From: "Martin J. Bligh" <mbligh@aracnet.com>
To: linux-kernel <linux-kernel@vger.kernel.org>
Cc: lse-tech <lse-tech@lists.sourceforge.net>
Subject: gcc 2.95 vs 3.21 performance
Date: Mon, 03 Feb 2003 15:05:06 -0800 [thread overview]
Message-ID: <336780000.1044313506@flay> (raw)
People keep extolling the virtues of gcc 3.2 to me, which I'm
reluctant to switch to, since it compiles so much slower. But
it supposedly generates better code, so I thought I'd compile
the kernel with both and compare the results. This is gcc 2.95
and 3.2.1 from debian unstable on a 16-way NUMA-Q. The kernbench
tests still use 2.95 for the compile-time stuff.
The results below leaves me distinctly unconvinced by the supposed
merits of modern gcc's. Not really better or worse, within experimental
error. But much slower to compile things with.
Kernbench-2: (make -j N vmlinux, where N = 2 x num_cpus)
Elapsed User System CPU
2.5.59 46.08 563.88 118.38 1480.00
2.5.59-gcc3.2 45.86 563.63 119.58 1489.33
Kernbench-16: (make -j N vmlinux, where N = 16 x num_cpus)
Elapsed User System CPU
2.5.59 47.45 568.02 143.17 1498.17
2.5.59-gcc3.2 47.15 567.41 143.72 1507.50
DISCLAIMER: SPEC(tm) and the benchmark name SDET(tm) are registered
trademarks of the Standard Performance Evaluation Corporation. This
benchmarking was performed for research purposes only, and the run results
are non-compliant and not-comparable with any published results.
Results are shown as percentages of the first set displayed
SDET 1 (see disclaimer)
Throughput Std. Dev
2.5.59 100.0% 0.8%
2.5.59-gcc3.2 95.3% 5.2%
SDET 2 (see disclaimer)
Throughput Std. Dev
2.5.59 100.0% 0.6%
2.5.59-gcc3.2 91.9% 7.1%
SDET 4 (see disclaimer)
Throughput Std. Dev
2.5.59 100.0% 5.7%
2.5.59-gcc3.2 98.8% 5.3%
SDET 8 (see disclaimer)
Throughput Std. Dev
2.5.59 100.0% 1.4%
2.5.59-gcc3.2 105.3% 4.7%
SDET 16 (see disclaimer)
Throughput Std. Dev
2.5.59 100.0% 1.7%
2.5.59-gcc3.2 103.1% 1.8%
SDET 32 (see disclaimer)
Throughput Std. Dev
2.5.59 100.0% 1.5%
2.5.59-gcc3.2 101.0% 1.6%
SDET 64 (see disclaimer)
Throughput Std. Dev
2.5.59 100.0% 0.7%
2.5.59-gcc3.2 103.1% 1.1%
SDET 128 (see disclaimer)
Throughput Std. Dev
NUMA schedbench 4:
AvgUser Elapsed TotalUser TotalSys
2.5.59 0.00 38.88 82.78 0.65
2.5.59-gcc3.2 0.00 41.80 107.76 0.73
NUMA schedbench 8:
AvgUser Elapsed TotalUser TotalSys
2.5.59 0.00 49.30 247.80 1.93
2.5.59-gcc3.2 0.00 38.00 229.83 2.11
NUMA schedbench 16:
AvgUser Elapsed TotalUser TotalSys
2.5.59 0.00 57.37 843.12 3.77
2.5.59-gcc3.2 0.00 57.28 839.21 2.85
NUMA schedbench 32:
AvgUser Elapsed TotalUser TotalSys
2.5.59 0.00 116.99 1805.79 6.05
2.5.59-gcc3.2 0.00 118.44 1788.09 6.25
NUMA schedbench 64:
AvgUser Elapsed TotalUser TotalSys
2.5.59 0.00 235.18 3632.73 15.45
2.5.59-gcc3.2 0.00 234.55 3633.76 15.02
------------------------------------------------------------------------------
And with the same kernel, comparing the compile times for gcc 2.95 to 3.2
Kernbench-2: (make -j N vmlinux, where N = 2 x num_cpus)
Elapsed User System CPU
gcc2.95 46.08 563.88 118.38 1480.00
gcc3.21 69.93 923.17 114.36 1483.17
Kernbench-16: (make -j N vmlinux, where N = 16 x num_cpus)
Elapsed User System CPU
gcc2.95 47.45 568.02 143.17 1498.17
gcc3.21 71.44 926.45 134.89 1485.33
pft.
