[Mesa-dev] [PATCH 0/3] i965: Use intel_try_pbo_upload for sub updates and 3D textures

[Neil Roberts]

I've made some small patches to try using the blorp blitter on a branch
here:

https://github.com/bpeel/mesa/tree/texture-array-opt

I've also made a little standalone performance test here:

https://github.com/bpeel/glut-fato/blob/master/src/fato-texture-array.c

The test creates a 512x512x64 2D array texture and a PBO with random
data at the same size. At every frame it copies random slices of the PBO
to random layers of the texture to replace every slice.

On my Haswell laptop, if I use the BLT blit engine then I see a 156%
increase in frame rate over the existing CPU blit path. If I use the
blorp blitter then I see 350%.

I'm told that using the blorp blitter also helps on BayTrail with the
SynMark test.

I'll try addressing the issues that Jason mentioned for the patches and
repost the whole branch with the blorp patches on top. I'll leave the
meta blitter path for Jason if he's already working on that.

Regards,
- Neil

Chris Forbes <chrisf at ijw.co.nz> writes:

> Are there some performance numbers to go with this?
>
> On Tue, Dec 23, 2014 at 12:08 PM, Neil Roberts <neil at linux.intel.com> wrote:
>> Here are some patches to make the i965 driver use the blit pipeline
>> for sub-texture updates and also for 3D textures and array textures
>> (including cube map arrays) when using PBOs. I've also posted some
>> patches to update the texsubimage test in Piglit to make sure these
>> codepaths are actually tested.
>>
>> http://cgit.freedesktop.org/piglit/commit/?id=5cfdc9d1d46fce9112845762
>> http://cgit.freedesktop.org/piglit/commit/?id=8433c118dfa6f03f615724d9
>> http://cgit.freedesktop.org/piglit/commit/?id=e4e88dfdb22be7b75150eae4
>> http://lists.freedesktop.org/archives/piglit/2014-December/013814.html
>> http://lists.freedesktop.org/archives/piglit/2014-December/013815.html
>> http://lists.freedesktop.org/archives/piglit/2014-December/013850.html
>>
>> I've tested the series with all tests on Piglit on Ivybridge and it
>> doesn't cause any regressions.
>>
>> _______________________________________________
>> mesa-dev mailing list
>> mesa-dev at lists.freedesktop.org
>> http://lists.freedesktop.org/mailman/listinfo/mesa-dev