commit | 92d04da38f03dfabd8cd9a7244588a49be9a2f41 | [log] [tgz] |
---|---|---|
author | mtklein <mtklein@chromium.org> | Mon Mar 23 12:01:45 2015 -0700 |
committer | Commit bot <commit-bot@chromium.org> | Mon Mar 23 12:01:46 2015 -0700 |
tree | 2acec1a590c4626774a2cc1927c59b116d39c848 | |
parent | b8a1d30a42d13ae83690b2d854a024d9b56e7b71 [diff] |
Replace _mm_cvtps_epi32(x) with _mm_cvttps_epi32(_mm_add_ps(0.5f), x). We don't have control over which way _mm_cvtps_epi32 rounds. - This makes the SSE SkPMFloat rounding consistent with _neon and _none. - Sk4f::cast<Sk4i>() is closer to (int)float's behavior. (Correct when >=0). Add tests that would fail at head. BUG=skia: Review URL: https://codereview.chromium.org/1029163002