[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <d3aae1c0-a9ac-b79d-fed9-0f57230167de@mellanox.com>
Date: Mon, 11 Feb 2019 15:38:22 +0000
From: Tariq Toukan <tariqt@...lanox.com>
To: Matthew Wilcox <willy@...radead.org>,
Tariq Toukan <tariqt@...lanox.com>
CC: Ilias Apalodimas <ilias.apalodimas@...aro.org>,
David Miller <davem@...emloft.net>,
"brouer@...hat.com" <brouer@...hat.com>,
"toke@...hat.com" <toke@...hat.com>,
"netdev@...r.kernel.org" <netdev@...r.kernel.org>,
"mgorman@...hsingularity.net" <mgorman@...hsingularity.net>,
"linux-mm@...ck.org" <linux-mm@...ck.org>
Subject: Re: [RFC, PATCH] net: page_pool: Don't use page->private to store
dma_addr_t
On 2/11/2019 2:12 PM, Matthew Wilcox wrote:
> On Mon, Feb 11, 2019 at 08:53:19AM +0000, Tariq Toukan wrote:
>> It's great to use the struct page to store its dma mapping, but I am
>> worried about extensibility.
>> page_pool is evolving, and it would need several more per-page fields.
>> One of them would be pageref_bias, a planned optimization to reduce the
>> number of the costly atomic pageref operations (and replace existing
>> code in several drivers).
>
> There's space for five words (20 or 40 bytes on 32/64 bit).
>
OK so this is good for now, and for the near future.
Fine by me.
Regards,
Tariq
Powered by blists - more mailing lists