Commit Graph
10 Commits
Author SHA1 Message Date
Nigel Tao dfab0d2b3d mimiclib: allow jpeg decoding via stb_image 2023-06-12 17:24:49 +10:00
Nigel Tao d5075c7e94 Push "int i" into "for (int i = etc; etc; etc)" 2021-11-02 14:07:58 +11:00
Nigel Tao 5dd7ede975 Add more png benchmarks for alternative mimic libs
On a mid-range x86_64 laptop:

"libpng" (Debian's /usr/lib/x86_64-linux-gnu/libpng16.so) numbers are
used as a 1.00x baseline for the other libraries.

(†): libpng's "simplified API" doesn't provide a way to ignore the
checksum. We copy the "verify_checksum" numbers for a 1.00x baseline.
Some of the other libraries don't have the option to ignore or verify
the checksum, always doing only one or the other.

----

libpng_decode_19k_8bpp                            58.0MB/s ± 0%  1.00x
libpng_decode_40k_24bpp                           73.1MB/s ± 0%  1.00x
libpng_decode_77k_8bpp                             177MB/s ± 0%  1.00x
libpng_decode_552k_32bpp_ignore_checksum           146MB/s ± 0%  (†)
libpng_decode_552k_32bpp_verify_checksum           146MB/s ± 0%  1.00x
libpng_decode_4002k_24bpp                          104MB/s ± 0%  1.00x

----

wuffs_decode_19k_8bpp/clang9                       131MB/s ± 0%  2.26x
wuffs_decode_40k_24bpp/clang9                      153MB/s ± 0%  2.09x
wuffs_decode_77k_8bpp/clang9                       472MB/s ± 0%  2.67x
wuffs_decode_552k_32bpp_ignore_checksum/clang9     370MB/s ± 0%  2.53x
wuffs_decode_552k_32bpp_verify_checksum/clang9     357MB/s ± 0%  2.45x
wuffs_decode_4002k_24bpp/clang9                    156MB/s ± 0%  1.50x

wuffs_decode_19k_8bpp/gcc10                        136MB/s ± 1%  2.34x
wuffs_decode_40k_24bpp/gcc10                       162MB/s ± 0%  2.22x
wuffs_decode_77k_8bpp/gcc10                        486MB/s ± 0%  2.75x
wuffs_decode_552k_32bpp_ignore_checksum/gcc10      388MB/s ± 0%  2.66x
wuffs_decode_552k_32bpp_verify_checksum/gcc10      373MB/s ± 0%  2.55x
wuffs_decode_4002k_24bpp/gcc10                     164MB/s ± 0%  1.58x

----

libspng_decode_19k_8bpp/clang9                    59.3MB/s ± 0%  1.02x
libspng_decode_40k_24bpp/clang9                   78.4MB/s ± 0%  1.07x
libspng_decode_77k_8bpp/clang9                     189MB/s ± 0%  1.07x
libspng_decode_552k_32bpp_ignore_checksum/clang9   236MB/s ± 0%  1.62x
libspng_decode_552k_32bpp_verify_checksum/clang9   203MB/s ± 0%  1.39x
libspng_decode_4002k_24bpp/clang9                  110MB/s ± 0%  1.06x

libspng_decode_19k_8bpp/gcc10                     59.6MB/s ± 0%  1.03x
libspng_decode_40k_24bpp/gcc10                    77.5MB/s ± 0%  1.06x
libspng_decode_77k_8bpp/gcc10                      189MB/s ± 0%  1.07x
libspng_decode_552k_32bpp_ignore_checksum/gcc10    223MB/s ± 0%  1.53x
libspng_decode_552k_32bpp_verify_checksum/gcc10    194MB/s ± 0%  1.33x
libspng_decode_4002k_24bpp/gcc10                   109MB/s ± 0%  1.05x

