From: Michal Hocko <mhocko@kernel.org>
To: Tetsuo Handa <penguin-kernel@i-love.sakura.ne.jp>
Cc: Andrey Ryabinin <aryabinin@virtuozzo.com>,
Thomas Lindroth <thomas.lindroth@gmail.com>,
linux-mm@kvack.org
Subject: Re: [BUG] kmemcg limit defeats __GFP_NOFAIL allocation
Date: Wed, 4 Sep 2019 13:59:33 +0200 [thread overview]
Message-ID: <20190904115933.GT3838@dhcp22.suse.cz> (raw)
In-Reply-To: <4d87d770-c110-224f-6c0c-d6fada90417d@i-love.sakura.ne.jp>
On Wed 04-09-19 20:32:25, Tetsuo Handa wrote:
> On 2019/09/04 20:25, Michal Hocko wrote:
> > On Wed 04-09-19 18:36:06, Tetsuo Handa wrote:
> > [...]
> >> The first bug is that __memcg_kmem_charge_memcg() in mm/memcontrol.c is
> >> failing to return 0 when it is a __GFP_NOFAIL allocation request.
> >> We should ignore limits when it is a __GFP_NOFAIL allocation request.
> >
> > OK, fixing that sounds like a reasonable thing to do.
> >
> >> If we force __memcg_kmem_charge_memcg() to return 0, then
> >>
> >> ----------
> >> struct page_counter *counter;
> >> int ret;
> >>
> >> + if (gfp & __GFP_NOFAIL)
> >> + return 0;
> >> +
> >> ret = try_charge(memcg, gfp, nr_pages);
> >> if (ret)
> >> return ret;
> >> ----------
> >
> > This should be more likely something like
> >
> > diff --git a/mm/memcontrol.c b/mm/memcontrol.c
> > index 9ec5e12486a7..05a4828edf9d 100644
> > --- a/mm/memcontrol.c
> > +++ b/mm/memcontrol.c
> > @@ -2820,7 +2820,8 @@ int __memcg_kmem_charge_memcg(struct page *page, gfp_t gfp, int order,
> > return ret;
> >
> > if (!cgroup_subsys_on_dfl(memory_cgrp_subsys) &&
> > - !page_counter_try_charge(&memcg->kmem, nr_pages, &counter)) {
> > + !page_counter_try_charge(&memcg->kmem, nr_pages, &counter) &&
> > + !(gfp_mask & __GFP_NOFAIL)) {
> > cancel_charge(memcg, nr_pages);
> > return -ENOMEM;
> > }
>
> Is it guaranteed that try_charge(__GFP_NOFAIL) never fails?
it enforces charges.
> >> the second bug that alloc_slabmgmt() in mm/slab.c is returning NULL
> >> when it is a __GFP_NOFAIL allocation request will appear.
> >> I don't know how to handle this.
> >
> > I am sorry, I do not follow, why would alloc_slabmgmt return NULL
> > with forcing gfp_nofail charges?
> >
>
> The reproducer is hitting
>
> @@ -2300,18 +2302,21 @@ static void *alloc_slabmgmt(struct kmem_cache *cachep,
> page->s_mem = addr + colour_off;
> page->active = 0;
>
> - if (OBJFREELIST_SLAB(cachep))
> + if (OBJFREELIST_SLAB(cachep)) {
> + BUG_ON(local_flags & __GFP_NOFAIL); // <= this condition
What does this bugon tries to say though. I am not an expert on slab bu
only OFF_SLAB(cachep) branch depends on an allocation. Others should
allocate object from the cache.
--
Michal Hocko
SUSE Labs
next prev parent reply other threads:[~2019-09-04 11:59 UTC|newest]
Thread overview: 29+ messages / expand[flat|nested] mbox.gz Atom feed top
2019-09-01 20:43 [BUG] Early OOM and kernel NULL pointer dereference in 4.19.69 Thomas Lindroth
2019-09-02 7:16 ` Michal Hocko
2019-09-02 7:27 ` Michal Hocko
2019-09-02 19:34 ` Thomas Lindroth
2019-09-03 7:41 ` Michal Hocko
2019-09-03 12:01 ` Thomas Lindroth
2019-09-03 12:05 ` Andrey Ryabinin
2019-09-03 12:22 ` Michal Hocko
2019-09-03 18:20 ` Thomas Lindroth
2019-09-03 19:36 ` Michal Hocko
[not found] ` <666dbcde-1b8a-9e2d-7d1f-48a117c78ae1@I-love.SAKURA.ne.jp>
2019-09-03 18:25 ` Thomas Lindroth
[not found] ` <4d0eda9a-319d-1a7d-1eed-71da90902367@i-love.sakura.ne.jp>
2019-09-04 11:25 ` [BUG] kmemcg limit defeats __GFP_NOFAIL allocation Michal Hocko
[not found] ` <4d87d770-c110-224f-6c0c-d6fada90417d@i-love.sakura.ne.jp>
2019-09-04 11:59 ` Michal Hocko [this message]
[not found] ` <0056063b-46ff-0ebd-ff0d-c96a1f9ae6b1@i-love.sakura.ne.jp>
2019-09-04 14:29 ` Michal Hocko
[not found] ` <405ce28b-c0b4-780c-c883-42d741ec60e0@i-love.sakura.ne.jp>
2019-09-05 23:11 ` Thomas Lindroth
2019-09-06 7:27 ` Michal Hocko
2019-09-06 10:54 ` Andrey Ryabinin
2019-09-06 11:29 ` Michal Hocko
2019-09-06 12:56 ` [PATCH] memcg, kmem: do not fail __GFP_NOFAIL charges Michal Hocko
2019-09-06 18:24 ` Shakeel Butt
2019-09-06 18:24 ` Shakeel Butt
2019-09-09 11:22 ` Michal Hocko
2019-09-11 12:00 ` Michal Hocko
2019-09-11 14:37 ` Andrew Morton
2019-09-11 15:16 ` Michal Hocko
2019-09-13 2:46 ` Shakeel Butt
2019-09-13 2:46 ` Shakeel Butt
2019-09-24 10:53 ` Michal Hocko
2019-09-24 23:06 ` Andrew Morton
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=20190904115933.GT3838@dhcp22.suse.cz \
--to=mhocko@kernel.org \
--cc=aryabinin@virtuozzo.com \
--cc=linux-mm@kvack.org \
--cc=penguin-kernel@i-love.sakura.ne.jp \
--cc=thomas.lindroth@gmail.com \
/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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.