public inbox for gcc-bugs@sourceware.org
help / color / mirror / Atom feed
From: "amacleod at redhat dot com" <gcc-bugzilla@gcc.gnu.org>
To: gcc-bugs@gcc.gnu.org
Subject: [Bug libstdc++/51798] [4.7 regression] libstdc++ atomicity performance regression due to __sync_fetch_and_add
Date: Tue, 24 Jan 2012 07:45:00 -0000 [thread overview]
Message-ID: <bug-51798-4-QalcKt5QHG@http.gcc.gnu.org/bugzilla/> (raw)
In-Reply-To: <bug-51798-4@http.gcc.gnu.org/bugzilla/>
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=51798
Andrew Macleod <amacleod at redhat dot com> changed:
What |Removed |Added
----------------------------------------------------------------------------
CC| |bkoz at redhat dot com
--- Comment #4 from Andrew Macleod <amacleod at redhat dot com> 2012-01-24 05:41:43 UTC ---
Looking at the old sync.md and rs6000.c files, it looks to me like
sync_fetch_and_add always issued a lwsync before the operation and an isync
afterwards.
With the new atomics, it looks like making these fetch_and_add operations use
acq_rel mode, we'd get exactly the same code. I suspect it would be safe to
convert those uses in atomicity.h to acq_rel. If acq_rel wasn't sufficient,
power would likely have had a defect somewhere over the years...
At least that would appear to be safe(r) to me on the surface... And it would
return the missing performance? I'd be more hesitant about relaxing the mode
any further than that.
next prev parent reply other threads:[~2012-01-24 5:42 UTC|newest]
Thread overview: 34+ messages / expand[flat|nested] mbox.gz Atom feed top
2012-01-09 14:31 [Bug libstdc++/51798] New: " dje at gcc dot gnu.org
2012-01-09 14:36 ` [Bug libstdc++/51798] " dje at gcc dot gnu.org
2012-01-09 14:38 ` dje at gcc dot gnu.org
2012-01-09 15:43 ` rguenth at gcc dot gnu.org
2012-01-09 17:12 ` dje at gcc dot gnu.org
2012-01-24 7:45 ` amacleod at redhat dot com [this message]
2012-01-24 16:46 ` dje at gcc dot gnu.org
2012-01-24 16:51 ` amacleod at redhat dot com
2012-01-25 14:27 ` jakub at gcc dot gnu.org
2012-01-25 15:39 ` dje at gcc dot gnu.org
2012-01-25 16:09 ` amacleod at redhat dot com
2012-01-25 16:17 ` dje at gcc dot gnu.org
2012-01-26 13:52 ` jakub at gcc dot gnu.org
2012-01-26 14:38 ` amacleod at redhat dot com
2012-01-26 15:50 ` dje at gcc dot gnu.org
2012-01-26 15:53 ` redi at gcc dot gnu.org
2012-01-26 21:50 ` dje at gcc dot gnu.org
2012-01-26 22:08 ` redi at gcc dot gnu.org
2012-01-27 10:09 ` jakub at gcc dot gnu.org
2012-01-27 11:37 ` jakub at gcc dot gnu.org
2012-01-27 13:03 ` amacleod at redhat dot com
2012-01-27 14:55 ` dje at gcc dot gnu.org
2012-01-27 15:01 ` jakub at gcc dot gnu.org
2012-01-27 15:26 ` dje at gcc dot gnu.org
2012-01-27 20:15 ` bkoz at gcc dot gnu.org
2012-01-27 21:05 ` jakub at gcc dot gnu.org
2012-01-27 21:08 ` jakub at gcc dot gnu.org
2012-02-09 8:46 ` bkoz at gcc dot gnu.org
2012-02-09 9:16 ` redi at gcc dot gnu.org
2012-02-09 20:08 ` bkoz at gcc dot gnu.org
2012-02-09 23:22 ` bkoz at gcc dot gnu.org
2012-02-10 18:21 ` bkoz at gcc dot gnu.org
2012-02-13 21:31 ` rth at gcc dot gnu.org
2012-02-17 21:03 ` bkoz 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-51798-4-QalcKt5QHG@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).