7 Dec
2010
7 Dec
'10
4 p.m.
On Tue, 7 Dec 2010 16:38:35 +0200 Peter Ujfalusi peter.ujfalusi@nokia.com wrote:
If this could happen, than IMHO it has to be handled by the omap_stop_dma, since other drivers could be hit by the same problem (there might be ERRATA for it already for OMAP4).
I agree with Peter that all DMA stopping related workarounds should be in omap_stop_dma. I'd expect that DMA is stopped when returning from there. Commit 9da65a9 was dealing with similar issue so probably this looping should be implemented somewhere around those lines reading & clearing the _CCR_EN bit?
--
Jarkko