(resend with adding Cc: containers and linux-mm)
Hi, Andrew.
Thank you for picking up this patch.
But this version has a small race problem. This is a fix-up patch.
===
From: Daisuke Nishimura nishim...@mxp.nes.nec.co.jp
We must clear mc.moving_task before waking up all waiters at the end
b1dd693e introduces, and deadlock which the commit fixed).
I'll test and resend this after new year holidays in Japan.
===
From: Daisuke Nishimura nishim...@mxp.nes.nec.co.jp
The commit b1dd693e(memcg: avoid deadlock between move charge and try_charge())
can cause another deadlock about mmap_sem
,
Daisuke Nishimura.
___
Containers mailing list
contain...@lists.linux-foundation.org
https://lists.linux-foundation.org/mailman/listinfo/containers
___
Devel mailing list
Devel@openvz.org
https://openvz.org/mailman
.
Can you take it into account too ?
Thanks,
Daisuke Nishimura.
___
Containers mailing list
contain...@lists.linux-foundation.org
https://lists.linux-foundation.org/mailman/listinfo/containers
___
Devel mailing
.
This patch limits the return value of vm_dirty_param() considring
global settings.
Changelog:
- fixed an argument mem int to u64
- fixed to use global available memory to cap memcg's value.
Signed-off-by: KAMEZAWA Hiroyuki kamezawa.hir...@jp.fujitsu.com
Acked-by: Daisuke Nishimura
memory that is dirty.
Signed-off-by: Greg Thelen gthe...@google.com
Signed-off-by: Andrea Righi ari...@develer.com
Reviewed-by: Daisuke Nishimura nishim...@mxp.nes.nec.co.jp
___
Containers mailing list
contain...@lists.linux-foundation.org
https
accounting.
--
1.7.1
Can you clarify whether we can limit the total dirty pages under hierarchy
in use_hierarchy==1 case ?
If we can, I think it would be better to note it in this documentation.
Thanks,
Daisuke Nishimura.
___
Containers mailing list
usage_in_bytes never
exceeds limit_in_bytes.
Thanks,
Daisuke Nishimura.
___
Containers mailing list
contain...@lists.linux-foundation.org
https://lists.linux-foundation.org/mailman/listinfo/containers
___
Devel
+static unsigned long long
+memcg_hierarchical_free_pages(struct mem_cgroup *mem)
+{
+ struct cgroup *cgroup;
+ unsigned long long min_free, free;
+
+ min_free = res_counter_read_u64(mem-res, RES_LIMIT) -
+ res_counter_read_u64(mem-res, RES_USAGE);
+ cgroup =
Thelen gthe...@google.com
Signed-off-by: Andrea Righi ari...@develer.com
Acked-by: Daisuke Nishimura nishim...@mxp.nes.nec.co.jp
___
Containers mailing list
contain...@lists.linux-foundation.org
https://lists.linux-foundation.org/mailman/listinfo/containers
...@develer.com
Signed-off-by: Balbir Singh bal...@linux.vnet.ibm.com
Signed-off-by: Greg Thelen gthe...@google.com
Acked-by: Daisuke Nishimura nishim...@mxp.nes.nec.co.jp
One question: shouldn't we return -EINVAL when writing to
dirty(_background)_limit_bytes
a bigger value than that of global
On Wed, 20 Oct 2010 12:31:10 +0900
Daisuke Nishimura nishim...@mxp.nes.nec.co.jp wrote:
On Mon, 18 Oct 2010 17:39:42 -0700
Greg Thelen gthe...@google.com wrote:
Add cgroupfs interface to memcg dirty page limits:
Direct write-out is controlled with:
- memory.dirty_ratio
On Wed, 20 Oct 2010 11:24:31 +0900
KAMEZAWA Hiroyuki kamezawa.hir...@jp.fujitsu.com wrote:
On Wed, 20 Oct 2010 10:14:21 +0900
KAMEZAWA Hiroyuki kamezawa.hir...@jp.fujitsu.com wrote:
On Wed, 20 Oct 2010 09:48:21 +0900
Daisuke Nishimura nishim...@mxp.nes.nec.co.jp wrote:
On Wed, 20
into global_dirty_limits().
