Nigel Tao
85a0283cfc
base: tweak swizzle_ycck ycc_upsampling parameter
...
This moves from a 2-valued "basic (box) or fancy (triangle)" upsampling
parameter to a 3-valued one, which can now represent the fact that
libjpeg's and libwebp's fancy (triangle) upsampling is ever-so-slightly
different (see script/script/replicate-libwebp-chroma-upsampling.go and
its comments).
This is not an optimization or pessimization per se, but for the record,
the performance impact is mixed.
name old speed new speed delta
wuffs_jpeg_decode_19k_8bpp/clang14 317MB/s ± 0% 314MB/s ± 0% -0.99% (p=0.008 n=5+5)
wuffs_jpeg_decode_30k_24bpp_progressive/clang14 318MB/s ± 0% 328MB/s ± 0% +3.13% (p=0.008 n=5+5)
wuffs_jpeg_decode_30k_24bpp_sequential/clang14 879MB/s ± 0% 871MB/s ± 0% -0.83% (p=0.008 n=5+5)
wuffs_jpeg_decode_77k_24bpp/clang14 719MB/s ± 0% 713MB/s ± 0% -0.89% (p=0.008 n=5+5)
wuffs_jpeg_decode_552k_24bpp_420/clang14 779MB/s ± 0% 771MB/s ± 0% -1.01% (p=0.008 n=5+5)
wuffs_jpeg_decode_552k_24bpp_444/clang14 902MB/s ± 0% 890MB/s ± 0% -1.32% (p=0.008 n=5+5)
wuffs_jpeg_decode_4002k_24bpp/clang14 784MB/s ± 0% 779MB/s ± 0% -0.66% (p=0.008 n=5+5)
wuffs_jpeg_decode_19k_8bpp/gcc12 358MB/s ± 0% 362MB/s ± 0% +0.93% (p=0.008 n=5+5)
wuffs_jpeg_decode_30k_24bpp_progressive/gcc12 305MB/s ± 0% 311MB/s ± 0% +1.91% (p=0.008 n=5+5)
wuffs_jpeg_decode_30k_24bpp_sequential/gcc12 912MB/s ± 0% 918MB/s ± 0% +0.73% (p=0.016 n=4+5)
wuffs_jpeg_decode_77k_24bpp/gcc12 749MB/s ± 0% 763MB/s ± 0% +1.83% (p=0.008 n=5+5)
wuffs_jpeg_decode_552k_24bpp_420/gcc12 813MB/s ± 0% 824MB/s ± 0% +1.37% (p=0.008 n=5+5)
wuffs_jpeg_decode_552k_24bpp_444/gcc12 969MB/s ± 0% 984MB/s ± 0% +1.62% (p=0.008 n=5+5)
wuffs_jpeg_decode_4002k_24bpp/gcc12 815MB/s ± 0% 832MB/s ± 0% +2.06% (p=0.008 n=5+5)
2026-07-03 21:38:28 +10:00
Nigel Tao
9a027b8016
base: add WUFFS_BASE__YCC_MODEL__ETC
2026-06-20 18:19:45 +10:00
Nigel Tao
53b3b0a9d0
std/jpeg: fix default-quality workbuf length check
2024-09-08 11:29:01 +10:00
Nigel Tao
e12f10eef0
std/jpeg: tweak ignore too-short marker code
...
name old speed new speed delta
wuffs_jpeg_decode_19k_8bpp/clang14 317MB/s ± 0% 325MB/s ± 0% +2.60% (p=0.008 n=5+5)
wuffs_jpeg_decode_30k_24bpp_progressive/clang14 322MB/s ± 0% 325MB/s ± 0% +0.94% (p=0.008 n=5+5)
wuffs_jpeg_decode_30k_24bpp_sequential/clang14 864MB/s ± 0% 880MB/s ± 0% +1.84% (p=0.008 n=5+5)
wuffs_jpeg_decode_77k_24bpp/clang14 720MB/s ± 0% 730MB/s ± 0% +1.39% (p=0.016 n=4+5)
wuffs_jpeg_decode_552k_24bpp_420/clang14 783MB/s ± 0% 794MB/s ± 0% +1.44% (p=0.008 n=5+5)
wuffs_jpeg_decode_552k_24bpp_444/clang14 884MB/s ± 0% 914MB/s ± 0% +3.39% (p=0.008 n=5+5)
wuffs_jpeg_decode_4002k_24bpp/clang14 781MB/s ± 0% 796MB/s ± 0% +1.90% (p=0.008 n=5+5)
wuffs_jpeg_decode_19k_8bpp/gcc12 359MB/s ± 0% 360MB/s ± 0% +0.12% (p=0.008 n=5+5)
wuffs_jpeg_decode_30k_24bpp_progressive/gcc12 311MB/s ± 0% 310MB/s ± 0% -0.30% (p=0.008 n=5+5)
wuffs_jpeg_decode_30k_24bpp_sequential/gcc12 909MB/s ± 0% 915MB/s ± 0% +0.66% (p=0.008 n=5+5)
wuffs_jpeg_decode_77k_24bpp/gcc12 747MB/s ± 0% 750MB/s ± 0% +0.34% (p=0.008 n=5+5)
wuffs_jpeg_decode_552k_24bpp_420/gcc12 817MB/s ± 0% 819MB/s ± 0% +0.30% (p=0.008 n=5+5)
wuffs_jpeg_decode_552k_24bpp_444/gcc12 971MB/s ± 0% 972MB/s ± 0% +0.11% (p=0.008 n=5+5)
wuffs_jpeg_decode_4002k_24bpp/gcc12 819MB/s ± 0% 821MB/s ± 0% +0.21% (p=0.008 n=5+5)
2024-08-25 23:03:06 +10:00
Nigel Tao
6693250aa9
lang: add io_reader.peek_u8_at
2024-08-25 15:07:29 +10:00
Nigel Tao
d4bab677b0
std/jpeg: skip 0xFF filler before restart search
2024-08-25 14:48:17 +10:00
Nigel Tao
6a9347843b
std/jpeg: ignore too-short COM or APPn markers
2024-08-25 13:46:16 +10:00
Nigel Tao
819742cfac
std/jpeg: add "#short SOS bitstream"
2024-08-24 17:01:28 +10:00
Nigel Tao
a824c55c3d
std/jpeg: allow DHT markers before SOF
2024-08-24 16:13:31 +10:00
Nigel Tao
a07eb7704b
std/jpeg: add QUIRK_REJECT_PROGRESSIVE_JPEGS
2024-08-11 10:15:13 +10:00
Nigel Tao
d270839e3f
std/jpeg: let use_lower_quality work with CMYK
2024-08-04 23:16:58 +10:00
Nigel Tao
c9fff5b207
std/jpeg: add and use WUFFS_BASE__QUIRK_QUALITY
2024-08-04 23:15:58 +10:00
Nigel Tao
42b070e487
std/jpeg: halve maximum scan count from 64 to 32
...
cjpeg defaults to using 1 (sequential) or 10 (progressive) scans. 32 is
still easily more than what's seen in practice.