----

lodepng_decode_19k_8bpp/clang9                    65.1MB/s ± 0%  1.12x
lodepng_decode_40k_24bpp/clang9                   72.1MB/s ± 0%  0.99x
lodepng_decode_77k_8bpp/clang9                     222MB/s ± 0%  1.25x
lodepng_decode_552k_32bpp_ignore_checksum/clang9               skipped
lodepng_decode_552k_32bpp_verify_checksum/clang9   162MB/s ± 0%  1.11x
lodepng_decode_4002k_24bpp/clang9                 70.5MB/s ± 0%  0.68x

lodepng_decode_19k_8bpp/gcc10                     61.1MB/s ± 0%  1.05x
lodepng_decode_40k_24bpp/gcc10                    62.5MB/s ± 1%  0.85x
lodepng_decode_77k_8bpp/gcc10                      176MB/s ± 0%  0.99x
lodepng_decode_552k_32bpp_ignore_checksum/gcc10                skipped
lodepng_decode_552k_32bpp_verify_checksum/gcc10    139MB/s ± 0%  0.95x
lodepng_decode_4002k_24bpp/gcc10                  62.3MB/s ± 0%  0.60x

----

stbimage_decode_19k_8bpp/clang9                   75.1MB/s ± 1%  1.29x
stbimage_decode_40k_24bpp/clang9                  84.6MB/s ± 0%  1.16x
stbimage_decode_77k_8bpp/clang9                    234MB/s ± 0%  1.32x
stbimage_decode_552k_32bpp_ignore_checksum/clang9  162MB/s ± 0%  1.11x
stbimage_decode_552k_32bpp_verify_checksum/clang9              skipped
stbimage_decode_4002k_24bpp/clang9                80.7MB/s ± 0%  0.78x

stbimage_decode_19k_8bpp/gcc10                    73.3MB/s ± 0%  1.26x
stbimage_decode_40k_24bpp/gcc10                   81.8MB/s ± 0%  1.12x
stbimage_decode_77k_8bpp/gcc10                     214MB/s ± 0%  1.21x
stbimage_decode_552k_32bpp_ignore_checksum/gcc10   145MB/s ± 0%  0.99x
stbimage_decode_552k_32bpp_verify_checksum/gcc10               skipped
stbimage_decode_4002k_24bpp/gcc10                 79.7MB/s ± 0%  0.77x

----

go_decode_19k_8bpp/go1.16                         39.3MB/s ± 1%  0.68x
go_decode_40k_24bpp/go1.16                        46.5MB/s ± 1%  0.64x
go_decode_77k_8bpp/go1.16                         78.3MB/s ± 0%  0.44x
go_decode_552k_32bpp_ignore_checksum/go1.16                    skipped
go_decode_552k_32bpp_verify_checksum/go1.16        120MB/s ± 1%  0.82x
go_decode_4002k_24bpp/go1.16                      50.7MB/s ± 0%  0.49x

----

rust_decode_19k_8bpp/rust1.48                     88.3MB/s ± 0%  1.52x
rust_decode_40k_24bpp/rust1.48                     133MB/s ± 0%  1.82x
rust_decode_77k_8bpp/rust1.48                      180MB/s ± 0%  1.02x
rust_decode_552k_32bpp_ignore_checksum/rust1.48                skipped
rust_decode_552k_32bpp_verify_checksum/rust1.48    146MB/s ± 0%  1.00x
rust_decode_4002k_24bpp/rust1.48                   134MB/s ± 0%  1.29x
2021-03-24 23:36:48 +11:00
Nigel Tao fd8025342e Be able to test and bench the libspng library
That's libspng with an 's', instead of libpng without an 's'.

Wuffs' std/png and std/zlib implementations have not been fully
optimized yet, but as a snapshot, here's how it compares with these
"mimic" libraries, libspng and libpng.

