* [PATCH 1/1] fs-writeback: Using spin_lock to check for work_list empty
@ 2011-08-31 5:11 Rajan Aggarwal
2011-08-31 6:46 ` Rajan Aggarwal
` (3 more replies)
0 siblings, 4 replies; 6+ messages in thread
From: Rajan Aggarwal @ 2011-08-31 5:11 UTC (permalink / raw)
To: linux-fsdevel; +Cc: linux-kernel, Rajan Aggarwal
The bdi_writeback_thread function does not use spin_lock to
see if the work_list is empty.
If the list is not empty, and if an interrupt happens before we
set the current->state to TASK_RUNNING then we could be stuck in
a schedule() due to kernel preemption.
This patch acquires and releases the wb_lock to avoid this scenario.
Signed-off-by: Rajan Aggarwal <rajan.aggarwal85@gmail.com>
---
fs/fs-writeback.c | 3 +++
1 files changed, 3 insertions(+), 0 deletions(-)
diff --git a/fs/fs-writeback.c b/fs/fs-writeback.c
index 04cf3b9..e333898 100644
--- a/fs/fs-writeback.c
+++ b/fs/fs-writeback.c
@@ -936,11 +936,14 @@ int bdi_writeback_thread(void *data)
if (pages_written)
wb->last_active = jiffies;
+ spin_lock_bh(&bdi->wb_lock);
set_current_state(TASK_INTERRUPTIBLE);
if (!list_empty(&bdi->work_list) || kthread_should_stop()) {
__set_current_state(TASK_RUNNING);
+ spin_unlock_bh(&bdi->wb_lock);
continue;
}
+ spin_unlock_bh(&bdi->wb_lock);
if (wb_has_dirty_io(wb) && dirty_writeback_interval)
schedule_timeout(msecs_to_jiffies(dirty_writeback_interval * 10));
--
1.7.4.1
^ permalink raw reply related [flat|nested] 6+ messages in thread
* Re: [PATCH 1/1] fs-writeback: Using spin_lock to check for work_list empty
2011-08-31 5:11 [PATCH 1/1] fs-writeback: Using spin_lock to check for work_list empty Rajan Aggarwal
@ 2011-08-31 6:46 ` Rajan Aggarwal
2011-08-31 6:51 ` Rajan Aggarwal
2011-08-31 21:27 ` Andrew Morton
` (2 subsequent siblings)
3 siblings, 1 reply; 6+ messages in thread
From: Rajan Aggarwal @ 2011-08-31 6:46 UTC (permalink / raw)
To: KAMEZAWA Hiroyuk; +Cc: linux-kernel, linux-fsdevel
Hi Kamezawa,
I noticed that you are responding to emails right now.
Can you please review the patch below and tell me whether it is
technically correct ?
Or, can you please suggest a suitable change for solving this ?
Thanks.
Kautuk
On Wed, Aug 31, 2011 at 10:41 AM, Rajan Aggarwal
<rajan.aggarwal85@gmail.com> wrote:
> The bdi_writeback_thread function does not use spin_lock to
> see if the work_list is empty.
>
> If the list is not empty, and if an interrupt happens before we
> set the current->state to TASK_RUNNING then we could be stuck in
> a schedule() due to kernel preemption.
>
> This patch acquires and releases the wb_lock to avoid this scenario.
>
> Signed-off-by: Rajan Aggarwal <rajan.aggarwal85@gmail.com>
> ---
> fs/fs-writeback.c | 3 +++
> 1 files changed, 3 insertions(+), 0 deletions(-)
>
> diff --git a/fs/fs-writeback.c b/fs/fs-writeback.c
> index 04cf3b9..e333898 100644
> --- a/fs/fs-writeback.c
> +++ b/fs/fs-writeback.c
> @@ -936,11 +936,14 @@ int bdi_writeback_thread(void *data)
> if (pages_written)
> wb->last_active = jiffies;
>
> + spin_lock_bh(&bdi->wb_lock);
> set_current_state(TASK_INTERRUPTIBLE);
> if (!list_empty(&bdi->work_list) || kthread_should_stop()) {
> __set_current_state(TASK_RUNNING);
> + spin_unlock_bh(&bdi->wb_lock);
> continue;
> }
> + spin_unlock_bh(&bdi->wb_lock);
>
> if (wb_has_dirty_io(wb) && dirty_writeback_interval)
> schedule_timeout(msecs_to_jiffies(dirty_writeback_interval * 10));
> --
> 1.7.4.1
>
>
--
To unsubscribe from this list: send the line "unsubscribe linux-fsdevel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: [PATCH 1/1] fs-writeback: Using spin_lock to check for work_list empty
2011-08-31 6:46 ` Rajan Aggarwal
@ 2011-08-31 6:51 ` Rajan Aggarwal
0 siblings, 0 replies; 6+ messages in thread
From: Rajan Aggarwal @ 2011-08-31 6:51 UTC (permalink / raw)
To: Andrew Morton; +Cc: linux-kernel, linux-fsdevel
Hi Andrew,
Can you please review the patch below and tell me whether it is
technically correct ?
