[dpdk-dev] bnx2x driver and 57800 versus 57810

2016-06-13 Thread Chas Williams
I finally tracked this down. On guests with MSI-X enabled on the VF devices, the bnx2x was determining the vector count incorrectly. Apparently, the PF needs this count set correctly during ACQUIRE to initialize the Interrupt Generation Units even though the DPDK driver does polling. http://www.

[dpdk-dev] bnx2x driver and 57800 versus 57810

2016-01-27 Thread Chas Williams
On Wed, 2016-01-27 at 07:32 +, Harish Patil wrote: > > > >I have to practically identical systems, same hypervisor on each > (Centos > >7.x).??In one, I have a 57800 card which works fine with DPDK with > >SRIOV.??In the other, I have a 57810 card which doesn't work with > SRIOV. > > > >For the

[dpdk-dev] bnx2x driver and 57800 versus 57810

2016-01-27 Thread Harish Patil
ched but it did not go thru? yet due to size restrictions. Your mail to 'dev' with the subject Re: [dpdk-dev] bnx2x driver and 57800 versus 57810 Is being held until the list moderator can review it for approval. The reason it is being held: Message

[dpdk-dev] bnx2x driver and 57800 versus 57810

2016-01-26 Thread Chas Williams
I have to practically identical systems, same hypervisor on each (Centos 7.x). ?In one, I have a 57800 card which works fine with DPDK with SRIOV. ?In the other, I have a 57810 card which doesn't work with SRIOV. For the 57810 I have tracked this down to the status block in the VF failing to be up