Commit Graph
1406 Commits
Author SHA1 Message Date
Nigel Tao 4ff64de78e Mark deflate's decode_huffman_fast as !? not !?? 2018-11-24 20:34:31 +11:00
Nigel Tao ec0fd9649d Don't skip derived I/O vars for !? methods 2018-11-24 17:52:04 +11:00
Nigel Tao cf51406fcf Mark deflate's init_fixed_huffman as !? not !?? 2018-11-24 16:49:52 +11:00
Nigel Tao 312cdc8a0a Mark deflate's init_huff as merely !? not !?? 2018-11-24 16:41:45 +11:00
Nigel Tao 88236105c9 Have cgen functions return empty_struct, not void 2018-11-24 16:00:46 +11:00
Nigel Tao 0482fadf41 Generate implicit wuffs_base__status for !? funcs 2018-11-24 15:41:56 +11:00
Nigel Tao 2434f0dee5 Remove some redundant cgen funk fields 2018-11-24 15:25:31 +11:00
Nigel Tao 5a43e931f9 Have cgen convert "ok" to "NULL" 2018-11-24 15:20:44 +11:00
Nigel Tao fcffb9676e Rename func header/footer as prologue/epilogue 2018-11-24 15:15:25 +11:00
Nigel Tao 762b320b5e Make the return value mandatory; add ok 2018-11-24 15:09:28 +11:00
Nigel Tao 24ace1c144 Type and bounds check the "x" in "return x" 2018-11-24 14:46:52 +11:00
Nigel Tao c84732f3d3 Quote built-in statuses 2018-11-24 13:42:18 +11:00
Nigel Tao 9e5aec5d25 Change "$foo" to base."$foo" 2018-11-24 13:36:09 +11:00
Nigel Tao 875362ea07 Change status "?foo" to just "?foo" 2018-11-24 12:36:42 +11:00
Nigel Tao 5f9d007b1c Make a struct's implicit reset method impure 2018-11-24 11:05:39 +11:00
Nigel Tao 8ab2a33aae Update GIF frame_dirty_rect more frequently
It's not an optimization per se, but for the record:

name                                    old speed     new speed     delta

wuffs_gif_decode_1k_bw/clang            260MB/s ± 1%  261MB/s ± 0%  +0.51%  (p=0.032 n=5+5)
wuffs_gif_decode_1k_color/clang         122MB/s ± 0%  123MB/s ± 1%    ~     (p=0.222 n=5+5)
wuffs_gif_decode_10k/clang              148MB/s ± 1%  148MB/s ± 0%    ~     (p=0.421 n=5+5)
wuffs_gif_decode_100k_artificial/clang  439MB/s ± 0%  443MB/s ± 1%  +0.99%  (p=0.008 n=5+5)
wuffs_gif_decode_100k_realistic/clang   174MB/s ± 1%  173MB/s ± 1%    ~     (p=0.151 n=5+5)
wuffs_gif_decode_1000k/clang            175MB/s ± 0%  175MB/s ± 0%  -0.30%  (p=0.016 n=4+5)
wuffs_gif_decode_anim_screencap/clang   941MB/s ± 1%  948MB/s ± 0%  +0.74%  (p=0.016 n=5+5)