Signed-off-by: Andrea Righi ari...@develer.com
Signed-off-by: Greg Thelen gthe...@google.com
Reviewed-by: Daisuke Nishimura nishim...@mxp.nes.nec.co.jp
___
Containers mailing list
contain...@lists.linux-foundation.org
https
-by: Greg Thelen gthe...@google.com
Signed-off-by: Balbir Singh bal...@linux.vnet.ibm.com
Acked-by: Daisuke Nishimura nishim...@mxp.nes.nec.co.jp
___
Containers mailing list
contain...@lists.linux-foundation.org
https://lists.linux-foundation.org/mailman
On Wed, 20 Oct 2010 13:18:57 +0900
KAMEZAWA Hiroyuki kamezawa.hir...@jp.fujitsu.com wrote:
On Tue, 19 Oct 2010 10:00:15 +0900
KAMEZAWA Hiroyuki kamezawa.hir...@jp.fujitsu.com wrote:
On Mon, 18 Oct 2010 17:39:44 -0700
Greg Thelen gthe...@google.com wrote:
If the current process is in
at different
levels.
Signed-off-by: Andrea Righi ari...@develer.com
Signed-off-by: Greg Thelen gthe...@google.com
Reviewed-by: Daisuke Nishimura nishim...@mxp.nes.nec.co.jp
___
Containers mailing list
contain...@lists.linux-foundation.org
https
gthe...@google.com
Acked-by: Daisuke Nishimura nishim...@mxp.nes.nec.co.jp
___
Containers mailing list
contain...@lists.linux-foundation.org
https://lists.linux-foundation.org/mailman/listinfo/containers
. However,
these more general interfaces allow for new statistics to be
more easily added. New statistics are added with memcg dirty
page accounting.
Signed-off-by: Greg Thelen gthe...@google.com
Signed-off-by: Andrea Righi ari...@develer.com
Acked-by: Daisuke Nishimura nishim
-off-by: Greg Thelen gthe...@google.com
This approach is more straightforward and easy to understand.
Acked-by: Daisuke Nishimura nishim...@mxp.nes.nec.co.jp
___
Containers mailing list
contain...@lists.linux-foundation.org
https://lists.linux
On Tue, 12 Oct 2010 14:48:01 +0900
KAMEZAWA Hiroyuki kamezawa.hir...@jp.fujitsu.com wrote:
On Tue, 12 Oct 2010 12:56:13 +0900
Daisuke Nishimura nishim...@mxp.nes.nec.co.jp wrote:
+err_out:
+ for (; mt info + num; mt++)
+ if (mt-type == MC_TARGET_PAGE
+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.
+ put_page
, but I agree we should describe it.
Thanks,
Daisuke Nishimura.
___
Containers mailing list
contain...@lists.linux-foundation.org
https://lists.linux-foundation.org/mailman/listinfo/containers
___
Devel mailing
it be better to remove -page, if possible ?
Thanks,
Daisuke Nishimura.
Signed-off-by: Balbir Singh bal...@linux.vnet.ibm.com
---
include/linux/page_cgroup.h |3 +++
mm/page_cgroup.c|1 +
2 files changed, 4 insertions(+), 0 deletions(-)
diff --git a/include/linux
? Or, we need re-organize current move-charge
implementation.
Thanks,
Daisuke Nishimura.
___
Containers mailing list
contain...@lists.linux-foundation.org
https://lists.linux-foundation.org/mailman/listinfo/containers
for reclaim?
We store page_cgroups in arrays now, so I suppose we can implement pc_to_pfn()
using the similar calculation as page_to_pfn() does.
IIRC, KAMEZAWA-san talked about it in another thread.
Thanks,
Daisuke Nishimura.
___
Containers mailing list
,
Daisuke Nishimura.
___
Containers mailing list
contain...@lists.linux-foundation.org
https://lists.linux-foundation.org/mailman/listinfo/containers
___
Devel mailing list
Devel@openvz.org
https://openvz.org
gthe...@google.com
Acked-by: Daisuke Nishimura nishim...@mxp.nes.nec.co.jp
Thanks,
Daisuke Nishimura.
