2
0
mirror of https://github.com/openvswitch/ovs synced 2025-10-25 15:07:05 +00:00
Files
openvswitch/lib
Ilya Maximets b59cc14e03 netdev-dpdk: Use instant sending instead of queueing of packets.
Current implementarion of TX packet's queueing is broken in several ways:

	* TX queue flushing implemented on receive assumes that all
	  core_id-s are sequential and starts from zero. This may lead
	  to situation when packets will stuck in queue forever and,
	  also, this influences on latency.

	* For a long time flushing logic depends on uninitialized
	  'txq_needs_locking', because it usually calculated after
	  'netdev_dpdk_alloc_txq' but used inside of this function
	  for initialization of 'flush_tx'.

Testing shows no performance difference with and without queueing.
Lets remove queueing at all because it doesn't work properly now and
also does not increase performance.

Signed-off-by: Ilya Maximets <i.maximets@samsung.com>
Acked-by: Daniele Di Proietto <diproiettod@vmware.com>
2016-07-06 15:46:35 -07:00
..
2016-05-16 17:46:55 -07:00
2016-05-16 17:46:55 -07:00
2016-04-26 23:28:59 -07:00
2015-12-21 17:23:43 -08:00
2016-06-29 18:19:40 -07:00
2016-06-24 09:17:00 -07:00
2016-06-24 09:17:00 -07:00
2016-07-02 21:19:22 -07:00
2016-06-24 09:17:00 -07:00
2016-06-24 09:17:00 -07:00
2016-04-27 15:00:37 -07:00
2016-04-27 15:00:37 -07:00
2016-07-02 11:35:55 -07:00
2016-04-26 23:28:59 -07:00
2016-04-26 23:28:59 -07:00
2016-07-02 11:35:55 -07:00
2016-06-24 09:17:00 -07:00
2016-06-24 09:17:00 -07:00
2016-06-03 13:08:31 -07:00
2016-07-02 11:35:55 -07:00
2016-05-18 19:39:18 -07:00
2016-07-02 11:35:55 -07:00
2016-04-18 08:49:59 -07:00
2016-04-18 08:49:59 -07:00
2016-03-30 13:01:21 -07:00
2016-06-09 13:34:59 +09:00
2015-12-14 05:12:34 -08:00
2016-03-30 13:01:21 -07:00
2016-03-28 17:36:40 -04:00
2016-07-02 11:35:55 -07:00
2016-05-18 19:39:18 -07:00
2016-06-26 20:31:28 -07:00
2016-06-26 20:31:28 -07:00
2016-02-10 13:36:04 -08:00
2016-06-26 20:31:28 -07:00