wuffs_gif_decode_1k_bw/gcc              266MB/s ± 0%  273MB/s ± 0%  +2.60%  (p=0.008 n=5+5)
wuffs_gif_decode_1k_color/gcc           137MB/s ± 0%  139MB/s ± 0%  +1.54%  (p=0.008 n=5+5)
wuffs_gif_decode_10k/gcc                163MB/s ± 0%  162MB/s ± 0%    ~     (p=0.151 n=5+5)
wuffs_gif_decode_100k_artificial/gcc    457MB/s ± 0%  456MB/s ± 0%    ~     (p=0.095 n=5+5)
wuffs_gif_decode_100k_realistic/gcc     184MB/s ± 0%  182MB/s ± 0%  -1.02%  (p=0.008 n=5+5)
wuffs_gif_decode_1000k/gcc              186MB/s ± 0%  184MB/s ± 0%  -1.04%  (p=0.016 n=5+4)
wuffs_gif_decode_anim_screencap/gcc     985MB/s ± 1%  980MB/s ± 0%    ~     (p=0.095 n=5+5)
2018-11-24 10:17:16 +11:00
Nigel Tao e84a716313 Add comment to consider status.is_complete 2018-11-24 10:10:39 +11:00
Nigel Tao c598e119e2 Add a gif.decoder.frame_dirty_rect test 2018-11-17 22:13:22 +11:00
Nigel Tao fc1002cb89 Implement gif.decoder.workbuf_len
Also relax decode_frame's workbuf length check from not-equals to
less-than.
2018-11-17 21:52:26 +11:00
Nigel Tao 4023294ba2 Add a gif.decoder.frame_dirty_rect method 2018-11-17 21:26:53 +11:00
Nigel Tao f06126c85a Rename color_u32argb; add YCoCg color 2018-11-17 11:43:34 +11:00
Nigel Tao e3486eec17 Remove unused gif.decoder.skip_frame variable 2018-11-17 10:58:39 +11:00
Nigel Tao c7bb8c3e09 Add commentary about local variable resumability 2018-11-17 09:34:21 +11:00
Nigel Tao 03b046d774 Add script/print-crc32-example.go 2018-11-17 08:29:48 +11:00
Nigel Tao 34fe051cac Add commentary about the CRC-32 implementation 2018-11-11 10:25:19 +11:00
Nigel Tao 7bdbab9e08 Remove obsolete gen.visitVars function 2018-11-10 17:24:11 +11:00
Nigel Tao 6ebe99ced9 Don't suspend/resume unnecessary local variables
name                                    old speed      new speed      delta

wuffs_adler32_10k/clang                 2.42GB/s ± 1%  2.41GB/s ± 0%  -0.43%  (p=0.000 n=49+10)
wuffs_adler32_100k/clang                2.42GB/s ± 1%  2.41GB/s ± 1%  -0.56%  (p=0.000 n=48+10)

wuffs_adler32_10k/gcc                   3.22GB/s ± 1%  3.21GB/s ± 0%  -0.35%  (p=0.000 n=48+10)
wuffs_adler32_100k/gcc                  3.22GB/s ± 0%  3.20GB/s ± 0%  -0.69%  (p=0.000 n=48+10)

wuffs_crc32_ieee_10k/clang              2.86GB/s ± 0%  2.85GB/s ± 0%  -0.44%  (p=0.000 n=48+9)
wuffs_crc32_ieee_100k/clang             2.88GB/s ± 1%  2.87GB/s ± 0%  -0.45%  (p=0.000 n=49+10)

wuffs_crc32_ieee_10k/gcc                3.35GB/s ± 1%  3.35GB/s ± 0%    ~     (p=0.094 n=48+9)
wuffs_crc32_ieee_100k/gcc               3.38GB/s ± 1%  3.38GB/s ± 1%    ~     (p=0.371 n=45+10)

wuffs_deflate_decode_1k/clang            147MB/s ± 1%   137MB/s ± 1%  -6.84%  (p=0.000 n=50+10)
wuffs_deflate_decode_10k/clang           241MB/s ± 0%   227MB/s ± 0%  -5.74%  (p=0.000 n=48+9)
wuffs_deflate_decode_100k/clang          269MB/s ± 0%   258MB/s ± 0%  -3.92%  (p=0.000 n=49+9)

wuffs_deflate_decode_1k/gcc              159MB/s ± 0%   161MB/s ± 1%  +1.51%  (p=0.000 n=49+10)
wuffs_deflate_decode_10k/gcc             269MB/s ± 0%   276MB/s ± 0%  +2.51%  (p=0.000 n=49+10)
wuffs_deflate_decode_100k/gcc            320MB/s ± 0%   316MB/s ± 0%  -1.29%  (p=0.000 n=46+10)

wuffs_gif_decode_1k_bw/clang             252MB/s ± 1%   264MB/s ± 0%  +4.56%  (p=0.000 n=49+10)
wuffs_gif_decode_1k_color/clang          127MB/s ± 1%   122MB/s ± 1%  -3.84%  (p=0.000 n=47+10)
wuffs_gif_decode_10k/clang               152MB/s ± 0%   148MB/s ± 0%  -3.12%  (p=0.000 n=47+9)
wuffs_gif_decode_100k_artificial/clang   431MB/s ± 0%   443MB/s ± 1%  +2.95%  (p=0.000 n=42+10)
wuffs_gif_decode_100k_realistic/clang    170MB/s ± 0%   173MB/s ± 0%  +1.39%  (p=0.000 n=48+10)
wuffs_gif_decode_1000k/clang             172MB/s ± 0%   174MB/s ± 0%  +1.27%  (p=0.000 n=48+10)
wuffs_gif_decode_anim_screencap/clang    917MB/s ± 1%   949MB/s ± 0%  +3.44%  (p=0.000 n=47+9)

