openjpeg/src/lib/openjp2/sparse_array.c, branch floating_point_irreversible_encoding Fork of openjpeg library for JPEG2000 encoding and decoding https://git.carlh.net/cgit/openjpeg/atom?h=floating_point_irreversible_encoding 2017-09-19T16:35:31Z Workaroudn cppcheck (correct) warning about same code in then and else branches 2017-09-19T16:35:31Z Even Rouault even.rouault@spatialys.com 2017-09-19T16:35:31Z urn:sha1:c22ea12219c711031ef9b5823f2d9eb1081261e9 Replace C++ style comments by C ones 2017-09-19T16:25:55Z Even Rouault even.rouault@spatialys.com 2017-09-19T16:25:55Z urn:sha1:ce8edf9bffb8caae5480b862a04b66e0e355394b Replace C++ style comments by C ones 2017-09-19T16:22:07Z Even Rouault even.rouault@spatialys.com 2017-09-19T16:22:07Z urn:sha1:d415723fd9f2ee415e506e94ee8be8127680fdcd Fix warnings and errors when compiling with a c++ compiler (#1021) 2017-09-19T10:46:20Z Even Rouault even.rouault@spatialys.com 2017-09-19T10:46:20Z urn:sha1:fdef69b43ce5222ac0f2a87f2fad1442ece340e8 Replace uses of size_t by OPJ_SIZE_T 2017-09-04T15:35:52Z Even Rouault even.rouault@spatialys.com 2017-09-04T15:35:52Z urn:sha1:579b8937eae7e6b6868b8b5c6286a742c10a5130 sparse_array: optimizations for lossy case 2017-09-01T20:23:29Z Even Rouault even.rouault@spatialys.com 2017-09-01T14:31:10Z urn:sha1:7017e67a01c378a7a1ee5e34dd544c793b5c23e4 Sub-tile decoding: speed up vertical pass in IDWT5x3 by processing 4 cols at a time 2017-09-01T14:31:00Z Even Rouault even.rouault@spatialys.com 2017-09-01T14:31:00Z urn:sha1:873004c615b1bed3ce780e869288602af86fdee5 Optimize reading/write into sparse array 2017-09-01T14:30:50Z Even Rouault even.rouault@spatialys.com 2017-09-01T14:30:50Z urn:sha1:b2cc8f7f81242f967b65e76de043e5e31663d793 Various changes to allow tile buffers of more than 4giga pixels 2017-09-01T14:30:44Z Even Rouault even.rouault@spatialys.com 2017-09-01T14:30:44Z urn:sha1:98b93103613c90753fb6c57696f9403f8ea0b1d6 Untested though, since that means a tile buffer of at least 16 GB. So there might be places where uint32 overflow on multiplication still occur... Sub-tile decoding: only allocate tile component buffer of the needed dimension 2017-09-01T14:30:29Z Even Rouault even.rouault@spatialys.com 2017-09-01T14:30:29Z urn:sha1:f9e9942330f476b66ac4a35d0ae521200878f343 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.