[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <CACKFLintU=LqY7MR-ZLT5QMWe8PY2x=rCb_rr7Wy4d-T2JLJ+Q@mail.gmail.com>
Date: Mon, 19 Sep 2016 09:39:08 -0700
From: Michael Chan <michael.chan@...adcom.com>
To: "Mintz, Yuval" <Yuval.Mintz@...ium.com>
Cc: "davem@...emloft.net" <davem@...emloft.net>,
"netdev@...r.kernel.org" <netdev@...r.kernel.org>
Subject: Re: [PATCH net-next 05/10] bnxt_en: Fix ethtool -l|-L inconsistent
channel counts.
On Mon, Sep 19, 2016 at 2:51 AM, Mintz, Yuval <Yuval.Mintz@...ium.com> wrote:
>>
>> If the user chooses 2 rx and 1 tx, he will use 3 msix vectors (3 completion rings,
>> etc). If the user chooses 2 combined (1 with rx/tx,
>> 1 with rx only), he will use 2 msix vectors (2 completion rings, etc).
>> With a large number of NPAR functions and SRIOV functions, the number of
>> rings available per function may not be symmetrical. We just want maximum
>> flexibility to make use of all available resources in 2 different modes (one that
>> uses more resources and one that uses less).
>
> Sounds like the user should chose 1 combined in 1 rx in this scenario.
We only support all combined or all rx/tx. Supporting arbitrary
combination of the 2 will add a lot more complication to the driver
code. I don't think it will add any value to the user either. In the
end, it's all the same and the user sees a simpler interface. If
combined is chosen, the driver logic will combine all available rx/tx
ring pairs. Any remaining rings will still be used to reach the
desired combined count.
Powered by blists - more mailing lists