Or, can you please suggest a suitable change for solving this ?
Thanks.
Rajan.
> On Wed, Aug 31, 2011 at 10:41 AM, Rajan Aggarwal
> <rajan.aggarwal85@gmail.com> wrote:
>> The bdi_writeback_thread function does not use spin_lock to
>> see if the work_list is empty.
>>
>> If the list is not empty, and if an interrupt happens before we
>> set the current->state to TASK_RUNNING then we could be stuck in
>> a schedule() due to kernel preemption.
>>
>> This patch acquires and releases the wb_lock to avoid this scenario.
>>
>> Signed-off-by: Rajan Aggarwal <rajan.aggarwal85@gmail.com>
>> ---
>> fs/fs-writeback.c | 3 +++
>> 1 files changed, 3 insertions(+), 0 deletions(-)
>>
>> diff --git a/fs/fs-writeback.c b/fs/fs-writeback.c
>> index 04cf3b9..e333898 100644
>> --- a/fs/fs-writeback.c
>> +++ b/fs/fs-writeback.c
>> @@ -936,11 +936,14 @@ int bdi_writeback_thread(void *data)
>> if (pages_written)
>> wb->last_active = jiffies;
>>
>> + spin_lock_bh(&bdi->wb_lock);
>> set_current_state(TASK_INTERRUPTIBLE);
>> if (!list_empty(&bdi->work_list) || kthread_should_stop()) {
>> __set_current_state(TASK_RUNNING);
>> + spin_unlock_bh(&bdi->wb_lock);
>> continue;
>> }
>> + spin_unlock_bh(&bdi->wb_lock);
>>
>> if (wb_has_dirty_io(wb) && dirty_writeback_interval)
>> schedule_timeout(msecs_to_jiffies(dirty_writeback_interval * 10));
>> --
>> 1.7.4.1
>>
>>
>
--
To unsubscribe from this list: send the line "unsubscribe linux-fsdevel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: [PATCH 1/1] fs-writeback: Using spin_lock to check for work_list empty
2011-08-31 5:11 [PATCH 1/1] fs-writeback: Using spin_lock to check for work_list empty Rajan Aggarwal
2011-08-31 6:46 ` Rajan Aggarwal
@ 2011-08-31 21:27 ` Andrew Morton
2011-09-01 7:54 ` KAMEZAWA Hiroyuki
2011-09-01 11:56 ` Peter Zijlstra
3 siblings, 0 replies; 6+ messages in thread
From: Andrew Morton @ 2011-08-31 21:27 UTC (permalink / raw)
To: Rajan Aggarwal; +Cc: linux-fsdevel, linux-kernel, Wu Fengguang
On Wed, 31 Aug 2011 10:41:49 +0530
Rajan Aggarwal <rajan.aggarwal85@gmail.com> wrote:
> The bdi_writeback_thread function does not use spin_lock to
> see if the work_list is empty.
>
> If the list is not empty, and if an interrupt happens before we
> set the current->state to TASK_RUNNING then we could be stuck in
> a schedule() due to kernel preemption.
>
> This patch acquires and releases the wb_lock to avoid this scenario.
>
> Signed-off-by: Rajan Aggarwal <rajan.aggarwal85@gmail.com>
> ---
> fs/fs-writeback.c | 3 +++
> 1 files changed, 3 insertions(+), 0 deletions(-)
>
> diff --git a/fs/fs-writeback.c b/fs/fs-writeback.c
> index 04cf3b9..e333898 100644
> --- a/fs/fs-writeback.c
> +++ b/fs/fs-writeback.c
> @@ -936,11 +936,14 @@ int bdi_writeback_thread(void *data)
> if (pages_written)
> wb->last_active = jiffies;
>
> + spin_lock_bh(&bdi->wb_lock);
> set_current_state(TASK_INTERRUPTIBLE);
> if (!list_empty(&bdi->work_list) || kthread_should_stop()) {
> __set_current_state(TASK_RUNNING);
> + spin_unlock_bh(&bdi->wb_lock);
> continue;
> }
> + spin_unlock_bh(&bdi->wb_lock);
>
> if (wb_has_dirty_io(wb) && dirty_writeback_interval)
> schedule_timeout(msecs_to_jiffies(dirty_writeback_interval * 10));
I don't see anything particularly wrong with the current code. If a
task gets preempted while in state TASK_INTERRUPTIBLE then it will
still be in that state when that task resumes running.
There might be some cross-CPU memory ordering issues in that code. If
so, the effects would be:
a) list_empty() falsely thought to return "false": the thread will
do one additional pointless loop and will then sleep.
b) list_empty() falsely thought to return "true": the thread will
prematurely attempt to go to sleep, introducing a teent bit of
additional latency in rare cases. But I think this is a "can't
happen" because of the memory barrier in
set_current_state(TASK_INTERRUPTIBLE): if the task made this mistake
running list_empty() then it will now be in state TASK_RUNNING and
the schedule() calls will fall straight through. I think.
