Commit Graph

30 Commits

Author SHA1 Message Date
Alexander Weinhold
c2f707aba7 change paths 2018-12-06 17:03:52 +01:00
Nigel Tao
46e4eb730a vector: use asm opcode mnemonics
There's no change in the binary output, just less mystery in the asm.

These mnemonics were introduced in Go 1.10:
https://golang.org/doc/go1.10#asm and https://golang.org/cl/75490

Current stable release (as of 2018-11-10) is Go 1.11, and
https://golang.org/doc/devel/release.html#policy says that Go 1.9 and
below are therefore no longer supported.

Change-Id: I1f9a63521bc8d5e8f8d395605f62bf7fb6a63bc5
Reviewed-on: https://go-review.googlesource.com/c/148997
Reviewed-by: Dmitri Shuralyov <dmitshur@golang.org>
2018-11-15 02:24:11 +00:00
Dmitri Shuralyov
4b0b9b997d vector: remove Go 1.8 support
This change reverts CL 32132. We no longer have to worry about
Go 1.5 support. This results in slightly simpler build constraints.

Change-Id: I021936646a6e9a82462a885b256894027fb58aeb
Reviewed-on: https://go-review.googlesource.com/c/148568
Reviewed-by: Brad Fitzpatrick <bradfitz@golang.org>
Run-TryBot: Brad Fitzpatrick <bradfitz@golang.org>
TryBot-Result: Gobot Gobot <gobot@golang.org>
2018-11-08 20:57:45 +00:00
Michael Munday
e20db36d77 vector: prevent fused multiply-add being used in floatingLineTo
The test code now produces bit-identical results across platforms.

Fixes golang/go#21460.

Change-Id: I58b72dacf3c7930e2804b4e37bd3ccd9ef3dc2f0
Reviewed-on: https://go-review.googlesource.com/61024
Run-TryBot: Michael Munday <mike.munday@ibm.com>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Reviewed-by: Nigel Tao <nigeltao@golang.org>
2017-09-07 23:27:51 +00:00
Nigel Tao
d1087bc575 vector: benchmark the floating point code path.
Specifically, this commit adds Benchmark.*1024.* functions, and then
gofmt's some other Benchmark functions.

There are existing floating point benchmarks in acc_test.go, but they
have narrower scope. The benchmarks in vector_test.go are closer to
end-to-end benchmarks for the package.

Change-Id: I5b4770bcefa9aa3ed4d4823f78993df7b80c7d87
Reviewed-on: https://go-review.googlesource.com/62130
Reviewed-by: David Symonds <dsymonds@golang.org>
2017-09-07 00:02:32 +00:00
Nigel Tao
507b1a44bd vector: measure the fixed point rasterizer's accuracy.
When commit 8874bef1 changed ϕ from 10 to 9, this total diff grew from:
Total diff = 3051, n = 208, avg = 14.66827 out of 65535, or 0.02238%.
to:
Total diff = 6412, n = 208, avg = 30.82692 out of 65535, or 0.04704%.

Also fix a comment typo.

Change-Id: I91cc861a3d3ab282b5431cfb141fb010fd1f8d14
Reviewed-on: https://go-review.googlesource.com/32771
Reviewed-by: David Crawshaw <crawshaw@golang.org>
2016-11-15 01:09:02 +00:00
Nigel Tao
98f3e4e74d vector: make args float32 pairs instead of f32.Vec2.
The f32.Vec2 type doesn't seem worth it.

Change-Id: I021c7e13d7e2dd261334f4aa7e867df4fd8f1c3e
Reviewed-on: https://go-review.googlesource.com/32772
Reviewed-by: David Crawshaw <crawshaw@golang.org>
2016-11-15 00:52:01 +00:00
Nigel Tao
c78039e8ce vector: tweak the asm register assignment.
This makes the fixed point and floating point code more similar.

Benchmarks don't show any significant change.

Change-Id: I723fa1605eaa248b40e784201b680c16cc3d26a2
Reviewed-on: https://go-review.googlesource.com/32134
Reviewed-by: David Crawshaw <crawshaw@golang.org>
2016-10-26 22:33:17 +00:00
Nigel Tao
0ed2caa453 vector: add go1.6 build tag for asm code.
Go 1.5 doesn't understand amd64 opcode mnemonics like PADDD or PMULLD.

Change-Id: Ic31ab4d2a8a222ca18288387de40db6ce673d7f1
Reviewed-on: https://go-review.googlesource.com/32132
Reviewed-by: David Crawshaw <crawshaw@golang.org>
2016-10-26 22:29:37 +00:00
Nigel Tao
93fad3647f vector: remove a shift from the inner loop.
PSHUFB already lets us pick certain bytes out of an XMM register.
There's no need to shift by 8 bits (1 byte) beforehand.