wuffs_gif_decode_1k_bw/gcc               265MB/s ± 1%   278MB/s ± 2%  +4.87%  (p=0.000 n=48+10)
wuffs_gif_decode_1k_color/gcc            135MB/s ± 0%   140MB/s ± 2%  +4.10%  (p=0.000 n=46+10)
wuffs_gif_decode_10k/gcc                 157MB/s ± 1%   166MB/s ± 0%  +5.84%  (p=0.000 n=48+8)
wuffs_gif_decode_100k_artificial/gcc     446MB/s ± 1%   459MB/s ± 1%  +2.89%  (p=0.000 n=48+10)
wuffs_gif_decode_100k_realistic/gcc      176MB/s ± 0%   186MB/s ± 2%  +5.82%  (p=0.000 n=46+10)
wuffs_gif_decode_1000k/gcc               178MB/s ± 0%   189MB/s ± 1%  +6.11%  (p=0.000 n=45+9)
wuffs_gif_decode_anim_screencap/gcc      962MB/s ± 1%   989MB/s ± 1%  +2.83%  (p=0.000 n=48+9)

wuffs_gzip_decode_10k/clang              212MB/s ± 0%   208MB/s ± 0%  -1.67%  (p=0.000 n=47+10)
wuffs_gzip_decode_100k/clang             240MB/s ± 0%   236MB/s ± 0%  -1.68%  (p=0.000 n=47+10)

wuffs_gzip_decode_10k/gcc                252MB/s ± 0%   246MB/s ± 0%  -2.41%  (p=0.000 n=47+10)
wuffs_gzip_decode_100k/gcc               290MB/s ± 0%   288MB/s ± 0%  -0.70%  (p=0.000 n=44+10)

wuffs_lzw_decode_20k/clang               202MB/s ± 0%   202MB/s ± 0%    ~     (p=0.069 n=48+8)
wuffs_lzw_decode_100k/clang              332MB/s ± 1%   334MB/s ± 1%  +0.54%  (p=0.000 n=46+10)

wuffs_lzw_decode_20k/gcc                 215MB/s ± 1%   223MB/s ± 0%  +3.60%  (p=0.000 n=49+9)
wuffs_lzw_decode_100k/gcc                361MB/s ± 1%   368MB/s ± 0%  +2.05%  (p=0.000 n=47+10)

wuffs_zlib_decode_10k/clang              218MB/s ± 0%   206MB/s ± 0%  -5.53%  (p=0.000 n=48+9)
wuffs_zlib_decode_100k/clang             244MB/s ± 0%   234MB/s ± 0%  -4.14%  (p=0.000 n=45+10)

wuffs_zlib_decode_10k/gcc                252MB/s ± 0%   252MB/s ± 0%    ~     (p=0.960 n=49+10)
wuffs_zlib_decode_100k/gcc               290MB/s ± 0%   292MB/s ± 0%  +0.52%  (p=0.000 n=49+10)
2018-11-10 17:04:40 +11:00
Nigel Tao 808d5c1e6c Re-order switch cases for consistency 2018-11-10 11:25:51 +11:00
Nigel Tao 48c7d1c030 Add script/crawl.go 2018-11-04 20:38:04 +11:00
Nigel Tao 8eb9781a9d Have gen/c code use relative includes
For example, gen/c/wuffs-std-gif.h now #include's "./wuffs-std-lzw.h"
instead of "std/lzw.h", which removes the need to configure an -I
directory when compiling it.
2018-11-03 11:03:11 +11:00
Nigel Tao 469ab1c9c1 Fix a 'wuffs gen' output formatting typo
The fix was just re-running 'wuffs gen'. I'm not sure why it was
mis-formatted in the first place.
2018-11-03 08:50:57 +11:00
Nigel Tao 3854ce3e89 Let test binaries run from any dir in the repo
Also spin out generate.WuffsRoot into its own package.
2018-11-03 08:46:31 +11:00
Nigel Tao e2d1f8e0c2 Generate const annotations for pure methods 2018-10-28 17:30:48 +11:00
Nigel Tao 92ecd0a901 Use a table, not a formula, to get low bits mask
It's not an optimization per se. It addresses the fact that, in the C
programming language, if u is an uint32_t value, then (u<<32) is
undefined behavior, so ((1<<32)-1) isn't necessarily 0xFFFFFFFF:

