0001 R11: 0001 R12:
>>> ea00103ef0c0
>>> [ 7691.367745] R13: 880424cd6000 R14: R15:
>>> 880424cd622c
>>> [ 7691.375738] FS: () GS:88043fc2()
>>> knlGS:
>>> [ 7
01 7e
[ 7691.528638] RIP [8119dcb4] mem_cgroup_move_account+0xf4/0x290
Add the required __this_cpu_read().
Sorry for my mistake and thanks for the fix up, it looks good to me.
Reviewed-by: Sha Zhengju handai@taobao.com
Thanks,
Sha
Signed-off-by: Johannes Weiner han
On Wed, Jul 10, 2013 at 2:03 PM, Konstantin Khlebnikov
wrote:
> Sha Zhengju wrote:
>>
>> Hi,
>>
>> On Mon, Jul 8, 2013 at 5:59 PM, Konstantin Khlebnikov
>> wrote:
>>>
>>> > This is proof of concept, just basic functionality for IO contro
On Wed, Jul 10, 2013 at 2:03 PM, Konstantin Khlebnikov
khlebni...@openvz.org wrote:
Sha Zhengju wrote:
Hi,
On Mon, Jul 8, 2013 at 5:59 PM, Konstantin Khlebnikov
khlebni...@openvz.org wrote:
This is proof of concept, just basic functionality for IO controller.
This cgroup will control
writeback: get rid of account_size argument in
> cancel_dirty_page()
> c575ef6 hugetlbfs: remove cancel_dirty_page() from truncate_huge_page()
> b720923 nfs: remove redundant cancel_dirty_page() from nfs_wb_page_cancel()
> 4c21e52 mm: remove redundant dirty pages check from __delete_f
cancel_dirty_page() from nfs_wb_page_cancel()
4c21e52 mm: remove redundant dirty pages check from __delete_from_page_cache()
Signed-off-by: Konstantin Khlebnikov khlebni...@openvz.org
Cc: cgro...@vger.kernel.org
Cc: de...@openvz.org
Cc: Michal Hocko mho...@suse.cz
Cc: Sha Zhengju handai@gmail.com
On Fri, May 3, 2013 at 5:11 PM, Michal Hocko wrote:
> On Wed 02-01-13 11:44:21, Michal Hocko wrote:
>> On Wed 26-12-12 01:26:07, Sha Zhengju wrote:
>> > From: Sha Zhengju
>> >
>> > This patch adds memcg routines to count dirty pages, which allows memo
On Fri, May 3, 2013 at 5:11 PM, Michal Hocko mho...@suse.cz wrote:
On Wed 02-01-13 11:44:21, Michal Hocko wrote:
On Wed 26-12-12 01:26:07, Sha Zhengju wrote:
From: Sha Zhengju handai@taobao.com
This patch adds memcg routines to count dirty pages, which allows memory
controller
Commit-ID: 1c3e826482ab698e418c7a894440e62c76aac893
Gitweb: http://git.kernel.org/tip/1c3e826482ab698e418c7a894440e62c76aac893
Author: Sha Zhengju
AuthorDate: Wed, 20 Feb 2013 17:14:38 +0800
Committer: Ingo Molnar
CommitDate: Wed, 20 Feb 2013 11:39:24 +0100
sched/core: Remove
From: Sha Zhengju
Signed-off-by: Sha Zhengju
---
include/linux/sched.h |1 -
kernel/sched/core.c | 22 ++
2 files changed, 2 insertions(+), 21 deletions(-)
diff --git a/include/linux/sched.h b/include/linux/sched.h
index 33cc421..f9ca237d 100644
--- a/include/linux
On Wed, Feb 20, 2013 at 4:31 PM, Xiaotian Feng wrote:
> On Wed, Feb 20, 2013 at 3:37 PM, Sha Zhengju wrote:
>> From: Sha Zhengju
>>
>> Signed-off-by: Sha Zhengju
>> ---
>> kernel/sched/core.c | 17 -
>> 1 file changed, 17 deletions(
From: Sha Zhengju
Signed-off-by: Sha Zhengju
---
kernel/sched/core.c | 22 ++
1 file changed, 2 insertions(+), 20 deletions(-)