The code generator is simpler. There's also a small win on the
benchmarks, especially for FixedAccumulateOpOverSIMD.

name                              old time/op  new time/op  delta
FixedAccumulateOpOverSIMD16-8      183ns ± 0%   174ns ± 0%  -4.92%    (p=0.000 n=8+8)
FixedAccumulateOpSrcSIMD16-8      87.0ns ± 1%  86.3ns ± 0%  -0.77%   (p=0.000 n=10+9)
FixedAccumulateMaskSIMD16-8       80.4ns ± 1%  81.2ns ± 1%  +1.01%  (p=0.000 n=10+10)
FloatingAccumulateOpOverSIMD16-8   250ns ± 1%   244ns ± 0%  -2.39%   (p=0.000 n=10+8)
FloatingAccumulateOpSrcSIMD16-8    176ns ± 1%   176ns ± 0%    ~      (p=0.142 n=10+8)
FloatingAccumulateMaskSIMD16-8     167ns ± 0%   167ns ± 0%    ~      (p=0.137 n=8+10)
FixedAccumulateOpOverSIMD64-8     2.73µs ± 1%  2.58µs ± 0%  -5.36%   (p=0.000 n=10+7)
FixedAccumulateOpSrcSIMD64-8      1.18µs ± 1%  1.17µs ± 0%  -0.33%   (p=0.003 n=10+9)
FixedAccumulateMaskSIMD64-8       1.09µs ± 0%  1.09µs ± 0%  -0.17%    (p=0.047 n=9+9)
FloatingAccumulateOpOverSIMD64-8  3.67µs ± 0%  3.61µs ± 1%  -1.47%   (p=0.000 n=7+10)
FloatingAccumulateOpSrcSIMD64-8   2.60µs ± 0%  2.61µs ± 0%  +0.19%    (p=0.003 n=8+8)
FloatingAccumulateMaskSIMD64-8    2.47µs ± 0%  2.46µs ± 0%    ~      (p=0.162 n=10+9)
GlyphAlpha16Over-8                2.99µs ± 0%  2.98µs ± 1%  -0.50%   (p=0.021 n=9+10)
GlyphAlpha16Src-8                 2.89µs ± 1%  2.89µs ± 0%    ~     (p=0.381 n=10+10)
GlyphAlpha32Over-8                4.53µs ± 0%  4.50µs ± 0%  -0.83%  (p=0.000 n=10+10)
GlyphAlpha32Src-8                 4.14µs ± 0%  4.13µs ± 0%  -0.21%   (p=0.026 n=9+10)
GlyphAlpha64Over-8                8.97µs ± 1%  8.80µs ± 0%  -1.85%   (p=0.000 n=10+9)
GlyphAlpha64Src-8                 7.42µs ± 1%  7.39µs ± 0%  -0.45%  (p=0.011 n=10+10)
GlyphAlpha128Over-8               21.8µs ± 0%  21.2µs ± 0%  -2.91%    (p=0.000 n=9+9)
GlyphAlpha128Src-8                15.6µs ± 0%  15.6µs ± 0%    ~      (p=0.982 n=10+7)
GlyphAlpha256Over-8               66.3µs ± 1%  63.7µs ± 0%  -3.84%   (p=0.000 n=10+9)
GlyphAlpha256Src-8                41.2µs ± 1%  41.2µs ± 1%    ~     (p=1.000 n=10+10)
GlyphRGBA16Over-8                 4.75µs ± 0%  4.75µs ± 1%    ~      (p=0.735 n=9+10)
GlyphRGBA16Src-8                  4.20µs ± 0%  4.20µs ± 0%    ~       (p=0.503 n=8+8)
GlyphRGBA32Over-8                 11.4µs ± 0%  11.4µs ± 0%    ~       (p=0.119 n=9+9)
GlyphRGBA32Src-8                  9.34µs ± 1%  9.32µs ± 0%    ~       (p=0.062 n=9+8)
GlyphRGBA64Over-8                 36.0µs ± 0%  36.1µs ± 0%    ~       (p=0.209 n=8+9)
GlyphRGBA64Src-8                  27.9µs ± 1%  27.8µs ± 0%    ~     (p=0.796 n=10+10)
GlyphRGBA128Over-8                 131µs ± 0%   131µs ± 0%    ~       (p=0.931 n=9+9)
GlyphRGBA128Src-8                 97.9µs ± 0%  97.7µs ± 1%    ~      (p=0.053 n=9+10)
GlyphRGBA256Over-8                 503µs ± 0%   503µs ± 1%    ~      (p=0.274 n=8+10)
GlyphRGBA256Src-8                  370µs ± 0%   369µs ± 0%    ~      (p=0.497 n=9+10)

