Commit Graph

  • 1b8030989f SIMD. sse only when needed Alexander 2020-01-17 03:30:42 +0300
  • be70e9eded SIMD Bands. access violation in getband Alexander 2020-01-17 03:19:06 +0300
  • d74ab21ac9 SIMD Reduce. sse4 implementation Alexander 2020-01-05 15:11:10 +0300
  • 33b11280b0 SIMD. Fix memory read error on some compilers Alexander 2020-01-07 02:35:42 +0300
  • f9c5589cac SIMD. Gather all SSE and AVX imports in ImPlatform.h Alexander 2020-01-15 16:46:47 +0300
  • e4ca18a9c8 SIMD Bands Alexander 2017-08-12 17:11:16 +0300
  • 8c3d6a32d1 SIMD Convert. Fix rounding error on RGB to L conversion Alexander 2020-01-16 00:05:03 +0300
  • 52cfbf339a SIMD Convert. Speedup other 2L convertions Alexander 2017-09-10 03:04:34 +0300
  • 4fbc9d1f37 SIMD Convert. RGB → L 2.2 times faster Alexander 2017-09-10 02:37:04 +0300
  • 7398f9e1c9 SIMD Convert. fix rounding and speedup a bit Alexander 2017-08-10 01:23:32 +0300
  • 3322b77c9a SIMD Convert. use 16bit arithmetics Alexander 2017-01-24 02:46:36 +0300
  • 03c150d0a6 SIMD Convert. sse4 version (still 1.4x faster than previous avx2 implementation) Alexander 2017-01-23 16:25:05 +0300
  • d82fe3dd74 SIMD Convert. use float div instead of gather Alexander 2017-01-23 13:52:21 +0300
  • 95a68340c9 SIMD Convert. fix RGBa → RGBA conversion on AVX2 homm 2016-08-16 15:22:54 +0200
  • f888dbe1cc SIMD Convert. RGBa → RGBA convert using gather homm 2016-05-29 23:30:38 +0300
  • 55b71f2edb SIMD Convert. avx2 implementation homm 2016-07-02 16:11:10 +0300
  • 6e91b2b41f SIMD Convert. faster implementation homm 2016-05-25 05:39:59 +0300
  • a7de792025 SIMD Convert. sse4 implementation homm 2016-05-25 05:25:22 +0300
  • a71b748bde SIMD Resample. Correct max coefficient calculation Alexander 2019-04-03 01:31:30 +0300
  • b3c28911c0 SIMD Resample. fix wrong usage of xmax for division compensation Alexander 2019-01-20 15:27:30 +0300
  • 247cba2e30 SIMD Resample. move files Alexander 2018-04-13 00:52:09 +0300
  • 7da807b8e4 SIMD Resample. add parentheses around var declarations Alexander 2017-09-23 17:19:54 +0300
  • 698e6a1698 SIMD Resample. optimize coefficients loading for horizontal pass Alexander 2017-02-25 11:26:24 +0300
  • 8796b40ef6 SIMD Resample. unrolled SSE4 & AVX2 homm 2016-10-04 02:49:07 +0300
  • b4a54840ee SIMD Filter. move files Alexander 2018-04-13 00:48:43 +0300
  • 416e580a77 SIMD Filter. fix memory access for: Alexander 2017-10-04 03:56:42 +0300
  • 1bd3fddf44 SIMD Filter. 5x5i_4u8 AVX2 Alexander 2017-09-13 23:57:42 +0300
  • 62dbbbdde6 SIMD Filter. advanced 5x5i_4u8 SSE4 Alexander 2017-09-13 21:14:38 +0300
  • b23fc5a958 SIMD Filter. 5x5i_4u8 SSE4 Alexander 2017-09-13 20:38:32 +0300
  • ccfc34a5ed SIMD Filter. finish 3x3i_4u8 Alexander 2017-09-13 12:05:21 +0300
  • a1ce80cb29 SIMD Filter. avx2 version Alexander 2017-09-13 05:37:40 +0300
  • 49f4de699f SIMD Filter. rearrange operations Alexander 2017-09-13 05:28:09 +0300
  • 3b67d8c516 SIMD Filter. reduce number of registers Alexander 2017-09-13 04:28:40 +0300
  • ab50dde9d2 SIMD Filter. Rearrange instruction for speedup Alexander 2017-09-13 03:22:03 +0300