---
include/linux/page_cgroup.h | 23 +++
1 files changed, 23 insertions(+), 0 deletions(-)
diff --git a/include/linux/page_cgroup.h b/include/linux/page_cgroup.h
-by: Daisuke Nishimura nishim...@mxp.nes.nec.co.jp
Thanks
Daisuke Nishimura.
---
Documentation/cgroups/memory.txt | 37 +
1 files changed, 37 insertions(+), 0 deletions(-)
diff --git a/Documentation/cgroups/memory.txt
b/Documentation/cgroups/memory.txt
.
Thanks,
Daisuke Nishimura.
___
Containers mailing list
contain...@lists.linux-foundation.org
https://lists.linux-foundation.org/mailman/listinfo/containers
___
Devel mailing list
Devel@openvz.org
https://openvz.org
On Wed, 14 Apr 2010 21:48:25 -0700, Greg Thelen gthe...@google.com wrote:
On Wed, Apr 14, 2010 at 7:40 PM, Daisuke Nishimura
nishim...@mxp.nes.nec.co.jp wrote:
On Wed, 14 Apr 2010 13:14:07 -0700, Greg Thelen gthe...@google.com wrote:
Vivek Goyal vgo...@redhat.com writes:
On Tue, Apr 13
. __remove_from_page_cache(), which will update
DIRTY, is called with irq disabled(iow, under mapping-tree_lock) but not in
interrupt context.
Anyway, I tend to agree with KAMEZAWA-san: use trylock always(except for
FILE_MAPPED),
or add some new interfaces(e.g. mem_cgroup_update_stat_locked/safe...).
Thanks,
Daisuke
account we don't have move-account-mapped-file, yet.
FILE_MAPPED is updated under pte lock. OTOH, move account is also done under
pte lock. page cgroup lock is held under pte lock in both cases, so move account
is not so problem as for FILE_MAPPED.
Thanks,
Daisuke Nishimura
speaking, it's not true
because global values can change. We need a hook in mem_cgroup_dirty_read()?).
Thanks,
Daisuke Nishimura.
6. Hierarchy support
--
1.6.3.3
___
Containers mailing list
contain...@lists.linux-foundation.org
https://lists.linux
On Tue, 16 Mar 2010 10:11:50 -0400
Vivek Goyal vgo...@redhat.com wrote:
On Tue, Mar 16, 2010 at 11:32:38AM +0900, Daisuke Nishimura wrote:
[..]
+ * mem_cgroup_page_stat() - get memory cgroup file cache statistics
+ * @item:memory statistic item exported to the kernel
))
+ continue;
cb-fill(cb, memcg_stat_strings[i].local_name, mystat.stat[i]);
}
not to show file stat in root cgroup ? It's meaningless value anyway.
Of course, you'd better mention it in [2/5] too.
Thanks,
Daisuke Nishimura.
/* per zone stat */
val
On Thu, 11 Mar 2010 15:15:11 +0900, KAMEZAWA Hiroyuki
kamezawa.hir...@jp.fujitsu.com wrote:
On Thu, 11 Mar 2010 14:13:00 +0900
KAMEZAWA Hiroyuki kamezawa.hir...@jp.fujitsu.com wrote:
On Thu, 11 Mar 2010 13:58:47 +0900
Daisuke Nishimura nishim...@mxp.nes.nec.co.jp wrote:
I'll consider
.
Signed-off-by: KAMEZAWA Hiroyuki kamezawa.hir...@jp.fujitsu.com
Looks good overall.
Thanks,
Daisuke Nishimura.
---
include/linux/memcontrol.h |7 ++-
include/linux/page_cgroup.h | 15 +++
mm/memcontrol.c | 92
+---
mm/rmap.c
/restore version to mixed version ?
Thanks,
Daisuke Nishimura.
===
include/linux/page_cgroup.h | 28 ++--
mm/memcontrol.c | 36 ++--
2 files changed, 44 insertions(+), 20 deletions(-)
diff --git a/include/linux/page_cgroup.h b
On Thu, 11 Mar 2010 13:49:08 +0900, KAMEZAWA Hiroyuki
kamezawa.hir...@jp.fujitsu.com wrote:
On Thu, 11 Mar 2010 13:31:23 +0900
Daisuke Nishimura nishim...@mxp.nes.nec.co.jp wrote:
On Wed, 10 Mar 2010 09:26:24 +0530, Balbir Singh
bal...@linux.vnet.ibm.com wrote:
* nishim
slower as I add additional codes. Using
local_irq_disable()/enable()
except in mem_cgroup_update_file_mapped(it can be the only candidate to be
called
with irq disabled in future) might be the choice.