https://stackoverflow.com/questions/7401888/why-doesnt-left-bit-shift-for-32-bit-integers-work-as-expected-when-used

name                                    old speed      new speed      delta

wuffs_deflate_decode_1k/clang            147MB/s ± 1%   147MB/s ± 0%     ~     (p=0.206 n=49+5)
wuffs_deflate_decode_10k/clang           234MB/s ± 1%   239MB/s ± 0%   +2.24%  (p=0.000 n=49+5)
wuffs_deflate_decode_100k/clang          261MB/s ± 0%   270MB/s ± 0%   +3.38%  (p=0.000 n=48+5)

wuffs_deflate_decode_1k/gcc              160MB/s ± 1%   160MB/s ± 0%     ~     (p=0.322 n=43+5)
wuffs_deflate_decode_10k/gcc             270MB/s ± 1%   272MB/s ± 0%   +0.93%  (p=0.000 n=46+5)
wuffs_deflate_decode_100k/gcc            313MB/s ± 1%   322MB/s ± 0%   +2.97%  (p=0.000 n=42+5)

wuffs_gif_decode_1k_bw/clang             253MB/s ± 1%   259MB/s ± 1%   +2.28%  (p=0.000 n=45+5)
wuffs_gif_decode_1k_color/clang          116MB/s ± 1%   128MB/s ± 0%  +11.20%  (p=0.000 n=45+5)
wuffs_gif_decode_10k/clang               135MB/s ± 1%   154MB/s ± 1%  +14.39%  (p=0.000 n=44+5)
wuffs_gif_decode_100k_artificial/clang   429MB/s ± 1%   426MB/s ± 0%   -0.70%  (p=0.000 n=48+5)
wuffs_gif_decode_100k_realistic/clang    162MB/s ± 0%   171MB/s ± 0%   +6.13%  (p=0.000 n=43+5)
wuffs_gif_decode_1000k/clang             163MB/s ± 1%   172MB/s ± 0%   +6.04%  (p=0.000 n=50+5)
wuffs_gif_decode_anim_screencap/clang    900MB/s ± 0%   909MB/s ± 0%   +1.06%  (p=0.000 n=45+5)

wuffs_gif_decode_1k_bw/gcc               268MB/s ± 1%   269MB/s ± 0%   +0.63%  (p=0.000 n=49+5)
wuffs_gif_decode_1k_color/gcc            133MB/s ± 1%   135MB/s ± 1%   +1.45%  (p=0.000 n=50+5)
wuffs_gif_decode_10k/gcc                 160MB/s ± 1%   158MB/s ± 0%   -1.24%  (p=0.000 n=48+5)
wuffs_gif_decode_100k_artificial/gcc     448MB/s ± 1%   446MB/s ± 0%   -0.40%  (p=0.039 n=50+5)
wuffs_gif_decode_100k_realistic/gcc      180MB/s ± 1%   177MB/s ± 0%   -1.48%  (p=0.000 n=50+5)
wuffs_gif_decode_1000k/gcc               182MB/s ± 1%   179MB/s ± 0%   -1.39%  (p=0.000 n=48+5)
wuffs_gif_decode_anim_screencap/gcc      963MB/s ± 1%   960MB/s ± 0%   -0.33%  (p=0.022 n=47+5)

wuffs_gzip_decode_10k/clang              215MB/s ± 1%   221MB/s ± 0%   +2.46%  (p=0.000 n=49+5)
wuffs_gzip_decode_100k/clang             239MB/s ± 1%   246MB/s ± 0%   +3.03%  (p=0.000 n=47+5)

wuffs_gzip_decode_10k/gcc                248MB/s ± 0%   248MB/s ± 0%     ~     (p=0.469 n=45+5)
wuffs_gzip_decode_100k/gcc               290MB/s ± 1%   291MB/s ± 0%   +0.34%  (p=0.007 n=49+5)

wuffs_lzw_decode_20k/clang               187MB/s ± 0%   199MB/s ± 2%   +6.45%  (p=0.000 n=49+5)
wuffs_lzw_decode_100k/clang              304MB/s ± 1%   329MB/s ± 1%   +8.44%  (p=0.000 n=46+5)

wuffs_lzw_decode_20k/gcc                 223MB/s ± 1%   208MB/s ± 2%   -6.78%  (p=0.000 n=46+5)
wuffs_lzw_decode_100k/gcc                364MB/s ± 1%   352MB/s ± 0%   -3.39%  (p=0.000 n=49+5)

