Em 22-07-2011 16:55, Jonathan Corbet escreveu: >>> The problem is that there's no convenient callback into the allocators >>> where the mapping and unmapping can be done now. So I'd have had to add a >>> couple of memops to do that. >> >> I think that some additional callbacks for allocators for synchronization >> buffer state will be required sooner or later anyway, so imho it is better >> to add them now to avoid massive fixing the drivers in the future. > > OK, I can certainly do a version of the patch along those lines. I'd > envision some sort of give_buffer_to_device() and give_buffer_to_cpu() > calls (with better names). It'll take me a little while to get it done, > though - travel and such are upon me. Hi Jon, I'm understanding that I can just mark those two patches as RFC at patchwork and wait for your version two of those patchsets. So, I'll just remove those two from my pending queue and move on ;) Thanks, mauro -- To unsubscribe from this list: send the line "unsubscribe linux-media" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
