swscale/x86/yuv2rgb_template: Add emms to MMX(EXT) functions

Fixes issue #22333. Note: These functions have been removed in commit 61e851381f, so the issue only affects releases 7.0 and older. Signed-off-by: Andreas Rheinhardt <andreas.rheinhardt@outlook.com> (cherry picked from commit 5e6c584b98cea9b9d78b037728b915887758666d)
swscale/output: Fix integer overflow in yuv2ya16_X_c_template()
2026-03-05 14:59:24 +01:00 · 2026-01-02 21:58:51 +00:00 · 2026-01-02 21:58:51 +00:00 · 2026-01-02 21:58:51 +00:00 · 2026-01-02 21:58:51 +00:00 · 2026-01-02 21:58:51 +00:00
533 changed files with 7755 additions and 4802 deletions
@@ -0,0 +1,23 @@
+exclude: ^tests/ref/
+
+repos:
+- repo: https://github.com/pre-commit/pre-commit-hooks
+  rev: v5.0.0
+  hooks:
+    - id: check-case-conflict
+    - id: check-executables-have-shebangs
+    - id: check-illegal-windows-names
+    - id: check-shebang-scripts-are-executable
+    - id: check-yaml
+    - id: end-of-file-fixer
+    - id: fix-byte-order-marker
+    - id: mixed-line-ending
+    - id: trailing-whitespace
+- repo: local
+  hooks:
+    - id: aarch64-asm-indent
+      name: fix aarch64 assembly indentation
+      files: ^.*/aarch64/.*\.S$
+      language: script
+      entry: ./tools/check_arm_indent.sh --apply
+      pass_filenames: false
@@ -0,0 +1,29 @@
+name: Lint
+
+on:
+  push:
+    branches:
+      - release/4.3
+  pull_request:
+
+jobs:
+  lint:
+    name: Pre-Commit
+    runs-on: utilities
+    steps:
+      - name: Checkout
+        uses: actions/checkout@v4
+      - name: Install pre-commit CI
+        id: install
+        run: |
+            python3 -m venv ~/pre-commit
+            ~/pre-commit/bin/pip install --upgrade pip setuptools
+            ~/pre-commit/bin/pip install pre-commit
+            echo "envhash=$({ python3 --version && cat .forgejo/pre-commit/config.yaml; } | sha256sum | cut -d' ' -f1)" >> $FORGEJO_OUTPUT
+      - name: Cache
+        uses: actions/cache@v4
+        with:
+          path: ~/.cache/pre-commit
+          key: pre-commit-${{ steps.install.outputs.envhash }}
+      - name: Run pre-commit CI
+        run: ~/pre-commit/bin/pre-commit run -c .forgejo/pre-commit/config.yaml --show-diff-on-failure --color=always --all-files
@@ -0,0 +1,80 @@
+name: Test
+
+on:
+  push:
+    branches:
+      - release/4.3
+  pull_request:
+
+jobs:
+  run_fate:
+    name: Fate (${{ matrix.runner }}, ${{ matrix.shared }}, ${{ matrix.bits }} bit)
+    strategy:
+      fail-fast: false
+      matrix:
+        runner: [linux-aarch64]
+        shared: ['static']
+        bits: ['64']
+        include:
+          - runner: linux-amd64
+            shared: 'static'
+            bits: '32'
+          - runner: linux-amd64
+            shared: 'shared'
+            bits: '64'
+    runs-on: ${{ matrix.runner }}
+    steps:
+      - name: Checkout
+        uses: actions/checkout@v4
+      - name: Configure
+        run: |
+          ./configure --enable-gpl --enable-nonfree --enable-memory-poisoning --assert-level=2 \
+              $([ "${{ matrix.bits }}" != "32" ] || echo --arch=x86_32 --extra-cflags=-m32 --extra-cxxflags=-m32 --extra-ldflags=-m32) \
+              $([ "${{ matrix.shared }}" != "shared" ] || echo --enable-shared --disable-static) \
+              || CFGRES=$? && CFGRES=$?
+          cat ffbuild/config.log
+          exit $CFGRES
+      - name: Build
+        run: make -j$(nproc)
+      - name: Restore Cached Fate-Suite
+        id: cache
+        uses: actions/cache/restore@v4
+        with:
+          path: fate-suite
+          key: fate-suite
+          restore-keys: |
+            fate-suite-
+      - name: Sync Fate-Suite
+        id: fate
+        run: |
+          make fate-rsync SAMPLES=$PWD/fate-suite
+          echo "hash=$(find fate-suite -type f -printf "%P %s %T@\n" | sort | sha256sum | cut -d' ' -f1)" >> $FORGEJO_OUTPUT
+      - name: Cache Fate-Suite
+        uses: actions/cache/save@v4
+        if: ${{ format('fate-suite-{0}', steps.fate.outputs.hash) != steps.cache.outputs.cache-matched-key }}
+        with:
+          path: fate-suite
+          key: fate-suite-${{ steps.fate.outputs.hash }}
+      - name: Run Fate
+        run: LD_LIBRARY_PATH="$(printf "%s:" "$PWD"/lib*)$PWD" make fate fate-build SAMPLES=$PWD/fate-suite -j$(nproc)
+  compile_only:
+    name: Fate (Win64, Build-Only)
+    strategy:
+      fail-fast: false
+      matrix:
+        image: ["ghcr.io/btbn/ffmpeg-builds/win64-gpl-4.3:latest"]
+    runs-on: linux-amd64
+    container: ${{ matrix.image }}
+    steps:
+      - name: Checkout
+        uses: actions/checkout@v4
+      - name: Configure
+        run: |
+          ./configure --pkg-config-flags="--static" $FFBUILD_TARGET_FLAGS $FF_CONFIGURE \
+              --cc="$CC" --cxx="$CXX" --ar="$AR" --ranlib="$RANLIB" --nm="$NM" \
+              --extra-cflags="$FF_CFLAGS" --extra-cxxflags="$FF_CXXFLAGS" \
+              --extra-libs="$FF_LIBS" --extra-ldflags="$FF_LDFLAGS" --extra-ldexeflags="$FF_LDEXEFLAGS"
+      - name: Build
+        run: make -j$(nproc)
+      - name: Run Fate
+        run: make -j$(nproc) fate-build
@@ -55,7 +55,7 @@ modified by someone else and passed on, the recipients should know
 that what they have is not the original version, so that the original
 author's reputation will not be affected by problems that might be
 introduced by others.
-
+
  Finally, software patents pose a constant threat to the existence of
 any free program.  We wish to make sure that a company cannot
 effectively restrict the users of a free program by obtaining a
@@ -111,7 +111,7 @@ modification follow.  Pay close attention to the difference between a
 "work based on the library" and a "work that uses the library".  The
 former contains code derived from the library, whereas the latter must
 be combined with the library in order to run.
-
+
                  GNU LESSER GENERAL PUBLIC LICENSE
   TERMS AND CONDITIONS FOR COPYING, DISTRIBUTION AND MODIFICATION

@@ -158,7 +158,7 @@ Library.
  You may charge a fee for the physical act of transferring a copy,
 and you may at your option offer warranty protection in exchange for a
 fee.
-
+
  2. You may modify your copy or copies of the Library or any portion
 of it, thus forming a work based on the Library, and copy and
 distribute such modifications or work under the terms of Section 1
@@ -216,7 +216,7 @@ instead of to this License.  (If a newer version than version 2 of the
 ordinary GNU General Public License has appeared, then you can specify
 that version instead if you wish.)  Do not make any other change in
 these notices.
-
+
  Once this change is made in a given copy, it is irreversible for
 that copy, so the ordinary GNU General Public License applies to all
 subsequent copies and derivative works made from that copy.
@@ -267,7 +267,7 @@ Library will still fall under Section 6.)
 distribute the object code for the work under the terms of Section 6.
 Any executables containing that work also fall under Section 6,
 whether or not they are linked directly with the Library itself.
-
+
  6. As an exception to the Sections above, you may also combine or
 link a "work that uses the Library" with the Library to produce a
 work containing portions of the Library, and distribute that work
@@ -329,7 +329,7 @@ restrictions of other proprietary libraries that do not normally
 accompany the operating system.  Such a contradiction means you cannot
 use both them and the Library together in an executable that you
 distribute.
-
+
  7. You may place library facilities that are a work based on the
 Library side-by-side in a single library together with other library
 facilities not covered by this License, and distribute such a combined
@@ -370,7 +370,7 @@ subject to these terms and conditions.  You may not impose any further
 restrictions on the recipients' exercise of the rights granted herein.
 You are not responsible for enforcing compliance by third parties with
 this License.
-
+
  11. If, as a consequence of a court judgment or allegation of patent
 infringement or for any other reason (not limited to patent issues),
 conditions are imposed on you (whether by court order, agreement or
@@ -422,7 +422,7 @@ conditions either of that version or of any later version published by
 the Free Software Foundation.  If the Library does not specify a
 license version number, you may choose any version ever published by
 the Free Software Foundation.
-
+
  14. If you wish to incorporate parts of the Library into other free
 programs whose distribution conditions are incompatible with these,
 write to the author to ask for permission.  For software which is
@@ -456,7 +456,7 @@ SUCH HOLDER OR OTHER PARTY HAS BEEN ADVISED OF THE POSSIBILITY OF SUCH
 DAMAGES.

                     END OF TERMS AND CONDITIONS
-
+
           How to Apply These Terms to Your New Libraries

  If you develop a new library, and you want it to be of the greatest
@@ -1,6 +1,6 @@
-See the Git history of the project (git://source.ffmpeg.org/ffmpeg) to
+See the Git history of the project (https://git.ffmpeg.org/ffmpeg) to
 get the names of people who have contributed to FFmpeg.

 To check the log, you can type the command "git log" in the FFmpeg
 source directory, or browse the online repository at
-http://source.ffmpeg.org.
+https://git.ffmpeg.org/ffmpeg
@@ -1,6 +1,922 @@
 Entries are sorted chronologically from oldest to youngest within each release,
 releases are sorted from youngest to oldest.

+
+version 4.3.9:
+ configure: update copyright year
+ avformat/hls: Partially revert "reduce default max reload to 3"
+ avformat/hls: Fix twitter
+ libavformat/hls: Be more restrictive on mpegts extensions
+ avformat/hls: .ts is always ok even if its a mov/mp4
+ avformat/hls: Print input format in error message
+ avformat/hls: Be more picky on extensions
+ avformat: add ff_match_url_ext()
+ avfilter/bwdif: account for chroma sub-sampling in min size calculation
+ avformat/iff: Check that we have a stream in read_dst_frame()
+ avformat/mlvdec: fix size checks
+ avformat/mxfdec: Check edit unit for overflow in mxf_set_current_edit_unit()
+ avcodec/h263dec: Check against previous dimensions instead of coded
+ avformat/mxfdec: Check avio_read() success in mxf_decrypt_triplet()
+ avcodec/huffyuvdec: Initialize whole output for decode_gray_bitstream()
+ avformat/ipmovie: Check signature_buffer read
+ avformat/wtvdec: Initialize buf
+ avcodec/cbs_vp9: Initialize VP9RawSuperframeIndex
+ avformat/vqf: Propagate errors from add_metadata()
+ avformat/vqf: Check avio_read() in add_metadata()
+ avformat/dashdec: Check whitelist
+ avutil/avstring: dont mess with NULL pointers in av_match_list()
+ avcodec/mpegvideo_enc: Check FLV1 resolution limits
+ avcodec/ffv1enc: Fix handling of 32bit unsigned symbols
+ avcodec/vc1dec: Clear block_index in vc1_decode_reset()
+ avcodec/aacsbr_template: Clear n_q on error
+ swscale/output: Fix undefined overflow in yuv2rgba64_full_X_c_template()
+ avfilter/af_pan: Fix sscanf() use
+ avfilter/vf_addroi: Add missing NULL termination to addroi_var_names[]()
+ avformat/rmdec: check that buf if completely filled
+ avcodec/hapdec: Clear tex buffer
+ avformat/mxfdec: Check that key was read sucessfull
+ avformat/rpl: Fix check for negative values
+ avformat/mlvdec: Check avio_read()
+ avcodec/utils: Fix block align overflow for ADPCM_IMA_WAV
+ avformat/matroskadec: Check pre_ns for overflow
+ avcodec/webp: Check ref_x/y
+ avcodec/ilbcdec: Initialize tempbuff2
+ avformat/dxa: check bpc
+ swscale/slice: clear allocated memory in alloc_lines()
+ avformat/icodec: fix integer overflow with nb_pal
+ doc/developer: Document relationship between git accounts and MAINTAINERS
+ avformat/vividas: Check avio_read() for failure
+ avformat/ilbc: Check avio_read() for failure
+ avformat/nistspheredec: Clear buffer
+ INSTALL: explain the circular dependency issue and solution
+ avformat/mpegts: Initialize predefined_SLConfigDescriptor_seen
+ avformat/mxfdec: Fix overflow in midpoint computation
+ swscale/output: used unsigned for bit accumulation
+ avcodec/rangecoder: only perform renorm check/loop for callers that need it
+ avcodec/ffv1dec: Fix end computation with ec=2
+ avcodec/ffv1enc: Prevent generation of files with broken slices
+ avformat/matroskadec: Check desc_bytes so bits fit in 64bit
+ avcodec/ffv1enc: Correct error message about unsupported version
+ avcodec/ffv1enc: Slice combination is unsupported
+ avcodec/ffv1enc: 2Pass mode is not possible with golomb coding
+ avcodec/ffv1enc: Fix >8bit context size
+ avcodec/xan: Add basic input size check
+ avcodec/svq3: Check for minimum size input
+ avcodec/eacmv: Check input size for intra frames
+ avcodec/jfdctint_template: use unsigned z* in row_fdct()
+ avformat/mxfdec: Check timecode for overflow
+ avformat/mxfdec: More offset_temp checks
+ swscale/output: Fix undefined integer overflow in yuv2rgba64_2_c_template()
+ swscale/swscale: Use unsigned operation to avoid undefined behavior
+ avcodec/vc2enc: basic sanity check on slice_max_bytes
+ avformat/mvdec: Check if name was fully read
+ avcodec/wmavoice: Do not use uninitialized pitch[0]
+ avcodec/notchlc: Check bytes left before reading
+ avcodec/vc1_block: propagate error codes
+ avformat/apetag: Check APETAGEX
+ avcodec/avcodec: Warn about data returned from get_buffer*()
+ avcodec/aic: Clear slice_data
+ avcodec/vc1dec: Clear mb_type_base and ttblk_base
+ avcodec/shorten: clear padding
+ avformat/mpeg: Check an avio_read() for failure
+ avcodec/mvha: Clear remaining space after inflate()
+ avformat/segafilm: Set keyframe
+ avcodec/dxva2: initialize hr in ff_dxva2_common_end_frame()
+ avcodec/dxva2: initialize validate
+ avcodec/dxva2: Initialize ConfigBitstreamRaw
+ avcodec/dxva2: Initialize dxva_size and check it
+ avfilter/vf_xfade: Compute w2, h2 with float
+ avfilter/vf_v360: Assert that vf was initialized
+ avfilter/vf_tonemap_opencl: Dereference after NULL check
+ avfilter/vf_xfade_opencl: Check ff_inlink_consume_frame() for failure
+ avformat/lmlm4: Eliminate some AVERROR(EIO)
+ avformat/wtvdec: Check length of read mpeg2_descriptor
+ avformat/wtvdec: clear sectors
+ vp9: recon: Use emulated edge to prevent buffer overflows
+ arm: vp9mc: Load only 12 pixels in the 4 pixel wide horizontal filter
+ aarch64: vp9mc: Load only 12 pixels in the 4 pixel wide horizontal filter
+ avformat/libzmq: fix check for zmq protocol prefix
+ configure: improve check for POSIX ioctl
+ configure: restore autodetection of v4l2 and fbdev
+ configure: use just the pkg-config for sndio
+ configure: enable ffnvcodec, nvenc, nvdec for FreeBSD
+ avutil/ppc/cpu: Also use the machdep.altivec sysctl on NetBSD
+ avutil/ppc/cpu: Use proper header for OpenBSD PPC CPU detection
+ lavd/v4l2: Use proper field type for second parameter of ioctl() with BSD's
+ configure: use pkg-config for sndio
+ libavcodec/arm/mlpdsp_armv5te: fix label format to work with binutils 2.43
+
+
+version 4.3.8:
+ avcodec/parser: ensure input padding is zeroed
+ avformat/img2dec: Clear padding data after EOF
+ avformat/wavdec: Check if there are 16 bytes before testing them
+ avcodec/snow: Fix off by 1 error in run_buffer
+ avcodec/utils: apply the same alignment to YUV410 as we do to YUV420 for snow
+ update for 4.3.8
+ avcodec/diracdsp: Remove unused variable
+ avcodec/vaapi_encode: Check hwctx
+ avcodec/proresdec: Consider negative bits left
+ avcodec/hevc/hevcdec: Do not allow slices to depend on failed slices
+ avutil/slicethread: Check pthread_*_init() for failure
+ avutil/frame: Check log2_crop_align
+ avutil/buffer: Check ff_mutex_init() for failure
+ avformat/xmv: Check this_packet_size
+ avformat/ty: rec_size seems to only need 32bit
+ avformat/tty: Check avio_size()
+ avformat/siff: Basic pkt_size check
+ avformat/sauce: Check avio_size() for failure
+ avformat/sapdec: Check ffurl_get_file_handle() for error
+ avformat/nsvdec: Check asize for PCM
+ avformat/mp3dec: Check header_filesize
+ avformat/mp3dec; Check for avio_size() failure
+ avformat/mov: Use 64bit for str_size
+ avformat/mm: Check length
+ avformat/hnm: Check *chunk_size
+ avformat/hlsenc: Check ret
+ avformat/bintext: Check avio_size() return
+ avformat/asfdec_o: Check size of index object
+ avfilter/scale_eval: Use 64bit, check values in ff_scale_adjust_dimensions()
+ avfilter/vf_lut3d: Check av_scanf()
+ avfilter/vf_deshake_opencl: Ensure that the first iteration initializes the best variables
+ swscale/output: Fix integer overflows in yuv2rgba64_X_c_template
+ avformat/mxfdec: Reorder elements of expression in bisect loop
+ avcodec/pnmdec: Use 64bit for input size check
+ avcodec/utvideoenc: Use unsigned shift to build flags
+ avcodec/vc2enc: Fix overflows with storing large values
+ avcodec/mpegvideo_enc: Do not duplicate pictures on shifting
+ avcodec/tiff: Check value on positive signed targets
+ avfilter/vf_bm3d: Dont round MSE2SSE to an integer
+ avdevice/dshow: Check device_filter_unique_name before use
+ avdevice/dshow_filter: Use wcscpy_s()
+ avcodec/flac_parser: Assert that we do not overrun the link_penalty array
+ avcodec/pixlet: Simplify pfx computation
+ avcodec/motion_est: Fix score squaring overflow
+ avcodec/loco: Check loco_get_rice() for failure
+ avcodec/loco: check get_ur_golomb_jpegls() for failure
+ avcodec/imm4: check cbphi for error
+ avcodec/iff: Use signed count
+ avcodec/golomb: Assert that k is in the supported range for get_ur/sr_golomb()
+ avcodec/golomb: Document return for get_ur_golomb_jpegls() and get_sr_golomb_flac()
+ avcodec/dxv: Fix type in get_opcodes()
+ avcodec/xsubdec: Check parse_timecode()
+ avutil/imgutils: av_image_check_size2() ensure width and height fit in 32bit
+ avcodec/proresenc_kostya: use unsigned alpha for rotation
+ avformat/rtmppkt: Simplify and deobfuscate amf_tag_skip() slightly
+ avformat/rmdec: use 64bit for audio_framesize checks
+ avutil/hwcontext_d3d11va: correct sizeof IDirect3DSurface9
+ avutil/hwcontext_d3d11va: correct sizeof AVD3D11FrameDescriptor
+ avformat/tls_schannel: Initialize ret
+ avformat/subfile: Assert that whence is a known case
+ avformat/subfile: Merge if into switch()
+ avformat/rtsp: Check that lower transport is handled in one of the if()
+ avformat/rtsp: initialize reply1
+ avformat/rtsp: use < 0 for error check
+ avformat/rtpenc_vc2hq: Check sizes
+ avfilter/af_aderivative: Free out on error
+ avfilter/af_pan: check nb_output_channels before use
+ cbs_av1: Reject thirty-two zero bits in uvlc code
+ tools/coverity: Phase 1 study of anti-halicogenic for coverity av_rescale()
+ avfilter/vf_avgblur: Check plane instead of AVFrame
+ avformat/rdt: Check pkt_len
+ avformat/mpeg: Check len in mpegps_probe()
+ avdevice/dshow: Check ICaptureGraphBuilder2_SetFiltergraph() for failure
+ avcodec/mfenc: check IMFSample_ConvertToContiguousBuffer() for failure
+ avcodec/vc1_loopfilter: Factor duplicate code in vc1_b_h_intfi_loop_filter()
+ avformat/img2dec: assert no pipe on ts_from_file
+ avcodec/cbs_jpeg: Try to move the read entity to one side in a test
+ avformat/mov: Check edit list for overflow
+ fftools/ffmpeg: Check read() for failure
+ swscale/output: Avoid undefined overflow in yuv2rgb_write_full()
+ swscale/output: alpha can become negative after scaling, use multiply
+ avcodec/targaenc: Allocate space for the palette
+ avcodec/r210enc: Use av_rescale for bitrate
+ avcodec/jfdctint_template: Fewer integer anomalies
+ avcodec/snowenc: MV limits due to mv_penalty table size
+ avformat/mxfdec: Check container_ul->desc before use
+ MAINTAINERS: Update the entries for the release maintainer for FFmpeg
+
+
+version 4.3.7:
+ avfilter/vf_rotate: Check ff_draw_init2() return value
+ avformat/matroskadec: Assert that num_levels is non negative
+ avformat/libzmq: Check av_strstart()
+ avformat/img2dec: Move DQT after unrelated if()
+ avdevice/xcbgrab: Check sscanf() return
+ fftools/cmdutils: Add protective () to FLAGS
+ avformat/sdp: Check before appending ","
+ avcodec/ilbcdec: Remove dead code
+ avcodec/vp8: Check cond init
+ avcodec/vp8: Check mutex init
+ avcodec/notchlc: Check init_get_bits8() for failure
+ avcodec/tests/dct: Use 64bit in intermediate for error computation
+ avcodec/scpr3: Check add_dec() for failure
+ avcodec/wavpackenc: Use unsigned for potential 31bit shift
+ avcodec/tests/jpeg2000dwt: Use 64bit in comparission
+ avcodec/tests/jpeg2000dwt: Use 64bit in err2 computation
+ avformat/fwse: Remove always false expression
+ avformat/asfdec_f: Use 64bit for preroll computation
+ avformat/ape: Use 64bit for final frame size
+ avcodec/tiff: Assert init_get_bits8() success in unpack_gray()
+ avcodec/tiff: Assert init_get_bits8() success in horizontal_fill()
+ swscale/yuv2rgb: Use 64bit for brightness computation
+ avutil/tests/opt: Check av_set_options_string() for failure
+ avutil/tests/dict: Check av_dict_set() before get for failure
+ avdevice/dshow: fix badly indented line
+ avcodec/mscc & mwsc: Check loop counts before use
+ avcodec/mpegvideo_enc: Fix potential overflow in RD
+ avcodec/mpeg4videodec: assert impossible wrap points
+ avcodec/vble: Check av_image_get_buffer_size() for failure
+ avcodec/vp3: Replace check by assert
+ avcodec/jpeg2000dec: remove ST=3 case
+ avcodec/qsvdec: Check av_image_get_buffer_size() for failure
+ avcodec/fmvc: remove dead assignment
+ avcodec/h264_slice: Remove dead sps check
+ avcodec/lpc: copy levenson coeffs only when they have been computed
+ avutil/tests/base64: Check with too short output array
+ libavutil/base64: Try not to write over the array end
+ avcodec/cbs_av1: Avoid shift overflow
+ doc/examples/demux_decode: Simplify loop
+ avcodec/mpegvideo_enc: Fix 1 line and one column images
+ swscale/output: Fix integer overflow in yuv2rgba64_full_1_c_template()
+ swscale/output: Fix integer overflow in yuv2rgba64_1_c_template
+ avformat/mxfdec: Check body_offset
+ avformat/kvag: Check sample_rate
+ avcodec/ac3_parser: Check init_get_bits8() for failure
+ avcodec/hevcdec: Check ref frame
+ doc/examples/vaapi_transcode: Simplify loop
+ avfilter/vf_thumbnail_cuda: Set ret before checking it
+ avfilter/signature_lookup: Dont copy uninitialized stuff around
+ avfilter/signature_lookup: Fix 2 differences to the refernce SW
+ avcodec/x86/vp3dsp_init: Set correct function pointer, fix crash
+ fate/subtitles: Ignore line endings for sub-scc test
+ avformat/mxfdec: Check index_edit_rate
+ swscale/utils: Fix xInc overflow
+ avformat/mxfdec: Make edit_unit_byte_count unsigned
+ avformat/movenc: Check that cts fits in 32bit
+ avformat/mxfdec: Check first case of offset_temp computation for overflow
+ avfilter/vf_signature: Dont crash on no frames
+ avformat/westwood_vqa: Fix 2g packets
+ avformat/matroskadec: Check timescale
+ avformat/wavdec: satuarte next_tag_ofs, data_end
+ avformat/sbgdec: Check for negative duration
+ avformat/rpl: Use 64bit for total_audio_size and check it
+ avformat/timecode: use 64bit for intermediate for rounding in fps_from_frame_rate()
+ avformat/jacosubdec: Use 64bit for abs
+ avformat/concatdec: Check user_duration sum
+ avcodec/truemotion1: Height not being a multiple of 4 is unsupported
+ avcodec/hcadec: do not set hfr_group_count to invalid values
+ avformat/concatdec: clip outpoint - inpoint overflow in get_best_effort_duration()
+ avformat/jacosubdec: clarify code
+ avformat/cafdec: Check that data chunk end fits within 64bit
+ avformat/iff: Saturate avio_tell() + 12
+ avformat/dxa: Adjust order of operations around block align
+ avformat/cafdec: dont seek beyond 64bit
+ avformat/id3v2: read_uslt() check for the amount read
+ avcodec/proresenc_kostya: Remove bug similarity text
+ avcodec/vorbisdec: Check remaining data in vorbis_residue_decode_internal()
+ libswscale/utils: Fix bayer to yuvj
+ swscale/swscale: Check srcSliceH for bayer
+ swscale/utils: Allocate more dithererror
+ avcodec/indeo3: Round dimensions up in allocate_frame_buffers()
+ avutil/rational: Document what is to be expected from av_d2q() of doubles representing rational numbers
+ avfilter/signature_lookup: Do not dereference NULL pointers after malloc failure
+ avfilter/signature_lookup: dont leave uncleared pointers in sll_free()
+ avcodec/mpegvideo_enc: Use ptrdiff_t for stride
+ libavformat/hlsenc.c: Populate OTI using AAC profile in write_codec_attr.
+ avcodec/mpegvideo_enc: Dont copy beyond the image
+ avfilter/vf_minterpolate: Check pts before division
+ avformat/flacdec: Avoid double AVERRORS
+ avfilter/vf_vidstabdetect: Avoid double AVERRORS
+ avfilter/vf_swaprect: round coordinates down
+ avfilter/vf_swaprect: Use height for vertical variables
+ avfilter/vf_swaprect: assert that rectangles are within memory
+ avfilter/af_alimiter: Check nextpos before use
+ avfilter/af_stereowiden: Check length
+ avfilter/vf_weave: Fix odd height handling
+ avfilter/vf_gradfun: Do not overread last line
+ avformat/mov: do not set sign bit for chunk_offsets
+ avcodec/jpeglsdec: Check Jpeg-LS LSE
+ configure: Enable section_data_rel_ro for FreeBSD and NetBSD aarch64 / arm
+ avformat/mov: Ignore duplicate ftyp
+ avformat/mov: Fix integer overflow in mov_read_packet().
+ seek: Fix crashes in ff_seek_frame_binary if built with latest Clang 14
+ avcodec/4xm: Check for cfrm exhaustion
+ avformat/mov: Disallow FTYP after streams
+ doc/html: fix styling issue with Texinfo 7.0
+ doc/html: support texinfo 7.0
+ doc/t2h.pm: fix missing TOC with texinfo 6.8 and above
+ doc/t2h.pm: fix missing CSS with texinfo 6.8 and above
+ avformat/matroskadec: Fix declaration-after-statement warnings
+ avformat/rtsp: Use rtsp_st->stream_index
+ avcodec/jpeg2000dec: Check image offset
+ avformat/mxfdec: Check klv offset
+ libavutil/ppc/cpu.c: check that AT_HWCAP2 is defined
+ avcodec/h2645_parse: Avoid EAGAIN
+ avcodec/xvididct: Make c* unsigned to avoid undefined overflows
+ avformat/tmv: Check video chunk size
+ avcodec/h264_parser: saturate dts a bit
+ avformat/asfdec_f: Saturate presentation time in marker
+ avformat/xwma: sanity check bits_per_coded_sample
+ avformat/matroskadec: Check prebuffered_ns for overflow
+ avformat/wavdec: Check left avio_tell for overflow
+ avformat/tta: Better totalframes check
+ avformat/rpl: Check for number_of_chunks overflow
+ avformat/mov: compute absolute dts difference without overflow in mov_find_next_sample()
+ avformat/jacosubdec: Check timeres
+ avformat/jacosubdec: avoid signed integer overflows in get_shift()
+ avformat/jacosubdec: Factorize code in get_shift() a bit
+ avcodec/escape124: Do not return random numbers
+ avformat/avs: Check if return code is representable
+ avcodec/lcldec: Make PNG filter addressing match the code afterwards
+ avformat/westwood_vqa: Check chunk size
+ avformat/sbgdec: Check for period overflow
+ avformat/concatdec: Check in/outpoint for overflow
+ avcodec/xvididct: Fix integer overflow in idct_row()
+ avcodec/celp_math: avoid overflow in shift
+ tools/target_dec_fuzzer: Adjust threshold for rtv1
+ avformat/hls: reduce default max reload to 3
+ avformat/format: Stop reading data at EOF during probing
+ avcodec/huffyuvdec: avoid undefined behavior with get_vlc2() failure
+ avcodec/cscd: Fix "CamStudio Lossless Codec 1.0" gzip files
+ avcodec/cscd: Check for CamStudio Lossless Codec 1.0 behavior in end check of LZO files
+ avcodec/hevcdec: Fix undefined memcpy()
+ avcodec/mpeg4videodec: more unsigned in amv computation
+ avcodec/tta: fix signed overflow in decorrelate
+ avcodec/apedec: Fix CRC for 24bps and bigendian
+ avcodec/xvididct: Fix integer overflow in idct_row()
+ avformat/avr: Check sample rate
+ avcodec/jpeg2000dec: Check for reduction factor and image offset
+ avutil/softfloat: Basic documentation for av_sincos_sf()
+ avutil/softfloat: fix av_sincos_sf()
+ avcodec/utils: fix 2 integer overflows in get_audio_frame_duration()
+ avcodec/hevcdec: Avoid null pointer dereferences in MC
+ avcodec/takdsp: Fix integer overflows
+ avcodec: Ignoring errors is only possible before the input end
+ avcodec/noise_bsf: Check for wrapped frames
+ avformat/oggparsetheora: clip duration within 64bit
+ avformat/wavdec: Check that smv block fits in available space
+ avcodec/tiff: add a zero DNG_LINEARIZATION_TABLE check
+ avcodec/tak: Check remaining bits in ff_tak_decode_frame_header()
+ avcodec/utils: the IFF_ILBM implementation assumes that there are a multiple of 16 allocated
+ avcodec/pngdec: Do not pass AVFrame into global header decode
+ avcodec/vorbisdec: Check codebook float values to be finite
+ avcodec/g2meet: Replace fake allocation avoidance for framebuf
+ avcodec/lcldec: More space for rgb24
+ avcodec/lcldec: Support 4:1:1 and 4:2:2 with odd width
+ libavcodec/lcldec: width and height should not be unsigned
+ avcodec/x86/mathops: clip constants used with shift instructions within inline assembly
+ lavc/pthread_frame: avoid leaving stale hwaccel state in worker threads
+ avformat/mov: Check if a key is longer than the atom containing it
+ avcodec/nvdec: reset bitstream_len/nb_slices when resetting bitstream pointer
+ avformat/mov: don't abort on duplicate Mastering Display Metadata boxes
+ avcodec/vdpau_mpeg4: fix order of quant matrix coefficients
+ avcodec/vdpau_mpeg12: fix order of quant matrix coefficients
+ avcodec/nvdec_mpeg4: fix order of quant matrix coefficients
+ avcodec/nvdec_mpeg2: fix order of quant matrix coefficients
+ avformat/nutdec: Add check for avformat_new_stream
+
+version 4.3.6:
+- avcodec/escape124: Check that blocks are allocated before use
+- avcodec/huffyuvdec: Fix undefined behavior with shift
+- avcodec/j2kenc: Replace RGB24 special case by generic test
+- avcodec/j2kenc: Fix funky bpno errors on decoding
+- avcodec/j2kenc: remove misleading pred value
+- avcodec/j2kenc: fix 5/3 DWT identifer
+- avcodec/vp3: Check width to avoid assertion failure
+- avcodec/g729postfilter: Limit shift in long term filter
+- configure: update copyright year
+- avcodec/tests/snowenc: Fix 2nd test
+- avcodec/tests/snowenc: return a failure if DWT/IDWT mismatches
+- avcodec/snowenc: Fix visual weight calculation
+- avcodec/tests/snowenc: unbreak DWT tests
+- avcodec/escape124: Fix some return codes
+- avcodec/escape124: fix signdness of end of input check
+- Use https for repository links
+- avcodec/motionpixels: Mask pixels to valid values
+- avcodec/xpmdec: Check size before allocation to avoid truncation
+- avcodec/bink: Avoid undefined out of array end pointers in binkb_decode_plane()
+- avcodec/bink: Fix off by 1 error in ref end
+- avcodec/utils: Ensure linesize for SVQ3
+- avcodec/utils: allocate a line more for VC1 and WMV3
+- avcodec/videodsp_template: Adjust pointers to avoid undefined pointer things
+- avcodec/pngdec: Check deloco index more exactly
+- avcodec/ffv1dec: Check that num h/v slices is supported
+- avformat/mov: Check samplesize and offset to avoid integer overflow
+- avcodec/pictordec: Remove mid exit branch
+- avcodec/eac3dec: avoid float noise in fixed mode addition to overflow
+- avcodec/utils: use 32pixel alignment for bink
+- avcodec/scpr3: Check bx
+- avcodec/012v: Order operations for odd size handling
+- avcodec/eatgq: : Check index increments in tgq_decode_block()
+- avcodec/scpr: Test bx before use
+- avformat/mxfdec: Use 64bit in remainder
+- avcodec/sunrast: Fix maplength check
+- avcodec/wavpack: Avoid undefined shift in get_tail()
+- avcodec/wavpack: Check for end of input in wv_unpack_dsd_high()
+- avformat/id3v2: Check taglen in read_uslt()
+- avcodec/tiff: Ignore tile_count
+- avcodec/ffv1dec: restructure slice coordinate reading a bit
+- avcodec/mlpdec: Check max matrix instead of max channel in noise check
+- swscale/input: Use more unsigned intermediates
+- avcodec/alsdec: The minimal block is at least 7 bits
+- avformat/replaygain: avoid undefined / negative abs
+- swscale/output: Bias 16bps output calculations to improve non overflowing range
+- avcodec/speedhq: Check buf_size to be big enough for DC
+- avcodec/ffv1dec: Fail earlier if prior context is corrupted
+- avfilter/vf_untile: swap the chroma shift values used for plane offsets
+- avcodec/vp3: Add missing check for av_malloc
+- avcodec/nvenc: fix vbv buffer size in cq mode
+- avcodec/mjpegenc: take into account component count when writing the SOF header size
+- swscale: aarch64: Fix yuv2rgb with negative strides
+
+version 4.3.5:
+ avformat/vividas: Check packet size
+ avcodec/dstdec: Check for overflow in build_filter()
+ avformat/spdifdec: Use 64bit to compute bit rate
+ avformat/rpl: Use 64bit for duration computation
+ avformat/xwma: Use av_rescale() for duration computation
+ avformat/sdsdec: Use av_rescale() to avoid intermediate overflow in duration calculation
+ avformat/sbgdec: Check ts_int in genrate_intervals
+ avformat/rmdec: check tag_size
+ avformat/nutdec: Check fields
+ avformat/flvdec: Use 64bit for sum_flv_tag_size
+ avformat/jacosubdec: Fix overflow in get_shift()
+ avformat/dxa: avoid bpc overflows
+ avformat/cafdec: Check that nb_frasmes fits within 64bit
+ avformat/asfdec_o: Limit packet offset
+ avformat/ape: Check frames size
+ avformat/icodec: Check nb_pal
+ avformat/aiffdec: Use 64bit for block_duration use
+ avformat/aiffdec: Check block_duration
+ avformat/mxfdec: only probe max run in
+ avformat/mxfdec: Check run_in is within 65536
+ avcodec/mjpegdec: Check for unsupported bayer case
+ avcodec/apedec: Fix integer overflow in filter_3800()
+ avcodec/tta: Check 24bit scaling for overflow
+ avcodec/tiff: Fix loop detection
+ libavformat/hls: Free keys
+ avcodec/fmvc: Move frame allocation to a later stage
+ avfilter/vf_showinfo: remove backspaces
+ avcodec/speedhq: Check width
+ avcodec/bink: disallow odd positioned scaled blocks
+ avformat/asfdec_o: limit recursion depth in asf_read_unknown()
+ doc/git-howto.texi: Document commit signing
+ libavcodec/8bps: Check that line lengths fit within the buffer
+ avcodec/midivid: Perform lzss_uncompress() before ff_reget_buffer()
+ libavformat/iff: Check for overflow in body_end calculation
+ avformat/avidec: Prevent entity expansion attacks
+ avcodec/h263dec: Sanity check against minimal I/P frame size
+ avcodec/hevcdec: Check s->ref in the md5 path similar to hwaccel
+ avformat/subviewerdec: Make read_ts() more flexible
+ avcodec/mjpegdec: bayer and rct are incompatible
+ MAINTAINERS: Add ED25519 key for signing my commits in the future
+ avcodec/hevc_filter: copy_CTB() only within width&height
+ avformat/flvdec: Check for EOF in index reading
+ avformat/nutdec: Check get_packetheader() in mainheader
+ avformat/asfdec_f: Use 64bit for packet start time
+ tools/target_dec_fuzzer: Adjust threshold for MMVIDEO
+ avcodec/lagarith: Check dst/src in zero run code
+ avcodec/h264dec: Skip late SEI
+ avcodec/sbrdsp_fixed: Fix integer overflows in sbr_qmf_deint_neg_c()
+ avfilter/vf_signature: Fix integer overflow in filter_frame()
+ avformat/rtsp: break on unknown protocols
+ avcodec/hevcdsp_template: stay within tables in sao_band_filter()
+ avcodec/tiff: Check pixel format types for dng
+ avcodec/qpeldsp: copy less for the mc0x cases
+ avcodec/ffv1dec: Limit golomb rice coded slices to width 8M
+ avformat/iff: simplify duration calculation
+ avcodec/wnv1: Check for width =1
+ avcodec/ffv1dec_template: fix indention
+ avformat/sctp: close socket on errors
+ avcodec/aasc: Fix indention
+ avcodec/qdrw: adjust max colors to array size
+ avcodec/alacdsp: Make intermediates unsigned
+ avformat/aiffdec: cleanup size handling for extreem cases
+ avcodec/jpeglsdec: fix end check for xfrm
+ avcodec/cdgraphics: limit scrolling to the line
+ avformat/aiffdec: avoid integer overflow in get_meta()
+ avformat/ape: more bits in size for less overflows
+ avformat/bfi: Check offsets better
+ avformat/asfdec_f: Check packet_frag_timestamp
+ avcodec/texturedspenc: Fix indexing in color distribution determination
+ avformat/act: Check ff_get_wav_header() for failure
+ avcodec/libxavs2: Improve r redundancy in occured
+ avformat/libzmq: Improve r redundancy in occured
+ avfilter/vsrc_mandelbrot: Check for malloc failure
+ avfilter/vf_frei0r: Copy to frame allocated according to frei0r requirements
+ avfilter/video: Add ff_default_get_video_buffer2() to set specific alignment
+ avformat/genh: Check sample rate
+ avcodec/pngenc: remove monowhite from apng formats
+ configure: bump year
+ configure: extend SDL check to accept all 2.x versions
+ lavf/tls_mbedtls: add support for mbedtls version 3
+
+version 4.3.4:
+ fate: update reference files after the recent dash manifest muxer changes
+ avformat/webmdashenc: fix on-demand profile string
+ Update for FFmpeg 4.3.4
+ avcodec/diracdec: avoid signed integer overflow in global mv
+ avcodec/takdsp: Fix integer overflow in decorrelate_sf()
+ avcodec/apedec: fix a integer overflow in long_filter_high_3800()
+ avfilter/vf_subtitles: pass storage size to libass
+ avformat/aqtitledec: Skip unrepresentable durations
+ avformat/cafdec: Do not store empty keys in read_info_chunk()
+ avformat/mxfdec: Do not clear array in mxf_read_strong_ref_array() before writing
+ avformat/mxfdec: Check for avio_read() failure in mxf_read_strong_ref_array()
+ avformat/mxfdec: Check count in mxf_read_strong_ref_array()
+ avformat/hls: Check target_duration
+ avcodec/pixlet: Avoid signed integer overflow in scaling in filterfn()
+ avformat/matroskadec: Check pre_ns
+ avcodec/sonic: Use unsigned for predictor_k to avoid undefined behavior
+ avcodec/mjpegbdec: Set buf_size
+ avformat/matroskadec: Use rounded down duration in get_cue_desc() check
+ avcodec/g729_parser: Check channels
+ avformat/avidec: Check height
+ avformat/rmdec: Better duplicate tags check
+ avformat/mov: Disallow empty sidx
+ avformat/matroskadec: Check duration
+ avformat/mov: Corner case encryption error cleanup in mov_read_senc()
+ avcodec/jpeglsdec: Fix if( code style
+ avcodec/jpeglsdec: Check get_ur_golomb_jpegls() for error
+ avcodec/motion_est: fix indention of ff_get_best_fcode()
+ avcodec/motion_est: Fix xy indexing on range violation in ff_get_best_fcode()
+ avcodec/jpeglsdec: Increase range for N in ls_get_code_runterm() by using unsigned
+ avformat/matroskadec: Check desc_bytes
+ avformat/utils: Fix invalid NULL pointer operation in ff_parse_key_value()
+ avformat/matroskadec: Fix infinite loop with bz decompression
+ avformat/mov: Check size before subtraction
+ avcodec/apedec: Fix integer overflows in predictor_update_3930()
+ avcodec/apedec: fix integer overflow in 8bit samples
+ avformat/flvdec: timestamps cannot use the full int64 range
+ avcodec/vqavideo: reset accounting on error
+ avcodec/alacdsp: fix integer overflow in decorrelate_stereo()
+ avformat/4xm: Check for duplicate track ids
+ avformat/4xm: Consider max_streams on reallocating tracks array
+ avformat/mov: Check next offset in mov_read_dref()
+ avformat/vivo: Favor setting fps from explicit fractions
+ avformat/vivo: Do not use the general expression evaluator for parsing a floating point value
+ avformat/mxfdec: Check for duplicate mxf_read_index_entry_array()
+ avcodec/apedec: Change avg to uint32_t
+ avformat/mov: Disallow duplicate smdm
+ avformat/mov: Check for EOF in mov_read_glbl()
+ avcodec/vp3: Check version in all cases when VP4 code is not built
+ avformat/mov: Check channels for mov_parse_stsd_audio()
+ avformat/avidec: Check read_odml_index() for failure
+ avformat/aiffdec: Use av_rescale() for bitrate
+ avformat/aiffdec: sanity check block_align
+ avformat/aiffdec: Check sample_rate
+ avcodec/flac_parser: Consider AV_INPUT_BUFFER_PADDING_SIZE
+ avcodec/libdav1d: free the Dav1dData packet on dav1d_send_data() failure
+ configure: Add missing libshine->mpegaudioheader dependency
+
+version 4.3.3:
+ avcodec/ttadsp: Fix integer overflows in tta_filter_process_c()
+ avutil/mathematics: Document av_rescale_rnd() behavior on non int64 results
+ avformat/matroskadec: Reset state also on failure in matroska_reset_status()
+ avformat/wavdec: Check smv_block_size
+ avformat/rmdec: Check for multiple audio_stream_info
+ avcodec/apedec: Use 64bit to avoid overflow
+ avcodec/apedec: Fix undefined integer overflow in long_filter_ehigh_3830()
+ oavformat/avidec: Check offset in odml
+ avformat/mpegts: use actually read packet size in mpegts_resync special case
+ avfilter/scale_npp: fix non-aligned output frame dimensions
+ Update for 4.3.3
+ swscale/alphablend: Fix slice handling
+ avcodec/apedec: Fix integer overflow in filter_fast_3320()
+ avcodec/mxpegdec: Check for AVDISCARD_ALL
+ avcodec/flicvideo: Check remaining bytes in FLI*COPY
+ avcodec/cbs_h265_syntax_template: Limit sps_num_palette_predictor_initializer_minus1 to 127
+ avcodec/snowdec: Maintain avmv buffer
+ avcodec/mpeg12dec: Do not put mpeg_f_code into an invalid state on error return
+ avcodec/mpegvideo_enc: Limit bitrate tolerance to the representable
+ avcodec/apedec: Fix integer overflow in intermediate
+ avformat/mvdec: Do not set invalid sample rate
+ avformat/sbgdec: Check for t0 overflow in expand_tseq()
+ avformat/rmdec: Use 64bit for intermediate for DEINT_ID_INT4
+ avformat/sbgdec: Check opt_duration and start for overflow
+ avformat/mov: Check for duplicate clli
+ avformat/jacosubdec: Check for min in t overflow in get_shift()
+ avformat/mxfdec: check channel number in mxf_get_d10_aes3_packet()
+ avcodec/utils: don't return negative values in av_get_audio_frame_duration()
+ avcodec/jpeg2000dec: Check that atom header is within bytsetream
+ avcodec/apedec: Fix 2 integer overflows in filter_3800()
+ avcodec/xpmdec: Move allocations down after more error checks
+ avformat/mov: Check dts for overflow in mov_read_trun()
+ avformat/avidec: Use 64bit for frame number in odml index parsing
+ avcodec/mjpegbdec: Skip SOS on AVDISCARD_ALL as does mjpeg
+ avcodec/mjpegdec: Check for bits left in mjpeg_decode_scan_progressive_ac()
+ avformat/adtsenc: return value check for init_get_bits in adts_decode_extradata
+ avcodec/webp: Check available space in loop in decode_entropy_coded_image()
+ avcodec/h264dec: use picture parameters in ff_print_debug_info2()
+ avcodec/vc1dec: ff_print_debug_info() does not support WMV3 field_mode
+ avcodec/frame_thread_encoder: Free AVCodecContext structure on error during init
+ avcodec/faxcompr: Check for end of input in cmode == 1 in decode_group3_2d_line()
+ avcodec/vc1dec: Disable error concealment for *IMAGE
+ avcodec/sbrdsp_fixed: Fix negation overflow in sbr_neg_odd_64_c()
+ avformat/wtvdec: Check for EOF before seeking back in parse_media_type()
+ avformat/mpc8: Check first keyframe position for overflow
+ avformat/wavdec: Use 64bit in new_pos computation
+ avformat/sbgdec: Check for overflow in timestamp preparation
+ avformat/dsicin: Check packet size for overflow
+ avformat/dsfdec: Change order of operations in bitrate computation
+ avformat/bfi: check nframes
+ avformat/avidec: fix position overflow in avi_load_index()
+ avformat/asfdec_f: Check sizeX against padding
+ avformat/aiffdec: Check for size overflow in header parsing
+ avcodec/aaccoder: Add minimal bias in search_for_ms()
+ avformat/mov: Avoid undefined overflow in time_offset calculation
+ avfilter/af_drmeter: Check that there is data
+ avfilter/vf_fftdnoiz: Use lrintf() in export_row8()
+ avfilter/vf_mestimate: Check b_count
+ avformat/mov: do not ignore errors in mov_metadata_hmmt()
+ avformat/mxfdec: Check size for shrinking
+ avcodec/dnxhddec: check and propagate function return value
+ swscale/slice: Fix wrong return on error
+ avcodec/aacdec_template: Avoid some invalid values to be set by decode_audio_specific_config_gb()
+ swscale/slice: Check slice for allocation failure
+ avformat/matroskadec: Fix handling of huge default durations
+ avcodec/lpc: check for zero err in normalization in compute_lpc_coefs()
+ avformat/ftp: Check for av_strtok() failure
+ tools/cws2fws: Check read() for failure
+ avcodec/cpia: Fix missing src_size update
+ avcodec/clearvideo: Check tile_size to be not too large
+ avcodec/utils: Use 64bit for intermediate in AV_CODEC_ID_ADPCM_THP* duration calculation
+ avformat/rmdec: Check old_format len for overflow
+ avformat/realtextdec: Check the pts difference before using it for the duration computation
+ avformat/qcp: Avoid negative nb_rates
+ avformat/nutdec: Check tmp_size
+ avformat/msf: Check that channels doesnt overflow during extradata construction
+ avformat/mpc8: Check for position overflow in mpc8_handle_chunk()
+ avformat/iff: Use 64bit in duration computation
+ avformat/dxa: Check fps to be within the supported range more precissely
+ avcodec/iff: Only write palette to plane 1 if its PAL8
+ avformat/tta: Check for EOF in index reading loop
+ Update missed irc links
+ avformat/rpl: The associative law doesnt hold for signed integers in C
+ avcodec/faxcompr: Check available bits in decode_uncompressed()
+ avcodec/faxcompr: Check if bits are available before reading in cmode == 9 || cmode == 10
+ avformat/utils: check dts/duration to be representable before using them
+ avcodec/utils: do "calc from frame_bytes, channels, and block_align" in 64bit
+ avcodec/ttadata: Add sentinel at the end of ff_tta_shift_1
+ avformat/mov: Check for duplicate mdcv
+ avfilter/vf_dctdnoiz: Check threads
+ avfilter/vf_ciescope: Fix undefined behavior in rgb_to_xy() with black
+ avformat/rpl: Check for EOF and zero framesize
+ avcodec/vc2enc: Check for non negative slice bounds
+ avformat/rpl: Use 64bit in bitrate computation and check it
+ avcodec/svq1enc: Do not print debug RD value before it has been computed
+ avcodec/aacpsy: Check bandwidth
+ avcodec/aacenc: Do not divide by lambda_count if it is 0
+ avcodec/aacenc: Use FLT_EPSILON for lambda minimum
+ avformat/cinedec: Fix index_entries size check
+ avfilter/vf_yadif: Fix handing of tiny images
+ avfilter/vf_vmafmotion: Check dimensions
+ avformat/movenc: Check pal_size before use
+ avcodec/lpc: Avoid floating point division by 0
+ avcodec/aacpsy: Avoid floating point division by 0 of norm_fac
+ avcodec/aacenc: Avoid 0 lambda
+ avcodec/exr: x/ymax cannot be INT_MAX
+ avformat/avio: Check av_opt_copy() for failure
+ avcodec/clearvideo: Check for 0 tile_shift
+ avcodec/vc1: Check remaining bits in ff_vc1_parse_frame_header()
+ avformat/mov: Ignore duplicate CoLL
+ avformat/mov: Limit nb_chapter_tracks to input size
+ avformat/utils: Use 64bit earlier in r_frame_rate check
+ avcodec/alsdec: Fix decoding error with mono audio files
+ avformat/mvdec: Check sample rate in parse_audio_var()
+ avcodec/faxcompr: Check for end of bitstream in decode_group3_1d_line() and decode_group3_2d_line()
+ avcodec/utils: treat PAL8 for jpegs similar to other colorspaces
+ avcodec/jpeglsdec: Set alpha plane in PAL8 so image is not 100% transparent
+ avformat/asfdec_o: Use ff_get_extradata()
+ avformat/id3v2: Check end for overflow in id3v2_parse()
+ avformat/wtvdec: Improve size overflow checks in parse_chunks()
+ avcodec/faxcompr: Check remaining bits on error in decode_group3_1d_line()
+ avcodec/utils: Check ima wav duration for overflow
+ avcodec/rv10: Execute whole size check earlier for rv20
+ avformat/cafdec: Check channels
+ avcodec/dpx: Check bits_per_color earlier
+ avformat/mvi: Check audio_data_size to be non negative
+ avcodec/pnm_parser: Check image size addition for overflow
+ avcodec/h265_metadata_bsf: Check nb_units before accessing the first in h265_metadata_update_fragment()
+ avformat/rmdec: use larger intermediate type for audio_framesize * sub_packet_h check
+ avcodec/h264_slice: Check input SPS in ff_h264_update_thread_context()
+ avformat/mpc8: check for size overflow in mpc8_get_chunk_header()
+ avformat/mov: Do not zero memory that is written too or unused
+ avcodec/mpegvideo: Update chroma_?_shift in ff_mpv_common_frame_size_change()
+ avformat/mov: Ignore multiple STSC / STCO
+ avformat/utils: Extend overflow check in dts wrap in compute_pkt_fields()
+ avfilter/vf_scale: Fix adding 0 to NULL (which is UB) in scale_slice()
+ avutil/common: Add FF_PTR_ADD()
+ avformat/wtvdec: Check size in SBE2_STREAM_DESC_EVENT / stream2_guid
+ tools/target_dec_fuzzer: Adjust threshold for H264
+ avformat/cafdec: Do not build an index if all packets are the same
+ avformat/vividas: Use equals check with n in read_sb_block()
+ avcodec/sonic: Use unsigned temporary in predictor_calc_error()
+ avformat/jacosubdec: Use 64bit intermediate for start/end timestamp shift
+ avformat/flvdec: Check array entry number
+ avcodec/h264_slice: Check sps in h264_slice_header_init()
+ avformat/movenc: Avoid loosing cluster array on failure
+ avformat/avidec: Check for dv streams before using priv_data in parse ##dc/##wb
+ avformat/mov: Check sample size for overflow in mov_parse_stsd_audio()
+ avformat/sbgdec: Check for overflow in last loop in expand_timestamps()
+ avcodec/ffwavesynth: Avoid signed integer overflow in phi_at()
+ avcodec/mpeg4videoenc: Check extradata malloc()
+ avcodec/speedhq: Width < 8 is not supported
+ avformat/matroskadec: Check for EOF in resync loop
+ avcodec/utils: Use more bits for intermediate for AV_CODEC_ID_ADPCM_MS
+ avcodec/jpegls: Check A[Q] for overflow in ff_jpegls_update_state_regular()
+ avformat/voc_packet: prevent remaining size from becoming negative in ff_voc_get_packet()
+ avutil/timecode: Avoid fps overflow
+ avformat/mvi: Check audio size for more overflows
+ avcodec/flacdec: Avoid undefined shift in error case
+ avcodec/ffv1dec: Check if trailer is available
+ avcodec/4xm: Check pre_gb in decode_i_block()
+ avformat/mvdec: Allocate extradata only once
+ avcodec/dcadsp: Fix integer overflow in dmix_add_c()
+ avformat/flvdec: Check double before cast in parse_keyframes_index()
+ avformat/paf: Check for EOF before allocation in read_header()
+ avcodec/aacdec_template: Avoid undefined negation in imdct_and_windowing_eld()
+ avformat/lxfdec: Fix multiple integer overflows related to track_size
+ avcodec/exr: skip bottom clearing loop when its outside the image
+ avutil/parseutils: Check sign in av_parse_time()
+ avformat/aiffdec: Check that SSND is at least 8 bytes
+ avformat/dcstr: Check sample rate
+ avcodec/alsdec: Check bitstream input in read_block()
+ avformat/mov: Extend data_size check in mov_read_udta_string()
+ avformat/aadec: Check for EOF while reading chapters
+ avformat/voc_packet: Add a basic check on max_size
+ avformat/microdvddec: use 64bit for durations
+ avcodec/h264_slice: clear old slice POC values on parsing failure
+ avfilter/overlay_cuda: check av_buffer_ref result
+ avfilter/overlay_cuda: hold explicit reference to hw_device_ctx
+ avformat/url: add ff_make_absolulte_url2 to be able to test windows path cases
+ avformat/url: fix ff_make_absolute_url with Windows file paths
+ Revert "avcodec: Add FF_CODEC_CAP_INIT_CLEANUP"
+ fftools/ffplay: do not write out of rdft visualization texture
+ avfilter/overlay_cuda: fix framesync with embedded PGS subtitle
+ avcodec/mxpegdec: Fix memleaks upon init failure
+ avcodec/mjpegdec: Fix memleak upon init failure
+ avfilter/af_headphone: Fix stack buffer overflow
+ avfilter/af_headphone: Don't overrun array
+ avfilter/af_headphone: Fix segfault when using very short streams
+ avfilter/af_headphone: Check for the existence of samples
+ avfilter/af_headphone: Remove always true check
+ avfilter/af_headphone: Don't use uninitialized buffer in log message
+ avformat/segment: Fix error messages
+ avformat/segment: Free SegmentListEntries in deinit, not write_trailer
+ avformat/segment: Fix leak and invalid free of AVIOContext
+ avformat/segment: Fix leak of string on error
+ avformat/segment: Fix segfault when error happens and segment list is output
+ avformat/segment: Fix segfault on allocation error, avoid allocation
+ avformat/segment: Fix leak of duration/framenumber lists upon error
+ avformat/segment: Don't overwrite AVCodecParameters after init
+ avformat/dashdec: Reset pointer to NULL after freeing it
+ libavformat/dashdec: Fix issue with dash on Windows
+ avformat/dashdec: Fix memleak on allocation error, avoid allocation
+ avformat/dashdec: Fix memleaks on error to add representation to dynarray
+ avformat/dashdec: Fix leak of representation languages
+ avformat/dashdec: Fix leak of string on error when parsing representation
+ avformat/dashdec: Fix leak of representation on error
+ avformat/dashdec: Remove unused index of representation
+ avformat/dashdec: Fix memleaks upon read_header failure
+ avformat/dashdec: Check allocation of AVProgram
+ avformat/dashdec, hls: Update correct pointer to AVDictionary
+ avformat/dashdec: Fix leak of AVDictionary on error
+ avformat/dashdec: Free subtitle representations on exit
+ avformat/dashdec: Free strings as soon as they aren't needed anymore
+ avformat/dashdec: Don't overwrite and leak old initialization fragments
+ avformat/dashdec: Don't leave representation in inconsistent state on error
+ avformat/dashdec: Remove dead code
+ avformat/spdifenc: Fix leak upon error
+ avformat/wavenc: Fix leak and segfault on reallocation error
+ avformat/mpegenc: Avoid adding invalid packet to queue
+ avformat/mpegenc: Fix leak in case trailer is never written
+ avformat/mpegenc: Ensure packet queue stays valid
+ avformat/mxfenc: Never set codec_ul UID to NULL
+ avcodec/frame_thread_encoder: Fix segfault on allocation error
+ avformat/utils: Add av_assert1 to preclude NULL + len, len != 0
+ avformat/utils: Fix undefined NULL + 0
+ avcodec/g722enc: Validate parameters before using them
+ avcodec/g722enc: Cleanup generically on init failure
+ avcodec/opusdec: Return error upon error
+ avcodec/wavpack: Fix leak on init failure
+ avcodec/pthread_slice: Don't use static variable, fix race
+ avcodec/a64multienc: Fix memleak upon init failure
+ avformat/flacenc: Fix memleak when writing attached pictures fails
+ avcodec/mpeg12enc: Always initialize MPEG-2 intra VLC table lengths
+ avcodec/fft_template, fft_init_table: Make ff_fft_init() thread-safe
+ avformat/asfdec_o: Don't segfault with lots of attached pics
+ avcodec/mss4: Fix memleaks upon allocation error
+ avformat/apngdec: Fix size/overflow checks
+ avformat/apngdec: Return error for incomplete header
+ avformat/jacosubdec: Fix unintended fallthrough
+ avcodec/bitstream: Consistently treat symbol as VLC_TYPE
+ avcodec/vp3: Check allocations of VLCs
+ avcodec/vp3: Fix memleak upon init failure
+ avcodec/movtextenc: Fix undefined left shifts outside the range of int
+ avcodec/movtextenc: Fix memleak on (re)allocation error
+ avcodec/movtextenc: Don't presume every style to have a font
+ avcodec/movtextenc: Reset array counter after freeing array
+ avcodec/movtextenc: Fix potential use of uninitialized value
+ avcodec/movtextenc: fix writing to bytestream on BE arches
+ avcodec/movtextdec: Fix leaks on (re)allocation failure
+ avcodec/movtextdec: Simplify checking for invalid extradata
+ avcodec/movtextdec: Fix leaks of strings upon reallocation failure
+ avcodec/movtextdec: Reset counter of fonts when freeing them
+ avcodec/sheervideo: Don't leave context in inconsistent state upon error
+ avfilter/vf_minterpolate: Fix left shift of negative value
+ avfilter/vf_minterpolate: Reject too small dimensions
+ avformat/movenc: Don't forget to free fragment buffers
+ avformat/movenc: Free old vos_data before overwriting it
+ avformat/movenc: Fix segfault when remuxing rtp hint stream
+ avcodec/dvenc: Fix undefined left shift of negative numbers
+ avformat/movenc: Fix stack overflow when remuxing timecode tracks
+ avcodec/utils: Also free encoder extradata on avcodec_open2() error
+ avcodec/utils: Don't forget cleaning up when allocating priv_data fails
+ avcodec/utils: Improve check for freeing codec private options
+ swresample/audioconvert: Fix left shift of negative value
+ avcodec/utils: Only call codec->close if init has been called
+ avformat/movenc: Fix segfault upon allocation error
+ avformat/swfdec: Reorder allocations/initializations
+ avformat/swfdec: Fix memleaks on error
+ avformat/dashdec: Avoid double free on error
+ avformat/tedcaptionsdec: Fix leak of AVBPrint upon error
+ avformat/swfenc: Fix memleak upon write_header error
+ avformat/rmdec: Fix potential crash on allocation failure
+ avformat/rmdec: Actually return value < 0 on read_header failure
+ avformat/avidec: Fix memleak when error happens after creating DV stream
+ avcodec/mpegaudiodec_template: Check return value of subdecoder
+ avcodec/j2kenc: Fix leaks on init failure
+ avcodec/flashsv2enc: Check allocations for success before usage
+ avcodec/flashsvenc: Avoid allocation of buffer, fix memleak
+ avcodec/ac3enc_template: Don't free uninitialized pointers on error
+ avcodec/svq3: Fix segfault on allocation error, avoid allocations
+ avcodec/sonic: Fix leaks upon allocation errors
+ avcodec/qtrleenc: Fix memleak upon allocation failure
+ avcodec/ffv1enc: Fix memleaks on init failure
+ avcodec/ffv1: Fix segfaults on allocation error
+ avcodec/mlpenc: Fix memleak upon init failure
+ avcodec/indeo5: Fix memleaks upon allocation error
+ avcodec/ivi: Fix segfault on allocation error
+ avcodec/magicyuvenc: Fix memleak upon init failure
+ avcodec/gif: Fix leaks upon allocation error
+ avcodec/avrndec: Fix memleak on error
+ avcodec/avrndec: Check allocation for success
+ avcodec/atrac1: Check allocation of AVFloatDSPContext
+ avfilter/vf_subtitles: Fix leaks on failure
+ avfilter/vf_uspp: Fix potential leak of dict on error
+ avfilter/lavfutils: Fix memleak when avformat_find_stream_info() fails
+ avfilter/lavfutils: Don't use uninitialized pointers for freeing
+ avcodec/binkaudio: Don't use static storage for context-dependent data
+ avcodec/bink: Fix memleak upon init failure
+ avcodec/wnv1: Use LE bitstream reader, avoid copying packet, fix memleak
+ avcodec/truemotion2: Avoid duplicating array, fix memleak
+ avfilter/af_headphone: Fix leak of channel layouts list on error
+ avfilter/af_headphone: Fix segfault upon allocation failure
+ avfilter/vf_signature: Fix leak of string upon error
+ avfilter/vf_signature: Fix leak of inpads' names
+ avfilter/af_aiir: Fix segfault and leak upon allocation failure
+ avfilter/vf_premultiply: Fix leak of names of inpads
+ avfilter/af_anequalizer: Fix memleak when inserting pad fails
+ avfilter/af_amerge: Fix segfault upon allocation failure
+ avfilter/af_aformat: Add uninit function
+ avfilter/vf_hwdownload: Fix leak of formats list upon error
+ avfilter/formats: Fix double frees and memleaks on error
+ avfilter/af_channelmap: Fix double-free of AVFilterChannelLayouts on error
+ avfilter/vf_alphamerge: Fix double-free of AVFilterFormats on error
+ avfilter/vf_overlay: Fix double-free of AVFilterFormats on error
+ avfilter/vf_remap: Fix double-free of AVFilterFormats on error
+ avfilter/vf_showpalette: Fix double-free of AVFilterFormats on error
+ avfilter/af_amix: Fix double-free of AVFilterChannelLayouts on error
+ avfilter/af_amix: Don't needlessly reallocate table
+ avfilter/vf_vpp_qsv: Fix leak of AVFilterFormats on error
+ avfilter/vf_paletteuse: Fix leaks of AVFilterFormats on error
+ avfilter/graphparser: Fix memleak when linking filters fails
+ fftools/ffmpeg: Fix leak of AVFilterInOut in case of error
+ avfilter/graphparser: Check allocations for success
+ avfilter/graphparser: Don't set pointer to one beyond '\0' of string
+ avfilter/graphparser: Fix leaks when parsing inputs fails
+ avfilter/af_afir: Fix leak of AVFilterChannelLayout in case of error
+ avformat/mlvdec: Only store dimensions after having validated them
+ avformat/mlvdec: Don't leak open AVIOContexts on error
+ avformat/mlvdec: Check for existence of AVIOContext before using it
+ avformat/hls: Fix memleak when url is empty
+ avfilter/formats: Fix heap-buffer overflow when merging channel layouts
+ Revert "lavfi/avfiltergraph: add check before free the format"
+ avfilter/formats: Leave lists' ownership unchanged upon merge failure
+ avformat/mpegts: Don't leave context in inconsistent state upon error
+ avformat/sierravmd: Don't return packets for non-existing stream
+ avformat/vividas: Check allocation for success
+ avformat/vividas: Check return value before storing it in smaller type
+ avformat/webmdashenc: Avoid allocations, fix memleak
+ avformat/webmdashenc: Fix segfault when no filename is given when live
+ avformat/matroskadec: Avoid undefined pointer arithmetic
+ avformat/mxfdec: Fix memleak upon repeating tags
+ avformat/mxfdec: Fix memleak when parsing tag fails
+ avformat/mxfdec: Fix memleak when adding element to array fails
+
+
 version 4.3.2:
 avcodec/hapdec: Change compressed_offset to unsigned 32bit
 avformat/rmdec: Check codec_length without overflow
@@ -15,3 +15,11 @@ NOTICE
 ------

