workqueue: Share the same PWQ for the CPUs of a pod and distribute max_active across pods

[PATCH 0/7] workqueue: Share the same PWQ for the CPUs of a pod and distribute max_active across pods

Lai Jiangshan posted 7 patches 1 year, 12 months ago

Download series mbox

include/linux/workqueue.h |  34 +++++-
kernel/workqueue.c        | 217 ++++++++++++++++++++++----------------
2 files changed, 157 insertions(+), 94 deletions(-)

Expand all Fold all

[PATCH 0/7] workqueue: Share the same PWQ for the CPUs of a pod and distribute max_active across pods

Posted by Lai Jiangshan 1 year, 12 months ago

From: Lai Jiangshan <jiangshan.ljs@antgroup.com>

A different approach to fix the misbehavior can easily be exposed as reported in
http://lkml.kernel.org/r/dbu6wiwu3sdhmhikb2w6lns7b27gbobfavhjj57kwi2quafgwl@htjcc5oikcr3.

Lai Jiangshan (6):
  workqueue: Reuse the default PWQ as much as possible
  workqueue: Share the same PWQ for the CPUs of a pod
  workqueue: Add pwq_calculate_max_active()
  workqueue: Wrap common code into wq_adjust_pwqs_max_active()
  workqueue: Addjust pwq's max_active when CPU online/offine
  workqueue: Rename wq->saved_max_active to wq->max_active

Tejun Heo (1):
  workqueue: Implement system-wide max_active enforcement for unbound
    workqueues

 include/linux/workqueue.h |  34 +++++-
 kernel/workqueue.c        | 217 ++++++++++++++++++++++----------------
 2 files changed, 157 insertions(+), 94 deletions(-)

-- 
2.19.1.6.gb485710b