On Thu, Sep 12, 2019 at 8:54 PM Shahed Shaikh <shsha...@marvell.com> wrote: > > First four patches are part of a fix for the ovs-dpdk issue > with 100Gb NIC [1]. > Fifth patch adds support for drop action in rte_flow. > > [1] > As per HW design of 100Gb mode, device internally uses 2 engines > (eng0 and eng1), and both engines need to be configured symmetrically. > Based on this requirement, driver design chose an approach > to allow user to allocate only even number of queues and split > those queues on both engines equally. > > This approach puts a limitation on number of queues to be allocated - > i.e. user can't configure odd number of queues on 100Gb mode. > OVS configures DPDK port with 1 rxq and 1 txq, which causes initialization > of qede port to fail. > > This patch series changes the implementation of queue allocation method > for 100Gb devices by removing above mentioned limitation and allowing > user to configure odd number of queues. > > Fix is split into 4 logical patches - > - First patch refactors Rx and Tx queue setup code to lay a foundation > for actual fix. > - Second patch actually implements a new approach to fix the issue. > - Third patch fixes RSS configuration w.r.t. new approach. > - Fourth patch fixes statistics code impacted due to new approach. > > change in v2: > Fixed compilation failure from patch 2 on clang compiler. > > Shahed Shaikh (5): > net/qede: refactor Rx and Tx queue setup > net/qede: fix ovs-dpdk failure when using odd number of queues on > 100Gb mode > net/qede: fix RSS configuration as per new 100Gb queue allocation > method > net/qede: fix stats flow as per new 100Gb queue allocation method > net/qede: implement rte_flow drop action
Fixed ./devtools/check-git-log.sh issues and Series applied to dpdk-next-net-mrvl/master. Thanks.