 - Non system dependencies (e.g. libx264, libvpx) are disabled by default.
+
+NOTICE for Package Maintainers
+------------------------------
+
+ - It is recommended to build FFmpeg twice, first with minimal external dependencies so
+   that 3rd party packages, which depend on FFmpegs libavutil/libavfilter/libavcodec/libavformat
+   can then be built. And last build FFmpeg with full dependancies (which may in turn depend on
+   some of these 3rd party packages). This avoids circular dependencies during build.
@@ -577,10 +577,12 @@ wm4
 Releases
 ========

+7.0                                     Michael Niedermayer
+6.1                                     Michael Niedermayer
+5.1                                     Michael Niedermayer
+4.4                                     Michael Niedermayer
+3.4                                     Michael Niedermayer
 2.8                                     Michael Niedermayer
-2.7                                     Michael Niedermayer
-2.6                                     Michael Niedermayer
-2.5                                     Michael Niedermayer

 If you want to maintain an older release, please contact us

@@ -610,6 +612,7 @@ Loren Merritt                 ABD9 08F4 C920 3F65 D8BE 35D7 1540 DAA7 060F 56DE
 Lou Logan (llogan)            7D68 DC73 CBEF EABB 671A B6CF 621C 2E28 82F8 DC3A
 Lynne                         FE50 139C 6805 72CA FD52 1F8D A2FE A5F0 3F03 4464
 Michael Niedermayer           9FF2 128B 147E F673 0BAD F133 611E C787 040B 0FAB
+                              DD1E C9E8 DE08 5C62 9B3E 1846 B18E 8928 B394 8D64
 Nicolas George                24CE 01CE 9ACC 5CEB 74D8 8D9D B063 D997 36E5 4C93
 Nikolay Aleksandrov           8978 1D8C FB71 588E 4B27 EAA8 C4F0 B5FC E011 13B1
 Panagiotis Issaris            6571 13A3 33D9 3726 F728 AA98 F643 B12E ECF3 E029
@@ -1 +1 @@
-4.3.2
+4.3.9
@@ -11,5 +11,5 @@

   We hope you will like this release as much as we enjoyed working on it, and
   as usual, if you have any questions about it, or any FFmpeg related topic,
-   feel free to join us on the #ffmpeg IRC channel (on irc.freenode.net) or ask
+   feel free to join us on the #ffmpeg IRC channel (on irc.libera.chat) or ask
   on the mailing-lists.
@@ -532,7 +532,7 @@ die(){

 If you think configure made a mistake, make sure you are using the latest
 version from Git.  If the latest version fails, report the problem to the
-ffmpeg-user@ffmpeg.org mailing list or IRC #ffmpeg on irc.freenode.net.
+ffmpeg-user@ffmpeg.org mailing list or IRC #ffmpeg on irc.libera.chat.
 EOF
    if disabled logging; then
        cat <<EOF
@@ -2330,6 +2330,7 @@ HAVE_LIST="
    opencl_vaapi_intel_media
    perl
    pod2man
+    posix_ioctl
    texi2html
 "

@@ -3236,7 +3237,7 @@ librav1e_encoder_deps="librav1e"
 librav1e_encoder_select="extract_extradata_bsf"
 librsvg_decoder_deps="librsvg"
 libshine_encoder_deps="libshine"
-libshine_encoder_select="audio_frame_queue"
+libshine_encoder_select="audio_frame_queue mpegaudioheader"
 libspeex_decoder_deps="libspeex"
 libspeex_encoder_deps="libspeex"
 libspeex_encoder_select="audio_frame_queue"
@@ -5330,6 +5331,7 @@ case $target_os in
        ;;
    netbsd)
        disable symver
+        enable section_data_rel_ro
        oss_indev_extralibs="-lossaudio"
        oss_outdev_extralibs="-lossaudio"
        enabled gcc || check_ldflags -Wl,-zmuldefs
@@ -5348,6 +5350,7 @@ case $target_os in
        disable symver
        ;;
    freebsd)
+        enable section_data_rel_ro
        ;;
    bsd/os)
        add_extralibs -lpoll -lgnugetopt
@@ -6493,7 +6496,7 @@ fi

 if enabled sdl2; then
    SDL2_CONFIG="${cross_prefix}sdl2-config"
-    test_pkg_config sdl2 "sdl2 >= 2.0.1 sdl2 < 2.1.0" SDL_events.h SDL_PollEvent
+    test_pkg_config sdl2 "sdl2 >= 2.0.1 sdl2 < 3.0.0" SDL_events.h SDL_PollEvent
    if disabled sdl2 && "${SDL2_CONFIG}" --version > /dev/null 2>&1; then
        sdl2_cflags=$("${SDL2_CONFIG}" --cflags)
        sdl2_extralibs=$("${SDL2_CONFIG}" --libs)
@@ -6539,11 +6542,13 @@ perl -v            > /dev/null 2>&1 && enable perl      || disable perl
 pod2man --help     > /dev/null 2>&1 && enable pod2man   || disable pod2man
 rsync --help 2> /dev/null | grep -q 'contimeout' && enable rsync_contimeout || disable rsync_contimeout

+check_headers linux/fb.h
+check_headers linux/videodev2.h
+test_code cc linux/videodev2.h "struct v4l2_frmsizeenum vfse; vfse.discrete.width = 0;" && enable_sanitized struct_v4l2_frmivalenum_discrete
+test_code cc sys/ioctl.h "int ioctl(int, int, ...)" && enable posix_ioctl
+
 # check V4L2 codecs available in the API
 if enabled v4l2_m2m; then
-    check_headers linux/fb.h
-    check_headers linux/videodev2.h
-    test_code cc linux/videodev2.h "struct v4l2_frmsizeenum vfse; vfse.discrete.width = 0;" && enable_sanitized struct_v4l2_frmivalenum_discrete
    check_cc v4l2_m2m linux/videodev2.h "int i = V4L2_CAP_VIDEO_M2M_MPLANE | V4L2_CAP_VIDEO_M2M | V4L2_BUF_FLAG_LAST;"
    check_cc vc1_v4l2_m2m linux/videodev2.h "int i = V4L2_PIX_FMT_VC1_ANNEX_G;"
    check_cc mpeg1_v4l2_m2m linux/videodev2.h "int i = V4L2_PIX_FMT_MPEG1;"
@@ -6588,7 +6593,7 @@ enabled alsa && { check_pkg_config alsa alsa "alsa/asoundlib.h" snd_pcm_htimesta
 enabled libjack &&
    require_pkg_config libjack jack jack/jack.h jack_port_get_latency_range

-enabled sndio && check_lib sndio sndio.h sio_open -lsndio
+enabled sndio && check_pkg_config sndio sndio sndio.h sio_open

 if enabled libcdio; then
    check_pkg_config libcdio libcdio_paranoia "cdio/cdda.h cdio/paranoia.h" cdio_cddap_open ||
@@ -6685,7 +6690,7 @@ enabled vulkan &&

 if enabled x86; then
    case $target_os in
-        mingw32*|mingw64*|win32|win64|linux|cygwin*)
+        freebsd|mingw32*|mingw64*|win32|win64|linux|cygwin*)
            ;;
        *)
            disable ffnvcodec cuvid nvdec nvenc
@@ -7513,7 +7518,7 @@ cat > $TMPH <<EOF
 #define FFMPEG_CONFIG_H
 #define FFMPEG_CONFIGURATION "$(c_escape $FFMPEG_CONFIGURATION)"
 #define FFMPEG_LICENSE "$(c_escape $license)"
-#define CONFIG_THIS_YEAR 2021
+#define CONFIG_THIS_YEAR 2025
 #define FFMPEG_DATADIR "$(eval c_escape $datadir)"
 #define AVCONV_DATADIR "$(eval c_escape $datadir)"
 #define CC_IDENT "$(c_escape ${cc_ident:-Unknown compiler})"
@@ -38,7 +38,7 @@ PROJECT_NAME           = FFmpeg
 # could be handy for archiving the generated documentation or if some version
 # control system is used.

-PROJECT_NUMBER         = 4.3.2
+PROJECT_NUMBER         = 4.3.9

 # Using the PROJECT_BRIEF tag one can provide an optional one line description
 # for a project that appears at the top of each page and should give viewer a
@@ -3,9 +3,9 @@
 The FFmpeg developers.

 For details about the authorship, see the Git history of the project
-(git://source.ffmpeg.org/ffmpeg), e.g. by typing the command
+(https://git.ffmpeg.org/ffmpeg), e.g. by typing the command
@command{git log} in the FFmpeg source directory, or browsing the
-online repository at @url{http://source.ffmpeg.org}.
+online repository at @url{https://git.ffmpeg.org/ffmpeg}.

 Maintainers for the specific components are listed in the file
@file{MAINTAINERS} in the source code tree.
@@ -63,4 +63,3 @@ make -j<num>
 make -k
    Continue build in case of errors, this is useful for the regression tests
    sometimes but note that it will still not run all reg tests.
-
@@ -327,6 +327,13 @@ segment index to start live streams at (negative values are from the end).
@item allowed_extensions
 ',' separated list of file extensions that hls is allowed to access.

+@item extension_picky
+This blocks disallowed extensions from probing
+It also requires all available segments to have matching extensions to the format
+except mpegts, which is always allowed.
+It is recommended to set the whitelists correctly instead of depending on extensions
+Enabled by default.
+
@item max_reload
 Maximum number of times a insufficient list is attempted to be reloaded.
 Default value is 1000.
@@ -762,6 +762,25 @@ In case you need finer control over how valgrind is invoked, use the
@code{--target-exec='valgrind <your_custom_valgrind_options>} option in
 your configure line instead.

+@anchor{Maintenance}
+@chapter Maintenance process
+
+@anchor{MAINTAINERS}
+@section MAINTAINERS
+
+The developers maintaining each part of the codebase are listed in @file{MAINTAINERS}.
+Being listed in @file{MAINTAINERS}, gives one the right to have git write access to
+the specific repository.
+
+@anchor{Becoming a maintainer}
+@section Becoming a maintainer
+
+People add themselves to @file{MAINTAINERS} by sending a patch like any other code
+change. These get reviewed by the community like any other patch. It is expected
+that, if someone has an objection to a new maintainer, she is willing to object
+in public with her full name and is willing to take over maintainership for the area.
+
+
@anchor{Release process}
@chapter Release process

@@ -137,11 +137,9 @@ static int decode_packet(AVCodecContext *dec, const AVPacket *pkt)
            ret = output_audio_frame(frame);

        av_frame_unref(frame);
-        if (ret < 0)
-            return ret;
    }

-    return 0;
+    return ret;
 }

 static int open_codec_context(int *stream_idx,
@@ -221,10 +221,8 @@ static int dec_enc(AVPacket *pkt, AVCodec *enc_codec)

 fail:
        av_frame_free(&frame);
-        if (ret < 0)
-            return ret;
    }
-    return 0;
+    return ret;
 }

 int main(int argc, char **argv)
@@ -53,7 +53,7 @@ Most distribution and operating system provide a package for it.
@section Cloning the source tree

@example
-git clone git://source.ffmpeg.org/ffmpeg <target>
+git clone https://git.ffmpeg.org/ffmpeg.git <target>
@end example

 This will put the FFmpeg sources into the directory @var{<target>}.
@@ -187,11 +187,18 @@ to make sure you don't have untracked files or deletions.
 git add [-i|-p|-A] <filenames/dirnames>
@end example

-Make sure you have told Git your name and email address
+Make sure you have told Git your name, email address and GPG key

@example
 git config --global user.name "My Name"
 git config --global user.email my@@email.invalid
+git config --global user.signingkey ABCDEF0123245
+@end example
+
+Enable signing all commits or use -S
+
+@example
+git config --global commit.gpgsign true
@end example

 Use @option{--global} to set the global configuration for all your Git checkouts.
@@ -393,6 +400,19 @@ git checkout -b svn_23456 $SHA1
 where @var{$SHA1} is the commit hash from the @command{git log} output.


+@chapter gpg key generation
+
+If you have no gpg key yet, we recommend that you create a ed25519 based key as it
+is small, fast and secure. Especially it results in small signatures in git.
+
+@example
+gpg --default-new-key-algo "ed25519/cert,sign+cv25519/encr" --quick-generate-key "human@@server.com"
+@end example
+
+When generating a key, make sure the email specified matches the email used in git as some sites like
+github consider mismatches a reason to declare such commits unverified. After generating a key you
+can add it to the MAINTAINER file and upload it to a keyserver.
+
@chapter Pre-push checklist

 Once you have a set of commits that you feel are ready for pushing,
@@ -157,4 +157,3 @@ PFD[32]   would for example be signed 32 bit little-endian IEEE float
@item XVID @tab non-compliant MPEG-4 generated by old Xvid
@item XVIX @tab non-compliant MPEG-4 generated by old Xvid with interlacing bug
@end multitable
-
@@ -20,8 +20,45 @@
 # License along with FFmpeg; if not, write to the Free Software
 # Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA

+# Texinfo 7.0 changed the syntax of various functions.
+# Provide a shim for older versions.
+sub ff_set_from_init_file($$) {
+    my $key = shift;
+    my $value = shift;
+    if (exists &{'texinfo_set_from_init_file'}) {
+        texinfo_set_from_init_file($key, $value);
+    } else {
+        set_from_init_file($key, $value);
+    }
+}
+
+sub ff_get_conf($) {
+    my $key = shift;
+    if (exists &{'texinfo_get_conf'}) {
+        texinfo_get_conf($key);
+    } else {
+        get_conf($key);
+    }
+}
+
+sub get_formatting_function($$) {
+    my $obj = shift;
+    my $func = shift;
+
+    my $sub = $obj->can('formatting_function');
+    if ($sub) {
+        return $obj->formatting_function($func);
+    } else {
+        return $obj->{$func};
+    }
+}
+
+# determine texinfo version
+my $program_version_num = version->declare(ff_get_conf('PACKAGE_VERSION'))->numify;
+my $program_version_6_8 = $program_version_num >= 6.008000;
+
 # no navigation elements
-set_from_init_file('HEADERS', 0);
+ff_set_from_init_file('HEADERS', 0);

 sub ffmpeg_heading_command($$$$$)
 {
@@ -55,7 +92,7 @@ sub ffmpeg_heading_command($$$$$)
        $element = $command->{'parent'};
    }
    if ($element) {
-        $result .= &{$self->{'format_element_header'}}($self, $cmdname,
+        $result .= &{get_formatting_function($self, 'format_element_header')}($self, $cmdname,
                                                       $command, $element);
    }

@@ -112,7 +149,11 @@ sub ffmpeg_heading_command($$$$$)
                $cmdname
                    = $Texinfo::Common::level_to_structuring_command{$cmdname}->[$heading_level];
            }
-            $result .= &{$self->{'format_heading_text'}}(
+            # format_heading_text expects an array of headings for texinfo >= 7.0
+            if ($program_version_num >= 7.000000) {
+                $heading = [$heading];
+            }
+            $result .= &{get_formatting_function($self,'format_heading_text')}(
                        $self, $cmdname, $heading,
                        $heading_level +
                        $self->get_conf('CHAPTER_HEADER_LEVEL') - 1, $command);
@@ -127,14 +168,18 @@ foreach my $command (keys(%Texinfo::Common::sectioning_commands), 'node') {
 }

 # print the TOC where @contents is used
-set_from_init_file('INLINE_CONTENTS', 1);
+if ($program_version_6_8) {
+    ff_set_from_init_file('CONTENTS_OUTPUT_LOCATION', 'inline');
+} else {
+    ff_set_from_init_file('INLINE_CONTENTS', 1);
+}

 # make chapters <h2>
-set_from_init_file('CHAPTER_HEADER_LEVEL', 2);
+ff_set_from_init_file('CHAPTER_HEADER_LEVEL', 2);

 # Do not add <hr>
-set_from_init_file('DEFAULT_RULE', '');
-set_from_init_file('BIG_RULE', '');
+ff_set_from_init_file('DEFAULT_RULE', '');
+ff_set_from_init_file('BIG_RULE', '');

 # Customized file beginning
 sub ffmpeg_begin_file($$$)
@@ -151,7 +196,18 @@ sub ffmpeg_begin_file($$$)
    my ($title, $description, $encoding, $date, $css_lines,
        $doctype, $bodytext, $copying_comment, $after_body_open,
        $extra_head, $program_and_version, $program_homepage,
-        $program, $generator) = $self->_file_header_informations($command);
+        $program, $generator);
+    if ($program_version_num >= 7.000000) {
+        ($title, $description, $encoding, $date, $css_lines,
+         $doctype, $bodytext, $copying_comment, $after_body_open,
+         $extra_head, $program_and_version, $program_homepage,
+         $program, $generator) = $self->_file_header_information($command);
+    } else {
+        ($title, $description, $encoding, $date, $css_lines,
+         $doctype, $bodytext, $copying_comment, $after_body_open,
+         $extra_head, $program_and_version, $program_homepage,
+         $program, $generator) = $self->_file_header_informations($command);
+    }

    my $links = $self->_get_links ($filename, $element);

@@ -184,7 +240,11 @@ EOT

    return $head1 . $head_title . $head2 . $head_title . $head3;
 }
-texinfo_register_formatting_function('begin_file', \&ffmpeg_begin_file);
+if ($program_version_6_8) {
+    texinfo_register_formatting_function('format_begin_file', \&ffmpeg_begin_file);
+} else {
+    texinfo_register_formatting_function('begin_file', \&ffmpeg_begin_file);
+}

 sub ffmpeg_program_string($)
 {
@@ -201,13 +261,17 @@ sub ffmpeg_program_string($)
      $self->gdt('This document was generated automatically.'));
  }
 }
-texinfo_register_formatting_function('program_string', \&ffmpeg_program_string);
+if ($program_version_6_8) {
+    texinfo_register_formatting_function('format_program_string', \&ffmpeg_program_string);
+} else {
+    texinfo_register_formatting_function('program_string', \&ffmpeg_program_string);
+}

 # Customized file ending
 sub ffmpeg_end_file($)
 {
    my $self = shift;
-    my $program_string = &{$self->{'format_program_string'}}($self);
+    my $program_string = &{get_formatting_function($self,'format_program_string')}($self);
    my $program_text = <<EOT;
      <p style="font-size: small;">
        $program_string
@@ -220,11 +284,15 @@ EOT
 EOT
    return $program_text . $footer;
 }
-texinfo_register_formatting_function('end_file', \&ffmpeg_end_file);
+if ($program_version_6_8) {
+    texinfo_register_formatting_function('format_end_file', \&ffmpeg_end_file);
+} else {
+    texinfo_register_formatting_function('end_file', \&ffmpeg_end_file);
+}

 # Dummy title command
 # Ignore title. Title is handled through ffmpeg_begin_file().
-set_from_init_file('USE_TITLEPAGE_FOR_TITLE', 1);
+ff_set_from_init_file('USE_TITLEPAGE_FOR_TITLE', 1);
 sub ffmpeg_title($$$$)
 {
    return '';
@@ -242,8 +310,14 @@ sub ffmpeg_float($$$$$)
    my $args = shift;
    my $content = shift;

