Re: [PATCH/RFC] dmaengine: add a slave parameter to __dma_request_channel()

From: Vinod Koul <vinod.koul@intel.com>
To: Guennadi Liakhovetski <g.liakhovetski@gmx.de>
Cc: Russell King - ARM Linux <linux@arm.linux.org.uk>,
	linux-kernel@vger.kernel.org,
	"'Jassi Brar'" <jassisinghbrar@gmail.com>,
	Linus Walleij <linus.walleij@linaro.org>,
	Magnus Damm <magnus.damm@gmail.com>,
	Paul Mundt <lethal@linux-sh.org>
Subject: Re: [PATCH/RFC] dmaengine: add a slave parameter to __dma_request_channel()
Date: Mon, 19 Mar 2012 19:30:04 +0530	[thread overview]
Message-ID: <1332165604.7180.24.camel@vkoul-udesk3> (raw)
In-Reply-To: <Pine.LNX.4.64.1203191436030.11516@axis700.grange>

On Mon, 2012-03-19 at 14:38 +0100, Guennadi Liakhovetski wrote:
> On Mon, 19 Mar 2012, Vinod Koul wrote:
> 
> > On Mon, 2012-03-19 at 12:47 +0100, Guennadi Liakhovetski wrote:
> > > As I wrote in a reply to Linus W - you need to pass information about the 
> > > requesting client to the dmaengine core to let it match it against mapping 
> > > tables. 
> > NO. 
> > The client only needs to say that he needs a channel for DMA_SLAVE
> 
> How?
Did you miss our earlier discussion on 
https://lkml.org/lkml/2012/3/8/26

Perhaps, I can enlighten you will below excerpt where we proposed to
add:

int dmaengine_add_channel_map(struct dmaengine_map *map, unsigned int num_entries)
{
	/* store this map into dmaengine and use for channle allocation */
}

So dmaengine knows client A can use DMAC P, channel 2 and 3.
So first request will receive ch 2 for A and second one will yield 3.

> 
> 	dma_request_channel(mask, filter_fn, filter_arg)
> 
> Where shall the client pass to dmaengine its identity info?
Only change I see for above to tell which client is requesting, so we
may have to add device pointer of client while requesting.
> 
> > DMAengine will know for this client, the platform channel map (already
> > given to it by platform) says that we can give it DMAC X, channel 4
> > only.
> 
> Some clients need multiple channels - Tx, Rx,...
> 
> Thanks
> Guennadi
> 
> > So see if it free, if so allocate it and give to client (while
> > doing usual stuff)
> > > You have to pass this information with the dma_request_channel() 
> > > function. So, either you need to add a parameter or you have to reuse one 
> > > of existing ones, e.g., deprecate the filter and use its argument for this 
> > > purpose. If you do this and as long as you pass that parameter further on 
> > > to the dmaengine device (controller) driver after whatever matching you 
> > > like to do in the core - I'm fine with that, that fits well with my 
> > > initial proposal.
> > I don't care about filter, it can go away if it is not required.
> > 
> > Passing slave_config is *enhancement* so for (hopefully) last time
> >  a) it has *nothing* to do with getting a channel, no role to play in
> > generic scheme of things
> >  b) it allows client to call one api for get+configure thats all! 
> > 
> > -- 
> > ~Vinod
> > 
> 
> ---
> Guennadi Liakhovetski, Ph.D.
> Freelance Open-Source Software Developer
> http://www.open-technology.de/
> --
> To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html
> Please read the FAQ at  http://www.tux.org/lkml/

-- 
~Vinod