^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: [PATCH 1/1] fs-writeback: Using spin_lock to check for work_list empty
2011-08-31 5:11 [PATCH 1/1] fs-writeback: Using spin_lock to check for work_list empty Rajan Aggarwal
2011-08-31 6:46 ` Rajan Aggarwal
2011-08-31 21:27 ` Andrew Morton
@ 2011-09-01 7:54 ` KAMEZAWA Hiroyuki
2011-09-01 11:56 ` Peter Zijlstra
3 siblings, 0 replies; 6+ messages in thread
From: KAMEZAWA Hiroyuki @ 2011-09-01 7:54 UTC (permalink / raw)
To: Rajan Aggarwal; +Cc: linux-fsdevel, linux-kernel
On Wed, 31 Aug 2011 10:41:49 +0530
Rajan Aggarwal <rajan.aggarwal85@gmail.com> wrote:
> The bdi_writeback_thread function does not use spin_lock to
> see if the work_list is empty.
>
> If the list is not empty, and if an interrupt happens before we
> set the current->state to TASK_RUNNING then we could be stuck in
> a schedule() due to kernel preemption.
>
> This patch acquires and releases the wb_lock to avoid this scenario.
>
> Signed-off-by: Rajan Aggarwal <rajan.aggarwal85@gmail.com>
Hmm, even if it sleeps, bdi_wakeup_flusher() will wake up the thread.
But it seems there is an useful function schedule_timeout_interruptible().
Then, how about this ? Your concern will go away with this ?
==
>From 5558d23b72004a890dc37411aa7515996c8592fa Mon Sep 17 00:00:00 2001
From: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Date: Thu, 1 Sep 2011 17:02:13 +0900
Subject: [PATCH] use schedule_timeout_interruptible() in bdi_writeback_thread
Use schedule_timeout_interruptible() rather than
set_current_state(TASK_INTERRUPTIBLE);
do some work
schedule_timeout()
Signed-off-by: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
---
fs/fs-writeback.c | 22 +++++++++-------------
1 files changed, 9 insertions(+), 13 deletions(-)
diff --git a/fs/fs-writeback.c b/fs/fs-writeback.c
index 04cf3b9..c538bda 100644
--- a/fs/fs-writeback.c
+++ b/fs/fs-writeback.c
@@ -936,22 +936,18 @@ int bdi_writeback_thread(void *data)
if (pages_written)
wb->last_active = jiffies;
- set_current_state(TASK_INTERRUPTIBLE);
- if (!list_empty(&bdi->work_list) || kthread_should_stop()) {
- __set_current_state(TASK_RUNNING);
+ if (!list_empty(&bdi->work_list) || kthread_should_stop())
continue;
- }
+ /*
+ * If we have nothing to do, we can go sleep without any
+ * timeout and save power. When a work is queued or
+ * something is made dirty - we will be woken up.
+ */
+ timeout = MAX_SCHEDULE_TIMEOUT;
if (wb_has_dirty_io(wb) && dirty_writeback_interval)
- schedule_timeout(msecs_to_jiffies(dirty_writeback_interval * 10));
- else {
- /*
- * We have nothing to do, so can go sleep without any
- * timeout and save power. When a work is queued or
- * something is made dirty - we will be woken up.
- */
- schedule();
- }
+ timeout = msecs_to_jiffies(dirty_writeback_interval * 10);
+ schedule_timeout_interruptible(timeout);
try_to_freeze();
}
--
1.7.4.1
^ permalink raw reply related [flat|nested] 6+ messages in thread
* Re: [PATCH 1/1] fs-writeback: Using spin_lock to check for work_list empty
2011-08-31 5:11 [PATCH 1/1] fs-writeback: Using spin_lock to check for work_list empty Rajan Aggarwal
` (2 preceding siblings ...)
2011-09-01 7:54 ` KAMEZAWA Hiroyuki
@ 2011-09-01 11:56 ` Peter Zijlstra
3 siblings, 0 replies; 6+ messages in thread
From: Peter Zijlstra @ 2011-09-01 11:56 UTC (permalink / raw)
To: Rajan Aggarwal; +Cc: linux-fsdevel, linux-kernel
On Wed, 2011-08-31 at 10:41 +0530, Rajan Aggarwal wrote:
>
> If the list is not empty, and if an interrupt happens before we
> set the current->state to TASK_RUNNING then we could be stuck in
> a schedule() due to kernel preemption.
No, look at PREEMPT_ACTIVE use in kernel/sched.c
^ permalink raw reply [flat|nested] 6+ messages in thread
end of thread, other threads:[~2011-09-01 11:56 UTC | newest]
Thread overview: 6+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2011-08-31 5:11 [PATCH 1/1] fs-writeback: Using spin_lock to check for work_list empty Rajan Aggarwal
2011-08-31 6:46 ` Rajan Aggarwal
2011-08-31 6:51 ` Rajan Aggarwal
2011-08-31 21:27 ` Andrew Morton
2011-09-01 7:54 ` KAMEZAWA Hiroyuki
2011-09-01 11:56 ` Peter Zijlstra
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).