next reply other threads:[~2003-02-03 23:03 UTC|newest]
Thread overview: 84+ messages / expand[flat|nested] mbox.gz Atom feed top
2003-02-03 23:05 Martin J. Bligh [this message]
2003-02-03 23:22 ` [Lse-tech] gcc 2.95 vs 3.21 performance Andi Kleen
2003-02-03 23:31 ` Richard B. Johnson
2003-02-04 0:43 ` J.A. Magallon
2003-02-04 13:42 ` Richard B. Johnson
2003-02-04 14:20 ` John Bradford
2003-02-04 6:54 ` Denis Vlasenko
2003-02-04 7:13 ` Martin J. Bligh
2003-02-04 12:25 ` Adrian Bunk
2003-02-04 15:51 ` Martin J. Bligh
2003-02-04 16:27 ` [Lse-tech] " Martin J. Bligh
2003-02-04 17:40 ` Patrick Mansfield
2003-02-04 17:55 ` Martin J. Bligh
2003-02-04 9:54 ` Bryan Andersen
2003-02-04 15:46 ` Martin J. Bligh
2003-02-04 19:09 ` Timothy D. Witham
2003-02-04 19:35 ` John Bradford
2003-02-04 19:44 ` Dave Jones
2003-02-04 20:11 ` John Bradford
2003-02-04 20:20 ` John Bradford
2003-02-04 20:45 ` Herman Oosthuysen
2003-02-04 21:44 ` Timothy D. Witham
2003-02-05 7:15 ` Denis Vlasenko
2003-02-05 10:36 ` Andreas Schwab
2003-02-05 11:41 ` Denis Vlasenko
2003-02-05 12:20 ` Dave Jones
2003-02-05 13:10 ` [Lse-tech] " Dipankar Sarma
2003-02-05 15:30 ` Martin J. Bligh
2003-02-04 21:38 ` Linus Torvalds
2003-02-04 21:54 ` John Bradford
2003-02-04 22:11 ` Linus Torvalds
2003-02-04 23:27 ` Timothy D. Witham
2003-02-04 23:21 ` Larry McVoy
2003-02-04 23:42 ` b_adlakha
2003-02-05 0:19 ` Andy Pfiffer
2003-02-04 23:51 ` Jakob Oestergaard
2003-02-05 1:03 ` Hugo Mills
2003-02-10 22:26 ` Andrea Arcangeli
2003-02-10 23:28 ` J.A. Magallon
2003-02-04 23:51 ` Eli Carter
2003-02-05 0:27 ` Larry McVoy
2003-02-06 20:42 ` Paul Jakma
2003-02-05 3:03 ` Tomas Szepe
2003-02-05 6:03 ` Mark Mielke
2003-02-07 16:09 ` Pavel Machek
2003-02-04 10:57 ` Padraig
2003-02-04 13:11 ` Helge Hafting
2003-02-04 13:29 ` Jörn Engel
2003-02-04 14:05 ` P
2003-02-04 20:36 ` Herman Oosthuysen
2003-02-04 12:20 ` [Lse-tech] " Dave Jones
2003-02-04 15:50 ` Martin J. Bligh
2003-02-10 12:13 ` Momchil Velikov
2003-02-06 15:42 ` gcc -O2 vs gcc -Os performance Martin J. Bligh
2003-02-06 15:51 ` [Lse-tech] " Andi Kleen
2003-02-06 17:48 ` Alan Cox
2003-02-06 17:06 ` Martin J. Bligh
2003-02-06 20:38 ` Martin J. Bligh
2003-02-06 21:32 ` John Bradford
2003-02-06 22:12 ` Linus Torvalds
2003-02-06 22:58 ` Martin J. Bligh
2003-02-06 23:16 ` Linus Torvalds
2003-02-06 23:59 ` Martin J. Bligh
2003-02-06 23:17 ` Roger Larsson
2003-02-06 23:33 ` Martin J. Bligh
[not found] <1044385759.1861.46.camel@localhost.localdomain.suse.lists.linux.kernel>
[not found] ` <200302041935.h14JZ69G002675@darkstar.example.net.suse.lists.linux.kernel>
[not found] ` <b1pbt8$2ll$1@penguin.transmeta.com.suse.lists.linux.kernel>
2003-02-04 22:05 ` gcc 2.95 vs 3.21 performance Andi Kleen
2003-02-04 22:14 ` Linus Torvalds
2003-02-05 10:04 ` Pavel Janík
2003-02-05 20:07 ` Linus Torvalds
2003-02-06 15:00 ` Horst von Brand
2003-02-04 22:59 ` Jeff Muizelaar
2003-02-04 23:12 ` b_adlakha
2003-02-05 8:41 ` Horst von Brand
2003-02-05 19:09 ` Linus Torvalds
2003-02-05 19:22 ` Randy.Dunlap
2003-02-05 19:24 ` John Bradford
2003-02-06 7:02 ` Neil Booth
[not found] ` <courier.3E423112.00007219@softhome.net>
[not found] ` <20030206212218.GA4891@daikokuya.co.uk>
2003-02-07 10:31 ` b_adlakha
2003-02-07 18:46 ` Horst von Brand
2003-02-07 21:49 ` Neil Booth
2003-02-10 2:14 ` Jeff Garzik
2003-02-10 9:19 ` Tomas Szepe
[not found] <120432836@toto.iv>
2003-02-05 2:45 ` Peter Chubb
[not found] <200302052021.h15KLrXv000881@darkstar.example.net>
2003-02-05 20:28 ` b_adlakha
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=336780000.1044313506@flay \
--to=mbligh@aracnet.com \
--cc=linux-kernel@vger.kernel.org \
--cc=lse-tech@lists.sourceforge.net \
/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).