From mboxrd@z Thu Jan 1 00:00:00 1970 From: Kirill Tkhai Subject: Re: [PATCH v2 4/5] mm/madvise: allow KSM hints for remote API Date: Fri, 17 Jan 2020 13:13:14 +0300 Message-ID: <37338e14-5a55-1926-b6c1-5f98b6a6fdb5@virtuozzo.com> References: <20200116235953.163318-1-minchan@kernel.org> <20200116235953.163318-5-minchan@kernel.org> Mime-Version: 1.0 Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: 7bit Return-path: In-Reply-To: <20200116235953.163318-5-minchan-DgEjT+Ai2ygdnm+yROfE0A@public.gmane.org> Content-Language: en-US Sender: linux-api-owner-u79uwXL29TY76Z2rM5mHXA@public.gmane.org To: Minchan Kim , Andrew Morton Cc: LKML , linux-mm , linux-api-u79uwXL29TY76Z2rM5mHXA@public.gmane.org, oleksandr-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org, Suren Baghdasaryan , Tim Murray , Daniel Colascione , Sandeep Patil , Sonny Rao , Brian Geffon , Michal Hocko , Johannes Weiner , Shakeel Butt , John Dias , christian.brauner-GeWIH/nMZzLQT0dZR+AlfA@public.gmane.org, sjpark-ebkRAfMGSJGzQB+pC5nmwQ@public.gmane.org, Minchan Kim List-Id: linux-api@vger.kernel.org On 17.01.2020 02:59, Minchan Kim wrote: > From: Oleksandr Natalenko > > It all began with the fact that KSM works only on memory that is marked > by madvise(). And the only way to get around that is to either: > > * use LD_PRELOAD; or > * patch the kernel with something like UKSM or PKSM. > > (i skip ptrace can of worms here intentionally) > > To overcome this restriction, lets employ a new remote madvise API. This > can be used by some small userspace helper daemon that will do auto-KSM > job for us. > > I think of two major consumers of remote KSM hints: > > * hosts, that run containers, especially similar ones and especially in > a trusted environment, sharing the same runtime like Node.js; > > * heavy applications, that can be run in multiple instances, not > limited to opensource ones like Firefox, but also those that cannot be > modified since they are binary-only and, maybe, statically linked. > > Speaking of statistics, more numbers can be found in the very first > submission, that is related to this one [1]. For my current setup with > two Firefox instances I get 100 to 200 MiB saved for the second instance > depending on the amount of tabs. > > 1 FF instance with 15 tabs: > > $ echo "$(cat /sys/kernel/mm/ksm/pages_sharing) * 4 / 1024" | bc > 410 > > 2 FF instances, second one has 12 tabs (all the tabs are different): > > $ echo "$(cat /sys/kernel/mm/ksm/pages_sharing) * 4 / 1024" | bc > 592 > > At the very moment I do not have specific numbers for containerised > workload, but those should be comparable in case the containers share > similar/same runtime. > > [1] https://lore.kernel.org/patchwork/patch/1012142/ > > Signed-off-by: Oleksandr Natalenko > Signed-off-by: Minchan Kim > --- > mm/madvise.c | 2 ++ > 1 file changed, 2 insertions(+) > > diff --git a/mm/madvise.c b/mm/madvise.c > index 84cffd0900f1..89557998d287 100644 > --- a/mm/madvise.c > +++ b/mm/madvise.c > @@ -1000,6 +1000,8 @@ process_madvise_behavior_valid(int behavior) > switch (behavior) { > case MADV_COLD: > case MADV_PAGEOUT: > + case MADV_MERGEABLE: > + case MADV_UNMERGEABLE: > return true; > default: > return false; Remote madvise on KSM parameters should be OK. One thing is madvise_behavior_valid() places MADV_MERGEABLE/UNMERGEABLE in #ifdef brackes, so -EINVAL is returned by madvise() syscall if KSM is not enabled. Here we should follow the same way for symmetry.