Kevin, As I pointed out the queuing model has changed between the existing CPPI3.0 DMA implementation in the kernel and the new CPPI4.1 DMA implementation. There are associated changes relating to initialization sequence.
Given this I think we can go ahead with the existing implementation from Sergei (for cppi4.1) and if need be in future merge the implementation into a single CPPI DMA implementation. Can we follow this approach in that way we will be able to enable the community to start using DA8xx platform sooner. regards swami ________________________________________ From: [email protected] [[email protected]] On Behalf Of Kevin Hilman [[email protected]] Sent: Tuesday, March 24, 2009 6:14 AM To: David Brownell Cc: [email protected]; [email protected] Subject: Re: [PATCH] DaVinci: add support for CPPI 4.1 David Brownell <[email protected]> writes: > On Monday 23 March 2009, Sergei Shtylyov wrote: >> I was just trying to outline the large difference between 3.0 and 4.1 >> that warrants a new driver, so I'm not sure why TI people started to argue... > > I think the pushback was on your statement that they're > so different as to be incompatible through the only > programming interface (musb_hdrc) that will use this. So, if I understand this correctly, they are indeed quite different, but none of those differences are actually relevant to the programming interface actually used. So, that begs the original question: Why can't the current version simply be extended instead of creating an entirely new driver? Kevin _______________________________________________ Davinci-linux-open-source mailing list [email protected] http://linux.davincidsp.com/mailman/listinfo/davinci-linux-open-source_______________________________________________ Davinci-linux-open-source mailing list [email protected] http://linux.davincidsp.com/mailman/listinfo/davinci-linux-open-source