Change-Id: I56651e70b258792b83ea2a74904756243c88bef4
Reviewed-on: https://go-review.googlesource.com/31537
Reviewed-by: David Crawshaw <crawshaw@golang.org>
2016-10-22 21:39:22 +00:00
Nigel Tao
8874bef159 vector: change ϕ from 10 to 9.
This slight loss in quality allows us to use int32 math exclusively
throughout raster_fixed.go, instead of occasionally dropping into int64
math. The change in ϕ doesn't affect the benchmarks noticably, but
staying in int32 does. The net effect:

name                              old time/op  new time/op  delta
GlyphAlpha16Over-8                3.36µs ± 0%  2.99µs ± 0%  -10.89%         (p=0.000 n=10+9)
GlyphAlpha16Src-8                 3.26µs ± 0%  2.89µs ± 1%  -11.34%         (p=0.000 n=9+10)
GlyphAlpha32Over-8                5.20µs ± 0%  4.53µs ± 0%  -12.76%         (p=0.000 n=8+10)
GlyphAlpha32Src-8                 4.81µs ± 1%  4.14µs ± 0%  -13.91%          (p=0.000 n=9+9)
GlyphAlpha64Over-8                10.2µs ± 0%   9.0µs ± 1%  -11.99%         (p=0.000 n=9+10)
GlyphAlpha64Src-8                 8.62µs ± 0%  7.42µs ± 1%  -13.89%         (p=0.000 n=9+10)
GlyphAlpha128Over-8               24.1µs ± 0%  21.8µs ± 0%   -9.32%          (p=0.000 n=9+9)
GlyphAlpha128Src-8                17.9µs ± 0%  15.6µs ± 0%  -12.68%         (p=0.000 n=9+10)
GlyphAlpha256Over-8               70.1µs ± 0%  66.3µs ± 1%   -5.44%        (p=0.000 n=10+10)
GlyphAlpha256Src-8                45.2µs ± 1%  41.2µs ± 1%   -8.92%        (p=0.000 n=10+10)
GlyphRGBA16Over-8                 5.12µs ± 0%  4.75µs ± 0%   -7.15%         (p=0.000 n=10+9)
GlyphRGBA16Src-8                  4.57µs ± 1%  4.20µs ± 0%   -8.18%          (p=0.000 n=9+8)
GlyphRGBA32Over-8                 12.1µs ± 0%  11.4µs ± 0%   -5.50%         (p=0.000 n=10+9)
GlyphRGBA32Src-8                  10.0µs ± 0%   9.3µs ± 1%   -6.80%         (p=0.000 n=10+9)
GlyphRGBA64Over-8                 37.2µs ± 0%  36.0µs ± 0%   -3.17%          (p=0.000 n=9+8)
GlyphRGBA64Src-8                  29.0µs ± 1%  27.9µs ± 1%   -4.05%         (p=0.000 n=9+10)
GlyphRGBA128Over-8                 134µs ± 1%   131µs ± 0%   -1.85%          (p=0.000 n=9+9)
GlyphRGBA128Src-8                  100µs ± 1%    98µs ± 0%   -2.27%         (p=0.000 n=10+9)
GlyphRGBA256Over-8                 506µs ± 0%   503µs ± 0%   -0.56%         (p=0.000 n=10+8)
GlyphRGBA256Src-8                  373µs ± 0%   370µs ± 0%   -1.01%         (p=0.000 n=10+9)

Change-Id: Ie02afac6fd6fa95f090bf3fe0a5c744799ea9dc5
Reviewed-on: https://go-review.googlesource.com/31532
Reviewed-by: David Crawshaw <crawshaw@golang.org>
2016-10-20 04:12:57 +00:00
Nigel Tao
fa54d6fa1c vector: simplify fixedLineTo computation.
name                              old time/op  new time/op   delta
GlyphAlpha16Over-8                3.38µs ± 0%   3.36µs ± 0%   -0.54%        (p=0.000 n=10+10)
GlyphAlpha16Src-8                 3.28µs ± 0%   3.26µs ± 0%   -0.69%         (p=0.000 n=10+9)
GlyphAlpha32Over-8                5.23µs ± 1%   5.20µs ± 0%   -0.58%         (p=0.000 n=10+8)
GlyphAlpha32Src-8                 4.83µs ± 0%   4.81µs ± 1%   -0.46%         (p=0.001 n=10+9)
GlyphAlpha64Over-8                10.2µs ± 0%   10.2µs ± 0%   -0.21%         (p=0.003 n=10+9)
GlyphAlpha64Src-8                 8.68µs ± 1%   8.62µs ± 0%   -0.70%         (p=0.000 n=10+9)
GlyphAlpha128Over-8               24.2µs ± 1%   24.1µs ± 0%   -0.58%         (p=0.001 n=10+9)
GlyphAlpha128Src-8                18.0µs ± 1%   17.9µs ± 0%   -0.61%         (p=0.001 n=10+9)
GlyphAlpha256Over-8               70.3µs ± 0%   70.1µs ± 0%   -0.37%        (p=0.019 n=10+10)
GlyphAlpha256Src-8                45.4µs ± 0%   45.2µs ± 1%   -0.38%         (p=0.041 n=8+10)
GlyphRGBA16Over-8                 5.14µs ± 0%   5.12µs ± 0%   -0.43%         (p=0.000 n=9+10)
GlyphRGBA16Src-8                  4.59µs ± 0%   4.57µs ± 1%   -0.43%          (p=0.005 n=9+9)
GlyphRGBA32Over-8                 12.2µs ± 1%   12.1µs ± 0%   -0.70%        (p=0.000 n=10+10)
GlyphRGBA32Src-8                  10.0µs ± 0%   10.0µs ± 0%     ~            (p=0.092 n=7+10)
GlyphRGBA64Over-8                 37.5µs ± 1%   37.2µs ± 0%   -0.75%         (p=0.000 n=10+9)
GlyphRGBA64Src-8                  29.1µs ± 0%   29.0µs ± 1%     ~            (p=0.243 n=10+9)
GlyphRGBA128Over-8                 135µs ± 0%    134µs ± 1%   -0.72%          (p=0.000 n=9+9)
GlyphRGBA128Src-8                  101µs ± 1%    100µs ± 1%     ~           (p=0.197 n=10+10)
GlyphRGBA256Over-8                 511µs ± 0%    506µs ± 0%   -0.97%        (p=0.000 n=10+10)
GlyphRGBA256Src-8                  374µs ± 0%    373µs ± 0%   -0.29%        (p=0.002 n=10+10)

