RE: mm: pages are not freed from lru_add_pvecs after process termination

From	"Odzioba, Lukasz" <lukasz.odzioba@intel.com>
Newsgroups	linux.kernel
Subject	RE: mm: pages are not freed from lru_add_pvecs after process termination
Date	2016-06-08 11:00 +0200
Message-ID	<rHHvc-6mU-9@gated-at.bofh.it> (permalink)
References	(6 earlier) <rvPHQ-7CY-25@gated-at.bofh.it> <rvQuf-8qB-13@gated-at.bofh.it> <rxxdL-3FS-7@gated-at.bofh.it> <rHlbj-Mt-1@gated-at.bofh.it> <rHnd7-1YN-7@gated-at.bofh.it>
Organization	linux.* mail to news gateway

Show all headers | View raw

On Tue 07-06-16 13:20:00, Michal Hocko wrote:
> I guess you want something like posix_memalign or start faulting in from
> an aligned address to guarantee you will fault 2MB pages. 

Good catch.

> Besides that I am really suspicious that this will be measurable at all.
> I would just go and spin a patch assuming you are still able to trigger
> OOM with the vanilla kernel. 

Yes, I am still able to trigger OOM, the tests I did are  more like sanity
checks rather than benchmarks. lru_cache_add takes very little time
so it was rather to look for some unexpected side effects.

Thank,
Lukas

Back to linux.kernel | Previous | Next — Previous in thread | Find similar | Unroll thread

Thread

RE: mm: pages are not freed from lru_add_pvecs after process  termination "Odzioba, Lukasz" <lukasz.odzioba@intel.com> - 2016-06-07 11:10 +0200
  Re: mm: pages are not freed from lru_add_pvecs after process  termination Michal Hocko <mhocko@kernel.org> - 2016-06-07 13:20 +0200
    RE: mm: pages are not freed from lru_add_pvecs after process  termination "Odzioba, Lukasz" <lukasz.odzioba@intel.com> - 2016-06-08 11:00 +0200

csiph-web