[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Date: Wed, 11 Nov 2015 00:44:26 +0000
From: Måns Rullgård <mans@...sr.com>
To: Andy Shevchenko <andy.shevchenko@...il.com>
Cc: "linux-kernel\@vger.kernel.org" <linux-kernel@...r.kernel.org>,
netdev <netdev@...r.kernel.org>, slash.tmp@...e.fr
Subject: Re: [PATCH v5] net: ethernet: add driver for Aurora VLSI NB8800 Ethernet controller
Andy Shevchenko <andy.shevchenko@...il.com> writes:
> On Wed, Nov 11, 2015 at 1:07 AM, Måns Rullgård <mans@...sr.com> wrote:
>> Andy Shevchenko <andy.shevchenko@...il.com> writes:
>
>>>>>> + nb8800_writel(priv, NB8800_TX_DESC_ADDR, txb->dma_desc);
>>>>>> + wmb(); /* ensure desc addr is written before starting DMA */
>>>>>
>>>>> Hm… Have I missed corresponding rmb() ? If it's about MMIO, perhaps mmiowb() ?
>>>>
>>>> Possibly.
>>>
>>> Standalone wmb() doesn't make sense.
>>
>> It does if you need to enforce ordering between normal and I/O memory.
>> In fact, since the descriptor is filled in using normal memory accesses,
>> my understanding is that mmiowb() would be insufficient here. The
>> comment could be improved, however.
>
> Can you then explain what exactly you are assured against in all cases
> where you are using wmb()s? It seems I don't recognize this part in
> some excerpts.
Certainly. I'll re-read memory-barriers.txt to make sure I'm doing the
right thing, then write a better comment.
--
Måns Rullgård
mans@...sr.com
--
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