From: Steven Rostedt <rostedt@goodmis.org>
To: Lai Jiangshan <laijs@cn.fujitsu.com>
Cc: Thomas Gleixner <tglx@linutronix.de>, Ingo Molnar <mingo@elte.hu>,
Peter Zijlstra <a.p.zijlstra@chello.nl>,
Andrew Morton <akpm@linux-foundation.org>,
Dave Young <hidave.darkstar@gmail.com>,
Darren Hart <dvhart@linux.intel.com>,
Namhyung Kim <namhyung@gmail.com>,
LKML <linux-kernel@vger.kernel.org>,
Linus Torvalds <torvalds@linux-foundation.org>
Subject: Re: [PATCH] rtmutex: ensure only the top waiter or higher priority task can take the lock and reduce unrelated boosting
Date: Wed, 15 Dec 2010 16:53:22 -0500 [thread overview]
Message-ID: <1292450002.5015.1903.camel@gandalf.stny.rr.com> (raw)
In-Reply-To: <4D0877D2.10000@cn.fujitsu.com>
On Wed, 2010-12-15 at 16:09 +0800, Lai Jiangshan wrote:
> /*
> @@ -543,11 +491,13 @@ static void remove_waiter(struct rt_mutex *lock,
>
> raw_spin_lock_irqsave(¤t->pi_lock, flags);
> plist_del(&waiter->list_entry, &lock->wait_list);
> - waiter->task = NULL;
> current->pi_blocked_on = NULL;
> raw_spin_unlock_irqrestore(¤t->pi_lock, flags);
>
> - if (first && owner != current) {
> + if (!owner)
> + return;
> +
> + if (first) {
This is a bug. There's a small chance that the mutex timed out, and at
that same time, the owner gave up the lock and woke this task up. Which
means this task is the new owner of the lock iff it was the
rt_mutex_top_waiter().
The fix is to do this:
if (!owner) {
if (first) {
ret = try_to_take_rt_mutex();
BUG_ON(!ret);
}
return first;
}
if (first) {
We need to make remove_waiter return 1 if it took the lock and 0 if it
did not, so it can pass this information back to the caller.
if (unlikely(ret)) {
if (remove_waiter(...))
ret = 0;
}
-- Steve
next prev parent reply other threads:[~2010-12-15 21:53 UTC|newest]
Thread overview: 46+ messages / expand[flat|nested] mbox.gz Atom feed top
2010-12-14 9:04 [PATCH] rtmutex: multiple candidate owners without " Lai Jiangshan
2010-12-14 14:01 ` Steven Rostedt
2010-12-14 16:44 ` Lai Jiangshan
2010-12-14 17:00 ` Steven Rostedt
2010-12-15 4:25 ` Lai Jiangshan
2010-12-14 20:07 ` Thomas Gleixner
2010-12-15 3:41 ` Lai Jiangshan
2010-12-15 4:16 ` Steven Rostedt
2010-12-15 8:09 ` [PATCH] rtmutex: ensure only the top waiter or higher priority task can take the lock and reduce " Lai Jiangshan
2010-12-15 12:07 ` Thomas Gleixner
2010-12-15 14:24 ` Lai Jiangshan
2010-12-15 14:52 ` Thomas Gleixner
2010-12-15 15:01 ` Steven Rostedt
2010-12-16 20:33 ` Darren Hart
2010-12-17 3:10 ` Lai Jiangshan
2010-12-17 3:17 ` Steven Rostedt
2010-12-17 3:30 ` Steven Rostedt
2010-12-15 15:04 ` Steven Rostedt
2010-12-23 9:07 ` Lai Jiangshan
2010-12-23 12:56 ` Steven Rostedt
2010-12-15 21:53 ` Steven Rostedt [this message]
2010-12-16 1:14 ` Lai Jiangshan
2010-12-16 13:56 ` Steven Rostedt
2010-12-16 14:07 ` Steven Rostedt
2011-01-06 14:50 ` Steven Rostedt
2011-01-10 11:37 ` Lai Jiangshan
2011-01-10 12:57 ` Steven Rostedt
2010-12-23 8:49 ` [PATCH V3] rtmutex: ensure only the top waiter or higher priority task can take the lock and remove " Lai Jiangshan
2011-01-12 17:03 ` Steven Rostedt
2011-01-12 17:04 ` Steven Rostedt
2011-01-12 17:05 ` Steven Rostedt
2011-01-14 9:09 ` [PATCH V4] " Lai Jiangshan
2011-01-21 17:34 ` Thomas Gleixner
2011-01-22 14:07 ` Steven Rostedt
2011-01-22 14:09 ` Steven Rostedt
2011-01-31 14:30 ` [tip:core/locking] rtmutex: Simplify PI algorithm and make highest prio task get lock tip-bot for Lai Jiangshan
2010-12-15 7:47 ` [PATCH] rtmutex: multiple candidate owners without unrelated boosting Thomas Gleixner
2010-12-16 20:55 ` Darren Hart
2010-12-23 7:25 ` Lai Jiangshan
2010-12-14 23:16 ` Steven Rostedt
2010-12-15 2:18 ` Steven Rostedt
2010-12-15 8:02 ` Thomas Gleixner
2010-12-15 14:02 ` Steven Rostedt
2010-12-15 14:16 ` Thomas Gleixner
2010-12-15 14:32 ` Steven Rostedt
2010-12-15 14:50 ` Thomas Gleixner
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=1292450002.5015.1903.camel@gandalf.stny.rr.com \
--to=rostedt@goodmis.org \
--cc=a.p.zijlstra@chello.nl \
--cc=akpm@linux-foundation.org \
--cc=dvhart@linux.intel.com \
--cc=hidave.darkstar@gmail.com \
--cc=laijs@cn.fujitsu.com \
--cc=linux-kernel@vger.kernel.org \
--cc=mingo@elte.hu \
--cc=namhyung@gmail.com \
--cc=tglx@linutronix.de \
--cc=torvalds@linux-foundation.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
all inboxes | Powered by JetHome®