[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <CAKgT0UeAJOdvqXfRw1xF+LnpGo3K7NQt_3DcJGUkiPMF+O8PTg@mail.gmail.com>
Date: Fri, 23 Mar 2018 11:31:34 -0700
From: Alexander Duyck <alexander.duyck@...il.com>
To: Sinan Kaya <okaya@...eaurora.org>
Cc: Jeff Kirsher <jeffrey.t.kirsher@...el.com>, sulrich@...eaurora.org,
Netdev <netdev@...r.kernel.org>,
Timur Tabi <timur@...eaurora.org>,
LKML <linux-kernel@...r.kernel.org>,
intel-wired-lan <intel-wired-lan@...ts.osuosl.org>,
linux-arm-msm@...r.kernel.org, linux-arm-kernel@...ts.infradead.org
Subject: Re: [Intel-wired-lan] [PATCH v6 7/7] ixgbevf: eliminate duplicate
barriers on weakly-ordered archs
On Fri, Mar 23, 2018 at 11:27 AM, Sinan Kaya <okaya@...eaurora.org> wrote:
> On 3/23/2018 2:25 PM, Alexander Duyck wrote:
>>> + /* We need this if more than one processor can write to our tail
>>> + * at a time, it synchronizes IO on IA64/Altix systems
>>> + */
>>> + mmiowb();
>>> }
>> The mmiowb shouldn't be needed for Rx. Only one CPU will be running
>> NAPI for the queue and we will synchronize this with a full writel
>> anyway when we re-enable the interrupts.
>>
>
> OK. I can fix this on the next version. I did a blanket search and replace for
> my writel_relaxed() changes as I don't know the code well enough.
>
> Please point me to the redundant ones.
So from what I can tell only this file and i40e needed any additional
mmiowb calls added. The rest are not needed.
- Alex
Powered by blists - more mailing lists