sched: Avoid creating superfluous NUMA domains on non-NUMA systems
authorPeter Zijlstra <a.p.zijlstra@chello.nl>
Wed, 20 Jul 2011 16:42:57 +0000 (18:42 +0200)
committerIngo Molnar <mingo@elte.hu>
Wed, 20 Jul 2011 16:54:33 +0000 (18:54 +0200)
commitd110235d2c331c4f79e0879f51104be79e17a469
treed56bb05a1274f9e7c6a20467d2d04713c185e7ce
parente3589f6c81e4764d32a25d2a2a0afe54fa344f5c
sched: Avoid creating superfluous NUMA domains on non-NUMA systems

When creating sched_domains, stop when we've covered the entire
target span instead of continuing to create domains, only to
later find they're redundant and throw them away again.

This avoids single node systems from touching funny NUMA
sched_domain creation code and reduces the risks of the new
SD_OVERLAP code.

Requested-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Anton Blanchard <anton@samba.org>
Cc: mahesh@linux.vnet.ibm.com
Cc: benh@kernel.crashing.org
Cc: linuxppc-dev@lists.ozlabs.org
Link: http://lkml.kernel.org/r/1311180177.29152.57.camel@twins
Signed-off-by: Ingo Molnar <mingo@elte.hu>
kernel/sched.c