Change-Id: Ic05a900935cb59e55711374db1e62b055d75c8e3
Reviewed-on: https://go-review.googlesource.com/31116
Reviewed-by: David Crawshaw <crawshaw@golang.org>
2016-10-16 03:06:30 +00:00
Nigel Tao
f72412cfe3 vector: fix overflow when rasterizing a 30 degree line.
There are some obvious TODOs, but they will be follow-up commits.

This is about correctness, not performance, but for the record:

name                              old time/op  new time/op   delta
GlyphAlpha16Over-8                3.16µs ± 0%   3.38µs ± 0%   +6.96%         (p=0.000 n=9+10)
GlyphAlpha16Src-8                 3.06µs ± 0%   3.28µs ± 0%   +7.21%        (p=0.000 n=10+10)
GlyphAlpha32Over-8                4.92µs ± 1%   5.23µs ± 1%   +6.24%        (p=0.000 n=10+10)
GlyphAlpha32Src-8                 4.53µs ± 1%   4.83µs ± 0%   +6.69%        (p=0.000 n=10+10)
GlyphAlpha64Over-8                9.60µs ± 1%  10.21µs ± 0%   +6.36%        (p=0.000 n=10+10)
GlyphAlpha64Src-8                 8.04µs ± 0%   8.68µs ± 1%   +7.99%         (p=0.000 n=9+10)
GlyphAlpha128Over-8               23.1µs ± 0%   24.2µs ± 1%   +5.08%         (p=0.000 n=9+10)
GlyphAlpha128Src-8                16.8µs ± 0%   18.0µs ± 1%   +6.76%        (p=0.000 n=10+10)
GlyphAlpha256Over-8               68.6µs ± 1%   70.3µs ± 0%   +2.50%        (p=0.000 n=10+10)
GlyphAlpha256Src-8                43.6µs ± 0%   45.4µs ± 0%   +4.08%         (p=0.000 n=10+8)
GlyphRGBA16Over-8                 4.92µs ± 0%   5.14µs ± 0%   +4.48%          (p=0.000 n=9+9)
GlyphRGBA16Src-8                  4.39µs ± 0%   4.59µs ± 0%   +4.60%          (p=0.000 n=8+9)
GlyphRGBA32Over-8                 11.8µs ± 0%   12.2µs ± 1%   +2.89%         (p=0.000 n=9+10)
GlyphRGBA32Src-8                  9.79µs ± 1%  10.03µs ± 0%   +2.49%         (p=0.000 n=10+7)
GlyphRGBA64Over-8                 36.7µs ± 1%   37.5µs ± 1%   +2.23%        (p=0.000 n=10+10)
GlyphRGBA64Src-8                  28.5µs ± 0%   29.1µs ± 0%   +2.09%        (p=0.000 n=10+10)
GlyphRGBA128Over-8                 133µs ± 0%    135µs ± 0%   +1.51%         (p=0.000 n=10+9)
GlyphRGBA128Src-8                 99.1µs ± 0%  100.5µs ± 1%   +1.47%         (p=0.000 n=9+10)
GlyphRGBA256Over-8                 505µs ± 0%    511µs ± 0%   +1.18%         (p=0.000 n=9+10)
GlyphRGBA256Src-8                  372µs ± 0%    374µs ± 0%   +0.69%         (p=0.000 n=9+10)

