From: Peter Zijlstra on
On Fri, 2010-08-06 at 00:10 +0800, Wu Fengguang wrote:

Acked-by: Peter Zijlstra <a.p.zijlstra(a)chello.nl>

> +/**
> + * bdi_dirty_limit - @bdi's share of dirty throttling threshold
> + *
> + * Allocate high/low dirty limits to fast/slow devices, in order to prevent
> + * - starving fast devices
> + * - piling up dirty pages (that will take long time to sync) on slow devices
> + *
> + * The bdi's share of dirty limit will be adapting to its throughput and
> + * bounded by the bdi->min_ratio and/or bdi->max_ratio parameters, if set.
> + */

Another thing solved by the introduction of per-bdi dirty limits (and
now per-bdi flushing) is the whole stacked-bdi writeout deadlock.

Although I'm not sure we want/need to mention that here.
--
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: Wu Fengguang on
On Fri, Aug 06, 2010 at 06:17:26PM +0800, Peter Zijlstra wrote:
> On Fri, 2010-08-06 at 00:10 +0800, Wu Fengguang wrote:
>
> Acked-by: Peter Zijlstra <a.p.zijlstra(a)chello.nl>
>
> > +/**
> > + * bdi_dirty_limit - @bdi's share of dirty throttling threshold
> > + *
> > + * Allocate high/low dirty limits to fast/slow devices, in order to prevent
> > + * - starving fast devices
> > + * - piling up dirty pages (that will take long time to sync) on slow devices
> > + *
> > + * The bdi's share of dirty limit will be adapting to its throughput and
> > + * bounded by the bdi->min_ratio and/or bdi->max_ratio parameters, if set.
> > + */
>
> Another thing solved by the introduction of per-bdi dirty limits (and
> now per-bdi flushing) is the whole stacked-bdi writeout deadlock.
>
> Although I'm not sure we want/need to mention that here.

The changelog looks like a suitable place :)

Thanks,
Fengguang
---
Subject: writeback: add comment to the dirty limits functions
From: Wu Fengguang <fengguang.wu(a)intel.com>
Date: Thu Jul 15 09:54:25 CST 2010

Document global_dirty_limits(), bdi_dirty_limit() and task_dirty_limit().

Note that another thing solved by the introduction of per-bdi dirty
limits (and now per-bdi flushing) is the whole stacked-bdi writeout
deadlock. -- Peter

Cc: Christoph Hellwig <hch(a)infradead.org>
Cc: Dave Chinner <david(a)fromorbit.com>
Cc: Jens Axboe <axboe(a)kernel.dk>
Acked-by: Peter Zijlstra <a.p.zijlstra(a)chello.nl>
Signed-off-by: Wu Fengguang <fengguang.wu(a)intel.com>
---
mm/page-writeback.c | 31 ++++++++++++++++++++++++++++---
1 file changed, 28 insertions(+), 3 deletions(-)

--- linux-next.orig/mm/page-writeback.c 2010-08-03 23:14:19.000000000 +0800
+++ linux-next/mm/page-writeback.c 2010-08-05 00:37:17.000000000 +0800
@@ -261,11 +261,18 @@ static inline void task_dirties_fraction
}

/*
- * scale the dirty limit
+ * task_dirty_limit - scale down dirty throttling threshold for one task
*
* task specific dirty limit:
*
* dirty -= (dirty/8) * p_{t}
+ *
+ * To protect light/slow dirtying tasks from heavier/fast ones, we start
+ * throttling individual tasks before reaching the bdi dirty limit.
+ * Relatively low thresholds will be allocated to heavy dirtiers. So when
+ * dirty pages grow large, heavy dirtiers will be throttled first, which will
+ * effectively curb the growth of dirty pages. Light dirtiers with high enough
+ * dirty threshold may never get throttled.
*/
static unsigned long task_dirty_limit(struct task_struct *tsk,
unsigned long bdi_dirty)
@@ -390,6 +397,15 @@ unsigned long determine_dirtyable_memory
return x + 1; /* Ensure that we never return 0 */
}

+/**
+ * global_dirty_limits - background-writeback and dirty-throttling thresholds
+ *
+ * Calculate the dirty thresholds based on sysctl parameters
+ * - vm.dirty_background_ratio or vm.dirty_background_bytes
+ * - vm.dirty_ratio or vm.dirty_bytes
+ * The dirty limits will be lifted by 1/4 for PF_LESS_THROTTLE (ie. nfsd) and
+ * runtime tasks.
+ */
void global_dirty_limits(unsigned long *pbackground, unsigned long *pdirty)
{
unsigned long background;
@@ -424,8 +440,17 @@ void global_dirty_limits(unsigned long *
*pdirty = dirty;
}

-unsigned long bdi_dirty_limit(struct backing_dev_info *bdi,
- unsigned long dirty)
+/**
+ * bdi_dirty_limit - @bdi's share of dirty throttling threshold
+ *
+ * Allocate high/low dirty limits to fast/slow devices, in order to prevent
+ * - starving fast devices
+ * - piling up dirty pages (that will take long time to sync) on slow devices
+ *
+ * The bdi's share of dirty limit will be adapting to its throughput and
+ * bounded by the bdi->min_ratio and/or bdi->max_ratio parameters, if set.
+ */
+unsigned long bdi_dirty_limit(struct backing_dev_info *bdi, unsigned long dirty)
{
u64 bdi_dirty;
long numerator, denominator;
--
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/