wuffs_zlib_decode_10k/clang              214MB/s ± 0%   216MB/s ± 1%   +0.69%  (p=0.026 n=49+5)
wuffs_zlib_decode_100k/clang             237MB/s ± 0%   242MB/s ± 0%   +2.37%  (p=0.000 n=48+5)

wuffs_zlib_decode_10k/gcc                247MB/s ± 1%   245MB/s ± 1%   -0.69%  (p=0.001 n=44+5)
wuffs_zlib_decode_100k/gcc               288MB/s ± 1%   288MB/s ± 1%     ~     (p=0.629 n=47+5)
2018-10-28 17:14:49 +11:00
Nigel Tao 717d98441e Re-order the release/c code
Have the order of implementation code match the order of header code.
2018-10-27 12:57:00 +11:00
Nigel Tao 61d03537c2 De-duplicate the gen/c/* files
The gen/c/* files are not checked in, due to the .gitignore file

In Wuffs version 0.1, they were checked in, and programs were expected
to #include them directly. Another goal was for them to be
self-contained, without any #include lines to other Wuffs files, so that
they would be easy to build, without having to configure an -I dir to a
C compiler.

They therefore inlined what would otherwise be an #include. For example,
the generated std/gif.h code would inline the contents of the std/lzw.h
file. The genrelease program would have to de-duplicate these inlines in
order to produce a clean, monolithic release/c/* file.

In Wuffs version 0.2 (the latest version), programs don't #include
gen/c/*. Instead, they #include the monolithic release/c/*. Avoiding the
need to configure an -I directory isn't as important as it used to be.

This change removes that inlining, using the simpler #include line, and
the genrelease program is then simpler because it doesn't need to
de-duplicate.
2018-10-27 12:50:01 +11:00
Nigel Tao 51a0b347ed Add WUFFS_BASE__LOW_BITS_MASK__UXX macros 2018-10-21 21:53:44 +11:00
Nigel Tao 94c054a341 Remove unused byte-swapping functions 2018-10-21 21:11:34 +11:00
Nigel Tao c3ef038c79 Fix missing operator for ast.Expr.Str 2018-10-21 18:12:26 +11:00
Nigel Tao f60723f624 Simplify cgen output for copy_from_slice 2018-10-21 18:03:54 +11:00
Nigel Tao b5c47e273f wuffs gen -version=0.2.0-alpha.22 2018-10-21 16:12:50 +11:00
Nigel Tao 479326cbad Rename a LZW decoder field 2018-10-21 16:07:36 +11:00
Nigel Tao b0c72867e6 Update std/lzw/README.md for algorithmic detail 2018-10-21 16:03:47 +11:00
Nigel Tao 3338d6d1bc Write LZW output 8 bytes at a time, not 1
The benchmarks are mixed, but I think it's still a net win.

----
name                                    old speed     new speed     delta

wuffs_gif_decode_1k_bw/clang            258MB/s ± 0%  253MB/s ± 1%    -2.04%  (p=0.008 n=5+5)
wuffs_gif_decode_1k_color/clang         159MB/s ± 0%  116MB/s ± 0%   -26.76%  (p=0.008 n=5+5)
wuffs_gif_decode_10k/clang              167MB/s ± 0%  138MB/s ± 0%   -17.42%  (p=0.008 n=5+5)
wuffs_gif_decode_100k_artificial/clang  205MB/s ± 0%  439MB/s ± 0%  +114.53%  (p=0.008 n=5+5)
wuffs_gif_decode_100k_realistic/clang   151MB/s ± 0%  164MB/s ± 0%    +8.41%  (p=0.008 n=5+5)
wuffs_gif_decode_1000k/clang            155MB/s ± 0%  165MB/s ± 0%    +6.50%  (p=0.008 n=5+5)
wuffs_gif_decode_anim_screencap/clang   255MB/s ± 0%  921MB/s ± 0%  +261.50%  (p=0.008 n=5+5)

wuffs_gif_decode_1k_bw/gcc              316MB/s ± 2%  267MB/s ± 1%   -15.78%  (p=0.008 n=5+5)
wuffs_gif_decode_1k_color/gcc           184MB/s ± 1%  134MB/s ± 0%   -27.57%  (p=0.008 n=5+5)
wuffs_gif_decode_10k/gcc                196MB/s ± 0%  161MB/s ± 0%   -17.50%  (p=0.008 n=5+5)
wuffs_gif_decode_100k_artificial/gcc    287MB/s ± 0%  449MB/s ± 1%   +56.22%  (p=0.008 n=5+5)
wuffs_gif_decode_100k_realistic/gcc     182MB/s ± 0%  182MB/s ± 1%      ~     (p=0.690 n=5+5)
wuffs_gif_decode_1000k/gcc              187MB/s ± 1%  185MB/s ± 1%      ~     (p=0.056 n=5+5)
wuffs_gif_decode_anim_screencap/gcc     397MB/s ± 0%  973MB/s ± 0%  +145.42%  (p=0.008 n=5+5)