Change-Id: Ice1d77de5bc2649f8cd88366bcae3c00e78d65c2
Reviewed-on: https://go-review.googlesource.com/31113
Reviewed-by: David Crawshaw <crawshaw@golang.org>
2016-10-14 22:25:13 +00:00
Nigel Tao
e78c45720c vector: add SIMD versions of xxxAccumulateMask.
name                              old time/op  new time/op  delta
GlyphAlphaLoose16Over-8           3.96µs ± 0%  3.64µs ± 1%   -8.08%   (p=0.000 n=8+10)
GlyphAlphaLoose16Src-8            3.64µs ± 0%  3.35µs ± 0%   -7.88%    (p=0.000 n=8+9)
GlyphAlphaLoose32Over-8           8.45µs ± 0%  6.74µs ± 0%  -20.22%    (p=0.000 n=8+9)
GlyphAlphaLoose32Src-8            7.24µs ± 0%  5.54µs ± 1%  -23.48%   (p=0.000 n=8+10)
GlyphAlphaLoose64Over-8           22.2µs ± 0%  17.4µs ± 0%  -21.67%    (p=0.000 n=9+9)
GlyphAlphaLoose64Src-8            17.6µs ± 1%  12.2µs ± 1%  -30.32%  (p=0.000 n=10+10)
GlyphAlphaLoose128Over-8          67.9µs ± 0%  53.3µs ± 1%  -21.53%  (p=0.000 n=10+10)
GlyphAlphaLoose128Src-8           48.2µs ± 0%  32.6µs ± 2%  -32.41%   (p=0.000 n=9+10)
GlyphAlphaLoose256Over-8           242µs ± 1%   187µs ± 1%  -22.96%    (p=0.000 n=9+9)
GlyphAlphaLoose256Src-8            163µs ± 0%   105µs ± 1%  -35.83%    (p=0.000 n=9+9)
GlyphRGBA16Over-8                 5.25µs ± 1%  4.95µs ± 0%   -5.78%    (p=0.000 n=9+9)
GlyphRGBA16Src-8                  4.72µs ± 0%  4.43µs ± 1%   -6.22%   (p=0.000 n=9+10)
GlyphRGBA32Over-8                 13.5µs ± 0%  11.9µs ± 1%  -12.19%   (p=0.000 n=9+10)
GlyphRGBA32Src-8                  11.5µs ± 1%   9.8µs ± 0%  -14.72%    (p=0.000 n=9+9)
GlyphRGBA64Over-8                 42.0µs ± 2%  36.9µs ± 1%  -12.19%  (p=0.000 n=10+10)
GlyphRGBA64Src-8                  34.1µs ± 1%  28.5µs ± 0%  -16.25%    (p=0.000 n=9+7)
GlyphRGBA128Over-8                 149µs ± 2%   133µs ± 1%  -10.24%   (p=0.000 n=10+9)
GlyphRGBA128Src-8                  115µs ± 1%    99µs ± 1%  -13.57%   (p=0.000 n=9+10)
GlyphRGBA256Over-8                 566µs ± 0%   511µs ± 1%   -9.85%   (p=0.000 n=9+10)
GlyphRGBA256Src-8                  435µs ± 0%   372µs ± 0%  -14.64%    (p=0.000 n=9+8)
GlyphNRGBA16Over-8                26.9µs ± 3%  26.0µs ± 3%   -3.55%   (p=0.000 n=10+9)
GlyphNRGBA16Src-8                 18.8µs ± 2%  18.4µs ± 2%   -2.21%   (p=0.000 n=9+10)
GlyphNRGBA32Over-8                99.1µs ± 2%  95.9µs ± 3%   -3.23%  (p=0.000 n=10+10)
GlyphNRGBA32Src-8                 65.6µs ± 3%  62.8µs ± 2%   -4.36%  (p=0.000 n=10+10)
GlyphNRGBA64Over-8                 376µs ± 4%   370µs ± 2%     ~     (p=0.063 n=10+10)
GlyphNRGBA64Src-8                  238µs ± 3%   233µs ± 1%   -2.21%   (p=0.000 n=9+10)
GlyphNRGBA128Over-8               1.52ms ± 2%  1.48ms ± 0%   -2.11%   (p=0.000 n=10+8)
GlyphNRGBA128Src-8                 951µs ± 3%   935µs ± 1%   -1.69%   (p=0.013 n=10+9)
GlyphNRGBA256Over-8               6.00ms ± 1%  5.87ms ± 3%   -2.12%  (p=0.002 n=10+10)
GlyphNRGBA256Src-8                3.94ms ± 2%  3.80ms ± 2%   -3.64%  (p=0.000 n=10+10)

A comparison of the non-SIMD and SIMD versions:

name                              time/op
FixedAccumulateMask16-8            237ns ± 0%
FixedAccumulateMaskSIMD16-8       80.0ns ± 1%
FloatingAccumulateMask16-8         413ns ± 2%
FloatingAccumulateMaskSIMD16-8     166ns ± 0%
FixedAccumulateMask64-8           3.42µs ± 0%
FixedAccumulateMaskSIMD64-8       1.09µs ± 0%
FloatingAccumulateMask64-8        6.92µs ± 0%
FloatingAccumulateMaskSIMD64-8    2.47µs ± 1%

