| lists.openwall.net | lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC | |
|
Open Source and information security mailing list archives
| ||
|
Message-Id: <20110623170811.16f4435f.kamezawa.hiroyu@jp.fujitsu.com>
Date: Thu, 23 Jun 2011 17:08:11 +0900
From: KAMEZAWA Hiroyuki <kamezawa.hiroyu@...fujitsu.com>
To: Michal Hocko <mhocko@...e.cz>
Cc: Christoph Hellwig <hch@...radead.org>, linux-mm@...ck.org,
linux-kernel@...r.kernel.org,
Andrew Morton <akpm@...ux-foundation.org>,
Hugh Dickins <hughd@...gle.com>,
Rik van Riel <riel@...hat.com>,
Michel Lespinasse <walken@...gle.com>,
Mel Gorman <mgorman@...e.de>, Lutz Vieweg <lvml@....de>
Subject: Re: [PATCH] memcg: unlock page before charging it. (WasRe: [PATCH
V2] mm: Do not keep page locked during page fault while charging it for
memcg
On Thu, 23 Jun 2011 09:41:33 +0200
Michal Hocko <mhocko@...e.cz> wrote:
> On Thu 23-06-11 15:08:42, KAMEZAWA Hiroyuki wrote:
> > On Wed, 22 Jun 2011 14:32:04 +0200
> > Michal Hocko <mhocko@...e.cz> wrote:
> >
> > > On Wed 22-06-11 08:15:16, Christoph Hellwig wrote:
> > > > > +
> > > > > + /* We have to drop the page lock here because memcg
> > > > > + * charging might block for unbound time if memcg oom
> > > > > + * killer is disabled.
> > > > > + */
> > > > > + unlock_page(vmf.page);
> > > > > + ret = mem_cgroup_newpage_charge(page, mm, GFP_KERNEL);
> > > > > + lock_page(vmf.page);
> > > >
> > > > This introduces a completely poinless unlock/lock cycle for non-memcg
> > > > pagefaults. Please make sure it only happens when actually needed.
> > >
> > > Fair point. Thanks!
> > > What about the following?
> > > I realize that pushing more memcg logic into mm/memory.c is not nice but
> > > I found it better than pushing the old page into mem_cgroup_newpage_charge.
> > > We could also check whether the old page is in the root cgroup because
> > > memcg oom killer is not active there but that would add more code into
> > > this hot path so I guess it is not worth it.
> > >
> > > Changes since v1
> > > - do not unlock page when memory controller is disabled.
> > >
> >
> > Great work. Then I confirmed Lutz' problem is fixed.
> >
> > But I like following style rather than additional lock/unlock.
> > How do you think ?
>
> Yes, I like it much more than the hairy way I did it. See comments bellow.
>
> > I tested this on the latest git tree and confirmed
> > the Lutz's livelock problem is fixed. And I think this should go stable tree.
> >
> >
> > ==
> > From 7e9250da9ff529958d4c1ff511458dbdac8e4b81 Mon Sep 17 00:00:00 2001
> > From: KAMEZAWA Hiroyuki <kamezawa.hiroyu@...fujitsu.com>
> > Date: Thu, 23 Jun 2011 15:05:57 +0900
> > Subject: [PATCH] memcg: unlock page before charging it.
> >
> > Currently we are keeping faulted page locked throughout whole __do_fault
> > call (except for page_mkwrite code path). If we do early COW we allocate a
> > new page which has to be charged for a memcg (mem_cgroup_newpage_charge).
> >
> > This function, however, might block for unbounded amount of time if memcg
> > oom killer is disabled or fork-bomb is running because the only way out of
> > the OOM situation is either an external event or OOM-situation fix.
> >
> > processes from faulting it in which is not good at all because we are
>
> Missing the beginning of the sentence?
>
Ah, yes...
> > basically punishing potentially an unrelated process for OOM condition
> > in a different group (I have seen stuck system because of ld-2.11.1.so being
> > locked).
> >
> > We can do test easily.
> > % cgcreate -g memory:A
> > % cgset -r memory.limit_in_bytes=64M A
> > % cgset -r memory.memsw.limit_in_bytes=64M A
> > % cd kernel_dir; cgexec -g memory:A make -j
> >
> > Then, the whole system will live-locked until you kill 'make -j'
> > by hands (or push reboot...) This is because some important
> > page in a shared library are locked and never released bcause of fork-bomb.
> >
> > This patch delays "charge" until unlock_page() called. There is
> > no problem as far as we keep reference on a page.
> > (memcg doesn't require page_lock()).
> >
> > Then, above livelock disappears.
> >
> > Reported-by: Lutz Vieweg <lvml@....de>
> > Original-idea-by: Michal Hocko <mhocko@...e.cz>
> > Signed-off-by: KAMEZAWA Hiroyuki <kamezawa.hiroyu@...fujitsu.com>
> > ---
> > mm/memory.c | 28 +++++++++++++++++++---------
> > 1 files changed, 19 insertions(+), 9 deletions(-)
> >
> > diff --git a/mm/memory.c b/mm/memory.c
> > index 87d9353..66442da 100644
> > --- a/mm/memory.c
> > +++ b/mm/memory.c
> > @@ -3129,7 +3129,7 @@ static int __do_fault(struct mm_struct *mm, struct vm_area_struct *vma,
> > struct page *page;
> > pte_t entry;
> > int anon = 0;
> > - int charged = 0;
> > + struct page *need_charge = NULL;
> > struct page *dirty_page = NULL;
> > struct vm_fault vmf;
> > int ret;
> > @@ -3177,12 +3177,7 @@ static int __do_fault(struct mm_struct *mm, struct vm_area_struct *vma,
> > ret = VM_FAULT_OOM;
> > goto out;
> > }
> > - if (mem_cgroup_newpage_charge(page, mm, GFP_KERNEL)) {
> > - ret = VM_FAULT_OOM;
> > - page_cache_release(page);
> > - goto out;
> > - }
> > - charged = 1;
> > + need_charge = page;
> > copy_user_highpage(page, vmf.page, address, vma);
> > __SetPageUptodate(page);
> > } else {
> > @@ -3251,12 +3246,11 @@ static int __do_fault(struct mm_struct *mm, struct vm_area_struct *vma,
> > /* no need to invalidate: a not-present page won't be cached */
> > update_mmu_cache(vma, address, page_table);
> > } else {
> > - if (charged)
> > - mem_cgroup_uncharge_page(page);
> > if (anon)
> > page_cache_release(page);
> > else
> > anon = 1; /* no anon but release faulted_page */
> > + need_charge = NULL;
> > }
> >
> > pte_unmap_unlock(page_table, ptl);
> > @@ -3268,6 +3262,17 @@ out:
> > if (set_page_dirty(dirty_page))
> > page_mkwrite = 1;
> > unlock_page(dirty_page);
> > + if (need_charge) {
> > + /*
> > + * charge this page before we drop refcnt.
> > + * memory cgroup returns OOM condition when
> > + * this task is killed. So, it's not necesasry
> > + * to undo.
> > + */
> > + if (mem_cgroup_newpage_charge(need_charge,
> > + mm, GFP_KERNEL))
> > + ret = VM_FAULT_OOM;
> > + }
>
> We do not need this hunk, don't we? dirty_page is set only if !anon so
> we never get to this path from COW.
>
You're right. (And Nishimura pointed out this, too)
> Other than that:
> Reviewed-by: Michal Hocko <mhocko@...e.cz>
>
I found the page is added to LRU before charging. (In this case,
memcg's LRU is ignored.) I'll post a new version with a fix.
Thanks,
-Kame
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
Powered by blists - more mailing lists