On a mid-range x86_64 laptop:

Wuffs/clang9:
wuffs_png_decode_image_19k_8bpp/clang9                    91.9MB/s ± 0%
wuffs_png_decode_image_40k_24bpp/clang9                    128MB/s ± 0%
wuffs_png_decode_image_77k_8bpp/clang9                     342MB/s ± 0%
wuffs_png_decode_image_552k_32bpp_ignore_checksum/clang9   324MB/s ± 0%
wuffs_png_decode_image_552k_32bpp_verify_checksum/clang9   258MB/s ± 0%
wuffs_png_decode_image_4002k_24bpp/clang9                  127MB/s ± 0%

Wuffs/gcc10:
wuffs_png_decode_image_19k_8bpp/gcc10                     94.3MB/s ± 0%
wuffs_png_decode_image_40k_24bpp/gcc10                     131MB/s ± 0%
wuffs_png_decode_image_77k_8bpp/gcc10                      348MB/s ± 1%
wuffs_png_decode_image_552k_32bpp_ignore_checksum/gcc10    319MB/s ± 0%
wuffs_png_decode_image_552k_32bpp_verify_checksum/gcc10    269MB/s ± 0%
wuffs_png_decode_image_4002k_24bpp/gcc10                   131MB/s ± 0%

libspng/clang9:
mimic_png_decode_image_19k_8bpp/clang9                    59.3MB/s ± 1%
mimic_png_decode_image_40k_24bpp/clang9                   78.4MB/s ± 0%
mimic_png_decode_image_77k_8bpp/clang9                     189MB/s ± 0%
mimic_png_decode_image_552k_32bpp_ignore_checksum/clang9   235MB/s ± 0%
mimic_png_decode_image_552k_32bpp_verify_checksum/clang9   203MB/s ± 0%
mimic_png_decode_image_4002k_24bpp/clang9                  110MB/s ± 0%

libspng/gcc10:
mimic_png_decode_image_19k_8bpp/gcc10                     59.5MB/s ± 0%
mimic_png_decode_image_40k_24bpp/gcc10                    77.6MB/s ± 0%
mimic_png_decode_image_77k_8bpp/gcc10                      188MB/s ± 0%
mimic_png_decode_image_552k_32bpp_ignore_checksum/gcc10    223MB/s ± 0%
mimic_png_decode_image_552k_32bpp_verify_checksum/gcc10    193MB/s ± 0%
mimic_png_decode_image_4002k_24bpp/gcc10                   109MB/s ± 0%

libpng (Debian_Testing_Bullseye /usr/lib/x86_64-linux-gnu/libpng.so):
mimic_png_decode_image_19k_8bpp                           58.3MB/s ± 0%
mimic_png_decode_image_40k_24bpp                          73.1MB/s ± 0%
mimic_png_decode_image_77k_8bpp                            177MB/s ± 0%
mimic_png_decode_image_552k_32bpp_ignore_checksum               skipped
mimic_png_decode_image_552k_32bpp_verify_checksum          146MB/s ± 0%
mimic_png_decode_image_4002k_24bpp                         104MB/s ± 0%
2021-01-22 13:15:42 +11:00
Nigel Tao 170a810486 Add std/png benches for ignore/verify checksums
wuffs_png_decode_image_552k_32bpp_ignore_checksum/clang9   324MB/s ± 0%
wuffs_png_decode_image_552k_32bpp_verify_checksum/clang9   258MB/s ± 0%

wuffs_png_decode_image_552k_32bpp_ignore_checksum/gcc10    319MB/s ± 0%
wuffs_png_decode_image_552k_32bpp_verify_checksum/gcc10    269MB/s ± 0%
2021-01-22 11:34:52 +11:00
Nigel Tao cf4aff9ef0 Have testlib image_decoder helpers take quirks 2021-01-21 22:29:30 +11:00
Nigel Tao aa2d41e0a0 Add a mimiclib scratch buffer 2020-12-29 21:38:52 +11:00
Nigel Tao 10e8ee5e79 Roll back 0c7a8cc0 re do_bench_image_decode
Commit 0c7a8cc029 is "Have
do_bench_image_decode pass a non-NULL dst". Adding a mandatory memcpy to
do_bench_image_decode noticably slowed down the bench numbers. Removing
the memcpy speeds it back up.

