[Mesa-dev] [PATCH] i965/tiled_memcpy: ytiled_to_linear a cache line at a time

Chris Wilson chris at chris-wilson.co.uk
Thu Feb 8 08:06:14 UTC 2018

Quoting Scott D Phillips (2018-02-08 00:16:34)
> Chris Wilson <chris at chris-wilson.co.uk> writes:
> > From: Scott D Phillips <scott.d.phillips at intel.com>
> >
> > Similar to the transformation applied to linear_to_ytiled, also align
> > each readback from the ytiled source to a cacheline (i.e. transfer a
> > whole cacheline from the source before moving on to the next column).
> > This will allow us to utilize movntqda (_mm_stream_si128) in a
> > subsequent patch to obtain near WB readback performance when accessing
> > the uncached ytiled memory, an order of magnitude improvement.
> lgtm, or R-b if you want to take the authorship.

It's your work, so please keep the authorship and land the
improvements :)

More information about the mesa-dev mailing list