Commit Graph
200 Commits
Author SHA1 Message Date
Nigel Tao 6ea3144379 Add test/data/checkerboard.*.png 2021-10-20 09:58:42 +11:00
Nigel Tao e6733aec19 Add test/data/red-blue-gradient.dcip3d65.png 2021-10-15 11:29:26 +11:00
Nigel Tao dbc209dd7e Add test/data/red-blue-gradient.*.png 2021-10-14 15:58:03 +11:00
Nigel Tao 62125d8967 Add test/nia-checksums-of-data.txt 2021-10-07 16:49:38 +11:00
Nigel Tao 6fb7697b4a Have scripts use WUFFS_CONFIG__STATIC_FUNCTIONS 2021-10-07 14:36:35 +11:00
Nigel Tao 75faea996d Add parse "122.416294033786585" test case 2021-08-22 16:02:44 +10:00
Nigel Tao 226c476994 Remove calls to Go's deprecated "io/ioutil" 2021-08-22 11:17:34 +10:00
Nigel Tao 788479dd64 Run "gofmt -w ." for new "//go:build" lines 2021-08-22 10:53:06 +10:00
Nigel Tao d3c7d103ca Update script/*.go module dependencies 2021-08-22 10:49:18 +10:00
Nigel Tao 816475b285 Add "b" (for "binary") to fopen calls
Updates #50
2021-07-07 20:28:35 +10:00
Nigel Tao a51867dc63 Fix "forseen" typo 2021-05-19 21:34:09 +10:00
Lander Brandt ab5f974b66 Use iterators for converting RGB to BGRA in the Rust PNG benchmark 2021-04-07 14:34:35 -07:00
Nigel Tao 61635213cd Add "‼" to "WUFFS C HEADER ENDS HERE" comment 2021-03-25 14:11:23 +11:00
Nigel Tao 7a73708a56 Have script/bench-*-png convert to BGRA
On a mid-range x86_64 laptop:

Compared to the previous commit:

go_decode_19k_8bpp                      39.3MB/s ± 1%  39.4MB/s ± 1%     ~     (p=0.548 n=5+5)
go_decode_40k_24bpp                     46.5MB/s ± 1%  46.7MB/s ± 1%     ~     (p=0.095 n=5+5)
go_decode_77k_8bpp                      78.3MB/s ± 0%  78.3MB/s ± 0%     ~     (p=0.286 n=4+5)
go_decode_552k_32bpp_verify_checksum     120MB/s ± 1%   118MB/s ± 0%     ~     (p=0.056 n=5+5)
go_decode_4002k_24bpp                   50.7MB/s ± 0%  50.5MB/s ± 0%   -0.58%  (p=0.016 n=5+4)

rust_decode_19k_8bpp                    88.3MB/s ± 0%  89.8MB/s ± 0%   +1.76%  (p=0.008 n=5+5)
rust_decode_40k_24bpp                    133MB/s ± 0%   122MB/s ± 0%   -8.58%  (p=0.008 n=5+5)
rust_decode_77k_8bpp                     180MB/s ± 0%   158MB/s ± 0%  -12.04%  (p=0.008 n=5+5)
rust_decode_552k_32bpp_verify_checksum   146MB/s ± 0%   136MB/s ± 0%   -6.59%  (p=0.016 n=4+5)
rust_decode_4002k_24bpp                  134MB/s ± 0%   122MB/s ± 0%   -9.02%  (p=0.008 n=5+5)

Compared to libpng (the C library):

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_verify_checksum                146MB/s ± 0%  1.00x
libpng_decode_4002k_24bpp                               104MB/s ± 0%  1.00x

go_decode_19k_8bpp/go1.16                              39.4MB/s ± 1%  0.68x
go_decode_40k_24bpp/go1.16                             46.7MB/s ± 1%  0.64x
go_decode_77k_8bpp/go1.16                              78.3MB/s ± 0%  0.44x
go_decode_552k_32bpp_verify_checksum/go1.16             118MB/s ± 0%  0.81x
go_decode_4002k_24bpp/go1.16                           50.5MB/s ± 0%  0.49x

