Skip to content

Backport "sched cpuset: Bring back cpuset_mutex"

Luiz Capitulino stable 1 version In review

Get series with b4
b4 am [email protected]

Versions newest first

  1. v1 Latest5 patches 3 repliesIn review

Patches in v1 No applies recorded yet

#Subject and filesSizeRepliesApplied
1/5cgroup/cpuset: Optimize cpuset_attach() on v2kernel/cgroup/cpuset.c+23 -10 replies
2/5cgroup/cpuset: no need to explicitly init a global static variablekernel/cgroup/cpuset.c+0 -20 replies
3/5cgroup/cpuset: Skip task update if hotplug doesn't affect current cpusetkernel/cgroup/cpuset.c+3 -00 replies
4/5cgroup/cpuset: Include offline CPUs when tasks' cpumasks in top_cpuset are updatedkernel/cgroup/cpuset.c+14 -90 replies
5/5sched/cpuset: Bring back cpuset_mutexinclude/linux/cpuset.h kernel/cgroup/cpuset.c kernel/sched/core.c+99 -900 replies

Reviews trailers per patch

3 of 5 patches have a Reviewed-by or Acked-by.

PatchAaron TomlinMukesh OjhaMichal KoutnýWaiman Long
1/5
2/5R-b A-b
3/5R-b
4/5
5/5R-b

Trailers given in replies are filled; those carried in the patches from earlier versions are outlined with dashes. Scroll the table sideways for all 4 reviewers.

Read the v1 thread

Cover letter [PATH 6.1.y 0/5] Backport "sched cpuset: Bring back cpuset_mutex"

Hi,

When using KVM on systems that require iTLB multihit mitigation enabled[1], we're observing very high latency (70ms+) in KVM_CREATE_VM ioctl() in 6.1 kernel in comparison to older stable kernels such as 5.10. This is true even when using favordynmods mount option.

We debugged this down to the cpuset controller trying to acquire cpuset_rwsem in cpuset_can_attach(). This happens because KVM creates a worker thread which calls cgroup_attach_task_all() during KVM_CREATE_VM. I don't know if favordynmods is supposed to cover this case or not, but removing cpuset_rwsem certainly solves the issue.

For the backport I tried to pick as many dependent commits as required to avoid conflicts. I would highly appreciate review from cgroup people.

Tests performed:
* Measured latency in KVM_CREATE_VM ioctl(), it goes down to less than 1ms
* Ran the cgroup kselftest tests, got same results with or without this series
* However, some tests such as test_memcontrol and test_kmem are failing in 6.1. This probably needs to be looked at
* To make test_cpuset_prs.sh work, I had to increase the timeout on line 592 to 1 second. With this change, the test runs and passes
* I run our downstream test suite against our downstream 6.1 kernel with this series applied, it passed

[1] For the case where the CPU is not vulnerable to iTLB multihit we can simply disable the iTLB multihit mitigation in KVM which avoids this whole situation. Disabling the mitigation is possible since upstream commit 0b210faf337 which I plan to backport soon

Daniel Vacek (1):
cgroup/cpuset: no need to explicitly init a global static variable

Juri Lelli (1):
sched/cpuset: Bring back cpuset_mutex

Waiman Long (3):
cgroup/cpuset: Optimize cpuset_attach() on v2 cgroup/cpuset: Skip task update if hotplug doesn't affect current
cpuset
cgroup/cpuset: Include offline CPUs when tasks' cpumasks in top_cpuset
are updated

 include/linux/cpuset.h |   8 +-
 kernel/cgroup/cpuset.c | 211 +++++++++++++++++++++++------------------
 kernel/sched/core.c    |  22 +++--
 3 files changed, 139 insertions(+), 102 deletions(-)
Signature · 1 line
2.40.1