[Mesa-dev] i965: slow glTexSubImage2D because of Y tiling

[Kenneth Graunke]

On 12/27/2013 05:56 AM, Tomash Brechko wrote:
> So I wonder if there are strong reasons to prefer Y tiling for textures,
> at least on gen == 6?  Is Y tiling preffered because of faster texel
> lookups (assuming glTexSubImage2D() is not called on each frame)?

Yes, I observed an ~8% improvement in frames per second on some
benchmarks.  As I understand it, Y-tiling is more efficient on the GPU.
 I can't remember whether I tested that on Gen6 or Gen7, though.

There are a bunch of people working on improving texture upload
performance.  It sounds like we need some faster paths for Y-tiled
buffers...

And perhaps we should add a new path for DrawPixels...

--Ken