Commit Graph
85 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 2434f0dee5 Remove some redundant cgen funk fields 2018-11-24 15:25:31 +11:00
Nigel Tao 762b320b5e Make the return value mandatory; add ok 2018-11-24 15:09:28 +11:00
Nigel Tao 875362ea07 Change status "?foo" to just "?foo" 2018-11-24 12:36:42 +11:00
Nigel Tao 4023294ba2 Add a gif.decoder.frame_dirty_rect method 2018-11-17 21:26:53 +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 9c04249bfe Add warnings as another status code category 2018-08-26 23:22:45 +10:00
Nigel Tao eec0fce647 Change error "?foo" to status "?foo" 2018-08-12 21:31:56 +10:00
Nigel Tao b840c80040 Make the status type a const char*, not a struct 2018-08-12 16:05:14 +10:00
Nigel Tao 2a32a86a6c Make the status type a struct, not an int32_t 2018-08-11 13:27:29 +10:00
Nigel Tao 9c0b2d2f5c Rename ioptr_etc cgen variables 2018-08-11 08:32:49 +10:00
Nigel Tao f2ee78aa2a Remove short_read_src cgen code path 2018-08-04 09:45:35 +10:00
Nigel Tao df19bcdb03 Require if and while conditions to be effect-free 2018-07-29 14:20:15 +10:00
Nigel Tao 5cca9471ec Have cgen look for !?? methods not ? methods 2018-07-29 14:01:57 +10:00
Nigel Tao 0a47966e72 Introduce finer-grained effects 2018-07-29 13:40:49 +10:00
Nigel Tao 9919d78245 Change func out-type from struct to bare type 2018-07-29 10:02:06 +10:00
Nigel Tao 85c4de47b9 Remove ProvenNotToSuspend and BoundsCheckOptimized 2018-07-08 21:48:50 +10:00
Nigel Tao ccb78f5db7 Rename ast.Node methods from Foo to AsFoo 2018-07-01 20:41:21 +10:00
Nigel Tao 50a955317a Remove the WUFFS_FOO__STATUS_ETC names 2018-06-10 22:05:06 +10:00
Nigel Tao d770e879ab Remove the wuffs_foo__status types 2018-06-10 21:40:43 +10:00
Nigel Tao 8474b473ca Refactor some read_uxx cgen code 2018-06-02 13:11:49 +10:00
Nigel Tao 20e459cfb2 Add read_u24be etc methods
Fixes #12
2018-06-02 13:03:24 +10:00
Nigel Tao 00203e564f Let ? methods return OK even without suspending 2018-05-27 19:09:58 +10:00
Nigel Tao eda326726a Allow table base.u8 as a type 2018-05-18 10:31:24 +10:00
Nigel Tao b5e45740c0 Rename o_0_ioptr1_v_w to o_0_ioptr_v_w 2018-05-17 10:03:51 +10:00
Nigel Tao 7452ec8035 Rename b_rptr_foo as ioptr_foo 2018-05-17 09:59:04 +10:00
Nigel Tao 2448b2ae7a Factor out writeBuiltinCallSuspendibles 2018-05-16 12:12:47 +10:00
Nigel Tao 6835c6e23c Remove more hard-coding in writeCallSuspendibles
Fixes #8
2018-05-16 11:51:33 +10:00
Nigel Tao 8b4a82ab83 Remove hard-coded output in writeCallSuspendibles
Updates #8
2018-05-16 11:41:48 +10:00
Nigel Tao 8071a2096c Remove parenthesesPolicy, part 3 of N 2018-05-14 20:48:31 +10:00
Nigel Tao 0870643c77 Remove parenthesesPolicy, part 2 of N 2018-05-14 20:42:39 +10:00
Nigel Tao f159bcdadc Remove parenthesesPolicy, part 1 of N 2018-05-14 20:33:52 +10:00
Nigel Tao f10ff8c896 Have std/gif pass through an intermediary 2018-05-13 15:50:51 +10:00
Nigel Tao ed1a8b4057 Update std/gif's pass-through uncompressed_wi 2018-05-13 13:09:58 +10:00
Nigel Tao a86023b290 Simplify temporary io_bind variables 2018-05-13 10:52:49 +10:00
Nigel Tao 92720594ac Distinguish io_bind(in.foo) versus io_bind(foo) 2018-05-13 10:45:07 +10:00
Nigel Tao 2d3b3b1dae Add an io_writer var to gif's decoder.decode_id 2018-05-13 10:40:44 +10:00
Nigel Tao bd12c54ab5 Rename mark to set_mark, part 1 of N 2018-05-12 14:17:26 +10:00
Nigel Tao b65fdee306 Make the else-length-1 iterate tail explicit 2018-05-11 10:22:24 +10:00
Nigel Tao e8eeee371a Raise built-in ident capacity from 96 to 768 2018-05-05 09:23:10 +10:00
Nigel Tao 1702c1e303 Rename iterate's step to length 2018-05-05 08:25:23 +10:00
Nigel Tao 76b74f9d3e Refactor gen.writeStatement 2018-05-04 14:28:50 +10:00
Nigel Tao 5c25b0e1d1 Factor out a writeIterateRound method 2018-05-04 14:19:58 +10:00
Nigel Tao 7bd37588b1 Factor out a writeIterate method 2018-05-04 14:03:47 +10:00
Nigel Tao 58cd282138 Unroll the std/crc32 loop
name                          old speed      new speed      delta
wuffs_crc32_ieee_10k/clang    2.30GB/s ± 1%  2.37GB/s ± 0%  +3.31%  (p=0.016 n=5+4)
wuffs_crc32_ieee_100k/clang   2.30GB/s ± 0%  2.35GB/s ± 0%  +2.39%  (p=0.008 n=5+5)
wuffs_crc32_ieee_10k/gcc      2.49GB/s ± 1%  2.72GB/s ± 0%  +9.19%  (p=0.008 n=5+5)
wuffs_crc32_ieee_100k/gcc     2.50GB/s ± 0%  2.70GB/s ± 1%  +8.18%  (p=0.008 n=5+5)
wuffs_gzip_decode_10k/clang    206MB/s ± 2%   207MB/s ± 0%    ~     (p=1.000 n=5+5)
wuffs_gzip_decode_100k/clang   231MB/s ± 2%   235MB/s ± 0%  +1.57%  (p=0.008 n=5+5)
wuffs_gzip_decode_10k/gcc      234MB/s ± 0%   235MB/s ± 1%    ~     (p=0.095 n=5+5)
wuffs_gzip_decode_100k/gcc     267MB/s ± 0%   266MB/s ± 2%    ~     (p=0.690 n=5+5)
2018-05-04 14:00:04 +10:00
Nigel Tao a23681d1ba Rename some generated iterate variables 2018-05-04 13:44:55 +10:00
Nigel Tao 96a549814d Change iterate variable types from ptr to slice 2018-05-04 13:20:34 +10:00