-    my ($caption, $prepended) = Texinfo::Common::float_name_caption($self,
-                                                                $command);
+    my ($caption, $prepended);
+    if ($program_version_num >= 7.000000) {
+        ($caption, $prepended) = Texinfo::Convert::Converter::float_name_caption($self,
+                                                                                 $command);
+    } else {
+        ($caption, $prepended) = Texinfo::Common::float_name_caption($self,
+                                                                     $command);
+    }
    my $caption_text = '';
    my $prepended_text;
    my $prepended_save = '';
@@ -315,8 +389,13 @@ sub ffmpeg_float($$$$$)
            $caption->{'args'}->[0], 'float caption');
    }
    if ($prepended_text.$caption_text ne '') {
-        $prepended_text = $self->_attribute_class('div','float-caption'). '>'
-                . $prepended_text;
+        if ($program_version_num >= 7.000000) {
+            $prepended_text = $self->html_attribute_class('div',['float-caption']). '>'
+                    . $prepended_text;
+        } else {
+            $prepended_text = $self->_attribute_class('div','float-caption'). '>'
+                    . $prepended_text;
+        }
        $caption_text .= '</div>';
    }
    my $html_class = '';
@@ -329,8 +408,13 @@ sub ffmpeg_float($$$$$)
        $prepended_text = '';
        $caption_text   = '';
    }
-    return $self->_attribute_class('div', $html_class). '>' . "\n" .
-        $prepended_text . $caption_text . $content . '</div>';
+    if ($program_version_num >= 7.000000) {
+        return $self->html_attribute_class('div', [$html_class]). '>' . "\n" .
+            $prepended_text . $caption_text . $content . '</div>';
+    } else {
+        return $self->_attribute_class('div', $html_class). '>' . "\n" .
+            $prepended_text . $caption_text . $content . '</div>';
+    }
 }

 texinfo_register_command_formatting('float',
@@ -44,4 +44,3 @@ a+b*c;
 here the reader knows that a,b,c are meant to be signed integers but for C
 standard compliance / to avoid undefined behavior they are stored in unsigned
 ints.
-
@@ -418,4 +418,4 @@ done:

 When all of this is done, you can submit your patch to the ffmpeg-devel
 mailing-list for review.  If you need any help, feel free to come on our IRC
-channel, #ffmpeg-devel on irc.freenode.net.
+channel, #ffmpeg-devel on irc.libera.chat.
@@ -1,3 +1,5 @@
+#!/bin/sh
+
 toupper(){
    echo "$@" | tr abcdefghijklmnopqrstuvwxyz ABCDEFGHIJKLMNOPQRSTUVWXYZ
 }
@@ -538,7 +538,7 @@ static const AVOption *opt_find(void *obj, const char *name, const char *unit,
    return o;
 }

-#define FLAGS (o->type == AV_OPT_TYPE_FLAGS && (arg[0]=='-' || arg[0]=='+')) ? AV_DICT_APPEND : 0
+#define FLAGS ((o->type == AV_OPT_TYPE_FLAGS && (arg[0]=='-' || arg[0]=='+')) ? AV_DICT_APPEND : 0)
 int opt_default(void *optctx, const char *opt, const char *arg)
 {
    const AVOption *o;
@@ -468,8 +468,9 @@ static int read_key(void)
        }
        //Read it
        if(nchars != 0) {
-            read(0, &ch, 1);
-            return ch;
+            if (read(0, &ch, 1) == 1)
+                return ch;
+            return 0;
        }else{
            return -1;
        }
@@ -528,6 +529,7 @@ static void ffmpeg_cleanup(int ret)
        for (j = 0; j < fg->nb_outputs; j++) {
            OutputFilter *ofilter = fg->outputs[j];

+            avfilter_inout_free(&ofilter->out_tmp);
            av_freep(&ofilter->name);
            av_freep(&ofilter->formats);
            av_freep(&ofilter->channel_layouts);
@@ -1151,6 +1151,8 @@ static void video_audio_display(VideoState *s)
        if (realloc_texture(&s->vis_texture, SDL_PIXELFORMAT_ARGB8888, s->width, s->height, SDL_BLENDMODE_NONE, 1) < 0)
            return;

+        if (s->xpos >= s->width)
+            s->xpos = 0;
        nb_display_channels= FFMIN(nb_display_channels, 2);
        if (rdft_bits != s->rdft_bits) {
            av_rdft_end(s->rdft);
@@ -1200,8 +1202,6 @@ static void video_audio_display(VideoState *s)
        }
        if (!s->paused)
            s->xpos++;
-        if (s->xpos >= s->width)
-            s->xpos= s->xleft;
    }
 }

@@ -131,8 +131,8 @@ static int zero12v_decode_frame(AVCodecContext *avctx, void *data,
            u = x/2 + (uint16_t *)(pic->data[1] + line * pic->linesize[1]);
            v = x/2 + (uint16_t *)(pic->data[2] + line * pic->linesize[2]);
            memcpy(y, y_temp, sizeof(*y) * (width - x));
-            memcpy(u, u_temp, sizeof(*u) * (width - x + 1) / 2);
-            memcpy(v, v_temp, sizeof(*v) * (width - x + 1) / 2);
+            memcpy(u, u_temp, sizeof(*u) * ((width - x + 1) / 2));
+            memcpy(v, v_temp, sizeof(*v) * ((width - x + 1) / 2));
        }

        line_end += stride;
@@ -498,8 +498,8 @@ static int decode_i_block(FourXContext *f, int16_t *block)
 {
    int code, i, j, level, val;

-    if (get_bits_left(&f->gb) < 2){
-        av_log(f->avctx, AV_LOG_ERROR, "%d bits left before decode_i_block()\n", get_bits_left(&f->gb));
+    if (get_bits_left(&f->pre_gb) < 2) {
+        av_log(f->avctx, AV_LOG_ERROR, "%d bits left before decode_i_block()\n", get_bits_left(&f->pre_gb));
        return AVERROR_INVALIDDATA;
    }

@@ -885,6 +885,8 @@ static int decode_frame(AVCodecContext *avctx, void *data,
        }

        if (i >= CFRAME_BUFFER_COUNT) {
+            if (free_index < 0)
+                return AVERROR_INVALIDDATA;
            i             = free_index;
            f->cfrm[i].id = id;
        }
@@ -70,6 +70,9 @@ static int decode_frame(AVCodecContext *avctx, void *data,
    unsigned char *planemap = c->planemap;
    int ret;

+    if (buf_size < planes * height *2)
+        return AVERROR_INVALIDDATA;
+
    if ((ret = ff_get_buffer(avctx, frame, 0)) < 0)
        return ret;

@@ -1174,7 +1174,7 @@ SKIPHEADERS-$(CONFIG_QSV)              += qsv.h qsv_internal.h
 SKIPHEADERS-$(CONFIG_QSVDEC)           += qsvdec.h
 SKIPHEADERS-$(CONFIG_QSVENC)           += qsvenc.h
 SKIPHEADERS-$(CONFIG_XVMC)             += xvmc.h
-SKIPHEADERS-$(CONFIG_VAAPI)            += vaapi_decode.h vaapi_encode.h
+SKIPHEADERS-$(CONFIG_VAAPI)            += vaapi_decode.h vaapi_hevc.h vaapi_encode.h
 SKIPHEADERS-$(CONFIG_VDPAU)            += vdpau.h vdpau_internal.h
 SKIPHEADERS-$(CONFIG_VIDEOTOOLBOX)     += videotoolbox.h vt_internal.h
 SKIPHEADERS-$(CONFIG_V4L2_M2M)         += v4l2_buffers.h v4l2_context.h v4l2_m2m.h
@@ -407,6 +407,7 @@ AVCodec ff_a64multi_encoder = {
    .close          = a64multi_close_encoder,
    .pix_fmts       = (const enum AVPixelFormat[]) {AV_PIX_FMT_GRAY8, AV_PIX_FMT_NONE},
    .capabilities   = AV_CODEC_CAP_DELAY,
+    .caps_internal  = FF_CODEC_CAP_INIT_CLEANUP,
 };
 #endif
 #if CONFIG_A64MULTI5_ENCODER
@@ -421,5 +422,6 @@ AVCodec ff_a64multi5_encoder = {
    .close          = a64multi_close_encoder,
    .pix_fmts       = (const enum AVPixelFormat[]) {AV_PIX_FMT_GRAY8, AV_PIX_FMT_NONE},
    .capabilities   = AV_CODEC_CAP_DELAY,
+    .caps_internal  = FF_CODEC_CAP_INIT_CLEANUP,
 };
 #endif
@@ -843,25 +843,25 @@ static void search_for_ms(AACEncContext *s, ChannelElement *cpe)
                                                    sce0->ics.swb_sizes[g],
                                                    sce0->sf_idx[w*16+g],
                                                    sce0->band_type[w*16+g],
-                                                    lambda / band0->threshold, INFINITY, &b1, NULL, 0);
+                                                    lambda / (band0->threshold + FLT_MIN), INFINITY, &b1, NULL, 0);
                        dist1 += quantize_band_cost(s, &sce1->coeffs[start + (w+w2)*128],
                                                    R34,
                                                    sce1->ics.swb_sizes[g],
                                                    sce1->sf_idx[w*16+g],
                                                    sce1->band_type[w*16+g],
-                                                    lambda / band1->threshold, INFINITY, &b2, NULL, 0);
+                                                    lambda / (band1->threshold + FLT_MIN), INFINITY, &b2, NULL, 0);
                        dist2 += quantize_band_cost(s, M,
                                                    M34,
                                                    sce0->ics.swb_sizes[g],
                                                    mididx,
                                                    midcb,
-                                                    lambda / minthr, INFINITY, &b3, NULL, 0);
+                                                    lambda / (minthr + FLT_MIN), INFINITY, &b3, NULL, 0);
                        dist2 += quantize_band_cost(s, S,
                                                    S34,
                                                    sce1->ics.swb_sizes[g],
                                                    sididx,
                                                    sidcb,
-                                                    mslambda / (minthr * bmax), INFINITY, &b4, NULL, 0);
+                                                    mslambda / (minthr * bmax + FLT_MIN), INFINITY, &b4, NULL, 0);
                        B0 += b1+b2;
                        B1 += b3+b4;
                        dist1 -= b1+b2;
@@ -974,14 +974,18 @@ static int decode_audio_specific_config_gb(AACContext *ac,
 {
    int i, ret;
    GetBitContext gbc = *gb;
+    MPEG4AudioConfig m4ac_bak = *m4ac;

-    if ((i = ff_mpeg4audio_get_config_gb(m4ac, &gbc, sync_extension, avctx)) < 0)
+    if ((i = ff_mpeg4audio_get_config_gb(m4ac, &gbc, sync_extension, avctx)) < 0) {
+        *m4ac = m4ac_bak;
        return AVERROR_INVALIDDATA;
+    }

    if (m4ac->sampling_index > 12) {
        av_log(avctx, AV_LOG_ERROR,
               "invalid sampling rate index %d\n",
               m4ac->sampling_index);
+        *m4ac = m4ac_bak;
        return AVERROR_INVALIDDATA;
    }
    if (m4ac->object_type == AOT_ER_AAC_LD &&
@@ -989,6 +993,7 @@ static int decode_audio_specific_config_gb(AACContext *ac,
        av_log(avctx, AV_LOG_ERROR,
               "invalid low delay sampling rate index %d\n",
               m4ac->sampling_index);
+        *m4ac = m4ac_bak;
        return AVERROR_INVALIDDATA;
    }

@@ -2807,7 +2812,7 @@ static void imdct_and_windowing_ld(AACContext *ac, SingleChannelElement *sce)

 static void imdct_and_windowing_eld(AACContext *ac, SingleChannelElement *sce)
 {
-    INTFLOAT *in    = sce->coeffs;
+    UINTFLOAT *in   = sce->coeffs;
    INTFLOAT *out   = sce->ret;
    INTFLOAT *saved = sce->saved;
    INTFLOAT *buf  = ac->buf_mdct;
@@ -28,6 +28,7 @@
 *              TODOs:
 * add sane pulse detection
 ***********************************/
+#include <float.h>

 #include "libavutil/libm.h"
 #include "libavutil/thread.h"
@@ -856,7 +857,7 @@ static int aac_encode_frame(AVCodecContext *avctx, AVPacket *avpkt,
                /* Not so fast though */
                ratio = sqrtf(ratio);
            }
-            s->lambda = FFMIN(s->lambda * ratio, 65536.f);
+            s->lambda = av_clipf(s->lambda * ratio, FLT_EPSILON, 65536.f);

            /* Keep iterating if we must reduce and lambda is in the sky */
            if (ratio > 0.9f && ratio < 1.1f) {
@@ -901,7 +902,7 @@ static av_cold int aac_encode_end(AVCodecContext *avctx)
 {
    AACEncContext *s = avctx->priv_data;

-    av_log(avctx, AV_LOG_INFO, "Qavg: %.3f\n", s->lambda_sum / s->lambda_count);
+    av_log(avctx, AV_LOG_INFO, "Qavg: %.3f\n", s->lambda_count ? s->lambda_sum / s->lambda_count : NAN);

    ff_mdct_end(&s->mdct1024);
    ff_mdct_end(&s->mdct128);
@@ -173,6 +173,7 @@ void ff_aac_search_for_tns(AACEncContext *s, SingleChannelElement *sce)
                      sce->ics.window_sequence[0] == LONG_START_SEQUENCE ? 0 : 2;
    const int sfb_len = sfb_end - sfb_start;
    const int coef_len = sce->ics.swb_offset[sfb_end] - sce->ics.swb_offset[sfb_start];
+    const int n_filt = is8 ? 1 : order != TNS_MAX_ORDER ? 2 : 3;

    if (coef_len <= 0 || sfb_len <= 0) {
        sce->tns.present = 0;
@@ -180,16 +181,30 @@ void ff_aac_search_for_tns(AACEncContext *s, SingleChannelElement *sce)
    }

    for (w = 0; w < sce->ics.num_windows; w++) {
-        float en[2] = {0.0f, 0.0f};
-        int oc_start = 0, os_start = 0;
+        float en[4] = {0.0f, 0.0f, 0.0f, 0.0f};
+        int oc_start = 0;
        int coef_start = sce->ics.swb_offset[sfb_start];

-        for (g = sfb_start; g < sce->ics.num_swb && g <= sfb_end; g++) {
-            FFPsyBand *band = &s->psy.ch[s->cur_channel].psy_bands[w*16+g];
-            if (g > sfb_start + (sfb_len/2))
-                en[1] += band->energy;
-            else
-                en[0] += band->energy;
+        if (n_filt == 2) {
+            for (g = sfb_start; g < sce->ics.num_swb && g <= sfb_end; g++) {
+                FFPsyBand *band = &s->psy.ch[s->cur_channel].psy_bands[w*16+g];
+                    if (g > sfb_start + (sfb_len/2))
+                        en[1] += band->energy; /* End */
+                    else
+                        en[0] += band->energy; /* Start */
+            }
+            en[2] = en[0];
+        } else {
+            for (g = sfb_start; g < sce->ics.num_swb && g <= sfb_end; g++) {
+                FFPsyBand *band = &s->psy.ch[s->cur_channel].psy_bands[w*16+g];
+                    if (g > sfb_start + (sfb_len/2) + (sfb_len/4))
+                        en[2] += band->energy; /* End */
+                    else if (g > sfb_start + (sfb_len/2) - (sfb_len/4))
+                        en[1] += band->energy; /* Middle */
+                    else
+                        en[0] += band->energy; /* Start */
+            }
+            en[3] = en[0];
        }

        /* LPC */
@@ -199,15 +214,14 @@ void ff_aac_search_for_tns(AACEncContext *s, SingleChannelElement *sce)
        if (!order || !isfinite(gain) || gain < TNS_GAIN_THRESHOLD_LOW || gain > TNS_GAIN_THRESHOLD_HIGH)
            continue;

-        tns->n_filt[w] = is8 ? 1 : order != TNS_MAX_ORDER ? 2 : 3;
+        tns->n_filt[w] = n_filt;
        for (g = 0; g < tns->n_filt[w]; g++) {
-            tns->direction[w][g] = slant != 2 ? slant : en[g] < en[!g];
-            tns->order[w][g] = g < tns->n_filt[w] ? order/tns->n_filt[w] : order - oc_start;
-            tns->length[w][g] = g < tns->n_filt[w] ? sfb_len/tns->n_filt[w] : sfb_len - os_start;
+            tns->direction[w][g] = slant != 2 ? slant : en[g] < en[g + 1];
+            tns->order[w][g] = order/tns->n_filt[w];
+            tns->length[w][g] = sfb_len/tns->n_filt[w];
            quantize_coefs(&coefs[oc_start], tns->coef_idx[w][g], tns->coef[w][g],
                            tns->order[w][g], c_bits);
            oc_start += tns->order[w][g];
-            os_start += tns->length[w][g];
        }
        count++;
    }
@@ -308,6 +308,9 @@ static av_cold int psy_3gpp_init(FFPsyContext *ctx) {
    const int bandwidth    = ctx->cutoff ? ctx->cutoff : AAC_CUTOFF(ctx->avctx);
    const float num_bark   = calc_bark((float)bandwidth);

+    if (bandwidth <= 0)
+        return AVERROR(EINVAL);
+
    ctx->model_priv_data = av_mallocz(sizeof(AacPsyContext));
    if (!ctx->model_priv_data)
        return AVERROR(ENOMEM);
@@ -794,7 +797,7 @@ static void psy_3gpp_analyze_channel(FFPsyContext *ctx, int channel,

        if (pe < 1.15f * desired_pe) {
            /* 6.6.1.3.6 "Final threshold modification by linearization" */
-            norm_fac = 1.0f / norm_fac;
+            norm_fac = norm_fac ? 1.0f / norm_fac : 0;
            for (w = 0; w < wi->num_windows*16; w += 16) {
                for (g = 0; g < num_bands; g++) {
                    AacPsyBand *band = &pch->band[w+g];
@@ -592,6 +592,7 @@ static int sbr_make_f_derived(AACContext *ac, SpectralBandReplication *sbr)

    if (sbr->n_q > 5) {
        av_log(ac->avctx, AV_LOG_ERROR, "Too many noise floor scale factors: %d\n", sbr->n_q);
+        sbr->n_q = 1;
        return -1;
    }

@@ -19,130 +19,130 @@
 #include "libavutil/aarch64/asm.S"

 function ff_ps_add_squares_neon, export=1
-1:      ld1         {v0.4S,v1.4S}, [x1], #32
-        fmul        v0.4S, v0.4S, v0.4S
-        fmul        v1.4S, v1.4S, v1.4S
-        faddp       v2.4S, v0.4S, v1.4S
-        ld1         {v3.4S}, [x0]
-        fadd        v3.4S, v3.4S, v2.4S
-        st1         {v3.4S}, [x0], #16
-        subs        w2, w2, #4
-        b.gt        1b
+1:      ld1             {v0.4s,v1.4s}, [x1], #32
+        fmul            v0.4s, v0.4s, v0.4s
+        fmul            v1.4s, v1.4s, v1.4s
+        faddp           v2.4s, v0.4s, v1.4s
+        ld1             {v3.4s}, [x0]
+        fadd            v3.4s, v3.4s, v2.4s
+        st1             {v3.4s}, [x0], #16
+        subs            w2, w2, #4
+        b.gt            1b
        ret
 endfunc

 function ff_ps_mul_pair_single_neon, export=1
-1:      ld1         {v0.4S,v1.4S}, [x1], #32
-        ld1         {v2.4S},       [x2], #16
-        zip1        v3.4S, v2.4S, v2.4S
-        zip2        v4.4S, v2.4S, v2.4S
-        fmul        v0.4S, v0.4S, v3.4S
-        fmul        v1.4S, v1.4S, v4.4S
-        st1         {v0.4S,v1.4S}, [x0], #32
-        subs        w3, w3, #4
-        b.gt        1b
+1:      ld1             {v0.4s,v1.4s}, [x1], #32
+        ld1             {v2.4s},       [x2], #16
+        zip1            v3.4s, v2.4s, v2.4s
+        zip2            v4.4s, v2.4s, v2.4s
+        fmul            v0.4s, v0.4s, v3.4s
+        fmul            v1.4s, v1.4s, v4.4s
+        st1             {v0.4s,v1.4s}, [x0], #32
+        subs            w3, w3, #4
+        b.gt            1b
        ret
 endfunc

 function ff_ps_stereo_interpolate_neon, export=1
-        ld1         {v0.4S}, [x2]
-        ld1         {v1.4S}, [x3]
-        zip1        v4.4S, v0.4S, v0.4S
-        zip2        v5.4S, v0.4S, v0.4S
-        zip1        v6.4S, v1.4S, v1.4S
-        zip2        v7.4S, v1.4S, v1.4S
-1:      ld1         {v2.2S}, [x0]
-        ld1         {v3.2S}, [x1]
-        fadd        v4.4S, v4.4S, v6.4S
-        fadd        v5.4S, v5.4S, v7.4S
-        mov         v2.D[1], v2.D[0]
-        mov         v3.D[1], v3.D[0]
-        fmul        v2.4S, v2.4S, v4.4S
-        fmla        v2.4S, v3.4S, v5.4S
-        st1         {v2.D}[0], [x0], #8
-        st1         {v2.D}[1], [x1], #8
-        subs        w4, w4, #1
-        b.gt        1b
+        ld1             {v0.4s}, [x2]
+        ld1             {v1.4s}, [x3]
+        zip1            v4.4s, v0.4s, v0.4s
+        zip2            v5.4s, v0.4s, v0.4s
+        zip1            v6.4s, v1.4s, v1.4s
+        zip2            v7.4s, v1.4s, v1.4s
+1:      ld1             {v2.2s}, [x0]
+        ld1             {v3.2s}, [x1]
+        fadd            v4.4s, v4.4s, v6.4s
+        fadd            v5.4s, v5.4s, v7.4s
+        mov             v2.d[1], v2.d[0]
+        mov             v3.d[1], v3.d[0]
+        fmul            v2.4s, v2.4s, v4.4s
+        fmla            v2.4s, v3.4s, v5.4s
+        st1             {v2.d}[0], [x0], #8
+        st1             {v2.d}[1], [x1], #8
+        subs            w4, w4, #1
+        b.gt            1b
        ret
 endfunc

 function ff_ps_stereo_interpolate_ipdopd_neon, export=1
-        ld1         {v0.4S,v1.4S}, [x2]
-        ld1         {v6.4S,v7.4S}, [x3]
-        fneg        v2.4S, v1.4S
-        fneg        v3.4S, v7.4S
-        zip1        v16.4S, v0.4S, v0.4S
-        zip2        v17.4S, v0.4S, v0.4S
-        zip1        v18.4S, v2.4S, v1.4S
-        zip2        v19.4S, v2.4S, v1.4S
-        zip1        v20.4S, v6.4S, v6.4S
-        zip2        v21.4S, v6.4S, v6.4S
-        zip1        v22.4S, v3.4S, v7.4S
-        zip2        v23.4S, v3.4S, v7.4S
-1:      ld1         {v2.2S}, [x0]
-        ld1         {v3.2S}, [x1]
-        fadd        v16.4S, v16.4S, v20.4S
-        fadd        v17.4S, v17.4S, v21.4S
-        mov         v2.D[1], v2.D[0]
-        mov         v3.D[1], v3.D[0]
-        fmul        v4.4S, v2.4S, v16.4S
-        fmla        v4.4S, v3.4S, v17.4S
-        fadd        v18.4S, v18.4S, v22.4S
-        fadd        v19.4S, v19.4S, v23.4S
-        ext         v2.16B, v2.16B, v2.16B, #4
-        ext         v3.16B, v3.16B, v3.16B, #4
-        fmla        v4.4S, v2.4S, v18.4S
-        fmla        v4.4S, v3.4S, v19.4S
-        st1         {v4.D}[0], [x0], #8
-        st1         {v4.D}[1], [x1], #8
-        subs        w4, w4, #1
-        b.gt        1b
+        ld1             {v0.4s,v1.4s}, [x2]
+        ld1             {v6.4s,v7.4s}, [x3]
+        fneg            v2.4s, v1.4s
+        fneg            v3.4s, v7.4s
+        zip1            v16.4s, v0.4s, v0.4s
+        zip2            v17.4s, v0.4s, v0.4s
+        zip1            v18.4s, v2.4s, v1.4s
+        zip2            v19.4s, v2.4s, v1.4s
+        zip1            v20.4s, v6.4s, v6.4s
+        zip2            v21.4s, v6.4s, v6.4s
+        zip1            v22.4s, v3.4s, v7.4s
+        zip2            v23.4s, v3.4s, v7.4s
+1:      ld1             {v2.2s}, [x0]
+        ld1             {v3.2s}, [x1]
+        fadd            v16.4s, v16.4s, v20.4s
+        fadd            v17.4s, v17.4s, v21.4s
+        mov             v2.d[1], v2.d[0]
+        mov             v3.d[1], v3.d[0]
+        fmul            v4.4s, v2.4s, v16.4s
+        fmla            v4.4s, v3.4s, v17.4s
+        fadd            v18.4s, v18.4s, v22.4s
+        fadd            v19.4s, v19.4s, v23.4s
+        ext             v2.16b, v2.16b, v2.16b, #4
+        ext             v3.16b, v3.16b, v3.16b, #4
+        fmla            v4.4s, v2.4s, v18.4s
+        fmla            v4.4s, v3.4s, v19.4s
+        st1             {v4.d}[0], [x0], #8
+        st1             {v4.d}[1], [x1], #8
+        subs            w4, w4, #1
+        b.gt            1b
        ret
 endfunc

 function ff_ps_hybrid_analysis_neon, export=1
-        lsl         x3, x3, #3
-        ld2         {v0.4S,v1.4S}, [x1], #32
-        ld2         {v2.2S,v3.2S}, [x1], #16
-        ld1         {v24.2S},      [x1], #8
-        ld2         {v4.2S,v5.2S}, [x1], #16
-        ld2         {v6.4S,v7.4S}, [x1]
-        rev64       v6.4S, v6.4S
-        rev64       v7.4S, v7.4S
-        ext         v6.16B, v6.16B, v6.16B, #8
-        ext         v7.16B, v7.16B, v7.16B, #8
-        rev64       v4.2S, v4.2S
-        rev64       v5.2S, v5.2S
-        mov         v2.D[1], v3.D[0]
-        mov         v4.D[1], v5.D[0]
-        mov         v5.D[1], v2.D[0]
-        mov         v3.D[1], v4.D[0]
-        fadd        v16.4S, v0.4S, v6.4S
-        fadd        v17.4S, v1.4S, v7.4S
-        fsub        v18.4S, v1.4S, v7.4S
-        fsub        v19.4S, v0.4S, v6.4S
-        fadd        v22.4S, v2.4S, v4.4S
-        fsub        v23.4S, v5.4S, v3.4S
-        trn1        v20.2D, v22.2D, v23.2D      // {re4+re8, re5+re7, im8-im4, im7-im5}
-        trn2        v21.2D, v22.2D, v23.2D      // {im4+im8, im5+im7, re4-re8, re5-re7}
-1:      ld2         {v2.4S,v3.4S}, [x2], #32
-        ld2         {v4.2S,v5.2S}, [x2], #16
-        ld1         {v6.2S},       [x2], #8
-        add         x2, x2, #8
-        mov         v4.D[1], v5.D[0]
-        mov         v6.S[1], v6.S[0]
-        fmul        v6.2S, v6.2S, v24.2S
-        fmul        v0.4S, v2.4S, v16.4S
-        fmul        v1.4S, v2.4S, v17.4S
-        fmls        v0.4S, v3.4S, v18.4S
-        fmla        v1.4S, v3.4S, v19.4S
-        fmla        v0.4S, v4.4S, v20.4S
-        fmla        v1.4S, v4.4S, v21.4S
-        faddp       v0.4S, v0.4S, v1.4S
-        faddp       v0.4S, v0.4S, v0.4S
-        fadd        v0.2S, v0.2S, v6.2S
-        st1         {v0.2S}, [x0], x3
-        subs        w4, w4, #1
-        b.gt        1b
+        lsl             x3, x3, #3
+        ld2             {v0.4s,v1.4s}, [x1], #32
+        ld2             {v2.2s,v3.2s}, [x1], #16
+        ld1             {v24.2s},      [x1], #8
+        ld2             {v4.2s,v5.2s}, [x1], #16
+        ld2             {v6.4s,v7.4s}, [x1]
+        rev64           v6.4s, v6.4s
+        rev64           v7.4s, v7.4s
+        ext             v6.16b, v6.16b, v6.16b, #8
+        ext             v7.16b, v7.16b, v7.16b, #8
+        rev64           v4.2s, v4.2s
+        rev64           v5.2s, v5.2s
+        mov             v2.d[1], v3.d[0]
+        mov             v4.d[1], v5.d[0]
+        mov             v5.d[1], v2.d[0]
+        mov             v3.d[1], v4.d[0]
+        fadd            v16.4s, v0.4s, v6.4s
+        fadd            v17.4s, v1.4s, v7.4s
+        fsub            v18.4s, v1.4s, v7.4s
+        fsub            v19.4s, v0.4s, v6.4s
+        fadd            v22.4s, v2.4s, v4.4s
+        fsub            v23.4s, v5.4s, v3.4s
+        trn1            v20.2d, v22.2d, v23.2d      // {re4+re8, re5+re7, im8-im4, im7-im5}
+        trn2            v21.2d, v22.2d, v23.2d      // {im4+im8, im5+im7, re4-re8, re5-re7}
+1:      ld2             {v2.4s,v3.4s}, [x2], #32
+        ld2             {v4.2s,v5.2s}, [x2], #16
+        ld1             {v6.2s},       [x2], #8
+        add             x2, x2, #8
+        mov             v4.d[1], v5.d[0]
+        mov             v6.s[1], v6.s[0]
+        fmul            v6.2s, v6.2s, v24.2s
+        fmul            v0.4s, v2.4s, v16.4s
+        fmul            v1.4s, v2.4s, v17.4s
+        fmls            v0.4s, v3.4s, v18.4s
+        fmla            v1.4s, v3.4s, v19.4s
+        fmla            v0.4s, v4.4s, v20.4s
+        fmla            v1.4s, v4.4s, v21.4s
+        faddp           v0.4s, v0.4s, v1.4s
+        faddp           v0.4s, v0.4s, v0.4s
+        fadd            v0.2s, v0.2s, v6.2s
+        st1             {v0.2s}, [x0], x3
+        subs            w4, w4, #1
+        b.gt            1b
        ret
 endfunc
@@ -353,18 +353,18 @@ function fft\n\()_neon, align=6
 endfunc
 .endm

-        def_fft    32,    16,     8
-        def_fft    64,    32,    16
-        def_fft   128,    64,    32
-        def_fft   256,   128,    64
-        def_fft   512,   256,   128
-        def_fft  1024,   512,   256
-        def_fft  2048,  1024,   512
-        def_fft  4096,  2048,  1024
-        def_fft  8192,  4096,  2048
-        def_fft 16384,  8192,  4096
-        def_fft 32768, 16384,  8192
-        def_fft 65536, 32768, 16384
+        def_fft         32,    16,     8
+        def_fft         64,    32,    16
+        def_fft         128,    64,    32
+        def_fft         256,   128,    64
+        def_fft         512,   256,   128
+        def_fft         1024,   512,   256
+        def_fft         2048,  1024,   512
+        def_fft         4096,  2048,  1024
+        def_fft         8192,  4096,  2048
+        def_fft         16384,  8192,  4096
+        def_fft         32768, 16384,  8192
+        def_fft         65536, 32768, 16384

 function ff_fft_calc_neon, export=1
        prfm            pldl1keep, [x1]
@@ -36,11 +36,11 @@ function ff_\type\()_\codec\()_chroma_mc8_neon, export=1
        lsl             w9,  w9,  #3
        lsl             w10, w10, #1
        add             w9,  w9,  w10
-        add             x6,  x6,  w9, UXTW
-        ld1r            {v22.8H}, [x6]
+        add             x6,  x6,  w9, uxtw
+        ld1r            {v22.8h}, [x6]
  .endif
  .ifc \codec,vc1
-        movi            v22.8H,   #28
+        movi            v22.8h,   #28
  .endif
        mul             w7,  w4,  w5
        lsl             w14, w5,  #3
@@ -53,139 +53,139 @@ function ff_\type\()_\codec\()_chroma_mc8_neon, export=1
        add             w4,  w4,  #64
        b.eq            2f

-        dup             v0.8B,  w4
-        dup             v1.8B,  w12
-        ld1             {v4.8B, v5.8B}, [x1], x2
-        dup             v2.8B,  w6
-        dup             v3.8B,  w7
-        ext             v5.8B,  v4.8B,  v5.8B,  #1
-1:      ld1             {v6.8B, v7.8B}, [x1], x2
-        umull           v16.8H, v4.8B,  v0.8B
-        umlal           v16.8H, v5.8B,  v1.8B
-        ext             v7.8B,  v6.8B,  v7.8B,  #1
-        ld1             {v4.8B, v5.8B}, [x1], x2
-        umlal           v16.8H, v6.8B,  v2.8B
+        dup             v0.8b,  w4
+        dup             v1.8b,  w12
+        ld1             {v4.8b, v5.8b}, [x1], x2
+        dup             v2.8b,  w6
+        dup             v3.8b,  w7
+        ext             v5.8b,  v4.8b,  v5.8b,  #1
+1:      ld1             {v6.8b, v7.8b}, [x1], x2
+        umull           v16.8h, v4.8b,  v0.8b
+        umlal           v16.8h, v5.8b,  v1.8b
+        ext             v7.8b,  v6.8b,  v7.8b,  #1
+        ld1             {v4.8b, v5.8b}, [x1], x2
+        umlal           v16.8h, v6.8b,  v2.8b
        prfm            pldl1strm, [x1]
-        ext             v5.8B,  v4.8B,  v5.8B,  #1
-        umlal           v16.8H, v7.8B,  v3.8B
-        umull           v17.8H, v6.8B,  v0.8B
+        ext             v5.8b,  v4.8b,  v5.8b,  #1
+        umlal           v16.8h, v7.8b,  v3.8b
+        umull           v17.8h, v6.8b,  v0.8b
        subs            w3,  w3,  #2
-        umlal           v17.8H, v7.8B, v1.8B
-        umlal           v17.8H, v4.8B, v2.8B
-        umlal           v17.8H, v5.8B, v3.8B
+        umlal           v17.8h, v7.8b, v1.8b
+        umlal           v17.8h, v4.8b, v2.8b
+        umlal           v17.8h, v5.8b, v3.8b
        prfm            pldl1strm, [x1, x2]
  .ifc \codec,h264
-        rshrn           v16.8B, v16.8H, #6
-        rshrn           v17.8B, v17.8H, #6
+        rshrn           v16.8b, v16.8h, #6
+        rshrn           v17.8b, v17.8h, #6
  .else
-        add             v16.8H, v16.8H, v22.8H
-        add             v17.8H, v17.8H, v22.8H
-        shrn            v16.8B, v16.8H, #6
-        shrn            v17.8B, v17.8H, #6
+        add             v16.8h, v16.8h, v22.8h
+        add             v17.8h, v17.8h, v22.8h
+        shrn            v16.8b, v16.8h, #6
+        shrn            v17.8b, v17.8h, #6
  .endif
  .ifc \type,avg
-        ld1             {v20.8B}, [x8], x2
-        ld1             {v21.8B}, [x8], x2
-        urhadd          v16.8B, v16.8B, v20.8B
-        urhadd          v17.8B, v17.8B, v21.8B
+        ld1             {v20.8b}, [x8], x2
+        ld1             {v21.8b}, [x8], x2
+        urhadd          v16.8b, v16.8b, v20.8b
+        urhadd          v17.8b, v17.8b, v21.8b
  .endif
-        st1             {v16.8B}, [x0], x2
-        st1             {v17.8B}, [x0], x2
+        st1             {v16.8b}, [x0], x2
+        st1             {v17.8b}, [x0], x2
        b.gt            1b
        ret

 2:      adds            w12, w12, w6
-        dup             v0.8B, w4
+        dup             v0.8b, w4
        b.eq            5f
        tst             w6,  w6
-        dup             v1.8B, w12
+        dup             v1.8b, w12
        b.eq            4f

-        ld1             {v4.8B}, [x1], x2
-3:      ld1             {v6.8B}, [x1], x2
-        umull           v16.8H, v4.8B,  v0.8B
-        umlal           v16.8H, v6.8B,  v1.8B
-        ld1             {v4.8B}, [x1], x2
-        umull           v17.8H, v6.8B,  v0.8B
-        umlal           v17.8H, v4.8B,  v1.8B
+        ld1             {v4.8b}, [x1], x2
+3:      ld1             {v6.8b}, [x1], x2
+        umull           v16.8h, v4.8b,  v0.8b
+        umlal           v16.8h, v6.8b,  v1.8b
+        ld1             {v4.8b}, [x1], x2
+        umull           v17.8h, v6.8b,  v0.8b
+        umlal           v17.8h, v4.8b,  v1.8b
        prfm            pldl1strm, [x1]
  .ifc \codec,h264
-        rshrn           v16.8B, v16.8H, #6
-        rshrn           v17.8B, v17.8H, #6
+        rshrn           v16.8b, v16.8h, #6
+        rshrn           v17.8b, v17.8h, #6
  .else
-        add             v16.8H, v16.8H, v22.8H
-        add             v17.8H, v17.8H, v22.8H
-        shrn            v16.8B, v16.8H, #6
-        shrn            v17.8B, v17.8H, #6
+        add             v16.8h, v16.8h, v22.8h
+        add             v17.8h, v17.8h, v22.8h
+        shrn            v16.8b, v16.8h, #6
+        shrn            v17.8b, v17.8h, #6
  .endif
        prfm            pldl1strm, [x1, x2]
  .ifc \type,avg
-        ld1             {v20.8B}, [x8], x2
-        ld1             {v21.8B}, [x8], x2
-        urhadd          v16.8B, v16.8B, v20.8B
-        urhadd          v17.8B, v17.8B, v21.8B
+        ld1             {v20.8b}, [x8], x2
+        ld1             {v21.8b}, [x8], x2
+        urhadd          v16.8b, v16.8b, v20.8b
+        urhadd          v17.8b, v17.8b, v21.8b
  .endif
        subs            w3,  w3,  #2
-        st1             {v16.8B}, [x0], x2
-        st1             {v17.8B}, [x0], x2
+        st1             {v16.8b}, [x0], x2
+        st1             {v17.8b}, [x0], x2
        b.gt            3b
        ret

-4:      ld1             {v4.8B, v5.8B}, [x1], x2
-        ld1             {v6.8B, v7.8B}, [x1], x2
-        ext             v5.8B,  v4.8B,  v5.8B,  #1
-        ext             v7.8B,  v6.8B,  v7.8B,  #1
+4:      ld1             {v4.8b, v5.8b}, [x1], x2
+        ld1             {v6.8b, v7.8b}, [x1], x2
+        ext             v5.8b,  v4.8b,  v5.8b,  #1
+        ext             v7.8b,  v6.8b,  v7.8b,  #1
        prfm            pldl1strm, [x1]
        subs            w3,  w3,  #2
-        umull           v16.8H, v4.8B, v0.8B
-        umlal           v16.8H, v5.8B, v1.8B
-        umull           v17.8H, v6.8B, v0.8B
-        umlal           v17.8H, v7.8B, v1.8B
+        umull           v16.8h, v4.8b, v0.8b
+        umlal           v16.8h, v5.8b, v1.8b
+        umull           v17.8h, v6.8b, v0.8b
+        umlal           v17.8h, v7.8b, v1.8b
        prfm            pldl1strm, [x1, x2]
  .ifc \codec,h264
-        rshrn           v16.8B, v16.8H, #6
-        rshrn           v17.8B, v17.8H, #6
+        rshrn           v16.8b, v16.8h, #6
+        rshrn           v17.8b, v17.8h, #6
  .else
-        add             v16.8H, v16.8H, v22.8H
-        add             v17.8H, v17.8H, v22.8H
-        shrn            v16.8B, v16.8H, #6
-        shrn            v17.8B, v17.8H, #6
+        add             v16.8h, v16.8h, v22.8h
+        add             v17.8h, v17.8h, v22.8h
+        shrn            v16.8b, v16.8h, #6
+        shrn            v17.8b, v17.8h, #6
  .endif
  .ifc \type,avg
-        ld1             {v20.8B}, [x8], x2
-        ld1             {v21.8B}, [x8], x2
-        urhadd          v16.8B, v16.8B, v20.8B
-        urhadd          v17.8B, v17.8B, v21.8B
+        ld1             {v20.8b}, [x8], x2
+        ld1             {v21.8b}, [x8], x2
+        urhadd          v16.8b, v16.8b, v20.8b
+        urhadd          v17.8b, v17.8b, v21.8b
  .endif
-        st1             {v16.8B}, [x0], x2
-        st1             {v17.8B}, [x0], x2
+        st1             {v16.8b}, [x0], x2
+        st1             {v17.8b}, [x0], x2
        b.gt            4b
        ret

-5:      ld1             {v4.8B}, [x1], x2
-        ld1             {v5.8B}, [x1], x2
+5:      ld1             {v4.8b}, [x1], x2
+        ld1             {v5.8b}, [x1], x2
        prfm            pldl1strm, [x1]
        subs            w3,  w3,  #2
-        umull           v16.8H, v4.8B, v0.8B
-        umull           v17.8H, v5.8B, v0.8B
+        umull           v16.8h, v4.8b, v0.8b
+        umull           v17.8h, v5.8b, v0.8b
        prfm            pldl1strm, [x1, x2]
  .ifc \codec,h264
-        rshrn           v16.8B, v16.8H, #6
-        rshrn           v17.8B, v17.8H, #6
+        rshrn           v16.8b, v16.8h, #6
+        rshrn           v17.8b, v17.8h, #6
  .else
-        add             v16.8H, v16.8H, v22.8H
-        add             v17.8H, v17.8H, v22.8H
-        shrn            v16.8B, v16.8H, #6
-        shrn            v17.8B, v17.8H, #6
+        add             v16.8h, v16.8h, v22.8h
+        add             v17.8h, v17.8h, v22.8h
+        shrn            v16.8b, v16.8h, #6
+        shrn            v17.8b, v17.8h, #6
  .endif
  .ifc \type,avg
-        ld1             {v20.8B}, [x8], x2
-        ld1             {v21.8B}, [x8], x2
-        urhadd          v16.8B, v16.8B, v20.8B
-        urhadd          v17.8B, v17.8B, v21.8B
+        ld1             {v20.8b}, [x8], x2
+        ld1             {v21.8b}, [x8], x2
+        urhadd          v16.8b, v16.8b, v20.8b
+        urhadd          v17.8b, v17.8b, v21.8b
  .endif
-        st1             {v16.8B}, [x0], x2
-        st1             {v17.8B}, [x0], x2
+        st1             {v16.8b}, [x0], x2
+        st1             {v17.8b}, [x0], x2
        b.gt            5b
        ret
 endfunc
@@ -206,11 +206,11 @@ function ff_\type\()_\codec\()_chroma_mc4_neon, export=1
        lsl             w9,  w9,  #3
        lsl             w10, w10, #1
        add             w9,  w9,  w10
-        add             x6,  x6,  w9, UXTW
-        ld1r            {v22.8H}, [x6]
+        add             x6,  x6,  w9, uxtw
+        ld1r            {v22.8h}, [x6]
  .endif
  .ifc \codec,vc1
-        movi            v22.8H,   #28
+        movi            v22.8h,   #28
  .endif
        mul             w7,  w4,  w5
        lsl             w14, w5,  #3
@@ -223,133 +223,133 @@ function ff_\type\()_\codec\()_chroma_mc4_neon, export=1
        add             w4,  w4,  #64
        b.eq            2f

