From: Minchan Kim <minchan@kernel.org> To: Andrew Morton <akpm@linux-foundation.org> Cc: LKML <linux-kernel@vger.kernel.org>, linux-mm <linux-mm@kvack.org>, linux-api@vger.kernel.org, oleksandr@redhat.com, Suren Baghdasaryan <surenb@google.com>, Tim Murray <timmurray@google.com>, Daniel Colascione <dancol@google.com>, Sandeep Patil <sspatil@google.com>, Sonny Rao <sonnyrao@google.com>, Brian Geffon <bgeffon@google.com>, Michal Hocko <mhocko@suse.com>, Johannes Weiner <hannes@cmpxchg.org>, Shakeel Butt <shakeelb@google.com>, John Dias <joaodias@google.com>, Minchan Kim <minchan@kernel.org> Subject: [PATCH 4/4] mm/madvise: allow KSM hints for remote API Date: Fri, 10 Jan 2020 13:34:33 -0800 [thread overview] Message-ID: <20200110213433.94739-5-minchan@kernel.org> (raw) In-Reply-To: <20200110213433.94739-1-minchan@kernel.org> From: Oleksandr Natalenko <oleksandr@redhat.com> 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 <oleksandr@redhat.com> Signed-off-by: Minchan Kim <minchan@kernel.org> --- mm/madvise.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/mm/madvise.c b/mm/madvise.c index eb42b2b7f49b..3aa9aec6bfd9 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; -- 2.25.0.rc1.283.g88dfdc4193-goog
WARNING: multiple messages have this Message-ID (diff)
From: Minchan Kim <minchan-DgEjT+Ai2ygdnm+yROfE0A@public.gmane.org> To: Andrew Morton <akpm-de/tnXTf+JLsfHDXvbKv3WD2FQJk+8+b@public.gmane.org> Cc: LKML <linux-kernel-u79uwXL29TY76Z2rM5mHXA@public.gmane.org>, linux-mm <linux-mm-Bw31MaZKKs3YtjvyW6yDsg@public.gmane.org>, linux-api-u79uwXL29TY76Z2rM5mHXA@public.gmane.org, oleksandr-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org, Suren Baghdasaryan <surenb-hpIqsD4AKlfQT0dZR+AlfA@public.gmane.org>, Tim Murray <timmurray-hpIqsD4AKlfQT0dZR+AlfA@public.gmane.org>, Daniel Colascione <dancol-hpIqsD4AKlfQT0dZR+AlfA@public.gmane.org>, Sandeep Patil <sspatil-hpIqsD4AKlfQT0dZR+AlfA@public.gmane.org>, Sonny Rao <sonnyrao-hpIqsD4AKlfQT0dZR+AlfA@public.gmane.org>, Brian Geffon <bgeffon-hpIqsD4AKlfQT0dZR+AlfA@public.gmane.org>, Michal Hocko <mhocko-IBi9RG/b67k@public.gmane.org>, Johannes Weiner <hannes-druUgvl0LCNAfugRpC6u6w@public.gmane.org>, Shakeel Butt <shakeelb-hpIqsD4AKlfQT0dZR+AlfA@public.gmane.org>, John Dias <joaodias-hpIqsD4AKlfQT0dZR+AlfA@public.gmane.org>, Minchan Kim <minchan-DgEjT+Ai2ygdnm+yROfE0A@public.gmane.org> Subject: [PATCH 4/4] mm/madvise: allow KSM hints for remote API Date: Fri, 10 Jan 2020 13:34:33 -0800 [thread overview] Message-ID: <20200110213433.94739-5-minchan@kernel.org> (raw) In-Reply-To: <20200110213433.94739-1-minchan-DgEjT+Ai2ygdnm+yROfE0A@public.gmane.org> From: Oleksandr Natalenko <oleksandr-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org> 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 <oleksandr-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org> Signed-off-by: Minchan Kim <minchan-DgEjT+Ai2ygdnm+yROfE0A@public.gmane.org> --- mm/madvise.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/mm/madvise.c b/mm/madvise.c index eb42b2b7f49b..3aa9aec6bfd9 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; -- 2.25.0.rc1.283.g88dfdc4193-goog
