On 2016-09-26 5:34 AM, Michel Dänzer wrote:
> On 26/09/16 11:23 AM, Keith Packard wrote:
>> A data-driven approach would be awesome here. Do we have a reasonable
>> performance metric? I'm no fan of gratuitous complexity, but text
>> performance is pretty important to me.
>
> Here are some data points with the radeon driver using glamor on
> radeonsi. This is comparing the default configuration with Option
> "TearFree" (which uses extents based damage tracking as described in
> my other post):
>
> 1: /tmp/baseline.txt
> 2: /tmp/tearfree.txt
>
> 1 2 Operation
> ------------ ------------------------- -------------------------
> 194000000.0 148000000.0 ( 0.763) Dot
> 9660000.0 9150000.0 ( 0.947) Char in 80-char aa line (Charter
> 10)
I've found that careful use of the _mm_max_epi16/_mm_min_epi16
intrinsics pushes the cost for calculating the bounds of Dot down into
the noise. Presumably the non-Intel platforms have something similar in
their SIMD instruction sets too, but I'm not sure how "pixman" you want
to go on this.
Peter Harris
--
Open Text Connectivity Solutions Group
Peter Harris http://connectivity.opentext.com/
Research and Development Phone: +1 905 762 6001
[email protected] Toll Free: 1 877 359 4866
_______________________________________________
[email protected]: X.Org development
Archives: http://lists.x.org/archives/xorg-devel
Info: https://lists.x.org/mailman/listinfo/xorg-devel