  • 74d95f5e9d SIMD Filter. better loading Alexander 2017-09-13 02:26:45 +0300
  • 0e17b8313b SIMD Filter. better macros Alexander 2017-09-13 01:58:43 +0300
  • eb57b5eca7 SIMD Filter. 3x3i Alexander 2017-09-13 01:37:18 +0300
  • b561752597 SIMD Filter. move ImagingFilterxxx functions to separate files Alexander 2017-09-12 18:17:03 +0300
  • 256eb37439 SIMD Filter. fix offset Alexander 2017-09-07 13:35:20 +0300
  • 92d392a99c SIMD Filter. 5x5 single channel SSE4 (tests failed) Alexander 2017-09-07 13:19:19 +0300
  • 92c591016e SIMD Filter. consider last pixel in AVX Alexander 2017-09-07 01:32:31 +0300
  • 9f53dfef57 SIMD Filter. unroll AVX (with no profit) Alexander 2017-09-04 17:36:37 +0300
  • efe722fce3 SIMD Filter. Macros for AVX Alexander 2017-09-04 16:20:49 +0300
  • 2329ee5605 SIMD Filter. unroll AVX 2 times Alexander 2017-09-04 14:16:08 +0300
  • 80dfbaf22d SIMD Filter. First AVX try Alexander 2017-09-04 13:05:59 +0300
  • f2e021aa94 SIMD Filter. 3x3 SSE4 singleband: 2 lines Alexander 2017-08-14 02:03:17 +0300
  • 901d668f0c SIMD Filter. reuse loaded values Alexander 2017-08-13 23:17:11 +0300
  • 05a5ff706a SIMD Filter. faster 3x3 singleband SSE4 Alexander 2017-08-13 22:34:34 +0300
  • 64ab983164 SIMD Filter. 3x3 SSE4 singleband Alexander 2017-08-13 22:17:27 +0300
  • 4da1f70050 SIMD Filter. use macros in 3x3 Alexander 2017-08-13 21:49:45 +0300
  • f747e7efc8 SIMD Filter. use macros Alexander 2017-08-13 21:34:19 +0300
  • 8e58436e44 SIMD Filter. rearrange 3x3 filter to match 5x5 Alexander 2017-08-13 20:10:47 +0300
  • c5b973b6ad SIMD Filter. improve locality in 5x5 filter Alexander 2017-08-13 20:02:11 +0300
  • a84ad68ca8 SIMD Filter. a bit faster 5x5 filter Alexander 2017-08-13 19:58:20 +0300
  • c07157fd4b SIMD Filter. fast 3x3 filter Alexander 2017-08-13 19:23:19 +0300
  • 2f4aad4d57 SIMD Filter. 5x5 implementation Alexander 2017-08-13 17:45:43 +0300
  • e9e2ed124a SIMD Filter. 3x3 implementation Alexander 2017-08-13 17:16:16 +0300
  • a794af7b00 SIMD BoxBlur. faster box blur for radius < 1 Alexander 2017-09-23 20:52:27 +0300
  • dede8035a3 SIMD BoxBlur. add parentheses around var declarations Alexander 2017-09-23 17:58:55 +0300
  • 9609a5849f SIMD BoxBlur. sse4 ImagingBoxBlur implementation homm 2015-07-16 02:33:12 +0300
  • 5a939e33f7 SIMD AlphaComposite. fast div aproximation Alexander 2017-01-23 04:56:37 +0300
  • a48b385426 SIMD AlphaComposite. move declarations to beginning of the blocks homm 2016-09-23 14:52:54 +0300
  • f7f4e9e448 SIMD AlphaComposite. fix bugs homm 2016-07-03 23:46:33 +0300
  • f64d6a19c1 SIMD AlphaComposite. speedup avx2 by using _mm256_mullo_epi16 instead of _mm256_mullo_epi32 homm 2016-06-19 16:46:10 +0300
  • 594b5d00cb SIMD AlphaComposite. speedup sse4 by using _mm_mullo_epi16 instead of _mm_mullo_epi32 homm 2016-06-19 15:55:58 +0300
  • 794a8054c4 SIMD AlphaComposite. increase precision homm 2016-06-12 21:05:06 +0300
  • e716f2125c SIMD AlphaComposite. avx2 implementation homm 2016-06-12 20:41:53 +0300
  • c7ce11f0f5 SIMD AlphaComposite. sse4 implementation homm 2016-06-12 19:40:56 +0300
