Add fate tests for palette act and field

.forgejo/CODEOWNERS

@@ -12,11 +12,14 @@ libavcodec/*ac3* @lynne
 libavcodec/*atrac9* @lynne
 libavcodec/*bitpacked* @lynne
 libavcodec/*dirac* @lynne
-libavcodec/*ffv1* @lynne
+libavcodec/*ffv1* @lynne @michaelni
+libavcodec/golomb* @michaelni
 libavcodec/*jpegxl* @lynne
 libavcodec/*jxl* @lynne
 libavcodec/*opus* @lynne
 libavcodec/*prores* @lynne
+libavcodec/rangecoder* @michaelni
+libavcodec/ratecontrol* @michaelni
 libavcodec/*siren* @lynne
 libavcodec/*vc2* @lynne
 
@@ -27,15 +30,21 @@ libavcodec/x86/* @lynne
 # avfilter
 # =======
 libavfilter/aarch64/* @mstorsjo
+libavfilter/vf_yadif* @michaelni
+libavfilter/vsrc_mandelbrot* @michaelni
 
 # avformat
 # =======
 libavformat/iamf* @jamrial
 
 # avutil
 # ======
-libavutil/*crc* @lynne
+libavutil/*crc* @lynne @michaelni
+libavutil/eval* @michaelni
 libavutil/iamf* @jamrial
+libavutil/mathematics* @michaelni
+libavutil/mem* @michaelni
+libavutil/rational* @michaelni
 libavutil/tx* @lynne
 
 libavutil/aarch64/* @lynne @mstorsjo
@@ -46,12 +55,17 @@ libavutil/x86/* @lynne
 # =======
 libswresample/aarch64/* @mstorsjo
 libswresample/arm/* @mstorsjo
+libswresample/* @michaelni
 
 # swscale
 # =======
 libswscale/aarch64/* @mstorsjo
 libswscale/arm/* @mstorsjo
 
+# doc
+# ===
+doc/* @GyanD
+
 # Frameworks
 # ==========
 *vulkan* @lynne

.forgejo/workflows/test.yml

@@ -30,7 +30,7 @@ jobs:
         id: fate
         run: |
           make fate-rsync SAMPLES=$PWD/fate-suite
-          echo "hash=$(find fate-suite -type f | sha256sum | cut -d' ' -f1)" >> $FORGEJO_OUTPUT
+          echo "hash=$(find fate-suite -type f | sort | sha256sum | cut -d' ' -f1)" >> $FORGEJO_OUTPUT
       - name: Cache Fate-Suite
         uses: actions/cache/save@v4
         if: ${{ format('fate-suite-{0}', steps.fate.outputs.hash) != steps.cache.outputs.cache-matched-key }}