Change-Id: Ib6980e5975ed2842ff2a372f76dd5f2e95c5526c
Reviewed-on: https://go-review.googlesource.com/30898
Reviewed-by: David Crawshaw <crawshaw@golang.org>
2016-10-14 04:48:47 +00:00
Nigel Tao
beb9675609 vector: fix overflow when rasterizing wide lines.
Change-Id: Iea92b74ca9533de2ef17534ee3acf4f40c3d03ef
Reviewed-on: https://go-review.googlesource.com/30899
Reviewed-by: David Crawshaw <crawshaw@golang.org>
2016-10-14 02:01:44 +00:00
Nigel Tao
8edbaf3f9e vector: add SIMD versions of xxxAccumulateOpOver.
name                             old time/op  new time/op  delta
GlyphAlpha16Over-8               3.55µs ± 0%  3.17µs ± 0%  -10.58%  (p=0.000 n=10+10)
GlyphAlpha32Over-8               6.73µs ± 1%  4.94µs ± 0%  -26.55%   (p=0.000 n=10+9)
GlyphAlpha64Over-8               16.4µs ± 0%   9.6µs ± 0%  -41.30%   (p=0.000 n=9+10)
GlyphAlpha128Over-8              47.3µs ± 0%  23.1µs ± 1%  -51.09%    (p=0.000 n=9+9)
GlyphAlpha256Over-8               159µs ± 0%    69µs ± 0%  -56.82%    (p=0.000 n=9+8)

A comparison of the non-SIMD and SIMD versions:

name                              time/op
FixedAccumulateOpOver16-8          579ns ± 0%
FixedAccumulateOpOverSIMD16-8      183ns ± 0%
FloatingAccumulateOpOver16-8       670ns ± 1%
FloatingAccumulateOpOverSIMD16-8   242ns ± 0%
FixedAccumulateOpOver64-8         9.61µs ± 0%
FixedAccumulateOpOverSIMD64-8     2.72µs ± 0%
FloatingAccumulateOpOver64-8      11.1µs ± 0%
FloatingAccumulateOpOverSIMD64-8  3.65µs ± 0%

Change-Id: I08273c40ac5445f39b77a88fb8b6b07fd3e5f84b
Reviewed-on: https://go-review.googlesource.com/30831
Reviewed-by: David Crawshaw <crawshaw@golang.org>
2016-10-13 03:04:36 +00:00
Nigel Tao
ae7296db94 vector: generate acc_amd64.s from a template.
There are currently 2 instantiations: fixed and floating math.

There will eventually be 6: 2 math types times 3 operators (OpOver,
OpSrc and Mask).

Change-Id: I16db8f3693abe30027ea2f822909980403a6142f
Reviewed-on: https://go-review.googlesource.com/30735
Reviewed-by: David Crawshaw <crawshaw@golang.org>
2016-10-11 07:55:43 +00:00
Nigel Tao
746988e7a2 vector: add SIMD versions of xxxAccumulateOpSrc.
name                         old time/op  new time/op  delta
GlyphAlpha16Src-8            3.37µs ± 0%  3.07µs ± 1%   -8.86%    (p=0.000 n=9+9)
GlyphAlpha32Src-8            6.01µs ± 1%  4.55µs ± 0%  -24.28%   (p=0.000 n=10+9)
GlyphAlpha64Src-8            13.2µs ± 0%   8.1µs ± 0%  -38.69%   (p=0.000 n=10+9)
GlyphAlpha128Src-8           32.9µs ± 0%  16.9µs ± 0%  -48.85%   (p=0.000 n=10+9)
GlyphAlpha256Src-8           98.0µs ± 0%  43.6µs ± 1%  -55.50%  (p=0.000 n=10+10)

A comparison of the non-SIMD and SIMD versions:

name                             time/op
FixedAccumulateOpSrc16-8          368ns ± 0%
FixedAccumulateOpSrcSIMD16-8     86.8ns ± 1%
FloatingAccumulateOpSrc16-8       434ns ± 0%
FloatingAccumulateOpSrcSIMD16-8   119ns ± 0%
FixedAccumulateOpSrc64-8         6.12µs ± 0%
FixedAccumulateOpSrcSIMD64-8     1.17µs ± 0%
FloatingAccumulateOpSrc64-8      7.15µs ± 0%
FloatingAccumulateOpSrcSIMD64-8  1.68µs ± 1%

Change-Id: I58e5c7a3ecd12e536aab8e765e94275453d0eac8
Reviewed-on: https://go-review.googlesource.com/30431
Reviewed-by: David Crawshaw <crawshaw@golang.org>
2016-10-10 07:32:32 +00:00
Nigel Tao
dc590effac vector: add some accumulate tests.
Change-Id: I199e1637698572e36b4a3bb83d652baf0d76717f
Reviewed-on: https://go-review.googlesource.com/30110
Reviewed-by: David Crawshaw <crawshaw@golang.org>
2016-10-05 23:42:26 +00:00
Nigel Tao
72141d56a2 vector: re-order some functions.
There are no code changes, just a re-ordering so that these files are
consistent with others in this package: OpOver, OpSrc, Mask.

