Skip to content

sched: Flatten the pick

Peter Zijlstra cgroups lkml regressions 3 versions – Applied

Status from this message: The following commit has been merged into the sched/core branch of tip:

Get series with b4
b4 am [email protected]

Versions newest first · linked by author and title

  1. v3 Latest7 patches 32 repliesApplied
  2. v210 patches 54 repliesApplied
  3. v1 RFC8 patches 3 repliesSuperseded

Changes in v3 from the cover letter

- merged debug and prep patches
- fixed update_entity_lag() on dequeue (Vincent)
- fixed throttle vs tick (Prateek)
- fixed wakeup_preempt_fair()
- rebased on tip/sched/core
- rewritten cgroup_mode changelogs
- reworked cgroup_mode concur
- added cgroup_mode tasks
- changed default cgroup_mode

[1] - https://lore.kernel.org/r/[email protected]

Patches in v3 Applied

#Subject and filesSizeRepliesApplied
1/7sched/fair: Add cgroup_mode switchkernel/sched/debug.c+74 -00 replies
2/7sched/fair: Add cgroup_mode: upkernel/sched/debug.c kernel/sched/fair.c kernel/sched/sched.h+34 -31 reply
3/7sched/fair: Add cgroup_mode: maxinclude/linux/cpuset.h kernel/cgroup/cpuset.c kernel/sched/debug.c +1 more+69 -55 replies
4/7sched/fair: Add cgroup_mode: concurkernel/sched/debug.c kernel/sched/fair.c kernel/sched/sched.h+40 -70 replies
5/7sched/fair: Add cgroup_mode: taskskernel/sched/debug.c kernel/sched/fair.c+17 -00 replies
6/7sched/fair: Change the default cgroup_mode to concurkernel/sched/debug.c kernel/sched/fair.c+3 -30 replies
7/7sched/eevdf: Move to a single runqueueinclude/linux/sched.h kernel/sched/core.c kernel/sched/debug.c +3 more+375 -47525 replies

Reviews trailers per patch

None of the 7 patches has a Reviewed-by or Acked-by.

PatchMihail Mihov
1/7
2/7
3/7
4/7
5/7
6/7
7/7T-b

Trailers given in replies are filled; those carried in the patches from earlier versions are outlined with dashes.

Read the v3 thread

Cover letter [PATCH v3 0/7] sched: Flatten the pick

Hi!

New version, same story [1]. TL;DR:

- Adds new cgroup_mode knob and implements new policies to address the hierarchy level weight mismatch.

- Builds upon that base to create a flat / single runqueue scheduler where the cgroup hierarchy is expressed through dynamic weight management.

I'm hoping to be able to merge these patches early in the next cycle (after 7.2-rc1).

Random benchmark:

Game vs 'for ((i=0; i<8; i++)) do nice ./spin.sh; done':

Lutris / GE-Proton10-34 / Steam Runtime 3 (sniper) Intel Core i7-2600K
AMD Radeon RX 580

Shadows Awakening (GOG)

default slice(*)

  FPS min   4.0   29.0
      avg  47.5   59.2
      max  83.7   83.7
  FT  min   9.3   10.2
      avg  34.0   17.0
      max 121.2   30.0

FPS (Frames Per Second)
FT (FrameTime)

  [*] Command prefix: 'chrt -o --sched-runtime 100000 0'

Changes since v2:

- merged debug and prep patches
- fixed update_entity_lag() on dequeue (Vincent)
- fixed throttle vs tick (Prateek)
- fixed wakeup_preempt_fair()
- rebased on tip/sched/core
- rewritten cgroup_mode changelogs
- reworked cgroup_mode concur
- added cgroup_mode tasks
- changed default cgroup_mode

[1] - https://lore.kernel.org/r/[email protected]

Can also be had:

git://git.kernel.org/pub/scm/linux/kernel/git/peterz/queue.git sched/flat

 include/linux/cpuset.h |    6 
 include/linux/sched.h  |    1 
 kernel/cgroup/cpuset.c |   15 
 kernel/sched/core.c    |    5 
 kernel/sched/debug.c   |   89 ++++
 kernel/sched/fair.c    |  943 ++++++++++++++++++++++++-------------------------
 kernel/sched/pelt.c    |    6 
 kernel/sched/sched.h   |   30 -
 8 files changed, 607 insertions(+), 488 deletions(-)