Messages in this thread | | | Date | Thu, 15 Jan 2015 10:56:10 +0300 | From | Vladimir Davydov <> | Subject | Re: [PATCH -mm] vmscan: move reclaim_state handling to shrink_slab |
| |
On Wed, Jan 14, 2015 at 03:34:49PM -0800, Andrew Morton wrote: > On Wed, 14 Jan 2015 16:55:36 +0300 Vladimir Davydov <vdavydov@parallels.com> wrote: > > This patch also makes shrink_slab() return the number of reclaimed slab > > pages (obtained from reclaim_state) instead of the number of reclaimed > > objects, because the latter is not of much use - it was only checked by > > drop_slab() to decide whether it should continue reclaim or abort. The > > number of reclaimed pages is more appropriate, because it also can be > > used by shrink_zone() to accumulate scan_control->nr_reclaimed. > > Not sure that this is a good change. If shrink_slab() managed to free > some objects but didn't free any pages then that's a good sign that > additional calls to shrink_slab() *will* free some pages. With this > change, drop_slab_node() can give up too early.
Fair enough. We'd better leave the return value intact then. I think we should add an additional argument to add the number of reclaimed slab pages to, as I intended to do initially. Will resend.
Thanks, Vladimir
| |