From: Andrew Morton <akpm@osdl.org>
To: Ingo Molnar <mingo@elte.hu>
Cc: geoff@linux.jf.intel.com, linux-kernel@vger.kernel.org,
kenneth.w.chen@intel.com
Subject: Re: [RFC] [PATCH] Performance of del_timer_sync
Date: Tue, 11 May 2004 13:11:37 -0700 [thread overview]
Message-ID: <20040511131137.2390ffa8.akpm@osdl.org> (raw)
In-Reply-To: <20040511195856.GA4958@elte.hu>
Ingo Molnar <mingo@elte.hu> wrote:
>
>
> * Andrew Morton <akpm@osdl.org> wrote:
>
> > Ingo, why is this not sufficient?
> >
> > @@ -331,6 +331,8 @@ int del_timer_sync(struct timer_list *ti
> >
> > del_again:
> > ret += del_timer(timer);
> > + if (!ret)
> > + return 0;
> >
> > for_each_cpu(i) {
> > base = &per_cpu(tvec_bases, i);
>
> it's not sufficient because a timer might be running on another CPU even
> if it has not been deleted. We delete a timer prior running it (so that
> the timer fn can re-activate the timer). So del_timer_sync() has to
> synchronize independently of whether the timer was removed or not.
>
Ah, OK, the timer handler may re-add itself. Really, that's a bug in the
caller: once they've decided to shoot down the timer the caller should have
made state changes which prevent the handler from re-adding the timer.
Still, too late to change that.
Neither AIO nor schedule_timeout() actually re-add the timer so they don't
need the full treatment, yes?
25-akpm/fs/aio.c | 2 +-
25-akpm/include/linux/timer.h | 2 ++
25-akpm/kernel/timer.c | 9 +++++++--
3 files changed, 10 insertions(+), 3 deletions(-)
diff -puN kernel/timer.c~del-timer-speedup kernel/timer.c
--- 25/kernel/timer.c~del-timer-speedup 2004-05-11 13:05:41.997859088 -0700
+++ 25-akpm/kernel/timer.c 2004-05-11 13:07:32.493061264 -0700
@@ -348,8 +348,13 @@ del_again:
return ret;
}
-
EXPORT_SYMBOL(del_timer_sync);
+
+int del_single_shot_timer(struct timer_struct *timer)
+{
+ if (del_timer(timer))
+ del_timer_sync(timer);
+}
#endif
static int cascade(tvec_base_t *base, tvec_t *tv, int index)
@@ -1109,7 +1114,7 @@ fastcall signed long __sched schedule_ti
add_timer(&timer);
schedule();
- del_timer_sync(&timer);
+ del_single_shot_timer(&timer);
timeout = expire - jiffies;
diff -puN include/linux/timer.h~del-timer-speedup include/linux/timer.h
--- 25/include/linux/timer.h~del-timer-speedup 2004-05-11 13:05:42.012856808 -0700
+++ 25-akpm/include/linux/timer.h 2004-05-11 13:07:01.905711248 -0700
@@ -88,8 +88,10 @@ static inline void add_timer(struct time
#ifdef CONFIG_SMP
extern int del_timer_sync(struct timer_list * timer);
+ extern int del_single_shot_timer(struct timer_struct *timer);
#else
# define del_timer_sync(t) del_timer(t)
+# define del_single_shot_timer(t) del_timer(t)
#endif
extern void init_timers(void);
diff -puN fs/aio.c~del-timer-speedup fs/aio.c
--- 25/fs/aio.c~del-timer-speedup 2004-05-11 13:05:42.028854376 -0700
+++ 25-akpm/fs/aio.c 2004-05-11 13:07:14.591782672 -0700
@@ -788,7 +788,7 @@ static inline void set_timeout(long star
static inline void clear_timeout(struct timeout *to)
{
- del_timer_sync(&to->timer);
+ del_single_shot_timer(&to->timer);
}
static int read_events(struct kioctx *ctx,
_
next prev parent reply other threads:[~2004-05-11 20:12 UTC|newest]
Thread overview: 22+ messages / expand[flat|nested] mbox.gz Atom feed top
2004-05-10 22:16 Geoff Gustafson
2004-05-11 7:45 ` Andrew Morton
2004-05-11 18:36 ` Geoff Gustafson
2004-05-11 19:11 ` Andrew Morton
2004-05-11 19:23 ` Chen, Kenneth W
2004-05-11 19:28 ` Andrew Morton
2004-05-11 19:30 ` Andrew Morton
2004-05-11 19:58 ` Ingo Molnar
2004-05-11 20:11 ` Andrew Morton [this message]
2004-05-11 20:19 ` Christoph Hellwig
2004-05-11 20:26 ` Andrew Morton
2004-05-11 20:32 ` Ingo Molnar
2004-05-11 20:55 ` Andrew Morton
2004-05-11 20:57 ` Ingo Molnar
2004-05-11 21:01 ` Chen, Kenneth W
2004-05-11 20:26 ` Ingo Molnar
2004-05-11 20:27 ` Chen, Kenneth W
2004-05-11 20:30 ` Andrew Morton
2004-05-11 20:45 ` Chen, Kenneth W
2004-05-11 20:38 ` Ingo Molnar
2004-05-11 22:46 Geoff Gustafson
[not found] <40A152A8.4080104@linux.intel.com>
2004-05-11 23:03 ` Andrew Morton
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=20040511131137.2390ffa8.akpm@osdl.org \
--to=akpm@osdl.org \
--cc=geoff@linux.jf.intel.com \
--cc=kenneth.w.chen@intel.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
all inboxes | Powered by JetHome®