[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20080213202402.22818482@siona>
Date: Wed, 13 Feb 2008 20:24:02 +0100
From: Haavard Skinnemoen <hskinnemoen@...el.com>
To: "Dan Williams" <dan.j.williams@...el.com>
Cc: linux-kernel@...r.kernel.org,
"Shannon Nelson" <shannon.nelson@...el.com>,
"David Brownell" <david-b@...bell.net>, kernel@...32linux.org,
"Francis Moreau" <francis.moro@...il.com>,
"Paul Mundt" <lethal@...ux-sh.org>,
"Vladimir A. Barinov" <vbarinov@...mvista.com>,
"Pierre Ossman" <drzeus-list@...eus.cx>
Subject: Re: [RFC v3 4/7] dmaengine: Add slave DMA interface
On Wed, 13 Feb 2008 12:07:26 -0700
"Dan Williams" <dan.j.williams@...el.com> wrote:
> > +struct dma_slave_descriptor {
> > + struct dma_async_tx_descriptor txd;
> > + struct list_head client_node;
> > +};
>
> Can you explain a bit why client_node is needed? I do not think we
> need dma_slave_descriptor if dma_unmap data / control is added to
> dma_async_tx_descriptor. Hmm?
Well, it's perhaps not needed for slave transfers as such. But the MMC
driver (and I suspect quite a few other users of the slave interface)
deals with scatterlists, so it needs a way to keep track of all the
descriptors it submits. Hence the list node.
But looking at your latest patch series, I guess we can use the new
"next" field instead. It's not like we really need the full
capabilities of list_head.
Haavard
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
Powered by blists - more mailing lists