The previous commit (f9c384d6 std/jpeg: add "expect multiple scans"
check) makes Wuffs more closely match libjpeg. Without that commit,
though, this commit halves the time taken to decode the oss-fuzz test
case (linked below) from 6 to 3 seconds, in an optimized build. For the
debug-sanitizer fuzzer build, without this commit, that test case times
out after 60 seconds but this commit should halve the time taken.
Updates https://bugs.chromium.org/p/oss-fuzz/issues/detail?id=70340
2024-07-24 22:09:25 +10:00
Nigel Tao
f9c384d6b2
std/jpeg: add "expect multiple scans" check
...
Updates https://bugs.chromium.org/p/oss-fuzz/issues/detail?id=70340
2024-07-23 21:53:32 +10:00
Nigel Tao
aae5997e12
lang: remove endwhile keyword
2024-07-08 17:15:48 +10:00
Nigel Tao
c77989d162
std: rename "var c" to "var c8"
2023-12-27 17:01:01 +11:00
Nigel Tao
eed1af7bee
base: rename history_retain_length
...
It was renamed to dst_history_retain_length.
It also was removed from image_decoders and token_decoders. It only
applies to io_transformers, as retaining history applies to the
destination (not source) IO buffer.
Also adjust example/bzcat and example/zcat to apply the history retain
length to the dst (not src) buffer.
2023-12-17 12:26:08 +11:00
Nigel Tao
82ea1985b4
base: add optional_u63 type
2023-11-19 12:02:24 +11:00
Patrick Georgi
f563f32f82
Add SPDX identifier to license headers
...
SPDX is an ISO standard for describing licenses in a computer readable
format, which is useful when applying license compliance tools and the
like to source code.
In addition to ensuring that the Wuffs C distribution is properly
annotated to the benefit of projects using Wuffs and SPDX-supporting
tools, add it everywhere else where there's a license in this repo for
consistency, too.
2023-10-06 20:52:36 +02:00
Nigel Tao
74a6ea2fcd
all: relicense from Apache2 to dual Apache2/MIT
...
The LICENSE has changed from a single license (Apache2) to a dual
license (Apache2 or MIT, at your option).
If you've previously used Wuffs under the Apache2 license, you are able
to keep doing so. No further action required by you.
The two files listed below (previous releases) are exceptions, as they
are older snapshots, frozen in time. Their license has not changed:
release/c/wuffs-v0.2.c
release/c/wuffs-v0.3.c
Every name in the AUTHORS file, other than Google Inc., has signed
Google's CLA (Contributor License Agreement).
2023-10-06 10:00:59 +11:00
Nigel Tao
5fb82e7d0a
std: add compact_retaining, history_retain_length
2023-10-05 13:24:44 +11:00
Nigel Tao
0f91f23bb6
std/jpeg: prefer decode_frame errors over swizzle
...
"g++ script/manual-test-truncated-input.cc && ./a.out test/data/*" is
now happy.
2023-10-04 22:27:41 +11:00
Nigel Tao
186bdc5fe2
Roll back 'decode truncated... on MCU boundaries'
...
See the previous commit for the rationale.
2023-09-16 20:53:15 +10:00
Nigel Tao
a3cecef69a
std/jpeg: decode truncated JPEGs on MCU boundaries
...
This only affects truncated (malformed) JPEGs. It moves Wuffs' std/jpeg
decoder closer to libjpeg-turbo, as seen by the "hash of the decoded
output" for test/data/peacock.default.truncated.jpeg now matching
test/data/peacock.default.truncated.decoded-by-djpeg.png instead of
test/data/peacock.default.truncated.decoded-by-wuffs.png
The decoding of test/data/peacock.progressive.truncated.jpeg is not yet
affected, requiring further investigation, but a future commit could
similarly make that match djpeg.
The two libraries don't use exactly the same algorithm because, for
performance, Wuffs' decoder reads from the bitstream 8 bytes at a time
where libjpeg-turbo reads only 1 byte at a time.
While this change moves Wuffs' output closer to libjpeg-turbo (to
repeat, only for truncated JPEGs; well-formed JPEGs are unaffected), the
benchmarks below show a negative performance impact, and visual
inspection of test/data/peacock.*.truncated.decoded-by-*.png shows that
matching libjpeg-turbo *exactly* on truncated JPEGs doesn't actually
always look better.
On balance, it doesn't seem worth it to keep this change (and spend more
time on that further investigation).
This commit will soon be followed by a rollback, but it is committed
anyway so that we can refer to these numbers in the git log.