Change-Id: Ib1d46a8e912dae0c760af655e919b77023688189
Reviewed-on: https://go-review.googlesource.com/30111
Reviewed-by: David Crawshaw <crawshaw@golang.org>
2016-10-02 02:10:30 +00:00
Nigel Tao
992afa5d48 vector: add a fixed point math implementation.
name                      old time/op  new time/op  delta

GlyphAlpha16Over-8        4.48µs ± 1%  3.56µs ± 0%  -20.70%   (p=0.000 n=9+10)
GlyphAlpha16Src-8         4.17µs ± 0%  3.38µs ± 1%  -19.09%  (p=0.000 n=10+10)
GlyphAlpha32Over-8        9.03µs ± 0%  6.74µs ± 0%  -25.33%   (p=0.000 n=9+10)
GlyphAlpha32Src-8         7.46µs ± 1%  5.98µs ± 0%  -19.80%   (p=0.000 n=10+9)
GlyphAlpha64Over-8        21.3µs ± 0%  16.4µs ± 0%  -22.84%  (p=0.000 n=10+10)
GlyphAlpha64Src-8         16.2µs ± 1%  13.1µs ± 0%  -19.33%  (p=0.000 n=10+10)
GlyphAlpha128Over-8       59.8µs ± 0%  47.2µs ± 0%  -21.11%    (p=0.000 n=9+9)
GlyphAlpha128Src-8        41.3µs ± 1%  33.0µs ± 0%  -20.26%   (p=0.000 n=9+10)
GlyphAlpha256Over-8        197µs ± 0%   158µs ± 0%  -19.44%   (p=0.000 n=9+10)
GlyphAlpha256Src-8         124µs ± 0%    98µs ± 0%  -21.17%    (p=0.000 n=9+9)

GlyphAlphaLoose16Over-8   4.73µs ± 0%  3.97µs ± 1%  -16.06%  (p=0.000 n=10+10)
GlyphAlphaLoose16Src-8    4.41µs ± 0%  3.64µs ± 1%  -17.50%  (p=0.000 n=10+10)
GlyphAlphaLoose32Over-8   9.62µs ± 0%  8.47µs ± 0%  -11.95%  (p=0.000 n=10+10)
GlyphAlphaLoose32Src-8    8.25µs ± 0%  7.19µs ± 0%  -12.88%    (p=0.000 n=9+9)
GlyphAlphaLoose64Over-8   25.6µs ± 0%  22.2µs ± 0%  -13.01%    (p=0.000 n=9+9)
GlyphAlphaLoose64Src-8    20.2µs ± 0%  17.2µs ± 1%  -14.98%  (p=0.000 n=10+10)
GlyphAlphaLoose128Over-8  83.4µs ± 1%  68.2µs ± 0%  -18.27%  (p=0.000 n=10+10)
GlyphAlphaLoose128Src-8   59.8µs ± 0%  47.4µs ± 0%  -20.77%   (p=0.000 n=10+9)
GlyphAlphaLoose256Over-8   273µs ± 1%   239µs ± 0%  -12.52%   (p=0.000 n=10+9)
GlyphAlphaLoose256Src-8    187µs ± 0%   155µs ± 1%  -16.91%   (p=0.000 n=9+10)

GlyphRGBA16Over-8         5.99µs ± 0%  5.24µs ± 1%  -12.60%   (p=0.000 n=9+10)
GlyphRGBA16Src-8          5.48µs ± 0%  4.68µs ± 0%  -14.68%   (p=0.000 n=9+10)
GlyphRGBA32Over-8         14.6µs ± 0%  13.5µs ± 0%   -7.60%    (p=0.000 n=9+9)
GlyphRGBA32Src-8          12.6µs ± 0%  11.4µs ± 0%   -9.62%    (p=0.000 n=9+9)
GlyphRGBA64Over-8         44.8µs ± 0%  42.2µs ± 0%   -5.69%    (p=0.000 n=9+9)
GlyphRGBA64Src-8          36.6µs ± 1%  33.5µs ± 1%   -8.55%    (p=0.000 n=9+9)
GlyphRGBA128Over-8         162µs ± 0%   148µs ± 1%   -8.85%   (p=0.000 n=10+9)
GlyphRGBA128Src-8          129µs ± 1%   114µs ± 0%  -11.61%   (p=0.000 n=9+10)
GlyphRGBA256Over-8         588µs ± 0%   573µs ± 0%   -2.53%   (p=0.000 n=9+10)
GlyphRGBA256Src-8          455µs ± 0%   426µs ± 1%   -6.51%   (p=0.000 n=9+10)

GlyphNRGBA16Over-8        27.0µs ± 4%  26.3µs ± 2%   -2.65%   (p=0.001 n=9+10)
GlyphNRGBA16Src-8         19.4µs ± 3%  18.6µs ± 1%   -4.35%   (p=0.000 n=9+10)
GlyphNRGBA32Over-8        97.4µs ± 3%  96.8µs ± 2%     ~      (p=0.447 n=9+10)
GlyphNRGBA32Src-8         66.6µs ± 3%  64.5µs ± 1%   -3.21%   (p=0.000 n=10+9)
GlyphNRGBA64Over-8         372µs ± 3%   368µs ± 1%     ~     (p=0.105 n=10+10)
GlyphNRGBA64Src-8          235µs ± 1%   234µs ± 1%     ~       (p=0.130 n=8+8)
GlyphNRGBA128Over-8       1.45ms ± 2%  1.48ms ± 3%   +2.06%    (p=0.014 n=9+9)
GlyphNRGBA128Src-8         926µs ± 3%   937µs ± 1%     ~      (p=0.113 n=10+9)
GlyphNRGBA256Over-8       5.76ms ± 2%  5.90ms ± 3%   +2.29%   (p=0.001 n=9+10)
GlyphNRGBA256Src-8        3.59ms ± 1%  3.86ms ± 1%   +7.46%   (p=0.000 n=9+10)

