Message-ID: <CALCETrVCvpi3KmS9n=WV8Gy3D89aqC-NKfH02LwgmGNGZtFWLw@mail.gmail.com>
Date:	Tue, 17 Dec 2013 14:25:11 -0800
From:	Andy Lutomirski <luto@...capital.net>
To:	Alex Thorlton <athorlton@....com>
Cc:	Andrew Morton <akpm@...ux-foundation.org>,
	"linux-mm@...ck.org" <linux-mm@...ck.org>,
	"Kirill A. Shutemov" <kirill.shutemov@...ux.intel.com>,
	Benjamin Herrenschmidt <benh@...nel.crashing.org>,
	Rik van Riel <riel@...hat.com>,
	Wanpeng Li <liwanp@...ux.vnet.ibm.com>,
	Mel Gorman <mgorman@...e.de>,
	Michel Lespinasse <walken@...gle.com>,
	Benjamin LaHaise <bcrl@...ck.org>,
	Oleg Nesterov <oleg@...hat.com>,
	"Eric W. Biederman" <ebiederm@...ssion.com>,
	Al Viro <viro@...iv.linux.org.uk>,
	David Rientjes <rientjes@...gle.com>,
	Zhang Yanfei <zhangyanfei@...fujitsu.com>,
	Peter Zijlstra <peterz@...radead.org>,
	Johannes Weiner <hannes@...xchg.org>,
	Michal Hocko <mhocko@...e.cz>,
	Jiang Liu <jiang.liu@...wei.com>,
	Cody P Schafer <cody@...ux.vnet.ibm.com>,
	Glauber Costa <glommer@...allels.com>,
	Kamezawa Hiroyuki <kamezawa.hiroyu@...fujitsu.com>,
	Naoya Horiguchi <n-horiguchi@...jp.nec.com>,
	"linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>,
	Andrea Arcangeli <aarcange@...hat.com>
Subject: Re: [RFC PATCH 0/3] Change how we determine when to hand out THPs

On Tue, Dec 17, 2013 at 9:47 AM, Alex Thorlton <athorlton@....com> wrote:
> On Tue, Dec 17, 2013 at 08:54:10AM -0800, Andy Lutomirski wrote:
>> On Tue, Dec 17, 2013 at 8:04 AM, Alex Thorlton <athorlton@....com> wrote:
>> > On Mon, Dec 16, 2013 at 05:43:40PM -0800, Andy Lutomirski wrote:
>> >> On Mon, Dec 16, 2013 at 9:12 AM, Alex Thorlton <athorlton@....com> wrote:
>> >> >> Please cc Andrea on this.
>> >> >
>> >> > I'm going to clean up a few small things for a v2 pretty soon, I'll be
>> >> > sure to cc Andrea there.
>> >> >
>> >> >> > My proposed solution to the problem is to allow users to set a
>> >> >> > threshold at which THPs will be handed out.  The idea here is that, when
>> >> >> > a user faults in a page in an area where they would usually be handed a
>> >> >> > THP, we pull 512 pages off the free list, as we would with a regular
>> >> >> > THP, but we only fault in single pages from that chunk, until the user
>> >> >> > has faulted in enough pages to pass the threshold we've set.  Once they
>> >> >> > pass the threshold, we do the necessary work to turn our 512 page chunk
>> >> >> > into a proper THP.  As it stands now, if the user tries to fault in
>> >> >> > pages from different nodes, we completely give up on ever turning a
>> >> >> > particular chunk into a THP, and just fault in the 4K pages as they're
>> >> >> > requested.  We may want to make this tunable in the future (i.e. allow
>> >> >> > them to fault in from only 2 different nodes).
>> >> >>
>> >> >> OK.  But all 512 pages reside on the same node, yes?  Whereas with thp
>> >> >> disabled those 512 pages would have resided closer to the CPUs which
>> >> >> instantiated them.
>> >> >
>> >> > As it stands right now, yes, since we're pulling a 512 page contiguous
>> >> > chunk off the free list, everything from that chunk will reside on the
>> >> > same node, but as I (stupidly) forgot to mention in my original e-mail,
>> >> > one piece I have yet to add is the functionality to put the remaining
>> >> > unfaulted pages from our chunk *back* on the free list after we give up
>> >> > on handing out a THP.  Once this is in there, things will behave more
>> >> > like they do when THP is turned completely off, i.e. pages will get
>> >> > faulted in closer to the CPU that first referenced them once we give up
>> >> > on handing out the THP.
>> >>
>> >> This sounds like it's almost the worst possible behavior wrt avoiding
>> >> memory fragmentation.  If userspace mmaps a very large region and then
>> >> starts accessing it randomly, it will allocate a bunch of contiguous
>> >> 512-page regions, claim one page from each, and return the other 511
>> >> pages to the free list.  Memory is now maximally fragmented from the
>> >> point of view of future THP allocations.
>> >
>> > Maybe I'm missing the point here to some degree, but the way I think
>> > about this is that if we trigger the behavior to return the pages to the
>> > free list, we don't *want* future THP allocations in that range of
>> > memory for the current process anyways.  So, having the memory be
>> > fragmented from the point of view of future THP allocations isn't an
>> > issue.
>> >
>>
>> Except that you're causing a problem for the whole system because one
>> process is triggering the "hugepages aren't helpful" heuristic.
>
> I do see where you're coming from here.  Do you have any good tests
> that can cause this type of memory fragmentation that I might be able to
> take a look at, to see how we might combat that issue in this case?
> It seems like something that could occur anyways, but my patch would
> create a situation where it could become a problem much more quickly.

mmap lots of space (comparable to total system memory).  Touch every
512th page.  (This will consume ~0.2% of memory with your patches.)

Now run any workload that benefits from THP (without unmapping the
first thing).  Make sure it still works well.

--Andy
--
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