name old speed new speed delta
wuffs_jpeg_decode_19k_8bpp/clang11 378MB/s ± 0% 364MB/s ± 0% -3.53% (p=0.008 n=5+5)
wuffs_jpeg_decode_30k_24bpp_progressive/clang11 325MB/s ± 0% 340MB/s ± 0% +4.62% (p=0.016 n=5+4)
wuffs_jpeg_decode_30k_24bpp_sequential/clang11 1.02GB/s ± 0% 0.99GB/s ± 0% -2.64% (p=0.008 n=5+5)
wuffs_jpeg_decode_77k_24bpp/clang11 842MB/s ± 0% 815MB/s ± 0% -3.16% (p=0.008 n=5+5)
wuffs_jpeg_decode_552k_24bpp_420/clang11 916MB/s ± 0% 887MB/s ± 0% -3.08% (p=0.008 n=5+5)
wuffs_jpeg_decode_552k_24bpp_444/clang11 1.10GB/s ± 0% 1.07GB/s ± 2% -3.39% (p=0.008 n=5+5)
wuffs_jpeg_decode_4002k_24bpp/clang11 919MB/s ± 0% 892MB/s ± 0% -2.95% (p=0.008 n=5+5)
wuffs_jpeg_decode_19k_8bpp/gcc10 360MB/s ± 0% 348MB/s ± 0% -3.34% (p=0.008 n=5+5)
wuffs_jpeg_decode_30k_24bpp_progressive/gcc10 290MB/s ± 0% 283MB/s ± 0% -2.46% (p=0.008 n=5+5)
wuffs_jpeg_decode_30k_24bpp_sequential/gcc10 925MB/s ± 0% 909MB/s ± 0% -1.70% (p=0.008 n=5+5)
wuffs_jpeg_decode_77k_24bpp/gcc10 758MB/s ± 0% 732MB/s ± 0% -3.43% (p=0.008 n=5+5)
wuffs_jpeg_decode_552k_24bpp_420/gcc10 829MB/s ± 0% 799MB/s ± 0% -3.54% (p=0.008 n=5+5)
wuffs_jpeg_decode_552k_24bpp_444/gcc10 981MB/s ± 0% 964MB/s ± 0% -1.72% (p=0.008 n=5+5)
wuffs_jpeg_decode_4002k_24bpp/gcc10 835MB/s ± 0% 811MB/s ± 0% -2.89% (p=0.008 n=5+5)
2023-09-16 20:49:46 +10:00
Nigel Tao
d89ec2d40b
std/jpeg: drain bitstream for truncated input
...
name old speed new speed delta
wuffs_jpeg_decode_19k_8bpp/clang11 367MB/s ± 0% 378MB/s ± 0% +2.91% (p=0.008 n=5+5)
wuffs_jpeg_decode_30k_24bpp_progressive/clang11 333MB/s ± 0% 325MB/s ± 0% -2.46% (p=0.008 n=5+5)
wuffs_jpeg_decode_30k_24bpp_sequential/clang11 1.00GB/s ± 0% 1.02GB/s ± 0% +1.37% (p=0.008 n=5+5)
wuffs_jpeg_decode_77k_24bpp/clang11 822MB/s ± 0% 842MB/s ± 0% +2.39% (p=0.008 n=5+5)
wuffs_jpeg_decode_552k_24bpp_420/clang11 898MB/s ± 0% 916MB/s ± 0% +1.98% (p=0.008 n=5+5)
wuffs_jpeg_decode_552k_24bpp_444/clang11 1.09GB/s ± 0% 1.10GB/s ± 0% +1.21% (p=0.008 n=5+5)
wuffs_jpeg_decode_4002k_24bpp/clang11 898MB/s ± 0% 919MB/s ± 0% +2.29% (p=0.008 n=5+5)
wuffs_jpeg_decode_19k_8bpp/gcc10 362MB/s ± 0% 360MB/s ± 0% -0.55% (p=0.008 n=5+5)
wuffs_jpeg_decode_30k_24bpp_progressive/gcc10 292MB/s ± 0% 290MB/s ± 0% -0.57% (p=0.008 n=5+5)
wuffs_jpeg_decode_30k_24bpp_sequential/gcc10 934MB/s ± 0% 925MB/s ± 0% -1.03% (p=0.008 n=5+5)
wuffs_jpeg_decode_77k_24bpp/gcc10 759MB/s ± 0% 758MB/s ± 0% -0.25% (p=0.008 n=5+5)
wuffs_jpeg_decode_552k_24bpp_420/gcc10 828MB/s ± 0% 829MB/s ± 0% ~ (p=0.222 n=5+5)
wuffs_jpeg_decode_552k_24bpp_444/gcc10 1.00GB/s ± 0% 0.98GB/s ± 0% -1.44% (p=0.008 n=5+5)
wuffs_jpeg_decode_4002k_24bpp/gcc10 837MB/s ± 0% 835MB/s ± 0% -0.29% (p=0.008 n=5+5)
2023-08-28 19:35:08 +10:00
Nigel Tao
dfe5bcd36e
std/jpeg: decode something for truncated input
2023-08-27 10:15:40 +10:00
Nigel Tao
80e847dceb
std/jpeg: stash mcu_blocks during progressive IDCT
2023-08-25 23:03:06 +10:00
Nigel Tao
4417dcea4e
std/jpeg: use default Huffman tables if missing
2023-08-24 16:18:11 +10:00
Nigel Tao
72b9c23120
std/jpeg: decode RGB (not just YCbCr) JPEGs
...
The benchmarks don't exercise the RGB code paths per se, but it does
affect the numbers. Especially for gcc, this essentially undoes the
performance impact of the previous (trivial) commit. This is all
presumably another code alignment coincidence per
https://news.ycombinator.com/item?id=23633583
name old speed new speed delta
wuffs_jpeg_decode_19k_8bpp/clang11 376MB/s ± 0% 376MB/s ± 0% ~ (p=1.000 n=5+5)
wuffs_jpeg_decode_30k_24bpp_progressive/clang11 338MB/s ± 0% 341MB/s ± 0% +0.98% (p=0.008 n=5+5)
wuffs_jpeg_decode_30k_24bpp_sequential/clang11 1.01GB/s ± 0% 1.01GB/s ± 0% +0.35% (p=0.008 n=5+5)
wuffs_jpeg_decode_77k_24bpp/clang11 841MB/s ± 0% 836MB/s ± 1% -0.65% (p=0.008 n=5+5)
wuffs_jpeg_decode_552k_24bpp_420/clang11 913MB/s ± 0% 912MB/s ± 0% -0.18% (p=0.016 n=5+5)
wuffs_jpeg_decode_552k_24bpp_444/clang11 1.11GB/s ± 0% 1.10GB/s ± 0% -0.34% (p=0.008 n=5+5)
wuffs_jpeg_decode_4002k_24bpp/clang11 920MB/s ± 0% 916MB/s ± 0% -0.42% (p=0.008 n=5+5)
wuffs_jpeg_decode_19k_8bpp/gcc10 349MB/s ± 0% 368MB/s ± 0% +5.60% (p=0.008 n=5+5)
wuffs_jpeg_decode_30k_24bpp_progressive/gcc10 328MB/s ± 0% 279MB/s ± 0% -14.82% (p=0.008 n=5+5)
wuffs_jpeg_decode_30k_24bpp_sequential/gcc10 925MB/s ± 0% 947MB/s ± 1% +2.33% (p=0.008 n=5+5)
wuffs_jpeg_decode_77k_24bpp/gcc10 732MB/s ± 0% 770MB/s ± 0% +5.23% (p=0.008 n=5+5)
wuffs_jpeg_decode_552k_24bpp_420/gcc10 798MB/s ± 0% 838MB/s ± 0% +4.99% (p=0.008 n=5+5)
wuffs_jpeg_decode_552k_24bpp_444/gcc10 962MB/s ± 0% 996MB/s ± 0% +3.54% (p=0.008 n=5+5)
wuffs_jpeg_decode_4002k_24bpp/gcc10 806MB/s ± 0% 847MB/s ± 0% +5.19% (p=0.008 n=5+5)
2023-08-21 16:12:21 +10:00
Nigel Tao
14a8bd0feb
std/jpeg: allow fewer-components sequential scans
...