Change-Id: I72f25193b5be4e57af09e9eea4eee50545a34cbf
Reviewed-on: https://go-review.googlesource.com/29972
Reviewed-by: David Crawshaw <crawshaw@golang.org>
2016-09-30 06:47:22 +00:00
Nigel Tao
8a20735b0e vector: sort benchmarks from slowest to fastest.
Change-Id: Ia6047f225c7ba4407fe831a49f90d4c74d203651
Reviewed-on: https://go-review.googlesource.com/29973
Reviewed-by: David Crawshaw <crawshaw@golang.org>
2016-09-30 06:02:01 +00:00
Nigel Tao
14bbbc8017 vector: add a fast path for RGBA dst images.
name                      old time/op  new time/op  delta
GlyphRGBA16Over-8         25.1µs ± 2%   6.0µs ± 0%  -76.09%  (p=0.000 n=10+10)
GlyphRGBA16Src-8          17.3µs ± 2%   5.4µs ± 0%  -68.68%  (p=0.000 n=10+10)
GlyphRGBA32Over-8         93.2µs ± 2%  14.5µs ± 0%  -84.41%  (p=0.000 n=10+10)
GlyphRGBA32Src-8          59.3µs ± 2%  12.4µs ± 0%  -79.04%    (p=0.000 n=9+9)
GlyphRGBA64Over-8          350µs ± 2%    45µs ± 0%  -87.21%    (p=0.000 n=9+9)
GlyphRGBA64Src-8           223µs ± 2%    37µs ± 0%  -83.58%    (p=0.000 n=9+9)
GlyphRGBA128Over-8        1.37ms ± 2%  0.16ms ± 1%  -88.19%   (p=0.000 n=9+10)
GlyphRGBA128Src-8          868µs ± 2%   128µs ± 0%  -85.21%   (p=0.000 n=10+9)
GlyphRGBA256Over-8        5.50ms ± 3%  0.59ms ± 0%  -89.31%  (p=0.000 n=10+10)
GlyphRGBA256Src-8         3.47ms ± 2%  0.45ms ± 0%  -86.91%   (p=0.000 n=9+10)

Change-Id: I3df60e6b7147872367715361c9d1ed52951b22e0
Reviewed-on: https://go-review.googlesource.com/29699
Reviewed-by: David Crawshaw <crawshaw@golang.org>
2016-09-27 04:00:57 +00:00
Nigel Tao
0080ac3d2c vector: add some more tests.
These tests are copied from the github.com/google/font-go prototype.

Change-Id: I4523193bd3453974633cbef0576c1203eb013a7d
Reviewed-on: https://go-review.googlesource.com/29697
Reviewed-by: David Crawshaw <crawshaw@golang.org>
2016-09-25 03:45:43 +00:00
Nigel Tao
7ea36498ac vector: allow dst and mask to have different strides.
Change-Id: If244dba2c681b11122a7a26b352f9411135f123f
Reviewed-on: https://go-review.googlesource.com/29696
Reviewed-by: David Crawshaw <crawshaw@golang.org>
2016-09-25 02:48:45 +00:00
Nigel Tao
b077ed42b3 vector: implement arbitrary dst and src image types.
Change-Id: Ib0d0961c69cabc3bb01d7b1d244796f0f713b819
Reviewed-on: https://go-review.googlesource.com/29691
Reviewed-by: David Crawshaw <crawshaw@golang.org>
2016-09-24 05:40:50 +00:00
Nigel Tao
50ce1ebf58 vector: add benchmarks.
Change-Id: I4946ccc3a915c7e9365cd5fe860da673a2270fbb
Reviewed-on: https://go-review.googlesource.com/29631
Reviewed-by: David Crawshaw <crawshaw@golang.org>
2016-09-23 02:06:24 +00:00
Nigel Tao
cd8486aac9 vector: implement DrawOp == draw.Over.
Some "TODO: the general case" lines remain.

Change-Id: If66e484a00d5ea3fce9db37d4ee493739648daa3
Reviewed-on: https://go-review.googlesource.com/29495
Reviewed-by: David Crawshaw <crawshaw@golang.org>
2016-09-23 01:58:13 +00:00
Nigel Tao
bb355ba442 vector: add CubeTo.
Change-Id: Ia64b5b077172303981538de99cc14c5bcc99a3e4
Reviewed-on: https://go-review.googlesource.com/28489
Reviewed-by: David Crawshaw <crawshaw@golang.org>
2016-09-08 11:11:11 +00:00
Nigel Tao
714f2e47f7 vector: new package for rasterizing 2-D graphics.
Updates golang/go#16904

Change-Id: I4e11e4e859c007c3444655a227ac935c27f3f784
Reviewed-on: https://go-review.googlesource.com/28347
Reviewed-by: David Crawshaw <crawshaw@golang.org>
2016-09-04 08:48:55 +00:00