Re: [PATCH 3/7] writeback: introduce smoothed global dirty limit

[Andrew Morton]

On Sun, 19 Jun 2011 23:55:47 +0800
Wu Fengguang <fengguang.wu@xxxxxxxxx> wrote:

> On Sun, Jun 19, 2011 at 11:36:37PM +0800, Christoph Hellwig wrote:
> > On Sun, Jun 19, 2011 at 11:01:11PM +0800, Wu Fengguang wrote:
> > > The start of a heavy weight application (ie. KVM) may instantly knock
> > > down determine_dirtyable_memory() and hence the global/bdi dirty
> > > thresholds.
> > > 
> > > So introduce global_dirty_limit for tracking the global dirty threshold
> > > with policies
> > > 
> > > - follow downwards slowly
> > > - follow up in one shot
> > > 
> > > global_dirty_limit can effectively mask out the impact of sudden drop of
> > > dirtyable memory. It will be used in the next patch for two new type of
> > > dirty limits.
> > 
> > This needs to be explained in more detail in comments near the actual
> > code.
> 
> Good point! This is the added comment for the update_dirty_limit() function.
> 
> /*
>  * The global dirtyable memory and dirty threshold could be suddenly knocked
>  * down by a large amount (eg. on the startup of KVM). This may throw the
>  * system into deep dirty exceeded state and throttled to "death" for a couple
>  * of seconds. The solution is to maintain global_dirty_limit for tracking
>  * slowly down to the knocked down dirty threshold.
>  */
> static void update_dirty_limit(unsigned long thresh,
>                                unsigned long dirty)
> 

Neither this nor the changelog explain things well. 

Looking at the code, KVM starts, allocates memory,
global_page_state(NR_FREE_PAGES) decreases by N and
global_reclaimable_pages() increases by N.  Nothing changed.

So what's going on here?
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/