name old speed new speed delta
wuffs_jpeg_decode_19k_8bpp/clang11 366MB/s ± 0% 369MB/s ± 0% +0.81% (p=0.008 n=5+5)
wuffs_jpeg_decode_30k_24bpp_progressive/clang11 302MB/s ± 0% 309MB/s ± 0% +2.28% (p=0.008 n=5+5)
wuffs_jpeg_decode_30k_24bpp_sequential/clang11 1.00GB/s ± 0% 1.00GB/s ± 0% -0.58% (p=0.008 n=5+5)
wuffs_jpeg_decode_77k_24bpp/clang11 820MB/s ± 0% 827MB/s ± 0% +0.80% (p=0.008 n=5+5)
wuffs_jpeg_decode_552k_24bpp_420/clang11 895MB/s ± 0% 897MB/s ± 0% +0.24% (p=0.008 n=5+5)
wuffs_jpeg_decode_552k_24bpp_444/clang11 1.11GB/s ± 0% 1.10GB/s ± 0% -0.30% (p=0.008 n=5+5)
wuffs_jpeg_decode_4002k_24bpp/clang11 897MB/s ± 0% 900MB/s ± 0% +0.30% (p=0.008 n=5+5)
wuffs_jpeg_decode_19k_8bpp/gcc10 370MB/s ± 0% 369MB/s ± 0% -0.26% (p=0.008 n=5+5)
wuffs_jpeg_decode_30k_24bpp_progressive/gcc10 309MB/s ± 0% 283MB/s ± 0% -8.35% (p=0.008 n=5+5)
wuffs_jpeg_decode_30k_24bpp_sequential/gcc10 948MB/s ± 0% 945MB/s ± 0% -0.27% (p=0.008 n=5+5)
wuffs_jpeg_decode_77k_24bpp/gcc10 769MB/s ± 0% 769MB/s ± 0% ~ (p=0.310 n=5+5)
wuffs_jpeg_decode_552k_24bpp_420/gcc10 841MB/s ± 0% 842MB/s ± 0% +0.13% (p=0.016 n=4+5)
wuffs_jpeg_decode_552k_24bpp_444/gcc10 1.00GB/s ± 0% 1.01GB/s ± 0% +0.75% (p=0.016 n=4+5)
wuffs_jpeg_decode_4002k_24bpp/gcc10 846MB/s ± 0% 848MB/s ± 0% +0.22% (p=0.008 n=5+5)
2023-08-19 17:22:47 +10:00
Nigel Tao
41d0c095c8
std/jpeg: implement block smoothing
...
"script/confirm-matches-djpeg.sh test/data/*.jpeg" is now all matches.
name old speed new speed delta
wuffs_jpeg_decode_19k_8bpp/clang11 369MB/s ± 0% 366MB/s ± 0% -0.79% (p=0.008 n=5+5)
wuffs_jpeg_decode_30k_24bpp_progressive/clang11 335MB/s ± 0% 302MB/s ± 0% -9.93% (p=0.008 n=5+5)
wuffs_jpeg_decode_30k_24bpp_sequential/clang11 1.01GB/s ± 0% 1.00GB/s ± 0% -1.07% (p=0.008 n=5+5)
wuffs_jpeg_decode_77k_24bpp/clang11 829MB/s ± 0% 820MB/s ± 0% -1.03% (p=0.008 n=5+5)
wuffs_jpeg_decode_552k_24bpp_420/clang11 903MB/s ± 0% 895MB/s ± 0% -0.93% (p=0.008 n=5+5)
wuffs_jpeg_decode_552k_24bpp_444/clang11 1.12GB/s ± 0% 1.11GB/s ± 0% -1.28% (p=0.008 n=5+5)
wuffs_jpeg_decode_4002k_24bpp/clang11 906MB/s ± 0% 897MB/s ± 0% -0.93% (p=0.008 n=5+5)
wuffs_jpeg_decode_19k_8bpp/gcc10 350MB/s ± 0% 370MB/s ± 0% +5.54% (p=0.016 n=4+5)
wuffs_jpeg_decode_30k_24bpp_progressive/gcc10 334MB/s ± 0% 309MB/s ± 0% -7.46% (p=0.008 n=5+5)
wuffs_jpeg_decode_30k_24bpp_sequential/gcc10 924MB/s ± 0% 948MB/s ± 0% +2.61% (p=0.008 n=5+5)
wuffs_jpeg_decode_77k_24bpp/gcc10 745MB/s ± 0% 769MB/s ± 0% +3.25% (p=0.008 n=5+5)
wuffs_jpeg_decode_552k_24bpp_420/gcc10 813MB/s ± 0% 841MB/s ± 0% +3.41% (p=0.029 n=4+4)
wuffs_jpeg_decode_552k_24bpp_444/gcc10 1.00GB/s ± 0% 1.00GB/s ± 0% +0.25% (p=0.029 n=4+4)
wuffs_jpeg_decode_4002k_24bpp/gcc10 821MB/s ± 0% 846MB/s ± 0% +3.00% (p=0.008 n=5+5)
2023-08-19 17:16:51 +10:00
Nigel Tao
2bff962cf1
std/jpeg: add decoder.decode_idct_x86_avx2
...
