From: David Gibson <david@gibson.dropbear.id.au>
To: Davidlohr Bueso <davidlohr@hp.com>
Cc: Joonsoo Kim <iamjoonsoo.kim@lge.com>,
Andrew Morton <akpm@linux-foundation.org>,
Rik van Riel <riel@redhat.com>, Mel Gorman <mgorman@suse.de>,
Michal Hocko <mhocko@suse.cz>,
"Aneesh Kumar K.V" <aneesh.kumar@linux.vnet.ibm.com>,
KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>,
Hugh Dickins <hughd@google.com>,
Davidlohr Bueso <davidlohr.bueso@hp.com>,
linux-mm@kvack.org, linux-kernel@vger.kernel.org,
Wanpeng Li <liwanp@linux.vnet.ibm.com>,
Naoya Horiguchi <n-horiguchi@ah.jp.nec.com>,
Hillf Danton <dhillf@gmail.com>
Subject: Re: [PATCH 17/18] mm, hugetlb: retry if we fail to allocate a hugepage with use_reserve
Date: Wed, 7 Aug 2013 11:03:12 +1000 [thread overview]
Message-ID: <20130807010312.GA17110@voom.redhat.com> (raw)
In-Reply-To: <1375834724.2134.49.camel@buesod1.americas.hpqcorp.net>
[-- Attachment #1: Type: text/plain, Size: 1718 bytes --]
On Tue, Aug 06, 2013 at 05:18:44PM -0700, Davidlohr Bueso wrote:
> On Mon, 2013-08-05 at 16:36 +0900, Joonsoo Kim wrote:
> > > Any mapping that doesn't use the reserved pool, not just
> > > MAP_NORESERVE. For example, if a process makes a MAP_PRIVATE mapping,
> > > then fork()s then the mapping is instantiated in the child, that will
> > > not draw from the reserved pool.
> > >
> > > > Should we ensure them to allocate the last hugepage?
> > > > They map a region with MAP_NORESERVE, so don't assume that their requests
> > > > always succeed.
> > >
> > > If the pages are available, people get cranky if it fails for no
> > > apparent reason, MAP_NORESERVE or not. They get especially cranky if
> > > it sometimes fails and sometimes doesn't due to a race condition.
> >
> > Hello,
> >
> > Hmm... Okay. I will try to implement another way to protect race condition.
> > Maybe it is the best to use a table mutex :)
> > Anyway, please give me a time, guys.
>
> So another option is to take the mutex table patchset for now as it
> *does* improve things a great deal, then, when ready, get rid of the
> instantiation lock all together.
We still don't have a solid proposal for doing that. Joonsoo Kim's
patchset misses cases (non reserved mappings). I'm also not certain
there aren't a few edge cases which can lead to even reserved mappings
failing, and if that happens the patches will lead to livelock.
Getting rid of the instantiation mutex is a lot harder than it appears.
--
David Gibson | I'll have my music baroque, and my code
david AT gibson.dropbear.id.au | minimalist, thank you. NOT _the_ _other_
| _way_ _around_!
http://www.ozlabs.org/~dgibson
[-- Attachment #2: Type: application/pgp-signature, Size: 198 bytes --]
next prev parent reply other threads:[~2013-08-07 1:32 UTC|newest]
Thread overview: 67+ messages / expand[flat|nested] mbox.gz Atom feed top
2013-07-29 5:31 [PATCH 00/18] mm, hugetlb: remove a hugetlb_instantiation_mutex Joonsoo Kim
2013-07-29 5:31 ` [PATCH 01/18] mm, hugetlb: protect reserved pages when softofflining requests the pages Joonsoo Kim
2013-07-29 7:24 ` Hillf Danton
2013-07-31 2:27 ` Joonsoo Kim
2013-07-31 2:49 ` Hillf Danton
2013-07-31 4:41 ` Joonsoo Kim
2013-07-31 6:21 ` Hillf Danton
2013-07-31 6:37 ` Joonsoo Kim
2013-07-31 15:25 ` Hillf Danton
2013-08-01 6:07 ` Joonsoo Kim
2013-08-01 16:17 ` Aneesh Kumar K.V
2013-08-04 5:10 ` Hillf Danton
2013-08-05 5:17 ` Aneesh Kumar K.V
2013-07-30 16:49 ` Aneesh Kumar K.V
2013-07-29 5:31 ` [PATCH 02/18] mm, hugetlb: change variable name reservations to resv Joonsoo Kim
