From: Peter Zijlstra <peterz@infradead.org>
To: Ingo Molnar <mingo@elte.hu>
Cc: Frederic Weisbecker <fweisbec@gmail.com>,
Jason Baron <jbaron@redhat.com>,
linux-kernel@vger.kernel.org, David Miller <davem@davemloft.net>,
Mike Galbraith <efault@gmx.de>
Subject: Re: [RFC][PATCH 6/7] perf: use jump_label to optimize the scheduler hooks
Date: Sun, 17 Oct 2010 12:16:16 +0200 [thread overview]
Message-ID: <1287310576.1998.151.camel@laptop> (raw)
In-Reply-To: <1287309159.1998.146.camel@laptop>
On Sun, 2010-10-17 at 11:52 +0200, Peter Zijlstra wrote:
> On Thu, 2010-10-14 at 22:34 +0200, Peter Zijlstra wrote:
> > +static inline void perf_event_task_sched_in(struct task_struct *task)
> > +{
> > + JUMP_LABEL(&perf_task_events, have_events);
> > + return;
> > +
> > +have_events:
> > + __perf_event_task_sched_in(task);
> > +}
>
> OK, so I hate the JUMP_LABEL() interface for it means we have to keep
> writing these silly stubs.
The below seems to actually compile too, awesome! ;-)
---
Index: linux-2.6/include/linux/jump_label.h
===================================================================
--- linux-2.6.orig/include/linux/jump_label.h
+++ linux-2.6/include/linux/jump_label.h
@@ -61,4 +61,14 @@ static inline int jump_label_text_reserv
#endif
+#define COND_STMT(key, stmt) \
+do { \
+ __label__ jl_enabled; \
+ JUMP_LABEL(key, jl_enabled); \
+ if (0) { \
+jl_enabled: \
+ stmt; \
+ } \
+} while (0)
+
#endif
Index: linux-2.6/include/linux/perf_event.h
===================================================================
--- linux-2.6.orig/include/linux/perf_event.h
+++ linux-2.6/include/linux/perf_event.h
@@ -901,21 +901,13 @@ extern atomic_t perf_task_events;
static inline void perf_event_task_sched_in(struct task_struct *task)
{
- JUMP_LABEL(&perf_task_events, have_events);
- return;
-
-have_events:
- __perf_event_task_sched_in(task);
+ COND_STMT(&perf_task_events, __perf_event_task_sched_in(task));
}
static inline
void perf_event_task_sched_out(struct task_struct *task, struct task_struct *next)
{
- JUMP_LABEL(&perf_task_events, have_events);
- return;
-
-have_events:
- __perf_event_task_sched_out(task, next);
+ COND_STMT(&perf_task_events, __perf_event_task_sched_out(task, next));
}
extern int perf_event_init_task(struct task_struct *child);
next prev parent reply other threads:[~2010-10-17 10:16 UTC|newest]
Thread overview: 32+ messages / expand[flat|nested] mbox.gz Atom feed top
2010-10-14 20:34 [RFC][PATCH 0/7] perf and jump_label bits Peter Zijlstra
2010-10-14 20:34 ` [RFC][PATCH 1/7] perf: Fix task refcount issues Peter Zijlstra
2010-10-15 18:14 ` Frederic Weisbecker
2010-10-15 20:02 ` Matt Helsley
2010-10-18 19:19 ` [tip:perf/core] perf: Fix task refcount bugs tip-bot for Peter Zijlstra
2010-10-14 20:34 ` [RFC][PATCH 2/7] perf: Find task before event alloc Peter Zijlstra
2010-10-18 19:20 ` [tip:perf/core] " tip-bot for Peter Zijlstra
2010-10-14 20:34 ` [RFC][PATCH 3/7] perf, hw_breakpoint: Fix crash in hw_breakpoint creation Peter Zijlstra
2010-10-15 13:47 ` Frederic Weisbecker
2010-10-15 13:52 ` Peter Zijlstra
2010-10-18 19:20 ` [tip:perf/core] " tip-bot for Peter Zijlstra
2010-10-14 20:34 ` [RFC][PATCH 4/7] jump_label: More consitent naming Peter Zijlstra
2010-10-18 19:21 ` [tip:perf/core] jump_label: Use more consistent naming tip-bot for Peter Zijlstra
2010-10-14 20:34 ` [RFC][PATCH 5/7] jump_label: atomic_t interface Peter Zijlstra
2010-10-15 14:01 ` Frederic Weisbecker
2010-10-18 19:21 ` [tip:perf/core] jump_label: Add " tip-bot for Peter Zijlstra
2010-10-18 19:21 ` [tip:perf/core] perf: Use jump_labels to optimize the scheduler hooks tip-bot for Peter Zijlstra
2010-10-14 20:34 ` [RFC][PATCH 6/7] perf: use jump_label " Peter Zijlstra
2010-10-15 13:59 ` Frederic Weisbecker
2010-10-17 9:52 ` Peter Zijlstra
2010-10-17 10:16 ` Peter Zijlstra [this message]
2010-10-14 20:34 ` [RFC][PATCH 7/7] perf: Optimize sw events Peter Zijlstra
2010-10-15 9:14 ` Peter Zijlstra
2010-10-15 14:18 ` Jason Baron
2010-10-15 14:57 ` Peter Zijlstra
2010-10-15 15:02 ` Jason Baron
2010-10-15 19:32 ` Steven Rostedt
2010-10-15 19:54 ` Peter Zijlstra
2010-10-16 6:27 ` Ingo Molnar
2010-10-15 14:04 ` Frederic Weisbecker
2010-10-15 14:08 ` Peter Zijlstra
2010-10-15 14:11 ` Frederic Weisbecker
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=1287310576.1998.151.camel@laptop \
--to=peterz@infradead.org \
--cc=davem@davemloft.net \
--cc=efault@gmx.de \
--cc=fweisbec@gmail.com \
--cc=jbaron@redhat.com \
--cc=linux-kernel@vger.kernel.org \
--cc=mingo@elte.hu \
/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