[PATCH v3 0/5] dma: cppi41: some trivial fixes and support for suspend/resume

[Daniel Mack <zonque@gmail.com>]

Here are some patches to teach the cppi41 DMA driver support for
suspend and resume. Patches 1-3 are simply cosmetic things that emerged
during my debugging sessions.

Patch #4 is actually a real bugfix which I would like Sebastian
Andrzej Siewior to have a look at. Quite frankly, the allocation scheme
in this driver and the logic to determine a descriptor's index number
seems quite complicated to me, but that's probably a different story.
Allocating and freeing the exact same pointer more than once is
certainly a bug.

Patch #5 adds support for suspend and resume. As the commit log says,
the code needed to achive this is the result of a trial-and-error
session. This is the minimum that's I'm left with now and which works
for me.

On a different note, I'm also working on patches for the musb core to
make it suspend and resume. Currently, I still have to rmmod/insmod
musb_dsps before/after the resume cycle, but I'm hoping to make
progress here soon.


Thanks,
Daniel

v2 -> v3:
	* Patch #5: kill DEV_PM_OPS macro hack and fix a typo
	  (Reported by Sergei Shtylyov)

v1 -> v2:
	* Patch #5: depend on CONFIG_PM_SLEEP rather than on
	  CONFIG_PM and use SIMPLE_DEV_PM_OPS
	  (Reported by Sergei Shtylyov)

Daniel Mack (5):
  dma: cppi41: pass around device instead of platform_device
  dma: cppi41: s/deinit_cpii41/deinit_cppi41/
  dma: cppi41: add shortcut to &pdev->dev in cppi41_dma_probe()
  dma: cppi41: only allocate descriptor memory once
  dma: cppi41: add support for suspend and resume

 drivers/dma/cppi41.c | 115 +++++++++++++++++++++++++++++++--------------------
 1 file changed, 70 insertions(+), 45 deletions(-)

-- 
1.8.3.1