Re: [PATCH 1/2] vmscan: shrink_slab() require number of lru_pages, not page order
From: KOSAKI Motohiro
Date: Sun Jun 27 2010 - 21:32:18 EST
> On Fri, 25 Jun 2010, KOSAKI Motohiro wrote:
> > Fix simple argument error. Usually 'order' is very small value than
> > lru_pages. then it can makes unnecessary icache dropping.
> This is going to reduce the delta that is added to shrinker->nr
> significantly thereby increasing the number of times that shrink_slab() is
Smaller shrink -> only makes retry
Bigger shrink -> makes unnecessary icache/dcache drop. it can bring
mysterious low performance.
> What does the "lru_pages" parameter do in shrink_slab()? Looks
> like its only role is as a divison factor in a complex calculation of
> pages to be scanned.
scanned/lru_pages ratio define basic shrink_slab puressure strength.
So, If you intentionally need bigger slab pressure, bigger scanned parameter
passing is better rather than mysterious 'order' parameter.
> do_try_to_free_pages passes 0 as "lru_pages" to shrink_slab() when trying
> to do cgroup lru scans. Why is that?
When cgroup lru scans, do_try_to_free_pages() don't call shrink_slab().
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/