.github/workflows/makefile.yml

@@ -0,0 +1,111 @@
+name: test
+run-name: ${{ github.workflow }} - ${{ github.sha }}
+on:
+  push:
+    branches: [ main, up ]
+  pull_request:
+    branches: [ main, up ]
+  workflow_dispatch:
+
+
+jobs:
+  ffvvc-test:
+    name: ffvvc-test / ${{ matrix.os.name }}/${{ matrix.compiler.name }}/${{ matrix.assembler.name }}
+    env:
+      configure_flags: --enable-ffmpeg --disable-everything --enable-decoder=vvc --enable-parser=vvc --enable-demuxer=vvc,mpegts --enable-protocol=file,pipe --enable-encoder=rawvideo,wrapped_avframe --enable-muxer=rawvideo,md5,null
+    strategy:
+      fail-fast: false
+      matrix:
+        os:
+          - { name: linux, runner: ubuntu-latest, shell: bash, runner_threads: 4 }
+          - { name: windows, runner: windows-latest, shell: 'msys2 {0}', runner_threads: 1 }
+        compiler:
+          - { name: gcc, flags: --cc=gcc }
+          - { name: clang, flags: --cc=clang }
+          - { name: msvc, flags: --toolchain=msvc }
+          - { name: clang-usan, flags: '--toolchain=clang-usan' }
+          - { name: clang-asan, flags: '--toolchain=clang-asan' }
+        assembler:
+          - { name: no asm, flags: --disable-asm }
+          - { name: nasm, flags: --as=nasm }
+        exclude:
+          # GitHub's Actions runners do not support AVX2.
+          - os: { name: linux, runner: ubuntu-latest, shell: bash, runner_threads: 4}
+            compiler: { name: msvc, flags: --toolchain=msvc }
+          - os: { name: linux, runner: ubuntu-latest, shell: bash, runner_threads: 4 }
+            assembler: { name: nasm, flags: --as=nasm }
+          # Address sanitizer cannot be run with handwritten assembly.
+          - compiler: { name: clang-asan, flags: '--toolchain=clang-asan' }
+            assembler: { name: nasm, flags: --as=nasm }
+          # Windows only supports MSVC.
+          - os: { name: windows, runner: windows-latest, shell: 'msys2 {0}', runner_threads: 1 }
+            compiler: { name: gcc, flags: --cc=gcc }
+          - os: { name: windows, runner: windows-latest, shell: 'msys2 {0}', runner_threads: 1 }
+            compiler: { name: clang, flags: --cc=clang }
+          - os: { name: windows, runner: windows-latest, shell: 'msys2 {0}', runner_threads: 1 }
+            compiler: { name: clang-usan, flags: '--toolchain=clang-usan' }
+          - os: { name: windows, runner: windows-latest, shell: 'msys2 {0}', runner_threads: 1 }
+            compiler: { name: clang-asan, flags: '--toolchain=clang-asan' }
+
+    runs-on: ${{ matrix.os.runner }}
+    defaults:
+      run:
+        shell: ${{ matrix.os.shell }}
+
+    steps:
+    - name: Get MSVC
+      if: ${{ matrix.compiler.name == 'msvc' && matrix.os.name == 'windows' }}
+      uses: ilammy/msvc-dev-cmd@v1
+
+    - name: Set up MSYS2
+      if: ${{ matrix.os.shell == 'msys2 {0}' }}
+      uses: msys2/setup-msys2@v2
+      with:
+        release: false
+        msystem: UCRT64
+        path-type: inherit
+        install: >-
+          make
+          diffutils
+
+    - name: Setup python package
+      run: python3 -m pip install tqdm pyyaml
+
+    - name: Get assembler
+      if: ${{ matrix.os.shell == 'msys2 {0}' && matrix.assembler.name != 'no asm' }}
+      run: pacman --noconfirm -S ${{ matrix.assembler.name }}
+
+    - name: Get source
+      uses: actions/checkout@v3
+      with:
+        path: FFmpeg
+
+    - name: Configure
+      run: cd FFmpeg && ./configure ${{ matrix.compiler.flags }} ${{ matrix.assembler.flags }} ${{ env.configure_flags }} || (tail ffbuild/config.log; false)
+
+    - name: Build
+      run: cd FFmpeg && make -j8
+
+    - name: Get tests
+      uses: actions/checkout@v3
+      with:
+        repository: ffvvc/tests
+        path: tests
+
+    - name: Unit test
+      run: python3 tests/tools/ffmpeg.py --threads ${{ matrix.os.runner_threads }} --ffmpeg-path=./FFmpeg/ffmpeg tests/conformance/passed
+
+    - name: DVB 40 frames test
+      run: python3 tests/tools/ffmpeg.py --threads 1 --ffmpeg-path=./FFmpeg/ffmpeg tests/conformance/dvb/40frames
+
+    - name: Check ASM
+      run: cd FFmpeg && make checkasm -j && ./tests/checkasm/checkasm
+
+    - name: Negative test
+      run: python3 tests/tools/ffmpeg.py --threads ${{ matrix.os.runner_threads }} --ffmpeg-path=./FFmpeg/ffmpeg tests/conformance/failed || true
+
+    - name: Check for fuzz regressions
+      run: python3 tests/tools/ffmpeg.py --threads ${{ matrix.os.runner_threads }} --ffmpeg-path=./FFmpeg/ffmpeg --fuzz tests/fuzz/passed
+
+    - name: Fuzz negative test
+      run: python3 tests/tools/ffmpeg.py --threads ${{ matrix.os.runner_threads }} --ffmpeg-path=./FFmpeg/ffmpeg --fuzz tests/fuzz/failed || tree

