On 09/29/2015 06:18 AM, Jon Hunter wrote:
On 28/09/15 16:07, Arnd Bergmann wrote:
On Monday 28 September 2015 15:57:46 Jon Hunter wrote:
...
Yes that makes sense, but I think that I have confused matters here a
bit and not thought through this entirely. So while we could configure
an audio interface, such as i2s, to use any adma-if port and hence any
dma channel with the appropriate hardware request signal, the mapping
between the adma-if port and request signal is fixed. For example,
adma-if rx1 port uses adma request signal 1
adma-if rx2 port uses adma request signal 2
adma-if rx3 port uses adma request signal 3
...
adma-if rx10 port uses adma request signal 10
and
adma-if tx1 port uses adma request signal 1
adma-if tx2 port uses adma request signal 2
adma-if tx3 port uses adma request signal 3
...
adma-if tx10 port uses adma request signal 10
What is connected to these adma-if tx and rx ports who knows but I think
that is where I was going wrong and made this more complex than it
should have been. So I think that the adma binding should have
#dma-cells = 1 and the adma-if binding have the following ...
admaif@0x702d0000 {
dmas = <&adma 1>, <&adma 1>, <&adma 2>, <&adma 2>,
...
dma-names = "rx1", "tx1", "rx2", "tx2",
...
};
Yes, that sounds about right.
--
To unsubscribe from this list: send the line "unsubscribe dmaengine" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
