Commit Graph
15 Commits
Author SHA1 Message Date
Nigel Tao f8146cf1c6 Remove more PUFFS_USE_NO_OP_PERFORMANCE_HACKS. 2017-11-03 10:15:43 +11:00
Nigel Tao e3822f9951 Re-introduce puffs_base__limit1.
It's not immediately obvious why this data structure performs differently, but
on net, the benchmark numbers seem better. It also won't require any
PUFFS_USE_NO_OP_PERFORMANCE_HACKS or "redundant but mysteriously performant"
use_limit fields.

name                             old speed      new speed      delta

puffs_adler32_10k/clang          2.46GB/s ± 0%  2.43GB/s ± 3%     ~     (p=0.513 n=5+10)
puffs_adler32_100k/clang         2.46GB/s ± 0%  2.44GB/s ± 2%     ~     (p=0.241 n=5+10)
puffs_flate_decode_1k/clang       122MB/s ± 1%   119MB/s ± 2%   -1.86%  (p=0.013 n=5+10)
puffs_flate_decode_10k/clang      147MB/s ± 0%   142MB/s ± 2%   -3.47%  (p=0.001 n=5+10)
puffs_flate_decode_100k/clang     161MB/s ± 0%   161MB/s ± 4%     ~     (p=0.635 n=4+10)
puffs_zlib_decode_10k/clang       139MB/s ± 0%   135MB/s ± 1%   -2.69%  (p=0.001 n=5+10)
puffs_zlib_decode_100k/clang      151MB/s ± 0%   152MB/s ± 3%     ~     (p=1.000 n=5+10)

puffs_adler32_10k/gcc            3.21GB/s ± 1%  3.18GB/s ± 2%     ~     (p=0.310 n=5+10)
puffs_adler32_100k/gcc           3.22GB/s ± 0%  3.19GB/s ± 2%     ~     (p=0.075 n=5+10)
puffs_flate_decode_1k/gcc         121MB/s ± 4%   126MB/s ± 1%   +4.54%  (p=0.002 n=5+8)
puffs_flate_decode_10k/gcc        144MB/s ± 1%   141MB/s ± 1%   -2.52%  (p=0.001 n=5+9)
puffs_flate_decode_100k/gcc       154MB/s ± 1%   158MB/s ± 0%   +2.39%  (p=0.001 n=5+9)
puffs_zlib_decode_10k/gcc         137MB/s ± 2%   135MB/s ± 1%     ~     (p=0.055 n=5+10)
puffs_zlib_decode_100k/gcc        146MB/s ± 2%   150MB/s ± 0%   +2.87%  (p=0.001 n=5+10)

puffs_gif_lzw_decode_20k/clang    132MB/s ± 0%   125MB/s ± 0%   -5.24%  (p=0.001 n=5+10)
puffs_gif_lzw_decode_100k/clang   176MB/s ± 0%   164MB/s ± 0%   -6.82%  (p=0.001 n=5+9)
puffs_gif_decode_1k/clang         363MB/s ± 0%   339MB/s ± 1%   -6.58%  (p=0.002 n=4+10)
puffs_gif_decode_10k/clang        138MB/s ± 0%   133MB/s ± 1%   -3.77%  (p=0.001 n=5+10)
puffs_gif_decode_100k/clang       125MB/s ± 0%   119MB/s ± 0%   -4.79%  (p=0.001 n=5+10)
puffs_gif_decode_1000k/clang      127MB/s ± 0%   121MB/s ± 0%   -5.04%  (p=0.001 n=5+10)

puffs_gif_lzw_decode_20k/gcc      141MB/s ± 0%   140MB/s ± 2%     ~     (p=1.000 n=5+10)
puffs_gif_lzw_decode_100k/gcc     182MB/s ± 0%   180MB/s ± 2%   -1.56%  (p=0.001 n=5+10)
puffs_gif_decode_1k/gcc           363MB/s ± 0%   409MB/s ± 3%  +12.71%  (p=0.001 n=5+10)
puffs_gif_decode_10k/gcc          127MB/s ± 0%   161MB/s ± 3%  +26.15%  (p=0.001 n=5+10)
puffs_gif_decode_100k/gcc         112MB/s ± 0%   136MB/s ± 2%  +21.95%  (p=0.001 n=5+10)
puffs_gif_decode_1000k/gcc        114MB/s ± 0%   140MB/s ± 2%  +22.78%  (p=0.001 n=5+10)
2017-11-03 10:09:57 +11:00
Nigel Tao e5db8233fc Re-introduce writer1.use_limit.
Nobody uses it yet, as we only decode and not encode, but it's there for
symmetry.

