timers/migration: Fix NUMA trees + cleanups

[PATCH 0/6] timers/migration: Fix NUMA trees + cleanups

Frederic Weisbecker posted 6 patches 3 months, 2 weeks ago

Download series mbox

kernel/time/timer_migration.c | 279 ++++++++++++++++++++++--------------------
1 file changed, 143 insertions(+), 136 deletions(-)

Expand all Fold all

[PATCH 0/6] timers/migration: Fix NUMA trees + cleanups

Posted by Frederic Weisbecker 3 months, 2 weeks ago

Hi,

A while ago I found out that NUMA tree layout had flaws but I thought it
was only about rare setups. Having had a closer look while thinking
about the big.LITTLE handling, I realized the NUMA timer tree are
actually imbalanced in many cases, except perhaps when 0-7 CPUs are lucky
enough to belong to the same node, and yet up to a certain amount of
CPUs.

Here is a fix proposal, plus a bunch of cleanups.

git://git.kernel.org/pub/scm/linux/kernel/git/frederic/linux-dynticks.git
	timers/core

HEAD: 4adecc004d35dbe8a48df3d23ed2b8cf42772716
Thanks,
	Frederic
---

Frederic Weisbecker (6):
      timers/migration: Convert "while" loops to use "for"
      timers/migration: Remove locking on group connection
      timers/migration: Fix imbalanced NUMA trees
      timers/migration: Assert that hotplug preparing CPU is part of stable active hierarchy
      timers/migration: Remove unused "cpu" parameter from tmigr_get_group()
      timers/migration: Remove dead code handling idle CPU checking for remote timers

 kernel/time/timer_migration.c | 279 ++++++++++++++++++++++--------------------
 1 file changed, 143 insertions(+), 136 deletions(-)