Whamcloud - gitweb
LU-5415 ldlm: high load because of negative timeout 14/11414/2
authorLi Xi <lixi@ddn.com>
Tue, 12 Aug 2014 14:58:53 +0000 (22:58 +0800)
committerOleg Drokin <oleg.drokin@intel.com>
Mon, 18 Aug 2014 03:39:17 +0000 (03:39 +0000)
commit82c866d41013b4e2512951e8cbec67221c01d113
tree72582aec6a9abc96ca45d2a27efd5aa75bcaeccf
parent65cf60544dbd040bf4591d97af1843da1fd0ae09
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.

Lustre-change: http://review.whamcloud.com/11227/
Lustre-commit: 381492e1fc3f93873bf495885859c4f63ce0f4fe

Signed-off-by: Li Xi <lixi@ddn.com>
Change-Id: I989255eca64d74dfcb2087060f627c2e72615248
Reviewed-on: http://review.whamcloud.com/11414
Tested-by: Jenkins
Tested-by: Maloo <hpdd-maloo@intel.com>
Reviewed-by: Oleg Drokin <oleg.drokin@intel.com>
lustre/ldlm/ldlm_pool.c