[Devel] Re: [PATCH v2] memcg: reduce lock time at move charge (Was Re: [PATCH 04/10] memcg: disable local interrupts in lock_page_cgroup()

2010-10-11 Thread KAMEZAWA Hiroyuki
On Tue, 12 Oct 2010 12:56:13 +0900 Daisuke Nishimura wrote: > > +err_out: > > + for (; mt < info + num; mt++) > > + if (mt->type == MC_TARGET_PAGE) { > > + putback_lru_page(mt->val.page); > Is this putback_lru_page() necessary ? > is_target_pte_for_mc() doesn't isola

[Devel] Re: [PATCH v2] memcg: reduce lock time at move charge (Was Re: [PATCH 04/10] memcg: disable local interrupts in lock_page_cgroup()

2010-10-11 Thread Daisuke Nishimura
> +err_out: > + for (; mt < info + num; mt++) > + if (mt->type == MC_TARGET_PAGE) { > + putback_lru_page(mt->val.page); Is this putback_lru_page() necessary ? is_target_pte_for_mc() doesn't isolate the page. Thanks, Daisuke Nishimura. > + p

[Devel] Re: [PATCH v2] memcg: reduce lock time at move charge (Was Re: [PATCH 04/10] memcg: disable local interrupts in lock_page_cgroup()

2010-10-11 Thread Balbir Singh
* KAMEZAWA Hiroyuki [2010-10-12 12:42:53]: > On Tue, 12 Oct 2010 09:09:15 +0530 > Balbir Singh wrote: > > > * KAMEZAWA Hiroyuki [2010-10-08 19:41:31]: > > > > > On Fri, 8 Oct 2010 14:12:01 +0900 > > > KAMEZAWA Hiroyuki wrote: > > > > > Sure. It walks the same data three times, potentially c

[Devel] Re: [PATCH v2] memcg: reduce lock time at move charge (Was Re: [PATCH 04/10] memcg: disable local interrupts in lock_page_cgroup()

2010-10-11 Thread KAMEZAWA Hiroyuki
On Tue, 12 Oct 2010 09:09:15 +0530 Balbir Singh wrote: > * KAMEZAWA Hiroyuki [2010-10-08 19:41:31]: > > > On Fri, 8 Oct 2010 14:12:01 +0900 > > KAMEZAWA Hiroyuki wrote: > > > > Sure. It walks the same data three times, potentially causing > > > > thrashing in the L1 cache. > > > > > > Hmm, m

[Devel] Re: [PATCH v2] memcg: reduce lock time at move charge (Was Re: [PATCH 04/10] memcg: disable local interrupts in lock_page_cgroup()

2010-10-11 Thread Balbir Singh
* KAMEZAWA Hiroyuki [2010-10-08 19:41:31]: > On Fri, 8 Oct 2010 14:12:01 +0900 > KAMEZAWA Hiroyuki wrote: > > > Sure. It walks the same data three times, potentially causing > > > thrashing in the L1 cache. > > > > Hmm, make this 2 times, at least. > > > How about this ? > == > From: KAMEZAWA

[Devel] Re: [PATCH v2] memcg: reduce lock time at move charge (Was Re: [PATCH 04/10] memcg: disable local interrupts in lock_page_cgroup()

2010-10-08 Thread KAMEZAWA Hiroyuki
On Fri, 8 Oct 2010 14:12:01 +0900 KAMEZAWA Hiroyuki wrote: > > Sure. It walks the same data three times, potentially causing > > thrashing in the L1 cache. > > Hmm, make this 2 times, at least. > How about this ? == From: KAMEZAWA Hiroyuki Presently, at task migration among cgroups, memory cg

[Devel] Re: [PATCH v2] memcg: reduce lock time at move charge (Was Re: [PATCH 04/10] memcg: disable local interrupts in lock_page_cgroup()

2010-10-07 Thread KAMEZAWA Hiroyuki
On Thu, 7 Oct 2010 21:55:56 -0700 Andrew Morton wrote: > On Fri, 8 Oct 2010 13:37:12 +0900 KAMEZAWA Hiroyuki > wrote: > > > On Thu, 7 Oct 2010 16:14:54 -0700 > > Andrew Morton wrote: > > > > > On Thu, 7 Oct 2010 17:04:05 +0900 > > > KAMEZAWA Hiroyuki wrote: > > > > > > > Now, at task migra

[Devel] Re: [PATCH v2] memcg: reduce lock time at move charge (Was Re: [PATCH 04/10] memcg: disable local interrupts in lock_page_cgroup()

2010-10-07 Thread Andrew Morton
On Fri, 8 Oct 2010 13:37:12 +0900 KAMEZAWA Hiroyuki wrote: > On Thu, 7 Oct 2010 16:14:54 -0700 > Andrew Morton wrote: > > > On Thu, 7 Oct 2010 17:04:05 +0900 > > KAMEZAWA Hiroyuki wrote: > > > > > Now, at task migration among cgroup, memory cgroup scans page table and > > > moving > > > acc

[Devel] Re: [PATCH v2] memcg: reduce lock time at move charge (Was Re: [PATCH 04/10] memcg: disable local interrupts in lock_page_cgroup()

2010-10-07 Thread KAMEZAWA Hiroyuki
On Thu, 7 Oct 2010 16:14:54 -0700 Andrew Morton wrote: > On Thu, 7 Oct 2010 17:04:05 +0900 > KAMEZAWA Hiroyuki wrote: > > > Now, at task migration among cgroup, memory cgroup scans page table and > > moving > > account if flags are properly set. > > > > The core code, mem_cgroup_move_charge_p

[Devel] Re: [PATCH v2] memcg: reduce lock time at move charge (Was Re: [PATCH 04/10] memcg: disable local interrupts in lock_page_cgroup()

2010-10-07 Thread Daisuke Nishimura
On Thu, 7 Oct 2010 16:14:54 -0700 Andrew Morton wrote: > On Thu, 7 Oct 2010 17:04:05 +0900 > KAMEZAWA Hiroyuki wrote: > > > Now, at task migration among cgroup, memory cgroup scans page table and > > moving > > account if flags are properly set. > > > > The core code, mem_cgroup_move_charge_p

[Devel] Re: [PATCH v2] memcg: reduce lock time at move charge (Was Re: [PATCH 04/10] memcg: disable local interrupts in lock_page_cgroup()

2010-10-07 Thread Andrew Morton
On Thu, 7 Oct 2010 17:04:05 +0900 KAMEZAWA Hiroyuki wrote: > Now, at task migration among cgroup, memory cgroup scans page table and moving > account if flags are properly set. > > The core code, mem_cgroup_move_charge_pte_range() does > > pte_offset_map_lock(); > for all ptes in a