I think the following change might be sufficient. DCB seems to use its own
init_port_dcb_config(), which would reset mq_mode. If this is good, I can
submit the patch. Thanks.
index b11eb2e..355db0f 100644
--- a/app/test-pmd/testpmd.c
+++ b/app/test-pmd/testpmd.c
@@ -1546,6 +1546,9 @@ init_port_config(void)
if (nb_rxq > 0) {
port->dev_conf.rx_adv_conf.rss_conf.rss_key = NULL;
port->dev_conf.rx_adv_conf.rss_conf.rss_hf = rss_hf;
+ if (nb_rxq > 1 && rss_hf != 0) {
+ port->dev_conf.rxmode.mq_mode =
ETH_MQ_RX_RSS;
+ }
} else {
port->dev_conf.rx_adv_conf.rss_conf.rss_key = NULL;
port->dev_conf.rx_adv_conf.rss_conf.rss_hf = 0;
On Thu, Jan 9, 2014 at 3:11 PM, Thomas Monjalon
<thomas.monjalon at 6wind.com>wrote:
> Hello,
>
> 09/01/2014 10:49, Daniel Kan :
> > The problem appears to be that rxmode.mq_mode was never set to
> ETH_MQ_RX_RSS
> > in testpmd.c; it?s initialized to 0.
>
> You're right. Its broken since the commit "ETH_MQ_RX_NONE should
> disable RSS":
>
> http://dpdk.org/browse/dpdk/commit/?id=243db2ddee3094a2cb39fdd4b17e26df4e7735e1
>
> > There should probably be a configuration for that, or should be set when
> > rxq > 1.
>
> RSS can be configured or disabled with testpmd options or commands.
> So it must be fixed in 2 places:
> - in app/test-pmd/parameters.c for options
> - in app/test-pmd/cmdline.c for commands
> When setting rss_hf, mq_mode must be set accordingly.
>
> Note that DCB feature can use mq_mode.
>
> Thanks for the report. Patch is welcome :)
> --
> Thomas
>
