When checking if we are DMA capable we have to check for the
IMX_SSI_DMA flag which is already set from platform_data instead
of setting it again when we want to do DMA.
Signed-off-by: NSascha Hauer <s.hauer@pengutronix.de>
Acked-by: NLiam Girdwood <lrg@Slimlogic.co.uk>
Signed-off-by: NMark Brown <broonie@opensource.wolfsonmicro.com>

This fixes a memory corruption when ASoC devices are used in
full-duplex mode. Specifically for pxa-ssp code, where this pointer
is dynamically allocated for each direction and destroyed upon each
stream start.

All other platforms are fixed blindly, I couldn't even compile-test
them. Sorry for any breakage I may have caused.

[Note that this is a backported version for 2.6.34.
 Upstream commit is fd23b7de]
Signed-off-by: NDaniel Mack <daniel@caiaq.de>
Reported-by: NSven Neumann <s.neumann@raumfeld.com>
Reported-by: NMichael Hirsch <m.hirsch@raumfeld.com>
Acked-by: NLiam Girdwood <lrg@slimlogic.co.uk>
Signed-off-by: NMark Brown <broonie@opensource.wolfsonmicro.com>

The clocks for the SSI block need handling before this can work.
Signed-off-by: NMark Brown <broonie@opensource.wolfsonmicro.com>

While dynamically allocated DAIs are the way forward the core doesn't
yet support anything except matching with a pointer to the actual DAI
so convert to doing that so that machine drivers don't have to jump
through hoops to register themselves.
Signed-off-by: NMark Brown <broonie@opensource.wolfsonmicro.com>
Acked-by: NJavier Martin <javier.martin@vista-silicon.com>

The old driver has the number of SSI units in the system hardcoded,
does not make use of the device model and works only on i.MX21/27.

This driver replaces it. It works in DMA mode on i.MX21/27 and using
an FIQ handler on other systems. It also supports AC97 mode of
the SSI units.
Signed-off-by: NSascha Hauer <s.hauer@pengutronix.de>
Acked-by: NJavier Martin <javier.martin@vista-silicon.com>
Acked-by: NLiam Girdwood <lrg@slimlogic.co.uk>
Signed-off-by: NMark Brown <broonie@opensource.wolfsonmicro.com>