From: Ingo Molnar <mingo@elte.hu>
To: David Singleton <dsingleton@mvista.com>
Cc: dino@in.ibm.com, linux-kernel@vger.kernel.org
Subject: Re: PI BUG with -rt13
Date: Fri, 18 Nov 2005 10:29:09 +0100 [thread overview]
Message-ID: <20051118092909.GC4858@elte.hu> (raw)
In-Reply-To: <437D0C59.1060607@mvista.com>
* David Singleton <dsingleton@mvista.com> wrote:
> >I was testing PI support in the -rt tree (-rt13) when I hit the BUG
> >below. I am using the BULL/Montavista glibc patches. However I would
> >think this can be reproduced using just plain FUTEX_WAKE/WAIT_ROBUST
> >APIs as well, though I havent tried. I can send out the test code
> >if anybody is interested. I have attached the .config below.
> >
> >
> If I make the lock in the timer_base_s struct a raw spinlock this
> BUG goes away.
that most likely just papers over the real bug. Given task-reference
count bug i fixed in the robust/PI-futexes code (see the patch below) i
suspect some more races and/or plain incorrect code.
[this patch below also converts the robust/PI-futex code to use RCU
instead of the tasklist_lock - which should remove a major latency
source from the futex code].
Ingo
Index: linux/kernel/rt.c
===================================================================
--- linux.orig/kernel/rt.c
+++ linux/kernel/rt.c
@@ -2939,15 +2939,20 @@ EXPORT_SYMBOL(rt_mutex_owned_by);
* and now own the lock, or negative values for failure, or positive
* values for the amount of time we waited before getting the lock.
*/
-int fastcall down_futex(struct rt_mutex *lock, unsigned long time, pid_t owner_pid)
+int fastcall
+down_futex(struct rt_mutex *lock, unsigned long time, pid_t owner_pid)
{
struct task_struct *owner_task = NULL;
#ifdef CONFIG_DEBUG_DEADLOCKS
unsigned long eip = CALLER_ADDR0;
#endif
- read_lock(&tasklist_lock);
+ int ret;
+
+ rcu_read_lock();
owner_task = find_task_by_pid(owner_pid);
- read_unlock(&tasklist_lock);
+ if (!get_task_struct_rcu(owner_task))
+ owner_task = NULL;
+ rcu_read_unlock();
if (!owner_task)
return -EOWNERDEAD;
@@ -2956,7 +2961,10 @@ int fastcall down_futex(struct rt_mutex
__down_mutex(lock __EIP__);
rt_mutex_set_owner(lock, owner_task->thread_info);
}
- return __down_interruptible(lock, time __EIP__);
+ ret = __down_interruptible(lock, time __EIP__);
+ put_task_struct(owner_task);
+
+ return ret;
}
EXPORT_SYMBOL(down_futex);
next prev parent reply other threads:[~2005-11-18 9:29 UTC|newest]
Thread overview: 23+ messages / expand[flat|nested] mbox.gz Atom feed top
2005-11-17 16:18 PI BUG with -rt13 Dinakar Guniguntala
2005-11-17 16:41 ` Dinakar Guniguntala
2005-11-17 23:03 ` David Singleton
2005-11-18 9:29 ` Ingo Molnar [this message]
2005-11-18 13:21 ` Dinakar Guniguntala
2005-11-18 13:27 ` Ingo Molnar
2005-11-18 17:05 ` david singleton
2005-11-18 17:44 ` Ingo Molnar
2005-11-21 19:48 ` David Singleton
2005-11-21 21:26 ` Ingo Molnar
2005-11-22 2:01 ` david singleton
2005-11-24 14:57 ` Ingo Molnar
2005-11-24 20:26 ` Dinakar Guniguntala
2005-11-24 20:32 ` Dinakar Guniguntala
2005-11-24 22:25 ` david singleton
2005-11-25 4:56 ` david singleton
2005-11-25 7:34 ` Dinakar Guniguntala
2005-11-25 8:04 ` Dinakar Guniguntala
2005-11-25 5:04 ` Ingo Molnar
2005-11-25 13:27 ` Dinakar Guniguntala
2005-11-25 8:03 ` Ingo Molnar
2005-11-22 2:13 ` david singleton
[not found] ` <8D664A17-5B07-11DA-A840-000A959BB91E@mvista.com>
2005-11-22 9:37 ` Dinakar Guniguntala
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=20051118092909.GC4858@elte.hu \
--to=mingo@elte.hu \
--cc=dino@in.ibm.com \
--cc=dsingleton@mvista.com \
--cc=linux-kernel@vger.kernel.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