wuffs_lzw_decode_20k/clang              170MB/s ± 0%  191MB/s ± 1%   +12.52%  (p=0.008 n=5+5)
wuffs_lzw_decode_100k/clang             242MB/s ± 0%  312MB/s ± 2%   +29.00%  (p=0.008 n=5+5)

wuffs_lzw_decode_20k/gcc                220MB/s ± 0%  227MB/s ± 0%    +3.31%  (p=0.008 n=5+5)
wuffs_lzw_decode_100k/gcc               317MB/s ± 0%  369MB/s ± 1%   +16.53%  (p=0.008 n=5+5)
----

LZW is the compression method used by GIF images. Roughly speaking, LZW
is a sequence of codes, and each code emits either 1 byte (a literal
code) or multiple bytes (a copy code). The speed gain in this commit
comes from, when emitting an N-byte copy code, this now requires N/8
iterations instead of N iterations, as each iteration now writes 8 bytes
instead of 1. Literal codes, both before and after this commit, requires
a single iteration.

Decoding should therefore get faster or slower depending on whether it's
copy-code-heavy or literal-code-heavy respectively. For what it's worth,
copy-code-heavy GIF images also usually compress better.

Running script/extract-giflzw.go (or examining the cached output at
test/data/analysis/giflzw-emission-lengths.txt) on the anim_screencap
(which got faster) and 1k_color (which got slower) cases (also known as
gifplayer-muybridge.gif and hippopotamus.regular.gif) shows that the
proportion of literal codes agrees with that explanation.

----
gifplayer-muybridge.gif (all frames)
 Percent   Count  Emission_Length_Bucket
   12.78   34658           1 (Literal Code)
   14.05   38106           2
   13.25   35937           3
   10.59   28707           4
    7.85   21286           5
    6.08   16481           6
    4.85   13156           7
    3.60    9764           8
    9.34   25318    9 ..  16
    4.89   13257   17 ..  32
    4.78   12954   33 ..  64
    4.72   12806   65 .. 128
    3.02    8190  127 .. 256
    0.20     550  256+

hippopotamus.regular.gif (all frames)
 Percent   Count  Emission_Length_Bucket
   82.66     672           1 (Literal Code)
   13.04     106           2
    2.83      23           3
    0.86       7           4
    0.37       3           5
    0.25       2           6
    0.00       0           7
    0.00       0           8
    0.00       0    9 ..  16
    0.00       0   17 ..  32
    0.00       0   33 ..  64
    0.00       0   65 .. 128
    0.00       0  127 .. 256
    0.00       0  256+
----

The new algorithm also adds more CPU and RAM overhead, regardless of
whether the input GIF image's LZW data is copy-code-heavy or
literal-code-heavy. For example, the suffix table grows by 8x, from 4KiB
to 32 KiB.

The gifplayer-muybridge.gif image is also larger (472 x 298) and the
hippopotamus.regular.gif image is also smaller (36 x 28). Higher fixed
costs may have larger (negative) relative effects on decoding smaller
images.

To repeat, I think it's still a net win.

An upcoming commit will update the std/lzw/README.md file to provide
more detail on the algorithm.

Thanks to Tushar Khandelwal for the idea.
2018-10-21 14:48:40 +11:00
Nigel Tao 8d7bd36179 Add more GIF benchmarks
The existing benchmarks decoded realistic photos. The new benchmarks
also decode artificially generated images, which compress better.