Thanks,
Daisuke Nishimura.
On Tue, 9 Mar 2010 10:20:58 +0530, Balbir Singh bal
On Mon, 8 Mar 2010 11:37:11 +0900, KAMEZAWA Hiroyuki
kamezawa.hir...@jp.fujitsu.com wrote:
On Mon, 8 Mar 2010 11:17:24 +0900
Daisuke Nishimura nishim...@mxp.nes.nec.co.jp wrote:
But IIRC, clear_writeback is done under treelock No ?
The place where NR_WRITEBACK is updated is out
On Mon, 8 Mar 2010 17:31:00 +0900, KAMEZAWA Hiroyuki
kamezawa.hir...@jp.fujitsu.com wrote:
On Mon, 8 Mar 2010 17:07:11 +0900
Daisuke Nishimura nishim...@mxp.nes.nec.co.jp wrote:
On Mon, 8 Mar 2010 11:37:11 +0900, KAMEZAWA Hiroyuki
kamezawa.hir...@jp.fujitsu.com wrote:
On Mon, 8 Mar
On Tue, 9 Mar 2010 09:20:54 +0900, KAMEZAWA Hiroyuki
kamezawa.hir...@jp.fujitsu.com wrote:
On Tue, 9 Mar 2010 09:18:45 +0900
Daisuke Nishimura nishim...@mxp.nes.nec.co.jp wrote:
On Mon, 8 Mar 2010 17:31:00 +0900, KAMEZAWA Hiroyuki
kamezawa.hir...@jp.fujitsu.com wrote:
On Mon, 8 Mar
On Tue, 9 Mar 2010 09:19:14 +0900, KAMEZAWA Hiroyuki
kamezawa.hir...@jp.fujitsu.com wrote:
On Tue, 9 Mar 2010 01:12:52 +0100
Andrea Righi ari...@develer.com wrote:
On Mon, Mar 08, 2010 at 05:31:00PM +0900, KAMEZAWA Hiroyuki wrote:
On Mon, 8 Mar 2010 17:07:11 +0900
Daisuke Nishimura
interrupt context.
This means lock_page_cgroup() is called under interrupt context, that is,
the case b) above can happen.
hmm... I don't have any good idea for now except disabling irq around page
cgroup lock
to avoid all of these mess things.
Thanks,
Daisuke Nishimura
On Mon, 8 Mar 2010 10:56:41 +0900, KAMEZAWA Hiroyuki
kamezawa.hir...@jp.fujitsu.com wrote:
On Mon, 8 Mar 2010 10:44:47 +0900
Daisuke Nishimura nishim...@mxp.nes.nec.co.jp wrote:
+/*
+ * mem_cgroup_update_page_stat_locked() - update memcg file cache's
accounting
+ * @page
/unlock_page_cgroup_migrate() can be called under irq
context
(e.g. end_page_writeback()), so I think we must local_irq_disable()/enable()
here.
Thanks,
Daisuke Nishimura.
/*
* We charges against to which may not have any tasks. Then, to
* can be under rmdir(). But in current
at rmdir, so I think it would be better to handle file pages anyway.
Thanks,
Daisuke Nishimura.
___
Containers mailing list
contain...@lists.linux-foundation.org
https://lists.linux-foundation.org/mailman/listinfo/containers
?).
FYI, I'm planning to add file-cache and shmem/tmpfs support for move_charge
feature
for 2.6.35.
But, hmm, it would be complicated if we try to move dirty account too.
Thanks,
Daisuke Nishimura.
___
Containers mailing list
contain...@lists.linux
PCG_FILE_MAPPED flag too ? I think this flag can be handled in the
same way as other flags you're trying to add, and we can change
if (page_mapped(page) !PageAnon(page)) to if (PageCgroupFileMapped(pc)
in __mem_cgroup_move_account(). It would be cleaner than current code, IMHO.
Thanks,
Daisuke
mapping-tree_lock.
hmm, we should call those mem_cgroup_update_state() outside mapping-tree_lock,
or add local_irq_save/restore() around lock/unlock_page_cgroup() to avoid
dead-lock.
Thanks,
Daisuke Nishimura.
On Mon, 1 Mar 2010 22:23:40 +0100, Andrea Righi ari...@develer.com wrote:
Apply
On Wed, 3 Mar 2010 12:29:06 +0900, KAMEZAWA Hiroyuki
kamezawa.hir...@jp.fujitsu.com wrote:
On Wed, 3 Mar 2010 11:12:38 +0900
Daisuke Nishimura nishim...@mxp.nes.nec.co.jp wrote:
diff --git a/mm/filemap.c b/mm/filemap.c
index fe09e51..f85acae 100644
--- a/mm/filemap.c
+++ b/mm
*/
- __this_cpu_add(mem-stat-count[MEM_CGROUP_STAT_FILE_MAPPED], val);
+ VM_BUG_ON(idx = MEM_CGROUP_STAT_NSTATS);
+ __this_cpu_add(mem-stat-count[idx], val);
done:
unlock_page_cgroup(pc);
Thanks,
Daisuke Nishimura
be better to name it __mem_cgroup_stat_set_safe.
And could you remove the definition of __mem_cgroup_stat_reset ?
Thanks,
Daisuke Nishimura.
static inline s64
__mem_cgroup_stat_read_local(struct mem_cgroup_stat_cpu *stat,
enum mem_cgroup_stat_index idx
);
+ }
+unlock:
+ rcu_read_unlock();
+}
+
Thanks,
Daisuke Nishimura.
___
Containers mailing list
contain...@lists.linux-foundation.org
https://lists.linux-foundation.org/mailman/listinfo/containers
On Sat, 12 Dec 2009 15:06:52 +0200
Kirill A. Shutemov kir...@shutemov.name wrote:
On Sat, Dec 12, 2009 at 5:50 AM, Daisuke Nishimura
d-nishim...@mtf.biglobe.ne.jp wrote:
Sorry, I disagree this change.
mem_cgroup_soft_limit_check() is used for checking how much current usage
exceeds
On Sat, 12 Dec 2009 21:46:08 +0200
Kirill A. Shutemov kir...@shutemov.name wrote:
On Sat, Dec 12, 2009 at 4:34 PM, Daisuke Nishimura
d-nishim...@mtf.biglobe.ne.jp wrote:
On Sat, 12 Dec 2009 15:06:52 +0200
Kirill A. Shutemov kir...@shutemov.name wrote:
On Sat, Dec 12, 2009 at 5:50 AM
to record to swap */
if (ctype != MEM_CGROUP_CHARGE_TYPE_SWAPOUT)
css_put(mem-css);
Can if (do_swap_account) check be moved into mem_cgroup_threshold ?
Thanks,
Daisuke Nishimura.
___
Containers mailing list
contain...@lists.linux
in
http://marc.info/?l=linux-mmm=126021128400687w=2.
I think you can change there as you want, I can change my patch
(http://marc.info/?l=linux-mmm=126023467303178w=2, it has not yet sent to
Andrew anyway) to check mem_cgroup_is_root() in mem_cgroup_update_tree().
Thanks,
Daisuke Nishimura.
On Sat, 12
on performance impact of these patches(w/o setting any
threshold)?
IMHO, it might be small enough to be ignored because KAMEZAWA-san's coalesce
charge/uncharge
patches have decreased charge/uncharge for res_counter itself, but I want to
know just to make sure.
Regards,
Daisuke Nishimura
to memory.limit_in_bytes.
max_usage_in_bytes means the max of the usage so far(you can reset it
by writing to it).
See Documentation/cgroups/memory.txt(it might be a bit old).
Thanks,
Daisuke Nishimura.
c) In Shell #1, run a memory consumer (in which, malloc() is called
to allocate memory
- under mem+swap controller, program works with no swap. Only global LRU
may make pages swapped-out.
- If swap-accounting-mode is off, swap can be used unlimitedly.
Hmm, sounds a bit differenct from what I want. How about others ?
Thanks,
Daisuke Nishimura
implement
prepare_attach_nosleep() which holds swap_lock.
Anyway, if unwinding is supported officially, I think we can find a way
to go.
I think so too.
Thanks,
Daisuke Nishimura.
___
Containers mailing list
[EMAIL PROTECTED]
https://lists.linux
On Mon, 14 Jul 2008 16:54:44 +0900, KAMEZAWA Hiroyuki [EMAIL PROTECTED] wrote:
On Mon, 14 Jul 2008 15:28:22 +0900
Daisuke Nishimura [EMAIL PROTECTED] wrote:
On Fri, 11 Jul 2008 09:20:58 +0900, KAMEZAWA Hiroyuki [EMAIL PROTECTED]
wrote:
Thank you for your effort.
On Wed, 9 Jul
Hi, Dave-san.
On Thu, 10 Jul 2008 13:35:36 -0700, Dave Hansen [EMAIL PROTECTED] wrote:
On Fri, 2008-07-04 at 15:17 +0900, Daisuke Nishimura wrote:
+config CGROUP_SWAP_RES_CTLR
+ bool Swap Resource Controller for Control Groups
+ depends on CGROUP_MEM_RES_CTLR SWAP
On Sat, 5 Jul 2008 13:29:44 +0900, KAMEZAWA Hiroyuki [EMAIL PROTECTED] wrote:
On Fri, 4 Jul 2008 21:33:01 +0900
Daisuke Nishimura [EMAIL PROTECTED] wrote:
On Fri, 4 Jul 2008 19:16:38 +0900, KAMEZAWA Hiroyuki [EMAIL PROTECTED]
wrote:
On Fri, 4 Jul 2008 15:24:23 +0900
Daisuke
Hi, Balbir-san.
On Sat, 05 Jul 2008 12:22:25 +0530, Balbir Singh [EMAIL PROTECTED] wrote:
Daisuke Nishimura wrote:
Hi.
This is new version of swapcgroup.
Major changes from previous version
- Rebased on 2.6.26-rc5-mm3.
The new -mm has been released, but these patches
can
where memory should be accounted but swap should not?
- hierarchy support
- move charges along with task
Both of them need more discussion.
Thanks,
Daisuke Nishimura.
___
Containers mailing list
[EMAIL PROTECTED]
https://lists.linux-foundation.org
of those files are the same as memory cgroup.
Change log
v2-v3
- Rebased on 2.6.26-rc5-mm3.
v1-v2
- Rebased on 2.6.26-rc2-mm1.
- Implemented as a add-on to memory cgroup.
Signed-off-by: Daisuke Nishimura [EMAIL PROTECTED]
---
init/Kconfig|7 +
mm/memcontrol.c | 67
-mm3
- add helper functions and removed #ifdef from sys_swapon()/sys_swapoff().
- add check on mem_cgroup_subsys.disabled
v1-v2
- Rebased on 2.6.26-rc2-mm1
- Implemented as a add-on to memory cgroup.
Signed-off-by: Daisuke Nishimura [EMAIL PROTECTED]
---
include/linux/memcontrol.h | 20
on 2.6.26-rc2-mm1
- Implemented as a add-on to memory cgroup.
Signed-off-by: Daisuke Nishimura [EMAIL PROTECTED]
---
include/linux/memcontrol.h | 17 ++
include/linux/swap.h |4 +-
mm/memcontrol.c| 53
mm/shmem.c
-mm3
- take into account global swap usage.
- change arg of vm_swap_full from page to memcg.
- add check on mem_cgroup_subsys.disabled
v1-v2
- new patch
Signed-off-by: Daisuke Nishimura [EMAIL PROTECTED]
---
include/linux/memcontrol.h | 12
include/linux/swap.h |4
before implementing this way,
hierarchy and move charges support are needed.
So I think this is enough for the first step.
Change log
v2-v3
- new patch
Signed-off-by: Daisuke Nishimura [EMAIL PROTECTED]
---
include/linux/memcontrol.h |5
mm/memcontrol.c| 47
be added !p-swap_map to check the device has not
been swapoff'ed.
Thanks,
Daisuke Nishimura.
+ swap_cgroup_uncharge(p, i);
+ spin_unlock(swap_lock);
+ }
+ spin_lock(swap_lock
.
find_next_to_unuse seems to have fragile assumptions and
can dereference p-swap_map as well.
You're right.
Thank you for pointing it out!
I'll consider more.
Thanks,
Daisuke Nishimura.
___
Containers mailing list
[EMAIL PROTECTED]
https://lists.linux
Hi, Kamezawa-san.
On Fri, 4 Jul 2008 18:58:45 +0900, KAMEZAWA Hiroyuki [EMAIL PROTECTED] wrote:
On Fri, 4 Jul 2008 15:22:44 +0900
Daisuke Nishimura [EMAIL PROTECTED] wrote:
/* Swap 50% full? Release swapcache more aggressively.. */
-#define vm_swap_full() (nr_swap_pages*2
On Fri, 4 Jul 2008 18:40:33 +0900, KAMEZAWA Hiroyuki [EMAIL PROTECTED] wrote:
On Fri, 4 Jul 2008 15:15:36 +0900
Daisuke Nishimura [EMAIL PROTECTED] wrote:
Hi.
This is new version of swapcgroup.
Major changes from previous version
- Rebased on 2.6.26-rc5-mm3.
The new -mm has
On Fri, 4 Jul 2008 19:16:38 +0900, KAMEZAWA Hiroyuki [EMAIL PROTECTED] wrote:
On Fri, 4 Jul 2008 15:24:23 +0900
Daisuke Nishimura [EMAIL PROTECTED] wrote:
This patch implements force_empty of swapcgroup.
Currently, it simply uncharges all the charges from the group.
I think
the usage as accurate as possible, that is
the size of memory used by processes in the group at the moment.
I agree that statistics is not the purpose of memcg(and swap),
but, IMHO, it's useful feature of memcg.
Administrators can know how busy or idle each groups are by it.
Thanks,
Daisuke
?
Currently no, but people who are developing or wants to develop new subsystems
may find this useful, for example Daisuke Nishimura-san's swap controller.
My current version is implemented as addon to memory controller,
so it can be used by just mounting memory controller :)
But anyway, this feature
-res.limit necessary)
+ rc = -ENOMEM;
I think it should be (newcg-res.limit necessary).
Thanks,
Daisuke Nishimura.
___
Containers mailing list
[EMAIL PROTECTED]
https://lists.linux-foundation.org/mailman/listinfo/containers
.
Thanks,
Daisuke Nishimura.
___
Containers mailing list
[EMAIL PROTECTED]
https://lists.linux-foundation.org/mailman/listinfo/containers
___
Devel mailing list
Devel@openvz.org
https://openvz.org/mailman/listinfo
mm-sem while move.
I preffer the latter.
Though it's expencive, I think moving a task would not happen
so offen.
Thanks,
Daisuke Nishimura.
___
Containers mailing list
[EMAIL PROTECTED]
https://lists.linux-foundation.org/mailman/listinfo/containers
,
Daisuke Nishimura.
while nishimura-san's one still seems to have a lot of todo,
it seems good enough as a start point to me.
so i'd like to withdraw mine.
nishimura-san, is it ok for you?
Of cource.
I'll work hard to make it better.
I would suggest that me merge the good parts from both
On 2008/05/27 22:18 +0900, Balbir Singh wrote:
Daisuke Nishimura wrote:
On 2008/05/27 16:42 +0900, Balbir Singh wrote:
YAMAMOTO Takashi wrote:
hi,
Thanks for looking into this. Yamamoto-San is also looking into a swap
controller. Is there a consensus on the approach?
Not yet, but I think
On 2008/05/27 22:46 +0900, Balbir Singh wrote:
Daisuke Nishimura wrote:
On 2008/05/27 22:18 +0900, Balbir Singh wrote:
Daisuke Nishimura wrote:
On 2008/05/27 16:42 +0900, Balbir Singh wrote:
YAMAMOTO Takashi wrote:
hi,
Thanks for looking into this. Yamamoto-San is also looking into a swap
On 2008/05/22 16:37 +0900, KAMEZAWA Hiroyuki wrote:
On Thu, 22 May 2008 15:20:05 +0900
Daisuke Nishimura [EMAIL PROTECTED] wrote:
+#ifdef CONFIG_CGROUP_SWAP_RES_CTLR
+int swap_cgroup_charge(struct page *page,
+struct swap_info_struct *si,
+unsigned
On 2008/05/22 21:32 +0900, KOSAKI Motohiro wrote:
perhaps, I don't understand your intention exactly.