diff --git a/kernel/sched/core.c b/kernel/sched/core.c
index 03d7784..b7b03cd 100644
--- a/kernel/sched/core.c
+++ b/kernel/sched/core.c
From: Sha Zhengju handai@taobao.com
Signed-off-by: Sha Zhengju handai@taobao.com
---
kernel/sched/core.c | 22 ++
1 file changed, 2 insertions(+), 20 deletions(-)
diff --git a/kernel/sched/core.c b/kernel/sched/core.c
index 03d7784..b7b03cd 100644
--- a/kernel
On Wed, Feb 20, 2013 at 4:31 PM, Xiaotian Feng xtf...@gmail.com wrote:
On Wed, Feb 20, 2013 at 3:37 PM, Sha Zhengju handai@gmail.com wrote:
From: Sha Zhengju handai@taobao.com
Signed-off-by: Sha Zhengju handai@taobao.com
---
kernel/sched/core.c | 17 -
1 file
From: Sha Zhengju handai@taobao.com
Signed-off-by: Sha Zhengju handai@taobao.com
---
include/linux/sched.h |1 -
kernel/sched/core.c | 22 ++
2 files changed, 2 insertions(+), 21 deletions(-)
diff --git a/include/linux/sched.h b/include/linux/sched.h
index
Commit-ID: 1c3e826482ab698e418c7a894440e62c76aac893
Gitweb: http://git.kernel.org/tip/1c3e826482ab698e418c7a894440e62c76aac893
Author: Sha Zhengju handai@taobao.com
AuthorDate: Wed, 20 Feb 2013 17:14:38 +0800
Committer: Ingo Molnar mi...@kernel.org
CommitDate: Wed, 20 Feb 2013 11:39
From: Sha Zhengju
Signed-off-by: Sha Zhengju
---
kernel/sched/core.c | 17 -
1 file changed, 17 deletions(-)
diff --git a/kernel/sched/core.c b/kernel/sched/core.c
index c0b07c3..792f6fc 100644
--- a/kernel/sched/core.c
+++ b/kernel/sched/core.c
@@ -1957,23 +1957,6
From: Sha Zhengju handai@taobao.com
Signed-off-by: Sha Zhengju handai@taobao.com
---
kernel/sched/core.c | 17 -
1 file changed, 17 deletions(-)
diff --git a/kernel/sched/core.c b/kernel/sched/core.c
index c0b07c3..792f6fc 100644
--- a/kernel/sched/core.c
+++ b/kernel
On Mon, Jan 14, 2013 at 10:55 PM, Glauber Costa wrote:
> On 01/14/2013 12:34 AM, Sha Zhengju wrote:
>>> + struct kernel_cpustat *kcpustat = this_cpu_ptr(ca->cpustat);
>>> > +
>>> > kcpustat = this_cpu_ptr(ca->cpu
On Mon, Jan 14, 2013 at 10:55 PM, Glauber Costa glom...@parallels.com wrote:
On 01/14/2013 12:34 AM, Sha Zhengju wrote:
+ struct kernel_cpustat *kcpustat = this_cpu_ptr(ca-cpustat);
+
kcpustat = this_cpu_ptr(ca-cpustat);
Is this reassignment unnecessary
On Wed, Jan 9, 2013 at 7:45 PM, Glauber Costa wrote:
> From: Tejun Heo
>
> cpuacct being on a separate hierarchy is one of the main cgroup
> related complaints from scheduler side and the consensus seems to be
>
> * Allowing cpuacct to be a separate controller was a mistake. In
> general
On Wed, Jan 9, 2013 at 7:45 PM, Glauber Costa glom...@parallels.com wrote:
From: Tejun Heo t...@kernel.org
cpuacct being on a separate hierarchy is one of the main cgroup
related complaints from scheduler side and the consensus seems to be
* Allowing cpuacct to be a separate controller was a
On Thu, Jan 10, 2013 at 9:39 PM, Michal Hocko wrote:
> On Thu 10-01-13 19:43:58, Sha Zhengju wrote:
>> From: Sha Zhengju
>
> THP are not swapped out because they are split before so this change
> doesn't make much sense to me.
Yes... I have been pu
From: Sha Zhengju
Signed-off-by: Sha Zhengju
---
mm/memcontrol.c | 13 ++---
1 file changed, 6 insertions(+), 7 deletions(-)
diff --git a/mm/memcontrol.c b/mm/memcontrol.c
index 3817460..674cf21 100644
--- a/mm/memcontrol.c
+++ b/mm/memcontrol.c
@@ -914,10 +914,9 @@ static long
On Thu, Jan 10, 2013 at 1:03 PM, Kamezawa Hiroyuki
wrote:
> (2013/01/10 13:26), Sha Zhengju wrote:
>
>> But this method also has its pros and cons(e.g. need lock nesting). So
>> I doubt whether the following is able to deal with these issues all
>> together:
>> (CP
On Thu, Jan 10, 2013 at 9:39 PM, Michal Hocko mho...@suse.cz wrote:
On Thu 10-01-13 19:43:58, Sha Zhengju wrote:
From: Sha Zhengju handai@taobao.com
THP are not swapped out because they are split before so this change
doesn't make much sense to me.
Yes... I have been puzzled by 'nr_pages
On Thu, Jan 10, 2013 at 1:03 PM, Kamezawa Hiroyuki
kamezawa.hir...@jp.fujitsu.com wrote:
(2013/01/10 13:26), Sha Zhengju wrote:
But this method also has its pros and cons(e.g. need lock nesting). So
I doubt whether the following is able to deal with these issues all
together:
(CPU-A does
From: Sha Zhengju handai@taobao.com
Signed-off-by: Sha Zhengju handai@taobao.com
---
mm/memcontrol.c | 13 ++---
1 file changed, 6 insertions(+), 7 deletions(-)
diff --git a/mm/memcontrol.c b/mm/memcontrol.c
index 3817460..674cf21 100644
--- a/mm/memcontrol.c
+++ b/mm
On Thu, Jan 10, 2013 at 10:16 AM, Kamezawa Hiroyuki
wrote:
> (2013/01/10 0:02), Sha Zhengju wrote:
>>
>> On Mon, Jan 7, 2013 at 3:25 PM, Kamezawa Hiroyuki
>> wrote:
>>>
>>> (2013/01/05 13:48), Sha Zhengju wrote:
>>>>
>>&g
On Mon, Jan 7, 2013 at 3:25 PM, Kamezawa Hiroyuki
wrote:
> (2013/01/05 13:48), Sha Zhengju wrote:
>>
>> On Wed, Jan 2, 2013 at 6:44 PM, Michal Hocko wrote:
>>>
>>> On Wed 26-12-12 01:26:07, Sha Zhengju wrote:
>>>>
>>>> From: Sha Zhengju
Hi Hugh,
On Mon, Jan 7, 2013 at 4:02 AM, Hugh Dickins wrote:
> On Sat, 5 Jan 2013, Sha Zhengju wrote:
>> On Wed, Jan 2, 2013 at 6:44 PM, Michal Hocko wrote:
>> >
>> > Maybe I have missed some other locking which would prevent this from
>> > happening
On Mon, Jan 7, 2013 at 4:07 AM, Greg Thelen wrote:
> On Tue, Dec 25 2012, Sha Zhengju wrote:
>
>> From: Sha Zhengju
>>
>> This patch adds memcg routines to count dirty pages, which allows memory
>> controller
>> to maintain an accurate view of the amount of
On Mon, Jan 7, 2013 at 4:07 AM, Greg Thelen wrote:
> On Tue, Dec 25 2012, Sha Zhengju wrote:
>
>> From: Sha Zhengju
>>
>> Similar to dirty page, we add per cgroup writeback pages accounting. The lock
>> rule still is:
>> mem_cgroup_begin_updat
On Mon, Jan 7, 2013 at 4:07 AM, Greg Thelen gthe...@google.com wrote:
On Tue, Dec 25 2012, Sha Zhengju wrote:
From: Sha Zhengju handai@taobao.com
Similar to dirty page, we add per cgroup writeback pages accounting. The lock
rule still is:
mem_cgroup_begin_update_page_stat
On Mon, Jan 7, 2013 at 4:07 AM, Greg Thelen gthe...@google.com wrote:
On Tue, Dec 25 2012, Sha Zhengju wrote:
From: Sha Zhengju handai@taobao.com
This patch adds memcg routines to count dirty pages, which allows memory
controller
to maintain an accurate view of the amount of its dirty
Hi Hugh,
On Mon, Jan 7, 2013 at 4:02 AM, Hugh Dickins hu...@google.com wrote:
On Sat, 5 Jan 2013, Sha Zhengju wrote:
On Wed, Jan 2, 2013 at 6:44 PM, Michal Hocko mho...@suse.cz wrote:
Maybe I have missed some other locking which would prevent this from
happening but the locking relations
On Mon, Jan 7, 2013 at 3:25 PM, Kamezawa Hiroyuki
kamezawa.hir...@jp.fujitsu.com wrote:
(2013/01/05 13:48), Sha Zhengju wrote:
On Wed, Jan 2, 2013 at 6:44 PM, Michal Hocko mho...@suse.cz wrote:
On Wed 26-12-12 01:26:07, Sha Zhengju wrote:
From: Sha Zhengju handai@taobao.com
This patch
On Thu, Jan 10, 2013 at 10:16 AM, Kamezawa Hiroyuki
kamezawa.hir...@jp.fujitsu.com wrote:
(2013/01/10 0:02), Sha Zhengju wrote:
On Mon, Jan 7, 2013 at 3:25 PM, Kamezawa Hiroyuki
kamezawa.hir...@jp.fujitsu.com wrote:
(2013/01/05 13:48), Sha Zhengju wrote:
On Wed, Jan 2, 2013 at 6:44 PM
On Fri, Dec 28, 2012 at 9:10 AM, Kamezawa Hiroyuki
wrote:
> (2012/12/26 2:28), Sha Zhengju wrote:
>> From: Sha Zhengju
>>
>> Signed-off-by: Sha Zhengju
>
> I don't think your words are bad but it may be better to sync with meminfo's
> text.
>
>> ---
>&
On Fri, Dec 28, 2012 at 9:45 AM, Kamezawa Hiroyuki
wrote:
> (2012/12/26 2:27), Sha Zhengju wrote:
>> From: Sha Zhengju
>>
>> It's inspired by a similar optimization from Glauber Costa
>> (memcg: make it suck faster; https://lkml.org/lkml/2012/9/25/154).
>>
On Wed, Jan 2, 2013 at 8:27 PM, Michal Hocko wrote:
> On Wed 26-12-12 01:27:27, Sha Zhengju wrote:
>> From: Sha Zhengju
>>
>> If memcg is enabled and no non-root memcg exists, all allocated pages
>> belongs to root_mem_cgroup and go through root memcg statistics ro
On Wed, Jan 2, 2013 at 8:27 PM, Michal Hocko mho...@suse.cz wrote:
On Wed 26-12-12 01:27:27, Sha Zhengju wrote:
From: Sha Zhengju handai@taobao.com
If memcg is enabled and no non-root memcg exists, all allocated pages
belongs to root_mem_cgroup and go through root memcg statistics
On Fri, Dec 28, 2012 at 9:45 AM, Kamezawa Hiroyuki
kamezawa.hir...@jp.fujitsu.com wrote:
(2012/12/26 2:27), Sha Zhengju wrote:
From: Sha Zhengju handai@taobao.com
It's inspired by a similar optimization from Glauber Costa
(memcg: make it suck faster; https://lkml.org/lkml/2012/9/25/154
On Fri, Dec 28, 2012 at 9:10 AM, Kamezawa Hiroyuki
kamezawa.hir...@jp.fujitsu.com wrote:
(2012/12/26 2:28), Sha Zhengju wrote:
From: Sha Zhengju handai@taobao.com
Signed-off-by: Sha Zhengju handai@taobao.com
I don't think your words are bad but it may be better to sync with meminfo's
On Fri, Dec 28, 2012 at 9:04 AM, Kamezawa Hiroyuki
wrote:
> (2012/12/26 2:27), Sha Zhengju wrote:
>> From: Sha Zhengju
>>
>> If memcg is enabled and no non-root memcg exists, all allocated pages
>> belongs to root_mem_cgroup and go through root memcg statistics ro
On Wed, Jan 2, 2013 at 6:44 PM, Michal Hocko wrote:
> On Wed 26-12-12 01:26:07, Sha Zhengju wrote:
>> From: Sha Zhengju
>>
>> This patch adds memcg routines to count dirty pages, which allows memory
>> controller
>> to maintain an accurate view of the a
Hi Michal,
Sorry for my late response, I'm just back from vocation. : )
On Wed, Jan 2, 2013 at 5:08 PM, Michal Hocko wrote:
> On Wed 26-12-12 01:22:36, Sha Zhengju wrote:
>> From: Sha Zhengju
>>
>> Commit a8e7d49a(Fix race in create_empty_buffers() vs
>&g
Hi Kame,
Sorry for the late response, I'm just back from vocation. : )
On Fri, Dec 28, 2012 at 8:39 AM, Kamezawa Hiroyuki
wrote:
> (2012/12/26 2:22), Sha Zhengju wrote:
>> From: Sha Zhengju
>>
>> Commit a8e7d49a(Fix race in create_empty_buffers() vs
>> __set_page_d
Hi Kame,
Sorry for the late response, I'm just back from vocation. : )
On Fri, Dec 28, 2012 at 8:39 AM, Kamezawa Hiroyuki
kamezawa.hir...@jp.fujitsu.com wrote:
(2012/12/26 2:22), Sha Zhengju wrote:
From: Sha Zhengju handai@taobao.com
Commit a8e7d49a(Fix race in create_empty_buffers() vs