-        dup             v24.8B,  w4
-        dup             v25.8B,  w12
-        ld1             {v4.8B}, [x1], x2
-        dup             v26.8B,  w6
-        dup             v27.8B,  w7
-        ext             v5.8B,  v4.8B,  v5.8B, #1
-        trn1            v0.2S,  v24.2S, v25.2S
-        trn1            v2.2S,  v26.2S, v27.2S
-        trn1            v4.2S,  v4.2S,  v5.2S
-1:      ld1             {v6.8B}, [x1], x2
-        ext             v7.8B,  v6.8B,  v7.8B, #1
-        trn1            v6.2S,  v6.2S,  v7.2S
-        umull           v18.8H, v4.8B,  v0.8B
-        umlal           v18.8H, v6.8B,  v2.8B
-        ld1             {v4.8B}, [x1], x2
-        ext             v5.8B,  v4.8B,  v5.8B, #1
-        trn1            v4.2S,  v4.2S,  v5.2S
+        dup             v24.8b,  w4
+        dup             v25.8b,  w12
+        ld1             {v4.8b}, [x1], x2
+        dup             v26.8b,  w6
+        dup             v27.8b,  w7
+        ext             v5.8b,  v4.8b,  v5.8b, #1
+        trn1            v0.2s,  v24.2s, v25.2s
+        trn1            v2.2s,  v26.2s, v27.2s
+        trn1            v4.2s,  v4.2s,  v5.2s
+1:      ld1             {v6.8b}, [x1], x2
+        ext             v7.8b,  v6.8b,  v7.8b, #1
+        trn1            v6.2s,  v6.2s,  v7.2s
+        umull           v18.8h, v4.8b,  v0.8b
+        umlal           v18.8h, v6.8b,  v2.8b
+        ld1             {v4.8b}, [x1], x2
+        ext             v5.8b,  v4.8b,  v5.8b, #1
+        trn1            v4.2s,  v4.2s,  v5.2s
        prfm            pldl1strm, [x1]
-        umull           v19.8H, v6.8B,  v0.8B
-        umlal           v19.8H, v4.8B,  v2.8B
-        trn1            v30.2D, v18.2D, v19.2D
-        trn2            v31.2D, v18.2D, v19.2D
-        add             v18.8H, v30.8H, v31.8H
+        umull           v19.8h, v6.8b,  v0.8b
+        umlal           v19.8h, v4.8b,  v2.8b
+        trn1            v30.2d, v18.2d, v19.2d
+        trn2            v31.2d, v18.2d, v19.2d
+        add             v18.8h, v30.8h, v31.8h
  .ifc \codec,h264
-        rshrn           v16.8B, v18.8H, #6
+        rshrn           v16.8b, v18.8h, #6
  .else
-        add             v18.8H, v18.8H, v22.8H
-        shrn            v16.8B, v18.8H, #6
+        add             v18.8h, v18.8h, v22.8h
+        shrn            v16.8b, v18.8h, #6
  .endif
        subs            w3,  w3,  #2
        prfm            pldl1strm, [x1, x2]
  .ifc \type,avg
-        ld1             {v20.S}[0], [x8], x2
-        ld1             {v20.S}[1], [x8], x2
-        urhadd          v16.8B, v16.8B, v20.8B
+        ld1             {v20.s}[0], [x8], x2
+        ld1             {v20.s}[1], [x8], x2
+        urhadd          v16.8b, v16.8b, v20.8b
  .endif
-        st1             {v16.S}[0], [x0], x2
-        st1             {v16.S}[1], [x0], x2
+        st1             {v16.s}[0], [x0], x2
+        st1             {v16.s}[1], [x0], x2
        b.gt            1b
        ret

 2:      adds            w12, w12, w6
-        dup             v30.8B, w4
+        dup             v30.8b, w4
        b.eq            5f
        tst             w6,  w6
-        dup             v31.8B, w12
-        trn1            v0.2S,  v30.2S, v31.2S
-        trn2            v1.2S,  v30.2S, v31.2S
+        dup             v31.8b, w12
+        trn1            v0.2s,  v30.2s, v31.2s
+        trn2            v1.2s,  v30.2s, v31.2s
        b.eq            4f

-        ext             v1.8B,  v0.8B,  v1.8B, #4
-        ld1             {v4.S}[0], [x1], x2
-3:      ld1             {v4.S}[1], [x1], x2
-        umull           v18.8H, v4.8B,  v0.8B
-        ld1             {v4.S}[0], [x1], x2
-        umull           v19.8H, v4.8B,  v1.8B
-        trn1            v30.2D, v18.2D, v19.2D
-        trn2            v31.2D, v18.2D, v19.2D
-        add             v18.8H, v30.8H, v31.8H
+        ext             v1.8b,  v0.8b,  v1.8b, #4
+        ld1             {v4.s}[0], [x1], x2
+3:      ld1             {v4.s}[1], [x1], x2
+        umull           v18.8h, v4.8b,  v0.8b
+        ld1             {v4.s}[0], [x1], x2
+        umull           v19.8h, v4.8b,  v1.8b
+        trn1            v30.2d, v18.2d, v19.2d
+        trn2            v31.2d, v18.2d, v19.2d
+        add             v18.8h, v30.8h, v31.8h
        prfm            pldl1strm, [x1]
  .ifc \codec,h264
-        rshrn           v16.8B, v18.8H, #6
+        rshrn           v16.8b, v18.8h, #6
  .else
-        add             v18.8H, v18.8H, v22.8H
-        shrn            v16.8B, v18.8H, #6
+        add             v18.8h, v18.8h, v22.8h
+        shrn            v16.8b, v18.8h, #6
  .endif
  .ifc \type,avg
-        ld1             {v20.S}[0], [x8], x2
-        ld1             {v20.S}[1], [x8], x2
-        urhadd          v16.8B, v16.8B, v20.8B
+        ld1             {v20.s}[0], [x8], x2
+        ld1             {v20.s}[1], [x8], x2
+        urhadd          v16.8b, v16.8b, v20.8b
  .endif
        subs            w3,  w3,  #2
        prfm            pldl1strm, [x1, x2]
-        st1             {v16.S}[0], [x0], x2
-        st1             {v16.S}[1], [x0], x2
+        st1             {v16.s}[0], [x0], x2
+        st1             {v16.s}[1], [x0], x2
        b.gt            3b
        ret

-4:      ld1             {v4.8B}, [x1], x2
-        ld1             {v6.8B}, [x1], x2
-        ext             v5.8B,  v4.8B,  v5.8B, #1
-        ext             v7.8B,  v6.8B,  v7.8B, #1
-        trn1            v4.2S,  v4.2S,  v5.2S
-        trn1            v6.2S,  v6.2S,  v7.2S
-        umull           v18.8H, v4.8B,  v0.8B
-        umull           v19.8H, v6.8B,  v0.8B
+4:      ld1             {v4.8b}, [x1], x2
+        ld1             {v6.8b}, [x1], x2
+        ext             v5.8b,  v4.8b,  v5.8b, #1
+        ext             v7.8b,  v6.8b,  v7.8b, #1
+        trn1            v4.2s,  v4.2s,  v5.2s
+        trn1            v6.2s,  v6.2s,  v7.2s
+        umull           v18.8h, v4.8b,  v0.8b
+        umull           v19.8h, v6.8b,  v0.8b
        subs            w3,  w3,  #2
-        trn1            v30.2D, v18.2D, v19.2D
-        trn2            v31.2D, v18.2D, v19.2D
-        add             v18.8H, v30.8H, v31.8H
+        trn1            v30.2d, v18.2d, v19.2d
+        trn2            v31.2d, v18.2d, v19.2d
+        add             v18.8h, v30.8h, v31.8h
        prfm            pldl1strm, [x1]
  .ifc \codec,h264
-        rshrn           v16.8B, v18.8H, #6
+        rshrn           v16.8b, v18.8h, #6
  .else
-        add             v18.8H, v18.8H, v22.8H
-        shrn            v16.8B, v18.8H, #6
+        add             v18.8h, v18.8h, v22.8h
+        shrn            v16.8b, v18.8h, #6
  .endif
  .ifc \type,avg
-        ld1             {v20.S}[0], [x8], x2
-        ld1             {v20.S}[1], [x8], x2
-        urhadd          v16.8B, v16.8B, v20.8B
+        ld1             {v20.s}[0], [x8], x2
+        ld1             {v20.s}[1], [x8], x2
+        urhadd          v16.8b, v16.8b, v20.8b
  .endif
        prfm            pldl1strm, [x1]
-        st1             {v16.S}[0], [x0], x2
-        st1             {v16.S}[1], [x0], x2
+        st1             {v16.s}[0], [x0], x2
+        st1             {v16.s}[1], [x0], x2
        b.gt            4b
        ret

-5:      ld1             {v4.S}[0], [x1], x2
-        ld1             {v4.S}[1], [x1], x2
-        umull           v18.8H, v4.8B,  v30.8B
+5:      ld1             {v4.s}[0], [x1], x2
+        ld1             {v4.s}[1], [x1], x2
+        umull           v18.8h, v4.8b,  v30.8b
        subs            w3,  w3,  #2
        prfm            pldl1strm, [x1]
  .ifc \codec,h264
-        rshrn           v16.8B, v18.8H, #6
+        rshrn           v16.8b, v18.8h, #6
  .else
-        add             v18.8H, v18.8H, v22.8H
-        shrn            v16.8B, v18.8H, #6
+        add             v18.8h, v18.8h, v22.8h
+        shrn            v16.8b, v18.8h, #6
  .endif
  .ifc \type,avg
-        ld1             {v20.S}[0], [x8], x2
-        ld1             {v20.S}[1], [x8], x2
-        urhadd          v16.8B, v16.8B, v20.8B
+        ld1             {v20.s}[0], [x8], x2
+        ld1             {v20.s}[1], [x8], x2
+        urhadd          v16.8b, v16.8b, v20.8b
  .endif
        prfm            pldl1strm, [x1]
-        st1             {v16.S}[0], [x0], x2
-        st1             {v16.S}[1], [x0], x2
+        st1             {v16.s}[0], [x0], x2
+        st1             {v16.s}[1], [x0], x2
        b.gt            5b
        ret
 endfunc
@@ -370,51 +370,51 @@ function ff_\type\()_h264_chroma_mc2_neon, export=1
        sub             w4,  w7,  w13
        sub             w4,  w4,  w14
        add             w4,  w4,  #64
-        dup             v0.8B,  w4
-        dup             v2.8B,  w12
-        dup             v1.8B,  w6
-        dup             v3.8B,  w7
-        trn1            v0.4H,  v0.4H,  v2.4H
-        trn1            v1.4H,  v1.4H,  v3.4H
+        dup             v0.8b,  w4
+        dup             v2.8b,  w12
+        dup             v1.8b,  w6
+        dup             v3.8b,  w7
+        trn1            v0.4h,  v0.4h,  v2.4h
+        trn1            v1.4h,  v1.4h,  v3.4h
 1:
-        ld1             {v4.S}[0],  [x1], x2
-        ld1             {v4.S}[1],  [x1], x2
-        rev64           v5.2S,  v4.2S
-        ld1             {v5.S}[1],  [x1]
-        ext             v6.8B,  v4.8B,  v5.8B,  #1
-        ext             v7.8B,  v5.8B,  v4.8B,  #1
-        trn1            v4.4H,  v4.4H,  v6.4H
-        trn1            v5.4H,  v5.4H,  v7.4H
-        umull           v16.8H, v4.8B,  v0.8B
-        umlal           v16.8H, v5.8B,  v1.8B
+        ld1             {v4.s}[0],  [x1], x2
+        ld1             {v4.s}[1],  [x1], x2
+        rev64           v5.2s,  v4.2s
+        ld1             {v5.s}[1],  [x1]
+        ext             v6.8b,  v4.8b,  v5.8b,  #1
+        ext             v7.8b,  v5.8b,  v4.8b,  #1
+        trn1            v4.4h,  v4.4h,  v6.4h
+        trn1            v5.4h,  v5.4h,  v7.4h
+        umull           v16.8h, v4.8b,  v0.8b
+        umlal           v16.8h, v5.8b,  v1.8b
  .ifc \type,avg
-        ld1             {v18.H}[0], [x0], x2
-        ld1             {v18.H}[2], [x0]
+        ld1             {v18.h}[0], [x0], x2
+        ld1             {v18.h}[2], [x0]
        sub             x0,  x0,  x2
  .endif
-        rev64           v17.4S, v16.4S
-        add             v16.8H, v16.8H, v17.8H
-        rshrn           v16.8B, v16.8H, #6
+        rev64           v17.4s, v16.4s
+        add             v16.8h, v16.8h, v17.8h
+        rshrn           v16.8b, v16.8h, #6
  .ifc \type,avg
-        urhadd          v16.8B, v16.8B, v18.8B
+        urhadd          v16.8b, v16.8b, v18.8b
  .endif
-        st1             {v16.H}[0], [x0], x2
-        st1             {v16.H}[2], [x0], x2
+        st1             {v16.h}[0], [x0], x2
+        st1             {v16.h}[2], [x0], x2
        subs            w3,  w3,  #2
        b.gt            1b
        ret

 2:
-        ld1             {v16.H}[0], [x1], x2
-        ld1             {v16.H}[1], [x1], x2
+        ld1             {v16.h}[0], [x1], x2
+        ld1             {v16.h}[1], [x1], x2
  .ifc \type,avg
-        ld1             {v18.H}[0], [x0], x2
-        ld1             {v18.H}[1], [x0]
+        ld1             {v18.h}[0], [x0], x2
+        ld1             {v18.h}[1], [x0]
        sub             x0,  x0,  x2
-        urhadd          v16.8B, v16.8B, v18.8B
+        urhadd          v16.8b, v16.8b, v18.8b
  .endif
-        st1             {v16.H}[0], [x0], x2
-        st1             {v16.H}[1], [x0], x2
+        st1             {v16.h}[0], [x0], x2
+        st1             {v16.h}[1], [x0], x2
        subs            w3,  w3,  #2
        b.gt            2b
        ret
@@ -27,114 +27,114 @@
 .macro  lowpass_const   r
        movz            \r, #20, lsl #16
        movk            \r, #5
-        mov             v6.S[0], \r
+        mov             v6.s[0], \r
 .endm

 //trashes v0-v5
 .macro  lowpass_8       r0,  r1,  r2,  r3,  d0,  d1,  narrow=1
-        ext             v2.8B,      \r0\().8B, \r1\().8B, #2
-        ext             v3.8B,      \r0\().8B, \r1\().8B, #3
-        uaddl           v2.8H,      v2.8B,     v3.8B
-        ext             v4.8B,      \r0\().8B, \r1\().8B, #1
-        ext             v5.8B,      \r0\().8B, \r1\().8B, #4
-        uaddl           v4.8H,      v4.8B,     v5.8B
-        ext             v1.8B,      \r0\().8B, \r1\().8B, #5
-        uaddl           \d0\().8H,  \r0\().8B, v1.8B
-        ext             v0.8B,      \r2\().8B, \r3\().8B, #2
-        mla             \d0\().8H,  v2.8H,     v6.H[1]
-        ext             v1.8B,      \r2\().8B, \r3\().8B, #3
-        uaddl           v0.8H,      v0.8B,     v1.8B
-        ext             v1.8B,      \r2\().8B, \r3\().8B, #1
-        mls             \d0\().8H,  v4.8H,     v6.H[0]
-        ext             v3.8B,      \r2\().8B, \r3\().8B, #4
-        uaddl           v1.8H,      v1.8B,     v3.8B
-        ext             v2.8B,      \r2\().8B, \r3\().8B, #5
-        uaddl           \d1\().8H,  \r2\().8B, v2.8B
-        mla             \d1\().8H,  v0.8H,     v6.H[1]
-        mls             \d1\().8H,  v1.8H,     v6.H[0]
+        ext             v2.8b,      \r0\().8b, \r1\().8b, #2
+        ext             v3.8b,      \r0\().8b, \r1\().8b, #3
+        uaddl           v2.8h,      v2.8b,     v3.8b
+        ext             v4.8b,      \r0\().8b, \r1\().8b, #1
+        ext             v5.8b,      \r0\().8b, \r1\().8b, #4
+        uaddl           v4.8h,      v4.8b,     v5.8b
+        ext             v1.8b,      \r0\().8b, \r1\().8b, #5
+        uaddl           \d0\().8h,  \r0\().8b, v1.8b
+        ext             v0.8b,      \r2\().8b, \r3\().8b, #2
+        mla             \d0\().8h,  v2.8h,     v6.h[1]
+        ext             v1.8b,      \r2\().8b, \r3\().8b, #3
+        uaddl           v0.8h,      v0.8b,     v1.8b
+        ext             v1.8b,      \r2\().8b, \r3\().8b, #1
+        mls             \d0\().8h,  v4.8h,     v6.h[0]
+        ext             v3.8b,      \r2\().8b, \r3\().8b, #4
+        uaddl           v1.8h,      v1.8b,     v3.8b
+        ext             v2.8b,      \r2\().8b, \r3\().8b, #5
+        uaddl           \d1\().8h,  \r2\().8b, v2.8b
+        mla             \d1\().8h,  v0.8h,     v6.h[1]
+        mls             \d1\().8h,  v1.8h,     v6.h[0]
  .if \narrow
-        sqrshrun        \d0\().8B,  \d0\().8H, #5
-        sqrshrun        \d1\().8B,  \d1\().8H, #5
+        sqrshrun        \d0\().8b,  \d0\().8h, #5
+        sqrshrun        \d1\().8b,  \d1\().8h, #5
  .endif
 .endm

 //trashes v0-v5, v7, v30-v31
 .macro  lowpass_8H      r0,  r1
-        ext             v0.16B,     \r0\().16B, \r0\().16B, #2
-        ext             v1.16B,     \r0\().16B, \r0\().16B, #3
-        uaddl           v0.8H,      v0.8B,      v1.8B
-        ext             v2.16B,     \r0\().16B, \r0\().16B, #1
-        ext             v3.16B,     \r0\().16B, \r0\().16B, #4
-        uaddl           v2.8H,      v2.8B,      v3.8B
-        ext             v30.16B,    \r0\().16B, \r0\().16B, #5
-        uaddl           \r0\().8H,  \r0\().8B,  v30.8B
-        ext             v4.16B,     \r1\().16B, \r1\().16B, #2
-        mla             \r0\().8H,  v0.8H,      v6.H[1]
-        ext             v5.16B,     \r1\().16B, \r1\().16B, #3
-        uaddl           v4.8H,      v4.8B,      v5.8B
-        ext             v7.16B,     \r1\().16B, \r1\().16B, #1
-        mls             \r0\().8H,  v2.8H,      v6.H[0]
-        ext             v0.16B,     \r1\().16B, \r1\().16B, #4
-        uaddl           v7.8H,      v7.8B,      v0.8B
-        ext             v31.16B,    \r1\().16B, \r1\().16B, #5
-        uaddl           \r1\().8H,  \r1\().8B,  v31.8B
-        mla             \r1\().8H,  v4.8H,      v6.H[1]
-        mls             \r1\().8H,  v7.8H,      v6.H[0]
+        ext             v0.16b,     \r0\().16b, \r0\().16b, #2
+        ext             v1.16b,     \r0\().16b, \r0\().16b, #3
+        uaddl           v0.8h,      v0.8b,      v1.8b
+        ext             v2.16b,     \r0\().16b, \r0\().16b, #1
+        ext             v3.16b,     \r0\().16b, \r0\().16b, #4
+        uaddl           v2.8h,      v2.8b,      v3.8b
+        ext             v30.16b,    \r0\().16b, \r0\().16b, #5
+        uaddl           \r0\().8h,  \r0\().8b,  v30.8b
+        ext             v4.16b,     \r1\().16b, \r1\().16b, #2
+        mla             \r0\().8h,  v0.8h,      v6.h[1]
+        ext             v5.16b,     \r1\().16b, \r1\().16b, #3
+        uaddl           v4.8h,      v4.8b,      v5.8b
+        ext             v7.16b,     \r1\().16b, \r1\().16b, #1
+        mls             \r0\().8h,  v2.8h,      v6.h[0]
+        ext             v0.16b,     \r1\().16b, \r1\().16b, #4
+        uaddl           v7.8h,      v7.8b,      v0.8b
+        ext             v31.16b,    \r1\().16b, \r1\().16b, #5
+        uaddl           \r1\().8h,  \r1\().8b,  v31.8b
+        mla             \r1\().8h,  v4.8h,      v6.h[1]
+        mls             \r1\().8h,  v7.8h,      v6.h[0]
 .endm

 // trashes v2-v5, v30
 .macro  lowpass_8_1     r0,  r1,  d0,  narrow=1
-        ext             v2.8B,     \r0\().8B, \r1\().8B, #2
-        ext             v3.8B,     \r0\().8B, \r1\().8B, #3
-        uaddl           v2.8H,     v2.8B,     v3.8B
-        ext             v4.8B,     \r0\().8B, \r1\().8B, #1
-        ext             v5.8B,     \r0\().8B, \r1\().8B, #4
-        uaddl           v4.8H,     v4.8B,     v5.8B
-        ext             v30.8B,    \r0\().8B, \r1\().8B, #5
-        uaddl           \d0\().8H, \r0\().8B, v30.8B
-        mla             \d0\().8H, v2.8H,     v6.H[1]
-        mls             \d0\().8H, v4.8H,     v6.H[0]
+        ext             v2.8b,     \r0\().8b, \r1\().8b, #2
+        ext             v3.8b,     \r0\().8b, \r1\().8b, #3
+        uaddl           v2.8h,     v2.8b,     v3.8b
+        ext             v4.8b,     \r0\().8b, \r1\().8b, #1
+        ext             v5.8b,     \r0\().8b, \r1\().8b, #4
+        uaddl           v4.8h,     v4.8b,     v5.8b
+        ext             v30.8b,    \r0\().8b, \r1\().8b, #5
+        uaddl           \d0\().8h, \r0\().8b, v30.8b
+        mla             \d0\().8h, v2.8h,     v6.h[1]
+        mls             \d0\().8h, v4.8h,     v6.h[0]
  .if \narrow
-        sqrshrun        \d0\().8B, \d0\().8H, #5
+        sqrshrun        \d0\().8b, \d0\().8h, #5
  .endif
 .endm

 // trashed v0-v7
 .macro  lowpass_8.16    r0,  r1,  r2
-        ext             v1.16B,     \r0\().16B, \r1\().16B, #4
-        ext             v0.16B,     \r0\().16B, \r1\().16B, #6
-        saddl           v5.4S,      v1.4H,      v0.4H
-        ext             v2.16B,     \r0\().16B, \r1\().16B, #2
-        saddl2          v1.4S,      v1.8H,      v0.8H
-        ext             v3.16B,     \r0\().16B, \r1\().16B, #8
-        saddl           v6.4S,      v2.4H,      v3.4H
-        ext             \r1\().16B, \r0\().16B, \r1\().16B, #10
-        saddl2          v2.4S,      v2.8H,      v3.8H
-        saddl           v0.4S,      \r0\().4H,  \r1\().4H
-        saddl2          v4.4S,      \r0\().8H,  \r1\().8H
+        ext             v1.16b,     \r0\().16b, \r1\().16b, #4
+        ext             v0.16b,     \r0\().16b, \r1\().16b, #6
+        saddl           v5.4s,      v1.4h,      v0.4h
+        ext             v2.16b,     \r0\().16b, \r1\().16b, #2
+        saddl2          v1.4s,      v1.8h,      v0.8h
+        ext             v3.16b,     \r0\().16b, \r1\().16b, #8
+        saddl           v6.4s,      v2.4h,      v3.4h
+        ext             \r1\().16b, \r0\().16b, \r1\().16b, #10
+        saddl2          v2.4s,      v2.8h,      v3.8h
+        saddl           v0.4s,      \r0\().4h,  \r1\().4h
+        saddl2          v4.4s,      \r0\().8h,  \r1\().8h

-        shl             v3.4S,  v5.4S,  #4
-        shl             v5.4S,  v5.4S,  #2
-        shl             v7.4S,  v6.4S,  #2
-        add             v5.4S,  v5.4S,  v3.4S
-        add             v6.4S,  v6.4S,  v7.4S
+        shl             v3.4s,  v5.4s,  #4
+        shl             v5.4s,  v5.4s,  #2
+        shl             v7.4s,  v6.4s,  #2
+        add             v5.4s,  v5.4s,  v3.4s
+        add             v6.4s,  v6.4s,  v7.4s

-        shl             v3.4S,  v1.4S,  #4
-        shl             v1.4S,  v1.4S,  #2
-        shl             v7.4S,  v2.4S,  #2
-        add             v1.4S,  v1.4S,  v3.4S
-        add             v2.4S,  v2.4S,  v7.4S
+        shl             v3.4s,  v1.4s,  #4
+        shl             v1.4s,  v1.4s,  #2
+        shl             v7.4s,  v2.4s,  #2
+        add             v1.4s,  v1.4s,  v3.4s
+        add             v2.4s,  v2.4s,  v7.4s

-        add             v5.4S,  v5.4S,  v0.4S
-        sub             v5.4S,  v5.4S,  v6.4S
+        add             v5.4s,  v5.4s,  v0.4s
+        sub             v5.4s,  v5.4s,  v6.4s

-        add             v1.4S,  v1.4S,  v4.4S
-        sub             v1.4S,  v1.4S,  v2.4S
+        add             v1.4s,  v1.4s,  v4.4s
+        sub             v1.4s,  v1.4s,  v2.4s

-        rshrn           v5.4H,  v5.4S,  #10
-        rshrn2          v5.8H,  v1.4S,  #10
+        rshrn           v5.4h,  v5.4s,  #10
+        rshrn2          v5.8h,  v1.4s,  #10

-        sqxtun          \r2\().8B,  v5.8H
+        sqxtun          \r2\().8b,  v5.8h
 .endm

 function put_h264_qpel16_h_lowpass_neon_packed
@@ -163,19 +163,19 @@ function \type\()_h264_qpel16_h_lowpass_neon
 endfunc

 function \type\()_h264_qpel8_h_lowpass_neon
-1:      ld1             {v28.8B, v29.8B}, [x1], x2
-        ld1             {v16.8B, v17.8B}, [x1], x2
+1:      ld1             {v28.8b, v29.8b}, [x1], x2
+        ld1             {v16.8b, v17.8b}, [x1], x2
        subs            x12, x12, #2
        lowpass_8       v28, v29, v16, v17, v28, v16
  .ifc \type,avg
-        ld1             {v2.8B},    [x0], x3
-        urhadd          v28.8B, v28.8B,  v2.8B
-        ld1             {v3.8B},    [x0]
-        urhadd          v16.8B, v16.8B, v3.8B
+        ld1             {v2.8b},    [x0], x3
+        urhadd          v28.8b, v28.8b,  v2.8b
+        ld1             {v3.8b},    [x0]
+        urhadd          v16.8b, v16.8b, v3.8b
        sub             x0,  x0,  x3
  .endif
-        st1             {v28.8B},    [x0], x3
-        st1             {v16.8B},    [x0], x3
+        st1             {v28.8b},    [x0], x3
+        st1             {v16.8b},    [x0], x3
        b.ne            1b
        ret
 endfunc
@@ -200,23 +200,23 @@ function \type\()_h264_qpel16_h_lowpass_l2_neon
 endfunc

 function \type\()_h264_qpel8_h_lowpass_l2_neon
-1:      ld1             {v26.8B, v27.8B}, [x1], x2
-        ld1             {v16.8B, v17.8B}, [x1], x2
-        ld1             {v28.8B},     [x3], x2
-        ld1             {v29.8B},     [x3], x2
+1:      ld1             {v26.8b, v27.8b}, [x1], x2
+        ld1             {v16.8b, v17.8b}, [x1], x2
+        ld1             {v28.8b},     [x3], x2
+        ld1             {v29.8b},     [x3], x2
        subs            x12, x12, #2
        lowpass_8       v26, v27, v16, v17, v26, v27
-        urhadd          v26.8B, v26.8B, v28.8B
-        urhadd          v27.8B, v27.8B, v29.8B
+        urhadd          v26.8b, v26.8b, v28.8b
+        urhadd          v27.8b, v27.8b, v29.8b
  .ifc \type,avg
-        ld1             {v2.8B},      [x0], x2
-        urhadd          v26.8B, v26.8B, v2.8B
-        ld1             {v3.8B},      [x0]
-        urhadd          v27.8B, v27.8B, v3.8B
+        ld1             {v2.8b},      [x0], x2
+        urhadd          v26.8b, v26.8b, v2.8b
+        ld1             {v3.8b},      [x0]
+        urhadd          v27.8b, v27.8b, v3.8b
        sub             x0,  x0,  x2
  .endif
-        st1             {v26.8B},     [x0], x2
-        st1             {v27.8B},     [x0], x2
+        st1             {v26.8b},     [x0], x2
+        st1             {v27.8b},     [x0], x2
        b.ne            1b
        ret
 endfunc
@@ -257,19 +257,19 @@ function \type\()_h264_qpel16_v_lowpass_neon
 endfunc

 function \type\()_h264_qpel8_v_lowpass_neon
-        ld1             {v16.8B}, [x1], x3
-        ld1             {v18.8B}, [x1], x3
-        ld1             {v20.8B}, [x1], x3
-        ld1             {v22.8B}, [x1], x3
-        ld1             {v24.8B}, [x1], x3
-        ld1             {v26.8B}, [x1], x3
-        ld1             {v28.8B}, [x1], x3
-        ld1             {v30.8B}, [x1], x3
-        ld1             {v17.8B}, [x1], x3
-        ld1             {v19.8B}, [x1], x3
-        ld1             {v21.8B}, [x1], x3
-        ld1             {v23.8B}, [x1], x3
-        ld1             {v25.8B}, [x1]
+        ld1             {v16.8b}, [x1], x3
+        ld1             {v18.8b}, [x1], x3
+        ld1             {v20.8b}, [x1], x3
+        ld1             {v22.8b}, [x1], x3
+        ld1             {v24.8b}, [x1], x3
+        ld1             {v26.8b}, [x1], x3
+        ld1             {v28.8b}, [x1], x3
+        ld1             {v30.8b}, [x1], x3
+        ld1             {v17.8b}, [x1], x3
+        ld1             {v19.8b}, [x1], x3
+        ld1             {v21.8b}, [x1], x3
+        ld1             {v23.8b}, [x1], x3
+        ld1             {v25.8b}, [x1]

        transpose_8x8B  v16, v18, v20, v22, v24, v26, v28, v30, v0,  v1
        transpose_8x8B  v17, v19, v21, v23, v25, v27, v29, v31, v0,  v1
@@ -280,33 +280,33 @@ function \type\()_h264_qpel8_v_lowpass_neon
        transpose_8x8B  v16, v17, v18, v19, v20, v21, v22, v23, v0,  v1

  .ifc \type,avg
-        ld1             {v24.8B},  [x0], x2
-        urhadd          v16.8B, v16.8B, v24.8B
-        ld1             {v25.8B}, [x0], x2
-        urhadd          v17.8B, v17.8B, v25.8B
-        ld1             {v26.8B}, [x0], x2
-        urhadd          v18.8B, v18.8B, v26.8B
-        ld1             {v27.8B}, [x0], x2
-        urhadd          v19.8B, v19.8B, v27.8B
-        ld1             {v28.8B}, [x0], x2
-        urhadd          v20.8B, v20.8B, v28.8B
-        ld1             {v29.8B}, [x0], x2
-        urhadd          v21.8B, v21.8B, v29.8B
-        ld1             {v30.8B}, [x0], x2
-        urhadd          v22.8B, v22.8B, v30.8B
-        ld1             {v31.8B}, [x0], x2
-        urhadd          v23.8B, v23.8B, v31.8B
+        ld1             {v24.8b},  [x0], x2
+        urhadd          v16.8b, v16.8b, v24.8b
+        ld1             {v25.8b}, [x0], x2
+        urhadd          v17.8b, v17.8b, v25.8b
+        ld1             {v26.8b}, [x0], x2
+        urhadd          v18.8b, v18.8b, v26.8b
+        ld1             {v27.8b}, [x0], x2
+        urhadd          v19.8b, v19.8b, v27.8b
+        ld1             {v28.8b}, [x0], x2
+        urhadd          v20.8b, v20.8b, v28.8b
+        ld1             {v29.8b}, [x0], x2
+        urhadd          v21.8b, v21.8b, v29.8b
+        ld1             {v30.8b}, [x0], x2
+        urhadd          v22.8b, v22.8b, v30.8b
+        ld1             {v31.8b}, [x0], x2
+        urhadd          v23.8b, v23.8b, v31.8b
        sub             x0,  x0,  x2,  lsl #3
  .endif

-        st1             {v16.8B}, [x0], x2
-        st1             {v17.8B}, [x0], x2
-        st1             {v18.8B}, [x0], x2
-        st1             {v19.8B}, [x0], x2
-        st1             {v20.8B}, [x0], x2
-        st1             {v21.8B}, [x0], x2
-        st1             {v22.8B}, [x0], x2
-        st1             {v23.8B}, [x0], x2
+        st1             {v16.8b}, [x0], x2
+        st1             {v17.8b}, [x0], x2
+        st1             {v18.8b}, [x0], x2
+        st1             {v19.8b}, [x0], x2
+        st1             {v20.8b}, [x0], x2
+        st1             {v21.8b}, [x0], x2
+        st1             {v22.8b}, [x0], x2
+        st1             {v23.8b}, [x0], x2

        ret
 endfunc
@@ -334,19 +334,19 @@ function \type\()_h264_qpel16_v_lowpass_l2_neon
 endfunc

 function \type\()_h264_qpel8_v_lowpass_l2_neon
-        ld1             {v16.8B}, [x1], x3
-        ld1             {v18.8B}, [x1], x3
-        ld1             {v20.8B}, [x1], x3
-        ld1             {v22.8B}, [x1], x3
-        ld1             {v24.8B}, [x1], x3
-        ld1             {v26.8B}, [x1], x3
-        ld1             {v28.8B}, [x1], x3
-        ld1             {v30.8B}, [x1], x3
-        ld1             {v17.8B}, [x1], x3
-        ld1             {v19.8B}, [x1], x3
-        ld1             {v21.8B}, [x1], x3
-        ld1             {v23.8B}, [x1], x3
-        ld1             {v25.8B}, [x1]
+        ld1             {v16.8b}, [x1], x3
+        ld1             {v18.8b}, [x1], x3
+        ld1             {v20.8b}, [x1], x3
+        ld1             {v22.8b}, [x1], x3
+        ld1             {v24.8b}, [x1], x3
+        ld1             {v26.8b}, [x1], x3
+        ld1             {v28.8b}, [x1], x3
+        ld1             {v30.8b}, [x1], x3
+        ld1             {v17.8b}, [x1], x3
+        ld1             {v19.8b}, [x1], x3
+        ld1             {v21.8b}, [x1], x3
+        ld1             {v23.8b}, [x1], x3
+        ld1             {v25.8b}, [x1]

        transpose_8x8B  v16, v18, v20, v22, v24, v26, v28, v30, v0,  v1
        transpose_8x8B  v17, v19, v21, v23, v25, v27, v29, v31, v0,  v1
@@ -356,51 +356,51 @@ function \type\()_h264_qpel8_v_lowpass_l2_neon
        lowpass_8       v28, v29, v30, v31, v22, v23
        transpose_8x8B  v16, v17, v18, v19, v20, v21, v22, v23, v0,  v1

-        ld1             {v24.8B},  [x12], x2
-        ld1             {v25.8B},  [x12], x2
-        ld1             {v26.8B},  [x12], x2
-        ld1             {v27.8B},  [x12], x2
-        ld1             {v28.8B},  [x12], x2
-        urhadd          v16.8B, v24.8B, v16.8B
-        urhadd          v17.8B, v25.8B, v17.8B
-        ld1             {v29.8B},  [x12], x2
-        urhadd          v18.8B, v26.8B, v18.8B
-        urhadd          v19.8B, v27.8B, v19.8B
-        ld1             {v30.8B}, [x12], x2
-        urhadd          v20.8B, v28.8B, v20.8B
-        urhadd          v21.8B, v29.8B, v21.8B
-        ld1             {v31.8B}, [x12], x2
-        urhadd          v22.8B, v30.8B, v22.8B
-        urhadd          v23.8B, v31.8B, v23.8B
+        ld1             {v24.8b},  [x12], x2
+        ld1             {v25.8b},  [x12], x2
+        ld1             {v26.8b},  [x12], x2
+        ld1             {v27.8b},  [x12], x2
+        ld1             {v28.8b},  [x12], x2
+        urhadd          v16.8b, v24.8b, v16.8b
+        urhadd          v17.8b, v25.8b, v17.8b
+        ld1             {v29.8b},  [x12], x2
+        urhadd          v18.8b, v26.8b, v18.8b
+        urhadd          v19.8b, v27.8b, v19.8b
+        ld1             {v30.8b}, [x12], x2
+        urhadd          v20.8b, v28.8b, v20.8b
+        urhadd          v21.8b, v29.8b, v21.8b
+        ld1             {v31.8b}, [x12], x2
+        urhadd          v22.8b, v30.8b, v22.8b
+        urhadd          v23.8b, v31.8b, v23.8b

  .ifc \type,avg
-        ld1             {v24.8B}, [x0], x3
-        urhadd          v16.8B, v16.8B, v24.8B
-        ld1             {v25.8B}, [x0], x3
-        urhadd          v17.8B, v17.8B, v25.8B
-        ld1             {v26.8B}, [x0], x3
-        urhadd          v18.8B, v18.8B, v26.8B
-        ld1             {v27.8B}, [x0], x3
-        urhadd          v19.8B, v19.8B, v27.8B
-        ld1             {v28.8B}, [x0], x3
-        urhadd          v20.8B, v20.8B, v28.8B
-        ld1             {v29.8B}, [x0], x3
-        urhadd          v21.8B, v21.8B, v29.8B
-        ld1             {v30.8B}, [x0], x3
-        urhadd          v22.8B, v22.8B, v30.8B
-        ld1             {v31.8B}, [x0], x3
-        urhadd          v23.8B, v23.8B, v31.8B
+        ld1             {v24.8b}, [x0], x3
+        urhadd          v16.8b, v16.8b, v24.8b
+        ld1             {v25.8b}, [x0], x3
+        urhadd          v17.8b, v17.8b, v25.8b
+        ld1             {v26.8b}, [x0], x3
+        urhadd          v18.8b, v18.8b, v26.8b
+        ld1             {v27.8b}, [x0], x3
+        urhadd          v19.8b, v19.8b, v27.8b
+        ld1             {v28.8b}, [x0], x3
+        urhadd          v20.8b, v20.8b, v28.8b
+        ld1             {v29.8b}, [x0], x3
+        urhadd          v21.8b, v21.8b, v29.8b
+        ld1             {v30.8b}, [x0], x3
+        urhadd          v22.8b, v22.8b, v30.8b
+        ld1             {v31.8b}, [x0], x3
+        urhadd          v23.8b, v23.8b, v31.8b
        sub             x0,  x0,  x3,  lsl #3
  .endif

-        st1             {v16.8B}, [x0], x3
-        st1             {v17.8B}, [x0], x3
-        st1             {v18.8B}, [x0], x3
-        st1             {v19.8B}, [x0], x3
-        st1             {v20.8B}, [x0], x3
-        st1             {v21.8B}, [x0], x3
-        st1             {v22.8B}, [x0], x3
-        st1             {v23.8B}, [x0], x3
+        st1             {v16.8b}, [x0], x3
+        st1             {v17.8b}, [x0], x3
+        st1             {v18.8b}, [x0], x3
+        st1             {v19.8b}, [x0], x3
+        st1             {v20.8b}, [x0], x3
+        st1             {v21.8b}, [x0], x3
+        st1             {v22.8b}, [x0], x3
+        st1             {v23.8b}, [x0], x3

        ret
 endfunc
@@ -411,19 +411,19 @@ endfunc

 function put_h264_qpel8_hv_lowpass_neon_top
        lowpass_const   w12
-        ld1             {v16.8H}, [x1], x3
-        ld1             {v17.8H}, [x1], x3
-        ld1             {v18.8H}, [x1], x3
-        ld1             {v19.8H}, [x1], x3
-        ld1             {v20.8H}, [x1], x3
-        ld1             {v21.8H}, [x1], x3
-        ld1             {v22.8H}, [x1], x3
-        ld1             {v23.8H}, [x1], x3
-        ld1             {v24.8H}, [x1], x3
-        ld1             {v25.8H}, [x1], x3
-        ld1             {v26.8H}, [x1], x3
-        ld1             {v27.8H}, [x1], x3
-        ld1             {v28.8H}, [x1]
+        ld1             {v16.8h}, [x1], x3
+        ld1             {v17.8h}, [x1], x3
+        ld1             {v18.8h}, [x1], x3
+        ld1             {v19.8h}, [x1], x3
+        ld1             {v20.8h}, [x1], x3
+        ld1             {v21.8h}, [x1], x3
+        ld1             {v22.8h}, [x1], x3
+        ld1             {v23.8h}, [x1], x3
+        ld1             {v24.8h}, [x1], x3
+        ld1             {v25.8h}, [x1], x3
+        ld1             {v26.8h}, [x1], x3
+        ld1             {v27.8h}, [x1], x3
+        ld1             {v28.8h}, [x1]
        lowpass_8H      v16, v17
        lowpass_8H      v18, v19
        lowpass_8H      v20, v21
@@ -447,7 +447,7 @@ function put_h264_qpel8_hv_lowpass_neon_top
        lowpass_8.16    v22, v30, v22
        lowpass_8.16    v23, v31, v23