  • 2dc15b8016 SIMD. version Alexander 2017-10-04 21:54:16 +0300
  • 3891eb94bf SIMD. Exclude tests from package Alexander 2018-10-17 17:15:02 +0300
  • 452aa6f6b9 SIMD. replace benchmarks with link to pillow-perf Alexander 2017-02-27 02:38:20 +0300
  • 9f511d459a SIMD. Rewritten the Pillow-SIMD readme Elijah 2016-10-07 16:54:22 +0500
  • 5628c31bb3 SIMD. pypi readme, update setup homm 2016-10-04 13:05:41 +0300
  • 033755777a SIMD. fix list (+6 squashed commits) homm 2016-04-30 01:16:00 +0300
  • 05b63366df 9.1.0.dev0 version bump Andrew Murray 2022-01-03 09:22:31 +1100
  • 71d4f05db4
    Merge pull request #239 from hugovk/main Andrew Murray 2022-01-03 09:09:43 +1100
  • 6eddd4bbb4 Add workflow_dispatch to allow triggering builds Hugo van Kemenade 2022-01-02 21:47:56 +0200
  • 3c1f014f8b Pillow -> 9.0.0 Andrew Murray 2022-01-02 21:00:06 +1100
  • 82541b6dec 9.0.0 version bump 9.0.0 Andrew Murray 2022-01-02 20:51:23 +1100
  • cae5ac495b
    Merge pull request #5924 from radarhere/cves Andrew Murray 2022-01-02 19:53:13 +1100
  • ed4cf78137 CVEs TBD Andrew Murray 2022-01-02 18:09:45 +1100
  • d7f60d1d5a
    Merge pull request #5923 from radarhere/imagemath_eval mergify[bot] 2022-01-02 07:04:41 +0000
  • 9e81644097
    Merge pull request #238 from radarhere/numpy_versions Hugo van Kemenade 2022-01-02 08:55:54 +0200
  • 8531b01d6c Restrict builtins for ImageMath.eval Andrew Murray 2022-01-02 17:23:49 +1100
  • 5b1d6a8753 Limit numpy on 32-bit to 1.21.5 for Python 3.8 and 3.9 Andrew Murray 2022-01-02 15:18:37 +1100
  • 1efb1d9fab
    Merge pull request #5922 from radarhere/releasenotes Hugo van Kemenade 2022-01-01 15:25:11 +0200
  • f6c78713a4 Added release notes for #5919, #5920 and #5921 Andrew Murray 2022-01-01 22:10:48 +1100
  • 032d2dc365 Update CHANGES.rst [ci skip] Andrew Murray 2022-01-01 22:00:12 +1100
  • baae9ec4b6
    Merge pull request #5921 from radarhere/jpeg_eoi mergify[bot] 2022-01-01 10:58:21 +0000
  • 1059eb5376 If appended EOI did not work, do not keep trying Andrew Murray 2022-01-01 21:04:32 +1100
  • 525c840b0f Updated path after pillow-wheels switched to libjpeg-turbo Andrew Murray 2022-01-01 20:15:06 +1100
  • 7291c3827f
    Merge pull request #5918 from radarhere/main Hugo van Kemenade 2022-01-01 10:41:07 +0200
  • 5543e4e2d4
    Merge pull request #5920 from radarhere/calloc Andrew Murray 2022-01-01 19:16:11 +1100
  • c48271ab35 Handle case where path count is zero Andrew Murray 2021-12-06 22:25:14 +1100
  • 1e092419b6 Initialize coordinates to zero Andrew Murray 2021-12-06 22:24:19 +1100
  • a8f90a3916
    Merge pull request #5919 from radarhere/duplicate Andrew Murray 2022-01-01 15:35:10 +1100
  • 7370a0b1cf Remove consecutive duplicates that only differ by their offset Andrew Murray 2022-01-01 14:54:23 +1100
  • 0bb3f87dcc
    Updated copyright year Andrew Murray 2022-01-01 10:40:57 +1100
  • 89d0d38796
    Merge pull request #5917 from radarhere/releasenotes Andrew Murray 2021-12-31 19:30:35 +1100
  • 949b431f03 Added release notes for pillow-wheels#237 Andrew Murray 2021-12-31 18:09:44 +1100