Skip to content

sched: Migrate disable support for RT

Thomas Gleixner lkml oe-kbuild-all oe-lkp 1 version In review

Get series with b4
b4 am [email protected]

Versions newest first

  1. v1 Latest10 patches 10 repliesIn review

Patches in v1 No applies recorded yet

#Subject and filesSizeRepliesApplied
1/10sched: Fix balance_callback()kernel/sched/core.c kernel/sched/sched.h+77 -440 replies
2/10sched/hotplug: Ensure only per-cpu kthreads run during hotplugkernel/sched/core.c kernel/sched/sched.h+115 -20 replies
3/10sched/core: Wait for tasks being pushed away on hotplugkernel/sched/core.c kernel/sched/sched.h+45 -30 replies
4/10sched/hotplug: Consolidate task migration on CPU unpluginclude/linux/cpuhotplug.h include/linux/sched/hotplug.h kernel/cpu.c +1 more+46 -1160 replies
5/10sched/core: Split __set_cpus_allowed_ptr()kernel/sched/core.c+38 -260 replies
6/10sched: Add task components for migration controlinclude/linux/sched.h init/init_task.c kernel/fork.c +1 more+35 -30 replies
7/10sched/core: Add mechanism to wait for affinity setting to completeinclude/linux/sched.h kernel/sched/core.c+95 -00 replies
8/10sched: Add update_migratory() callback to scheduler classeskernel/sched/deadline.c kernel/sched/rt.c kernel/sched/sched.h+24 -00 replies
9/10sched/core: Add migrate_disable/enable()include/linux/preempt.h include/linux/sched.h kernel/sched/core.c +1 more+169 -210 replies
10/10sched/core: Make migrate disable and CPU hotplug cooperativekernel/sched/core.c kernel/sched/sched.h+38 -40 replies

Reviews trailers per patch

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

PatchValentin SchneiderScott WoodDaniel Bristot de Oliveira
1/10R-b
2/10
3/10
4/10
5/10
6/10
7/10
8/10Originally-by
9/10
10/10Originally-by

Trailers given in replies are filled; those carried in the patches from earlier versions are outlined with dashes. Scroll the table sideways for all 3 reviewers.

Read the v1 thread

Cover letter [patch 00/10] sched: Migrate disable support for RT

On RT enabled kernels most of the code including spin/rw lock held sections are preemptible, which also makes the tasks migrateable. That violates the per CPU constraints. RT needs therefore a mechanism to control migration independent of preemption.

This series is providing a mostly rewritten version of the RT code based on the newest scheduler bits and includes and depends on Peter's 'let tasks kick themself off the outgoing CPU' feature.

While migrate disable/enable are trivial in principle the tricky part is the coordination and serialization against concurrent affinity changes, e.g. sched_setaffinity(2), and CPU hotplug.

Thanks,

tglx