rust_decode_19k_8bpp/rust1.48                          89.8MB/s ± 0%  1.55x
rust_decode_40k_24bpp/rust1.48                          122MB/s ± 0%  1.67x
rust_decode_77k_8bpp/rust1.48                           158MB/s ± 0%  0.89x
rust_decode_552k_32bpp_verify_checksum/rust1.48         136MB/s ± 0%  0.93x
rust_decode_4002k_24bpp/rust1.48                        122MB/s ± 0%  1.17x
2021-03-25 12:38:02 +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 85bdcd1e1e Fix script/benchstat-ratio.go when there's no '/' 2021-03-13 22:40:37 +11:00
Nigel Tao ee62582609 Tweak some mimic bench names for benchstat-ratio 2021-03-13 13:31:41 +11:00
Nigel Tao 9e4c87b6ab Add script/benchstat-ratio.go 2021-03-13 00:11:50 +11:00
Nigel Tao 79c7ba0ae9 Add script/print-crc32-x86-sse42-magic-numbers.go 2021-03-07 22:18:52 +11:00
Nigel Tao 3c627574a7 Fix and tweak hand-written decode_huffman_fast 2021-02-10 16:02:05 +11:00
Nigel Tao f89c3d81bd Build fix for hand-written decode_huffman_fast 2021-01-29 09:08:25 +11:00
Nigel Tao 2f78804e62 Use more inclusive language 2021-01-23 19:50:47 +11:00
Nigel Tao 5477132298 Have std/png verify PLTE chunk checksums 2021-01-21 20:21:33 +11:00
Nigel Tao a1c22caa15 Rename load and store to peek and poke 2021-01-17 22:44:39 +11:00
Nigel Tao f473ffa050 Deprecate script/preprocess-wuffs.go 2020-12-18 22:10:16 +11:00
Nigel Tao 6775c05c08 Have etc-parse-number-f64.cc test fast_float 2020-10-24 17:44:06 +11:00
Nigel Tao 93f9550720 Update manual-test-parse-number-f64.cc data repo 2020-10-23 18:46:12 +11:00
Nigel Tao ba3818cb6b Add more parse_number_f64 test cases 2020-09-28 12:51:45 +10:00
Nigel Tao 6c2fb9a2fd Tweak test/data/artif*/gif-transparent-index.gif 2020-09-27 00:09:25 +10:00
Nigel Tao 344123b179 Add test/data/artificial/gif-transparent-index.gif 2020-09-25 23:01:43 +10:00
Nigel Tao afe7f27ab7 Trim Eisel-Lemire powers-of-10 tables
This tracks the upstream github.com/lemire/fast_double_parser

On a mid-range x86_64 laptop, processing the 181 MiB citylots.json file
from github.com/zemirco/sf-city-lots-json:

$ g++ -O3 script/process-json-numbers.c
$ time ./a.out -parse-number-f64 < citylots.json
Before/After:
real    0m1.035s
real    0m1.003s

Binary size, before:
29544 gen/lib/c/clang-9-dynamic/wuffs-base-floatconv.lo
29648 gen/lib/c/clang-9-static/wuffs-base-floatconv.o
31192 gen/lib/c/gcc-dynamic/wuffs-base-floatconv.lo
31160 gen/lib/c/gcc-static/wuffs-base-floatconv.o

After:
27160 gen/lib/c/clang-9-dynamic/wuffs-base-floatconv.lo
27264 gen/lib/c/clang-9-static/wuffs-base-floatconv.o
28760 gen/lib/c/gcc-dynamic/wuffs-base-floatconv.lo
28760 gen/lib/c/gcc-static/wuffs-base-floatconv.o
2020-09-23 16:18:27 +10:00
Nigel Tao b6d855210c Use a formula to calculate Eisel-Lemire power-of-2
On a mid-range x86_64 laptop, processing the 181 MiB citylots.json file
from github.com/zemirco/sf-city-lots-json:

$ g++ -O3 script/process-json-numbers.c
$ time ./a.out -parse-number-f64 < citylots.json
Before/After:
real    0m0.992s
real    0m1.035s
2020-09-23 15:23:12 +10:00
Nigel Tao cd4cbc9102 Tweak JSON trailing comment/new-line quirks 2020-09-22 23:03:49 +10:00
Nigel Tao f2be4dd7cf Trim private_implementation__powers_of_10 array
Binary size, before:
30344 gen/lib/c/clang-9-dynamic/wuffs-base-floatconv.lo
30488 gen/lib/c/clang-9-static/wuffs-base-floatconv.o
31992 gen/lib/c/gcc-dynamic/wuffs-base-floatconv.lo
31992 gen/lib/c/gcc-static/wuffs-base-floatconv.o

After:
29512 gen/lib/c/clang-9-dynamic/wuffs-base-floatconv.lo
29656 gen/lib/c/clang-9-static/wuffs-base-floatconv.o
31160 gen/lib/c/gcc-dynamic/wuffs-base-floatconv.lo
31160 gen/lib/c/gcc-static/wuffs-base-floatconv.o
2020-09-15 22:14:30 +10:00
Nigel Tao ddac2c50ee Add more {parse,render}_number_f64 test cases 2020-09-15 21:17:33 +10:00
Nigel Tao 57d47c633c Fix etc__high_prec_dec__to_f64 shift amount
Shifting 0.299 left by 2 doesn't put us in [½ .. 1].
2020-09-08 16:43:31 +10:00
Nigel Tao 1210340e75 Add script/manual-test-parse-number-f64.cc 2020-09-07 23:11:07 +10:00
Nigel Tao 510c230cf4 Rename USE_FAST_DOUBLE_PARSER macro 2020-09-07 22:59:45 +10:00
Nigel Tao a20a2bbe56 Fix "stand-alone C++" typo 2020-09-07 21:06:58 +10:00
Nigel Tao 6344181b96 Add look-up table for JSON ASCII escapes
On a mid-range x86_64 laptop, processing the 181 MiB citylots.json file
from github.com/zemirco/sf-city-lots-json:

