test suite reviews and discussions
 help / color / mirror / Atom feed
* [dts] The semantics of rte_eth_rx_burst and rte_eth_rx_queue_count
@ 2015-07-27 18:04 Masoud Moshref Javadi
  2015-07-27 22:47 ` Masoud Moshref Javadi
  0 siblings, 1 reply; 2+ messages in thread
From: Masoud Moshref Javadi @ 2015-07-27 18:04 UTC (permalink / raw)
  To: dts

Hi,

I don't understand the semantics of these two methods (rte_eth_rx_burst 
and rte_eth_rx_queue_count).

For rte_eth_rx_burst with burst of B.

  * Does that mean it will not return until it receive at least B
    packets? or it just returns with whatever packets that are in the
    NIC's queue?
  * Does that mean it will wait at least x clocks before returning 0 to
    B packets?

For rte_eth_rx_queue_count:

  * Does it go to the NIC and poll the # filled entries at its queue? if
    yes, how much is its overhead?
  * Do we need to call rte_eth_rx_burst (or similar methods) to make it
    updated?


Regards

^ permalink raw reply	[flat|nested] 2+ messages in thread

end of thread, other threads:[~2015-07-27 22:45 UTC | newest]

Thread overview: 2+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2015-07-27 18:04 [dts] The semantics of rte_eth_rx_burst and rte_eth_rx_queue_count Masoud Moshref Javadi
2015-07-27 22:47 ` Masoud Moshref Javadi

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).