-        transpose_8x8B v16, v17, v18, v19, v20, v21, v22, v23, v0,  v1
+        transpose_8x8B  v16, v17, v18, v19, v20, v21, v22, v23, v0,  v1

        ret
 endfunc
@@ -457,33 +457,33 @@ function \type\()_h264_qpel8_hv_lowpass_neon
        mov             x10, x30
        bl              put_h264_qpel8_hv_lowpass_neon_top
  .ifc \type,avg
-        ld1             {v0.8B},      [x0], x2
-        urhadd          v16.8B, v16.8B, v0.8B
-        ld1             {v1.8B},      [x0], x2
-        urhadd          v17.8B, v17.8B, v1.8B
-        ld1             {v2.8B},      [x0], x2
-        urhadd          v18.8B, v18.8B, v2.8B
-        ld1             {v3.8B},      [x0], x2
-        urhadd          v19.8B, v19.8B, v3.8B
-        ld1             {v4.8B},      [x0], x2
-        urhadd          v20.8B, v20.8B, v4.8B
-        ld1             {v5.8B},      [x0], x2
-        urhadd          v21.8B, v21.8B, v5.8B
-        ld1             {v6.8B},      [x0], x2
-        urhadd          v22.8B, v22.8B, v6.8B
-        ld1             {v7.8B},      [x0], x2
-        urhadd          v23.8B, v23.8B, v7.8B
+        ld1             {v0.8b},      [x0], x2
+        urhadd          v16.8b, v16.8b, v0.8b
+        ld1             {v1.8b},      [x0], x2
+        urhadd          v17.8b, v17.8b, v1.8b
+        ld1             {v2.8b},      [x0], x2
+        urhadd          v18.8b, v18.8b, v2.8b
+        ld1             {v3.8b},      [x0], x2
+        urhadd          v19.8b, v19.8b, v3.8b
+        ld1             {v4.8b},      [x0], x2
+        urhadd          v20.8b, v20.8b, v4.8b
+        ld1             {v5.8b},      [x0], x2
+        urhadd          v21.8b, v21.8b, v5.8b
+        ld1             {v6.8b},      [x0], x2
+        urhadd          v22.8b, v22.8b, v6.8b
+        ld1             {v7.8b},      [x0], x2
+        urhadd          v23.8b, v23.8b, v7.8b
        sub             x0,  x0,  x2,  lsl #3
  .endif

-        st1             {v16.8B},     [x0], x2
-        st1             {v17.8B},     [x0], x2
-        st1             {v18.8B},     [x0], x2
-        st1             {v19.8B},     [x0], x2
-        st1             {v20.8B},     [x0], x2
-        st1             {v21.8B},     [x0], x2
-        st1             {v22.8B},     [x0], x2
-        st1             {v23.8B},     [x0], x2
+        st1             {v16.8b},     [x0], x2
+        st1             {v17.8b},     [x0], x2
+        st1             {v18.8b},     [x0], x2
+        st1             {v19.8b},     [x0], x2
+        st1             {v20.8b},     [x0], x2
+        st1             {v21.8b},     [x0], x2
+        st1             {v22.8b},     [x0], x2
+        st1             {v23.8b},     [x0], x2

        ret             x10
 endfunc
@@ -497,45 +497,45 @@ function \type\()_h264_qpel8_hv_lowpass_l2_neon
        mov             x10, x30
        bl              put_h264_qpel8_hv_lowpass_neon_top

-        ld1             {v0.8B, v1.8B},  [x2], #16
-        ld1             {v2.8B, v3.8B},  [x2], #16
-        urhadd          v0.8B,  v0.8B,  v16.8B
-        urhadd          v1.8B,  v1.8B,  v17.8B
-        ld1             {v4.8B, v5.8B},  [x2], #16
-        urhadd          v2.8B,  v2.8B,  v18.8B
-        urhadd          v3.8B,  v3.8B,  v19.8B
-        ld1             {v6.8B, v7.8B},  [x2], #16
-        urhadd          v4.8B,  v4.8B,  v20.8B
-        urhadd          v5.8B,  v5.8B,  v21.8B
-        urhadd          v6.8B,  v6.8B,  v22.8B
-        urhadd          v7.8B,  v7.8B,  v23.8B
+        ld1             {v0.8b, v1.8b},  [x2], #16
+        ld1             {v2.8b, v3.8b},  [x2], #16
+        urhadd          v0.8b,  v0.8b,  v16.8b
+        urhadd          v1.8b,  v1.8b,  v17.8b
+        ld1             {v4.8b, v5.8b},  [x2], #16
+        urhadd          v2.8b,  v2.8b,  v18.8b
+        urhadd          v3.8b,  v3.8b,  v19.8b
+        ld1             {v6.8b, v7.8b},  [x2], #16
+        urhadd          v4.8b,  v4.8b,  v20.8b
+        urhadd          v5.8b,  v5.8b,  v21.8b
+        urhadd          v6.8b,  v6.8b,  v22.8b
+        urhadd          v7.8b,  v7.8b,  v23.8b
  .ifc \type,avg
-        ld1             {v16.8B},     [x0], x3
-        urhadd          v0.8B,  v0.8B,  v16.8B
-        ld1             {v17.8B},     [x0], x3
-        urhadd          v1.8B,  v1.8B,  v17.8B
-        ld1             {v18.8B},     [x0], x3
-        urhadd          v2.8B,  v2.8B,  v18.8B
-        ld1             {v19.8B},     [x0], x3
-        urhadd          v3.8B,  v3.8B,  v19.8B
-        ld1             {v20.8B},     [x0], x3
-        urhadd          v4.8B,  v4.8B,  v20.8B
-        ld1             {v21.8B},     [x0], x3
-        urhadd          v5.8B,  v5.8B,  v21.8B
-        ld1             {v22.8B},     [x0], x3
-        urhadd          v6.8B,  v6.8B,  v22.8B
-        ld1             {v23.8B},     [x0], x3
-        urhadd          v7.8B,  v7.8B,  v23.8B
+        ld1             {v16.8b},     [x0], x3
+        urhadd          v0.8b,  v0.8b,  v16.8b
+        ld1             {v17.8b},     [x0], x3
+        urhadd          v1.8b,  v1.8b,  v17.8b
+        ld1             {v18.8b},     [x0], x3
+        urhadd          v2.8b,  v2.8b,  v18.8b
+        ld1             {v19.8b},     [x0], x3
+        urhadd          v3.8b,  v3.8b,  v19.8b
+        ld1             {v20.8b},     [x0], x3
+        urhadd          v4.8b,  v4.8b,  v20.8b
+        ld1             {v21.8b},     [x0], x3
+        urhadd          v5.8b,  v5.8b,  v21.8b
+        ld1             {v22.8b},     [x0], x3
+        urhadd          v6.8b,  v6.8b,  v22.8b
+        ld1             {v23.8b},     [x0], x3
+        urhadd          v7.8b,  v7.8b,  v23.8b
        sub             x0,  x0,  x3,  lsl #3
  .endif
-        st1             {v0.8B},      [x0], x3
-        st1             {v1.8B},      [x0], x3
-        st1             {v2.8B},      [x0], x3
-        st1             {v3.8B},      [x0], x3
-        st1             {v4.8B},      [x0], x3
-        st1             {v5.8B},      [x0], x3
-        st1             {v6.8B},      [x0], x3
-        st1             {v7.8B},      [x0], x3
+        st1             {v0.8b},      [x0], x3
+        st1             {v1.8b},      [x0], x3
+        st1             {v2.8b},      [x0], x3
+        st1             {v3.8b},      [x0], x3
+        st1             {v4.8b},      [x0], x3
+        st1             {v5.8b},      [x0], x3
+        st1             {v6.8b},      [x0], x3
+        st1             {v7.8b},      [x0], x3

        ret             x10
 endfunc
@@ -579,8 +579,8 @@ function \type\()_h264_qpel16_hv_lowpass_l2_neon
 endfunc
 .endm

-        h264_qpel16_hv put
-        h264_qpel16_hv avg
+        h264_qpel16_hv  put
+        h264_qpel16_hv  avg

 .macro  h264_qpel8      type
 function ff_\type\()_h264_qpel8_mc10_neon, export=1
@@ -758,8 +758,8 @@ function ff_\type\()_h264_qpel8_mc33_neon, export=1
 endfunc
 .endm

-        h264_qpel8 put
-        h264_qpel8 avg
+        h264_qpel8      put
+        h264_qpel8      avg

 .macro  h264_qpel16     type
 function ff_\type\()_h264_qpel16_mc10_neon, export=1
@@ -930,5 +930,5 @@ function ff_\type\()_h264_qpel16_mc33_neon, export=1
 endfunc
 .endm

-        h264_qpel16 put
-        h264_qpel16 avg
+        h264_qpel16     put
+        h264_qpel16     avg
@@ -26,295 +26,295 @@
  .if \avg
        mov             x12, x0
  .endif
-1:      ld1             {v0.16B},  [x1], x2
-        ld1             {v1.16B},  [x1], x2
-        ld1             {v2.16B},  [x1], x2
-        ld1             {v3.16B},  [x1], x2
+1:      ld1             {v0.16b},  [x1], x2
+        ld1             {v1.16b},  [x1], x2
+        ld1             {v2.16b},  [x1], x2
+        ld1             {v3.16b},  [x1], x2
  .if \avg
-        ld1             {v4.16B},  [x12], x2
-        urhadd          v0.16B,  v0.16B,  v4.16B
-        ld1             {v5.16B},  [x12], x2
-        urhadd          v1.16B,  v1.16B,  v5.16B
-        ld1             {v6.16B},  [x12], x2
-        urhadd          v2.16B,  v2.16B,  v6.16B
-        ld1             {v7.16B},  [x12], x2
-        urhadd          v3.16B,  v3.16B,  v7.16B
+        ld1             {v4.16b},  [x12], x2
+        urhadd          v0.16b,  v0.16b,  v4.16b
+        ld1             {v5.16b},  [x12], x2
+        urhadd          v1.16b,  v1.16b,  v5.16b
+        ld1             {v6.16b},  [x12], x2
+        urhadd          v2.16b,  v2.16b,  v6.16b
+        ld1             {v7.16b},  [x12], x2
+        urhadd          v3.16b,  v3.16b,  v7.16b
  .endif
        subs            w3,  w3,  #4
-        st1             {v0.16B},  [x0], x2
-        st1             {v1.16B},  [x0], x2
-        st1             {v2.16B},  [x0], x2
-        st1             {v3.16B},  [x0], x2
+        st1             {v0.16b},  [x0], x2
+        st1             {v1.16b},  [x0], x2
+        st1             {v2.16b},  [x0], x2
+        st1             {v3.16b},  [x0], x2
        b.ne            1b
        ret
 .endm

 .macro  pixels16_x2     rnd=1, avg=0
-1:      ld1             {v0.16B, v1.16B}, [x1], x2
-        ld1             {v2.16B, v3.16B}, [x1], x2
+1:      ld1             {v0.16b, v1.16b}, [x1], x2
+        ld1             {v2.16b, v3.16b}, [x1], x2
        subs            w3,  w3,  #2
-        ext             v1.16B,  v0.16B,  v1.16B,  #1
-        avg             v0.16B,  v0.16B,  v1.16B
-        ext             v3.16B,  v2.16B,  v3.16B,  #1
-        avg             v2.16B,  v2.16B,  v3.16B
+        ext             v1.16b,  v0.16b,  v1.16b,  #1
+        avg             v0.16b,  v0.16b,  v1.16b
+        ext             v3.16b,  v2.16b,  v3.16b,  #1
+        avg             v2.16b,  v2.16b,  v3.16b
  .if \avg
-        ld1             {v1.16B}, [x0], x2
-        ld1             {v3.16B}, [x0]
-        urhadd          v0.16B,  v0.16B,  v1.16B
-        urhadd          v2.16B,  v2.16B,  v3.16B
+        ld1             {v1.16b}, [x0], x2
+        ld1             {v3.16b}, [x0]
+        urhadd          v0.16b,  v0.16b,  v1.16b
+        urhadd          v2.16b,  v2.16b,  v3.16b
        sub             x0,  x0,  x2
  .endif
-        st1             {v0.16B}, [x0], x2
-        st1             {v2.16B}, [x0], x2
+        st1             {v0.16b}, [x0], x2
+        st1             {v2.16b}, [x0], x2
        b.ne            1b
        ret
 .endm

 .macro  pixels16_y2     rnd=1, avg=0
        sub             w3,  w3,  #2
-        ld1             {v0.16B}, [x1], x2
-        ld1             {v1.16B}, [x1], x2
+        ld1             {v0.16b}, [x1], x2
+        ld1             {v1.16b}, [x1], x2
 1:      subs            w3,  w3,  #2
-        avg             v2.16B,  v0.16B,  v1.16B
-        ld1             {v0.16B}, [x1], x2
-        avg             v3.16B,  v0.16B,  v1.16B
-        ld1             {v1.16B}, [x1], x2
+        avg             v2.16b,  v0.16b,  v1.16b
+        ld1             {v0.16b}, [x1], x2
+        avg             v3.16b,  v0.16b,  v1.16b
+        ld1             {v1.16b}, [x1], x2
  .if \avg
-        ld1             {v4.16B}, [x0], x2
-        ld1             {v5.16B}, [x0]
-        urhadd          v2.16B,  v2.16B,  v4.16B
-        urhadd          v3.16B,  v3.16B,  v5.16B
+        ld1             {v4.16b}, [x0], x2
+        ld1             {v5.16b}, [x0]
+        urhadd          v2.16b,  v2.16b,  v4.16b
+        urhadd          v3.16b,  v3.16b,  v5.16b
        sub             x0,  x0,  x2
  .endif
-        st1             {v2.16B}, [x0], x2
-        st1             {v3.16B}, [x0], x2
+        st1             {v2.16b}, [x0], x2
+        st1             {v3.16b}, [x0], x2
        b.ne            1b

-        avg             v2.16B,  v0.16B,  v1.16B
-        ld1             {v0.16B}, [x1], x2
-        avg             v3.16B,  v0.16B,  v1.16B
+        avg             v2.16b,  v0.16b,  v1.16b
+        ld1             {v0.16b}, [x1], x2
+        avg             v3.16b,  v0.16b,  v1.16b
  .if \avg
-        ld1             {v4.16B}, [x0], x2
-        ld1             {v5.16B}, [x0]
-        urhadd          v2.16B,  v2.16B,  v4.16B
-        urhadd          v3.16B,  v3.16B,  v5.16B
+        ld1             {v4.16b}, [x0], x2
+        ld1             {v5.16b}, [x0]
+        urhadd          v2.16b,  v2.16b,  v4.16b
+        urhadd          v3.16b,  v3.16b,  v5.16b
        sub             x0,  x0,  x2
  .endif
-        st1             {v2.16B},     [x0], x2
-        st1             {v3.16B},     [x0], x2
+        st1             {v2.16b},     [x0], x2
+        st1             {v3.16b},     [x0], x2

        ret
 .endm

 .macro  pixels16_xy2    rnd=1, avg=0
        sub             w3,  w3,  #2
-        ld1             {v0.16B, v1.16B}, [x1], x2
-        ld1             {v4.16B, v5.16B}, [x1], x2
+        ld1             {v0.16b, v1.16b}, [x1], x2
+        ld1             {v4.16b, v5.16b}, [x1], x2
 NRND    movi            v26.8H, #1
-        ext             v1.16B,  v0.16B,  v1.16B,  #1
-        ext             v5.16B,  v4.16B,  v5.16B,  #1
-        uaddl           v16.8H,  v0.8B,   v1.8B
-        uaddl2          v20.8H,  v0.16B,  v1.16B
-        uaddl           v18.8H,  v4.8B,   v5.8B
-        uaddl2          v22.8H,  v4.16B,  v5.16B
+        ext             v1.16b,  v0.16b,  v1.16b,  #1
+        ext             v5.16b,  v4.16b,  v5.16b,  #1
+        uaddl           v16.8h,  v0.8b,   v1.8b
+        uaddl2          v20.8h,  v0.16b,  v1.16b
+        uaddl           v18.8h,  v4.8b,   v5.8b
+        uaddl2          v22.8h,  v4.16b,  v5.16b
 1:      subs            w3,  w3,  #2
-        ld1             {v0.16B, v1.16B}, [x1], x2
-        add             v24.8H,  v16.8H,  v18.8H
+        ld1             {v0.16b, v1.16b}, [x1], x2
+        add             v24.8h,  v16.8h,  v18.8h
 NRND    add             v24.8H,  v24.8H,  v26.8H
-        ext             v30.16B, v0.16B,  v1.16B,  #1
-        add             v1.8H,   v20.8H,  v22.8H
-        mshrn           v28.8B,  v24.8H,  #2
+        ext             v30.16b, v0.16b,  v1.16b,  #1
+        add             v1.8h,   v20.8h,  v22.8h
+        mshrn           v28.8b,  v24.8h,  #2
 NRND    add             v1.8H,   v1.8H,   v26.8H
-        mshrn2          v28.16B, v1.8H,   #2
+        mshrn2          v28.16b, v1.8h,   #2
  .if \avg
-        ld1             {v16.16B},        [x0]
-        urhadd          v28.16B, v28.16B, v16.16B
+        ld1             {v16.16b},        [x0]
+        urhadd          v28.16b, v28.16b, v16.16b
  .endif
-        uaddl           v16.8H,  v0.8B,   v30.8B
-        ld1             {v2.16B, v3.16B}, [x1], x2
-        uaddl2          v20.8H,  v0.16B,  v30.16B
-        st1             {v28.16B},        [x0], x2
-        add             v24.8H,  v16.8H,  v18.8H
+        uaddl           v16.8h,  v0.8b,   v30.8b
+        ld1             {v2.16b, v3.16b}, [x1], x2
+        uaddl2          v20.8h,  v0.16b,  v30.16b
+        st1             {v28.16b},        [x0], x2
+        add             v24.8h,  v16.8h,  v18.8h
 NRND    add             v24.8H,  v24.8H,  v26.8H
-        ext             v3.16B,  v2.16B,  v3.16B,  #1
-        add             v0.8H,   v20.8H,  v22.8H
-        mshrn           v30.8B,  v24.8H,  #2
+        ext             v3.16b,  v2.16b,  v3.16b,  #1
+        add             v0.8h,   v20.8h,  v22.8h
+        mshrn           v30.8b,  v24.8h,  #2
 NRND    add             v0.8H,   v0.8H,   v26.8H
-        mshrn2          v30.16B, v0.8H,   #2
+        mshrn2          v30.16b, v0.8h,   #2
  .if \avg
-        ld1             {v18.16B},        [x0]
-        urhadd          v30.16B, v30.16B, v18.16B
+        ld1             {v18.16b},        [x0]
+        urhadd          v30.16b, v30.16b, v18.16b
  .endif
-        uaddl           v18.8H,   v2.8B,  v3.8B
-        uaddl2          v22.8H,   v2.16B, v3.16B
-        st1             {v30.16B},        [x0], x2
+        uaddl           v18.8h,   v2.8b,  v3.8b
+        uaddl2          v22.8h,   v2.16b, v3.16b
+        st1             {v30.16b},        [x0], x2
        b.gt            1b

-        ld1             {v0.16B, v1.16B}, [x1], x2
-        add             v24.8H,  v16.8H,  v18.8H
+        ld1             {v0.16b, v1.16b}, [x1], x2
+        add             v24.8h,  v16.8h,  v18.8h
 NRND    add             v24.8H,  v24.8H,  v26.8H
-        ext             v30.16B, v0.16B,  v1.16B,  #1
-        add             v1.8H,   v20.8H,  v22.8H
-        mshrn           v28.8B,  v24.8H,  #2
+        ext             v30.16b, v0.16b,  v1.16b,  #1
+        add             v1.8h,   v20.8h,  v22.8h
+        mshrn           v28.8b,  v24.8h,  #2
 NRND    add             v1.8H,   v1.8H,   v26.8H
-        mshrn2          v28.16B, v1.8H,   #2
+        mshrn2          v28.16b, v1.8h,   #2
  .if \avg
-        ld1             {v16.16B},        [x0]
-        urhadd          v28.16B, v28.16B, v16.16B
+        ld1             {v16.16b},        [x0]
+        urhadd          v28.16b, v28.16b, v16.16b
  .endif
-        uaddl           v16.8H,  v0.8B,   v30.8B
-        uaddl2          v20.8H,  v0.16B,  v30.16B
-        st1             {v28.16B},        [x0], x2
-        add             v24.8H,  v16.8H,  v18.8H
+        uaddl           v16.8h,  v0.8b,   v30.8b
+        uaddl2          v20.8h,  v0.16b,  v30.16b
+        st1             {v28.16b},        [x0], x2
+        add             v24.8h,  v16.8h,  v18.8h
 NRND    add             v24.8H,  v24.8H,  v26.8H
-        add             v0.8H,   v20.8H,  v22.8H
-        mshrn           v30.8B,  v24.8H,  #2
+        add             v0.8h,   v20.8h,  v22.8h
+        mshrn           v30.8b,  v24.8h,  #2
 NRND    add             v0.8H,   v0.8H,   v26.8H
-        mshrn2          v30.16B, v0.8H,   #2
+        mshrn2          v30.16b, v0.8h,   #2
  .if \avg
-        ld1             {v18.16B},        [x0]
-        urhadd          v30.16B, v30.16B, v18.16B
+        ld1             {v18.16b},        [x0]
+        urhadd          v30.16b, v30.16b, v18.16b
  .endif
-        st1             {v30.16B},        [x0], x2
+        st1             {v30.16b},        [x0], x2

        ret
 .endm

 .macro  pixels8         rnd=1, avg=0
-1:      ld1             {v0.8B}, [x1], x2
-        ld1             {v1.8B}, [x1], x2
-        ld1             {v2.8B}, [x1], x2
-        ld1             {v3.8B}, [x1], x2
+1:      ld1             {v0.8b}, [x1], x2
+        ld1             {v1.8b}, [x1], x2
+        ld1             {v2.8b}, [x1], x2
+        ld1             {v3.8b}, [x1], x2
  .if \avg
-        ld1             {v4.8B}, [x0], x2
-        urhadd          v0.8B,  v0.8B,  v4.8B
-        ld1             {v5.8B}, [x0], x2
-        urhadd          v1.8B,  v1.8B,  v5.8B
-        ld1             {v6.8B}, [x0], x2
-        urhadd          v2.8B,  v2.8B,  v6.8B
-        ld1             {v7.8B}, [x0], x2
-        urhadd          v3.8B,  v3.8B,  v7.8B
+        ld1             {v4.8b}, [x0], x2
+        urhadd          v0.8b,  v0.8b,  v4.8b
+        ld1             {v5.8b}, [x0], x2
+        urhadd          v1.8b,  v1.8b,  v5.8b
+        ld1             {v6.8b}, [x0], x2
+        urhadd          v2.8b,  v2.8b,  v6.8b
+        ld1             {v7.8b}, [x0], x2
+        urhadd          v3.8b,  v3.8b,  v7.8b
        sub             x0,  x0,  x2,  lsl #2
  .endif
        subs            w3,  w3,  #4
-        st1             {v0.8B}, [x0], x2
-        st1             {v1.8B}, [x0], x2
-        st1             {v2.8B}, [x0], x2
-        st1             {v3.8B}, [x0], x2
+        st1             {v0.8b}, [x0], x2
+        st1             {v1.8b}, [x0], x2
+        st1             {v2.8b}, [x0], x2
+        st1             {v3.8b}, [x0], x2
        b.ne            1b
        ret
 .endm

 .macro  pixels8_x2      rnd=1, avg=0
-1:      ld1             {v0.8B, v1.8B}, [x1], x2
-        ext             v1.8B,  v0.8B,  v1.8B,  #1
-        ld1             {v2.8B, v3.8B}, [x1], x2
-        ext             v3.8B,  v2.8B,  v3.8B,  #1
+1:      ld1             {v0.8b, v1.8b}, [x1], x2
+        ext             v1.8b,  v0.8b,  v1.8b,  #1
+        ld1             {v2.8b, v3.8b}, [x1], x2
+        ext             v3.8b,  v2.8b,  v3.8b,  #1
        subs            w3,  w3,  #2
-        avg             v0.8B,   v0.8B,   v1.8B
-        avg             v2.8B,   v2.8B,   v3.8B
+        avg             v0.8b,   v0.8b,   v1.8b
+        avg             v2.8b,   v2.8b,   v3.8b
  .if \avg
-        ld1             {v4.8B},     [x0], x2
-        ld1             {v5.8B},     [x0]
-        urhadd          v0.8B,   v0.8B,   v4.8B
-        urhadd          v2.8B,   v2.8B,   v5.8B
+        ld1             {v4.8b},     [x0], x2
+        ld1             {v5.8b},     [x0]
+        urhadd          v0.8b,   v0.8b,   v4.8b
+        urhadd          v2.8b,   v2.8b,   v5.8b
        sub             x0,  x0,  x2
  .endif
-        st1             {v0.8B}, [x0], x2
-        st1             {v2.8B}, [x0], x2
+        st1             {v0.8b}, [x0], x2
+        st1             {v2.8b}, [x0], x2
        b.ne            1b
        ret
 .endm

 .macro  pixels8_y2      rnd=1, avg=0
        sub             w3,  w3,  #2
-        ld1             {v0.8B},  [x1], x2
-        ld1             {v1.8B},  [x1], x2
+        ld1             {v0.8b},  [x1], x2
+        ld1             {v1.8b},  [x1], x2
 1:      subs            w3,  w3,  #2
-        avg             v4.8B,  v0.8B,  v1.8B
-        ld1             {v0.8B},  [x1], x2
-        avg             v5.8B,  v0.8B,  v1.8B
-        ld1             {v1.8B},  [x1], x2
+        avg             v4.8b,  v0.8b,  v1.8b
+        ld1             {v0.8b},  [x1], x2
+        avg             v5.8b,  v0.8b,  v1.8b
+        ld1             {v1.8b},  [x1], x2
  .if \avg
-        ld1             {v2.8B},     [x0], x2
-        ld1             {v3.8B},     [x0]
-        urhadd          v4.8B,  v4.8B,  v2.8B
-        urhadd          v5.8B,  v5.8B,  v3.8B
+        ld1             {v2.8b},     [x0], x2
+        ld1             {v3.8b},     [x0]
+        urhadd          v4.8b,  v4.8b,  v2.8b
+        urhadd          v5.8b,  v5.8b,  v3.8b
        sub             x0,  x0,  x2
  .endif
-        st1             {v4.8B},     [x0], x2
-        st1             {v5.8B},     [x0], x2
+        st1             {v4.8b},     [x0], x2
+        st1             {v5.8b},     [x0], x2
        b.ne            1b

-        avg             v4.8B,  v0.8B,  v1.8B
-        ld1             {v0.8B},  [x1], x2
-        avg             v5.8B,  v0.8B,  v1.8B
+        avg             v4.8b,  v0.8b,  v1.8b
+        ld1             {v0.8b},  [x1], x2
+        avg             v5.8b,  v0.8b,  v1.8b
  .if \avg
-        ld1             {v2.8B},     [x0], x2
-        ld1             {v3.8B},     [x0]
-        urhadd          v4.8B,  v4.8B,  v2.8B
-        urhadd          v5.8B,  v5.8B,  v3.8B
+        ld1             {v2.8b},     [x0], x2
+        ld1             {v3.8b},     [x0]
+        urhadd          v4.8b,  v4.8b,  v2.8b
+        urhadd          v5.8b,  v5.8b,  v3.8b
        sub             x0,  x0,  x2
  .endif
-        st1             {v4.8B},     [x0], x2
-        st1             {v5.8B},     [x0], x2
+        st1             {v4.8b},     [x0], x2
+        st1             {v5.8b},     [x0], x2

        ret
 .endm

 .macro  pixels8_xy2     rnd=1, avg=0
        sub             w3,  w3,  #2
-        ld1             {v0.16B},     [x1], x2
-        ld1             {v1.16B},     [x1], x2
+        ld1             {v0.16b},     [x1], x2
+        ld1             {v1.16b},     [x1], x2
 NRND    movi            v19.8H, #1
-        ext             v4.16B,  v0.16B,  v4.16B,  #1
-        ext             v6.16B,  v1.16B,  v6.16B,  #1
-        uaddl           v16.8H,  v0.8B,  v4.8B
-        uaddl           v17.8H,  v1.8B,  v6.8B
+        ext             v4.16b,  v0.16b,  v4.16b,  #1
+        ext             v6.16b,  v1.16b,  v6.16b,  #1
+        uaddl           v16.8h,  v0.8b,  v4.8b
+        uaddl           v17.8h,  v1.8b,  v6.8b
 1:      subs            w3,  w3,  #2
-        ld1             {v0.16B},     [x1], x2
-        add             v18.8H, v16.8H,  v17.8H
-        ext             v4.16B,  v0.16B,  v4.16B,  #1
+        ld1             {v0.16b},     [x1], x2
+        add             v18.8h, v16.8h,  v17.8h
+        ext             v4.16b,  v0.16b,  v4.16b,  #1
 NRND    add             v18.8H, v18.8H, v19.8H
-        uaddl           v16.8H,  v0.8B,  v4.8B
-        mshrn           v5.8B,  v18.8H, #2
-        ld1             {v1.16B},     [x1], x2
-        add             v18.8H, v16.8H,  v17.8H
+        uaddl           v16.8h,  v0.8b,  v4.8b
+        mshrn           v5.8b,  v18.8h, #2
+        ld1             {v1.16b},     [x1], x2
+        add             v18.8h, v16.8h,  v17.8h
  .if \avg
-        ld1             {v7.8B},     [x0]
-        urhadd          v5.8B,  v5.8B,  v7.8B
+        ld1             {v7.8b},     [x0]
+        urhadd          v5.8b,  v5.8b,  v7.8b
  .endif
 NRND    add             v18.8H, v18.8H, v19.8H
-        st1             {v5.8B},     [x0], x2
-        mshrn           v7.8B,  v18.8H, #2
+        st1             {v5.8b},     [x0], x2
+        mshrn           v7.8b,  v18.8h, #2
  .if \avg
-        ld1             {v5.8B},     [x0]
-        urhadd          v7.8B,  v7.8B,  v5.8B
+        ld1             {v5.8b},     [x0]
+        urhadd          v7.8b,  v7.8b,  v5.8b
  .endif
-        ext             v6.16B,  v1.16B,  v6.16B,  #1
-        uaddl           v17.8H,  v1.8B,   v6.8B
-        st1             {v7.8B},     [x0], x2
+        ext             v6.16b,  v1.16b,  v6.16b,  #1
+        uaddl           v17.8h,  v1.8b,   v6.8b
+        st1             {v7.8b},     [x0], x2
        b.gt            1b

-        ld1             {v0.16B},     [x1], x2
-        add             v18.8H, v16.8H, v17.8H
-        ext             v4.16B, v0.16B, v4.16B,  #1
+        ld1             {v0.16b},     [x1], x2
+        add             v18.8h, v16.8h, v17.8h
+        ext             v4.16b, v0.16b, v4.16b,  #1
 NRND    add             v18.8H, v18.8H, v19.8H
-        uaddl           v16.8H,  v0.8B, v4.8B
-        mshrn           v5.8B,  v18.8H, #2
-        add             v18.8H, v16.8H, v17.8H
+        uaddl           v16.8h,  v0.8b, v4.8b
+        mshrn           v5.8b,  v18.8h, #2
+        add             v18.8h, v16.8h, v17.8h
  .if \avg
-        ld1             {v7.8B},     [x0]
-        urhadd          v5.8B,  v5.8B,  v7.8B
+        ld1             {v7.8b},     [x0]
+        urhadd          v5.8b,  v5.8b,  v7.8b
  .endif
 NRND    add             v18.8H, v18.8H, v19.8H
-        st1             {v5.8B},     [x0], x2
-        mshrn           v7.8B,  v18.8H, #2
+        st1             {v5.8b},     [x0], x2
+        mshrn           v7.8b,  v18.8h, #2
  .if \avg
-        ld1             {v5.8B},     [x0]
-        urhadd          v7.8B,  v7.8B,  v5.8B
+        ld1             {v5.8b},     [x0]
+        urhadd          v7.8b,  v7.8b,  v5.8b
  .endif
-        st1             {v7.8B},     [x0], x2
+        st1             {v7.8b},     [x0], x2

        ret
 .endm
@@ -19,6 +19,7 @@
 #ifndef AVCODEC_AARCH64_IDCT_H
 #define AVCODEC_AARCH64_IDCT_H

+#include <stddef.h>
 #include <stdint.h>

 void ff_simple_idct_neon(int16_t *data);
@@ -17,133 +17,133 @@
 */

 .macro  transpose_8x8B  r0, r1, r2, r3, r4, r5, r6, r7, r8, r9
-        trn1            \r8\().8B,  \r0\().8B,  \r1\().8B
-        trn2            \r9\().8B,  \r0\().8B,  \r1\().8B
-        trn1            \r1\().8B,  \r2\().8B,  \r3\().8B
-        trn2            \r3\().8B,  \r2\().8B,  \r3\().8B
-        trn1            \r0\().8B,  \r4\().8B,  \r5\().8B
-        trn2            \r5\().8B,  \r4\().8B,  \r5\().8B
-        trn1            \r2\().8B,  \r6\().8B,  \r7\().8B
-        trn2            \r7\().8B,  \r6\().8B,  \r7\().8B
+        trn1            \r8\().8b,  \r0\().8b,  \r1\().8b
+        trn2            \r9\().8b,  \r0\().8b,  \r1\().8b
+        trn1            \r1\().8b,  \r2\().8b,  \r3\().8b
+        trn2            \r3\().8b,  \r2\().8b,  \r3\().8b
+        trn1            \r0\().8b,  \r4\().8b,  \r5\().8b
+        trn2            \r5\().8b,  \r4\().8b,  \r5\().8b
+        trn1            \r2\().8b,  \r6\().8b,  \r7\().8b
+        trn2            \r7\().8b,  \r6\().8b,  \r7\().8b

-        trn1            \r4\().4H,  \r0\().4H,  \r2\().4H
-        trn2            \r2\().4H,  \r0\().4H,  \r2\().4H
-        trn1            \r6\().4H,  \r5\().4H,  \r7\().4H
-        trn2            \r7\().4H,  \r5\().4H,  \r7\().4H
-        trn1            \r5\().4H,  \r9\().4H,  \r3\().4H
-        trn2            \r9\().4H,  \r9\().4H,  \r3\().4H
-        trn1            \r3\().4H,  \r8\().4H,  \r1\().4H
-        trn2            \r8\().4H,  \r8\().4H,  \r1\().4H
+        trn1            \r4\().4h,  \r0\().4h,  \r2\().4h
+        trn2            \r2\().4h,  \r0\().4h,  \r2\().4h
+        trn1            \r6\().4h,  \r5\().4h,  \r7\().4h
+        trn2            \r7\().4h,  \r5\().4h,  \r7\().4h
+        trn1            \r5\().4h,  \r9\().4h,  \r3\().4h
+        trn2            \r9\().4h,  \r9\().4h,  \r3\().4h
+        trn1            \r3\().4h,  \r8\().4h,  \r1\().4h
+        trn2            \r8\().4h,  \r8\().4h,  \r1\().4h

-        trn1            \r0\().2S,  \r3\().2S,  \r4\().2S
-        trn2            \r4\().2S,  \r3\().2S,  \r4\().2S
+        trn1            \r0\().2s,  \r3\().2s,  \r4\().2s
+        trn2            \r4\().2s,  \r3\().2s,  \r4\().2s

-        trn1            \r1\().2S,  \r5\().2S,  \r6\().2S
-        trn2            \r5\().2S,  \r5\().2S,  \r6\().2S
+        trn1            \r1\().2s,  \r5\().2s,  \r6\().2s
+        trn2            \r5\().2s,  \r5\().2s,  \r6\().2s

-        trn2            \r6\().2S,  \r8\().2S,  \r2\().2S
-        trn1            \r2\().2S,  \r8\().2S,  \r2\().2S
+        trn2            \r6\().2s,  \r8\().2s,  \r2\().2s
+        trn1            \r2\().2s,  \r8\().2s,  \r2\().2s

-        trn1            \r3\().2S,  \r9\().2S,  \r7\().2S
-        trn2            \r7\().2S,  \r9\().2S,  \r7\().2S
+        trn1            \r3\().2s,  \r9\().2s,  \r7\().2s
+        trn2            \r7\().2s,  \r9\().2s,  \r7\().2s
 .endm

 .macro  transpose_8x16B r0, r1, r2, r3, r4, r5, r6, r7, t0, t1
-        trn1            \t0\().16B, \r0\().16B, \r1\().16B
-        trn2            \t1\().16B, \r0\().16B, \r1\().16B
-        trn1            \r1\().16B, \r2\().16B, \r3\().16B
-        trn2            \r3\().16B, \r2\().16B, \r3\().16B
-        trn1            \r0\().16B, \r4\().16B, \r5\().16B
-        trn2            \r5\().16B, \r4\().16B, \r5\().16B
-        trn1            \r2\().16B, \r6\().16B, \r7\().16B
-        trn2            \r7\().16B, \r6\().16B, \r7\().16B
+        trn1            \t0\().16b, \r0\().16b, \r1\().16b
+        trn2            \t1\().16b, \r0\().16b, \r1\().16b
+        trn1            \r1\().16b, \r2\().16b, \r3\().16b
+        trn2            \r3\().16b, \r2\().16b, \r3\().16b
+        trn1            \r0\().16b, \r4\().16b, \r5\().16b
+        trn2            \r5\().16b, \r4\().16b, \r5\().16b
+        trn1            \r2\().16b, \r6\().16b, \r7\().16b
+        trn2            \r7\().16b, \r6\().16b, \r7\().16b

-        trn1            \r4\().8H,  \r0\().8H,  \r2\().8H
-        trn2            \r2\().8H,  \r0\().8H,  \r2\().8H
-        trn1            \r6\().8H,  \r5\().8H,  \r7\().8H
-        trn2            \r7\().8H,  \r5\().8H,  \r7\().8H
-        trn1            \r5\().8H,  \t1\().8H,  \r3\().8H
-        trn2            \t1\().8H,  \t1\().8H,  \r3\().8H
-        trn1            \r3\().8H,  \t0\().8H,  \r1\().8H
-        trn2            \t0\().8H,  \t0\().8H,  \r1\().8H
+        trn1            \r4\().8h,  \r0\().8h,  \r2\().8h
+        trn2            \r2\().8h,  \r0\().8h,  \r2\().8h
+        trn1            \r6\().8h,  \r5\().8h,  \r7\().8h
+        trn2            \r7\().8h,  \r5\().8h,  \r7\().8h
+        trn1            \r5\().8h,  \t1\().8h,  \r3\().8h
+        trn2            \t1\().8h,  \t1\().8h,  \r3\().8h
+        trn1            \r3\().8h,  \t0\().8h,  \r1\().8h
+        trn2            \t0\().8h,  \t0\().8h,  \r1\().8h

-        trn1            \r0\().4S,  \r3\().4S,  \r4\().4S
-        trn2            \r4\().4S,  \r3\().4S,  \r4\().4S
+        trn1            \r0\().4s,  \r3\().4s,  \r4\().4s
+        trn2            \r4\().4s,  \r3\().4s,  \r4\().4s

-        trn1            \r1\().4S,  \r5\().4S,  \r6\().4S
-        trn2            \r5\().4S,  \r5\().4S,  \r6\().4S
+        trn1            \r1\().4s,  \r5\().4s,  \r6\().4s
+        trn2            \r5\().4s,  \r5\().4s,  \r6\().4s

-        trn2            \r6\().4S,  \t0\().4S,  \r2\().4S
-        trn1            \r2\().4S,  \t0\().4S,  \r2\().4S
+        trn2            \r6\().4s,  \t0\().4s,  \r2\().4s
+        trn1            \r2\().4s,  \t0\().4s,  \r2\().4s

-        trn1            \r3\().4S,  \t1\().4S,  \r7\().4S
-        trn2            \r7\().4S,  \t1\().4S,  \r7\().4S
+        trn1            \r3\().4s,  \t1\().4s,  \r7\().4s
+        trn2            \r7\().4s,  \t1\().4s,  \r7\().4s
 .endm

 .macro  transpose_4x16B r0, r1, r2, r3, t4, t5, t6, t7
-        trn1            \t4\().16B, \r0\().16B,  \r1\().16B
-        trn2            \t5\().16B, \r0\().16B,  \r1\().16B
-        trn1            \t6\().16B, \r2\().16B,  \r3\().16B
-        trn2            \t7\().16B, \r2\().16B,  \r3\().16B
+        trn1            \t4\().16b, \r0\().16b,  \r1\().16b
+        trn2            \t5\().16b, \r0\().16b,  \r1\().16b
+        trn1            \t6\().16b, \r2\().16b,  \r3\().16b
+        trn2            \t7\().16b, \r2\().16b,  \r3\().16b

-        trn1            \r0\().8H,  \t4\().8H,  \t6\().8H
-        trn2            \r2\().8H,  \t4\().8H,  \t6\().8H
-        trn1            \r1\().8H,  \t5\().8H,  \t7\().8H
-        trn2            \r3\().8H,  \t5\().8H,  \t7\().8H
+        trn1            \r0\().8h,  \t4\().8h,  \t6\().8h
+        trn2            \r2\().8h,  \t4\().8h,  \t6\().8h
+        trn1            \r1\().8h,  \t5\().8h,  \t7\().8h
+        trn2            \r3\().8h,  \t5\().8h,  \t7\().8h
 .endm

 .macro  transpose_4x8B  r0, r1, r2, r3, t4, t5, t6, t7
-        trn1            \t4\().8B,  \r0\().8B,  \r1\().8B
-        trn2            \t5\().8B,  \r0\().8B,  \r1\().8B
-        trn1            \t6\().8B,  \r2\().8B,  \r3\().8B
-        trn2            \t7\().8B,  \r2\().8B,  \r3\().8B
+        trn1            \t4\().8b,  \r0\().8b,  \r1\().8b
+        trn2            \t5\().8b,  \r0\().8b,  \r1\().8b
+        trn1            \t6\().8b,  \r2\().8b,  \r3\().8b
+        trn2            \t7\().8b,  \r2\().8b,  \r3\().8b

-        trn1            \r0\().4H,  \t4\().4H,  \t6\().4H
-        trn2            \r2\().4H,  \t4\().4H,  \t6\().4H
-        trn1            \r1\().4H,  \t5\().4H,  \t7\().4H
-        trn2            \r3\().4H,  \t5\().4H,  \t7\().4H
+        trn1            \r0\().4h,  \t4\().4h,  \t6\().4h
+        trn2            \r2\().4h,  \t4\().4h,  \t6\().4h
+        trn1            \r1\().4h,  \t5\().4h,  \t7\().4h
+        trn2            \r3\().4h,  \t5\().4h,  \t7\().4h
 .endm

 .macro  transpose_4x4H  r0, r1, r2, r3, r4, r5, r6, r7
-        trn1            \r4\().4H,  \r0\().4H,  \r1\().4H
-        trn2            \r5\().4H,  \r0\().4H,  \r1\().4H
-        trn1            \r6\().4H,  \r2\().4H,  \r3\().4H
-        trn2            \r7\().4H,  \r2\().4H,  \r3\().4H
-        trn1            \r0\().2S,  \r4\().2S,  \r6\().2S
-        trn2            \r2\().2S,  \r4\().2S,  \r6\().2S
-        trn1            \r1\().2S,  \r5\().2S,  \r7\().2S
-        trn2            \r3\().2S,  \r5\().2S,  \r7\().2S
+        trn1            \r4\().4h,  \r0\().4h,  \r1\().4h
+        trn2            \r5\().4h,  \r0\().4h,  \r1\().4h
+        trn1            \r6\().4h,  \r2\().4h,  \r3\().4h
+        trn2            \r7\().4h,  \r2\().4h,  \r3\().4h
+        trn1            \r0\().2s,  \r4\().2s,  \r6\().2s
+        trn2            \r2\().2s,  \r4\().2s,  \r6\().2s
+        trn1            \r1\().2s,  \r5\().2s,  \r7\().2s
+        trn2            \r3\().2s,  \r5\().2s,  \r7\().2s
 .endm