wuffs_gif_decode_1k_bw/clang9                278MB/s ± 0%   373MB/s ± 0%  +34.42%  (p=0.008 n=5+5)
wuffs_gif_decode_1k_color_full_init/clang9  85.0MB/s ± 0%  91.1MB/s ± 0%   +7.27%  (p=0.008 n=5+5)
wuffs_gif_decode_1k_color_part_init/clang9   104MB/s ± 0%   114MB/s ± 0%   +9.51%  (p=0.008 n=5+5)
wuffs_gif_decode_10k_bgra/clang9             434MB/s ± 0%   449MB/s ± 0%   +3.59%  (p=0.008 n=5+5)
wuffs_gif_decode_10k_indexed/clang9          115MB/s ± 0%   120MB/s ± 0%   +4.63%  (p=0.008 n=5+5)
wuffs_gif_decode_20k/clang9                  140MB/s ± 1%   145MB/s ± 0%   +3.16%  (p=0.008 n=5+5)
wuffs_gif_decode_100k_artificial/clang9      301MB/s ± 0%   312MB/s ± 0%   +3.62%  (p=0.016 n=5+4)
wuffs_gif_decode_100k_realistic/clang9       125MB/s ± 0%   127MB/s ± 0%   +1.40%  (p=0.008 n=5+5)
wuffs_gif_decode_1000k_full_init/clang9      128MB/s ± 0%   129MB/s ± 0%   +0.78%  (p=0.008 n=5+5)
wuffs_gif_decode_1000k_part_init/clang9      128MB/s ± 0%   128MB/s ± 1%     ~     (p=0.151 n=5+5)
wuffs_gif_decode_anim_screencap/clang9       586MB/s ± 0%   671MB/s ± 0%  +14.60%  (p=0.016 n=5+4)

wuffs_gif_decode_1k_bw/gcc10                 271MB/s ± 0%   322MB/s ± 0%  +18.81%  (p=0.008 n=5+5)
wuffs_gif_decode_1k_color_full_init/gcc10   94.8MB/s ± 0%  99.2MB/s ± 1%   +4.64%  (p=0.008 n=5+5)
wuffs_gif_decode_1k_color_part_init/gcc10    116MB/s ± 0%   123MB/s ± 0%   +5.90%  (p=0.008 n=5+5)
wuffs_gif_decode_10k_bgra/gcc10              374MB/s ± 0%   382MB/s ± 1%   +2.12%  (p=0.008 n=5+5)
wuffs_gif_decode_10k_indexed/gcc10           116MB/s ± 1%   119MB/s ± 0%   +3.09%  (p=0.008 n=5+5)
wuffs_gif_decode_20k/gcc10                   137MB/s ± 0%   139MB/s ± 0%   +1.46%  (p=0.008 n=5+5)
wuffs_gif_decode_100k_artificial/gcc10       286MB/s ± 0%   293MB/s ± 0%   +2.49%  (p=0.008 n=5+5)
wuffs_gif_decode_100k_realistic/gcc10        118MB/s ± 0%   119MB/s ± 0%   +0.80%  (p=0.016 n=5+4)
wuffs_gif_decode_1000k_full_init/gcc10       121MB/s ± 0%   121MB/s ± 0%   +0.33%  (p=0.008 n=5+5)
wuffs_gif_decode_1000k_part_init/gcc10       121MB/s ± 0%   121MB/s ± 0%   +0.28%  (p=0.016 n=5+5)
wuffs_gif_decode_anim_screencap/gcc10        582MB/s ± 0%   659MB/s ± 0%  +13.25%  (p=0.008 n=5+5)

