fix in-gamut check in SkRasterPipeline::append_constant_color()
We're checking that we can fit in a byte, but not that r≤a, etc.
Alone, this will kick us into the float pipeline in a few
places where we thought incorrectly we were able to use the
lowp pipeline before, mostly in --config narrow.
I hope this will also let us strength-reduce some of the clamping
we do in SkRasterPipelineBlitter (clamp_gamut). Today I'm finding
some of the assumptions I was making there are actually wrong.
Change-Id: I65274d0c30d2228a0337dd763e35bf5c47b6a957
Reviewed-on: https://skia-review.googlesource.com/152584
Reviewed-by: Brian Osman <brianosman@google.com>
Commit-Queue: Mike Klein <mtklein@google.com>
1 file changed