openjpeg/src/lib/openjp2/dwt.c, branch v2.2.0 Fork of openjpeg library for JPEG2000 encoding and decoding https://git.carlh.net/cgit/openjpeg/atom?h=v2.2.0 2017-07-06T10:11:37Z Comment fix 2017-07-06T10:11:37Z Even Rouault even.rouault@spatialys.com 2017-07-06T10:11:37Z urn:sha1:60f8ddf577b20ccbb3dc0559b624115b9ba522ba IDWT 5x3: fix bug in AVX2 implementation (#953, #957) 2017-06-29T22:03:05Z Even Rouault even.rouault@spatialys.com 2017-06-29T22:03:05Z urn:sha1:8fa405ee15b5151df40d8d96460257ee5b510135 IDWT 5x3: generalize SSE2 version for AVX2 2017-06-21T10:12:58Z Even Rouault even.rouault@spatialys.com 2017-06-21T10:12:58Z urn:sha1:fd0dc535ad9ae0d369d1039aaf56235583ca64ea Thanks to our macros that abstract SSE use, the functions can use AVX2 when available (at compile time) This brings an extra 23% speed improvement on bench_dwt in 64bit builds with AVX2 compared to SSE2. dwt.c: small cleanup 2017-06-20T23:07:56Z Even Rouault even.rouault@spatialys.com 2017-06-20T23:07:56Z urn:sha1:f6e3475cc9328e5378e96e72b5adf29fb07c5feb Improve performance of inverse DWT 5x3 (#953) 2017-06-20T16:01:34Z Even Rouault even.rouault@spatialys.com 2017-06-20T15:56:25Z urn:sha1:fa55b52d19f39765d7f108d654a8f87035a5772f * Use single-pass lifting inverse wavelet transform. * For vertical pass, use SSE2 when available so as to process 8 columns in parallel. This is the most beneficial improvement, since the vertical pass involves a lot of cache trashing. With the bench_dwt utility with default arguments (16383x16383 image), time goes from 4.064 s to 1.212 s. Fix astyle issue 2017-06-17T14:37:56Z Even Rouault even.rouault@spatialys.com 2017-06-17T14:37:56Z urn:sha1:32b20b93e04d38303f736e975eb5f483fda01ca5 Fix warnings with recent GCC versions 2017-06-17T12:09:31Z Even Rouault even.rouault@spatialys.com 2017-06-17T12:09:31Z urn:sha1:cc07aec6c7b50090c3f471ca2c68d0ffa8aa4ff1 Reformat whole codebase with astyle.options (#128) 2017-05-09T18:46:20Z Even Rouault even.rouault@spatialys.com 2017-05-09T13:44:46Z urn:sha1:563bd8499e63db976ca8358216138647593354bc Remove TODO for overflow check (#842) 2016-09-15T21:51:34Z Matthieu Darbois mayeut@users.noreply.github.com 2016-09-15T21:51:34Z urn:sha1:6e7616c83cb01d9b7a5a35e093ff9ff6741acc74 The check was already done. It’s been simplified. Reformat to get consistent style throughout the functions. Add overflow checks for opj_aligned_malloc (#841) 2016-09-14T23:57:53Z Matthieu Darbois mayeut@users.noreply.github.com 2016-09-14T23:57:53Z urn:sha1:9a07ccb3d0f076388e4da684a3bfd4327125c721 See https://pdfium.googlesource.com/pdfium/+/b20ab6c7acb3be1393461eb650ca8fa4660c937e/third_party/libopenjpeg20/0020-opj_aligned_malloc.patch