name old speed new speed delta
wuffs_jpeg_decode_19k_8bpp/clang11 237MB/s ± 0% 370MB/s ± 0% +55.94% (p=0.008 n=5+5)
wuffs_jpeg_decode_30k_24bpp_progressive/clang11 205MB/s ± 0% 231MB/s ± 0% +12.80% (p=0.008 n=5+5)
wuffs_jpeg_decode_30k_24bpp_sequential/clang11 379MB/s ± 0% 470MB/s ± 0% +23.92% (p=0.008 n=5+5)
wuffs_jpeg_decode_77k_24bpp/clang11 332MB/s ± 0% 418MB/s ± 0% +25.95% (p=0.008 n=5+5)
wuffs_jpeg_decode_552k_24bpp_420/clang11 345MB/s ± 0% 435MB/s ± 0% +25.99% (p=0.008 n=5+5)
wuffs_jpeg_decode_552k_24bpp_444/clang11 483MB/s ± 0% 620MB/s ± 0% +28.53% (p=0.008 n=5+5)
wuffs_jpeg_decode_4002k_24bpp/clang11 349MB/s ± 0% 441MB/s ± 0% +26.47% (p=0.008 n=5+5)
wuffs_jpeg_decode_19k_8bpp/gcc10 231MB/s ± 0% 352MB/s ± 0% +52.78% (p=0.008 n=5+5)
wuffs_jpeg_decode_30k_24bpp_progressive/gcc10 227MB/s ± 0% 296MB/s ± 0% +30.44% (p=0.008 n=5+5)
wuffs_jpeg_decode_30k_24bpp_sequential/gcc10 500MB/s ± 0% 666MB/s ± 0% +33.21% (p=0.008 n=5+5)
wuffs_jpeg_decode_77k_24bpp/gcc10 417MB/s ± 0% 565MB/s ± 0% +35.32% (p=0.008 n=5+5)
wuffs_jpeg_decode_552k_24bpp_420/gcc10 440MB/s ± 0% 608MB/s ± 0% +38.37% (p=0.008 n=5+5)
wuffs_jpeg_decode_552k_24bpp_444/gcc10 478MB/s ± 0% 624MB/s ± 0% +30.70% (p=0.008 n=5+5)
wuffs_jpeg_decode_4002k_24bpp/gcc10 445MB/s ± 0% 616MB/s ± 0% +38.42% (p=0.008 n=5+5)
2023-07-18 09:17:08 +10:00
Nigel Tao
0af626b850
std/jpeg: grow quant_tables' elements, s/u8/u16/
2023-07-17 10:27:49 +10:00
Nigel Tao
4dfd70928b
std: add "get_quirk(key: u32) u64"
2023-07-16 12:04:09 +10:00
Nigel Tao
49f50cca91
std/jpeg: cap the implicit-zero bitstream padding
...
Updates https://bugs.chromium.org/p/oss-fuzz/issues/detail?id=59966
name old speed new speed delta
wuffs_jpeg_decode_19k_8bpp/clang11 241MB/s ± 0% 237MB/s ± 0% -1.57% (p=0.008 n=5+5)
wuffs_jpeg_decode_30k_24bpp_progressive/clang11 213MB/s ± 0% 212MB/s ± 0% -0.43% (p=0.016 n=5+4)
wuffs_jpeg_decode_30k_24bpp_sequential/clang11 387MB/s ± 0% 386MB/s ± 0% -0.08% (p=0.008 n=5+5)
wuffs_jpeg_decode_77k_24bpp/clang11 341MB/s ± 0% 340MB/s ± 0% -0.50% (p=0.016 n=4+5)
wuffs_jpeg_decode_552k_24bpp_420/clang11 356MB/s ± 0% 353MB/s ± 0% -0.89% (p=0.008 n=5+5)
wuffs_jpeg_decode_552k_24bpp_444/clang11 484MB/s ± 0% 480MB/s ± 0% -0.96% (p=0.008 n=5+5)
wuffs_jpeg_decode_4002k_24bpp/clang11 364MB/s ± 0% 360MB/s ± 0% -1.12% (p=0.008 n=5+5)
wuffs_jpeg_decode_19k_8bpp/gcc10 244MB/s ± 0% 247MB/s ± 0% +0.88% (p=0.008 n=5+5)
wuffs_jpeg_decode_30k_24bpp_progressive/gcc10 254MB/s ± 0% 254MB/s ± 0% -0.20% (p=0.008 n=5+5)
wuffs_jpeg_decode_30k_24bpp_sequential/gcc10 508MB/s ± 0% 512MB/s ± 0% +0.61% (p=0.008 n=5+5)
wuffs_jpeg_decode_77k_24bpp/gcc10 434MB/s ± 0% 436MB/s ± 0% +0.39% (p=0.008 n=5+5)
wuffs_jpeg_decode_552k_24bpp_420/gcc10 456MB/s ± 0% 458MB/s ± 0% +0.41% (p=0.016 n=4+5)
wuffs_jpeg_decode_552k_24bpp_444/gcc10 481MB/s ± 0% 483MB/s ± 0% +0.49% (p=0.008 n=5+5)
wuffs_jpeg_decode_4002k_24bpp/gcc10 463MB/s ± 0% 466MB/s ± 0% +0.57% (p=0.016 n=4+5)
2023-06-23 21:38:01 +10:00
Nigel Tao
8d9d6c1292
Performance mystery: load-bearing "no-op continue"
...
There is no semantic difference in this commit. There's no further code
after the if-else chain in the loop body. Nonetheless, adding a "no-op
continue" impacts performance: bad for clang and good for gcc.
Profiling suggests we spend less than 5% overall in
wuffs_jpeg__decoder__fill_bitstream.
