Even Rouault [Sun, 11 Feb 2018 12:31:04 +0000 (13:31 +0100)]
Avoid out-of-bounds write overflow due to uint32 overflow computation on images with huge dimensions. Credit to Google Autofuzz project for providing test case
Even Rouault [Mon, 8 Jan 2018 08:38:44 +0000 (09:38 +0100)]
opj_t2_encode_packet(): disable setting empty packet header bit to 1 when there is an empty packet
This effectively reverts commit 2609fb8077125b5b31f1bcc2f98c12ff1e6572d7
since it has been reported that
such packets cause decoding issues with cinema J2K hardware
decoders: https://groups.google.com/forum/#!topic/openjpeg/M7M_fLX_Bco
Even Rouault [Thu, 30 Nov 2017 13:48:34 +0000 (14:48 +0100)]
opj_j2k_read_cod: remove check for 'No more than one COD marker per tile' (fixes #1043)
This check was added per https://github.com/uclouvain/openjpeg/commit/daed8cc9195555e101ab708a501af2dfe6d5e001
to fix https://github.com/uclouvain/openjpeg/issues/476 , but it does not seem
to be necessary with latest master (issue476.jp2 doesn't cause memory issues),
and breaks reading legit files.
Even Rouault [Mon, 9 Oct 2017 09:40:43 +0000 (11:40 +0200)]
Unix build: fix regression of 2.3.0 where a shared-only or static-only build lacks the installation target for the library (#1019, fixes regression introduced by 3dfc6ca2bcf06fd1adb6b6b4cecc6c092f08ba0b)
Even Rouault [Fri, 8 Sep 2017 08:56:49 +0000 (10:56 +0200)]
opj_tcd_mct_decode(): avoid heap buffer overflow when components have not the same number of resolutions. Also fixes an issue with subtile decoding. Fixes https://bugs.chromium.org/p/oss-fuzz/issues/detail?id=3331. Credit to OSS Fuzz
Even Rouault [Fri, 8 Sep 2017 07:16:51 +0000 (09:16 +0200)]
Use opj_image_data_free() where appropriate (adapted from https://github.com/uclouvain/openjpeg/pull/1015/commits/dab9db0723a5bb9f3d745f9dd7a0b8b3b18b8054, #1014)
Even Rouault [Wed, 6 Sep 2017 13:59:19 +0000 (15:59 +0200)]
Fix null pointer dereference on partial tile decoding when they are empty. Fixes https://bugs.chromium.org/p/oss-fuzz/issues/detail?id=3297 (master only)
For some reason, the OPJ_CI_ARCH=x86_64 OPJ_CI_BUILD_CONFIGURATION=Release OPJ_NUM_THREADS=2
configuration fails once PR1010 has been merged in master
( https://travis-ci.org/uclouvain/openjpeg/jobs/272219011 ) whereas (almost) the same
code in my branch didn't fail per https://travis-ci.org/rouault/openjpeg/jobs/271738113
The errors we get are the same as with the other x86_64 compilers, so nothing alarming here.
Even Rouault [Fri, 1 Sep 2017 18:43:39 +0000 (20:43 +0200)]
Replace error message 'Not enough memory for tile data' by 'Size of tile data exceeds system limits' (refs https://github.com/uclouvain/openjpeg/pull/730#issuecomment-326654188)
Even Rouault [Fri, 1 Sep 2017 17:16:35 +0000 (19:16 +0200)]
opj_compress help: revert 32572617765cb9d77302384653a48d793b8f657f and indicate 1 again as being the value to get lossless for -r. In opj_j2k_setup_encoder(), make sure that ll rates[] <= 1.0 are set to 0. Document 0 as being lossless for -q / tcp_distoratio (#1009)
Even Rouault [Fri, 1 Sep 2017 14:30:48 +0000 (16:30 +0200)]
Allow several repeated calls to opj_set_decode_area() and opj_decode() for single-tiled images
* Only works for single-tiled images --> will error out cleanly, as currently
in other cases
* Save re-reading the codestream for the tile, and re-use code-blocks of the
previous decoding pass.
* Future improvements might involve improving opj_decompress, and the image writing logic,
to use this strategy.
Even Rouault [Fri, 1 Sep 2017 14:30:37 +0000 (16:30 +0200)]
Remove limitation that prevents from opening images bigger than 4 billion pixels
However the intermediate buffer for decoding must still be smaller than 4
billion pixels, so this is useful for decoding at a lower resolution level,
or subtile decoding.
Even Rouault [Fri, 1 Sep 2017 14:30:29 +0000 (16:30 +0200)]
Sub-tile decoding: only allocate tile component buffer of the needed dimension
Instead of being the full tile size.
* Use a sparse array mechanism to store code-blocks and intermediate stages of
IDWT.
* IDWT, DC level shift and MCT stages are done just on that smaller array.
* Improve copy of tile component array to final image, by saving an intermediate
buffer.
* For full-tile decoding at reduced resolution, only allocate the tile buffer to
the reduced size, instead of the full-resolution size.