Jerin Jacob c3acd92746 ring: fix single consumer dequeue performance ... Use of rte_smb_wmb() instead of rte_smb_rmb() in sc dequeue function creates the additional overhead of waiting for all the STOREs to be completed to local buffer from ring buffer memory. The sc dequeue function demands only LOAD-STORE barrier where LOADs from ring buffer memory needs to be completed before tail pointer update. Changing to rte_smb_rmb() to enable the required LOAD-STORE barrier. Fixes: ecc7d10e44 ("ring: guarantee dequeue ordering before tail update") Signed-off-by: Jerin Jacob <jerin.jacob@caviumnetworks.com> Acked-by: Konstantin Ananyev <konstantin.ananyev@intel.com>		2016-07-25 17:55:12 +02:00
..
Makefile	mem: move librte_malloc to eal/common	2015-07-16 13:44:48 +02:00
rte_ring_version.map	ring: support freeing	2015-11-04 00:49:59 +01:00
rte_ring.c	mem: avoid memzone/mempool/ring name truncation	2016-05-19 14:40:14 +02:00
rte_ring.h	ring: fix single consumer dequeue performance	2016-07-25 17:55:12 +02:00