 .macro  transpose_8x8H  r0, r1, r2, r3, r4, r5, r6, r7, r8, r9
-        trn1            \r8\().8H,  \r0\().8H,  \r1\().8H
-        trn2            \r9\().8H,  \r0\().8H,  \r1\().8H
-        trn1            \r1\().8H,  \r2\().8H,  \r3\().8H
-        trn2            \r3\().8H,  \r2\().8H,  \r3\().8H
-        trn1            \r0\().8H,  \r4\().8H,  \r5\().8H
-        trn2            \r5\().8H,  \r4\().8H,  \r5\().8H
-        trn1            \r2\().8H,  \r6\().8H,  \r7\().8H
-        trn2            \r7\().8H,  \r6\().8H,  \r7\().8H
+        trn1            \r8\().8h,  \r0\().8h,  \r1\().8h
+        trn2            \r9\().8h,  \r0\().8h,  \r1\().8h
+        trn1            \r1\().8h,  \r2\().8h,  \r3\().8h
+        trn2            \r3\().8h,  \r2\().8h,  \r3\().8h
+        trn1            \r0\().8h,  \r4\().8h,  \r5\().8h
+        trn2            \r5\().8h,  \r4\().8h,  \r5\().8h
+        trn1            \r2\().8h,  \r6\().8h,  \r7\().8h
+        trn2            \r7\().8h,  \r6\().8h,  \r7\().8h

-        trn1            \r4\().4S,  \r0\().4S,  \r2\().4S
-        trn2            \r2\().4S,  \r0\().4S,  \r2\().4S
-        trn1            \r6\().4S,  \r5\().4S,  \r7\().4S
-        trn2            \r7\().4S,  \r5\().4S,  \r7\().4S
-        trn1            \r5\().4S,  \r9\().4S,  \r3\().4S
-        trn2            \r9\().4S,  \r9\().4S,  \r3\().4S
-        trn1            \r3\().4S,  \r8\().4S,  \r1\().4S
-        trn2            \r8\().4S,  \r8\().4S,  \r1\().4S
+        trn1            \r4\().4s,  \r0\().4s,  \r2\().4s
+        trn2            \r2\().4s,  \r0\().4s,  \r2\().4s
+        trn1            \r6\().4s,  \r5\().4s,  \r7\().4s
+        trn2            \r7\().4s,  \r5\().4s,  \r7\().4s
+        trn1            \r5\().4s,  \r9\().4s,  \r3\().4s
+        trn2            \r9\().4s,  \r9\().4s,  \r3\().4s
+        trn1            \r3\().4s,  \r8\().4s,  \r1\().4s
+        trn2            \r8\().4s,  \r8\().4s,  \r1\().4s

-        trn1            \r0\().2D,  \r3\().2D,  \r4\().2D
-        trn2            \r4\().2D,  \r3\().2D,  \r4\().2D
+        trn1            \r0\().2d,  \r3\().2d,  \r4\().2d
+        trn2            \r4\().2d,  \r3\().2d,  \r4\().2d

-        trn1            \r1\().2D,  \r5\().2D,  \r6\().2D
-        trn2            \r5\().2D,  \r5\().2D,  \r6\().2D
+        trn1            \r1\().2d,  \r5\().2d,  \r6\().2d
+        trn2            \r5\().2d,  \r5\().2d,  \r6\().2d

-        trn2            \r6\().2D,  \r8\().2D,  \r2\().2D
-        trn1            \r2\().2D,  \r8\().2D,  \r2\().2D
+        trn2            \r6\().2d,  \r8\().2d,  \r2\().2d
+        trn1            \r2\().2d,  \r8\().2d,  \r2\().2d

-        trn1            \r3\().2D,  \r9\().2D,  \r7\().2D
-        trn2            \r7\().2D,  \r9\().2D,  \r7\().2D
+        trn1            \r3\().2d,  \r9\().2d,  \r7\().2d
+        trn2            \r7\().2d,  \r9\().2d,  \r7\().2d

 .endm
@@ -33,81 +33,81 @@ const tab_x2, align=4
 endconst

 function ff_opus_deemphasis_neon, export=1
-        movrel  x4, tab_st
-        ld1    {v4.4s}, [x4]
-        movrel  x4, tab_x0
-        ld1    {v5.4s}, [x4]
-        movrel  x4, tab_x1
-        ld1    {v6.4s}, [x4]
-        movrel  x4, tab_x2
-        ld1    {v7.4s}, [x4]
+        movrel          x4, tab_st
+        ld1             {v4.4s}, [x4]
+        movrel          x4, tab_x0
+        ld1             {v5.4s}, [x4]
+        movrel          x4, tab_x1
+        ld1             {v6.4s}, [x4]
+        movrel          x4, tab_x2
+        ld1             {v7.4s}, [x4]

-        fmul v0.4s, v4.4s, v0.s[0]
+        fmul            v0.4s, v4.4s, v0.s[0]

-1:      ld1  {v1.4s, v2.4s}, [x1], #32
+1:      ld1             {v1.4s, v2.4s}, [x1], #32

-        fmla v0.4s, v5.4s, v1.s[0]
-        fmul v3.4s, v7.4s, v2.s[2]
+        fmla            v0.4s, v5.4s, v1.s[0]
+        fmul            v3.4s, v7.4s, v2.s[2]

-        fmla v0.4s, v6.4s, v1.s[1]
-        fmla v3.4s, v6.4s, v2.s[1]
+        fmla            v0.4s, v6.4s, v1.s[1]
+        fmla            v3.4s, v6.4s, v2.s[1]

-        fmla v0.4s, v7.4s, v1.s[2]
-        fmla v3.4s, v5.4s, v2.s[0]
+        fmla            v0.4s, v7.4s, v1.s[2]
+        fmla            v3.4s, v5.4s, v2.s[0]

-        fadd v1.4s, v1.4s, v0.4s
-        fadd v2.4s, v2.4s, v3.4s
+        fadd            v1.4s, v1.4s, v0.4s
+        fadd            v2.4s, v2.4s, v3.4s

-        fmla v2.4s, v4.4s, v1.s[3]
+        fmla            v2.4s, v4.4s, v1.s[3]

-        st1  {v1.4s, v2.4s}, [x0], #32
-        fmul v0.4s, v4.4s, v2.s[3]
+        st1             {v1.4s, v2.4s}, [x0], #32
+        fmul            v0.4s, v4.4s, v2.s[3]

-        subs w2, w2, #8
-        b.gt 1b
+        subs            w2, w2, #8
+        b.gt            1b

-        mov s0, v2.s[3]
+        mov             s0, v2.s[3]

        ret
 endfunc

 function ff_opus_postfilter_neon, export=1
-        ld1 {v0.4s}, [x2]
-        dup v1.4s, v0.s[1]
-        dup v2.4s, v0.s[2]
-        dup v0.4s, v0.s[0]
+        ld1             {v0.4s}, [x2]
+        dup             v1.4s, v0.s[1]
+        dup             v2.4s, v0.s[2]
+        dup             v0.4s, v0.s[0]

-        add w1, w1, #2
-        sub x1, x0, x1, lsl #2
+        add             w1, w1, #2
+        sub             x1, x0, x1, lsl #2

-        ld1 {v3.4s}, [x1]
-        fmul v3.4s, v3.4s, v2.4s
+        ld1             {v3.4s}, [x1]
+        fmul            v3.4s, v3.4s, v2.4s

-1:      add x1, x1, #4
-        ld1 {v4.4s}, [x1]
-        add x1, x1, #4
-        ld1 {v5.4s}, [x1]
-        add x1, x1, #4
-        ld1 {v6.4s}, [x1]
-        add x1, x1, #4
-        ld1 {v7.4s}, [x1]
+1:      add             x1, x1, #4
+        ld1             {v4.4s}, [x1]
+        add             x1, x1, #4
+        ld1             {v5.4s}, [x1]
+        add             x1, x1, #4
+        ld1             {v6.4s}, [x1]
+        add             x1, x1, #4
+        ld1             {v7.4s}, [x1]

-        fmla v3.4s, v7.4s, v2.4s
-        fadd v6.4s, v6.4s, v4.4s
+        fmla            v3.4s, v7.4s, v2.4s
+        fadd            v6.4s, v6.4s, v4.4s

-        ld1 {v4.4s}, [x0]
-        fmla v4.4s, v5.4s, v0.4s
+        ld1             {v4.4s}, [x0]
+        fmla            v4.4s, v5.4s, v0.4s

-        fmul v6.4s, v6.4s, v1.4s
-        fadd v6.4s, v6.4s, v3.4s
+        fmul            v6.4s, v6.4s, v1.4s
+        fadd            v6.4s, v6.4s, v3.4s

-        fadd v4.4s, v4.4s, v6.4s
-        fmul v3.4s, v7.4s, v2.4s
+        fadd            v4.4s, v4.4s, v6.4s
+        fmul            v3.4s, v7.4s, v2.4s

-        st1  {v4.4s}, [x0], #16
+        st1             {v4.4s}, [x0], #16

-        subs w3, w3, #4
-        b.gt 1b
+        subs            w3, w3, #4
+        b.gt            1b

        ret
 endfunc
@@ -46,49 +46,49 @@ function ff_sbr_sum64x5_neon, export=1
        add             x3, x0, #192*4
        add             x4, x0, #256*4
        mov             x5, #64
-1:      ld1             {v0.4S}, [x0]
-        ld1             {v1.4S}, [x1], #16
-        fadd            v0.4S, v0.4S, v1.4S
-        ld1             {v2.4S}, [x2], #16
-        fadd            v0.4S, v0.4S, v2.4S
-        ld1             {v3.4S}, [x3], #16
-        fadd            v0.4S, v0.4S, v3.4S
-        ld1             {v4.4S}, [x4], #16
-        fadd            v0.4S, v0.4S, v4.4S
-        st1             {v0.4S}, [x0], #16
+1:      ld1             {v0.4s}, [x0]
+        ld1             {v1.4s}, [x1], #16
+        fadd            v0.4s, v0.4s, v1.4s
+        ld1             {v2.4s}, [x2], #16
+        fadd            v0.4s, v0.4s, v2.4s
+        ld1             {v3.4s}, [x3], #16
+        fadd            v0.4s, v0.4s, v3.4s
+        ld1             {v4.4s}, [x4], #16
+        fadd            v0.4s, v0.4s, v4.4s
+        st1             {v0.4s}, [x0], #16
        subs            x5, x5, #4
        b.gt            1b
        ret
 endfunc

 function ff_sbr_sum_square_neon, export=1
-        movi            v0.4S, #0
-1:      ld1             {v1.4S}, [x0], #16
-        fmla            v0.4S, v1.4S, v1.4S
+        movi            v0.4s, #0
+1:      ld1             {v1.4s}, [x0], #16
+        fmla            v0.4s, v1.4s, v1.4s
        subs            w1, w1, #2
        b.gt            1b
-        faddp           v0.4S, v0.4S, v0.4S
-        faddp           v0.4S, v0.4S, v0.4S
+        faddp           v0.4s, v0.4s, v0.4s
+        faddp           v0.4s, v0.4s, v0.4s
        ret
 endfunc

 function ff_sbr_neg_odd_64_neon, export=1
        mov             x1, x0
-        movi            v5.4S, #1<<7, lsl #24
-        ld2             {v0.4S, v1.4S}, [x0], #32
-        eor             v1.16B, v1.16B, v5.16B
-        ld2             {v2.4S, v3.4S}, [x0], #32
+        movi            v5.4s, #1<<7, lsl #24
+        ld2             {v0.4s, v1.4s}, [x0], #32
+        eor             v1.16b, v1.16b, v5.16b
+        ld2             {v2.4s, v3.4s}, [x0], #32
 .rept 3
-        st2             {v0.4S, v1.4S}, [x1], #32
-        eor             v3.16B, v3.16B, v5.16B
-        ld2             {v0.4S, v1.4S}, [x0], #32
-        st2             {v2.4S, v3.4S}, [x1], #32
-        eor             v1.16B, v1.16B, v5.16B
-        ld2             {v2.4S, v3.4S}, [x0], #32
+        st2             {v0.4s, v1.4s}, [x1], #32
+        eor             v3.16b, v3.16b, v5.16b
+        ld2             {v0.4s, v1.4s}, [x0], #32
+        st2             {v2.4s, v3.4s}, [x1], #32
+        eor             v1.16b, v1.16b, v5.16b
+        ld2             {v2.4s, v3.4s}, [x0], #32
 .endr
-        eor             v3.16B, v3.16B, v5.16B
-        st2             {v0.4S, v1.4S}, [x1], #32
-        st2             {v2.4S, v3.4S}, [x1], #32
+        eor             v3.16b, v3.16b, v5.16b
+        st2             {v0.4s, v1.4s}, [x1], #32
+        st2             {v2.4s, v3.4s}, [x1], #32
        ret
 endfunc

@@ -97,26 +97,26 @@ function ff_sbr_qmf_pre_shuffle_neon, export=1
        add             x2, x0, #64*4
        mov             x3, #-16
        mov             x4, #-4
-        movi            v6.4S, #1<<7, lsl #24
-        ld1             {v0.2S}, [x0], #8
-        st1             {v0.2S}, [x2], #8
+        movi            v6.4s, #1<<7, lsl #24
+        ld1             {v0.2s}, [x0], #8
+        st1             {v0.2s}, [x2], #8
 .rept 7
-        ld1             {v1.4S}, [x1], x3
-        ld1             {v2.4S}, [x0], #16
-        eor             v1.16B, v1.16B, v6.16B
-        rev64           v1.4S, v1.4S
-        ext             v1.16B, v1.16B, v1.16B, #8
-        st2             {v1.4S, v2.4S}, [x2], #32
+        ld1             {v1.4s}, [x1], x3
+        ld1             {v2.4s}, [x0], #16
+        eor             v1.16b, v1.16b, v6.16b
+        rev64           v1.4s, v1.4s
+        ext             v1.16b, v1.16b, v1.16b, #8
+        st2             {v1.4s, v2.4s}, [x2], #32
 .endr
        add             x1, x1, #8
-        ld1             {v1.2S}, [x1], x4
-        ld1             {v2.2S}, [x0], #8
-        ld1             {v1.S}[3], [x1]
-        ld1             {v2.S}[2], [x0]
-        eor             v1.16B, v1.16B, v6.16B
-        rev64           v1.4S, v1.4S
-        st2             {v1.2S, v2.2S}, [x2], #16
-        st2             {v1.S, v2.S}[2], [x2]
+        ld1             {v1.2s}, [x1], x4
+        ld1             {v2.2s}, [x0], #8
+        ld1             {v1.s}[3], [x1]
+        ld1             {v2.s}[2], [x0]
+        eor             v1.16b, v1.16b, v6.16b
+        rev64           v1.4s, v1.4s
+        st2             {v1.2s, v2.2s}, [x2], #16
+        st2             {v1.s, v2.s}[2], [x2]
        ret
 endfunc

@@ -124,13 +124,13 @@ function ff_sbr_qmf_post_shuffle_neon, export=1
        add             x2, x1, #60*4
        mov             x3, #-16
        mov             x4, #32
-        movi            v6.4S, #1<<7, lsl #24
-1:      ld1             {v0.4S}, [x2], x3
-        ld1             {v1.4S}, [x1], #16
-        eor             v0.16B, v0.16B, v6.16B
-        rev64           v0.4S, v0.4S
-        ext             v0.16B, v0.16B, v0.16B, #8
-        st2             {v0.4S, v1.4S}, [x0], #32
+        movi            v6.4s, #1<<7, lsl #24
+1:      ld1             {v0.4s}, [x2], x3
+        ld1             {v1.4s}, [x1], #16
+        eor             v0.16b, v0.16b, v6.16b
+        rev64           v0.4s, v0.4s
+        ext             v0.16b, v0.16b, v0.16b, #8
+        st2             {v0.4s, v1.4s}, [x0], #32
        subs            x4, x4, #4
        b.gt            1b
        ret
@@ -141,13 +141,13 @@ function ff_sbr_qmf_deint_neg_neon, export=1
        add             x2, x0, #60*4
        mov             x3, #-32
        mov             x4, #32
-        movi            v2.4S, #1<<7, lsl #24
-1:      ld2             {v0.4S, v1.4S}, [x1], x3
-        eor             v0.16B, v0.16B, v2.16B
-        rev64           v1.4S, v1.4S
-        ext             v1.16B, v1.16B, v1.16B, #8
-        st1             {v0.4S}, [x2]
-        st1             {v1.4S}, [x0], #16
+        movi            v2.4s, #1<<7, lsl #24
+1:      ld2             {v0.4s, v1.4s}, [x1], x3
+        eor             v0.16b, v0.16b, v2.16b
+        rev64           v1.4s, v1.4s
+        ext             v1.16b, v1.16b, v1.16b, #8
+        st1             {v0.4s}, [x2]
+        st1             {v1.4s}, [x0], #16
        sub             x2, x2, #16
        subs            x4, x4, #4
        b.gt            1b
@@ -159,16 +159,16 @@ function ff_sbr_qmf_deint_bfly_neon, export=1
        add             x3, x0, #124*4
        mov             x4, #64
        mov             x5, #-16
-1:      ld1             {v0.4S}, [x1], #16
-        ld1             {v1.4S}, [x2], x5
-        rev64           v2.4S, v0.4S
-        ext             v2.16B, v2.16B, v2.16B, #8
-        rev64           v3.4S, v1.4S
-        ext             v3.16B, v3.16B, v3.16B, #8
-        fadd            v1.4S, v1.4S, v2.4S
-        fsub            v0.4S, v0.4S, v3.4S
-        st1             {v0.4S}, [x0], #16
-        st1             {v1.4S}, [x3], x5
+1:      ld1             {v0.4s}, [x1], #16
+        ld1             {v1.4s}, [x2], x5
+        rev64           v2.4s, v0.4s
+        ext             v2.16b, v2.16b, v2.16b, #8
+        rev64           v3.4s, v1.4s
+        ext             v3.16b, v3.16b, v3.16b, #8
+        fadd            v1.4s, v1.4s, v2.4s
+        fsub            v0.4s, v0.4s, v3.4s
+        st1             {v0.4s}, [x0], #16
+        st1             {v1.4s}, [x3], x5
        subs            x4, x4, #4
        b.gt            1b
        ret
@@ -178,32 +178,32 @@ function ff_sbr_hf_gen_neon, export=1
        sxtw            x4, w4
        sxtw            x5, w5
        movrel          x6, factors
-        ld1             {v7.4S}, [x6]
-        dup             v1.4S, v0.S[0]
-        mov             v2.8B, v1.8B
-        mov             v2.S[2], v7.S[0]
-        mov             v2.S[3], v7.S[0]
-        fmul            v1.4S, v1.4S, v2.4S
-        ld1             {v0.D}[0], [x3]
-        ld1             {v0.D}[1], [x2]
-        fmul            v0.4S, v0.4S, v1.4S
-        fmul            v1.4S, v0.4S, v7.4S
-        rev64           v0.4S, v0.4S
+        ld1             {v7.4s}, [x6]
+        dup             v1.4s, v0.s[0]
+        mov             v2.8b, v1.8b
+        mov             v2.s[2], v7.s[0]
+        mov             v2.s[3], v7.s[0]
+        fmul            v1.4s, v1.4s, v2.4s
+        ld1             {v0.d}[0], [x3]
+        ld1             {v0.d}[1], [x2]
+        fmul            v0.4s, v0.4s, v1.4s
+        fmul            v1.4s, v0.4s, v7.4s
+        rev64           v0.4s, v0.4s
        sub             x7, x5, x4
        add             x0, x0, x4, lsl #3
        add             x1, x1, x4, lsl #3
        sub             x1, x1, #16
-1:      ld1             {v2.4S}, [x1], #16
-        ld1             {v3.2S}, [x1]
-        fmul            v4.4S, v2.4S, v1.4S
-        fmul            v5.4S, v2.4S, v0.4S
-        faddp           v4.4S, v4.4S, v4.4S
-        faddp           v5.4S, v5.4S, v5.4S
-        faddp           v4.4S, v4.4S, v4.4S
-        faddp           v5.4S, v5.4S, v5.4S
-        mov             v4.S[1], v5.S[0]
-        fadd            v4.2S, v4.2S, v3.2S
-        st1             {v4.2S}, [x0], #8
+1:      ld1             {v2.4s}, [x1], #16
+        ld1             {v3.2s}, [x1]
+        fmul            v4.4s, v2.4s, v1.4s
+        fmul            v5.4s, v2.4s, v0.4s
+        faddp           v4.4s, v4.4s, v4.4s
+        faddp           v5.4s, v5.4s, v5.4s
+        faddp           v4.4s, v4.4s, v4.4s
+        faddp           v5.4s, v5.4s, v5.4s
+        mov             v4.s[1], v5.s[0]
+        fadd            v4.2s, v4.2s, v3.2s
+        st1             {v4.2s}, [x0], #8
        sub             x1, x1, #8
        subs            x7, x7, #1
        b.gt            1b
@@ -215,10 +215,10 @@ function ff_sbr_hf_g_filt_neon, export=1
        sxtw            x4, w4
        mov             x5, #40*2*4
        add             x1, x1, x4, lsl #3
-1:      ld1             {v0.2S}, [x1], x5
-        ld1             {v1.S}[0], [x2], #4
-        fmul            v2.4S, v0.4S, v1.S[0]
-        st1             {v2.2S}, [x0], #8
+1:      ld1             {v0.2s}, [x1], x5
+        ld1             {v1.s}[0], [x2], #4
+        fmul            v2.4s, v0.4s, v1.s[0]
+        st1             {v2.2s}, [x0], #8
        subs            x3, x3, #1
        b.gt            1b
        ret
@@ -227,46 +227,46 @@ endfunc
 function ff_sbr_autocorrelate_neon, export=1
        mov             x2, #38
        movrel          x3, factors
-        ld1             {v0.4S}, [x3]
-        movi            v1.4S, #0
-        movi            v2.4S, #0
-        movi            v3.4S, #0
-        ld1             {v4.2S}, [x0], #8
-        ld1             {v5.2S}, [x0], #8
-        fmul            v16.2S, v4.2S, v4.2S
-        fmul            v17.2S, v5.2S, v4.S[0]
-        fmul            v18.2S, v5.2S, v4.S[1]
-1:      ld1             {v5.D}[1], [x0], #8
-        fmla            v1.2S, v4.2S, v4.2S
-        fmla            v2.4S, v5.4S, v4.S[0]
-        fmla            v3.4S, v5.4S, v4.S[1]
-        mov             v4.D[0], v5.D[0]
-        mov             v5.D[0], v5.D[1]
+        ld1             {v0.4s}, [x3]
+        movi            v1.4s, #0
+        movi            v2.4s, #0
+        movi            v3.4s, #0
+        ld1             {v4.2s}, [x0], #8
+        ld1             {v5.2s}, [x0], #8
+        fmul            v16.2s, v4.2s, v4.2s
+        fmul            v17.2s, v5.2s, v4.s[0]
+        fmul            v18.2s, v5.2s, v4.s[1]
+1:      ld1             {v5.d}[1], [x0], #8
+        fmla            v1.2s, v4.2s, v4.2s
+        fmla            v2.4s, v5.4s, v4.s[0]
+        fmla            v3.4s, v5.4s, v4.s[1]
+        mov             v4.d[0], v5.d[0]
+        mov             v5.d[0], v5.d[1]
        subs            x2, x2, #1
        b.gt            1b
-        fmul            v19.2S, v4.2S, v4.2S
-        fmul            v20.2S, v5.2S, v4.S[0]
-        fmul            v21.2S, v5.2S, v4.S[1]
-        fadd            v22.4S, v2.4S, v20.4S
-        fsub            v22.4S, v22.4S, v17.4S
-        fadd            v23.4S, v3.4S, v21.4S
-        fsub            v23.4S, v23.4S, v18.4S
-        rev64           v23.4S, v23.4S
-        fmul            v23.4S, v23.4S, v0.4S
-        fadd            v22.4S, v22.4S, v23.4S
-        st1             {v22.4S}, [x1], #16
-        fadd            v23.2S, v1.2S, v19.2S
-        fsub            v23.2S, v23.2S, v16.2S
-        faddp           v23.2S, v23.2S, v23.2S
-        st1             {v23.S}[0], [x1]
+        fmul            v19.2s, v4.2s, v4.2s
+        fmul            v20.2s, v5.2s, v4.s[0]
+        fmul            v21.2s, v5.2s, v4.s[1]
+        fadd            v22.4s, v2.4s, v20.4s
+        fsub            v22.4s, v22.4s, v17.4s
+        fadd            v23.4s, v3.4s, v21.4s
+        fsub            v23.4s, v23.4s, v18.4s
+        rev64           v23.4s, v23.4s
+        fmul            v23.4s, v23.4s, v0.4s
+        fadd            v22.4s, v22.4s, v23.4s
+        st1             {v22.4s}, [x1], #16
+        fadd            v23.2s, v1.2s, v19.2s
+        fsub            v23.2s, v23.2s, v16.2s
+        faddp           v23.2s, v23.2s, v23.2s
+        st1             {v23.s}[0], [x1]
        add             x1, x1, #8
-        rev64           v3.2S, v3.2S
-        fmul            v3.2S, v3.2S, v0.2S
-        fadd            v2.2S, v2.2S, v3.2S
-        st1             {v2.2S}, [x1]
+        rev64           v3.2s, v3.2s
+        fmul            v3.2s, v3.2s, v0.2s
+        fadd            v2.2s, v2.2s, v3.2s
+        st1             {v2.2s}, [x1]
        add             x1, x1, #16
-        faddp           v1.2S, v1.2S, v1.2S
-        st1             {v1.S}[0], [x1]
+        faddp           v1.2s, v1.2s, v1.2s
+        st1             {v1.s}[0], [x1]
        ret
 endfunc

@@ -278,25 +278,25 @@ endfunc
 1:      and             x3, x3, #0x1ff
        add             x8, x7, x3, lsl #3
        add             x3, x3, #2
-        ld1             {v2.4S}, [x0]
-        ld1             {v3.2S}, [x1], #8
-        ld1             {v4.2S}, [x2], #8
-        ld1             {v5.4S}, [x8]
-        mov             v6.16B, v2.16B
-        zip1            v3.4S, v3.4S, v3.4S
-        zip1            v4.4S, v4.4S, v4.4S
-        fmla            v6.4S, v1.4S, v3.4S
-        fmla            v2.4S, v5.4S, v4.4S
-        fcmeq           v7.4S, v3.4S, #0
-        bif             v2.16B, v6.16B, v7.16B
-        st1             {v2.4S}, [x0], #16
+        ld1             {v2.4s}, [x0]
+        ld1             {v3.2s}, [x1], #8
+        ld1             {v4.2s}, [x2], #8
+        ld1             {v5.4s}, [x8]
+        mov             v6.16b, v2.16b
+        zip1            v3.4s, v3.4s, v3.4s
+        zip1            v4.4s, v4.4s, v4.4s
+        fmla            v6.4s, v1.4s, v3.4s
+        fmla            v2.4s, v5.4s, v4.4s
+        fcmeq           v7.4s, v3.4s, #0
+        bif             v2.16b, v6.16b, v7.16b
+        st1             {v2.4s}, [x0], #16
        subs            x5, x5, #2
        b.gt            1b
 .endm

 function ff_sbr_hf_apply_noise_0_neon, export=1
        movrel          x9, phi_noise_0
-        ld1             {v1.4S}, [x9]
+        ld1             {v1.4s}, [x9]
        apply_noise_common
        ret
 endfunc
@@ -305,14 +305,14 @@ function ff_sbr_hf_apply_noise_1_neon, export=1
        movrel          x9, phi_noise_1
        and             x4, x4, #1
        add             x9, x9, x4, lsl #4
-        ld1             {v1.4S}, [x9]
+        ld1             {v1.4s}, [x9]
        apply_noise_common
        ret
 endfunc

 function ff_sbr_hf_apply_noise_2_neon, export=1
        movrel          x9, phi_noise_2
-        ld1             {v1.4S}, [x9]
+        ld1             {v1.4s}, [x9]
        apply_noise_common
        ret
 endfunc
@@ -321,7 +321,7 @@ function ff_sbr_hf_apply_noise_3_neon, export=1
        movrel          x9, phi_noise_3
        and             x4, x4, #1
        add             x9, x9, x4, lsl #4
-        ld1             {v1.4S}, [x9]
+        ld1             {v1.4s}, [x9]
        apply_noise_common
        ret
 endfunc
@@ -54,7 +54,7 @@ endconst
        prfm            pldl1keep, [\data]
        mov             x10, x30
        movrel          x3, idct_coeff_neon
-        ld1             {v0.2D}, [x3]
+        ld1             {v0.2d}, [x3]
 .endm

 .macro idct_end
@@ -74,146 +74,146 @@ endconst
 .endm

 .macro idct_col4_top y1, y2, y3, y4, i, l
-        smull\i         v7.4S,  \y3\l, z2
-        smull\i         v16.4S, \y3\l, z6
-        smull\i         v17.4S, \y2\l, z1
-        add             v19.4S, v23.4S, v7.4S
-        smull\i         v18.4S, \y2\l, z3
-        add             v20.4S, v23.4S, v16.4S
-        smull\i         v5.4S,  \y2\l, z5
-        sub             v21.4S, v23.4S, v16.4S
-        smull\i         v6.4S,  \y2\l, z7
-        sub             v22.4S, v23.4S, v7.4S
+        smull\i         v7.4s,  \y3\l, z2
+        smull\i         v16.4s, \y3\l, z6
+        smull\i         v17.4s, \y2\l, z1
+        add             v19.4s, v23.4s, v7.4s
+        smull\i         v18.4s, \y2\l, z3
+        add             v20.4s, v23.4s, v16.4s
+        smull\i         v5.4s,  \y2\l, z5
+        sub             v21.4s, v23.4s, v16.4s
+        smull\i         v6.4s,  \y2\l, z7
+        sub             v22.4s, v23.4s, v7.4s

-        smlal\i         v17.4S, \y4\l, z3
-        smlsl\i         v18.4S, \y4\l, z7
-        smlsl\i         v5.4S,  \y4\l, z1
-        smlsl\i         v6.4S,  \y4\l, z5
+        smlal\i         v17.4s, \y4\l, z3
+        smlsl\i         v18.4s, \y4\l, z7
+        smlsl\i         v5.4s,  \y4\l, z1
+        smlsl\i         v6.4s,  \y4\l, z5
 .endm

 .macro idct_row4_neon y1, y2, y3, y4, pass
-        ld1             {\y1\().2D,\y2\().2D}, [x2], #32
-        movi            v23.4S, #1<<2, lsl #8
-        orr             v5.16B, \y1\().16B, \y2\().16B
-        ld1             {\y3\().2D,\y4\().2D}, [x2], #32
-        orr             v6.16B, \y3\().16B, \y4\().16B
-        orr             v5.16B, v5.16B, v6.16B
-        mov             x3, v5.D[1]
-        smlal           v23.4S, \y1\().4H, z4
+        ld1             {\y1\().2d,\y2\().2d}, [x2], #32
+        movi            v23.4s, #1<<2, lsl #8
+        orr             v5.16b, \y1\().16b, \y2\().16b
+        ld1             {\y3\().2d,\y4\().2d}, [x2], #32
+        orr             v6.16b, \y3\().16b, \y4\().16b
+        orr             v5.16b, v5.16b, v6.16b
+        mov             x3, v5.d[1]
+        smlal           v23.4s, \y1\().4h, z4

-        idct_col4_top   \y1, \y2, \y3, \y4, 1, .4H
+        idct_col4_top   \y1, \y2, \y3, \y4, 1, .4h

        cmp             x3, #0
        b.eq            \pass\()f

-        smull2          v7.4S, \y1\().8H, z4
-        smlal2          v17.4S, \y2\().8H, z5
-        smlsl2          v18.4S, \y2\().8H, z1
-        smull2          v16.4S, \y3\().8H, z2
-        smlal2          v5.4S, \y2\().8H, z7
-        add             v19.4S, v19.4S, v7.4S
-        sub             v20.4S, v20.4S, v7.4S
-        sub             v21.4S, v21.4S, v7.4S
-        add             v22.4S, v22.4S, v7.4S
-        smlal2          v6.4S, \y2\().8H, z3
-        smull2          v7.4S, \y3\().8H, z6
-        smlal2          v17.4S, \y4\().8H, z7
-        smlsl2          v18.4S, \y4\().8H, z5
-        smlal2          v5.4S, \y4\().8H, z3
-        smlsl2          v6.4S, \y4\().8H, z1
-        add             v19.4S, v19.4S, v7.4S
-        sub             v20.4S, v20.4S, v16.4S
-        add             v21.4S, v21.4S, v16.4S
-        sub             v22.4S, v22.4S, v7.4S
+        smull2          v7.4s, \y1\().8h, z4
+        smlal2          v17.4s, \y2\().8h, z5
+        smlsl2          v18.4s, \y2\().8h, z1
+        smull2          v16.4s, \y3\().8h, z2
+        smlal2          v5.4s, \y2\().8h, z7
+        add             v19.4s, v19.4s, v7.4s
+        sub             v20.4s, v20.4s, v7.4s
+        sub             v21.4s, v21.4s, v7.4s
+        add             v22.4s, v22.4s, v7.4s
+        smlal2          v6.4s, \y2\().8h, z3
+        smull2          v7.4s, \y3\().8h, z6
+        smlal2          v17.4s, \y4\().8h, z7
+        smlsl2          v18.4s, \y4\().8h, z5
+        smlal2          v5.4s, \y4\().8h, z3
+        smlsl2          v6.4s, \y4\().8h, z1
+        add             v19.4s, v19.4s, v7.4s
+        sub             v20.4s, v20.4s, v16.4s
+        add             v21.4s, v21.4s, v16.4s
+        sub             v22.4s, v22.4s, v7.4s

 \pass:  add             \y3\().4S, v19.4S, v17.4S
-        add             \y4\().4S, v20.4S, v18.4S
-        shrn            \y1\().4H, \y3\().4S, #ROW_SHIFT
-        shrn            \y2\().4H, \y4\().4S, #ROW_SHIFT
-        add             v7.4S, v21.4S, v5.4S
-        add             v16.4S, v22.4S, v6.4S
-        shrn            \y3\().4H, v7.4S, #ROW_SHIFT
-        shrn            \y4\().4H, v16.4S, #ROW_SHIFT
-        sub             v22.4S, v22.4S, v6.4S
-        sub             v19.4S, v19.4S, v17.4S
-        sub             v21.4S, v21.4S, v5.4S
-        shrn2           \y1\().8H, v22.4S, #ROW_SHIFT
-        sub             v20.4S, v20.4S, v18.4S
-        shrn2           \y2\().8H, v21.4S, #ROW_SHIFT
-        shrn2           \y3\().8H, v20.4S, #ROW_SHIFT
-        shrn2           \y4\().8H, v19.4S, #ROW_SHIFT
+        add             \y4\().4s, v20.4s, v18.4s
+        shrn            \y1\().4h, \y3\().4s, #ROW_SHIFT
+        shrn            \y2\().4h, \y4\().4s, #ROW_SHIFT
+        add             v7.4s, v21.4s, v5.4s
+        add             v16.4s, v22.4s, v6.4s
+        shrn            \y3\().4h, v7.4s, #ROW_SHIFT
+        shrn            \y4\().4h, v16.4s, #ROW_SHIFT
+        sub             v22.4s, v22.4s, v6.4s
+        sub             v19.4s, v19.4s, v17.4s
+        sub             v21.4s, v21.4s, v5.4s
+        shrn2           \y1\().8h, v22.4s, #ROW_SHIFT
+        sub             v20.4s, v20.4s, v18.4s
+        shrn2           \y2\().8h, v21.4s, #ROW_SHIFT
+        shrn2           \y3\().8h, v20.4s, #ROW_SHIFT
+        shrn2           \y4\().8h, v19.4s, #ROW_SHIFT

-        trn1            v16.8H, \y1\().8H, \y2\().8H
-        trn2            v17.8H, \y1\().8H, \y2\().8H
-        trn1            v18.8H, \y3\().8H, \y4\().8H
-        trn2            v19.8H, \y3\().8H, \y4\().8H
-        trn1            \y1\().4S, v16.4S, v18.4S
-        trn1            \y2\().4S, v17.4S, v19.4S
-        trn2            \y3\().4S, v16.4S, v18.4S
-        trn2            \y4\().4S, v17.4S, v19.4S
+        trn1            v16.8h, \y1\().8h, \y2\().8h
+        trn2            v17.8h, \y1\().8h, \y2\().8h
+        trn1            v18.8h, \y3\().8h, \y4\().8h
+        trn2            v19.8h, \y3\().8h, \y4\().8h
+        trn1            \y1\().4s, v16.4s, v18.4s
+        trn1            \y2\().4s, v17.4s, v19.4s
+        trn2            \y3\().4s, v16.4s, v18.4s
+        trn2            \y4\().4s, v17.4s, v19.4s
 .endm

 .macro declare_idct_col4_neon i, l
 function idct_col4_neon\i
-        dup             v23.4H, z4c
+        dup             v23.4h, z4c
 .if \i == 1
-        add             v23.4H, v23.4H, v24.4H
+        add             v23.4h, v23.4h, v24.4h
 .else
-        mov             v5.D[0], v24.D[1]
-        add             v23.4H, v23.4H, v5.4H
+        mov             v5.d[0], v24.d[1]
+        add             v23.4h, v23.4h, v5.4h
 .endif
-        smull           v23.4S, v23.4H, z4
+        smull           v23.4s, v23.4h, z4

        idct_col4_top   v24, v25, v26, v27, \i, \l

-        mov             x4, v28.D[\i - 1]
-        mov             x5, v29.D[\i - 1]
+        mov             x4, v28.d[\i - 1]
+        mov             x5, v29.d[\i - 1]
        cmp             x4, #0
        b.eq            1f

-        smull\i         v7.4S,  v28\l,  z4
-        add             v19.4S, v19.4S, v7.4S
-        sub             v20.4S, v20.4S, v7.4S
-        sub             v21.4S, v21.4S, v7.4S
-        add             v22.4S, v22.4S, v7.4S
+        smull\i         v7.4s,  v28\l,  z4
+        add             v19.4s, v19.4s, v7.4s
+        sub             v20.4s, v20.4s, v7.4s
+        sub             v21.4s, v21.4s, v7.4s
+        add             v22.4s, v22.4s, v7.4s

-1:      mov             x4, v30.D[\i - 1]
+1:      mov             x4, v30.d[\i - 1]
        cmp             x5, #0
        b.eq            2f

-        smlal\i         v17.4S, v29\l, z5
-        smlsl\i         v18.4S, v29\l, z1
-        smlal\i         v5.4S,  v29\l, z7
-        smlal\i         v6.4S,  v29\l, z3
+        smlal\i         v17.4s, v29\l, z5
+        smlsl\i         v18.4s, v29\l, z1
+        smlal\i         v5.4s,  v29\l, z7
+        smlal\i         v6.4s,  v29\l, z3

-2:      mov             x5, v31.D[\i - 1]
+2:      mov             x5, v31.d[\i - 1]
        cmp             x4, #0
        b.eq            3f

-        smull\i         v7.4S,  v30\l, z6
-        smull\i         v16.4S, v30\l, z2
-        add             v19.4S, v19.4S, v7.4S
-        sub             v22.4S, v22.4S, v7.4S
-        sub             v20.4S, v20.4S, v16.4S
-        add             v21.4S, v21.4S, v16.4S
+        smull\i         v7.4s,  v30\l, z6
+        smull\i         v16.4s, v30\l, z2
+        add             v19.4s, v19.4s, v7.4s
+        sub             v22.4s, v22.4s, v7.4s
+        sub             v20.4s, v20.4s, v16.4s
+        add             v21.4s, v21.4s, v16.4s

 3:      cmp             x5, #0
        b.eq            4f

-        smlal\i         v17.4S, v31\l, z7
-        smlsl\i         v18.4S, v31\l, z5
-        smlal\i         v5.4S,  v31\l, z3
-        smlsl\i         v6.4S,  v31\l, z1
+        smlal\i         v17.4s, v31\l, z7
+        smlsl\i         v18.4s, v31\l, z5
+        smlal\i         v5.4s,  v31\l, z3
+        smlsl\i         v6.4s,  v31\l, z1

-4:      addhn           v7.4H, v19.4S, v17.4S
-        addhn2          v7.8H, v20.4S, v18.4S
-        subhn           v18.4H, v20.4S, v18.4S
-        subhn2          v18.8H, v19.4S, v17.4S
+4:      addhn           v7.4h, v19.4s, v17.4s
+        addhn2          v7.8h, v20.4s, v18.4s
+        subhn           v18.4h, v20.4s, v18.4s
+        subhn2          v18.8h, v19.4s, v17.4s

-        addhn           v16.4H, v21.4S, v5.4S
-        addhn2          v16.8H, v22.4S, v6.4S
-        subhn           v17.4H, v22.4S, v6.4S
-        subhn2          v17.8H, v21.4S, v5.4S
+        addhn           v16.4h, v21.4s, v5.4s
+        addhn2          v16.8h, v22.4s, v6.4s
+        subhn           v17.4h, v22.4s, v6.4s
+        subhn2          v17.8h, v21.4s, v5.4s

        ret
 endfunc
@@ -229,33 +229,33 @@ function ff_simple_idct_put_neon, export=1
        idct_row4_neon  v28, v29, v30, v31, 2
        bl              idct_col4_neon1

-        sqshrun         v1.8B,  v7.8H, #COL_SHIFT-16
-        sqshrun2        v1.16B, v16.8H, #COL_SHIFT-16
-        sqshrun         v3.8B,  v17.8H, #COL_SHIFT-16
-        sqshrun2        v3.16B, v18.8H, #COL_SHIFT-16
+        sqshrun         v1.8b,  v7.8h, #COL_SHIFT-16
+        sqshrun2        v1.16b, v16.8h, #COL_SHIFT-16
+        sqshrun         v3.8b,  v17.8h, #COL_SHIFT-16
+        sqshrun2        v3.16b, v18.8h, #COL_SHIFT-16

        bl              idct_col4_neon2

-        sqshrun         v2.8B,  v7.8H, #COL_SHIFT-16
-        sqshrun2        v2.16B, v16.8H, #COL_SHIFT-16
-        sqshrun         v4.8B,  v17.8H, #COL_SHIFT-16
-        sqshrun2        v4.16B, v18.8H, #COL_SHIFT-16
+        sqshrun         v2.8b,  v7.8h, #COL_SHIFT-16
+        sqshrun2        v2.16b, v16.8h, #COL_SHIFT-16
+        sqshrun         v4.8b,  v17.8h, #COL_SHIFT-16
+        sqshrun2        v4.16b, v18.8h, #COL_SHIFT-16

