Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]

Groups > linux.kernel > #1611719 > unrolled thread

[RFC][PATCHv2 3/8] printk: offload printing from wake_up_klogd_work_func()

Back to article view | Back to linux.kernel

This discussion starts older than the indexed window; earlier articles aren't shown. The article labeled Started by below is the oldest one visible, not the original post.

Contents

  [RFC][PATCHv2 3/8] printk: offload printing from wake_up_klogd_work_func() Sergey Senozhatsky <sergey.senozhatsky@gmail.com> - 2017-03-29 11:30 +0200
    Re: [RFC][PATCHv2 3/8] printk: offload printing from  wake_up_klogd_work_func() Petr Mladek <pmladek@suse.com> - 2017-03-31 17:00 +0200
      Re: [RFC][PATCHv2 3/8] printk: offload printing from  wake_up_klogd_work_func() Sergey Senozhatsky <sergey.senozhatsky@gmail.com> - 2017-04-04 17:20 +0200

#1611719 — [RFC][PATCHv2 3/8] printk: offload printing from wake_up_klogd_work_func()

Offload printing of printk_deferred() messages from IRQ context
to a schedulable printing kthread, when possible (the same way
we do it in vprintk_emit()). Otherwise, console_unlock() can
force the printing CPU to spend unbound amount of time flushing
kernel messages from IRQ context.

Signed-off-by: Sergey Senozhatsky <sergey.senozhatsky@gmail.com>
---
 kernel/printk/printk.c | 12 ++++++++++--
 1 file changed, 10 insertions(+), 2 deletions(-)

diff --git a/kernel/printk/printk.c b/kernel/printk/printk.c
index ab6b3b2a68c6..1927b5cb5cbe 100644
--- a/kernel/printk/printk.c
+++ b/kernel/printk/printk.c
@@ -2741,8 +2741,16 @@ static void wake_up_klogd_work_func(struct irq_work *irq_work)
 		 * If trylock fails, someone else is doing the printing.
 		 * PRINTK_PENDING_OUTPUT bit is cleared by console_unlock().
 		 */
-		if (console_trylock())
-			console_unlock();
+		if (printk_kthread_enabled()) {
+			wake_up_process(printk_kthread);
+		} else {
+			/*
+			 * If trylock fails, someone else is doing the
+			 * printing
+			 */
+			if (console_trylock())
+				console_unlock();
+		}
 	}
 
 	if (test_and_clear_bit(PRINTK_PENDING_WAKEUP, &printk_pending))
-- 
2.12.2

[toc] | [next] | [standalone]

#1614088 — Re: [RFC][PATCHv2 3/8] printk: offload printing from wake_up_klogd_work_func()

On Wed 2017-03-29 18:25:06, Sergey Senozhatsky wrote:
> Offload printing of printk_deferred() messages from IRQ context
> to a schedulable printing kthread, when possible (the same way
> we do it in vprintk_emit()). Otherwise, console_unlock() can
> force the printing CPU to spend unbound amount of time flushing
> kernel messages from IRQ context.
> 
> Signed-off-by: Sergey Senozhatsky <sergey.senozhatsky@gmail.com>
> ---
>  kernel/printk/printk.c | 12 ++++++++++--
>  1 file changed, 10 insertions(+), 2 deletions(-)
> 
> diff --git a/kernel/printk/printk.c b/kernel/printk/printk.c
> index ab6b3b2a68c6..1927b5cb5cbe 100644
> --- a/kernel/printk/printk.c
> +++ b/kernel/printk/printk.c
> @@ -2741,8 +2741,16 @@ static void wake_up_klogd_work_func(struct irq_work *irq_work)
>  		 * If trylock fails, someone else is doing the printing.
>  		 * PRINTK_PENDING_OUTPUT bit is cleared by console_unlock().
>  		 */
> -		if (console_trylock())
> -			console_unlock();
> +		if (printk_kthread_enabled()) {
> +			wake_up_process(printk_kthread);

Note that the relation between printk_kthread_enabled()
and wake_up_process() is racy. The conditions might change
between these two calls. It looks fine here, well almost.

The critical point is in vprintk_emit(). It must use the emergency
mode (call the consoles directly) when it is called from a process
that started the emergency mode.

We could be more relaxed here. IMHO, the only sensitive situation
is if printk_deferred() is used in the emergency context.
We might want to use the emergency mode here as well but
it is not guaranteed. printk_emergency might get cleared
in the meantime.

A solution might be to add one more bit, e.g.
PRINTK_PENDING_EMERGENCY_OUTPUT. We should force the emergency mode
here when it is set. It should be cleared together with the normal
PRINTK_PENDING_OUTPUT.

Or do you think that this is a corner case that we could
ignore for now?

Otherwise, it looks fine to me.

Best Regards,
Petr

[toc] | [prev] | [next] | [standalone]

#1616092 — Re: [RFC][PATCHv2 3/8] printk: offload printing from wake_up_klogd_work_func()

Hi Petr,

sorry for the delay.

On (03/31/17 16:56), Petr Mladek wrote:
[..]
> > diff --git a/kernel/printk/printk.c b/kernel/printk/printk.c
> > index ab6b3b2a68c6..1927b5cb5cbe 100644
> > --- a/kernel/printk/printk.c
> > +++ b/kernel/printk/printk.c
> > @@ -2741,8 +2741,16 @@ static void wake_up_klogd_work_func(struct irq_work *irq_work)
> >  		 * If trylock fails, someone else is doing the printing.
> >  		 * PRINTK_PENDING_OUTPUT bit is cleared by console_unlock().
> >  		 */
> > -		if (console_trylock())
> > -			console_unlock();
> > +		if (printk_kthread_enabled()) {
> > +			wake_up_process(printk_kthread);
> 
> Note that the relation between printk_kthread_enabled()
> and wake_up_process() is racy. The conditions might change
> between these two calls. It looks fine here, well almost.
> 
> The critical point is in vprintk_emit(). It must use the emergency
> mode (call the consoles directly) when it is called from a process
> that started the emergency mode.

hm, we don't guarantee this. printk(), both in threaded and in
emergency modes, can fail to acquire console_sem.

> We could be more relaxed here. IMHO, the only sensitive situation
> is if printk_deferred() is used in the emergency context.
> We might want to use the emergency mode here as well but
> it is not guaranteed.

hm, I don't think any path does

	printk_emergency_begin()
	printk_deferred()
	printk_emergency_end()

and expects logbuf output to be flushed by the time it does
printk_emergency_end(). it's most likely something like this

	printk_emergency_begin()
	printk()
	printk_emergency_end()

the expectations here are more reasonable, but still, no
guarantees are provided (even in non-kthreaded printk mode).

> A solution might be to add one more bit, e.g.
> PRINTK_PENDING_EMERGENCY_OUTPUT. We should force the emergency mode
> here when it is set. It should be cleared together with the normal
> PRINTK_PENDING_OUTPUT.
> 
> Or do you think that this is a corner case that we could
> ignore for now?

hm, I guess we don't really count on irq_work in emergency
situations. but I need more time to think. good questions, Petr.

	-ss

[toc] | [prev] | [standalone]

Back to top | Article view | linux.kernel

csiph-web