VM Rx rte_mbuf allocation for single packet isn't efficient

Bug #1748080 reported by Yi-y-yang
6
This bug affects 1 person
Affects Status Importance Assigned to Milestone
Juniper Openstack
Status tracked in Trunk
Trunk
Fix Committed
Undecided
Unassigned
OpenContrail
New
Undecided
Unassigned

Bug Description

It isn't efficient to allocate rte_mbuf by rte_pktmbuf_alloc for every VM Rx, rte_pktmbuf_alloc_bulk will be more efficient, but we must make sure we have some rte_mbufs left for KNI, otherwise VM traffic will make KNI hungry and result in Out of Memory.

Revision history for this message
OpenContrail Admin (ci-admin-f) wrote : [Review update] master

Review in progress for https://review.opencontrail.org/39557
Submitter: Yi Yang (<email address hidden>)

Revision history for this message
OpenContrail Admin (ci-admin-f) wrote : A change has been merged

Reviewed: https://review.opencontrail.org/39557
Committed: http://github.com/Juniper/contrail-vrouter/commit/cd70202a4b596be44f10e2ca775db4658472b83c
Submitter: Zuul v3 CI (<email address hidden>)
Branch: master

commit cd70202a4b596be44f10e2ca775db4658472b83c
Author: Yi Yang <email address hidden>
Date: Fri Feb 23 11:18:32 2018 +0800

Optimize VM Rx by batch processing

Closes-bug: #1748080

Change-Id: Iec7324b69687e93ca03e38aed015f34624049053
Signed-off-by: Yi Yang <email address hidden>

To post a comment you must log in.
This report contains Public information  
Everyone can see this information.

Other bug subscribers

Remote bug watches

Bug watches keep track of this bug in other bug trackers.