Re: [PATCH] dmaengine: delete redundant chan_id and chancntinitialization in dma drivers

[Barry Song]

2011/9/22 Jassi Brar <jassisinghbrar@xxxxxxxxx>:
> On Thu, Sep 22, 2011 at 12:01 PM, Barry Song <21cnbao@xxxxxxxxx> wrote:
>> 2011/9/22 Jassi Brar <jassisinghbrar@xxxxxxxxx>:
>>> On Thu, Sep 22, 2011 at 11:11 AM, Jassi Brar <jassisinghbrar@xxxxxxxxx> wrote:
>>>>
>>>>> client drivers can request a specific chan_id just like it can request a gpio.
>>>> No dear. That's not gonna fly if we are to have common client drivers
>>>> across SoCs.
>>>>
>>> Btw, I meant you can't simply have clients specify a channel number and
>>> get the channel in a portable manner. Clients need to specify the capabilities
>>> they need of the channel.
>>
>> if the dmac is flexible in design, what you said is right. but the
>> problem is that there are some chips which actually fix one channel to
>> one peripheral.
> Nevertheless it is better to assume the dmac's req-signals could be hacked
> to re-route at even runtime.
> Also it won't work for client drivers that run the same controller on different
> SoC with a different dmac.
>
you mean there is a mapping table in dmac driver. otherwise, there
will always require some changes in client drivers.
-barry
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/