-        zip1            v16.4S, v1.4S, v2.4S
-        zip2            v17.4S, v1.4S, v2.4S
+        zip1            v16.4s, v1.4s, v2.4s
+        zip2            v17.4s, v1.4s, v2.4s

-        st1             {v16.D}[0], [x0], x1
-        st1             {v16.D}[1], [x0], x1
+        st1             {v16.d}[0], [x0], x1
+        st1             {v16.d}[1], [x0], x1

-        zip1            v18.4S, v3.4S, v4.4S
-        zip2            v19.4S, v3.4S, v4.4S
+        zip1            v18.4s, v3.4s, v4.4s
+        zip2            v19.4s, v3.4s, v4.4s

-        st1             {v17.D}[0], [x0], x1
-        st1             {v17.D}[1], [x0], x1
-        st1             {v18.D}[0], [x0], x1
-        st1             {v18.D}[1], [x0], x1
-        st1             {v19.D}[0], [x0], x1
-        st1             {v19.D}[1], [x0], x1
+        st1             {v17.d}[0], [x0], x1
+        st1             {v17.d}[1], [x0], x1
+        st1             {v18.d}[0], [x0], x1
+        st1             {v18.d}[1], [x0], x1
+        st1             {v19.d}[0], [x0], x1
+        st1             {v19.d}[1], [x0], x1

        idct_end
 endfunc
@@ -267,59 +267,59 @@ function ff_simple_idct_add_neon, export=1
        idct_row4_neon  v28, v29, v30, v31, 2
        bl              idct_col4_neon1

-        sshr            v1.8H, v7.8H, #COL_SHIFT-16
-        sshr            v2.8H, v16.8H, #COL_SHIFT-16
-        sshr            v3.8H, v17.8H, #COL_SHIFT-16
-        sshr            v4.8H, v18.8H, #COL_SHIFT-16
+        sshr            v1.8h, v7.8h, #COL_SHIFT-16
+        sshr            v2.8h, v16.8h, #COL_SHIFT-16
+        sshr            v3.8h, v17.8h, #COL_SHIFT-16
+        sshr            v4.8h, v18.8h, #COL_SHIFT-16

        bl              idct_col4_neon2

-        sshr            v7.8H, v7.8H, #COL_SHIFT-16
-        sshr            v16.8H, v16.8H, #COL_SHIFT-16
-        sshr            v17.8H, v17.8H, #COL_SHIFT-16
-        sshr            v18.8H, v18.8H, #COL_SHIFT-16
+        sshr            v7.8h, v7.8h, #COL_SHIFT-16
+        sshr            v16.8h, v16.8h, #COL_SHIFT-16
+        sshr            v17.8h, v17.8h, #COL_SHIFT-16
+        sshr            v18.8h, v18.8h, #COL_SHIFT-16

        mov             x9,  x0
-        ld1             {v19.D}[0], [x0], x1
-        zip1            v23.2D, v1.2D, v7.2D
-        zip2            v24.2D, v1.2D, v7.2D
-        ld1             {v19.D}[1], [x0], x1
-        zip1            v25.2D, v2.2D, v16.2D
-        zip2            v26.2D, v2.2D, v16.2D
-        ld1             {v20.D}[0], [x0], x1
-        zip1            v27.2D, v3.2D, v17.2D
-        zip2            v28.2D, v3.2D, v17.2D
-        ld1             {v20.D}[1], [x0], x1
-        zip1            v29.2D, v4.2D, v18.2D
-        zip2            v30.2D, v4.2D, v18.2D
-        ld1             {v21.D}[0], [x0], x1
-        uaddw           v23.8H, v23.8H, v19.8B
-        uaddw2          v24.8H, v24.8H, v19.16B
-        ld1             {v21.D}[1], [x0], x1
-        sqxtun          v23.8B, v23.8H
-        sqxtun2         v23.16B, v24.8H
-        ld1             {v22.D}[0], [x0], x1
-        uaddw           v24.8H, v25.8H, v20.8B
-        uaddw2          v25.8H, v26.8H, v20.16B
-        ld1             {v22.D}[1], [x0], x1
-        sqxtun          v24.8B, v24.8H
-        sqxtun2         v24.16B, v25.8H
-        st1             {v23.D}[0], [x9], x1
-        uaddw           v25.8H, v27.8H, v21.8B
-        uaddw2          v26.8H, v28.8H, v21.16B
-        st1             {v23.D}[1], [x9], x1
-        sqxtun          v25.8B, v25.8H
-        sqxtun2         v25.16B, v26.8H
-        st1             {v24.D}[0], [x9], x1
-        uaddw           v26.8H, v29.8H, v22.8B
-        uaddw2          v27.8H, v30.8H, v22.16B
-        st1             {v24.D}[1], [x9], x1
-        sqxtun          v26.8B, v26.8H
-        sqxtun2         v26.16B, v27.8H
-        st1             {v25.D}[0], [x9], x1
-        st1             {v25.D}[1], [x9], x1
-        st1             {v26.D}[0], [x9], x1
-        st1             {v26.D}[1], [x9], x1
+        ld1             {v19.d}[0], [x0], x1
+        zip1            v23.2d, v1.2d, v7.2d
+        zip2            v24.2d, v1.2d, v7.2d
+        ld1             {v19.d}[1], [x0], x1
+        zip1            v25.2d, v2.2d, v16.2d
+        zip2            v26.2d, v2.2d, v16.2d
+        ld1             {v20.d}[0], [x0], x1
+        zip1            v27.2d, v3.2d, v17.2d
+        zip2            v28.2d, v3.2d, v17.2d
+        ld1             {v20.d}[1], [x0], x1
+        zip1            v29.2d, v4.2d, v18.2d
+        zip2            v30.2d, v4.2d, v18.2d
+        ld1             {v21.d}[0], [x0], x1
+        uaddw           v23.8h, v23.8h, v19.8b
+        uaddw2          v24.8h, v24.8h, v19.16b
+        ld1             {v21.d}[1], [x0], x1
+        sqxtun          v23.8b, v23.8h
+        sqxtun2         v23.16b, v24.8h
+        ld1             {v22.d}[0], [x0], x1
+        uaddw           v24.8h, v25.8h, v20.8b
+        uaddw2          v25.8h, v26.8h, v20.16b
+        ld1             {v22.d}[1], [x0], x1
+        sqxtun          v24.8b, v24.8h
+        sqxtun2         v24.16b, v25.8h
+        st1             {v23.d}[0], [x9], x1
+        uaddw           v25.8h, v27.8h, v21.8b
+        uaddw2          v26.8h, v28.8h, v21.16b
+        st1             {v23.d}[1], [x9], x1
+        sqxtun          v25.8b, v25.8h
+        sqxtun2         v25.16b, v26.8h
+        st1             {v24.d}[0], [x9], x1
+        uaddw           v26.8h, v29.8h, v22.8b
+        uaddw2          v27.8h, v30.8h, v22.16b
+        st1             {v24.d}[1], [x9], x1
+        sqxtun          v26.8b, v26.8h
+        sqxtun2         v26.16b, v27.8h
+        st1             {v25.d}[0], [x9], x1
+        st1             {v25.d}[1], [x9], x1
+        st1             {v26.d}[0], [x9], x1
+        st1             {v26.d}[1], [x9], x1

        idct_end
 endfunc
@@ -333,30 +333,30 @@ function ff_simple_idct_neon, export=1
        sub             x2, x2, #128
        bl              idct_col4_neon1

-        sshr            v1.8H, v7.8H, #COL_SHIFT-16
-        sshr            v2.8H, v16.8H, #COL_SHIFT-16
-        sshr            v3.8H, v17.8H, #COL_SHIFT-16
-        sshr            v4.8H, v18.8H, #COL_SHIFT-16
+        sshr            v1.8h, v7.8h, #COL_SHIFT-16
+        sshr            v2.8h, v16.8h, #COL_SHIFT-16
+        sshr            v3.8h, v17.8h, #COL_SHIFT-16
+        sshr            v4.8h, v18.8h, #COL_SHIFT-16

        bl              idct_col4_neon2

-        sshr            v7.8H, v7.8H, #COL_SHIFT-16
-        sshr            v16.8H, v16.8H, #COL_SHIFT-16
-        sshr            v17.8H, v17.8H, #COL_SHIFT-16
-        sshr            v18.8H, v18.8H, #COL_SHIFT-16
+        sshr            v7.8h, v7.8h, #COL_SHIFT-16
+        sshr            v16.8h, v16.8h, #COL_SHIFT-16
+        sshr            v17.8h, v17.8h, #COL_SHIFT-16
+        sshr            v18.8h, v18.8h, #COL_SHIFT-16

-        zip1            v23.2D, v1.2D, v7.2D
-        zip2            v24.2D, v1.2D, v7.2D
-        st1             {v23.2D,v24.2D}, [x2], #32
-        zip1            v25.2D, v2.2D, v16.2D
-        zip2            v26.2D, v2.2D, v16.2D
-        st1             {v25.2D,v26.2D}, [x2], #32
-        zip1            v27.2D, v3.2D, v17.2D
-        zip2            v28.2D, v3.2D, v17.2D
-        st1             {v27.2D,v28.2D}, [x2], #32
-        zip1            v29.2D, v4.2D, v18.2D
-        zip2            v30.2D, v4.2D, v18.2D
-        st1             {v29.2D,v30.2D}, [x2], #32
+        zip1            v23.2d, v1.2d, v7.2d
+        zip2            v24.2d, v1.2d, v7.2d
+        st1             {v23.2d,v24.2d}, [x2], #32
+        zip1            v25.2d, v2.2d, v16.2d
+        zip2            v26.2d, v2.2d, v16.2d
+        st1             {v25.2d,v26.2d}, [x2], #32
+        zip1            v27.2d, v3.2d, v17.2d
+        zip2            v28.2d, v3.2d, v17.2d
+        st1             {v27.2d,v28.2d}, [x2], #32
+        zip1            v29.2d, v4.2d, v18.2d
+        zip2            v30.2d, v4.2d, v18.2d
+        st1             {v29.2d,v30.2d}, [x2], #32

        idct_end
 endfunc
@@ -330,32 +330,32 @@ endfunc
        //   v17: hev

        // convert to signed value:
-        eor            v3.16b, v3.16b, v21.16b           // PS0 = P0 ^ 0x80
-        eor            v4.16b, v4.16b, v21.16b           // QS0 = Q0 ^ 0x80
+        eor             v3.16b, v3.16b, v21.16b           // PS0 = P0 ^ 0x80
+        eor             v4.16b, v4.16b, v21.16b           // QS0 = Q0 ^ 0x80

-        movi           v20.8h, #3
-        ssubl          v18.8h, v4.8b,  v3.8b             // QS0 - PS0
-        ssubl2         v19.8h, v4.16b, v3.16b            //   (widened to 16bit)
-        eor            v2.16b, v2.16b, v21.16b           // PS1 = P1 ^ 0x80
-        eor            v5.16b, v5.16b, v21.16b           // QS1 = Q1 ^ 0x80
-        mul            v18.8h, v18.8h, v20.8h            // w = 3 * (QS0 - PS0)
-        mul            v19.8h, v19.8h, v20.8h
+        movi            v20.8h, #3
+        ssubl           v18.8h, v4.8b,  v3.8b             // QS0 - PS0
+        ssubl2          v19.8h, v4.16b, v3.16b            //   (widened to 16bit)
+        eor             v2.16b, v2.16b, v21.16b           // PS1 = P1 ^ 0x80
+        eor             v5.16b, v5.16b, v21.16b           // QS1 = Q1 ^ 0x80
+        mul             v18.8h, v18.8h, v20.8h            // w = 3 * (QS0 - PS0)
+        mul             v19.8h, v19.8h, v20.8h

-        sqsub          v20.16b, v2.16b, v5.16b           // clamp(PS1-QS1)
-        movi           v22.16b, #4
-        movi           v23.16b, #3
+        sqsub           v20.16b, v2.16b, v5.16b           // clamp(PS1-QS1)
+        movi            v22.16b, #4
+        movi            v23.16b, #3
    .if \inner
-        and            v20.16b, v20.16b, v17.16b         // if(hev) w += clamp(PS1-QS1)
+        and             v20.16b, v20.16b, v17.16b         // if(hev) w += clamp(PS1-QS1)
    .endif
-        saddw          v18.8h,  v18.8h, v20.8b           // w += clamp(PS1-QS1)
-        saddw2         v19.8h,  v19.8h, v20.16b
-        sqxtn          v18.8b,  v18.8h                   // narrow result back into v18
-        sqxtn2         v18.16b, v19.8h
+        saddw           v18.8h,  v18.8h, v20.8b           // w += clamp(PS1-QS1)
+        saddw2          v19.8h,  v19.8h, v20.16b
+        sqxtn           v18.8b,  v18.8h                   // narrow result back into v18
+        sqxtn2          v18.16b, v19.8h
    .if !\inner && !\simple
-        eor            v1.16b,  v1.16b,  v21.16b         // PS2 = P2 ^ 0x80
-        eor            v6.16b,  v6.16b,  v21.16b         // QS2 = Q2 ^ 0x80
+        eor             v1.16b,  v1.16b,  v21.16b         // PS2 = P2 ^ 0x80
+        eor             v6.16b,  v6.16b,  v21.16b         // QS2 = Q2 ^ 0x80
    .endif
-        and            v18.16b, v18.16b, v16.16b         // w &= normal_limit
+        and             v18.16b, v18.16b, v16.16b         // w &= normal_limit

        // registers used at this point..
        //   v0 -> P3  (don't corrupt)
@@ -375,44 +375,44 @@ endfunc
        //   P0 = s2u(PS0 + c2);

    .if \simple
-        sqadd          v19.16b, v18.16b, v22.16b           // c1 = clamp((w&hev)+4)
-        sqadd          v20.16b, v18.16b, v23.16b           // c2 = clamp((w&hev)+3)
-        sshr           v19.16b, v19.16b, #3                // c1 >>= 3
-        sshr           v20.16b, v20.16b, #3                // c2 >>= 3
-        sqsub          v4.16b,  v4.16b,  v19.16b           // QS0 = clamp(QS0-c1)
-        sqadd          v3.16b,  v3.16b,  v20.16b           // PS0 = clamp(PS0+c2)
-        eor            v4.16b,  v4.16b,  v21.16b           // Q0 = QS0 ^ 0x80
-        eor            v3.16b,  v3.16b,  v21.16b           // P0 = PS0 ^ 0x80
-        eor            v5.16b,  v5.16b,  v21.16b           // Q1 = QS1 ^ 0x80
-        eor            v2.16b,  v2.16b,  v21.16b           // P1 = PS1 ^ 0x80
+        sqadd           v19.16b, v18.16b, v22.16b           // c1 = clamp((w&hev)+4)
+        sqadd           v20.16b, v18.16b, v23.16b           // c2 = clamp((w&hev)+3)
+        sshr            v19.16b, v19.16b, #3                // c1 >>= 3
+        sshr            v20.16b, v20.16b, #3                // c2 >>= 3
+        sqsub           v4.16b,  v4.16b,  v19.16b           // QS0 = clamp(QS0-c1)
+        sqadd           v3.16b,  v3.16b,  v20.16b           // PS0 = clamp(PS0+c2)
+        eor             v4.16b,  v4.16b,  v21.16b           // Q0 = QS0 ^ 0x80
+        eor             v3.16b,  v3.16b,  v21.16b           // P0 = PS0 ^ 0x80
+        eor             v5.16b,  v5.16b,  v21.16b           // Q1 = QS1 ^ 0x80
+        eor             v2.16b,  v2.16b,  v21.16b           // P1 = PS1 ^ 0x80
    .elseif \inner
        // the !is4tap case of filter_common, only used for inner blocks
        //   c3 = ((c1&~hev) + 1) >> 1;
        //   Q1 = s2u(QS1 - c3);
        //   P1 = s2u(PS1 + c3);
-        sqadd          v19.16b, v18.16b, v22.16b           // c1 = clamp((w&hev)+4)
-        sqadd          v20.16b, v18.16b, v23.16b           // c2 = clamp((w&hev)+3)
-        sshr           v19.16b, v19.16b, #3                // c1 >>= 3
-        sshr           v20.16b, v20.16b, #3                // c2 >>= 3
-        sqsub          v4.16b,  v4.16b,  v19.16b           // QS0 = clamp(QS0-c1)
-        sqadd          v3.16b,  v3.16b,  v20.16b           // PS0 = clamp(PS0+c2)
-        bic            v19.16b, v19.16b, v17.16b           // c1 & ~hev
-        eor            v4.16b,  v4.16b,  v21.16b           // Q0 = QS0 ^ 0x80
-        srshr          v19.16b, v19.16b, #1                // c3 >>= 1
-        eor            v3.16b,  v3.16b,  v21.16b           // P0 = PS0 ^ 0x80
-        sqsub          v5.16b,  v5.16b,  v19.16b           // QS1 = clamp(QS1-c3)
-        sqadd          v2.16b,  v2.16b,  v19.16b           // PS1 = clamp(PS1+c3)
-        eor            v5.16b,  v5.16b,  v21.16b           // Q1 = QS1 ^ 0x80
-        eor            v2.16b,  v2.16b,  v21.16b           // P1 = PS1 ^ 0x80
+        sqadd           v19.16b, v18.16b, v22.16b           // c1 = clamp((w&hev)+4)
+        sqadd           v20.16b, v18.16b, v23.16b           // c2 = clamp((w&hev)+3)
+        sshr            v19.16b, v19.16b, #3                // c1 >>= 3
+        sshr            v20.16b, v20.16b, #3                // c2 >>= 3
+        sqsub           v4.16b,  v4.16b,  v19.16b           // QS0 = clamp(QS0-c1)
+        sqadd           v3.16b,  v3.16b,  v20.16b           // PS0 = clamp(PS0+c2)
+        bic             v19.16b, v19.16b, v17.16b           // c1 & ~hev
+        eor             v4.16b,  v4.16b,  v21.16b           // Q0 = QS0 ^ 0x80
+        srshr           v19.16b, v19.16b, #1                // c3 >>= 1
+        eor             v3.16b,  v3.16b,  v21.16b           // P0 = PS0 ^ 0x80
+        sqsub           v5.16b,  v5.16b,  v19.16b           // QS1 = clamp(QS1-c3)
+        sqadd           v2.16b,  v2.16b,  v19.16b           // PS1 = clamp(PS1+c3)
+        eor             v5.16b,  v5.16b,  v21.16b           // Q1 = QS1 ^ 0x80
+        eor             v2.16b,  v2.16b,  v21.16b           // P1 = PS1 ^ 0x80
    .else
-        and            v20.16b, v18.16b, v17.16b           // w & hev
-        sqadd          v19.16b, v20.16b, v22.16b           // c1 = clamp((w&hev)+4)
-        sqadd          v20.16b, v20.16b, v23.16b           // c2 = clamp((w&hev)+3)
-        sshr           v19.16b, v19.16b, #3                // c1 >>= 3
-        sshr           v20.16b, v20.16b, #3                // c2 >>= 3
-        bic            v18.16b, v18.16b, v17.16b           // w &= ~hev
-        sqsub          v4.16b,  v4.16b,  v19.16b           // QS0 = clamp(QS0-c1)
-        sqadd          v3.16b,  v3.16b,  v20.16b           // PS0 = clamp(PS0+c2)
+        and             v20.16b, v18.16b, v17.16b           // w & hev
+        sqadd           v19.16b, v20.16b, v22.16b           // c1 = clamp((w&hev)+4)
+        sqadd           v20.16b, v20.16b, v23.16b           // c2 = clamp((w&hev)+3)
+        sshr            v19.16b, v19.16b, #3                // c1 >>= 3
+        sshr            v20.16b, v20.16b, #3                // c2 >>= 3
+        bic             v18.16b, v18.16b, v17.16b           // w &= ~hev
+        sqsub           v4.16b,  v4.16b,  v19.16b           // QS0 = clamp(QS0-c1)
+        sqadd           v3.16b,  v3.16b,  v20.16b           // PS0 = clamp(PS0+c2)

        // filter_mbedge:
        //   a = clamp((27*w + 63) >> 7);
@@ -424,35 +424,35 @@ endfunc
        //   a = clamp((9*w + 63) >> 7);
        //   Q2 = s2u(QS2 - a);
        //   P2 = s2u(PS2 + a);
-        movi           v17.8h,  #63
-        sshll          v22.8h,  v18.8b, #3
-        sshll2         v23.8h,  v18.16b, #3
-        saddw          v22.8h,  v22.8h, v18.8b
-        saddw2         v23.8h,  v23.8h, v18.16b
-        add            v16.8h,  v17.8h, v22.8h
-        add            v17.8h,  v17.8h, v23.8h           //  9*w + 63
-        add            v19.8h,  v16.8h, v22.8h
-        add            v20.8h,  v17.8h, v23.8h           // 18*w + 63
-        add            v22.8h,  v19.8h, v22.8h
-        add            v23.8h,  v20.8h, v23.8h           // 27*w + 63
-        sqshrn         v16.8b,  v16.8h,  #7
-        sqshrn2        v16.16b, v17.8h, #7              // clamp(( 9*w + 63)>>7)
-        sqshrn         v19.8b,  v19.8h, #7
-        sqshrn2        v19.16b, v20.8h, #7              // clamp((18*w + 63)>>7)
-        sqshrn         v22.8b,  v22.8h, #7
-        sqshrn2        v22.16b, v23.8h, #7              // clamp((27*w + 63)>>7)
-        sqadd          v1.16b,  v1.16b,  v16.16b        // PS2 = clamp(PS2+a)
-        sqsub          v6.16b,  v6.16b,  v16.16b        // QS2 = clamp(QS2-a)
-        sqadd          v2.16b,  v2.16b,  v19.16b        // PS1 = clamp(PS1+a)
-        sqsub          v5.16b,  v5.16b,  v19.16b        // QS1 = clamp(QS1-a)
-        sqadd          v3.16b,  v3.16b,  v22.16b        // PS0 = clamp(PS0+a)
-        sqsub          v4.16b,  v4.16b,  v22.16b        // QS0 = clamp(QS0-a)
-        eor            v3.16b,  v3.16b,  v21.16b        // P0 = PS0 ^ 0x80
-        eor            v4.16b,  v4.16b,  v21.16b        // Q0 = QS0 ^ 0x80
-        eor            v2.16b,  v2.16b,  v21.16b        // P1 = PS1 ^ 0x80
-        eor            v5.16b,  v5.16b,  v21.16b        // Q1 = QS1 ^ 0x80
-        eor            v1.16b,  v1.16b,  v21.16b        // P2 = PS2 ^ 0x80
-        eor            v6.16b,  v6.16b,  v21.16b        // Q2 = QS2 ^ 0x80
+        movi            v17.8h,  #63
+        sshll           v22.8h,  v18.8b, #3
+        sshll2          v23.8h,  v18.16b, #3
+        saddw           v22.8h,  v22.8h, v18.8b
+        saddw2          v23.8h,  v23.8h, v18.16b
+        add             v16.8h,  v17.8h, v22.8h
+        add             v17.8h,  v17.8h, v23.8h           //  9*w + 63
+        add             v19.8h,  v16.8h, v22.8h
+        add             v20.8h,  v17.8h, v23.8h           // 18*w + 63
+        add             v22.8h,  v19.8h, v22.8h
+        add             v23.8h,  v20.8h, v23.8h           // 27*w + 63
+        sqshrn          v16.8b,  v16.8h,  #7
+        sqshrn2         v16.16b, v17.8h, #7              // clamp(( 9*w + 63)>>7)
+        sqshrn          v19.8b,  v19.8h, #7
+        sqshrn2         v19.16b, v20.8h, #7              // clamp((18*w + 63)>>7)
+        sqshrn          v22.8b,  v22.8h, #7
+        sqshrn2         v22.16b, v23.8h, #7              // clamp((27*w + 63)>>7)
+        sqadd           v1.16b,  v1.16b,  v16.16b        // PS2 = clamp(PS2+a)
+        sqsub           v6.16b,  v6.16b,  v16.16b        // QS2 = clamp(QS2-a)
+        sqadd           v2.16b,  v2.16b,  v19.16b        // PS1 = clamp(PS1+a)
+        sqsub           v5.16b,  v5.16b,  v19.16b        // QS1 = clamp(QS1-a)
+        sqadd           v3.16b,  v3.16b,  v22.16b        // PS0 = clamp(PS0+a)
+        sqsub           v4.16b,  v4.16b,  v22.16b        // QS0 = clamp(QS0-a)
+        eor             v3.16b,  v3.16b,  v21.16b        // P0 = PS0 ^ 0x80
+        eor             v4.16b,  v4.16b,  v21.16b        // Q0 = QS0 ^ 0x80
+        eor             v2.16b,  v2.16b,  v21.16b        // P1 = PS1 ^ 0x80
+        eor             v5.16b,  v5.16b,  v21.16b        // Q1 = QS1 ^ 0x80
+        eor             v1.16b,  v1.16b,  v21.16b        // P2 = PS2 ^ 0x80
+        eor             v6.16b,  v6.16b,  v21.16b        // Q2 = QS2 ^ 0x80
    .endif
 .endm

@@ -507,48 +507,48 @@ function ff_vp8_v_loop_filter8uv\name\()_neon, export=1
        sub             x0,  x0,  x2,  lsl #2
        sub             x1,  x1,  x2,  lsl #2
        // Load pixels:
-        ld1          {v0.d}[0],     [x0], x2  // P3
-        ld1          {v0.d}[1],     [x1], x2  // P3
-        ld1          {v1.d}[0],     [x0], x2  // P2
-        ld1          {v1.d}[1],     [x1], x2  // P2
-        ld1          {v2.d}[0],     [x0], x2  // P1
-        ld1          {v2.d}[1],     [x1], x2  // P1
-        ld1          {v3.d}[0],     [x0], x2  // P0
-        ld1          {v3.d}[1],     [x1], x2  // P0
-        ld1          {v4.d}[0],     [x0], x2  // Q0
-        ld1          {v4.d}[1],     [x1], x2  // Q0
-        ld1          {v5.d}[0],     [x0], x2  // Q1
-        ld1          {v5.d}[1],     [x1], x2  // Q1
-        ld1          {v6.d}[0],     [x0], x2  // Q2
-        ld1          {v6.d}[1],     [x1], x2  // Q2
-        ld1          {v7.d}[0],     [x0]      // Q3
-        ld1          {v7.d}[1],     [x1]      // Q3
+        ld1             {v0.d}[0],     [x0], x2  // P3
+        ld1             {v0.d}[1],     [x1], x2  // P3
+        ld1             {v1.d}[0],     [x0], x2  // P2
+        ld1             {v1.d}[1],     [x1], x2  // P2
+        ld1             {v2.d}[0],     [x0], x2  // P1
+        ld1             {v2.d}[1],     [x1], x2  // P1
+        ld1             {v3.d}[0],     [x0], x2  // P0
+        ld1             {v3.d}[1],     [x1], x2  // P0
+        ld1             {v4.d}[0],     [x0], x2  // Q0
+        ld1             {v4.d}[1],     [x1], x2  // Q0
+        ld1             {v5.d}[0],     [x0], x2  // Q1
+        ld1             {v5.d}[1],     [x1], x2  // Q1
+        ld1             {v6.d}[0],     [x0], x2  // Q2
+        ld1             {v6.d}[1],     [x1], x2  // Q2
+        ld1             {v7.d}[0],     [x0]      // Q3
+        ld1             {v7.d}[1],     [x1]      // Q3

-        dup          v22.16b, w3                 // flim_E
-        dup          v23.16b, w4                 // flim_I
+        dup             v22.16b, w3                 // flim_E
+        dup             v23.16b, w4                 // flim_I

        vp8_loop_filter inner=\inner, hev_thresh=w5

        // back up to P2:  u,v -= stride * 6
-        sub          x0,  x0,  x2,  lsl #2
-        sub          x1,  x1,  x2,  lsl #2
-        sub          x0,  x0,  x2,  lsl #1
-        sub          x1,  x1,  x2,  lsl #1
+        sub             x0,  x0,  x2,  lsl #2
+        sub             x1,  x1,  x2,  lsl #2
+        sub             x0,  x0,  x2,  lsl #1
+        sub             x1,  x1,  x2,  lsl #1

        // Store pixels:

-        st1          {v1.d}[0],     [x0], x2  // P2
-        st1          {v1.d}[1],     [x1], x2  // P2
-        st1          {v2.d}[0],     [x0], x2  // P1
-        st1          {v2.d}[1],     [x1], x2  // P1
-        st1          {v3.d}[0],     [x0], x2  // P0
-        st1          {v3.d}[1],     [x1], x2  // P0
-        st1          {v4.d}[0],     [x0], x2  // Q0
-        st1          {v4.d}[1],     [x1], x2  // Q0
-        st1          {v5.d}[0],     [x0], x2  // Q1
-        st1          {v5.d}[1],     [x1], x2  // Q1
-        st1          {v6.d}[0],     [x0]      // Q2
-        st1          {v6.d}[1],     [x1]      // Q2
+        st1             {v1.d}[0],     [x0], x2  // P2
+        st1             {v1.d}[1],     [x1], x2  // P2
+        st1             {v2.d}[0],     [x0], x2  // P1
+        st1             {v2.d}[1],     [x1], x2  // P1
+        st1             {v3.d}[0],     [x0], x2  // P0
+        st1             {v3.d}[1],     [x1], x2  // P0
+        st1             {v4.d}[0],     [x0], x2  // Q0
+        st1             {v4.d}[1],     [x1], x2  // Q0
+        st1             {v5.d}[0],     [x0], x2  // Q1
+        st1             {v5.d}[1],     [x1], x2  // Q1
+        st1             {v6.d}[0],     [x0]      // Q2
+        st1             {v6.d}[1],     [x1]      // Q2

        ret
 endfunc
@@ -579,7 +579,7 @@ function ff_vp8_h_loop_filter16\name\()_neon, export=1
        ld1             {v6.d}[1], [x0], x1
        ld1             {v7.d}[1], [x0], x1

-        transpose_8x16B   v0,  v1,  v2,  v3,  v4,  v5,  v6,  v7, v30, v31
+        transpose_8x16B v0,  v1,  v2,  v3,  v4,  v5,  v6,  v7, v30, v31

        dup             v22.16b, w2                 // flim_E
    .if !\simple
@@ -590,7 +590,7 @@ function ff_vp8_h_loop_filter16\name\()_neon, export=1

        sub             x0,  x0,  x1, lsl #4    // backup 16 rows

-        transpose_8x16B   v0,  v1,  v2,  v3,  v4,  v5,  v6,  v7, v30, v31
+        transpose_8x16B v0,  v1,  v2,  v3,  v4,  v5,  v6,  v7, v30, v31

        // Store pixels:
        st1             {v0.d}[0], [x0], x1
@@ -624,24 +624,24 @@ function ff_vp8_h_loop_filter8uv\name\()_neon, export=1
        sub             x1,  x1,  #4

        // Load pixels:
-        ld1          {v0.d}[0],     [x0], x2 // load u
-        ld1          {v0.d}[1],     [x1], x2 // load v
-        ld1          {v1.d}[0],     [x0], x2
-        ld1          {v1.d}[1],     [x1], x2
-        ld1          {v2.d}[0],     [x0], x2
-        ld1          {v2.d}[1],     [x1], x2
-        ld1          {v3.d}[0],     [x0], x2
-        ld1          {v3.d}[1],     [x1], x2
-        ld1          {v4.d}[0],     [x0], x2
-        ld1          {v4.d}[1],     [x1], x2
-        ld1          {v5.d}[0],     [x0], x2
-        ld1          {v5.d}[1],     [x1], x2
-        ld1          {v6.d}[0],     [x0], x2
-        ld1          {v6.d}[1],     [x1], x2
-        ld1          {v7.d}[0],     [x0], x2
-        ld1          {v7.d}[1],     [x1], x2
+        ld1             {v0.d}[0],     [x0], x2 // load u
+        ld1             {v0.d}[1],     [x1], x2 // load v
+        ld1             {v1.d}[0],     [x0], x2
+        ld1             {v1.d}[1],     [x1], x2
+        ld1             {v2.d}[0],     [x0], x2
+        ld1             {v2.d}[1],     [x1], x2
+        ld1             {v3.d}[0],     [x0], x2
+        ld1             {v3.d}[1],     [x1], x2
+        ld1             {v4.d}[0],     [x0], x2
+        ld1             {v4.d}[1],     [x1], x2
+        ld1             {v5.d}[0],     [x0], x2
+        ld1             {v5.d}[1],     [x1], x2
+        ld1             {v6.d}[0],     [x0], x2
+        ld1             {v6.d}[1],     [x1], x2
+        ld1             {v7.d}[0],     [x0], x2
+        ld1             {v7.d}[1],     [x1], x2

-        transpose_8x16B   v0,  v1,  v2,  v3,  v4,  v5,  v6,  v7, v30, v31
+        transpose_8x16B v0,  v1,  v2,  v3,  v4,  v5,  v6,  v7, v30, v31

        dup             v22.16b, w3                 // flim_E
        dup             v23.16b, w4                 // flim_I
@@ -651,25 +651,25 @@ function ff_vp8_h_loop_filter8uv\name\()_neon, export=1
        sub             x0,  x0,  x2, lsl #3    // backup u 8 rows
        sub             x1,  x1,  x2, lsl #3    // backup v 8 rows

-        transpose_8x16B   v0,  v1,  v2,  v3,  v4,  v5,  v6,  v7, v30, v31
+        transpose_8x16B v0,  v1,  v2,  v3,  v4,  v5,  v6,  v7, v30, v31

        // Store pixels:
-        st1          {v0.d}[0],     [x0], x2 // load u
-        st1          {v0.d}[1],     [x1], x2 // load v
-        st1          {v1.d}[0],     [x0], x2
-        st1          {v1.d}[1],     [x1], x2
-        st1          {v2.d}[0],     [x0], x2
-        st1          {v2.d}[1],     [x1], x2
-        st1          {v3.d}[0],     [x0], x2
-        st1          {v3.d}[1],     [x1], x2
-        st1          {v4.d}[0],     [x0], x2
-        st1          {v4.d}[1],     [x1], x2
-        st1          {v5.d}[0],     [x0], x2
-        st1          {v5.d}[1],     [x1], x2
-        st1          {v6.d}[0],     [x0], x2
-        st1          {v6.d}[1],     [x1], x2
-        st1          {v7.d}[0],     [x0]
-        st1          {v7.d}[1],     [x1]
+        st1             {v0.d}[0],     [x0], x2 // load u
+        st1             {v0.d}[1],     [x1], x2 // load v
+        st1             {v1.d}[0],     [x0], x2
+        st1             {v1.d}[1],     [x1], x2
+        st1             {v2.d}[0],     [x0], x2
+        st1             {v2.d}[1],     [x1], x2
+        st1             {v3.d}[0],     [x0], x2
+        st1             {v3.d}[1],     [x1], x2
+        st1             {v4.d}[0],     [x0], x2
+        st1             {v4.d}[1],     [x1], x2
+        st1             {v5.d}[0],     [x0], x2
+        st1             {v5.d}[1],     [x1], x2
+        st1             {v6.d}[0],     [x0], x2
+        st1             {v6.d}[1],     [x1], x2
+        st1             {v7.d}[0],     [x0]
+        st1             {v7.d}[1],     [x1]

        ret

@@ -230,6 +230,9 @@ function \type\()_8tap_\size\()h_\idx1\idx2
        // reduced dst stride
 .if \size >= 16
        sub             x1,  x1,  x5
+.elseif \size == 4
+        add             x12, x2,  #8
+        add             x13, x7,  #8
 .endif
        // size >= 16 loads two qwords and increments x2,
        // for size 4/8 it's enough with one qword and no
@@ -248,9 +251,14 @@ function \type\()_8tap_\size\()h_\idx1\idx2
 .if \size >= 16
        ld1             {v4.8b,  v5.8b,  v6.8b},  [x2], #24
        ld1             {v16.8b, v17.8b, v18.8b}, [x7], #24
-.else
+.elseif \size == 8
        ld1             {v4.8b,  v5.8b},  [x2]
        ld1             {v16.8b, v17.8b}, [x7]
+.else // \size == 4
+        ld1             {v4.8b},  [x2]
+        ld1             {v16.8b}, [x7]
+        ld1             {v5.s}[0],  [x12], x3
+        ld1             {v17.s}[0], [x13], x3
 .endif
        uxtl            v4.8h,  v4.8b
        uxtl            v5.8h,  v5.8b
@@ -104,26 +104,26 @@ static int aasc_decode_frame(AVCodecContext *avctx,
        ff_msrle_decode(avctx, s->frame, 8, &s->gb);
        break;
    case MKTAG('A', 'A', 'S', 'C'):
-    switch (compr) {
-    case 0:
-        stride = (avctx->width * psize + psize) & ~psize;
-        if (buf_size < stride * avctx->height)
+        switch (compr) {
+        case 0:
+            stride = (avctx->width * psize + psize) & ~psize;
+            if (buf_size < stride * avctx->height)
+                return AVERROR_INVALIDDATA;
+            for (i = avctx->height - 1; i >= 0; i--) {
+                memcpy(s->frame->data[0] + i * s->frame->linesize[0], buf, avctx->width * psize);
+                buf += stride;
+                buf_size -= stride;
+            }
+            break;
+        case 1:
+            bytestream2_init(&s->gb, buf, buf_size);
+            ff_msrle_decode(avctx, s->frame, 8, &s->gb);
+            break;
+        default:
+            av_log(avctx, AV_LOG_ERROR, "Unknown compression type %d\n", compr);
            return AVERROR_INVALIDDATA;
-        for (i = avctx->height - 1; i >= 0; i--) {
-            memcpy(s->frame->data[0] + i * s->frame->linesize[0], buf, avctx->width * psize);
-            buf += stride;
-            buf_size -= stride;
        }
        break;
-    case 1:
-        bytestream2_init(&s->gb, buf, buf_size);
-        ff_msrle_decode(avctx, s->frame, 8, &s->gb);
-        break;
-    default:
-        av_log(avctx, AV_LOG_ERROR, "Unknown compression type %d\n", compr);
-        return AVERROR_INVALIDDATA;
-    }
-        break;
    default:
        av_log(avctx, AV_LOG_ERROR, "Unknown FourCC: %X\n", avctx->codec_tag);
        return -1;
@@ -75,6 +75,7 @@
 #define AC3_DYNAMIC_RANGE1      0

 typedef int                     INTFLOAT;
+typedef unsigned int            UINTFLOAT;
 typedef int16_t                 SHORTFLOAT;

 #else /* USE_FIXED */
@@ -94,6 +95,7 @@ typedef int16_t                 SHORTFLOAT;
 #define AC3_DYNAMIC_RANGE1      1.0f

 typedef float                   INTFLOAT;
+typedef float                   UINTFLOAT;
 typedef float                   SHORTFLOAT;

 #endif /* USE_FIXED */
@@ -179,7 +179,9 @@ int av_ac3_parse_header(const uint8_t *buf, size_t size,
    AC3HeaderInfo hdr;
    int err;

-    init_get_bits8(&gb, buf, size);
+    err = init_get_bits8(&gb, buf, size);
+    if (err < 0)
+        return AVERROR_INVALIDDATA;
    err = ff_ac3_parse_header(&gb, &hdr);
    if (err < 0)
        return AVERROR_INVALIDDATA;
@@ -43,7 +43,7 @@ int AC3_NAME(allocate_sample_buffers)(AC3EncodeContext *s)

    FF_ALLOC_OR_GOTO(s->avctx, s->windowed_samples, AC3_WINDOW_SIZE *
                     sizeof(*s->windowed_samples), alloc_fail);
-    FF_ALLOC_ARRAY_OR_GOTO(s->avctx, s->planar_samples, s->channels, sizeof(*s->planar_samples),
+    FF_ALLOCZ_ARRAY_OR_GOTO(s->avctx, s->planar_samples, s->channels, sizeof(*s->planar_samples),
                     alloc_fail);
    for (ch = 0; ch < s->channels; ch++) {
        FF_ALLOCZ_OR_GOTO(s->avctx, s->planar_samples[ch],
@@ -470,8 +470,7 @@ static av_cold int aic_decode_init(AVCodecContext *avctx)
        }
    }

-    ctx->slice_data = av_malloc_array(ctx->slice_width, AIC_BAND_COEFFS
-                                * sizeof(*ctx->slice_data));
+    ctx->slice_data = av_calloc(ctx->slice_width, AIC_BAND_COEFFS * sizeof(*ctx->slice_data));
    if (!ctx->slice_data) {
        av_log(avctx, AV_LOG_ERROR, "Error allocating slice buffer\n");

@@ -29,12 +29,12 @@ static void decorrelate_stereo(int32_t *buffer[2], int nb_samples,
    int i;

    for (i = 0; i < nb_samples; i++) {
-        int32_t a, b;
+        uint32_t a, b;

        a = buffer[0][i];
        b = buffer[1][i];

-        a -= (b * decorr_left_weight) >> decorr_shift;
+        a -= (int)(b * decorr_left_weight) >> decorr_shift;
        b += a;

        buffer[0][i] = b;
@@ -1016,6 +1016,10 @@ static int read_block(ALSDecContext *ctx, ALSBlockData *bd)
    ALSSpecificConfig *sconf = &ctx->sconf;

    *bd->shift_lsbs = 0;
+
+    if (get_bits_left(gb) < 7)
+        return AVERROR_INVALIDDATA;
+
    // read block type flag and read the samples accordingly
    if (get_bits1(gb)) {
        ret = read_var_block_data(ctx, bd);
@@ -1628,7 +1632,7 @@ static int read_frame_data(ALSDecContext *ctx, unsigned int ra_frame)
    AVCodecContext *avctx    = ctx->avctx;
    GetBitContext *gb = &ctx->gb;
    unsigned int div_blocks[32];                ///< block sizes.
-    unsigned int c;
+    int c;
    unsigned int js_blocks[2];
    uint32_t bs_info = 0;
    int ret;
@@ -1806,14 +1810,17 @@ static int decode_frame(AVCodecContext *avctx, void *data, int *got_frame_ptr,
    else
        ctx->cur_frame_length = sconf->frame_length;

-    ctx->highest_decoded_channel = 0;
+    ctx->highest_decoded_channel = -1;
    // decode the frame data
    if ((invalid_frame = read_frame_data(ctx, ra_frame)) < 0)
        av_log(ctx->avctx, AV_LOG_WARNING,
               "Reading frame data failed. Skipping RA unit.\n");

-    if (ctx->highest_decoded_channel == 0)
+    if (ctx->highest_decoded_channel == -1) {
+        av_log(ctx->avctx, AV_LOG_WARNING,
+               "No channel data decoded.\n");
        return AVERROR_INVALIDDATA;
+    }

    ctx->frame_id++;

@@ -2109,8 +2116,8 @@ static av_cold int decode_init(AVCodecContext *avctx)
        ctx->nbits  = av_malloc_array(ctx->cur_frame_length, sizeof(*ctx->nbits));
        ctx->mlz    = av_mallocz(sizeof(*ctx->mlz));

