From: Naoya Horiguchi <n-horiguchi@ah.jp.nec.com>
To: Balbir Singh <bsingharora@gmail.com>
Cc: Laurent Dufour <ldufour@linux.vnet.ibm.com>,
"linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>,
"linux-mm@kvack.org" <linux-mm@kvack.org>,
"akpm@linux-foundation.org" <akpm@linux-foundation.org>
Subject: Re: [PATCH v2 1/2] mm: Uncharge poisoned pages
Date: Wed, 26 Apr 2017 02:34:11 +0000 [thread overview]
Message-ID: <20170426023410.GA11619@hori1.linux.bs1.fc.nec.co.jp> (raw)
In-Reply-To: <1493171698.4828.1.camel@gmail.com>
On Wed, Apr 26, 2017 at 11:54:58AM +1000, Balbir Singh wrote:
> On Tue, 2017-04-25 at 16:27 +0200, Laurent Dufour wrote:
> > When page are poisoned, they should be uncharged from the root memory
> > cgroup.
> >
> > This is required to avoid a BUG raised when the page is onlined back:
> > BUG: Bad page state in process mem-on-off-test pfn:7ae3b
> > page:f000000001eb8ec0 count:0 mapcount:0 mapping: (null)
> > index:0x1
> > flags: 0x3ffff800200000(hwpoison)
> > raw: 003ffff800200000 0000000000000000 0000000000000001
> > 00000000ffffffff
> > raw: 5deadbeef0000100 5deadbeef0000200 0000000000000000
> > c0000007fe055800
> > page dumped because: page still charged to cgroup
> > page->mem_cgroup:c0000007fe055800
> > Modules linked in: pseries_rng rng_core vmx_crypto virtio_balloon
> > ip_tables x_tables autofs4 virtio_blk virtio_net virtio_pci
> > virtio_ring virtio
> > CPU: 34 PID: 5946 Comm: mem-on-off-test Tainted: G B 4.11.0-rc7-hwp
> > Call Trace:
> > [c0000007e4a737f0] [c000000000958e8c] dump_stack+0xb0/0xf0
> > (unreliable)
> > [c0000007e4a73830] [c00000000021588c] bad_page+0x11c/0x190
> > [c0000007e4a738c0] [c00000000021757c] free_pcppages_bulk+0x46c/0x600
> > [c0000007e4a73990] [c00000000021924c] free_hot_cold_page+0x2ec/0x320
> > [c0000007e4a739e0] [c0000000002a6440] generic_online_page+0x50/0x70
> > [c0000007e4a73a10] [c0000000002a6184] online_pages_range+0x94/0xe0
> > [c0000007e4a73a70] [c00000000005a2b0] walk_system_ram_range+0xe0/0x120
> > [c0000007e4a73ac0] [c0000000002cce44] online_pages+0x2b4/0x6b0
> > [c0000007e4a73b60] [c000000000600558] memory_subsys_online+0x218/0x270
> > [c0000007e4a73bf0] [c0000000005dec84] device_online+0xb4/0x110
> > [c0000007e4a73c30] [c000000000600f00] store_mem_state+0xc0/0x190
> > [c0000007e4a73c70] [c0000000005da1d4] dev_attr_store+0x34/0x60
> > [c0000007e4a73c90] [c000000000377c70] sysfs_kf_write+0x60/0xa0
> > [c0000007e4a73cb0] [c0000000003769fc] kernfs_fop_write+0x16c/0x240
> > [c0000007e4a73d00] [c0000000002d1b0c] __vfs_write+0x3c/0x1b0
> > [c0000007e4a73d90] [c0000000002d34dc] vfs_write+0xcc/0x230
> > [c0000007e4a73de0] [c0000000002d50e0] SyS_write+0x60/0x110
> > [c0000007e4a73e30] [c00000000000b760] system_call+0x38/0xfc
> >
> > Signed-off-by: Laurent Dufour <ldufour@linux.vnet.ibm.com>
> > ---
> > mm/memory-failure.c | 3 +++
> > 1 file changed, 3 insertions(+)
> >
> > diff --git a/mm/memory-failure.c b/mm/memory-failure.c
> > index 27f7210e7fab..22bd22eb25cb 100644
> > --- a/mm/memory-failure.c
> > +++ b/mm/memory-failure.c
> > @@ -529,6 +529,9 @@ static const char * const action_page_types[] = {
> > */
> > static int delete_from_lru_cache(struct page *p)
> > {
> > + if (memcg_kmem_enabled())
> > + memcg_kmem_uncharge(p, 0);
> > +
>
> The changelog is not quite clear, so we are uncharging a page using
> memcg_kmem_uncharge for a page in swap cache/page cache?
Hi Balbir,
Yes, in the normal page lifecycle, uncharge is done in page free time.
But in memory error handling case, in-use pages (i.e. swap cache and page
cache) are removed from normal path and they don't pass page freeing code.
So I think that this change is to keep the consistent charging for such a case.
- Naoya Horiguchi
next prev parent reply other threads:[~2017-04-26 3:09 UTC|newest]
Thread overview: 34+ messages / expand[flat|nested] mbox.gz Atom feed top
2017-04-25 14:27 [PATCH v2 0/2] BUG raised when onlining HWPoisoned page Laurent Dufour
2017-04-25 14:27 ` [PATCH v2 1/2] mm: Uncharge poisoned pages Laurent Dufour
2017-04-25 23:48 ` Naoya Horiguchi
2017-04-26 1:54 ` Balbir Singh
2017-04-26 2:34 ` Naoya Horiguchi [this message]
2017-04-26 3:45 ` Balbir Singh
2017-04-26 4:46 ` Naoya Horiguchi
2017-04-26 8:59 ` Balbir Singh
2017-04-28 9:32 ` Laurent Dufour
2017-04-27 14:37 ` Michal Hocko
2017-04-27 20:51 ` Andi Kleen
2017-04-28 6:07 ` Michal Hocko
2017-04-28 7:31 ` Michal Hocko
2017-04-28 9:17 ` Laurent Dufour
2017-04-28 13:48 ` Michal Hocko
2017-05-02 14:59 ` Laurent Dufour
2017-05-02 18:55 ` Michal Hocko
2017-05-03 11:34 ` Laurent Dufour
2017-05-04 1:21 ` Balbir Singh
2017-05-08 10:42 ` Laurent Dufour
2017-05-09 1:41 ` Balbir Singh
2017-05-08 2:58 ` Naoya Horiguchi
2017-05-09 9:18 ` Michal Hocko
2017-05-09 22:59 ` Naoya Horiguchi
2017-04-25 14:27 ` [PATCH v2 2/2] mm: skip HWPoisoned pages when onlining pages Laurent Dufour
2017-04-26 2:10 ` Balbir Singh
2017-04-26 3:13 ` Naoya Horiguchi
2017-04-28 2:51 ` Balbir Singh
2017-04-28 6:30 ` Michal Hocko
2017-04-28 6:50 ` Michal Hocko
2017-04-28 6:51 ` Michal Hocko
2017-05-10 7:41 ` Michal Hocko
2018-01-17 23:03 ` Andrew Morton
2018-01-23 18:15 ` Laurent Dufour
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20170426023410.GA11619@hori1.linux.bs1.fc.nec.co.jp \
--to=n-horiguchi@ah.jp.nec.com \
--cc=akpm@linux-foundation.org \
--cc=bsingharora@gmail.com \
--cc=ldufour@linux.vnet.ibm.com \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mm@kvack.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).