Changelog

@@ -26,6 +26,7 @@ version <next>:
 - OpenHarmony hardware decoder/encoder
 - Colordetect filter
 - Add vf_scale_d3d11 filter
+- No longer disabling GCC autovectorization, on X86, ARM and AArch64
 
 
 version 7.1:

configure

@@ -4925,7 +4925,7 @@ msvc_common_flags(){
             -lstdc++)             ;;
             -l*)                  echo ${flag#-l}.lib ;;
             -LARGEADDRESSAWARE)   echo $flag ;;
-            -L*)                  echo -libpath:${flag#-L} ;;
+            -L*) [ "$_flags_type" = "link" ] && echo -libpath:${flag#-L} ;;
             -Wl,*)                ;;
             *)                    echo $flag ;;
         esac
@@ -4947,6 +4947,12 @@ msvc_flags(){
     done
 }
 
+msvc_flags_link(){
+    _flags_type=link
+    msvc_flags "$@"
+    unset _flags_type
+}
+
 icl_flags(){
     msvc_common_flags "$@"
     for flag; do
@@ -5173,12 +5179,13 @@ probe_cc(){
         _cflags_noopt="-O1"
         if $_cc -nologo- 2>&1 | grep -q Linker; then
             _ld_o='-out:$@'
+            _flags_filter=msvc_flags_link
         else
             _ld_o='-Fe$@'
+            _flags_filter=msvc_flags
         fi
         _cc_o='-Fo$@'
         _cc_e='-P -Fi$@'
-        _flags_filter=msvc_flags
         _ld_lib='lib%.a'
         _ld_path='-libpath:'
         _flags='-nologo'
@@ -5277,6 +5284,12 @@ else
     ar_o='$@'
 fi
 
+# Treat unrecognized flags as errors on MSVC
+test_cpp_condition windows.h "_MSC_FULL_VER >= 193030705" &&
+    check_cflags -options:strict
+test_host_cpp_condition windows.h "_MSC_FULL_VER >= 193030705" &&
+    check_host_cflags -options:strict
+
 add_cflags $extra_cflags
 add_cxxflags $extra_cxxflags
 add_objcflags $extra_objcflags
@@ -5298,6 +5311,7 @@ if test "$cpu" = host; then
     case "$cc_type" in
         gcc|llvm_gcc)
             check_native(){
+                : > $TMPC
                 $cc $1=native -v -c -o $TMPO $TMPC >$TMPE 2>&1 || return
                 sed -n "/cc1.*$1=/{
                             s/.*$1=\\([^ ]*\\).*/\\1/
@@ -5309,6 +5323,7 @@ if test "$cpu" = host; then
         ;;
         clang)
             check_native(){
+                : > $TMPC
                 $cc $1=native -v -c -o $TMPO $TMPC >$TMPE 2>&1 || return
                 sed -n "/cc1.*-target-cpu /{
                             s/.*-target-cpu \\([^ ]*\\).*/\\1/
@@ -7714,7 +7729,25 @@ if enabled icc; then
             disable aligned_stack
     fi
 elif enabled gcc; then
-    check_optflags -fno-tree-vectorize
+    gcc_version=$($cc -dumpversion)
+    major_version=${gcc_version%%.*}
+    if [ $major_version -lt 13 ]; then
+        # Disable tree-vectorize for GCC <13 - it has historically been buggy.
+        check_optflags -fno-tree-vectorize
+    else
+        case $arch in
+        x86|arm|aarch64)
+            # Allow the default of having tree-vectorize enabled on well tested
+            # architectures.
+            ;;
+        *)
+            # Disable tree-vectorize on potentially less tested
+            # architectures. Known issues:
+            # - https://gcc.gnu.org/PR121064 on Loongarch
+            check_optflags -fno-tree-vectorize
+            ;;
+        esac
+    fi
     check_cflags -Werror=format-security
     check_cflags -Werror=implicit-function-declaration
     check_cflags -Werror=missing-prototypes

