibm.com
Subject: Re: [PATCH 7/9] math: Remove powerpc e_hypot
Date: Tue, 12 Oct 2021 07:17:35 -0500 [thread overview]
Message-ID: <20211012121735.GM243632@li-24c3614c-2adc-11b2-a85c-85f334518bdb.ibm.com> (raw)
In-Reply-To: <bace65d8-a815-5163-428d-38e59f117226@linux.ibm.com>
On Wed, Oct 06, 2021 at 02:51:50PM -0500, Paul E Murphy via Libc-alpha wrote:
> On 10/6/21 2:46 PM, Adhemerval Zanella wrote:
> > On 06/10/2021 16:43, Paul E Murphy wrote:
> > > On 10/6/21 1:05 PM, Adhemerval Zanella via Libc-alpha wrote:
> > > > power9 master:
> > > >
> > > > "hypot": {
> > > > "workload-random": {
> > > > "duration": 5.20744e+08,
> > > > "iterations": 5.6e+07,
> > > > "reciprocal-throughput": 9.27604,
> > > > "latency": 9.32194,
> > > > "max-throughput": 1.07805e+08,
> > > > "min-throughput": 1.07274e+08
> > > > }
> > > > }
> > > >
> > > > power9 patched:
> > > >
> > > > "hypot": {
> > > > "workload-random": {
> > > > "duration": 5.23124e+08,
> > > > "iterations": 5.4e+07,
> > > > "reciprocal-throughput": 9.51892,
> > > > "latency": 9.85606,
> > > > "max-throughput": 1.05054e+08,
> > > > "min-throughput": 1.0146e+08
> > > > }
> > > > }
> > > >
> > > > Checked on powerpc64-linux-gnu (power8) and powerpc64le-linux-gnu
> > > > (power9).
> > > > ---
> > >
> > > This also removes the ppc specific hypotf too. Have you had a chance to measure the difference?
> > >
> > > I think it is OK if the performance difference is negligible. Though, I don't speak for the other ppc contributors in that regard. They may want to double-check the results.
> >
> > I would expect the performance difference to be similar to 64-bit,
> > at least on the hardware I checked (since most of the time would
> > be spent in FPU calculations).
> >
> > In any case, I don't see much gain in keeping a 32-bit powerpc
> > implementation that deviates from generic implementation.
> >
>
> Oops, I read this patch in isolation to the series. There probably isn't any
> difference. Sorry for the noise.
>
> Anyhow, in the context of the series, this looks OK to me.
[Accidentally dropped the mailing list from this post. Resending...]
The results above showed over 5% performance degradation.
I'm not sure how rigorous those results are, and while arguably "close",
they were a bit concerning.
I ran the results on a local Power9 system, and see much better performance
with the new code. I thought I should share in case anyone else had similar
concerns.
glibc-master on Power9 with patch 1/9 (to get "workload-random"):
"hypot": {
"workload-random": {
"duration": 5.1525e+08,
"iterations": 8e+07,
"reciprocal-throughput": 6.2956,
"latency": 6.58564,
"max-throughput": 1.58841e+08,
"min-throughput": 1.51846e+08
}
}
glibc-master on Power9 with patches 1-9/9:
"hypot": {
"workload-random": {
"duration": 5.20747e+08,
"iterations": 9.8e+07,
"reciprocal-throughput": 5.10696,
"latency": 5.52053,
"max-throughput": 1.95811e+08,
"min-throughput": 1.81142e+08
}
}
Tested-by: Paul A. Clarke <pc@us.ibm.com>
PC
next prev parent reply other threads:[~2021-10-12 12:17 UTC|newest]
Thread overview: 31+ messages / expand[flat|nested] mbox.gz Atom feed top
2021-10-06 18:05 [PATCH 0/9] Improve hypot() Adhemerval Zanella
2021-10-06 18:05 ` [PATCH 1/9] benchtests: Make hypot input random Adhemerval Zanella
2021-10-07 9:48 ` Paul Zimmermann
2021-10-07 11:20 ` Adhemerval Zanella
2021-10-07 17:08 ` Joseph Myers
2021-10-06 18:05 ` [PATCH 2/9] benchtests: Add hypotf Adhemerval Zanella
2021-10-07 10:40 ` Paul Zimmermann
2021-10-06 18:05 ` [PATCH 3/9] math: Simplify hypotf implementation Adhemerval Zanella
2021-10-07 9:44 ` Paul Zimmermann
2021-10-07 11:37 ` Adhemerval Zanella
2021-10-07 12:08 ` Paul Zimmermann
2021-10-06 18:05 ` [PATCH 4/9] math: Use an improved algorithm for hypot (dbl-64) Adhemerval Zanella
2021-10-06 18:05 ` [PATCH 5/9] math: Use an improved algorithm for hypotl (ldbl-96) Adhemerval Zanella
2021-10-06 18:05 ` [PATCH 6/9] math: Use an improved algorithm for hypotl (ldbl-128) Adhemerval Zanella
2021-10-06 18:05 ` [PATCH 7/9] math: Remove powerpc e_hypot Adhemerval Zanella
2021-10-06 19:43 ` Paul E Murphy
2021-10-06 19:46 ` Adhemerval Zanella
2021-10-06 19:51 ` Paul E Murphy
2021-10-12 12:17 ` Paul A. Clarke [this message]
2021-10-06 18:05 ` [PATCH 8/9] i386: Move hypot implementation to C Adhemerval Zanella
2021-10-06 18:37 ` Joseph Myers
2021-10-06 19:19 ` Adhemerval Zanella
2021-10-06 19:20 ` Adhemerval Zanella
2021-10-06 19:52 ` Joseph Myers
2021-10-07 8:28 ` Paul Zimmermann
2021-10-07 17:05 ` Joseph Myers
2021-10-06 18:05 ` [PATCH 9/9] math: Remove the error handling wrapper from hypot and hypotf Adhemerval Zanella
2021-10-06 18:38 ` Joseph Myers
2021-10-06 19:23 ` Adhemerval Zanella
2021-10-07 11:03 ` [PATCH 0/9] Improve hypot() Paul Zimmermann
2021-10-07 12:39 ` Adhemerval Zanella
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=20211012121735.GM243632@li-24c3614c-2adc-11b2-a85c-85f334518bdb.ibm.com \
--to=pc@us.ibm.com \
--cc=adhemerval.zanella@linaro.org \
--cc=libc-alpha@sourceware.org \
--cc=murphyp@linux.ibm.com \
/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).