name old speed new speed delta
wuffs_jpeg_decode_19k_8bpp/clang11 241MB/s ± 0% 229MB/s ± 0% -5.02% (p=0.008 n=5+5)
wuffs_jpeg_decode_30k_24bpp_progressive/clang11 213MB/s ± 0% 205MB/s ± 0% -4.13% (p=0.008 n=5+5)
wuffs_jpeg_decode_30k_24bpp_sequential/clang11 386MB/s ± 0% 375MB/s ± 0% -2.88% (p=0.008 n=5+5)
wuffs_jpeg_decode_77k_24bpp/clang11 341MB/s ± 0% 326MB/s ± 0% -4.58% (p=0.008 n=5+5)
wuffs_jpeg_decode_552k_24bpp_420/clang11 356MB/s ± 0% 339MB/s ± 0% -4.90% (p=0.008 n=5+5)
wuffs_jpeg_decode_552k_24bpp_444/clang11 484MB/s ± 0% 475MB/s ± 0% -1.96% (p=0.008 n=5+5)
wuffs_jpeg_decode_4002k_24bpp/clang11 364MB/s ± 0% 343MB/s ± 0% -5.83% (p=0.008 n=5+5)
wuffs_jpeg_decode_19k_8bpp/gcc10 234MB/s ± 0% 245MB/s ± 0% +4.68% (p=0.008 n=5+5)
wuffs_jpeg_decode_30k_24bpp_progressive/gcc10 245MB/s ± 0% 254MB/s ± 0% +3.35% (p=0.008 n=5+5)
wuffs_jpeg_decode_30k_24bpp_sequential/gcc10 500MB/s ± 0% 514MB/s ± 0% +2.81% (p=0.008 n=5+5)
wuffs_jpeg_decode_77k_24bpp/gcc10 417MB/s ± 0% 434MB/s ± 0% +4.18% (p=0.008 n=5+5)
wuffs_jpeg_decode_552k_24bpp_420/gcc10 442MB/s ± 0% 459MB/s ± 0% +3.76% (p=0.008 n=5+5)
wuffs_jpeg_decode_552k_24bpp_444/gcc10 477MB/s ± 0% 486MB/s ± 0% +1.99% (p=0.008 n=5+5)
wuffs_jpeg_decode_4002k_24bpp/gcc10 450MB/s ± 0% 467MB/s ± 0% +3.70% (p=0.008 n=5+5)
It is a mystery why this happens. Note though that the "continue" in the
Wuffs code becomes a "goto" in the generated C code, because (in the
general case) Wuffs has "labeled continue/break", which can jump out of
a not-inner-most loop, and C99 does not. Wuffs' "generate C code" thus
uses a C "goto". A follow-up commit may try a C "continue" instead.
To reproduce:
----
$ git clone https://github.com/google/wuffs.git
$ cd wuffs
$ gcc --version | head -n 1
gcc (Debian 10.2.1-6) 10.2.1 20210110
$ # Check out this commit.
$ git checkout --quiet THE_HASH_OF_THIS_COMMIT
$ gcc -O3 -Wall -std=c99 test/c/std/jpeg.c -DWUFFS_MIMIC -ljpeg
$ ./a.out -bench -focus=wuffs_jpeg_decode_77k_24bpp | grep MB/s
Benchmarkwuffs_jpeg_decode_77k_24bpp/gcc10 3000 176988 ns/op 433.927 MB/s
Benchmarkwuffs_jpeg_decode_77k_24bpp/gcc10 3000 176849 ns/op 434.268 MB/s
Benchmarkwuffs_jpeg_decode_77k_24bpp/gcc10 3000 176800 ns/op 434.389 MB/s
Benchmarkwuffs_jpeg_decode_77k_24bpp/gcc10 3000 176844 ns/op 434.279 MB/s
Benchmarkwuffs_jpeg_decode_77k_24bpp/gcc10 3000 176869 ns/op 434.218 MB/s
$ # Check out the parent commit.
$ git checkout --quiet THE_HASH_OF_THIS_COMMIT^
$ gcc -O3 -Wall -std=c99 test/c/std/jpeg.c -DWUFFS_MIMIC -ljpeg
$ ./a.out -bench -focus=wuffs_jpeg_decode_77k_24bpp | grep MB/s
Benchmarkwuffs_jpeg_decode_77k_24bpp/gcc10 3000 183848 ns/op 417.734 MB/s
Benchmarkwuffs_jpeg_decode_77k_24bpp/gcc10 3000 183870 ns/op 417.684 MB/s
Benchmarkwuffs_jpeg_decode_77k_24bpp/gcc10 3000 183833 ns/op 417.770 MB/s
Benchmarkwuffs_jpeg_decode_77k_24bpp/gcc10 3000 183873 ns/op 417.679 MB/s
Benchmarkwuffs_jpeg_decode_77k_24bpp/gcc10 3000 183843 ns/op 417.746 MB/s
----
2023-06-23 17:02:14 +10:00
Nigel Tao
40fca0f03d
std/jpeg: roll back fill_bitstream inner loop
...
Commit 723c604a "std/jpeg: tweak decoder.fill_bitstream inner loop"
inadvertently changed "increment wi" behavior in the rare code paths.
name old speed new speed delta
wuffs_jpeg_decode_19k_8bpp/clang11 237MB/s ± 0% 241MB/s ± 0% +1.79% (p=0.008 n=5+5)
wuffs_jpeg_decode_30k_24bpp_progressive/clang11 213MB/s ± 0% 213MB/s ± 0% +0.36% (p=0.008 n=5+5)
wuffs_jpeg_decode_30k_24bpp_sequential/clang11 384MB/s ± 0% 386MB/s ± 0% +0.50% (p=0.008 n=5+5)
wuffs_jpeg_decode_77k_24bpp/clang11 338MB/s ± 0% 341MB/s ± 0% +0.95% (p=0.008 n=5+5)
wuffs_jpeg_decode_552k_24bpp_420/clang11 353MB/s ± 0% 356MB/s ± 0% +0.99% (p=0.008 n=5+5)
wuffs_jpeg_decode_552k_24bpp_444/clang11 481MB/s ± 0% 484MB/s ± 0% +0.61% (p=0.008 n=5+5)
wuffs_jpeg_decode_4002k_24bpp/clang11 359MB/s ± 0% 364MB/s ± 0% +1.40% (p=0.008 n=5+5)
wuffs_jpeg_decode_19k_8bpp/gcc10 245MB/s ± 0% 234MB/s ± 0% -4.45% (p=0.008 n=5+5)
wuffs_jpeg_decode_30k_24bpp_progressive/gcc10 247MB/s ± 0% 245MB/s ± 0% -0.59% (p=0.008 n=5+5)
wuffs_jpeg_decode_30k_24bpp_sequential/gcc10 510MB/s ± 0% 500MB/s ± 0% -1.94% (p=0.008 n=5+5)
wuffs_jpeg_decode_77k_24bpp/gcc10 432MB/s ± 0% 417MB/s ± 0% -3.62% (p=0.008 n=5+5)
wuffs_jpeg_decode_552k_24bpp_420/gcc10 456MB/s ± 0% 442MB/s ± 0% -3.14% (p=0.008 n=5+5)
wuffs_jpeg_decode_552k_24bpp_444/gcc10 484MB/s ± 0% 477MB/s ± 0% -1.49% (p=0.008 n=5+5)
wuffs_jpeg_decode_4002k_24bpp/gcc10 465MB/s ± 0% 450MB/s ± 0% -3.28% (p=0.008 n=5+5)
2023-06-23 16:29:16 +10:00
Nigel Tao
bedf2bd7c3
lang: let skip_u32_fast imply can_undo_byte
2023-06-20 23:02:01 +10:00
Nigel Tao
723c604a5b
std/jpeg: tweak decoder.fill_bitstream inner loop
...