Also adjust script/extract-giflzw.go to optionally analyze and print a
histogram of LZW code emission lengths. "Compresses better" is highly
correlated with "a lower proportion of literal (1 length) codes".
2018-10-21 11:46:37 +11:00
Nigel Tao 6945bdc75c Fill the LZW bit buffer less often
name                             old speed     new speed     delta

wuffs_gif_decode_1k_bw/clang     292MB/s ± 1%  256MB/s ± 1%  -12.29%  (p=0.008 n=5+5)
wuffs_gif_decode_1k_color/clang  143MB/s ± 0%  159MB/s ± 0%  +10.95%  (p=0.008 n=5+5)
wuffs_gif_decode_10k/clang       150MB/s ± 1%  165MB/s ± 0%  +10.24%  (p=0.008 n=5+5)
wuffs_gif_decode_100k/clang      152MB/s ± 0%  150MB/s ± 0%   -1.19%  (p=0.016 n=4+5)
wuffs_gif_decode_1000k/clang     156MB/s ± 1%  153MB/s ± 1%   -1.60%  (p=0.008 n=5+5)

wuffs_gif_decode_1k_bw/gcc       250MB/s ± 1%  330MB/s ± 0%  +31.99%  (p=0.008 n=5+5)
wuffs_gif_decode_1k_color/gcc    172MB/s ± 1%  184MB/s ± 1%   +7.40%  (p=0.008 n=5+5)
wuffs_gif_decode_10k/gcc         189MB/s ± 0%  193MB/s ± 0%   +2.31%  (p=0.008 n=5+5)
wuffs_gif_decode_100k/gcc        166MB/s ± 0%  180MB/s ± 0%   +8.56%  (p=0.008 n=5+5)
wuffs_gif_decode_1000k/gcc       171MB/s ± 0%  186MB/s ± 0%   +8.19%  (p=0.008 n=5+5)

wuffs_lzw_decode_20k/clang       168MB/s ± 3%  168MB/s ± 0%     ~     (p=0.222 n=5+5)
wuffs_lzw_decode_100k/clang      252MB/s ± 1%  239MB/s ± 0%   -4.89%  (p=0.008 n=5+5)

wuffs_lzw_decode_20k/gcc         209MB/s ± 0%  217MB/s ± 0%   +4.28%  (p=0.008 n=5+5)
wuffs_lzw_decode_100k/gcc        307MB/s ± 0%  314MB/s ± 0%   +2.27%  (p=0.008 n=5+5)
2018-10-20 07:22:23 +11:00
Nigel Tao 24e7ee0104 Flush LZW output if stalled on input
name                             old speed     new speed     delta

wuffs_gif_decode_1k_bw/clang     295MB/s ± 0%  292MB/s ± 1%   -0.89%  (p=0.008 n=5+5)
wuffs_gif_decode_1k_color/clang  163MB/s ± 0%  143MB/s ± 0%  -12.15%  (p=0.008 n=5+5)
wuffs_gif_decode_10k/clang       179MB/s ± 0%  150MB/s ± 1%  -16.37%  (p=0.008 n=5+5)
wuffs_gif_decode_100k/clang      155MB/s ± 0%  152MB/s ± 0%   -1.84%  (p=0.016 n=5+4)
wuffs_gif_decode_1000k/clang     159MB/s ± 0%  156MB/s ± 1%   -2.26%  (p=0.008 n=5+5)

wuffs_gif_decode_1k_bw/gcc       313MB/s ± 0%  250MB/s ± 1%  -20.20%  (p=0.008 n=5+5)
wuffs_gif_decode_1k_color/gcc    173MB/s ± 0%  172MB/s ± 1%   -1.01%  (p=0.008 n=5+5)
wuffs_gif_decode_10k/gcc         184MB/s ± 1%  189MB/s ± 0%   +2.34%  (p=0.008 n=5+5)
wuffs_gif_decode_100k/gcc        173MB/s ± 1%  166MB/s ± 0%   -3.88%  (p=0.008 n=5+5)
wuffs_gif_decode_1000k/gcc       179MB/s ± 1%  171MB/s ± 0%   -4.07%  (p=0.008 n=5+5)

wuffs_lzw_decode_20k/clang       185MB/s ± 0%  168MB/s ± 3%   -9.34%  (p=0.008 n=5+5)
wuffs_lzw_decode_100k/clang      260MB/s ± 0%  252MB/s ± 1%   -3.36%  (p=0.008 n=5+5)

wuffs_lzw_decode_20k/gcc         211MB/s ± 0%  209MB/s ± 0%   -1.01%  (p=0.008 n=5+5)
wuffs_lzw_decode_100k/gcc        300MB/s ± 0%  307MB/s ± 0%   +2.16%  (p=0.008 n=5+5)
2018-10-14 22:35:42 +11:00
Nigel Tao ff92d22ea6 Read LZW input bits 4 bytes at a time, not 1
name                             old speed     new speed     delta

