On Mon, Dec 08, 2014 at 02:23:21PM +0000, Russell King - ARM Linux wrote: > On Mon, Dec 08, 2014 at 06:37:27PM +0530, Vinod Koul wrote: > > I actually like the split model, you can also prepare txn ahead of time and > > submit them when needed. > > Actually, you can't - that's not permitted. I have email(s) from Dan > explicitly stating that it is permitted for a driver to take a spinlock > in their prepare callback, and release it when the descriptor is > submitted. Several DMA engine drivers (particularly those in for > async_tx) do exactly that. > > The reason that submit is separate from prepare is to allow DMA engine > users to set a callback - if it weren't for that, there wouldn't be a > submit step, prepare would have done everything. Yes thats right. Do you mind pointing to thread Dan replied, I would like to add these bits and anything else missing to Documentation Thanks -- ~Vinod -- To unsubscribe from this list: send the line "unsubscribe dmaengine" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
