The Linux Kernel Mailing List
 help / color / mirror / Atom feed
* [PATCH 1/3] genirq: Get rid of unneeded force parameter in irq_finalize_oneshot()
@ 2012-03-19 14:58 Alexander Gordeev
  2012-03-22 10:32 ` Thomas Gleixner
  0 siblings, 1 reply; 4+ messages in thread
From: Alexander Gordeev @ 2012-03-19 14:58 UTC (permalink / raw)
  To: Thomas Gleixner; +Cc: linux-kernel

The only place irq_finalize_oneshot() is called with force parameter set
is the threaded handler error exit path. But IRQTF_RUNTHREAD is dropped
at this point and irq_wake_thread() is not going to reset it again,
since PF_EXITING is set for this thread already. So we will drop the
thread's bit in threads_oneshot anyway and the force parameter is
superfluous.

Also, the orded or checking PF_EXITING and IRQTF_RUNTHREAD flags in
irq_wake_thread() is important now, so escalate it more visibly in the
source code.
---
 kernel/irq/handle.c |   16 ++++++++++------
 kernel/irq/manage.c |   10 +++++-----
 2 files changed, 15 insertions(+), 11 deletions(-)

diff --git a/kernel/irq/handle.c b/kernel/irq/handle.c
index 6ff84e6..bdb1803 100644
--- a/kernel/irq/handle.c
+++ b/kernel/irq/handle.c
@@ -54,14 +54,18 @@ static void warn_no_thread(unsigned int irq, struct irqaction *action)
 static void irq_wake_thread(struct irq_desc *desc, struct irqaction *action)
 {
 	/*
-	 * Wake up the handler thread for this action. In case the
-	 * thread crashed and was killed we just pretend that we
-	 * handled the interrupt. The hardirq handler has disabled the
-	 * device interrupt, so no irq storm is lurking. If the
+	 * In case the thread crashed and was killed we just pretend that
+	 * we handled the interrupt. The hardirq handler has disabled the
+	 * device interrupt, so no irq storm is lurking.
+	 */
+	if (action->thread->flags & PF_EXITING)
+		return;
+
+	/*
+	 * Wake up the handler thread for this action. If the
 	 * RUNTHREAD bit is already set, nothing to do.
 	 */
-	if ((action->thread->flags & PF_EXITING) ||
-	    test_and_set_bit(IRQTF_RUNTHREAD, &action->thread_flags))
+	if (test_and_set_bit(IRQTF_RUNTHREAD, &action->thread_flags))
 		return;
 
 	/*
diff --git a/kernel/irq/manage.c b/kernel/irq/manage.c
index b0ccd1a..bf606a5 100644
--- a/kernel/irq/manage.c
+++ b/kernel/irq/manage.c
@@ -645,7 +645,7 @@ static int irq_wait_for_interrupt(struct irqaction *action)
  * is marked MASKED.
  */
 static void irq_finalize_oneshot(struct irq_desc *desc,
-				 struct irqaction *action, bool force)
+				 struct irqaction *action)
 {
 	if (!(desc->istate & IRQS_ONESHOT))
 		return;
@@ -679,7 +679,7 @@ again:
 	 * we would clear the threads_oneshot bit of this thread which
 	 * was just set.
 	 */
-	if (!force && test_bit(IRQTF_RUNTHREAD, &action->thread_flags))
+	if (test_bit(IRQTF_RUNTHREAD, &action->thread_flags))
 		goto out_unlock;
 
 	desc->threads_oneshot &= ~action->thread_mask;
@@ -739,7 +739,7 @@ irq_forced_thread_fn(struct irq_desc *desc, struct irqaction *action)
 
 	local_bh_disable();
 	ret = action->thread_fn(action->irq, action->dev_id);
-	irq_finalize_oneshot(desc, action, false);
+	irq_finalize_oneshot(desc, action);
 	local_bh_enable();
 	return ret;
 }
@@ -755,7 +755,7 @@ static irqreturn_t irq_thread_fn(struct irq_desc *desc,
 	irqreturn_t ret;
 
 	ret = action->thread_fn(action->irq, action->dev_id);
-	irq_finalize_oneshot(desc, action, false);
+	irq_finalize_oneshot(desc, action);
 	return ret;
 }
 
@@ -844,7 +844,7 @@ void exit_irq_thread(void)
 		wake_threads_waitq(desc);
 
 	/* Prevent a stale desc->threads_oneshot */
-	irq_finalize_oneshot(desc, action, true);
+	irq_finalize_oneshot(desc, action);
 }
 
 static void irq_setup_forced_threading(struct irqaction *new)
-- 
1.7.7.6


^ permalink raw reply related	[flat|nested] 4+ messages in thread

* Re: [PATCH 1/3] genirq: Get rid of unneeded force parameter in irq_finalize_oneshot()
  2012-03-19 14:58 [PATCH 1/3] genirq: Get rid of unneeded force parameter in irq_finalize_oneshot() Alexander Gordeev
@ 2012-03-22 10:32 ` Thomas Gleixner
  2012-03-22 11:06   ` Alexander Gordeev
  0 siblings, 1 reply; 4+ messages in thread
From: Thomas Gleixner @ 2012-03-22 10:32 UTC (permalink / raw)
  To: Alexander Gordeev; +Cc: linux-kernel

On Mon, 19 Mar 2012, Alexander Gordeev wrote:

> The only place irq_finalize_oneshot() is called with force parameter set
> is the threaded handler error exit path. But IRQTF_RUNTHREAD is dropped
> at this point and irq_wake_thread() is not going to reset it again,
> since PF_EXITING is set for this thread already. So we will drop the
> thread's bit in threads_oneshot anyway and the force parameter is
> superfluous.