Hi Michal,
Sorry for my late response, I'm just back from vocation. : )
On Wed, Jan 2, 2013 at 5:08 PM, Michal Hocko mho...@suse.cz wrote:
On Wed 26-12-12 01:22:36, Sha Zhengju wrote:
From: Sha Zhengju handai@taobao.com
Commit a8e7d49a(Fix race in create_empty_buffers() vs
On Wed, Jan 2, 2013 at 6:44 PM, Michal Hocko mho...@suse.cz wrote:
On Wed 26-12-12 01:26:07, Sha Zhengju wrote:
From: Sha Zhengju handai@taobao.com
This patch adds memcg routines to count dirty pages, which allows memory
controller
to maintain an accurate view of the amount of its dirty
On Fri, Dec 28, 2012 at 9:04 AM, Kamezawa Hiroyuki
kamezawa.hir...@jp.fujitsu.com wrote:
(2012/12/26 2:27), Sha Zhengju wrote:
From: Sha Zhengju handai@taobao.com
If memcg is enabled and no non-root memcg exists, all allocated pages
belongs to root_mem_cgroup and go through root memcg
From: Sha Zhengju
If memcg is enabled and no non-root memcg exists, all allocated pages
belongs to root_mem_cgroup and go through root memcg statistics routines
which brings some overheads. So for the sake of performance, we can give
up accounting stats of root memcg
From: Sha Zhengju
Similar to dirty page, we add per cgroup writeback pages accounting. The lock
rule still is:
mem_cgroup_begin_update_page_stat()
modify page WRITEBACK stat
mem_cgroup_update_page_stat()
mem_cgroup_end_update_page_stat()
There're two writeback
From: Sha Zhengju
Signed-off-by: Sha Zhengju
---
Documentation/cgroups/memory.txt |2 ++
1 file changed, 2 insertions(+)
diff --git a/Documentation/cgroups/memory.txt b/Documentation/cgroups/memory.txt
index addb1f1..2828164 100644
--- a/Documentation/cgroups/memory.txt
+++ b
From: Sha Zhengju
It's inspired by a similar optimization from Glauber Costa
(memcg: make it suck faster; https://lkml.org/lkml/2012/9/25/154).
Here we use jump label to patch the memcg page stat accounting code
in or out when not used. when the first non-root memcg comes to
life the code
From: Sha Zhengju
While accounting memcg page stat, it's not worth to use MEMCG_NR_FILE_MAPPED
as an extra layer of indirection because of the complexity and presumed
performance overhead. We can use MEM_CGROUP_STAT_FILE_MAPPED directly.
Signed-off-by: Sha Zhengju
Acked-by: KAMEZAWA Hiroyuki
From: Sha Zhengju
This patch adds memcg routines to count dirty pages, which allows memory
controller
to maintain an accurate view of the amount of its dirty memory and can provide
some
info for users while cgroup's direct reclaim is working.
After Kame's commit 89c06bd5(memcg: use new logic
From: Sha Zhengju
Following we will treat SetPageDirty and dirty page accounting as an integrated
operation. Filesystems had better use vfs interface directly to avoid those
details.
Signed-off-by: Sha Zhengju
Acked-by: Sage Weil
---
fs/buffer.c |3 ++-
fs/ceph/addr.c
From: Sha Zhengju
Commit a8e7d49a(Fix race in create_empty_buffers() vs
__set_page_dirty_buffers())
extracts TestSetPageDirty from __set_page_dirty and is far away from
account_page_dirtied. But it's better to make the two operations in one single
function to keep modular. So in order to avoid
comed. : )
Change log:
v3 <--v2
1. change lock order of mapping->tree_lock and memcg->move_lock
2. performance optimization in 6/8 and 7/8
v2 <-- v1:
1. add test numbers
2. some small fix and comments
Sha Zhengju (8):
memcg-remove-MEMCG_NR_FILE_MAPP
1. change lock order of mapping-tree_lock and memcg-move_lock
2. performance optimization in 6/8 and 7/8
v2 -- v1:
1. add test numbers
2. some small fix and comments
Sha Zhengju (8):
memcg-remove-MEMCG_NR_FILE_MAPPED.patch
Make-TestSetPageDirty-and-dirty
From: Sha Zhengju handai@taobao.com
Commit a8e7d49a(Fix race in create_empty_buffers() vs
__set_page_dirty_buffers())
extracts TestSetPageDirty from __set_page_dirty and is far away from
account_page_dirtied. But it's better to make the two operations in one single
function to keep modular
From: Sha Zhengju handai@taobao.com
Following we will treat SetPageDirty and dirty page accounting as an integrated
operation. Filesystems had better use vfs interface directly to avoid those
details.
Signed-off-by: Sha Zhengju handai@taobao.com
Acked-by: Sage Weil s...@inktank.com
From: Sha Zhengju handai@taobao.com
This patch adds memcg routines to count dirty pages, which allows memory
controller
to maintain an accurate view of the amount of its dirty memory and can provide
some
info for users while cgroup's direct reclaim is working.
After Kame's commit 89c06bd5
From: Sha Zhengju handai@taobao.com
While accounting memcg page stat, it's not worth to use MEMCG_NR_FILE_MAPPED
as an extra layer of indirection because of the complexity and presumed
performance overhead. We can use MEM_CGROUP_STAT_FILE_MAPPED directly.
Signed-off-by: Sha Zhengju handai
From: Sha Zhengju handai@taobao.com
It's inspired by a similar optimization from Glauber Costa
(memcg: make it suck faster; https://lkml.org/lkml/2012/9/25/154).
Here we use jump label to patch the memcg page stat accounting code
in or out when not used. when the first non-root memcg comes
From: Sha Zhengju handai@taobao.com
Signed-off-by: Sha Zhengju handai@taobao.com
---
Documentation/cgroups/memory.txt |2 ++
1 file changed, 2 insertions(+)
diff --git a/Documentation/cgroups/memory.txt b/Documentation/cgroups/memory.txt
index addb1f1..2828164 100644
From: Sha Zhengju handai@taobao.com
Similar to dirty page, we add per cgroup writeback pages accounting. The lock
rule still is:
mem_cgroup_begin_update_page_stat()
modify page WRITEBACK stat
mem_cgroup_update_page_stat()
mem_cgroup_end_update_page_stat
From: Sha Zhengju handai@taobao.com
If memcg is enabled and no non-root memcg exists, all allocated pages
belongs to root_mem_cgroup and go through root memcg statistics routines
which brings some overheads. So for the sake of performance, we can give
up accounting stats of root memcg
On 11/09/2012 06:50 PM, Michal Hocko wrote:
On Fri 09-11-12 18:23:07, Sha Zhengju wrote:
On 11/09/2012 12:25 AM, Michal Hocko wrote:
On Thu 08-11-12 23:52:47, Sha Zhengju wrote:
[...]
+ for (i = 0; i< MEM_CGROUP_STAT_NSTATS; i++) {
+ long long val
On 11/09/2012 04:12 PM, Kamezawa Hiroyuki wrote:
> (2012/11/09 0:52), Sha Zhengju wrote:
>> From: Sha Zhengju
>>
>> Current when a memcg oom is happening the oom dump messages is still global
>> state and provides few useful info for users. This patch prints more point
On 11/09/2012 12:25 AM, Michal Hocko wrote:
On Thu 08-11-12 23:52:47, Sha Zhengju wrote:
[...]
(2) After change
[ 269.225628] mal invoked oom-killer: gfp_mask=0xd0, order=0, oom_score_adj=0
[ 269.225633] mal cpuset=/ mems_allowed=0-1
[ 269.225636] Pid: 4616, comm: mal Not tainted 3.6.0+ #25
On 11/09/2012 12:25 AM, Michal Hocko wrote:
On Thu 08-11-12 23:52:47, Sha Zhengju wrote:
[...]
(2) After change
[ 269.225628] mal invoked oom-killer: gfp_mask=0xd0, order=0, oom_score_adj=0
[ 269.225633] mal cpuset=/ mems_allowed=0-1
[ 269.225636] Pid: 4616, comm: mal Not tainted 3.6.0+ #25
On 11/09/2012 04:12 PM, Kamezawa Hiroyuki wrote:
(2012/11/09 0:52), Sha Zhengju wrote:
From: Sha Zhengju handai@taobao.com
Current when a memcg oom is happening the oom dump messages is still global
state and provides few useful info for users. This patch prints more pointed
memcg page
On 11/09/2012 06:50 PM, Michal Hocko wrote:
On Fri 09-11-12 18:23:07, Sha Zhengju wrote:
On 11/09/2012 12:25 AM, Michal Hocko wrote:
On Thu 08-11-12 23:52:47, Sha Zhengju wrote:
[...]
+ for (i = 0; i MEM_CGROUP_STAT_NSTATS; i++) {
+ long long val = 0
From: Sha Zhengju
Current when a memcg oom is happening the oom dump messages is still global
state and provides few useful info for users. This patch prints more pointed
memcg page statistics for memcg-oom.
We set up a simple cgroup hierarchy for test:
root_memcg
On 11/08/2012 06:34 AM, Michal Hocko wrote:
On Wed 07-11-12 16:41:59, Sha Zhengju wrote:
From: Sha Zhengju
If memcg oom happening, don't scan all system tasks to dump memory state of
eligible tasks, instead we iterates only over the process attached to the oom
memcg and avoid the rcu lock
On 11/08/2012 05:07 PM, Kamezawa Hiroyuki wrote:
> (2012/11/07 17:41), Sha Zhengju wrote:
>> From: Sha Zhengju
>>
>> Current, when a memcg oom is happening the oom dump messages is still global
>> state and provides few useful info for users. This patch prints
On 11/08/2012 06:17 AM, Michal Hocko wrote:
On Wed 07-11-12 16:41:36, Sha Zhengju wrote:
From: Sha Zhengju
Current, when a memcg oom is happening the oom dump messages is still global
state and provides few useful info for users. This patch prints more pointed
memcg page statistics for memcg
On 11/08/2012 02:02 AM, David Rientjes wrote:
On Wed, 7 Nov 2012, Sha Zhengju wrote:
diff --git a/mm/memcontrol.c b/mm/memcontrol.c
index 0eab7d5..2df5e72 100644
--- a/mm/memcontrol.c
+++ b/mm/memcontrol.c
@@ -118,6 +118,14 @@ static const char * const mem_cgroup_events_names
On 11/08/2012 02:02 AM, David Rientjes wrote:
On Wed, 7 Nov 2012, Sha Zhengju wrote:
diff --git a/mm/memcontrol.c b/mm/memcontrol.c
index 0eab7d5..2df5e72 100644
--- a/mm/memcontrol.c
+++ b/mm/memcontrol.c
@@ -118,6 +118,14 @@ static const char * const mem_cgroup_events_names
On 11/08/2012 06:17 AM, Michal Hocko wrote:
On Wed 07-11-12 16:41:36, Sha Zhengju wrote:
From: Sha Zhengjuhandai@taobao.com
Current, when a memcg oom is happening the oom dump messages is still global
state and provides few useful info for users. This patch prints more pointed
memcg page
On 11/08/2012 05:07 PM, Kamezawa Hiroyuki wrote:
(2012/11/07 17:41), Sha Zhengju wrote:
From: Sha Zhengju handai@taobao.com
Current, when a memcg oom is happening the oom dump messages is still global
state and provides few useful info for users. This patch prints more pointed
memcg page
On 11/08/2012 06:34 AM, Michal Hocko wrote:
On Wed 07-11-12 16:41:59, Sha Zhengju wrote:
From: Sha Zhengjuhandai@taobao.com
If memcg oom happening, don't scan all system tasks to dump memory state of
eligible tasks, instead we iterates only over the process attached to the oom
memcg
From: Sha Zhengju handai@taobao.com
Current when a memcg oom is happening the oom dump messages is still global
state and provides few useful info for users. This patch prints more pointed
memcg page statistics for memcg-oom.
We set up a simple cgroup hierarchy for test:
root_memcg
From: Sha Zhengju
If memcg oom happening, don't scan all system tasks to dump memory state of
eligible tasks, instead we iterates only over the process attached to the oom
memcg and avoid the rcu lock.
