+16
−120
+1
−19
Loading
Previously, a pool of events per socket was created for NUMA locality. However, when passing a message between threads, the event would be taken from one pool and put back into another, resulting in imbalances. At this time, I do not see an efficient way to allocate the events so that they remain NUMA local. We'll work on that over time. However, for correctness right now, go down to a single global pool of events. Change-Id: I09f8b0c5c928777e8274c53c6dce21b9c346e2a5 Signed-off-by:Ben Walker <benjamin.walker@intel.com> Reviewed-on: https://review.gerrithub.io/433519 Tested-by:
SPDK CI Jenkins <sys_sgci@intel.com> Reviewed-by:
Jim Harris <james.r.harris@intel.com> Reviewed-by:
Changpeng Liu <changpeng.liu@intel.com> Reviewed-by:
Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com> Reviewed-by:
Darek Stojaczyk <dariusz.stojaczyk@intel.com> Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com>