From: Steven Rostedt <rostedt@goodmis.org>
To: Ingo Molnar <mingo@elte.hu>
Cc: "Paul E. McKenney" <paulmck@us.ibm.com>,
Karsten Wiese <annabellesgarden@yahoo.de>,
dwalker@mvista.com, george anzinger <george@mvista.com>,
Adrian Bunk <bunk@stusta.de>,
Sven-Thorsten Dietrich <sven@mvista.com>,
LKML <linux-kernel@vger.kernel.org>,
Thomas Gleixner <tglx@linutronix.de>
Subject: Re: [RFC] RT-patch update to remove the global pi_lock
Date: Thu, 25 Aug 2005 22:23:30 -0400 [thread overview]
Message-ID: <1125023010.5365.4.camel@localhost.localdomain> (raw)
In-Reply-To: <1125000563.6264.10.camel@localhost.localdomain>
On Thu, 2005-08-25 at 16:09 -0400, Steven Rostedt wrote:
> A word of caution (aka. disclaimer). This is still new. I still expect
> there are some cases in the code that was missed and can cause a dead
> lock or other bad side effect. Hopefully, we can iron these all out.
> Also, I noticed that since the task takes it's own pi_lock for most of
> the code, if something locks up and a NMI goes off, the down_trylock in
> printk will also lock when it tries to take it's own pi_lock.
OK, found my first bug :-)
Just so everyone knows. In rt.c, all pi_waiter access (reading or
writing) must be protected by the task's pi_lock, and all access to the
lock's wait_list must be protected by the lock's wait_lock. The magic
is in the locking order :-).
-- Steve
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
Index: linux_realtime_goliath/kernel/rt.c
===================================================================
--- linux_realtime_goliath/kernel/rt.c (revision 306)
+++ linux_realtime_goliath/kernel/rt.c (working copy)
@@ -671,6 +671,7 @@
struct rt_mutex_waiter *w;
struct plist *curr1;
+ __raw_spin_lock(old_owner->task->pi_lock);
TRACE_WARN_ON_LOCKED(plist_empty(&waiter->pi_list));
TRACE_WARN_ON_LOCKED(lock_owner(lock));
@@ -681,6 +682,7 @@
}
TRACE_WARN_ON_LOCKED(1);
ok:
+ __raw_spin_unlock(old_owner->task->pi_lock);
return;
}
@@ -734,6 +736,8 @@
if (old_owner == new_owner)
return;
+ TRACE_BUG_ON_LOCKED(!spin_is_locked(&old_owner->task->pi_lock));
+ TRACE_BUG_ON_LOCKED(!spin_is_locked(&new_owner->task->pi_lock));
plist_for_each_safe(curr1, next1, &old_owner->task->pi_waiters) {
w = plist_entry(curr1, struct rt_mutex_waiter, pi_list);
if (w->lock == lock) {
@@ -932,6 +936,8 @@
/*
* Add SCHED_NORMAL tasks to the end of the waitqueue (FIFO):
*/
+ TRACE_BUG_ON_LOCKED(!spin_is_locked(&task->pi_lock));
+ TRACE_BUG_ON_LOCKED(!spin_is_locked(&lock->wait_lock));
#ifndef ALL_TASKS_PI
if (!rt_task(task)) {
plist_add(&waiter->list, &lock->wait_list);
@@ -939,6 +945,7 @@
return;
}
#endif
+ __raw_spin_lock(&lock_owner(lock)->task->pi_lock);
plist_add(&waiter->pi_list, &lock_owner(lock)->task->pi_waiters);
/*
* Add RT tasks to the head:
@@ -949,11 +956,9 @@
* If the waiter has higher priority than the owner
* then temporarily boost the owner:
*/
- if (task->prio < lock_owner(lock)->task->prio) {
- __raw_spin_lock(&lock_owner(lock)->task->pi_lock);
+ if (task->prio < lock_owner(lock)->task->prio)
pi_setprio(lock, lock_owner(lock)->task, task->prio);
- __raw_spin_unlock(&lock_owner(lock)->task->pi_lock);
- }
+ __raw_spin_unlock(&lock_owner(lock)->task->pi_lock);
}
/*
next prev parent reply other threads:[~2005-08-26 2:24 UTC|newest]
Thread overview: 76+ messages / expand[flat|nested] mbox.gz Atom feed top
2005-08-16 12:18 2.6.13-rc6-rt3 Ingo Molnar
2005-08-16 15:31 ` 2.6.13-rc6-rt5 Steven Rostedt
2005-08-16 15:44 ` 2.6.13-rc6-rt5 Steven Rostedt
2005-08-16 16:08 ` 2.6.13-rc6-rt5 Steven Rostedt
2005-08-16 16:16 ` 2.6.13-rc6-rt5 Ingo Molnar
2005-08-16 16:22 ` 2.6.13-rc6-rt5 Ingo Molnar
2005-08-16 16:32 ` 2.6.13-rc6-rt5 Ingo Molnar
2005-08-16 16:37 ` 2.6.13-rc6-rt5 Ingo Molnar
2005-08-16 16:52 ` 2.6.13-rc6-rt5 Ingo Molnar
