sched: Wholesale removal of sd_idle logic
authorVenkatesh Pallipadi <venki@google.com>
Mon, 14 Feb 2011 22:38:50 +0000 (14:38 -0800)
committerIngo Molnar <mingo@elte.hu>
Wed, 16 Feb 2011 12:33:20 +0000 (13:33 +0100)
commit46e49b3836c7cd2ae5b5fe76fa981d0d292a52fe
tree25c9a3fa6ad6f45fb8553e0ebe52b973a02f4ef6
parent48fa4b8ecf683f5e411303553da9e186e8b8406e
sched: Wholesale removal of sd_idle logic

sd_idle logic was introduced way back in 2005 (commit 5969fe06),
as an HT optimization.

As per the discussion in the thread here:

  lkml - sched: Resolve sd_idle and first_idle_cpu Catch-22 - v1
  https://patchwork.kernel.org/patch/532501/

The capacity based logic in the load balancer right now handles this
in a much cleaner way, handling more than 2 SMT siblings etc, and sd_idle
does not seem to bring any additional benefits. sd_idle logic also has
some bugs that has performance impact. Here is the patch that removes
the sd_idle logic altogether.

Also, there was a dependency of sched_mc_power_savings == 2, with sd_idle
logic.

Signed-off-by: Venkatesh Pallipadi <venki@google.com>
Acked-by: Vaidyanathan Srinivasan <svaidy@linux.vnet.ibm.com>
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
LKML-Reference: <1297723130-693-1-git-send-email-venki@google.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
kernel/sched_fair.c