| commit | cbf4fba43933302a846872e4c5ce8f1adb8b325e | [log] [tgz] |
|---|---|---|
| author | mtklein <mtklein@chromium.org> | Tue Nov 17 14:19:52 2015 -0800 |
| committer | Commit bot <commit-bot@chromium.org> | Tue Nov 17 14:19:52 2015 -0800 |
| tree | 96dad6cc0a2241544a0cf52cccdc7a0fbe89f9b1 | |
| parent | 56847a65648af4d06da9c26c55242949a1bf31ab [diff] |
div255(x) as ((x+128)*257)>>16 with SSE _mm_mulhi_epu16 makes the (...*257)>>16 part simple. This seems to speed up every transfermode that uses div255(), in the 7-25% range. It even appears to obviate the need for approxMulDiv255() on SSE. I'm not sure about NEON yet, so I'll keep approxMulDiv255() for now. Should be no pixels change: https://gold.skia.org/search2?issue=1452903004&unt=true&query=source_type%3Dgm&master=false BUG=skia: CQ_EXTRA_TRYBOTS=client.skia:Test-Ubuntu-GCC-GCE-CPU-AVX2-x86_64-Release-SKNX_NO_SIMD-Trybot Review URL: https://codereview.chromium.org/1452903004