[Intel-gfx] Stolen memory, again.

Chris Wilson chris at chris-wilson.co.uk
Tue Sep 4 22:02:52 CEST 2012


This addresses all the comments I've received so far, except for the
handling of pwrite/pread through additional obj->ops as I think the cure
was worse than the disease in that case (due to the fiddling around
required for sg iterator and tracking individual struct page operations).
-Chris




More information about the Intel-gfx mailing list