Hi Felipe, On Wed, May 19, 2010 at 7:50 PM, Felipe Contreras <felipe.contreras@xxxxxxxxx> wrote: > I don't know what VM_IO means, but essentially we don't want to go > find each and page when we know the memory is contiguous. Although the > fact that the memory is writecombine would help us avoid unnecessary > flushes as well :) Yes, I guess you can remove those calls from the application :) > > A few updates: > 1) libomxil-ti doesn't use the flush/inv ioctls at all... all buffers > are mapped/unmapped I'm not following libomxil-ti but I guess they count on the flushing that happens when the buffers are mapped (as part of the get_user_pages call) > 2) I said only 4 buffers are mapped at a time. Actually, it's more > like 8 * 3 + 1. Most of these are 4K, although only a few bytes are > used. Ok, thanks for the update. This still sounds like a small number that doesn't require overkill data structures. > 4) I'm already doing the changes needed in gst-dsp. Great! I briefly looked at the patches and they seem very nice. Thanks a lot, Ohad. > > Cheers. > > -- > Felipe Contreras > -- 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
