From: John Kacur on
On Wed, Feb 3, 2010 at 9:09 PM, Steven Rostedt <rostedt(a)goodmis.org> wrote:
> t On Wed, 2010-02-03 at 20:53 +0100, John Kacur wrote:
>> Balance local_irq_disable() and local_irq_enable() as well as
>> spin_lock_irq() and spin_lock_unlock_irq
>>
>> Signed-off-by: John Kacur <jkacur(a)redhat.com>
>> ---
>> �mm/vmscan.c | � �3 ++-
>> �1 files changed, 2 insertions(+), 1 deletions(-)
>>
>> diff --git a/mm/vmscan.c b/mm/vmscan.c
>> index c26986c..b895025 100644
>> --- a/mm/vmscan.c
>> +++ b/mm/vmscan.c
>> @@ -1200,8 +1200,9 @@ static unsigned long shrink_inactive_list(unsigned long max_scan,
>> � � � � � � � if (current_is_kswapd())
>> � � � � � � � � � � � __count_vm_events(KSWAPD_STEAL, nr_freed);
>> � � � � � � � __count_zone_vm_events(PGSTEAL, zone, nr_freed);
>> + � � � � � � local_irq_enable();
>>
>> - � � � � � � spin_lock(&zone->lru_lock);
>> + � � � � � � spin_lock_irq(&zone->lru_lock);
>> � � � � � � � /*
>> � � � � � � � �* Put back any unfreeable pages.
>> � � � � � � � �*/
>
>
> The above looks wrong. I don't know the code, but just by looking at
> where the locking and interrupts are, I can take a guess.
>
> Lets add a little more of the code:
>
> � � � � � � � �local_irq_disable();
> � � � � � � � �if (current_is_kswapd())
> � � � � � � � � � � � �__count_vm_events(KSWAPD_STEAL, nr_freed);
> � � � � � � � �__count_zone_vm_events(PGSTEAL, zone, nr_freed);
>
> � � � � � � � �spin_lock(&zone->lru_lock);
> � � � � � � � �/*
>
> I'm guessing the __count_zone_vm_events and friends need interrupts
> disabled here, probably due to per cpu stuff. But if you enable
> interrupts before the spin_lock() you may let an interrupt come in and
> invalidate what was done above it.
>
> So no, I do not think enabling interrupts here is a good thing.
>

okay, and since we have already done local_irq_disable(), then that is
why we only need the spin_lock() and not the spin_lock_irq() flavour?
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo(a)vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
From: John Kacur on
On Thu, Feb 4, 2010 at 1:22 AM, KOSAKI Motohiro
<kosaki.motohiro(a)jp.fujitsu.com> wrote:
>> On Wed, Feb 3, 2010 at 9:09 PM, Steven Rostedt <rostedt(a)goodmis.org> wrote:
>> > t On Wed, 2010-02-03 at 20:53 +0100, John Kacur wrote:
>> >> Balance local_irq_disable() and local_irq_enable() as well as
>> >> spin_lock_irq() and spin_lock_unlock_irq
>> >>
>> >> Signed-off-by: John Kacur <jkacur(a)redhat.com>
>> >> ---
>> >> �mm/vmscan.c | � �3 ++-
>> >> �1 files changed, 2 insertions(+), 1 deletions(-)
>> >>
>> >> diff --git a/mm/vmscan.c b/mm/vmscan.c
>> >> index c26986c..b895025 100644
>> >> --- a/mm/vmscan.c
>> >> +++ b/mm/vmscan.c
>> >> @@ -1200,8 +1200,9 @@ static unsigned long shrink_inactive_list(unsigned long max_scan,
>> >> � � � � � � � if (current_is_kswapd())
>> >> � � � � � � � � � � � __count_vm_events(KSWAPD_STEAL, nr_freed);
>> >> � � � � � � � __count_zone_vm_events(PGSTEAL, zone, nr_freed);
>> >> + � � � � � � local_irq_enable();
>> >>
>> >> - � � � � � � spin_lock(&zone->lru_lock);
>> >> + � � � � � � spin_lock_irq(&zone->lru_lock);
>> >> � � � � � � � /*
>> >> � � � � � � � �* Put back any unfreeable pages.
>> >> � � � � � � � �*/
>> >
>> >
>> > The above looks wrong. I don't know the code, but just by looking at
>> > where the locking and interrupts are, I can take a guess.
>> >
>> > Lets add a little more of the code:
>> >
>> > � � � � � � � �local_irq_disable();
>> > � � � � � � � �if (current_is_kswapd())
>> > � � � � � � � � � � � �__count_vm_events(KSWAPD_STEAL, nr_freed);
>> > � � � � � � � �__count_zone_vm_events(PGSTEAL, zone, nr_freed);
>> >
>> > � � � � � � � �spin_lock(&zone->lru_lock);
>> > � � � � � � � �/*
>> >
>> > I'm guessing the __count_zone_vm_events and friends need interrupts
>> > disabled here, probably due to per cpu stuff. But if you enable
>> > interrupts before the spin_lock() you may let an interrupt come in and
>> > invalidate what was done above it.
>> >
>> > So no, I do not think enabling interrupts here is a good thing.
>> >
>>
>> okay, and since we have already done local_irq_disable(), then that is
>> why we only need the spin_lock() and not the spin_lock_irq() flavour?
>
> Yes, spin_lock_irq() is equivalent to spin_lock() + irq_disable().
> Now, we already disabled irq. then, we only need spin_lock().
>
> So, I don't think shrink_inactive_list need any fix.
>

Thanks for the explanation!
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo(a)vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/