2013-07-30 16:50 ` Aneesh Kumar K.V
2013-07-29 5:31 ` [PATCH 03/18] mm, hugetlb: unify region structure handling Joonsoo Kim
2013-07-30 17:27 ` Aneesh Kumar K.V
2013-07-31 2:36 ` Joonsoo Kim
2013-07-29 5:31 ` [PATCH 04/18] mm, hugetlb: region manipulation functions take resv_map rather list_head Joonsoo Kim
2013-07-29 5:31 ` [PATCH 05/18] mm, hugetlb: protect region tracking via newly introduced resv_map lock Joonsoo Kim
2013-07-29 8:58 ` Hillf Danton
2013-07-31 2:41 ` Joonsoo Kim
2013-07-29 18:53 ` Davidlohr Bueso
2013-07-31 2:43 ` Joonsoo Kim
2013-07-29 5:31 ` [PATCH 06/18] mm, hugetlb: remove vma_need_reservation() Joonsoo Kim
2013-07-29 17:52 ` Naoya Horiguchi
2013-07-31 4:53 ` Joonsoo Kim
2013-07-30 17:49 ` Aneesh Kumar K.V
2013-07-31 4:56 ` Joonsoo Kim
2013-07-29 5:31 ` [PATCH 07/18] mm, hugetlb: pass has_reserve to dequeue_huge_page_vma() Joonsoo Kim
2013-07-29 5:31 ` [PATCH 08/18] mm, hugetlb: do hugepage_subpool_get_pages() when avoid_reserve Joonsoo Kim
2013-07-29 18:05 ` Naoya Horiguchi
2013-07-31 5:02 ` Joonsoo Kim
2013-07-31 20:55 ` Naoya Horiguchi
2013-07-29 5:32 ` [PATCH 09/18] mm, hugetlb: unify has_reserve and avoid_reserve to use_reserve Joonsoo Kim
2013-07-29 5:32 ` [PATCH 10/18] mm, hugetlb: call vma_has_reserve() before entering alloc_huge_page() Joonsoo Kim
2013-07-29 18:27 ` Naoya Horiguchi
2013-07-31 5:06 ` Joonsoo Kim
2013-07-29 5:32 ` [PATCH 11/18] mm, hugetlb: move down outside_reserve check Joonsoo Kim
2013-07-29 18:39 ` Naoya Horiguchi
2013-07-31 5:08 ` Joonsoo Kim
2013-07-31 20:46 ` Naoya Horiguchi
2013-07-29 5:32 ` [PATCH 12/18] mm, hugetlb: remove a check for return value of alloc_huge_page() Joonsoo Kim
2013-07-29 5:32 ` [PATCH 13/18] mm, hugetlb: grab a page_table_lock after page_cache_release Joonsoo Kim
2013-07-29 18:50 ` Naoya Horiguchi
2013-07-29 5:32 ` [PATCH 14/18] mm, hugetlb: clean-up error handling in hugetlb_cow() Joonsoo Kim
2013-07-29 5:32 ` [PATCH 15/18] mm, hugetlb: move up anon_vma_prepare() Joonsoo Kim
2013-07-29 19:05 ` Naoya Horiguchi
2013-07-29 19:19 ` Naoya Horiguchi
2013-07-31 5:12 ` Joonsoo Kim
2013-07-31 16:43 ` Naoya Horiguchi
2013-07-29 5:32 ` [PATCH 16/18] mm, hugetlb: return a reserved page to a reserved pool if failed Joonsoo Kim
2013-07-29 20:19 ` Naoya Horiguchi
2013-07-31 5:21 ` Joonsoo Kim
2013-07-29 5:32 ` [PATCH 17/18] mm, hugetlb: retry if we fail to allocate a hugepage with use_reserve Joonsoo Kim
2013-07-29 7:28 ` David Gibson
2013-07-31 5:37 ` Joonsoo Kim
2013-08-03 10:43 ` David Gibson
2013-08-05 7:36 ` Joonsoo Kim
2013-08-07 0:18 ` Davidlohr Bueso
2013-08-07 1:03 ` David Gibson [this message]
2013-08-07 1:38 ` Davidlohr Bueso
2013-08-07 9:18 ` Joonsoo Kim
2013-08-09 0:02 ` David Gibson
2013-08-09 9:37 ` Joonsoo Kim
2013-07-29 5:32 ` [PATCH 18/18] mm, hugetlb: remove a hugetlb_instantiation_mutex Joonsoo Kim
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=20130807010312.GA17110@voom.redhat.com \
--to=david@gibson.dropbear.id.au \
--cc=akpm@linux-foundation.org \
--cc=aneesh.kumar@linux.vnet.ibm.com \
--cc=davidlohr.bueso@hp.com \
--cc=davidlohr@hp.com \
--cc=dhillf@gmail.com \
--cc=hughd@google.com \
--cc=iamjoonsoo.kim@lge.com \
--cc=kamezawa.hiroyu@jp.fujitsu.com \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=liwanp@linux.vnet.ibm.com \
--cc=mgorman@suse.de \
--cc=mhocko@suse.cz \
--cc=n-horiguchi@ah.jp.nec.com \
--cc=riel@redhat.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 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).