Thanks David and Flavio. Patch worked like a charm.
On Tue, Feb 4, 2020 at 3:05 PM Ravi Kerur <rke...@gmail.com> wrote:
> David/Flavio, I will try the patch and let you know. Thanks for the quick
> help.
>
> Thanks.
>
> On Tue, Feb 4, 2020 at 12:35 PM Flavio Leitner <f...@sysclose.org> wrote:
>
>> On Tue, Feb 04, 2020 at 08:41:37PM +0100, David Marchand wrote:
>> > On Tue, Feb 4, 2020 at 7:07 PM Ravi Kerur <rke...@gmail.com> wrote:
>> > > *2020-02-04T17:42:52.976Z|00084|dpdk|ERR|Ethdev port_id=0 requested Tx
>> > > offloads 0x2a doesn't match Tx offloads capabilities 0xe in
>> > > rte_eth_dev_configure()*
>> >
>> > I'd say we have a little issue with the introduction of TSO.
>> > How about:
>>
>> Yeah, looks like the card support IPv4 and TCP cksum but not TSO yet.
>> The patch below looks good to me.
>> Thanks
>> fbl
>>
>> >
>> > diff --git a/lib/netdev-dpdk.c b/lib/netdev-dpdk.c
>> > index d6c87f8d4..4888d1cae 100644
>> > --- a/lib/netdev-dpdk.c
>> > +++ b/lib/netdev-dpdk.c
>> > @@ -1132,7 +1132,7 @@ dpdk_eth_dev_init(struct netdev_dpdk *dev)
>> > dev->hw_ol_features &= ~NETDEV_RX_HW_SCATTER;
>> > }
>> >
>> > - if (info.tx_offload_capa & tx_tso_offload_capa) {
>> > + if ((info.tx_offload_capa & tx_tso_offload_capa) ==
>> tx_tso_offload_capa) {
>> > dev->hw_ol_features |= NETDEV_TX_TSO_OFFLOAD;
>> > } else {
>> > dev->hw_ol_features &= ~NETDEV_TX_TSO_OFFLOAD;
>> >
>> >
>> >
>> > --
>> > David Marchand
>> >
>> > _______________________________________________
>> > dev mailing list
>> > d...@openvswitch.org
>> > https://mail.openvswitch.org/mailman/listinfo/ovs-dev
>>
>> --
>> fbl
>>
>
_______________________________________________
dev mailing list
d...@openvswitch.org
https://mail.openvswitch.org/mailman/listinfo/ovs-dev