Signed-off-by: Sha Zhengju
Cc: Michal Hocko
Cc: KAMEZAWA Hiroyuki
Cc: David Rientjes
Cc
From: Sha Zhengju
Current, when a memcg oom is happening the oom dump messages is still global
state and provides few useful info for users. This patch prints more pointed
memcg page statistics for memcg-oom.
Signed-off-by: Sha Zhengju
Cc: Michal Hocko
Cc: KAMEZAWA Hiroyuki
Cc: David
From: Sha Zhengju
When memcg oom is happening the current memcg related dump information
is limited for debugging. The patches provide more detailed memcg page
statistics
and also take hierarchy into consideration.
The previous primitive version can be reached here:
https://lkml.org/lkml/2012
From: Sha Zhengju handai@taobao.com
When memcg oom is happening the current memcg related dump information
is limited for debugging. The patches provide more detailed memcg page
statistics
and also take hierarchy into consideration.
The previous primitive version can be reached here:
https
From: Sha Zhengju handai@taobao.com
Current, when a memcg oom is happening the oom dump messages is still global
state and provides few useful info for users. This patch prints more pointed
memcg page statistics for memcg-oom.
Signed-off-by: Sha Zhengju handai@taobao.com
Cc: Michal Hocko
From: Sha Zhengju handai@taobao.com
If memcg oom happening, don't scan all system tasks to dump memory state of
eligible tasks, instead we iterates only over the process attached to the oom
memcg and avoid the rcu lock.
Signed-off-by: Sha Zhengju handai@taobao.com
Cc: Michal Hocko mho
On 10/23/2012 11:35 AM, Qiang Gao wrote:
information about the system is in the attach file "information.txt"
I can not reproduce it in the upstream 3.6.0 kernel..
On Sat, Oct 20, 2012 at 12:04 AM, Michal Hocko wrote:
On Wed 17-10-12 18:23:34, gaoqiang wrote:
I looked up nothing useful with
On 10/23/2012 11:35 AM, Qiang Gao wrote:
information about the system is in the attach file information.txt
I can not reproduce it in the upstream 3.6.0 kernel..
On Sat, Oct 20, 2012 at 12:04 AM, Michal Hockomho...@suse.cz wrote:
On Wed 17-10-12 18:23:34, gaoqiang wrote:
I looked up nothing
On 10/18/2012 11:32 PM, Michal Hocko wrote:
On Thu 18-10-12 21:51:57, Sha Zhengju wrote:
On 10/18/2012 07:56 PM, Michal Hocko wrote:
On Wed 17-10-12 01:14:48, Sha Zhengju wrote:
On Tuesday, October 16, 2012, Michal Hocko wrote:
[...]
Could you be more specific about the motivation
On 10/18/2012 07:56 PM, Michal Hocko wrote:
On Wed 17-10-12 01:14:48, Sha Zhengju wrote:
On Tuesday, October 16, 2012, Michal Hocko wrote:
[...]
Could you be more specific about the motivation for this patch? Is it
"let's be consistent with the global oom" or you have a rea
On 10/18/2012 11:32 PM, Michal Hocko wrote:
On Thu 18-10-12 21:51:57, Sha Zhengju wrote:
On 10/18/2012 07:56 PM, Michal Hocko wrote:
On Wed 17-10-12 01:14:48, Sha Zhengju wrote:
On Tuesday, October 16, 2012, Michal Hockomho...@suse.cz wrote:
[...]
Could you be more specific about
On 10/18/2012 07:56 PM, Michal Hocko wrote:
On Wed 17-10-12 01:14:48, Sha Zhengju wrote:
On Tuesday, October 16, 2012, Michal Hockomho...@suse.cz wrote:
[...]
Could you be more specific about the motivation for this patch? Is it
let's be consistent with the global oom or you have a real use
On 10/16/2012 06:20 PM, Ni zhan Chen wrote:
On 10/16/2012 06:12 PM, Sha Zhengju wrote:
From: Sha Zhengju
Sysctl oom_kill_allocating_task enables or disables killing the
OOM-triggering
task in out-of-memory situations, but it only works on overall
system-wide oom.
But it's also a useful
From: Sha Zhengju
Sysctl oom_kill_allocating_task enables or disables killing the OOM-triggering
task in out-of-memory situations, but it only works on overall system-wide oom.
But it's also a useful indication in memcg so we take it into consideration
while oom happening in memcg. Other sysctl
1 - 100 of 132 matches
Mail list logo