$ time gen/bin/example-jsonptr < citylots.json > /dev/null
Before this commit:
real    0m1.799s
After:
real    0m1.765s
Ratio: 1.02x
2020-08-21 14:11:31 +10:00
Nigel Tao 1adbc9547f Split std/json's QUIRK_ALLOW_BACKSLASH_X in two 2020-08-06 23:28:07 +10:00
Nigel Tao 1e96ef200c Add test/data/cbor-rfc-7049-examples.* 2020-07-27 22:00:28 +10:00
Nigel Tao 0054a053f5 Add comment link to github.com/cbor/test-vectors 2020-07-24 22:47:39 +10:00
Nigel Tao 271680333e Have std/cbor parse tags 2020-07-24 15:54:44 +10:00
Nigel Tao c9d4e34dfc Split ETC__VBC__INLINE_INTEGER into {,un}signed 2020-07-21 15:31:49 +10:00
Nigel Tao 4e193599ac Add std/cbor 2020-07-15 14:08:37 +10:00
Nigel Tao fcb6416c16 Add script/extract-cbor-rfc-7049-examples.go 2020-07-10 22:56:47 +10:00
Nigel Tao 27766aedf9 Let process-json-numbers use fast_double_parser 2020-07-09 10:59:54 +10:00
Nigel Tao 6b7ce30e32 Add wuffs_base__parse_number_xxx options arg 2020-07-07 16:19:46 +10:00
Nigel Tao 86093c8194 Have parse_number_f64 use the Eisel algorithm
The algorithm is discussed at
https://lemire.me/blog/2020/03/10/fast-float-parsing-in-practice/

----

Binary size, before:
27848 gen/lib/c/clang-5.0-dynamic/wuffs-base-f64conv.lo
28040 gen/lib/c/clang-5.0-static/wuffs-base-f64conv.o
30392 gen/lib/c/gcc-dynamic/wuffs-base-f64conv.lo
30392 gen/lib/c/gcc-static/wuffs-base-f64conv.o

After:
29416 gen/lib/c/clang-5.0-dynamic/wuffs-base-f64conv.lo
29688 gen/lib/c/clang-5.0-static/wuffs-base-f64conv.o
31792 gen/lib/c/gcc-dynamic/wuffs-base-f64conv.lo
31792 gen/lib/c/gcc-static/wuffs-base-f64conv.o

The binary size difference does not account for the powers-of-10 table,
which was expanded in a previous commit in anticipation of this one. A
future commit will also remove some of the other f64 parser code used
before implementing the Eisel algorithm.

----

Processing citylots.json from github.com/zemirco/sf-city-lots-json
$ g++ -O3 script/process-json-numbers.c
$ time ./a.out -parse-number-f64 < citylots.json
Before/After:
real    0m1.231s
real	0m1.031s

The difference with and without -parse-number-f64 measures how much time
is spent in the parse_number_f64 function. Before: 424ms, after: 224ms.
$ time ./a.out                   < citylots.json
real	0m0.807s

Similarly, for canada.json from github.com/miloyip/nativejson-benchmark
$ time ./a.out -parse-number-f64 < canada.json
Before/After:
real	0m0.023s
real	0m0.017s
$ time ./a.out                   < canada.json
real	0m0.013s
Time spent in parse_number_f64, before: 10ms, after: 4ms.

----

name                                                old time/op  new time/op   delta
wuffs_strconv_parse_number_f64_1_lsh53_add0/clang5  77.0ns ± 0%   37.0ns ± 0%  -51.95%  (p=0.029 n=4+4)
wuffs_strconv_parse_number_f64_1_lsh53_add1/clang5   690ns ± 0%    725ns ± 2%   +4.95%  (p=0.008 n=5+5)
wuffs_strconv_parse_number_f64_pi_long/clang5       92.0ns ± 0%  136.0ns ± 2%  +47.83%  (p=0.016 n=4+5)
wuffs_strconv_parse_number_f64_pi_short/clang5      35.0ns ± 3%   18.0ns ± 0%  -48.57%  (p=0.008 n=5+5)
wuffs_strconv_parse_number_f64_1_lsh53_add0/gcc7    67.0ns ± 0%   34.0ns ± 0%  -49.25%  (p=0.000 n=5+4)
wuffs_strconv_parse_number_f64_1_lsh53_add1/gcc7     712ns ± 0%    760ns ± 2%   +6.72%  (p=0.008 n=5+5)
wuffs_strconv_parse_number_f64_pi_long/gcc7         88.4ns ± 1%  116.2ns ± 3%  +31.45%  (p=0.008 n=5+5)
wuffs_strconv_parse_number_f64_pi_short/gcc7        32.6ns ± 2%   21.4ns ± 7%  -34.36%  (p=0.008 n=5+5)
2020-07-07 15:27:51 +10:00