linux-kernel - [PATCH 0/4] Optimize cgroup context switch

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [thread-next>] [day] [month] [year] [list]

Message-Id: <1556549045-71814-1-git-send-email-kan.liang@linux.intel.com>
Date:   Mon, 29 Apr 2019 07:44:01 -0700
From:   kan.liang@...ux.intel.com
To:     peterz@...radead.org, tglx@...utronix.de, mingo@...hat.com,
        linux-kernel@...r.kernel.org
Cc:     eranian@...gle.com, tj@...nel.org, ak@...ux.intel.com,
        Kan Liang <kan.liang@...ux.intel.com>
Subject: [PATCH 0/4] Optimize cgroup context switch

From: Kan Liang <kan.liang@...ux.intel.com>

On systems with very high context switch rates between cgroups,
there are high overhead using cgroup perf.

Current codes have two issues.
- System-wide events are mistakenly switched in cgroup
  context switch. It causes system-wide events miscounting,
  and brings avoidable overhead.
  Patch 1 fixes the issue.
- The cgroup context switch sched_in is low efficient.
  All cgroup events share the same per-cpu pinned/flexible groups.
  The RB trees for pinned/flexible groups don't understand cgroup.
  Current code has to traverse all events, and use event_filter_match()
  to filter the events for specific cgroup.
  Patch 2-4 adds a fast path for cgroup context switch sched_in by
  training the RB tree to understand cgroup. The extra filtering
  can be avoided.


Here is test with 6 cgroups running.
Each cgroup has a specjbb benchmark running.
The perf command is as below.
   perf stat -e cycles,instructions -e cycles,instructions
   -e cycles,instructions -e cycles,instructions 
   -e cycles,instructions -e cycles,instructions
   -G cgroup1,cgroup1,cgroup2,cgroup2,cgroup3,cgroup3
   -G cgroup4,cgroup4,cgroup5,cgroup5,cgroup6,cgroup6
   -a -e cycles,instructions -I 1000

The average RT (Response Time) reported from specjbb is
used as key performance metrics. (The lower the better)

                                        RT(us)              Overhead
Baseline (no perf stat):                4286.9
Use cgroup perf, no patches:            4483.6                4.6%
Use cgroup perf, apply patch 1:         4369.2                1.9%
Use cgroup perf, apple all patches:     4335.3                1.1%

Kan Liang (4):
  perf: Fix system-wide events miscounting during cgroup monitoring
  perf: Add filter_match() as a parameter for pinned/flexible_sched_in()
  perf cgroup: Add cgroup ID as a key of RB tree
  perf cgroup: Add fast path for cgroup switch

 include/linux/perf_event.h |   7 ++
 kernel/events/core.c       | 171 +++++++++++++++++++++++++++++++++++++++------
 2 files changed, 157 insertions(+), 21 deletions(-)

-- 
2.7.4