Re: [Intel-gfx] [PATCH 3/6] drm/i915: Support write only mappings

2011-09-20 Thread Chris Wilson
On Mon, 19 Sep 2011 21:25:03 -0700, Ben Widawsky wrote: > When doing a GTT mapping, the pages are not backed until taking a fault. > If we know that the object is write only, when the fault happens we do not > need > to make the pages coherent with the CPU. This allows for semi fast prefaults >

Re: [Intel-gfx] [PATCH 3/6] drm/i915: Support write only mappings

2011-09-19 Thread Ben Widawsky
On Mon, Sep 19, 2011 at 10:29:15PM -0700, Keith Packard wrote: > On Mon, 19 Sep 2011 21:25:03 -0700, Ben Widawsky wrote: > > When doing a GTT mapping, the pages are not backed until taking a fault. > > If we know that the object is write only, when the fault happens we do not > > need > > to make

Re: [Intel-gfx] [PATCH 3/6] drm/i915: Support write only mappings

2011-09-19 Thread Keith Packard
On Mon, 19 Sep 2011 21:25:03 -0700, Ben Widawsky wrote: > When doing a GTT mapping, the pages are not backed until taking a fault. > If we know that the object is write only, when the fault happens we do not > need > to make the pages coherent with the CPU. This allows for semi fast prefaults >

[Intel-gfx] [PATCH 3/6] drm/i915: Support write only mappings

2011-09-19 Thread Ben Widawsky
When doing a GTT mapping, the pages are not backed until taking a fault. If we know that the object is write only, when the fault happens we do not need to make the pages coherent with the CPU. This allows for semi fast prefaults to occur in libdrm at map time. For the non-GTT case, there isn't m