Wrong.

CPU 0				CPU 1

    do_exit()
				check(PF_EXITING)
				set_bit(IRQTF_RUNTHREAD)
				...
    set(PF_EXITING)
    exit_irq_thread()

    if (test_bit(IRQTF_RUNTHREAD))
	goto out;

	-> FAIL

Thanks,

	tglx

^ permalink raw reply	[flat|nested] 4+ messages in thread

* Re: [PATCH 1/3] genirq: Get rid of unneeded force parameter in irq_finalize_oneshot()
  2012-03-22 10:32 ` Thomas Gleixner
@ 2012-03-22 11:06   ` Alexander Gordeev
  2012-03-22 14:15     ` Thomas Gleixner
  0 siblings, 1 reply; 4+ messages in thread
From: Alexander Gordeev @ 2012-03-22 11:06 UTC (permalink / raw)
  To: Thomas Gleixner; +Cc: linux-kernel

On Thu, Mar 22, 2012 at 11:32:26AM +0100, Thomas Gleixner wrote:
> Wrong.
> 
> CPU 0				CPU 1
> 
>     do_exit()
> 				check(PF_EXITING)
> 				set_bit(IRQTF_RUNTHREAD)
> 				...
>     set(PF_EXITING)
>     exit_irq_thread()

Thomas,

Did not you pick this code with 7140ea1 commit?

@@ -845,6 +842,13 @@ void exit_irq_thread(void)
 
 	desc = irq_to_desc(action->irq);
 
+	/*
+	 * If IRQTF_RUNTHREAD is set, we need to decrement
+	 * desc->threads_active and wake possible waiters.
+	 */
+	if (test_and_clear_bit(IRQTF_RUNTHREAD, &action->thread_flags))
+		wake_threads_waitq(desc);
+
 	/* Prevent a stale desc->threads_oneshot */
 	irq_finalize_oneshot(desc, action, true);
 }

> 
>     if (test_bit(IRQTF_RUNTHREAD))
> 	goto out;
> 
> 	-> FAIL
> 
> Thanks,
> 
> 	tglx

-- 
Regards,
Alexander Gordeev
agordeev@redhat.com

^ permalink raw reply	[flat|nested] 4+ messages in thread

* Re: [PATCH 1/3] genirq: Get rid of unneeded force parameter in irq_finalize_oneshot()
  2012-03-22 11:06   ` Alexander Gordeev
@ 2012-03-22 14:15     ` Thomas Gleixner
  0 siblings, 0 replies; 4+ messages in thread
From: Thomas Gleixner @ 2012-03-22 14:15 UTC (permalink / raw)
  To: Alexander Gordeev; +Cc: linux-kernel

On Thu, 22 Mar 2012, Alexander Gordeev wrote:

> On Thu, Mar 22, 2012 at 11:32:26AM +0100, Thomas Gleixner wrote:
> > Wrong.
> > 
> > CPU 0				CPU 1
> > 
> >     do_exit()
> > 				check(PF_EXITING)
> > 				set_bit(IRQTF_RUNTHREAD)
> > 				...
> >     set(PF_EXITING)
> >     exit_irq_thread()
> 
> Thomas,
> 
> Did not you pick this code with 7140ea1 commit?
> 
> @@ -845,6 +842,13 @@ void exit_irq_thread(void)
>  
>  	desc = irq_to_desc(action->irq);
>  
> +	/*
> +	 * If IRQTF_RUNTHREAD is set, we need to decrement
> +	 * desc->threads_active and wake possible waiters.
> +	 */
> +	if (test_and_clear_bit(IRQTF_RUNTHREAD, &action->thread_flags))
> +		wake_threads_waitq(desc);
> +

Gah crap. Was looking on the wrong branch. That's what you get for
looking at patches before coffee hit brain.

Thanks,

	tglx

^ permalink raw reply	[flat|nested] 4+ messages in thread

end of thread, other threads:[~2012-03-22 14:15 UTC | newest]

Thread overview: 4+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2012-03-19 14:58 [PATCH 1/3] genirq: Get rid of unneeded force parameter in irq_finalize_oneshot() Alexander Gordeev
2012-03-22 10:32 ` Thomas Gleixner
2012-03-22 11:06   ` Alexander Gordeev
2012-03-22 14:15     ` Thomas Gleixner

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox