[PATCH 1/7] zram: fix race between reset and flushing pending work

[PATCH 1/7] zram: fix race between reset and flushing pending work

[Minchan Kim]

Dan and Sergey reported that there is a racy between reset and
flushing of pending work so that it could make oops by freeing
zram->meta in reset while zram_slot_free can access zram->meta
if new request is adding during the race window.

This patch moves flush after taking init_lock so it prevents
new request so that it closes the race.

Cc: stable <stable@vger.kernel.org>
Signed-off-by: Minchan Kim <minchan@kernel.org>
---
 drivers/block/zram/zram_drv.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/block/zram/zram_drv.c b/drivers/block/zram/zram_drv.c
index f9711c5..213dfc1 100644
--- a/drivers/block/zram/zram_drv.c
+++ b/drivers/block/zram/zram_drv.c
@@ -553,14 +553,14 @@ static void zram_reset_device(struct zram *zram, bool reset_capacity)
 	size_t index;
 	struct zram_meta *meta;
 
-	flush_work(&zram->free_work);
-
 	down_write(&zram->init_lock);
 	if (!zram->init_done) {
 		up_write(&zram->init_lock);
 		return;
 	}
 
+	flush_work(&zram->free_work);
+
 	meta = zram->meta;
 	zram->init_done = 0;
 
-- 
1.8.4.3

Re: [PATCH 1/7] zram: fix race between reset and flushing pending work

[Andrew Morton]

On Mon, 13 Jan 2014 20:18:56 +0900 Minchan Kim <minchan@kernel.org> wrote:

> Dan and Sergey reported that there is a racy between reset and
> flushing of pending work so that it could make oops by freeing
> zram->meta in reset while zram_slot_free can access zram->meta
> if new request is adding during the race window.
> 
> This patch moves flush after taking init_lock so it prevents
> new request so that it closes the race.
> 
> ..
>
> --- a/drivers/block/zram/zram_drv.c
> +++ b/drivers/block/zram/zram_drv.c
> @@ -553,14 +553,14 @@ static void zram_reset_device(struct zram *zram, bool reset_capacity)
>  	size_t index;
>  	struct zram_meta *meta;
>  
> -	flush_work(&zram->free_work);
> -
>  	down_write(&zram->init_lock);
>  	if (!zram->init_done) {
>  		up_write(&zram->init_lock);
>  		return;
>  	}
>  
> +	flush_work(&zram->free_work);
> +
>  	meta = zram->meta;
>  	zram->init_done = 0;

This makes zram.lock nest inside zram.init_lock, which afaict is new
behaviour.

That all seems OK and logical - has it been well tested with lockdep?

Re: [PATCH 1/7] zram: fix race between reset and flushing pending work

[Minchan Kim]

On Mon, Jan 13, 2014 at 03:55:27PM -0800, Andrew Morton wrote:
> On Mon, 13 Jan 2014 20:18:56 +0900 Minchan Kim <minchan@kernel.org> wrote:
> 
> > Dan and Sergey reported that there is a racy between reset and
> > flushing of pending work so that it could make oops by freeing
> > zram->meta in reset while zram_slot_free can access zram->meta
> > if new request is adding during the race window.
> > 
> > This patch moves flush after taking init_lock so it prevents
> > new request so that it closes the race.
> > 
> > ..
> >
> > --- a/drivers/block/zram/zram_drv.c
> > +++ b/drivers/block/zram/zram_drv.c
> > @@ -553,14 +553,14 @@ static void zram_reset_device(struct zram *zram, bool reset_capacity)
> >  	size_t index;
> >  	struct zram_meta *meta;
> >  
> > -	flush_work(&zram->free_work);
> > -
> >  	down_write(&zram->init_lock);
> >  	if (!zram->init_done) {
> >  		up_write(&zram->init_lock);
> >  		return;
> >  	}
> >  
> > +	flush_work(&zram->free_work);
> > +
> >  	meta = zram->meta;
> >  	zram->init_done = 0;
> 
> This makes zram.lock nest inside zram.init_lock, which afaict is new
> behaviour.

Originally, it was nested so it's not new. :)
Look at zram_make_request which hold init_lock and then zram_bvec_rw
hold zram->lock.

> 
> That all seems OK and logical - has it been well tested with lockdep?

Yeb.

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

-- 
Kind regards,
Minchan Kim