[babl] Created tag BABL_0_1_90

[Øyvind "pippin" Kolås <ok src gnome org>]

The unsigned tag 'BABL_0_1_90' was created.

Tagger: Øyvind Kolås <pippin gimp org>
Date: 1645482868 +0100

    0.1.90

Changes since the last tag 'BABL_0_1_88_':

Andrzej Hunt (1):
      babl_format_new: also free doc to plug leak

Jehan (2):
      tools: use rand() as fallback of random().
      tools: use _putenv_s() instead of setenv() on Windows.

Mingye Wang (4):
      oklab, initial ver
      expose chromatic adaptation matrix
      oklab: chad back to D65
      quick fix

Tomasz Golinski (2):
      Add CIE Lab alpha u8 format
      Add CIE Lab alpha u16 format

Øyvind Kolås (71):
      post-release version bump
      gitlab-ci: remove nonexistent meson option enable-sse3
      export-symbols: add babl_chromatic_adaptation_matrix
      oklab: add to meson, register model conversions
      babl: adjust sanity criteria for babl_model_is_symmetric
      oklab: add double variant of code for model conversions
      Oklab: disable only Oklch not Oklab
      tests: add a xyz to lab test
      babl: set ecx to 0 in cpuid
      gitlab-ci: disable paccache, trying to unbreak build
      babl: sync cpuaccel with GEGL
      build: enable -ftree-vectorize always, and -mfpu=neon on aarch64
      base: x86-64-v2 and x86-64-v3 variants
      babl: reduce number of pixels tested for paths to 1024, include some out of bounds pixels in start
      multipass build, with SIMD variants for x86_64 and arm
      extensions: cairo remove a broken conversion, fix RGBA u8 to cairo reference
      tools: babl-verify make verify do space free conversions by default, it is more useful
      extensions: reduce amount of competing conversions
      extensions/cairo: add a correct enough for 8bit cairo32 to cairo24 conversion
      babl: add missing files babl-trc.[ch]
      babl: add more missing SIMD build related files
      babl: cpuaccel fix arm neon detect
      babl: cpuaccel fix arm neon detect
      babl: remove allocations in u8 variants of RGB space conversions
      babl: adjust default path parameters to a path_length of 3
      babl: add u8 code paths to trc/matrix space conversion
      tools: rework babl-benchmark to operate on sets resembling workingspace + GEGL + display
      babl: add lut for some u8 conversions
      babl: remove some now dead u8 code
      babl: use LUT for 16bit to 8bit conversions
      babl: LUT cleanups
      babl: add sse2 flags when building RGB converter
      babl: avoid using half initialized LUTs when used from threads
      babl: lut, tidy u8 code - disable u16/half variant
      babl: free LUTs that have been unused for more than 5 minutes
      babl: LUTs fix typo in timelimit check
      ctx: factor lut processing out of path_process
      tools: add babl-lut-verify
      build: append list of one element rather than string
      make trampoline for lut processing
      build: fix append of sse2 cflags
      Revert "make trampoline for lut processing"
      babl: lut remove unused u16 code
      babl: also do auto LUTs for 3-3 bpp and 3-4bpp paths
      tools: add RGB variants to babl-benchmark
      babl: do LUTs for 1 and 2 byte encodings
      babl: LUTs for 16byte destination encodings
      babl: measure performance of LUTs and only replace if faster
      babl: add LUT debugging configurable through BABL_LUT_INFO environment variable
      babl: remove optional instrumentation
      new symbol babl_fish_get_process
      babl: only show active luts during gc on info_level>=4
      babl: add support for 2,2 LUTs
      babl: improve LUT debug info
      babl: make unified alpha conversion branchless
      babl: initialize bpp after reading cost in fish cache
      babl: do luts for 2->16 byte fishes
      babl: clean up compiler warnings
      babl: fix fix bpp initialization when loaded from cache
      sse2-float: correctly handle associted alpha in remainder
      tools: inhibit cache use in lut-verify and babl-benchmark
      babl: do not load cache if BABL_INHIBIT_CACHE is set in environment
      tools/babl-benchmark: print LUT/REF markers
      babl: add 4:8 luts
      tools: include CMYK encodings in babl-benchmark
      tools: rewrite babl-lut-verify to be more generic
      tools: include more conversions in babl-lut-verify
      babl: fix mutex lock/unlocks symmetry in babl_fish()
      babl: drop a now deprecated const
      NEWS: update
      release babl 0.1.90