wuffs_gif_decode_1k_bw/clang     306MB/s ± 0%  295MB/s ± 0%   -3.62%  (p=0.008 n=5+5)
wuffs_gif_decode_1k_color/clang  156MB/s ± 0%  163MB/s ± 0%   +4.64%  (p=0.008 n=5+5)
wuffs_gif_decode_10k/clang       153MB/s ± 0%  179MB/s ± 0%  +17.34%  (p=0.008 n=5+5)
wuffs_gif_decode_100k/clang      147MB/s ± 0%  155MB/s ± 0%   +5.13%  (p=0.008 n=5+5)
wuffs_gif_decode_1000k/clang     151MB/s ± 1%  159MB/s ± 0%   +5.34%  (p=0.008 n=5+5)

wuffs_gif_decode_1k_bw/gcc       319MB/s ± 0%  313MB/s ± 0%   -1.79%  (p=0.008 n=5+5)
wuffs_gif_decode_1k_color/gcc    165MB/s ± 0%  173MB/s ± 0%   +4.82%  (p=0.008 n=5+5)
wuffs_gif_decode_10k/gcc         171MB/s ± 1%  184MB/s ± 1%   +7.48%  (p=0.008 n=5+5)
wuffs_gif_decode_100k/gcc        161MB/s ± 1%  173MB/s ± 1%   +7.32%  (p=0.008 n=5+5)
wuffs_gif_decode_1000k/gcc       166MB/s ± 0%  179MB/s ± 1%   +7.53%  (p=0.008 n=5+5)

wuffs_lzw_decode_20k/clang       167MB/s ± 0%  185MB/s ± 0%  +10.85%  (p=0.008 n=5+5)
wuffs_lzw_decode_100k/clang      239MB/s ± 0%  260MB/s ± 0%   +8.96%  (p=0.008 n=5+5)

wuffs_lzw_decode_20k/gcc         192MB/s ± 0%  211MB/s ± 0%   +9.47%  (p=0.008 n=5+5)
wuffs_lzw_decode_100k/gcc        295MB/s ± 0%  300MB/s ± 0%   +1.65%  (p=0.008 n=5+5)
2018-10-14 22:14:02 +11:00
Nigel Tao 4e27b1280e Write LZW decoder output in batches
name                             old speed     new speed     delta

wuffs_gif_decode_1k_bw/clang     309MB/s ± 0%  306MB/s ± 0%   -0.93%  (p=0.008 n=5+5)
wuffs_gif_decode_1k_color/clang  160MB/s ± 0%  156MB/s ± 0%   -2.26%  (p=0.008 n=5+5)
wuffs_gif_decode_10k/clang       145MB/s ± 0%  153MB/s ± 0%   +5.21%  (p=0.008 n=5+5)
wuffs_gif_decode_100k/clang      131MB/s ± 1%  147MB/s ± 0%  +12.15%  (p=0.008 n=5+5)
wuffs_gif_decode_1000k/clang     134MB/s ± 0%  151MB/s ± 1%  +12.83%  (p=0.008 n=5+5)

wuffs_gif_decode_1k_bw/gcc       339MB/s ± 0%  319MB/s ± 0%   -5.99%  (p=0.008 n=5+5)
wuffs_gif_decode_1k_color/gcc    186MB/s ± 0%  165MB/s ± 0%  -10.93%  (p=0.008 n=5+5)
wuffs_gif_decode_10k/gcc         164MB/s ± 0%  171MB/s ± 1%   +4.19%  (p=0.008 n=5+5)
wuffs_gif_decode_100k/gcc        143MB/s ± 0%  161MB/s ± 1%  +12.32%  (p=0.008 n=5+5)
wuffs_gif_decode_1000k/gcc       146MB/s ± 0%  166MB/s ± 0%  +13.61%  (p=0.008 n=5+5)

wuffs_lzw_decode_20k/clang       142MB/s ± 0%  167MB/s ± 0%  +17.51%  (p=0.008 n=5+5)
wuffs_lzw_decode_100k/clang      186MB/s ± 1%  239MB/s ± 0%  +28.68%  (p=0.008 n=5+5)

wuffs_lzw_decode_20k/gcc         162MB/s ± 0%  192MB/s ± 0%  +18.63%  (p=0.008 n=5+5)
wuffs_lzw_decode_100k/gcc        235MB/s ± 1%  295MB/s ± 0%  +25.59%  (p=0.008 n=5+5)
2018-10-14 17:53:20 +11:00
Nigel Tao 9573b88a02 Remove unused appendFact code path 2018-10-14 13:11:58 +11:00