2005-08-16 17:08 ` 2.6.13-rc6-rt6 Ingo Molnar
2005-08-16 17:50 ` 2.6.13-rc6-rt6 Steven Rostedt
2005-08-16 18:07 ` 2.6.13-rc6-rt6 Steven Rostedt
2005-08-16 18:50 ` 2.6.13-rc6-rt6 Steven Rostedt
2005-08-17 4:20 ` 2.6.13-rc6-rt6 Steven Rostedt
2005-08-17 5:46 ` 2.6.13-rc6-rt6 Steven Rostedt
2005-08-17 6:47 ` 2.6.13-rc6-rt6 Ingo Molnar
2005-08-17 14:05 ` 2.6.13-rc6-rt6 Steven Rostedt
2005-08-17 14:24 ` 2.6.13-rc6-rt6 Steven Rostedt
2005-08-17 16:13 ` 2.6.13-rc6-rt6 Steven Rostedt
2005-08-17 16:23 ` 2.6.13-rc6-rt6 Ingo Molnar
2005-08-17 17:10 ` 2.6.13-rc6-rt6 K.R. Foley
2005-08-17 18:31 ` 2.6.13-rc6-rt6 Steven Rostedt
2005-08-17 19:31 ` 2.6.13-rc6-rt6 Steven Rostedt
2005-08-18 0:02 ` 2.6.13-rc6-rt6 Steven Rostedt
2005-08-18 2:44 ` 2.6.13-rc6-rt6 Steven Rostedt
[not found] ` <20050822075012.GB19386@elte.hu>
[not found] ` <1124704837.5208.22.camel@localhost.localdomain>
[not found] ` <20050822101632.GA28803@elte.hu>
[not found] ` <1124710309.5208.30.camel@localhost.localdomain>
[not found] ` <20050822113858.GA1160@elte.hu>
[not found] ` <1124715755.5647.4.camel@localhost.localdomain>
[not found] ` <20050822183355.GB13888@elte.hu>
2005-08-22 19:40 ` 2.6.13-rc6-rt6 Steven Rostedt
2005-08-22 19:44 ` [RFC] RT-patch update to remove the global pi_lock Steven Rostedt
2005-08-22 22:19 ` Daniel Walker
2005-08-23 0:26 ` Steven Rostedt
2005-08-23 0:51 ` Daniel Walker
2005-08-23 1:32 ` Steven Rostedt
2005-08-23 3:38 ` Steven Rostedt
[not found] ` <1124908080.5604.22.camel@localhost.localdomain>
[not found] ` <1124917003.5711.8.camel@localhost.localdomain>
2005-08-24 21:05 ` Thomas Gleixner
2005-08-25 1:13 ` Steven Rostedt
2005-08-25 1:38 ` Daniel Walker
2005-08-25 1:48 ` Steven Rostedt
2005-08-25 6:31 ` Ingo Molnar
2005-08-25 6:35 ` Ingo Molnar
2005-08-25 16:15 ` Steven Rostedt
2005-08-25 19:34 ` Ingo Molnar
2005-08-25 19:46 ` Steven Rostedt
2005-08-23 5:29 ` Ingo Molnar
2005-08-25 14:47 ` Steven Rostedt
2005-08-25 15:06 ` Steven Rostedt
2005-08-25 17:47 ` Ingo Molnar
2005-08-25 20:09 ` Steven Rostedt
2005-08-25 21:32 ` Daniel Walker
2005-08-26 2:23 ` Steven Rostedt [this message]
2005-08-26 13:52 ` Steven Rostedt
2005-08-30 15:00 ` Steven Rostedt
2005-08-30 15:52 ` Steven Rostedt
2005-08-30 23:08 ` Steven Rostedt
2005-08-31 15:01 ` [FYI] 2.6.13-rt3 and a nanosleep jitter test Steven Rostedt
2005-08-31 15:12 ` Daniel Walker
2005-08-31 15:30 ` Steven Rostedt
2005-08-31 15:13 ` Daniel Walker
2005-08-31 15:19 ` Steven Rostedt
2005-08-31 15:30 ` Daniel Walker
2005-08-23 5:46 ` 2.6.13-rc6-rt6 Ingo Molnar
2005-08-19 21:22 ` 2.6.13-rc6-rt6 Steven Rostedt
2005-08-19 22:47 ` 2.6.13-rc6-rt6 Paul E. McKenney
2005-08-19 23:02 ` 2.6.13-rc6-rt6 Steven Rostedt
2005-08-19 23:12 ` 2.6.13-rc6-rt6 Paul E. McKenney
2005-08-19 23:20 ` 2.6.13-rc6-rt6 Steven Rostedt
2005-08-19 23:44 ` 2.6.13-rc6-rt6 Paul E. McKenney
2005-08-22 7:53 ` 2.6.13-rc6-rt6 Ingo Molnar
2005-08-17 19:27 ` 2.6.13-rc6-rt6 Ingo Molnar
2005-08-17 19:39 ` 2.6.13-rc6-rt6 Steven Rostedt
2005-08-17 17:32 ` 2.6.13-rc6-rt6 Steven Rostedt
2005-08-17 19:34 ` 2.6.13-rc6-rt6 Steven Rostedt
2005-08-17 5:59 ` 2.6.13-rc6-rt6 Ingo Molnar
2005-08-17 20:01 ` 2.6.13-rc6-rt8 Peter Bortas
2005-08-23 6:14 ` 2.6.13-rc6-rt8 Ingo Molnar
2005-08-28 20:36 ` 2.6.13-rc6-rt8 Peter Bortas
2005-08-18 9:57 ` 2.6.13-rc6-rt3 Alistair John Strachan
2005-08-18 10:00 ` 2.6.13-rc6-rt3 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=1125023010.5365.4.camel@localhost.localdomain \
--to=rostedt@goodmis.org \
--cc=annabellesgarden@yahoo.de \
--cc=bunk@stusta.de \
--cc=dwalker@mvista.com \
--cc=george@mvista.com \
--cc=linux-kernel@vger.kernel.org \
--cc=mingo@elte.hu \
--cc=paulmck@us.ibm.com \
--cc=sven@mvista.com \
--cc=tglx@linutronix.de \
/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