[Intel-gfx] [PATCH 1/2] drm/i915: Prefault the entire object on first page fault

[Chris Wilson]

On Wed, Jun 11, 2014 at 01:41:45PM -0700, Volkin, Bradley D wrote:
> On Tue, Jun 10, 2014 at 04:14:40AM -0700, Chris Wilson wrote:
> > Inserting additional PTEs has no side-effect for us as the pfn are fixed
> > for the entire time the object is resident in the global GTT. The
> > downside is that we pay the entire cost of faulting the object upon the
> > first hit, for which we in return receive the benefit of removing the
> > per-page faulting overhead.
> > 
> > On an Ivybridge i7-3720qm with 1600MHz DDR3, with 32 fences,
using i-g-t/gem_fence_upload
> > Upload rate for 2 linear surfaces:	8127MiB/s -> 8134MiB/s
> > Upload rate for 2 tiled surfaces:	8607MiB/s -> 8625MiB/s
> > Upload rate for 4 linear surfaces:	8127MiB/s -> 8127MiB/s
> > Upload rate for 4 tiled surfaces:	8611MiB/s -> 8602MiB/s
> > Upload rate for 8 linear surfaces:	8114MiB/s -> 8124MiB/s
> > Upload rate for 8 tiled surfaces:	8601MiB/s -> 8603MiB/s
> > Upload rate for 16 linear surfaces:	8110MiB/s -> 8123MiB/s
> > Upload rate for 16 tiled surfaces:	8595MiB/s -> 8606MiB/s
> > Upload rate for 32 linear surfaces:	8104MiB/s -> 8121MiB/s
> > Upload rate for 32 tiled surfaces:	8589MiB/s -> 8605MiB/s
> > Upload rate for 64 linear surfaces:	8107MiB/s -> 8121MiB/s
> > Upload rate for 64 tiled surfaces:	2013MiB/s -> 3017MiB/s
> > 
Testcase: i-g-t/gem_fence_upload
> > Signed-off-by: Chris Wilson <chris at chris-wilson.co.uk>
> > Cc: "Goel, Akash" <akash.goel at intel.com>
> 
> For reproducibility it would be nice to have the testcase info, assuming
> it's something from i-g-t. Other than that, I think this change looks good.

It was a proposed test case along with the last set of patches. I should
have referenced it properly in the commit.
-Chris

-- 
Chris Wilson, Intel Open Source Technology Centre