doc/filters.texi

@@ -16348,6 +16348,36 @@ Render frames with rounded corners. The value, given as a float ranging from
 square to fully circular. In other words, it gives the radius divided by half
 the smaller side length. Defaults to @code{0.0}.
 
+@item lut
+Specifies a custom LUT (in Adobe .cube format) to apply to the colors
+as part of color conversion. The exact interpretation depends on the value
+of @option{lut_type}.
+
+@item lut_type
+Controls the interpretation of color values fed to and from the LUT
+specified as @option{lut}. Valid values are:
+
+@table @samp
+@item auto
+Chooses the interpretation of the LUT automatically from tagged
+metadata, and otherwise falls back to @samp{native}. (Default)
+
+@item native
+Applied to raw image contents in its native RGB colorspace (non-linear
+light), before conversion to the output color space.
+
+@item normalized
+Applied to the normalized RGB image contents, in linear light, before
+conversion to the output color space.
+
+@item conversion
+Fully replaces the conversion from the image color space to the output
+color space. If such a LUT is present, it has the highest priority, and
+overrides any ICC profiles, as well as options related to tone mapping
+and output colorimetry (@option{color_primaries}, @option{color_trc}).
+
+@end table
+
 @item extra_opts
 Pass extra libplacebo internal configuration options. These can be specified
 as a list of @var{key}=@var{value} pairs separated by ':'. The following example

libavcodec/dsicinaudio.c

@@ -81,6 +81,8 @@ static av_cold int cinaudio_decode_init(AVCodecContext *avctx)
     cin->initial_decode_frame = 1;
     cin->delta                = 0;
     avctx->sample_fmt         = AV_SAMPLE_FMT_S16;
+    if (!avctx->sample_rate)
+        avctx->sample_rate = 8000;
     av_channel_layout_uninit(&avctx->ch_layout);
     avctx->ch_layout          = (AVChannelLayout)AV_CHANNEL_LAYOUT_MONO;
 

libavcodec/evrcdec.c

@@ -239,6 +239,8 @@ static av_cold int evrc_decode_init(AVCodecContext *avctx)
     av_channel_layout_uninit(&avctx->ch_layout);
     avctx->ch_layout = (AVChannelLayout)AV_CHANNEL_LAYOUT_MONO;
     avctx->sample_fmt     = AV_SAMPLE_FMT_FLT;
+    if (!avctx->sample_rate)
+        avctx->sample_rate = 8000;
 
     for (i = 0; i < FILTER_ORDER; i++) {
         e->prev_lspf[i] = (i + 1) * 0.048;

libavcodec/g722dec.c

@@ -63,6 +63,8 @@ static av_cold int g722_decode_init(AVCodecContext * avctx)
     av_channel_layout_uninit(&avctx->ch_layout);
     avctx->ch_layout      = (AVChannelLayout)AV_CHANNEL_LAYOUT_MONO;
     avctx->sample_fmt     = AV_SAMPLE_FMT_S16;
+    if (!avctx->sample_rate)
+        avctx->sample_rate = 16000;
 
     c->band[0].scale_factor = 8;
     c->band[1].scale_factor = 2;