No benchmark changes, as nobody uses it yet.
2017-11-02 16:46:03 +11:00
Nigel Tao 7d8085f891 Re-introduce reader1.use_limit.
The std/flate benchmarks get better. The std/gif benchmarks get worse,
but that's possibly because the PUFFS_USE_NO_OP_PERFORMANCE_HACKS hacks
are fragile.

name                             old speed      new speed      delta

puffs_adler32_10k/clang          2.45GB/s ± 0%  2.45GB/s ± 0%     ~     (p=0.310 n=5+5)
puffs_adler32_100k/clang         2.46GB/s ± 0%  2.46GB/s ± 0%     ~     (p=0.690 n=5+5)
puffs_flate_decode_1k/clang       113MB/s ± 0%   113MB/s ± 0%     ~     (p=0.690 n=5+5)
puffs_flate_decode_10k/clang      140MB/s ± 0%   140MB/s ± 0%     ~     (p=0.222 n=5+5)
puffs_flate_decode_100k/clang     151MB/s ± 0%   151MB/s ± 0%     ~     (p=0.095 n=5+5)
puffs_zlib_decode_10k/clang       132MB/s ± 1%   132MB/s ± 0%     ~     (p=0.548 n=5+5)
puffs_zlib_decode_100k/clang      143MB/s ± 0%   143MB/s ± 0%     ~     (p=0.730 n=5+4)

puffs_adler32_10k/gcc            3.26GB/s ± 0%  3.26GB/s ± 0%     ~     (p=0.548 n=5+5)
puffs_adler32_100k/gcc           3.27GB/s ± 0%  3.27GB/s ± 0%     ~     (p=0.841 n=5+5)
puffs_flate_decode_1k/gcc         108MB/s ± 0%   125MB/s ± 0%  +14.97%  (p=0.008 n=5+5)
puffs_flate_decode_10k/gcc        128MB/s ± 0%   148MB/s ± 0%  +15.48%  (p=0.016 n=4+5)
puffs_flate_decode_100k/gcc       136MB/s ± 0%   158MB/s ± 0%  +16.40%  (p=0.008 n=5+5)
puffs_zlib_decode_10k/gcc         123MB/s ± 0%   141MB/s ± 0%  +14.70%  (p=0.016 n=4+5)
puffs_zlib_decode_100k/gcc        131MB/s ± 0%   151MB/s ± 0%  +15.60%  (p=0.016 n=5+4)

puffs_gif_lzw_decode_20k/clang    131MB/s ± 1%   128MB/s ± 3%   -2.04%  (p=0.032 n=5+5)
puffs_gif_lzw_decode_100k/clang   172MB/s ± 1%   173MB/s ± 1%     ~     (p=0.151 n=5+5)
puffs_gif_decode_1k/clang         351MB/s ± 3%   341MB/s ± 5%     ~     (p=0.310 n=5+5)
puffs_gif_decode_10k/clang        138MB/s ± 1%   134MB/s ± 3%   -3.34%  (p=0.008 n=5+5)
puffs_gif_decode_100k/clang       122MB/s ± 2%   122MB/s ± 2%     ~     (p=1.000 n=5+5)
puffs_gif_decode_1000k/clang      126MB/s ± 1%   122MB/s ± 4%   -3.07%  (p=0.016 n=5+5)

