Skip to content

RFC sched/fair: Complete EEVDF

Peter Zijlstra lkml 1 version In review

Get series with b4
b4 am [email protected]

Versions newest first

  1. v1 RFC Latest10 patches 66 repliesIn review

Patches in v1 No applies recorded yet

#Subject and filesSizeRepliesApplied
1/10sched/eevdf: Add feature commentskernel/sched/features.h+7 -00 replies
2/10sched/eevdf: Remove min_vruntime_copykernel/sched/fair.c kernel/sched/sched.h+2 -70 replies
3/10sched/fair: Cleanup pick_task_fair() vs throttlekernel/sched/fair.c+3 -31 reply
4/10sched/fair: Cleanup pick_task_fair()s currkernel/sched/fair.c+2 -80 replies
5/10sched/fair: Unify pick_{,next_}_task_fair()kernel/sched/fair.c+8 -521 reply
6/10sched: Allow sched_class::dequeue_task() to failkernel/sched/core.c kernel/sched/deadline.c kernel/sched/fair.c +4 more+19 -80 replies
7/10sched/fair: Re-organize dequeue_task_fair()kernel/sched/fair.c+55 -270 replies
8/10sched/fair: Implement delayed dequeueinclude/linux/sched.h kernel/sched/core.c kernel/sched/fair.c +2 more+167 -1851 replies
9/10sched/eevdf: Allow shorter slices to wakeup-preemptkernel/sched/fair.c kernel/sched/features.h+12 -30 replies
10/10sched/eevdf: Use sched_attr::sched_runtime to set request/slice suggestioninclude/linux/sched.h kernel/sched/core.c kernel/sched/debug.c +1 more+36 -1112 replies

Reviews trailers per patch

1 of 10 patches has a Reviewed-by or Acked-by.

PatchMike GalbraithBen Segall
1/10
2/10
3/10R-b
4/10
5/10
6/10
7/10
8/10
9/10T-b
10/10

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

Read the v1 thread

Cover letter [RFC][PATCH 00/10] sched/fair: Complete EEVDF

Hi all,

I'm slowly crawling back out of the hole and trying to get back to work. Availability is still low on my end, but I'll try and respond to some email.

Anyway, in order to get my feet wet again with sitting behind a computer, find here a few patches that should functionally complete the EEVDF journey.

This very much includes the new interface that exposes the extra parameter that EEVDF has. I've chosen to use sched_attr::sched_runtime for this over a nice-like value because some workloads actually know their slice length (can be dynamically measured in the same way as for deadline using CLOCK_THREAD_CPUTIME_ID) and using the real request size is much more effective than some relative measure.

[[ using too short a request size will increase job preemption overhead,
using too long a request size will decrease timeliness ]]

The whole delayed-dequeue thing is I think a fundamental thing that was missing from the EEVDF paper. Without something like this EEVDF will simply not work right. IIRC this was mentioned to me many years ago when people worked on BFQ iosched and ran into this same issue. Time had erased the critical aspect of this note and I had to re-discover it again.

Also, I think Ben expressed concern that preserving lag over long periods doesn't make sense a while back.

The implementation presented here is one that should work with our cgroup mess and keeps most of the ugly inside fair.c unlike previous versions which puked all over the core scheduler code.

Critically cfs-cgroup throttling is not tested, and cgroups are only tested in so far that a systemd infected machine now boots (took a bit).

Other than that, it works well enough to build the next kernel and it passes the few trivial latency-slice tests I ran.

Anyway, please have a poke and let me know...