[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <CAGXr9JH-iEE-d=pZvR1KOsK+URCxyXL-GW4JtaF49DEGZO56eg@mail.gmail.com>
Date: Wed, 25 Mar 2015 15:53:44 -0700
From: Petri Gynther <pgynther@...gle.com>
To: Florian Fainelli <f.fainelli@...il.com>
Cc: netdev <netdev@...r.kernel.org>,
David Miller <davem@...emloft.net>,
Jaedon Shin <jaedon.shin@...il.com>
Subject: Re: [PATCH net-next 0/7] net: bcmgenet: multiple Rx queues support
Hi Florian,
On Wed, Mar 25, 2015 at 3:32 PM, Florian Fainelli <f.fainelli@...il.com> wrote:
> On 25/03/15 12:35, Petri Gynther wrote:
>> Final patch set to add support for multiple Rx queues:
>> 1. remove priv->int0_mask and priv->int1_mask
>> 2. modify Tx ring int_enable and int_disable vectors
>> 3. simplify bcmgenet_init_dma()
>> 4. tweak init_umac()
>> 5. rework Tx NAPI code
>> 6. rework Rx NAPI code
>> 7. add support for multiple Rx queues
>>
>
> Looks very good, thanks for getting these change out! Now I suppose what
> we would need is some sort of ethtool rx-nfc API to allow flow
> configuration towards particular RX queues?
> --
> Florian
Thanks for the quick review. Yes, as the next step, we would need
ethtool rx-nfc API to allow Rx flow configuration to various Rx
queues. Also, Rx priority queues are still off by default:
[GENET_V3] = {
.tx_queues = 4,
.tx_bds_per_q = 32,
.rx_queues = 0,
.rx_bds_per_q = 0,
[GENET_V4] = {
.tx_queues = 4,
.tx_bds_per_q = 32,
.rx_queues = 0,
.rx_bds_per_q = 0,
-- Petri
--
To unsubscribe from this list: send the line "unsubscribe netdev" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Powered by blists - more mailing lists