next prev parent reply other threads:[~2020-01-10 21:34 UTC|newest] Thread overview: 41+ messages / expand[flat|nested] mbox.gz Atom feed top 2020-01-10 21:34 [PATCH 0/4] introduce memory hinting API for external process Minchan Kim 2020-01-10 21:34 ` [PATCH 1/4] mm: factor out madvise's core functionality Minchan Kim 2020-01-11 7:37 ` SeongJae Park 2020-01-11 7:37 ` SeongJae Park 2020-01-13 18:11 ` Minchan Kim 2020-01-13 18:22 ` SeongJae Park 2020-01-10 21:34 ` [PATCH 2/4] mm: introduce external memory hinting API Minchan Kim 2020-01-11 7:34 ` SeongJae Park 2020-01-11 7:34 ` SeongJae Park 2020-01-13 18:02 ` Minchan Kim 2020-01-13 8:47 ` Kirill Tkhai 2020-01-13 8:47 ` Kirill Tkhai 2020-01-13 10:42 ` Christian Brauner 2020-01-13 10:42 ` Christian Brauner 2020-01-13 18:44 ` Minchan Kim 2020-01-13 18:44 ` Minchan Kim 2020-01-13 19:10 ` Christian Brauner 2020-01-13 19:27 ` Daniel Colascione 2020-01-13 19:27 ` Daniel Colascione 2020-01-13 19:27 ` Daniel Colascione 2020-01-13 20:42 ` Christian Brauner 2020-01-13 21:04 ` Daniel Colascione 2020-01-13 21:04 ` Daniel Colascione 2020-01-13 21:04 ` Daniel Colascione 2020-01-14 19:20 ` Christian Brauner 2020-01-14 18:59 ` Minchan Kim 2020-01-14 18:59 ` Minchan Kim 2020-01-14 19:22 ` Christian Brauner 2020-01-14 19:22 ` Christian Brauner 2020-01-13 18:39 ` Minchan Kim 2020-01-13 19:18 ` Daniel Colascione 2020-01-13 19:18 ` Daniel Colascione 2020-01-13 19:18 ` Daniel Colascione 2020-01-14 8:39 ` Kirill Tkhai 2020-01-14 8:39 ` Kirill Tkhai 2020-01-14 19:12 ` Minchan Kim 2020-01-15 9:38 ` Kirill Tkhai 2020-01-10 21:34 ` [PATCH 3/4] mm/madvise: employ mmget_still_valid for write lock Minchan Kim 2020-01-10 21:34 ` Minchan Kim [this message] 2020-01-10 21:34 ` [PATCH 4/4] mm/madvise: allow KSM hints for remote API Minchan Kim 2020-01-11 7:42 ` SeongJae Park
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=20200110213433.94739-5-minchan@kernel.org \ --to=minchan@kernel.org \ --cc=akpm@linux-foundation.org \ --cc=bgeffon@google.com \ --cc=dancol@google.com \ --cc=hannes@cmpxchg.org \ --cc=joaodias@google.com \ --cc=linux-api@vger.kernel.org \ --cc=linux-kernel@vger.kernel.org \ --cc=linux-mm@kvack.org \ --cc=mhocko@suse.com \ --cc=oleksandr@redhat.com \ --cc=shakeelb@google.com \ --cc=sonnyrao@google.com \ --cc=sspatil@google.com \ --cc=surenb@google.com \ --cc=timmurray@google.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: linkBe sure your reply has a Subject: header at the top and a blank line before the message body.
This is an external index of several public inboxes, see mirroring instructions on how to clone and mirror all data and code used by this external index.