Whamcloud - gitweb
LU-5415 ldlm: high load because of negative timeout 27/11227/5
authorLi Xi <lixi@ddn.com>
Fri, 25 Jul 2014 03:08:03 +0000 (11:08 +0800)
committerOleg Drokin <oleg.drokin@intel.com>
Tue, 12 Aug 2014 14:13:59 +0000 (14:13 +0000)
commit381492e1fc3f93873bf495885859c4f63ce0f4fe
tree6b037169de459d9c93f630e09bf7911964051bf1
parenta888b6e9dc7e85b750cabb994987379ff2e92a4a
LU-5415 ldlm: high load because of negative timeout

When the time of LRU resizing exceeds waiting period of
recalculation, the ldlm daemon will keep on resizing without any
interval of rest. That will cause high CPU load.

This patch fixes the problem by setting the recalculation timestamp
after LRU resizing finishes rather than before it. What is more, an
interval of one second is enforced between each recalculation.

Signed-off-by: Li Xi <lixi@ddn.com>
Change-Id: I63b1951882754952e8ea8d921f42ddad1e6daa33
Reviewed-on: http://review.whamcloud.com/11227
Tested-by: Jenkins
Tested-by: Maloo <hpdd-maloo@intel.com>
Reviewed-by: Dmitry Eremin <dmitry.eremin@intel.com>
Reviewed-by: Bobi Jam <bobijam@gmail.com>
Reviewed-by: Lai Siyao <lai.siyao@intel.com>
Reviewed-by: Oleg Drokin <oleg.drokin@intel.com>
lustre/ldlm/ldlm_pool.c