On 02/19/2012 05:52 PM, Grazvydas Ignotas wrote: >> Might be an idea to do this in the McBSP driver - have it do the switch >> transparently, then flip back when the port is brought up again? > > Maybe, but I think pandora is the only one using external clock in > mainline I have patch on top of the mcbsp merge series which allows users (developers) to switch between McBSP2 master/slave configuration on Beagle. It will have two PCM: 0 is the current configuration (twl4030 master, mcbsp2 slave) 1 is the same as with pandora (twl4030 slave, mcbsp2 master - CLKS pin is the source for the SRG). With this I can help to track down the suspend issue you see on Pandora. I hope. > and mcbsp currently doesn't track this state, just does a > register write. I was also thinking that this should be handled by the mcbsp driver. I'm really not sure why we need to do this - it might be clock/hwmod issue at the end. We could do this unconditionally when all streams has been stopped on the mcbsp port IMHO. > If you still prefer it on mcbsp side, I can do it > after Peter's mcbsp merge work is finished I guess. I'll wait for Janusz for OMAP1 results before I send the v2 series. So far so good: now the Pandora like configuration also works. -- Péter -- To unsubscribe from this list: send the line "unsubscribe linux-omap" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
