9e3ec89cf2
[libaom] Use fuzzer plugin source from libaom repository ( #2515 )
...
Removed av1_dec_fuzzer.cc from projects/libaom and instead using
from libaom repository
BUG=https://crbug.com/aomedia/2257
2019-06-14 08:29:20 -07:00
3049c50d48
Migrate projects using -lFuzzingEngine to $LIB_FUZZING_ENGINE ( #2325 )
...
Migrate from -lFuzzingEngine to $LIB_FUZZING_ENGINE where possible and not causing breakage
2019-05-01 11:09:55 -04:00
23dd904766
[libaom] use ADD to retrieve file ( #1964 )
...
* [libaom] use ADD to retrieve file
libwebp uses ADD here which appears to be the canonical way to download a file.
* Update Dockerfile
2018-11-20 10:50:19 -08:00
72adedc68e
av1_dec_fuzzer: vary thread count in range [2, 64] ( #1840 )
...
* av1_dec_fuzzer: vary thread count in range [2, 64]
use at most the first 100 bytes after skipping the file header to
calculate a hash used to create the thread count.
+ add missing includes, delete unused string.h
* av1_dec_fuzzer: use the first 32 bytes as a hash
This is the minimum required by this function as it's read and interpreted as the IVF file header
* av1_dec_fuzzer: use 1 header byte for thread count
this is less complex than using a hash and produces the same coverage, similar to:
https://github.com/ImageMagick/ImageMagick/blob/master/Magick++/fuzz/rotate_fuzzer.cc#L9-L16
https://cs.chromium.org/chromium/src/base/json/json_reader_fuzzer.cc?type=cs&q=json_reader_fuzzer&sq=package:chromium&g=0&l=20
2018-10-04 08:36:32 -07:00
014d5b9ac6
av1_dec_fuzzer: set cfg.allow_lowbitdepth to CONFIG_LOWBITDEPTH ( #1818 )
...
If cfg.allow_lowbitdepth is 0, then seq_params->use_highbitdepth is
unconditionally set to 1 by the following code in av1_read_color_config():
seq_params->use_highbitdepth =
seq_params->bit_depth > AOM_BITS_8 || !allow_lowbitdepth;
aomdec.c sets cfg.allow_lowbitdepth to CONFIG_LOWBITDEPTH. (We usually
pass -DCONFIG_LOWBITDEPTH=1 to cmake.) Chromium defines
CONFIG_LOWBITDEPTH as 1 and sets cfg.allow_lowbitdepth to 1:
https://chromium-review.googlesource.com/c/chromium/src/+/1178938
Our fuzzing test binary should match the behavior of aomdec and Chromium.
2018-09-15 14:30:16 -07:00
445361ce6e
av1_dec_fuzzer: rm unused variables ( #1816 )
2018-09-15 14:29:05 -07:00
f246e63b27
libaom: Add README explaining how to create a PR. ( #1651 )
...
* libaom: Add README explaining how to create a PR.
This may be a standard process for oss-fuzz and/or github in general,
but it's non-trivial for folks on libaom team, and has been found
to be essential by others in the team.
* Escape angle brackets and capital
2018-07-25 10:28:59 -07:00
5e067e6f3d
Miscellaneous cleanups. ( #1648 )
...
- Close 'file' before returning on errors.
- Don't call aom_codec_destroy() after an aom_codec_dec_init() failure.
- Remove a cast to unsigned int* that's no longer necessary.
- Change NULL to nullptr.
2018-07-24 10:53:34 +10:00
9a383175d2
Use a smaller AOM_MAX_ALLOCABLE_MEMORY for msan. ( #1647 )
...
The test cases for bug 9497 and bug 9499 cause out-of-memory errors
under msan but not under asan and ubsan. (This seems to imply msan has a
higher memory overhead.) Use a smaller AOM_MAX_ALLOCABLE_MEMORY value of
512 MB (as opposed to 1 GB) for msan.
BUG=oss-fuzz:9497
BUG=oss-fuzz:9499
2018-07-24 10:52:16 +10:00
8dbf55aa24
[libaom] build.sh: disable AV1 encoder ( #1631 )
...
Only the decoder is needed by the fuzz target. This will help prune unnecessary files from the coverage report.
BUG=chromium:838380
2018-07-18 19:31:55 -07:00
4ef7d05498
[libaom] build.sh: s/CONFIG_UNIT_TESTS/ENABLE_TESTS/ ( #1630 )
...
This change was made in:
d9647f214 cmake: Improve testing.
This has no effect on the fuzz target, but will decrease the build time.
2018-07-18 16:14:04 -07:00
90f03e3afa
[libaom] Add MemorySanitizer (MSan) support. ( #1627 )
...
Build libaom with all assembly code disabled for MemorySanitizer.
2018-07-17 17:41:55 -07:00
fb776a2881
[libaom] Run cmake in a subdirectory under $WORK. ( #1625 )
...
Before we run cmake, we remove everything in the directory where we run
cmake. This will remove /work/msan if we run cmake in $WORK. To avoid
that, create a $WORK/build directory and run cmake in that directory.
2018-07-17 15:41:08 -07:00
6ec00106a5
Rename dictionary file to match conventions ( #1612 )
2018-07-12 20:25:32 -07:00
b2dd9ab7ed
libaom: Add dictionary file. ( #1611 )
...
* libaom: Add dictionary file.
* Add byte position for fourCC too
2018-07-13 12:19:04 +10:00
cd9acd02f9
[libaom] Enable asserts in build.sh ( #1599 )
2018-07-06 15:11:43 -07:00
22119c0252
[libaom] Enable DO_RANGE_CHECK_CLAMP config. ( #1598 )
...
This is to avoid some noise of integer overflow sanitizer errors in
transform functions, which are actually working as intended.
BUG=oss-fuzz:9065
2018-07-06 09:34:53 -07:00
f8071907bd
libaom: Enable per-allocation memory limit. ( #1588 )
...
This avoids OOM fuzzer failures due to 2 GB limit set by oss-fuzz.
BUG=oss-fuzz:9222,9224
2018-07-02 17:31:39 -07:00
203510bc49
libaom: call codec_destroy at the end. ( #1583 )
...
This call was mistakenly missing earlier.
BUG=oss-fuzz:8849
BUG=oss-fuzz:8853
2018-06-29 16:19:45 -07:00
73886ea3b6
libaom,project.yaml: add wtc@ to auto_cc ( #1582 )
2018-06-29 12:24:34 -07:00
6948f92282
[libaom] Add Jai to CC list. ( #1549 )
2018-06-21 16:41:28 -07:00
5d4631a9fd
[libaom] av1_dec_fuzzer.cc: correct ivfdec.h include ( #1541 )
...
ivfdec.h is under common/, remove the aom prefix which would require an include path to $SRC in addition to $SRC/aom for the others.
2018-06-20 08:00:55 -07:00
ee91487456
[libaom] Add Hui to auto-ccs. ( #1512 )
2018-06-12 14:44:35 -07:00
befad94d0d
libaom: Remove fuzzer name suffix for serial mode. ( #1511 )
2018-06-12 13:00:26 -07:00
47b1086228
[libaom] Add a threaded mode for the fuzzer. ( #1509 )
2018-06-12 12:41:00 -07:00
ea7d5938fa
[libaom] Add project ( #1502 )
2018-06-11 18:27:33 -07:00
Harish Mahendrakar