wuffs_png_decode_19k_8bpp/clang9            88.7MB/s ± 0%  94.2MB/s ± 0%   +6.28%  (p=0.008 n=5+5)
wuffs_png_decode_40k_24bpp/clang9            104MB/s ± 0%   108MB/s ± 0%   +3.96%  (p=0.008 n=5+5)
wuffs_png_decode_77k_8bpp/clang9             335MB/s ± 0%   352MB/s ± 0%   +5.29%  (p=0.008 n=5+5)
wuffs_png_decode_552k_32bpp/clang9           118MB/s ± 0%   119MB/s ± 0%   +0.54%  (p=0.008 n=5+5)
wuffs_png_decode_4002k_24bpp/clang9          109MB/s ± 0%   112MB/s ± 0%   +2.36%  (p=0.008 n=5+5)

wuffs_png_decode_19k_8bpp/gcc10             96.7MB/s ± 0%  98.0MB/s ± 0%   +1.36%  (p=0.029 n=4+4)
wuffs_png_decode_40k_24bpp/gcc10             114MB/s ± 0%   115MB/s ± 0%   +0.79%  (p=0.008 n=5+5)
wuffs_png_decode_77k_8bpp/gcc10              316MB/s ± 0%   320MB/s ± 0%   +1.18%  (p=0.016 n=5+4)
wuffs_png_decode_552k_32bpp/gcc10            149MB/s ± 0%   149MB/s ± 0%   +0.17%  (p=0.016 n=5+4)
wuffs_png_decode_4002k_24bpp/gcc10           116MB/s ± 0%   116MB/s ± 0%   +0.09%  (p=0.032 n=5+5)
2020-12-29 21:29:32 +11:00
Nigel Tao 0c7a8cc029 Have do_bench_image_decode pass a non-NULL dst 2020-12-28 15:54:15 +11:00
Nigel Tao 88616cc969 Add std/png mimic benches
Wuffs' std/png and std/zlib implementations have not been optimized yet,
but even at this early stage, they are roughly on par with their
"mimics": equivalent production-quality C libraries (libpng and zlib).

On a mid-range x86_64 laptop:

wuffs_png_decode_19k_8bpp/clang9     95.8MB/s ± 0%
wuffs_png_decode_40k_24bpp/clang9     100MB/s ± 1%
wuffs_png_decode_77k_8bpp/clang9      359MB/s ± 0%
wuffs_png_decode_552k_32bpp/clang9    119MB/s ± 0%
wuffs_png_decode_4002k_24bpp/clang9   101MB/s ± 0%

wuffs_png_decode_19k_8bpp/gcc10      97.2MB/s ± 0%
wuffs_png_decode_40k_24bpp/gcc10      111MB/s ± 0%
wuffs_png_decode_77k_8bpp/gcc10       316MB/s ± 0%
wuffs_png_decode_552k_32bpp/gcc10     132MB/s ± 0%
wuffs_png_decode_4002k_24bpp/gcc10    116MB/s ± 0%

mimic_png_decode_19k_8bpp            58.1MB/s ± 0%
mimic_png_decode_40k_24bpp           73.3MB/s ± 0%
mimic_png_decode_77k_8bpp             177MB/s ± 0%
mimic_png_decode_552k_32bpp           147MB/s ± 0%
mimic_png_decode_4002k_24bpp          104MB/s ± 0%

wuffs_zlib_decode_10k/clang9   126MB/s ± 0%
wuffs_zlib_decode_100k/clang9  136MB/s ± 0%

wuffs_zlib_decode_10k/gcc10    131MB/s ± 0%
wuffs_zlib_decode_100k/gcc10   159MB/s ± 0%

mimic_zlib_decode_10k          133MB/s ± 0%
mimic_zlib_decode_100k         160MB/s ± 0%
2020-12-28 15:52:46 +11:00