puffs_gif_lzw_decode_20k/gcc      140MB/s ± 3%   140MB/s ± 3%     ~     (p=0.421 n=5+5)
puffs_gif_lzw_decode_100k/gcc     187MB/s ± 2%   184MB/s ± 1%     ~     (p=0.056 n=5+5)
puffs_gif_decode_1k/gcc           401MB/s ± 1%   357MB/s ± 1%  -10.88%  (p=0.008 n=5+5)
puffs_gif_decode_10k/gcc          154MB/s ± 1%   126MB/s ± 1%  -18.42%  (p=0.008 n=5+5)
puffs_gif_decode_100k/gcc         131MB/s ± 2%   109MB/s ± 0%  -17.37%  (p=0.008 n=5+5)
puffs_gif_decode_1000k/gcc        135MB/s ± 1%   111MB/s ± 1%  -17.65%  (p=0.008 n=5+5)
2017-11-02 16:35:15 +11:00
Nigel Tao ed6ce6fd54 Add PUFFS_USE_NO_OP_PERFORMANCE_HACKS.
name                             old speed      new speed      delta

puffs_adler32_10k/gcc            3.26GB/s ± 0%  3.26GB/s ± 0%     ~     (p=0.222 n=5+5)
puffs_adler32_100k/gcc           3.27GB/s ± 0%  3.27GB/s ± 0%   -0.12%  (p=0.032 n=5+5)
puffs_flate_decode_1k/gcc         108MB/s ± 0%   108MB/s ± 0%     ~     (p=0.841 n=5+5)
puffs_flate_decode_10k/gcc        128MB/s ± 0%   128MB/s ± 0%     ~     (p=0.057 n=4+4)
puffs_flate_decode_100k/gcc       136MB/s ± 0%   136MB/s ± 0%     ~     (p=0.222 n=5+5)
puffs_zlib_decode_10k/gcc         123MB/s ± 0%   123MB/s ± 0%     ~     (p=0.730 n=5+4)
puffs_zlib_decode_100k/gcc        130MB/s ± 0%   131MB/s ± 0%     ~     (p=1.000 n=5+5)

puffs_gif_lzw_decode_20k/gcc      137MB/s ± 4%   140MB/s ± 3%     ~     (p=0.095 n=5+5)
puffs_gif_lzw_decode_100k/gcc     177MB/s ± 3%   187MB/s ± 2%   +5.75%  (p=0.008 n=5+5)
puffs_gif_decode_1k/gcc           348MB/s ± 0%   401MB/s ± 1%  +15.34%  (p=0.016 n=4+5)
puffs_gif_decode_10k/gcc          121MB/s ± 5%   154MB/s ± 1%  +27.90%  (p=0.008 n=5+5)
puffs_gif_decode_100k/gcc         106MB/s ± 4%   131MB/s ± 2%  +24.35%  (p=0.008 n=5+5)
puffs_gif_decode_1000k/gcc        109MB/s ± 2%   135MB/s ± 1%  +24.03%  (p=0.008 n=5+5)
2017-11-02 14:51:29 +11:00
Nigel Tao aa396e7992 Change cgen's limit from a count to a pointer. 2017-10-27 10:28:35 +11:00
Nigel Tao 3485950a3f Remove the obsolete puffs_base__limit1 concept. 2017-10-27 07:51:57 +11:00
Nigel Tao 8ff453c752 Re-do how limits work. 2017-10-26 16:59:03 +11:00
Nigel Tao 21bbbd422f Generate code for "mark in.dst".
The generated code is ugly and the language design is not finalized, but
the flate_backref_crosses_blocks test (testing a correctness bug) now
passes. We can fix the ugliness in subsequent commits.
2017-10-22 14:44:53 +11:00
Nigel Tao f29c4e278f Follow the pkg__type__method convention for base. 2017-10-21 15:59:08 +11:00
Nigel Tao d4372cb5b8 Change license to Apache 2.
Also remove explicit PATENTS file. The new LICENSE file already
discusses patents.

Note that the Puffs project has not yet been publicly released.
2017-10-12 11:17:41 +11:00
Nigel Tao 6f132a74e2 Remove the 'mark' concept, no longer necessary. 2017-10-02 14:02:13 +11:00
Nigel Tao fd5d44b3fc Add a mark method and type. 2017-09-23 13:17:37 +10:00
Nigel Tao 519ec373f2 First cut at a "[] u8" slice type. 2017-09-08 16:53:59 +10:00
Nigel Tao fdddcc8e1d Merge multiple puffs-xxx-c binaries into one. 2017-08-24 09:43:57 +10:00