Why can't you make wrapper function?
e.g.
vm_swap_full(page_to_memcg(page))
OK.
I'll try it.
___
Containers mailing list
[EMAIL PROTECTED]
this feature or not, and avoid overhead for users
not using this feature.
- move charges along with task move between cgroups.
*1
https://lists.linux-foundation.org/pipermail/containers/2008-March/010216.html
Thanks,
Daisuke Nishimura.
___
Containers mailing
-by: Daisuke Nishimura [EMAIL PROTECTED]
---
init/Kconfig|7 +
mm/memcontrol.c | 67 +++
2 files changed, 74 insertions(+), 0 deletions(-)
diff --git a/init/Kconfig b/init/Kconfig
index 6135d07..aabbb83 100644
--- a/init/Kconfig
+++ b
This patch add a member to swap_info_struct for cgroup.
This member, array of pointers to mem_cgroup, is used to
remember to which cgroup each swap entries are charged.
The memory for this array of pointers is allocated on swapon,
and freed on swapoff.
Signed-off-by: Daisuke Nishimura [EMAIL
?
I added to swap_info_struct a member 'struct swap_cgroup **',
array of pointer to which swap_cgroup the swap entry is
charged.
Signed-off-by: Daisuke Nishimura [EMAIL PROTECTED]
---
include/linux/memcontrol.h | 21 +++
include/linux/swap.h |4 +-
mm
of Rik's splitlru patchset:
[patch 04/14] free swap space on swap-in/activation
but, I should verify whether this change to vm_swap_full()
is valid.
Signed-off-by: Daisuke Nishimura [EMAIL PROTECTED]
---
include/linux/memcontrol.h |3 +++
include/linux/swap.h |6 +-
mm
Hi.
On 2008/05/22 16:23 +0900, KAMEZAWA Hiroyuki wrote:
On Thu, 22 May 2008 15:18:51 +0900
Daisuke Nishimura [EMAIL PROTECTED] wrote:
This patch add a member to swap_info_struct for cgroup.
This member, array of pointers to mem_cgroup, is used to
remember to which cgroup each swap entries
.
this is memory(or swap) cgroup operation.
this argument is slightly odd.
But what callers of vm_swap_full() know is page,
not mem_cgroup.
I don't want to add to callers something like:
pc = get_page_cgroup(page);
mem = pc-mem_cgroup;
vm_swap_full(mem);
Thanks,
Daisuke Nishimura
to number of page
ret = (usage * 2 limit) || (nr_swap_pages * 2 total_swap_pages)
:
Thanks,
Daisuke Nishimura.
___
Containers mailing list
[EMAIL PROTECTED]
https://lists.linux-foundation.org/mailman/listinfo/containers
On 2008/05/22 16:44 +0900, KAMEZAWA Hiroyuki wrote:
On Thu, 22 May 2008 15:13:41 +0900
Daisuke Nishimura [EMAIL PROTECTED] wrote:
Hi.
I updated my swapcgroup patch.
seems good in general.
Thanks :-)
ToDo:
- handle force_empty.
Without this, we can do rmdir() against cgroup
(sorry, I sent the previous mail before completing it)
On 2008/05/23 11:10 +0900, Daisuke Nishimura wrote:
On 2008/05/22 16:44 +0900, KAMEZAWA Hiroyuki wrote:
(snip)
- make it possible for users to select if they use
this feature or not, and avoid overhead for users
not using this feature
Hi.
On 2008/05/23 6:27 +0900, Balbir Singh wrote:
Daisuke Nishimura wrote:
Hi.
I updated my swapcgroup patch.
Major changes from previous version(*1):
- Rebased on 2.6.26-rc2-mm1 + KAMEZAWA-san's performance
improvement patchset v4.
- Implemented as a add-on to memory cgroup.
So
1 - 100 of 114 matches
Mail list logo
