From: Andi Kleen <andi@firstfloor.org>
To: Steven Rostedt <rostedt@goodmis.org>
Cc: Andi Kleen <andi@firstfloor.org>,
Peter Zijlstra <peterz@infradead.org>,
Jiri Olsa <jolsa@redhat.com>,
"Paul E. McKenney" <paulmck@linux.vnet.ibm.com>,
linux-kernel@vger.kernel.org, Ingo Molnar <mingo@kernel.org>,
Josh Triplett <josh@joshtriplett.org>,
Jan Stancek <jstancek@redhat.com>
Subject: Re: [BUG] msr-trace.h:42 suspicious rcu_dereference_check() usage!
Date: Mon, 21 Nov 2016 12:44:45 -0800 [thread overview]
Message-ID: <20161121204445.GX26852@two.firstfloor.org> (raw)
In-Reply-To: <20161121141517.31936dbe@gandalf.local.home>
> > http://halobates.de/tracepoint-trace
>
> There's a lot of push and pop regs due to function calling. There's
> places that inlines can still improve things, and perhaps even some
> likely unlikelys well placed.
Assuming you avoid all the push/pop and all the call/ret this would only be
~25% of the total instructions. There is just far too much logic and
computation in there.
% awk ' { print $5 } ' tracepoint-trace | sort | uniq -c | sort -rn
222 mov
57 push
57 pop
35 test
34 cmp
32 and
28 jz
25 jnz
21 ret
20 call
16 lea
11 add
-Andi
next prev parent reply other threads:[~2016-11-21 20:44 UTC|newest]
Thread overview: 48+ messages / expand[flat|nested] mbox.gz Atom feed top
2016-11-21 0:53 [BUG] msr-trace.h:42 suspicious rcu_dereference_check() usage! Jiri Olsa
2016-11-21 9:02 ` Paul E. McKenney
2016-11-21 9:43 ` Peter Zijlstra
2016-11-21 11:00 ` Paul E. McKenney
2016-11-21 9:28 ` Peter Zijlstra
2016-11-21 9:34 ` Jiri Olsa
2016-11-21 9:42 ` Peter Zijlstra
2016-11-21 11:22 ` Jiri Olsa
2016-11-21 11:31 ` Peter Zijlstra
2016-11-21 12:49 ` Jiri Olsa
2016-11-21 12:58 ` Peter Zijlstra
2016-11-21 14:15 ` Steven Rostedt
2016-11-21 14:37 ` Peter Zijlstra
2016-11-21 15:35 ` Borislav Petkov
2016-11-21 15:41 ` Peter Zijlstra
2016-11-21 16:06 ` Borislav Petkov
2016-11-29 13:16 ` Borislav Petkov
2016-11-29 13:59 ` Thomas Gleixner
2016-11-30 8:48 ` Borislav Petkov
2016-11-30 8:54 ` Thomas Gleixner
2016-11-30 9:07 ` Borislav Petkov
2016-11-30 9:14 ` Thomas Gleixner
2016-11-29 14:04 ` Jiri Olsa
2016-11-21 14:20 ` Steven Rostedt
2016-11-21 17:06 ` Andi Kleen
2016-11-21 17:18 ` Peter Zijlstra
2016-11-21 17:45 ` Andi Kleen
2016-11-21 18:01 ` Steven Rostedt
2016-11-21 18:06 ` Andi Kleen
2016-11-21 18:22 ` Steven Rostedt
2016-11-21 18:37 ` Andi Kleen
2016-11-21 19:06 ` Steven Rostedt
2016-11-21 19:15 ` Steven Rostedt
2016-11-21 20:44 ` Andi Kleen [this message]
2016-11-22 8:19 ` Paul E. McKenney
2016-11-22 14:39 ` Steven Rostedt
2016-11-22 19:05 ` Andi Kleen
2016-11-24 2:06 ` Steven Rostedt
2016-11-28 21:48 ` Andi Kleen
2016-11-28 22:46 ` Steven Rostedt
2016-11-28 23:10 ` Andi Kleen
2016-11-21 17:55 ` Steven Rostedt
2016-11-21 18:24 ` Steven Rostedt
2016-11-21 20:12 ` Paul E. McKenney
2017-02-23 12:24 ` Jiri Olsa
2017-02-23 17:11 ` Andi Kleen
2017-02-23 17:28 ` Peter Zijlstra
2017-02-23 17:56 ` Borislav Petkov
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=20161121204445.GX26852@two.firstfloor.org \
--to=andi@firstfloor.org \
--cc=jolsa@redhat.com \
--cc=josh@joshtriplett.org \
--cc=jstancek@redhat.com \
--cc=linux-kernel@vger.kernel.org \
--cc=mingo@kernel.org \
--cc=paulmck@linux.vnet.ibm.com \
--cc=peterz@infradead.org \
--cc=rostedt@goodmis.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 NNTP newsgroup(s).