public inbox for gcc-bugs@sourceware.org
help / color / mirror / Atom feed
From: "dominiq at lps dot ens.fr" <gcc-bugzilla@gcc.gnu.org>
To: gcc-bugs@gcc.gnu.org
Subject: [Bug fortran/48636] Enable more inlining with -O2 and higher
Date: Sun, 17 Apr 2011 10:23:00 -0000 [thread overview]
Message-ID: <bug-48636-4-h8elOSpi8G@http.gcc.gnu.org/bugzilla/> (raw)
In-Reply-To: <bug-48636-4@http.gcc.gnu.org/bugzilla/>
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=48636
--- Comment #2 from Dominique d'Humieres <dominiq at lps dot ens.fr> 2011-04-17 10:23:03 UTC ---
As shown by the following results it seems that --param max-inline-insns-auto=*
is the way to go.
Date & Time : 17 Apr 2011 11:22:05
Test Name : pbharness
Compile Command : gfc %n.f90 -Ofast -funroll-loops -ftree-loop-linear
-fomit-frame-pointer --param max-inline-insns-auto=400 -fwhole-program -flto
-fstack-arrays -o %n
Benchmarks : ac aermod air capacita channel doduc fatigue gas_dyn induct
linpk mdbx nf protein rnflow test_fpu tfft
Maximum Times : 300.0
Target Error % : 0.200
Minimum Repeats : 2
Maximum Repeats : 5
Benchmark Compile Executable Ave Run Number Estim
Name (secs) (bytes) (secs) Repeats Err %
--------- ------- ---------- ------- ------- ------
ac 8.07 54576 8.11 2 0.0062
aermod 175.22 1472624 18.83 2 0.1647
air 25.65 89992 6.78 5 0.1871
capacita 14.02 109536 40.36 2 0.0483
channel 3.11 34448 2.94 5 0.6012
doduc 29.46 224584 27.44 2 0.0437
fatigue 9.85 77032 2.74 2 0.0365
gas_dyn 26.09 144112 4.68 5 0.6928
induct 24.32 189696 14.24 2 0.1193
linpk 3.13 21536 21.69 2 0.0254
mdbx 9.18 84776 12.55 2 0.0678
nf 34.14 124640 18.38 2 0.1034
protein 28.14 155624 35.48 2 0.0789
rnflow 43.93 204176 26.70 2 0.0262
test_fpu 21.90 141696 11.18 2 0.0045
tfft 1.71 22072 3.29 5 0.1369
Geometric Mean Execution Time = 11.60 seconds
================================================================================
Date & Time : 17 Apr 2011 11:50:20
Test Name : pbharness
Compile Command : gfc %n.f90 -Ofast -funroll-loops -ftree-loop-linear
-fomit-frame-pointer -finline-limit=600 -fwhole-program -flto -fstack-arrays -o
%n
Benchmarks : ac aermod air capacita channel doduc fatigue gas_dyn induct
linpk mdbx nf protein rnflow test_fpu tfft
Maximum Times : 300.0
Target Error % : 0.200
Minimum Repeats : 2
Maximum Repeats : 5
Benchmark Compile Executable Ave Run Number Estim
Name (secs) (bytes) (secs) Repeats Err %
--------- ------- ---------- ------- ------- ------
ac 8.06 54576 8.11 2 0.0062
aermod 175.54 1480632 18.92 2 0.0106
air 25.36 89992 6.76 2 0.0740
capacita 13.95 109536 40.32 2 0.0161
channel 3.13 34448 2.95 5 0.1703
doduc 27.31 212280 27.18 2 0.0331
fatigue 9.82 77032 2.74 2 0.0182
gas_dyn 24.86 144112 4.67 5 0.3052
induct 24.25 189696 14.21 2 0.0035
linpk 2.55 21536 21.69 2 0.0023
mdbx 9.17 84776 12.53 2 0.0239