name old speed new speed delta
wuffs_jpeg_decode_19k_8bpp/clang11 229MB/s ± 0% 225MB/s ± 0% -1.78% (p=0.008 n=5+5)
wuffs_jpeg_decode_30k_24bpp_progressive/clang11 212MB/s ± 0% 212MB/s ± 0% ~ (p=0.421 n=5+5)
wuffs_jpeg_decode_30k_24bpp_sequential/clang11 380MB/s ± 0% 377MB/s ± 0% -0.56% (p=0.008 n=5+5)
wuffs_jpeg_decode_77k_24bpp/clang11 331MB/s ± 0% 328MB/s ± 0% -1.00% (p=0.008 n=5+5)
wuffs_jpeg_decode_552k_24bpp_420/clang11 344MB/s ± 0% 341MB/s ± 0% -0.97% (p=0.008 n=5+5)
wuffs_jpeg_decode_552k_24bpp_444/clang11 476MB/s ± 0% 473MB/s ± 0% -0.66% (p=0.008 n=5+5)
wuffs_jpeg_decode_4002k_24bpp/clang11 350MB/s ± 0% 348MB/s ± 0% -0.64% (p=0.008 n=5+5)
wuffs_jpeg_decode_19k_8bpp/gcc10 215MB/s ± 0% 225MB/s ± 0% +4.62% (p=0.008 n=5+5)
wuffs_jpeg_decode_30k_24bpp_progressive/gcc10 246MB/s ± 0% 247MB/s ± 0% +0.35% (p=0.008 n=5+5)
wuffs_jpeg_decode_30k_24bpp_sequential/gcc10 485MB/s ± 0% 499MB/s ± 0% +2.84% (p=0.008 n=5+5)
wuffs_jpeg_decode_77k_24bpp/gcc10 394MB/s ± 0% 408MB/s ± 0% +3.63% (p=0.008 n=5+5)
wuffs_jpeg_decode_552k_24bpp_420/gcc10 414MB/s ± 0% 428MB/s ± 0% +3.30% (p=0.008 n=5+5)
wuffs_jpeg_decode_552k_24bpp_444/gcc10 462MB/s ± 0% 470MB/s ± 0% +1.81% (p=0.016 n=4+5)
wuffs_jpeg_decode_4002k_24bpp/gcc10 420MB/s ± 0% 433MB/s ± 0% +3.13% (p=0.008 n=5+5)
2023-06-20 22:21:10 +10:00
Nigel Tao
9b97c27ec6
std/jpeg: tweak UNZIG array
...
We'll hopefully claw back the performance in a follow-up commit.
name old speed new speed delta
wuffs_jpeg_decode_19k_8bpp/clang11 222MB/s ± 0% 224MB/s ± 0% +1.11% (p=0.008 n=5+5)
wuffs_jpeg_decode_30k_24bpp_progressive/clang11 213MB/s ± 0% 210MB/s ± 0% -1.53% (p=0.008 n=5+5)
wuffs_jpeg_decode_30k_24bpp_sequential/clang11 376MB/s ± 0% 378MB/s ± 0% +0.37% (p=0.008 n=5+5)
wuffs_jpeg_decode_77k_24bpp/clang11 325MB/s ± 0% 328MB/s ± 0% +0.66% (p=0.008 n=5+5)
wuffs_jpeg_decode_552k_24bpp_420/clang11 339MB/s ± 0% 340MB/s ± 0% +0.46% (p=0.008 n=5+5)
wuffs_jpeg_decode_552k_24bpp_444/clang11 468MB/s ± 0% 469MB/s ± 0% +0.27% (p=0.008 n=5+5)
wuffs_jpeg_decode_4002k_24bpp/clang11 344MB/s ± 0% 346MB/s ± 0% +0.37% (p=0.008 n=5+5)
wuffs_jpeg_decode_19k_8bpp/gcc10 212MB/s ± 0% 213MB/s ± 0% +0.46% (p=0.008 n=5+5)
wuffs_jpeg_decode_30k_24bpp_progressive/gcc10 246MB/s ± 0% 221MB/s ± 0% -9.98% (p=0.008 n=5+5)
wuffs_jpeg_decode_30k_24bpp_sequential/gcc10 484MB/s ± 0% 484MB/s ± 0% ~ (p=0.056 n=5+5)
wuffs_jpeg_decode_77k_24bpp/gcc10 390MB/s ± 0% 392MB/s ± 0% +0.47% (p=0.008 n=5+5)
wuffs_jpeg_decode_552k_24bpp_420/gcc10 415MB/s ± 0% 414MB/s ± 0% -0.17% (p=0.008 n=5+5)
wuffs_jpeg_decode_552k_24bpp_444/gcc10 463MB/s ± 0% 459MB/s ± 0% -0.78% (p=0.008 n=5+5)
wuffs_jpeg_decode_4002k_24bpp/gcc10 420MB/s ± 0% 419MB/s ± 0% -0.19% (p=0.008 n=5+5)
2023-06-18 20:57:50 +10:00
Nigel Tao
143991a301
std/jpeg: check up-front for sufficient AC bits
...
