Re: [PATCH 06/13] DMAENGINE: driver for the ARM PL080/PL081PrimeCells

From: Russell King - ARM Linux
Date: Wed Dec 22 2010 - 07:23:32 EST


On Tue, Dec 21, 2010 at 06:20:37PM +0000, Russell King - ARM Linux wrote:
> Having just looked at this while trying to undo the DMA API abuses
> in the PL011 UART driver, I'm getting rather frustrated with this
> code.

Right, just tried this on the Versatile PB/926, which has a PL080.
The result is DMA errors. This turns out to be the hard-coding of
which AHB bus is used.

You can't hard-code this information into the driver - it's part of
the bus matrix configuration. On Versatile PB/926, the two AHB
buses have different memory maps - see DUI0224 page 3-13:

* DMA0 (which is DMA AHB M1) has access to the APB peripherals
but not the system memory.
* DMA1 (which is DMA AHB M2) has access to the system memory but
none of the APB peripherals.

Throwing in #ifdef's to sort this out resolves the problem on the PB/926.

Looking at the driver code too, I'm having a hard time understanding it,
probably because of the use of "master" and "slave" - I don't think this
has anything to do with which AHB master is used. The comments against
pl08x_choose_master_bus() seem to imply that it controls which AHB master
is used, but it has no apparant effect on that. I don't think this
function does anything like that anymore.
--
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/