nf 34.21 124640 18.41 4 0.1634
protein 28.01 155624 35.46 2 0.0310
rnflow 38.11 183696 26.74 2 0.0037
test_fpu 19.63 141720 10.84 2 0.0323
tfft 1.69 22072 3.29 2 0.0152
Geometric Mean Execution Time = 11.57 seconds
================================================================================
next prev parent reply other threads:[~2011-04-17 10:23 UTC|newest]
Thread overview: 46+ messages / expand[flat|nested] mbox.gz Atom feed top
2011-04-15 21:05 [Bug fortran/48636] New: " tkoenig at gcc dot gnu.org
2011-04-16 11:22 ` [Bug fortran/48636] " rguenth at gcc dot gnu.org
2011-04-17 10:23 ` dominiq at lps dot ens.fr [this message]
2011-04-17 10:44 ` hubicka at gcc dot gnu.org
2011-04-17 13:32 ` tkoenig at gcc dot gnu.org
2011-04-17 14:12 ` dominiq at lps dot ens.fr
2011-04-20 11:22 ` jb at gcc dot gnu.org
2011-04-20 12:29 ` burnus at gcc dot gnu.org
2011-04-20 13:10 ` jb at gcc dot gnu.org
2011-04-20 15:42 ` burnus at gcc dot gnu.org
2011-04-20 16:41 ` tkoenig at gcc dot gnu.org
2011-04-20 18:15 ` jb at gcc dot gnu.org
2011-05-04 16:23 ` hubicka at gcc dot gnu.org
2011-05-04 17:31 ` burnus at gcc dot gnu.org
2011-06-04 18:08 ` hubicka at gcc dot gnu.org
2012-07-03 17:44 ` jamborm at gcc dot gnu.org
2012-08-11 10:50 ` jamborm at gcc dot gnu.org
2012-08-21 6:54 ` hubicka at gcc dot gnu.org
2012-08-21 8:15 ` hubicka at gcc dot gnu.org
2012-09-12 21:52 ` hubicka at gcc dot gnu.org
2012-10-16 16:39 ` hubicka at gcc dot gnu.org
2012-10-16 17:58 ` dominiq at lps dot ens.fr
2012-10-16 20:59 ` dominiq at lps dot ens.fr
2012-10-17 12:20 ` jakub at gcc dot gnu.org
2012-10-17 13:13 ` dominiq at lps dot ens.fr
2012-10-17 14:06 ` dominiq at lps dot ens.fr
2012-10-19 8:45 ` vincenzo.innocente at cern dot ch
2012-10-20 10:35 ` hubicka at gcc dot gnu.org
2012-10-20 11:22 ` dominiq at lps dot ens.fr
2012-10-20 12:11 ` tkoenig at gcc dot gnu.org
2012-10-28 10:08 ` hubicka at gcc dot gnu.org
2012-10-28 10:11 ` hubicka at gcc dot gnu.org
2012-10-28 11:27 ` vincenzo.innocente at cern dot ch
2012-11-07 9:34 ` hubicka at gcc dot gnu.org
2012-11-07 11:18 ` hubicka at gcc dot gnu.org
2012-11-08 16:46 ` hubicka at gcc dot gnu.org
2012-11-11 18:15 ` hubicka at gcc dot gnu.org
2012-11-12 12:16 ` hubicka at gcc dot gnu.org
2012-11-12 12:45 ` izamyatin at gmail dot com
2012-11-14 23:22 ` hubicka at gcc dot gnu.org
2012-11-14 23:54 ` Jan Hubicka
2012-11-14 23:55 ` hubicka at ucw dot cz
2012-11-15 2:29 ` hubicka at gcc dot gnu.org
2012-11-16 14:43 ` dominiq at lps dot ens.fr
2013-03-01 17:49 ` wschmidt at gcc dot gnu.org
2013-03-04 17:54 ` wschmidt at gcc dot gnu.org
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=bug-48636-4-h8elOSpi8G@http.gcc.gnu.org/bugzilla/ \
--to=gcc-bugzilla@gcc.gnu.org \
--cc=gcc-bugs@gcc.gnu.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).