The micro-benchmarks are mixed for now, but this sets us up to remove an
inner-loop bounds check in a future commit.
name old speed new speed delta
wuffs_jpeg_decode_19k_8bpp/clang11 226MB/s ± 0% 224MB/s ± 1% -1.03% (p=0.029 n=4+4)
wuffs_jpeg_decode_30k_24bpp_progressive/clang11 210MB/s ± 0% 206MB/s ± 0% -1.59% (p=0.029 n=4+4)
wuffs_jpeg_decode_30k_24bpp_sequential/clang11 379MB/s ± 0% 374MB/s ± 0% -1.25% (p=0.016 n=4+5)
wuffs_jpeg_decode_77k_24bpp/clang11 326MB/s ± 0% 319MB/s ± 0% -2.35% (p=0.029 n=4+4)
wuffs_jpeg_decode_552k_24bpp_420/clang11 338MB/s ± 0% 329MB/s ± 0% -2.53% (p=0.008 n=5+5)
wuffs_jpeg_decode_552k_24bpp_444/clang11 468MB/s ± 0% 465MB/s ± 0% -0.58% (p=0.008 n=5+5)
wuffs_jpeg_decode_4002k_24bpp/clang11 347MB/s ± 0% 332MB/s ± 0% -4.41% (p=0.016 n=4+5)
wuffs_jpeg_decode_19k_8bpp/gcc10 174MB/s ± 0% 176MB/s ± 0% +1.18% (p=0.008 n=5+5)
wuffs_jpeg_decode_30k_24bpp_progressive/gcc10 247MB/s ± 0% 246MB/s ± 0% -0.55% (p=0.008 n=5+5)
wuffs_jpeg_decode_30k_24bpp_sequential/gcc10 477MB/s ± 0% 472MB/s ± 0% -0.90% (p=0.008 n=5+5)
wuffs_jpeg_decode_77k_24bpp/gcc10 328MB/s ± 0% 334MB/s ± 0% +1.69% (p=0.008 n=5+5)
wuffs_jpeg_decode_552k_24bpp_420/gcc10 346MB/s ± 0% 351MB/s ± 0% +1.57% (p=0.008 n=5+5)
wuffs_jpeg_decode_552k_24bpp_444/gcc10 424MB/s ± 0% 425MB/s ± 0% +0.39% (p=0.008 n=5+5)
wuffs_jpeg_decode_4002k_24bpp/gcc10 350MB/s ± 0% 349MB/s ± 0% -0.23% (p=0.008 n=5+5)
2023-06-18 17:37:12 +10:00
Nigel Tao
1f0f323414
std/jpeg: memoize Huffman table selectors
2023-06-12 16:23:53 +10:00
Nigel Tao
c0d7d7d046
std/jpeg: remove decode_idct block-selector arg
...
name old speed new speed delta
wuffs_jpeg_decode_19k_8bpp/clang11 210MB/s ± 0% 213MB/s ± 0% +1.54% (p=0.008 n=5+5)
wuffs_jpeg_decode_30k_24bpp_progressive/clang11 200MB/s ± 0% 197MB/s ± 0% -1.31% (p=0.008 n=5+5)
wuffs_jpeg_decode_30k_24bpp_sequential/clang11 362MB/s ± 0% 373MB/s ± 0% +3.04% (p=0.008 n=5+5)
wuffs_jpeg_decode_77k_24bpp/clang11 307MB/s ± 0% 318MB/s ± 0% +3.82% (p=0.008 n=5+5)
wuffs_jpeg_decode_552k_24bpp_420/clang11 318MB/s ± 0% 331MB/s ± 0% +4.09% (p=0.008 n=5+5)
wuffs_jpeg_decode_552k_24bpp_444/clang11 434MB/s ± 0% 440MB/s ± 0% +1.34% (p=0.016 n=4+5)
wuffs_jpeg_decode_4002k_24bpp/clang11 320MB/s ± 0% 340MB/s ± 0% +6.29% (p=0.008 n=5+5)
wuffs_jpeg_decode_19k_8bpp/gcc10 165MB/s ± 0% 172MB/s ± 0% +4.32% (p=0.008 n=5+5)
wuffs_jpeg_decode_30k_24bpp_progressive/gcc10 224MB/s ± 0% 227MB/s ± 0% +1.30% (p=0.008 n=5+5)
wuffs_jpeg_decode_30k_24bpp_sequential/gcc10 454MB/s ± 0% 465MB/s ± 0% +2.40% (p=0.008 n=5+5)
wuffs_jpeg_decode_77k_24bpp/gcc10 311MB/s ± 0% 324MB/s ± 0% +4.38% (p=0.008 n=5+5)
wuffs_jpeg_decode_552k_24bpp_420/gcc10 327MB/s ± 0% 342MB/s ± 0% +4.52% (p=0.008 n=5+5)
wuffs_jpeg_decode_552k_24bpp_444/gcc10 384MB/s ± 0% 393MB/s ± 0% +2.34% (p=0.008 n=5+5)
wuffs_jpeg_decode_4002k_24bpp/gcc10 324MB/s ± 0% 345MB/s ± 1% +6.26% (p=0.008 n=5+5)
2023-06-10 14:12:08 +10:00
Nigel Tao
a3e8531ecb
std/jpeg: add swizzle_ycck_scratch_buffer_2k
2023-06-06 22:25:30 +10:00
Nigel Tao
2865c5bfbd
std/jpeg: call slice.bulk_{load,save}_host_endian
...
Updates https://bugs.chromium.org/p/oss-fuzz/issues/detail?id=59540
name old speed new speed delta
wuffs_jpeg_decode_30k_24bpp_progressive/clang11 131MB/s ± 0% 168MB/s ± 0% +28.44% (p=0.008 n=5+5)
wuffs_jpeg_decode_30k_24bpp_progressive/gcc10 167MB/s ± 0% 175MB/s ± 0% +4.80% (p=0.008 n=5+5)
2023-06-05 12:51:50 +10:00
Nigel Tao
4a164cda64
std/jpeg: tweak apply_progressive_idct
2023-06-05 09:29:30 +10:00
Nigel Tao
238aadd8eb
std/jpeg: factor out calculate_multiple_competc
2023-06-04 21:27:23 +10:00
Nigel Tao
2f18dc61d0
std/jpeg: call slice.bulk_memset
2023-06-04 21:22:14 +10:00
Nigel Tao
5edfe45e15
std/jpeg: add "#unsupported scan count"
...
Updates https://bugs.chromium.org/p/oss-fuzz/issues/detail?id=59540
2023-06-04 15:46:18 +10:00
Nigel Tao
0b18d43e4f
std/jpeg: avoid unnecessary IDCT applications
...
Updates https://bugs.chromium.org/p/oss-fuzz/issues/detail?id=59540
name old speed new speed delta
wuffs_jpeg_decode_30k_24bpp_progressive/clang11 84.6MB/s ± 0% 136.4MB/s ± 0% +61.33% (p=0.008 n=5+5)
wuffs_jpeg_decode_30k_24bpp_progressive/gcc10 93.8MB/s ± 0% 166.5MB/s ± 0% +77.49% (p=0.008 n=5+5)
2023-06-04 15:30:20 +10:00