-        if (!ctx->mlz || !ctx->acf || !ctx->shift_value || !ctx->last_shift_value
-            || !ctx->last_acf_mantissa || !ctx->raw_mantissa) {
+        if (!ctx->larray || !ctx->nbits || !ctx->mlz || !ctx->acf || !ctx->shift_value
+            || !ctx->last_shift_value || !ctx->last_acf_mantissa || !ctx->raw_mantissa) {
            av_log(avctx, AV_LOG_ERROR, "Allocating buffer memory failed.\n");
            ret = AVERROR(ENOMEM);
            goto fail;
@@ -2121,6 +2128,10 @@ static av_cold int decode_init(AVCodecContext *avctx)

        for (c = 0; c < avctx->channels; ++c) {
            ctx->raw_mantissa[c] = av_mallocz_array(ctx->cur_frame_length, sizeof(**ctx->raw_mantissa));
+            if (!ctx->raw_mantissa[c]) {
+                av_log(avctx, AV_LOG_ERROR, "Allocating buffer memory failed.\n");
+                return AVERROR(ENOMEM);
+            }
        }
    }

@@ -102,7 +102,7 @@ typedef struct APEFilter {
    int16_t *historybuffer; ///< filter memory
    int16_t *delay;         ///< filtered values

-    int avg;
+    uint32_t avg;
 } APEFilter;

 typedef struct APERice {
@@ -852,7 +852,7 @@ static av_always_inline int filter_fast_3320(APEPredictor *p,
    }

    predictionA = p->buf[delayA] * 2U - p->buf[delayA - 1];
-    p->lastA[filter] = decoded + ((int32_t)(predictionA  * p->coeffsA[filter][0]) >> 9);
+    p->lastA[filter] = decoded + (unsigned)((int32_t)(predictionA  * p->coeffsA[filter][0]) >> 9);

    if ((decoded ^ predictionA) > 0)
        p->coeffsA[filter][0]++;
@@ -882,8 +882,8 @@ static av_always_inline int filter_3800(APEPredictor *p,
        return predictionA;
    }
    d2 =  p->buf[delayA];
-    d1 = (p->buf[delayA] - p->buf[delayA - 1]) * 2U;
-    d0 =  p->buf[delayA] + ((p->buf[delayA - 2] - p->buf[delayA - 1]) * 8U);
+    d1 = (p->buf[delayA] - (unsigned)p->buf[delayA - 1]) * 2;
+    d0 =  p->buf[delayA] + ((p->buf[delayA - 2] - (unsigned)p->buf[delayA - 1]) * 8);
    d3 =  p->buf[delayB] * 2U - p->buf[delayB - 1];
    d4 =  p->buf[delayB];

@@ -903,7 +903,7 @@ static av_always_inline int filter_3800(APEPredictor *p,
    p->coeffsB[filter][0] += (((d3 >> 29) & 4) - 2) * sign;
    p->coeffsB[filter][1] -= (((d4 >> 30) & 2) - 1) * sign;

-    p->filterB[filter] = p->lastA[filter] + (predictionB >> shift);
+    p->filterB[filter] = p->lastA[filter] + (unsigned)(predictionB >> shift);
    p->filterA[filter] = p->filterB[filter] + (unsigned)((int)(p->filterA[filter] * 31U) >> 5);

    return p->filterA[filter];
@@ -928,7 +928,7 @@ static void long_filter_high_3800(int32_t *buffer, int order, int shift, int len
            dotprod += delay[j] * (unsigned)coeffs[j];
            coeffs[j] += ((delay[j] >> 31) | 1) * sign;
        }
-        buffer[i] -= dotprod >> shift;
+        buffer[i] -= (unsigned)(dotprod >> shift);
        for (j = 0; j < order - 1; j++)
            delay[j] = delay[j + 1];
        delay[order - 1] = buffer[i];
@@ -952,7 +952,7 @@ static void long_filter_ehigh_3830(int32_t *buffer, int length)
        for (j = 7; j > 0; j--)
            delay[j] = delay[j - 1];
        delay[0] = buffer[i];
-        buffer[i] -= dotprod >> 9;
+        buffer[i] -= (unsigned)(dotprod >> 9);
    }
 }

@@ -1061,13 +1061,13 @@ static av_always_inline int predictor_update_3930(APEPredictor *p,
                                                  const int delayA)
 {
    int32_t predictionA, sign;
-    int32_t d0, d1, d2, d3;
+    uint32_t d0, d1, d2, d3;

    p->buf[delayA]     = p->lastA[filter];
    d0 = p->buf[delayA    ];
-    d1 = p->buf[delayA    ] - p->buf[delayA - 1];
-    d2 = p->buf[delayA - 1] - p->buf[delayA - 2];
-    d3 = p->buf[delayA - 2] - p->buf[delayA - 3];
+    d1 = p->buf[delayA    ] - (unsigned)p->buf[delayA - 1];
+    d2 = p->buf[delayA - 1] - (unsigned)p->buf[delayA - 2];
+    d3 = p->buf[delayA - 2] - (unsigned)p->buf[delayA - 3];

    predictionA = d0 * p->coeffsA[filter][0] +
                  d1 * p->coeffsA[filter][1] +
@@ -1078,10 +1078,10 @@ static av_always_inline int predictor_update_3930(APEPredictor *p,
    p->filterA[filter] = p->lastA[filter] + ((int)(p->filterA[filter] * 31U) >> 5);

    sign = APESIGN(decoded);
-    p->coeffsA[filter][0] += ((d0 < 0) * 2 - 1) * sign;
-    p->coeffsA[filter][1] += ((d1 < 0) * 2 - 1) * sign;
-    p->coeffsA[filter][2] += ((d2 < 0) * 2 - 1) * sign;
-    p->coeffsA[filter][3] += ((d3 < 0) * 2 - 1) * sign;
+    p->coeffsA[filter][0] += (((int32_t)d0 < 0) * 2 - 1) * sign;
+    p->coeffsA[filter][1] += (((int32_t)d1 < 0) * 2 - 1) * sign;
+    p->coeffsA[filter][2] += (((int32_t)d2 < 0) * 2 - 1) * sign;
+    p->coeffsA[filter][3] += (((int32_t)d3 < 0) * 2 - 1) * sign;

    return p->filterA[filter];
 }
@@ -1309,7 +1309,7 @@ static void do_apply_filter(APEContext *ctx, int version, APEFilter *f,
            absres = res < 0 ? -(unsigned)res : res;
            if (absres)
                *f->adaptcoeffs = APESIGN(res) *
-                                  (8 << ((absres > f->avg * 3) + (absres > f->avg * 4 / 3)));
+                                  (8 << ((absres > f->avg * 3LL) + (absres > (f->avg + f->avg / 3))));
                /* equivalent to the following code
                    if (absres <= f->avg * 4 / 3)
                        *f->adaptcoeffs = APESIGN(res) * 8;
@@ -1559,7 +1559,7 @@ static int ape_decode_frame(AVCodecContext *avctx, void *data,
        for (ch = 0; ch < s->channels; ch++) {
            sample8 = (uint8_t *)frame->data[ch];
            for (i = 0; i < blockstodecode; i++)
-                *sample8++ = (s->decoded[ch][i] + 0x80) & 0xff;
+                *sample8++ = (s->decoded[ch][i] + 0x80U) & 0xff;
        }
        break;
    case 16:
@@ -1581,13 +1581,24 @@ static int ape_decode_frame(AVCodecContext *avctx, void *data,
    s->samples -= blockstodecode;

    if (avctx->err_recognition & AV_EF_CRCCHECK &&
-        s->fileversion >= 3900 && s->bps < 24) {
+        s->fileversion >= 3900) {
        uint32_t crc = s->CRC_state;
        const AVCRC *crc_tab = av_crc_get_table(AV_CRC_32_IEEE_LE);
+        int stride = s->bps == 24 ? 4 : (s->bps>>3);
+        int offset = s->bps == 24;
+        int bytes  = s->bps >> 3;
+
        for (i = 0; i < blockstodecode; i++) {
            for (ch = 0; ch < s->channels; ch++) {
-                uint8_t *smp = frame->data[ch] + (i*(s->bps >> 3));
-                crc = av_crc(crc_tab, crc, smp, s->bps >> 3);
+#if HAVE_BIGENDIAN
+                uint8_t *smp_native = frame->data[ch] + i*stride;
+                uint8_t smp[4];
+                for(int j = 0; j<stride; j++)
+                    smp[j] = smp_native[stride-j-1];
+#else
+                uint8_t *smp = frame->data[ch] + i*stride;
+#endif
+                crc = av_crc(crc_tab, crc, smp+offset, bytes);
            }
        }

@@ -48,4 +48,3 @@ function ff_scalarproduct_int16_neon, export=1
        vmov.32         r0,  d3[0]
        bx              lr
 endfunc
-
@@ -229,7 +229,7 @@ A .endif
  .endif

        // Begin loop
-01:
+1:
  .if TOTAL_TAPS == 0
        // Things simplify a lot in this case
        // In fact this could be pipelined further if it's worth it...
@@ -241,7 +241,7 @@ A .endif
        str     ST0, [PST, #-4]!
        str     ST0, [PST, #4 * (MAX_BLOCKSIZE + MAX_FIR_ORDER)]
        str     ST0, [PSAMP], #4 * MAX_CHANNELS
-        bne     01b
+        bne     1b
  .else
    .if \fir_taps & 1
      .set LOAD_REG, 1
@@ -333,7 +333,7 @@ T       orr     AC0, AC0, AC1
        str     ST3, [PST, #-4]!
        str     ST2, [PST, #4 * (MAX_BLOCKSIZE + MAX_FIR_ORDER)]
        str     ST3, [PSAMP], #4 * MAX_CHANNELS
-        bne     01b
+        bne     1b
  .endif
        b       99f

@@ -279,11 +279,13 @@ function \type\()_8tap_\size\()h_\idx1\idx2
        sub             r1,  r1,  r5
 .endif
        @ size >= 16 loads two qwords and increments r2,
-        @ for size 4/8 it's enough with one qword and no
-        @ postincrement
+        @ size 4 loads 1 d word, increments r2 and loads 1 32-bit lane
+        @ for size 8 it's enough with one qword and no postincrement
 .if \size >= 16
        sub             r3,  r3,  r5
        sub             r3,  r3,  #8
+.elseif \size == 4
+        sub             r3,  r3,  #8
 .endif
        @ Load the filter vector
        vld1.16         {q0},  [r12,:128]
@@ -295,9 +297,14 @@ function \type\()_8tap_\size\()h_\idx1\idx2
 .if \size >= 16
        vld1.8          {d18, d19, d20}, [r2]!
        vld1.8          {d24, d25, d26}, [r7]!
-.else
+.elseif \size == 8
        vld1.8          {q9},  [r2]
        vld1.8          {q12}, [r7]
+.else @ size == 4
+        vld1.8          {d18}, [r2]!
+        vld1.8          {d24}, [r7]!
+        vld1.32         {d19[0]}, [r2]
+        vld1.32         {d25[0]}, [r7]
 .endif
        vmovl.u8        q8,  d18
        vmovl.u8        q9,  d19
@@ -362,6 +362,10 @@ static av_cold int atrac1_decode_init(AVCodecContext *avctx)
    ff_atrac_generate_tables();

    q->fdsp = avpriv_float_dsp_alloc(avctx->flags & AV_CODEC_FLAG_BITEXACT);
+    if (!q->fdsp) {
+        atrac1_decode_end(avctx);
+        return AVERROR(ENOMEM);
+    }

    q->bands[0] = q->low;
    q->bands[1] = q->mid;
@@ -1294,6 +1294,10 @@ typedef struct AVCodecContext {
     *   this callback and filled with the extra buffers if there are more
     *   buffers than buf[] can hold. extended_buf will be freed in
     *   av_frame_unref().
+     *   Decoders will generally initialize the whole buffer before it is output
+     *   but it can in rare error conditions happen that uninitialized data is passed
+     *   through. \important The buffers returned by get_buffer* should thus not contain sensitive
+     *   data.
     *
     * If AV_CODEC_CAP_DR1 is not set then get_buffer2() must call
     * avcodec_default_get_buffer2() instead of providing buffers allocated by
@@ -54,6 +54,8 @@ static av_cold int init(AVCodecContext *avctx)
        }

        a->mjpeg_avctx = avcodec_alloc_context3(codec);
+        if (!a->mjpeg_avctx)
+            return AVERROR(ENOMEM);

        av_dict_set(&thread_opt, "threads", "1", 0); // Is this needed ?
        a->mjpeg_avctx->refcounted_frames = 1;
@@ -169,5 +171,5 @@ AVCodec ff_avrn_decoder = {
    .close          = end,
    .decode         = decode_frame,
    .max_lowres     = 3,
-    .caps_internal  = FF_CODEC_CAP_INIT_THREADSAFE,
+    .caps_internal  = FF_CODEC_CAP_INIT_THREADSAFE | FF_CODEC_CAP_INIT_CLEANUP,
 };
@@ -867,7 +867,7 @@ static int binkb_decode_plane(BinkContext *c, AVFrame *frame, GetBitContext *gb,

    binkb_init_bundles(c);
    ref_start = frame->data[plane_idx];
-    ref_end   = frame->data[plane_idx] + (bh * frame->linesize[plane_idx] + bw) * 8;
+    ref_end   = frame->data[plane_idx] + ((bh - 1) * frame->linesize[plane_idx] + bw - 1) * 8;

    for (i = 0; i < 64; i++)
        coordmap[i] = (i & 7) + (i >> 3) * stride;
@@ -923,7 +923,7 @@ static int binkb_decode_plane(BinkContext *c, AVFrame *frame, GetBitContext *gb,
                xoff = binkb_get_value(c, BINKB_SRC_X_OFF);
                yoff = binkb_get_value(c, BINKB_SRC_Y_OFF) + ybias;
                ref = dst + xoff + yoff * stride;
-                if (ref < ref_start || ref + 8*stride > ref_end) {
+                if (ref < ref_start || ref > ref_end) {
                    av_log(c->avctx, AV_LOG_WARNING, "Reference block is out of bounds\n");
                } else if (ref + 8*stride < dst || ref >= dst + 8*stride) {
                    c->hdsp.put_pixels_tab[1][0](dst, ref, stride, 8);
@@ -939,7 +939,7 @@ static int binkb_decode_plane(BinkContext *c, AVFrame *frame, GetBitContext *gb,
                xoff = binkb_get_value(c, BINKB_SRC_X_OFF);
                yoff = binkb_get_value(c, BINKB_SRC_Y_OFF) + ybias;
                ref = dst + xoff + yoff * stride;
-                if (ref < ref_start || ref + 8 * stride > ref_end) {
+                if (ref < ref_start || ref > ref_end) {
                    av_log(c->avctx, AV_LOG_WARNING, "Reference block is out of bounds\n");
                } else if (ref + 8*stride < dst || ref >= dst + 8*stride) {
                    c->hdsp.put_pixels_tab[1][0](dst, ref, stride, 8);
@@ -971,7 +971,7 @@ static int binkb_decode_plane(BinkContext *c, AVFrame *frame, GetBitContext *gb,
                xoff = binkb_get_value(c, BINKB_SRC_X_OFF);
                yoff = binkb_get_value(c, BINKB_SRC_Y_OFF) + ybias;
                ref = dst + xoff + yoff * stride;
-                if (ref < ref_start || ref + 8 * stride > ref_end) {
+                if (ref < ref_start || ref > ref_end) {
                    av_log(c->avctx, AV_LOG_WARNING, "Reference block is out of bounds\n");
                } else if (ref + 8*stride < dst || ref >= dst + 8*stride) {
                    c->hdsp.put_pixels_tab[1][0](dst, ref, stride, 8);
@@ -1084,7 +1084,7 @@ static int bink_decode_plane(BinkContext *c, AVFrame *frame, GetBitContext *gb,
        for (bx = 0; bx < bw; bx++, dst += 8, prev += 8) {
            blk = get_value(c, BINK_SRC_BLOCK_TYPES);
            // 16x16 block type on odd line means part of the already decoded block, so skip it
-            if ((by & 1) && blk == SCALED_BLOCK) {
+            if (((by & 1) || (bx & 1)) && blk == SCALED_BLOCK) {
                bx++;
                dst  += 8;
                prev += 8;
@@ -1381,10 +1381,8 @@ static av_cold int decode_init(AVCodecContext *avctx)
    ff_hpeldsp_init(&c->hdsp, avctx->flags);
    ff_binkdsp_init(&c->binkdsp);

-    if ((ret = init_bundles(c)) < 0) {
-        free_bundles(c);
+    if ((ret = init_bundles(c)) < 0)
        return ret;
-    }

    if (c->version == 'b') {
        if (!binkb_initialised) {
@@ -1424,4 +1422,5 @@ AVCodec ff_bink_decoder = {
    .decode         = decode_frame,
    .flush          = flush,
    .capabilities   = AV_CODEC_CAP_DR1,
+    .caps_internal  = FF_CODEC_CAP_INIT_CLEANUP,
 };
@@ -40,8 +40,6 @@
 #include "rdft.h"
 #include "wma_freqs.h"

-static float quant_table[96];
-
 #define MAX_CHANNELS 2
 #define BINK_BLOCK_MAX_SIZE (MAX_CHANNELS << 11)

@@ -58,6 +56,7 @@ typedef struct BinkAudioContext {
    float root;
    DECLARE_ALIGNED(32, FFTSample, coeffs)[BINK_BLOCK_MAX_SIZE];
    float previous[MAX_CHANNELS][BINK_BLOCK_MAX_SIZE / 16];  ///< coeffs from previous audio block
+    float quant_table[96];
    AVPacket *pkt;
    union {
        RDFTContext rdft;
@@ -116,7 +115,7 @@ static av_cold int decode_init(AVCodecContext *avctx)
        s->root = s->frame_len / (sqrt(s->frame_len) * 32768.0);
    for (i = 0; i < 96; i++) {
        /* constant is result of 0.066399999/log10(M_E) */
-        quant_table[i] = expf(i * 0.15289164787221953823f) * s->root;
+        s->quant_table[i] = expf(i * 0.15289164787221953823f) * s->root;
    }

    /* calculate number of bands */
@@ -197,7 +196,7 @@ static int decode_block(BinkAudioContext *s, float **out, int use_dct)
            return AVERROR_INVALIDDATA;
        for (i = 0; i < s->num_bands; i++) {
            int value = get_bits(gb, 8);
-            quant[i]  = quant_table[FFMIN(value, 95)];
+            quant[i]  = s->quant_table[FFMIN(value, 95)];
        }

        k = 0;
@@ -129,7 +129,7 @@ static int alloc_table(VLC *vlc, int size, int use_static)

 typedef struct VLCcode {
    uint8_t bits;
-    uint16_t symbol;
+    VLC_TYPE symbol;
    /** codeword, with the first bit-to-be-read in the msb
     * (even if intended for a little-endian bitstream reader) */
    uint32_t code;
@@ -36,7 +36,7 @@ static int cbs_av1_read_uvlc(CodedBitstreamContext *ctx, GetBitContext *gbc,
        position = get_bits_count(gbc);

    zeroes = 0;
-    while (1) {
+    while (zeroes < 32) {
        if (get_bits_left(gbc) < 1) {
            av_log(ctx->log_ctx, AV_LOG_ERROR, "Invalid uvlc code at "
                   "%s: bitstream ended.\n", name);
@@ -49,7 +49,18 @@ static int cbs_av1_read_uvlc(CodedBitstreamContext *ctx, GetBitContext *gbc,
    }

    if (zeroes >= 32) {
-        value = MAX_UINT_BITS(32);
+        // The spec allows at least thirty-two zero bits followed by a
+        // one to mean 2^32-1, with no constraint on the number of
+        // zeroes.  The libaom reference decoder does not match this,
+        // instead reading thirty-two zeroes but not the following one
+        // to mean 2^32-1.  These two interpretations are incompatible
+        // and other implementations may follow one or the other.
+        // Therefore we reject thirty-two zeroes because the intended
+        // behaviour is not clear.
+        av_log(ctx->log_ctx, AV_LOG_ERROR, "Thirty-two zero bits in "
+               "%s uvlc code: considered invalid due to conflicting "
+               "standard and reference decoder behaviour.\n", name);
+        return AVERROR_INVALIDDATA;
    } else {
        if (get_bits_left(gbc) < zeroes) {
            av_log(ctx->log_ctx, AV_LOG_ERROR, "Invalid uvlc code at "
@@ -383,7 +394,7 @@ static int cbs_av1_write_increment(CodedBitstreamContext *ctx, PutBitContext *pb
    }

    if (len > 0)
-        put_bits(pbc, len, (1 << len) - 1 - (value != range_max));
+        put_bits(pbc, len, (1U << len) - 1 - (value != range_max));

    return 0;
 }
@@ -728,7 +728,7 @@ static int FUNC(sps_scc_extension)(CodedBitstreamContext *ctx, RWContext *rw,

        flag(sps_palette_predictor_initializer_present_flag);
        if (current->sps_palette_predictor_initializer_present_flag) {
-            ue(sps_num_palette_predictor_initializer_minus1, 0, 128);
+            ue(sps_num_palette_predictor_initializer_minus1, 0, 127);
            for (comp = 0; comp < (current->chroma_format_idc ? 3 : 1); comp++) {
                int bit_depth = comp == 0 ? current->bit_depth_luma_minus8 + 8
                                          : current->bit_depth_chroma_minus8 + 8;
@@ -166,13 +166,13 @@ static int cbs_jpeg_split_fragment(CodedBitstreamContext *ctx,
            }
        } else {
            i = start;
-            if (i + 2 > frag->data_size) {
+            if (i > frag->data_size - 2) {
                av_log(ctx->log_ctx, AV_LOG_ERROR, "Invalid JPEG image: "
                       "truncated at %02x marker.\n", marker);
                return AVERROR_INVALIDDATA;
            }
            length = AV_RB16(frag->data + i);
-            if (i + length > frag->data_size) {
+            if (length > frag->data_size - i) {
                av_log(ctx->log_ctx, AV_LOG_ERROR, "Invalid JPEG image: "
                       "truncated at %02x marker segment.\n", marker);
                return AVERROR_INVALIDDATA;
@@ -422,7 +422,7 @@ static int cbs_vp9_split_fragment(CodedBitstreamContext *ctx,
    superframe_header = frag->data[frag->data_size - 1];

    if ((superframe_header & 0xe0) == 0xc0) {
-        VP9RawSuperframeIndex sfi;
+        VP9RawSuperframeIndex sfi = {0};
        GetBitContext gbc;
        size_t index_size, pos;
        int i;
@@ -239,7 +239,7 @@ static void cdg_scroll(CDGraphicsContext *cc, uint8_t *data,
    for (y = FFMAX(0, vinc); y < FFMIN(CDG_FULL_HEIGHT + vinc, CDG_FULL_HEIGHT); y++)
        memcpy(out + FFMAX(0, hinc) + stride * y,
               in + FFMAX(0, hinc) - hinc + (y - vinc) * stride,
-               FFMIN(stride + hinc, stride));
+               FFABS(stride) - FFABS(hinc));

    if (vinc > 0)
        cdg_fill_wrapper(0, 0, out,
@@ -78,7 +78,7 @@ int64_t ff_dot_product(const int16_t *a, const int16_t *b, int length);
 *
 * @return value << offset, if offset>=0; value >> -offset - otherwise
 */
-static inline int bidir_sal(int value, int offset)
+static inline unsigned bidir_sal(unsigned value, int offset)
 {
    if(offset < 0) return value >> -offset;
    else           return value <<  offset;
@@ -665,8 +665,8 @@ static av_cold int clv_decode_init(AVCodecContext *avctx)
    }

    c->tile_shift = av_log2(c->tile_size);
-    if (1U << c->tile_shift != c->tile_size) {
-        av_log(avctx, AV_LOG_ERROR, "Tile size: %d, is not power of 2.\n", c->tile_size);
+    if (1U << c->tile_shift != c->tile_size || c->tile_shift < 1 || c->tile_shift > 30) {
+        av_log(avctx, AV_LOG_ERROR, "Tile size: %d, is not power of 2 > 1 and < 2^31\n", c->tile_size);
        return AVERROR_INVALIDDATA;
    }

@@ -91,4 +91,3 @@ AVCodec ff_cljr_decoder = {
    .decode         = decode_frame,
    .capabilities   = AV_CODEC_CAP_DR1,
 };
-
@@ -111,6 +111,7 @@ static int cpia_decode_frame(AVCodecContext *avctx,
        // Read line length, two byte little endian
        linelength = AV_RL16(src);
        src += 2;
+        src_size -= 2;

        if (src_size < linelength) {
            frame->decode_error_flags = FF_DECODE_ERROR_INVALID_BITSTREAM;
@@ -71,6 +71,9 @@ static int decode_frame(AVCodecContext *avctx, void *data, int *got_frame,
    int buf_size = avpkt->size;
    CamStudioContext *c = avctx->priv_data;
    int ret;
+    int bpp = avctx->bits_per_coded_sample / 8;
+    int bugdelta = FFALIGN(avctx->width * bpp, 4)       * avctx->height
+                 -        (avctx->width     & ~3) * bpp * avctx->height;

    if (buf_size < 2) {
        av_log(avctx, AV_LOG_ERROR, "coded frame too small\n");
@@ -84,7 +87,7 @@ static int decode_frame(AVCodecContext *avctx, void *data, int *got_frame,
    switch ((buf[0] >> 1) & 7) {
    case 0: { // lzo compression
        int outlen = c->decomp_size, inlen = buf_size - 2;
-        if (av_lzo1x_decode(c->decomp_buf, &outlen, &buf[2], &inlen) || outlen) {
+        if (av_lzo1x_decode(c->decomp_buf, &outlen, &buf[2], &inlen) || (outlen && outlen != bugdelta)) {
            av_log(avctx, AV_LOG_ERROR, "error during lzo decompression\n");
            return AVERROR_INVALIDDATA;
        }
@@ -93,7 +96,7 @@ static int decode_frame(AVCodecContext *avctx, void *data, int *got_frame,
    case 1: { // zlib compression
 #if CONFIG_ZLIB
        unsigned long dlen = c->decomp_size;
-        if (uncompress(c->decomp_buf, &dlen, &buf[2], buf_size - 2) != Z_OK || dlen != c->decomp_size) {
+        if (uncompress(c->decomp_buf, &dlen, &buf[2], buf_size - 2) != Z_OK || (dlen != c->decomp_size && dlen != c->decomp_size - bugdelta)) {
            av_log(avctx, AV_LOG_ERROR, "error during zlib decompression\n");
            return AVERROR_INVALIDDATA;
        }
@@ -328,7 +328,7 @@ static void dmix_add_c(int32_t *dst, const int32_t *src, int coeff, ptrdiff_t le
    int i;

    for (i = 0; i < len; i++)
-        dst[i] += mul15(src[i], coeff);
+        dst[i] += (unsigned)mul15(src[i], coeff);
 }

 static void dmix_scale_c(int32_t *dst, int scale, ptrdiff_t len)
@@ -1431,8 +1431,8 @@ static void global_mv(DiracContext *s, DiracBlock *block, int x, int y, int ref)
    int *c      = s->globalmc[ref].perspective;

    int64_t m   = (1<<ep) - (c[0]*(int64_t)x + c[1]*(int64_t)y);
-    int64_t mx  = m * (int64_t)((A[0][0] * (int64_t)x + A[0][1]*(int64_t)y) + (1LL<<ez) * b[0]);
-    int64_t my  = m * (int64_t)((A[1][0] * (int64_t)x + A[1][1]*(int64_t)y) + (1LL<<ez) * b[1]);
+    int64_t mx  = m * (uint64_t)((A[0][0] * (int64_t)x + A[0][1]*(int64_t)y) + (1LL<<ez) * b[0]);
+    int64_t my  = m * (uint64_t)((A[1][0] * (int64_t)x + A[1][1]*(int64_t)y) + (1LL<<ez) * b[1]);

    block->u.mv[ref][0] = (mx + (1<<(ez+ep))) >> (ez+ep);
    block->u.mv[ref][1] = (my + (1<<(ez+ep))) >> (ez+ep);
@@ -195,7 +195,7 @@ static void dequant_subband_ ## PX ## _c(uint8_t *src, uint8_t *dst, ptrdiff_t s
 {                                                                                          \
    int i, y;                                                                              \
    for (y = 0; y < tot_v; y++) {                                                          \
-        PX c, sign, *src_r = (PX *)src, *dst_r = (PX *)dst;                                \
+        PX c, *src_r = (PX *)src, *dst_r = (PX *)dst;                                      \
        for (i = 0; i < tot_h; i++) {                                                      \
            c = *src_r++;                                                                  \
            if     (c < 0) c = -((-(unsigned)c*qf + qs) >> 2);                             \
@@ -110,6 +110,7 @@ static av_cold int dnxhd_decode_init(AVCodecContext *avctx)

 static int dnxhd_init_vlc(DNXHDContext *ctx, uint32_t cid, int bitdepth)
 {
+    int ret;
    if (cid != ctx->cid) {
        int index;

@@ -129,19 +130,26 @@ static int dnxhd_init_vlc(DNXHDContext *ctx, uint32_t cid, int bitdepth)
        ff_free_vlc(&ctx->dc_vlc);
        ff_free_vlc(&ctx->run_vlc);

-        init_vlc(&ctx->ac_vlc, DNXHD_VLC_BITS, 257,
+        if ((ret = init_vlc(&ctx->ac_vlc, DNXHD_VLC_BITS, 257,
                 ctx->cid_table->ac_bits, 1, 1,
-                 ctx->cid_table->ac_codes, 2, 2, 0);
-        init_vlc(&ctx->dc_vlc, DNXHD_DC_VLC_BITS, bitdepth > 8 ? 14 : 12,
+                 ctx->cid_table->ac_codes, 2, 2, 0)) < 0)
+            goto out;
+        if ((ret = init_vlc(&ctx->dc_vlc, DNXHD_DC_VLC_BITS, bitdepth > 8 ? 14 : 12,
                 ctx->cid_table->dc_bits, 1, 1,
-                 ctx->cid_table->dc_codes, 1, 1, 0);
-        init_vlc(&ctx->run_vlc, DNXHD_VLC_BITS, 62,
+                 ctx->cid_table->dc_codes, 1, 1, 0)) < 0)
+            goto out;
+        if ((ret = init_vlc(&ctx->run_vlc, DNXHD_VLC_BITS, 62,
                 ctx->cid_table->run_bits, 1, 1,
-                 ctx->cid_table->run_codes, 2, 2, 0);
+                 ctx->cid_table->run_codes, 2, 2, 0)) < 0)
+            goto out;

        ctx->cid = cid;
    }
-    return 0;
+    ret = 0;
+out:
+    if (ret < 0)
+        av_log(ctx->avctx, AV_LOG_ERROR, "init_vlc failed\n");
+    return ret;
 }

 static int dnxhd_get_profile(int cid)
@@ -206,6 +206,9 @@ static int decode_frame(AVCodecContext *avctx,
        return AVERROR_PATCHWELCOME;
    }

+    if (bits_per_color > 32)
+        return AVERROR_INVALIDDATA;
+
    buf += 820;
    avctx->sample_aspect_ratio.num = read32(&buf, endian);
    avctx->sample_aspect_ratio.den = read32(&buf, endian);
@@ -214,7 +214,7 @@ static uint8_t prob_dst_x_bit(int c)
    return (ff_reverse[c & 127] >> 1) + 1;
 }

-static void build_filter(int16_t table[DST_MAX_ELEMENTS][16][256], const Table *fsets)
+static int build_filter(int16_t table[DST_MAX_ELEMENTS][16][256], const Table *fsets)
 {
    int i, j, k, l;

@@ -225,14 +225,17 @@ static void build_filter(int16_t table[DST_MAX_ELEMENTS][16][256], const Table *
            int total = av_clip(length - j * 8, 0, 8);

            for (k = 0; k < 256; k++) {
-                int v = 0;
+                int64_t v = 0;

                for (l = 0; l < total; l++)
                    v += (((k >> l) & 1) * 2 - 1) * fsets->coeff[i][j * 8 + l];
+                if ((int16_t)v != v)
+                    return AVERROR_INVALIDDATA;
                table[i][j][k] = v;
            }
        }
    }
+    return 0;
 }

 static int decode_frame(AVCodecContext *avctx, void *data,
@@ -328,7 +331,9 @@ static int decode_frame(AVCodecContext *avctx, void *data,
        return AVERROR_INVALIDDATA;
    ac_init(ac, gb);

-    build_filter(s->filter, &s->fsets);
+    ret = build_filter(s->filter, &s->fsets);
+    if (ret < 0)
+        return ret;

    memset(s->status, 0xAA, sizeof(s->status));
    memset(dsd, 0, frame->nb_samples * 4 * channels);
@@ -259,4 +259,3 @@ av_cold int ff_dvvideo_init(AVCodecContext *avctx)

    return 0;
 }
-
@@ -338,4 +338,3 @@ const AVDVProfile *av_dv_codec_profile2(int width, int height,

    return p;
 }
-
@@ -215,8 +215,8 @@ static av_always_inline int dv_guess_dct_mode(DVVideoContext *s, uint8_t *data,
    if (s->avctx->flags & AV_CODEC_FLAG_INTERLACED_DCT) {
        int ps = s->ildct_cmp(NULL, data, NULL, linesize, 8) - 400;
        if (ps > 0) {
-            int is = s->ildct_cmp(NULL, data,            NULL, linesize << 1, 4) +
-                     s->ildct_cmp(NULL, data + linesize, NULL, linesize << 1, 4);
+            int is = s->ildct_cmp(NULL, data,            NULL, linesize * 2, 4) +
+                     s->ildct_cmp(NULL, data + linesize, NULL, linesize * 2, 4);
            return ps > is;
        }
    }
@@ -511,7 +511,7 @@ static av_always_inline int dv_init_enc_block(EncBlockInfo* bi, uint8_t *data, i

    if (data) {
        if (DV_PROFILE_IS_HD(s->sys)) {
-            s->get_pixels(blk, data, linesize << bi->dct_mode);
+            s->get_pixels(blk, data, linesize * (1 << bi->dct_mode));
            s->fdct[0](blk);
        } else {
            bi->dct_mode = dv_guess_dct_mode(s, data, linesize);
@@ -860,7 +860,7 @@ static int dv_encode_video_segment(AVCodecContext *avctx, void *arg)

        qnos[mb_index] = DV_PROFILE_IS_HD(s->sys) ? 1 : 15;

-        y_ptr    = s->frame->data[0] + ((mb_y * s->frame->linesize[0] + mb_x) << 3);
+        y_ptr    = s->frame->data[0] + (mb_y * s->frame->linesize[0] + mb_x) * 8;
        linesize = s->frame->linesize[0];

        if (s->sys->height == 1080 && mb_y < 134)
@@ -874,12 +874,12 @@ static int dv_encode_video_segment(AVCodecContext *avctx, void *arg)
        if ((s->sys->pix_fmt == AV_PIX_FMT_YUV420P)                      ||
            (s->sys->pix_fmt == AV_PIX_FMT_YUV411P && mb_x >= (704 / 8)) ||
            (s->sys->height >= 720 && mb_y != 134)) {
-            y_stride = s->frame->linesize[0] << (3*!enc_blk->dct_mode);
+            y_stride = s->frame->linesize[0] * (1 << (3*!enc_blk->dct_mode));
        } else {
            y_stride = 16;
        }
        y_ptr    = s->frame->data[0] +
-                   ((mb_y * s->frame->linesize[0] + mb_x) << 3);
+                   (mb_y * s->frame->linesize[0] + mb_x) * 8;
        linesize = s->frame->linesize[0];

        if (s->sys->video_stype == 4) { /* SD 422 */
@@ -898,17 +898,17 @@ static int dv_encode_video_segment(AVCodecContext *avctx, void *arg)
        enc_blk += 4;

        /* initializing chrominance blocks */
-        c_offset = (((mb_y >>  (s->sys->pix_fmt == AV_PIX_FMT_YUV420P)) * s->frame->linesize[1] +
-                     (mb_x >> ((s->sys->pix_fmt == AV_PIX_FMT_YUV411P) ? 2 : 1))) << 3);
+        c_offset = ((mb_y >>  (s->sys->pix_fmt == AV_PIX_FMT_YUV420P)) * s->frame->linesize[1] +
+                    (mb_x >> ((s->sys->pix_fmt == AV_PIX_FMT_YUV411P) ? 2 : 1))) * 8;
        for (j = 2; j; j--) {
            uint8_t *c_ptr = s->frame->data[j] + c_offset;
            linesize = s->frame->linesize[j];
-            y_stride = (mb_y == 134) ? 8 : (s->frame->linesize[j] << (3*!enc_blk->dct_mode));
+            y_stride = (mb_y == 134) ? 8 : (s->frame->linesize[j] * (1 << (3*!enc_blk->dct_mode)));
            if (s->sys->pix_fmt == AV_PIX_FMT_YUV411P && mb_x >= (704 / 8)) {
                uint8_t *d;
                uint8_t *b = scratch;
                for (i = 0; i < 8; i++) {
-                    d      = c_ptr + (linesize << 3);
+                    d      = c_ptr + linesize * 8;
                    b[0]   = c_ptr[0];
                    b[1]   = c_ptr[1];
                    b[2]   = c_ptr[2];
@@ -440,7 +440,7 @@ static int get_opcodes(GetByteContext *gb, uint32_t *table, uint8_t *dst, int op

    size_in_bits = bytestream2_get_le32(gb);
    endoffset = ((size_in_bits + 7) >> 3) - 4;
-    if (endoffset <= 0 || bytestream2_get_bytes_left(gb) < endoffset)
+    if ((int)endoffset <= 0 || bytestream2_get_bytes_left(gb) < endoffset)
        return AVERROR_INVALIDDATA;

    offset = endoffset;
@@ -111,7 +111,7 @@ static int dxva_get_decoder_configuration(AVCodecContext *avctx,

    for (i = 0; i < cfg_count; i++) {
        unsigned score;
-        UINT ConfigBitstreamRaw;
+        UINT ConfigBitstreamRaw = 0;
        GUID guidConfigBitstreamEncryption;

 #if CONFIG_D3D11VA
@@ -262,7 +262,7 @@ static int dxva_get_decoder_guid(AVCodecContext *avctx, void *service, void *sur
    *decoder_guid = ff_GUID_NULL;
    for (i = 0; dxva_modes[i].guid; i++) {
        const dxva_mode *mode = &dxva_modes[i];
-        int validate;
+        int validate = 0;
        if (!dxva_check_codec_compatibility(avctx, mode))
            continue;

@@ -794,7 +794,7 @@ int ff_dxva2_commit_buffer(AVCodecContext *avctx,
                           unsigned type, const void *data, unsigned size,
                           unsigned mb_count)
 {
-    void     *dxva_data;
+    void     *dxva_data = NULL;
    unsigned dxva_size;
    int      result;
    HRESULT hr = 0;
@@ -816,7 +816,7 @@ int ff_dxva2_commit_buffer(AVCodecContext *avctx,
               type, (unsigned)hr);
        return -1;
    }
-    if (size <= dxva_size) {
+    if (dxva_data && size <= dxva_size) {
        memcpy(dxva_data, data, size);

 #if CONFIG_D3D11VA
@@ -894,7 +894,7 @@ int ff_dxva2_common_end_frame(AVCodecContext *avctx, AVFrame *frame,
 #endif
    DECODER_BUFFER_DESC             *buffer = NULL, *buffer_slice = NULL;
    int result, runs = 0;
-    HRESULT hr;
+    HRESULT hr = -1;
    unsigned type;
    FFDXVASharedContext *sctx = DXVA_SHARED_CONTEXT(avctx);

@@ -139,9 +139,11 @@ static void ff_eac3_apply_spectral_extension(AC3DecodeContext *s)
            // spx_noise_blend and spx_signal_blend are both FP.23
            nscale *= 1.0 / (1<<23);
            sscale *= 1.0 / (1<<23);
+            if (nscale < -1.0)
+                nscale = -1.0;
 #endif
            for (i = 0; i < s->spx_band_sizes[bnd]; i++) {
-                float noise  = nscale * (int32_t)av_lfg_get(&s->dith_state);
+                UINTFLOAT noise = (INTFLOAT)(nscale * (int32_t)av_lfg_get(&s->dith_state));
                s->transform_coeffs[ch][bin]   *= sscale;
                s->transform_coeffs[ch][bin++] += noise;
            }
@@ -198,12 +198,15 @@ static int cmv_decode_frame(AVCodecContext *avctx,
    if ((ret = av_image_check_size(s->width, s->height, 0, s->avctx)) < 0)
        return ret;

+    buf += EA_PREAMBLE_SIZE;
+    if (!(buf[0]&1) && buf_end - buf < s->width * s->height * (int64_t)(100 - s->avctx->discard_damaged_percentage) / 100)
+        return AVERROR_INVALIDDATA;
+
    if ((ret = ff_get_buffer(avctx, frame, AV_GET_BUFFER_FLAG_REF)) < 0)
        return ret;

    memcpy(frame->data[1], s->palette, AVPALETTE_SIZE);

-    buf += EA_PREAMBLE_SIZE;
    if ((buf[0]&1)) {  // subtype
        cmv_decode_inter(s, frame, buf+2, buf_end);
        frame->key_frame = 0;
@@ -58,7 +58,7 @@ static av_cold int tgq_decode_init(AVCodecContext *avctx)
    return 0;
 }

-static void tgq_decode_block(TgqContext *s, int16_t block[64], GetBitContext *gb)
+static int tgq_decode_block(TgqContext *s, int16_t block[64], GetBitContext *gb)
 {
    uint8_t *perm = s->scantable.permutated;
    int i, j, value;
@@ -66,6 +66,8 @@ static void tgq_decode_block(TgqContext *s, int16_t block[64], GetBitContext *gb
    for (i = 1; i < 64;) {
        switch (show_bits(gb, 3)) {
        case 4:
+            if (i >= 63)
+                return AVERROR_INVALIDDATA;
            block[perm[i++]] = 0;
        case 0:
            block[perm[i++]] = 0;
@@ -75,6 +77,8 @@ static void tgq_decode_block(TgqContext *s, int16_t block[64], GetBitContext *gb
        case 1:
            skip_bits(gb, 2);
            value = get_bits(gb, 6);
+            if (value > 64 - i)
+                return AVERROR_INVALIDDATA;
            for (j = 0; j < value; j++)
                block[perm[i++]] = 0;
            break;
@@ -102,6 +106,7 @@ static void tgq_decode_block(TgqContext *s, int16_t block[64], GetBitContext *gb
        }
    }
    block[0] += 128 << 4;
+    return 0;
 }

 static void tgq_idct_put_mb(TgqContext *s, int16_t (*block)[64], AVFrame *frame,
@@ -161,8 +166,11 @@ static int tgq_decode_mb(TgqContext *s, AVFrame *frame, int mb_y, int mb_x)
        if (ret < 0)
            return ret;

-        for (i = 0; i < 6; i++)
-            tgq_decode_block(s, s->block[i], &gb);
+        for (i = 0; i < 6; i++) {
+            int ret = tgq_decode_block(s, s->block[i], &gb);
+            if (ret < 0)
+                return ret;
+        }
        tgq_idct_put_mb(s, s->block, frame, mb_x, mb_y);
        bytestream2_skip(&s->gb, mode);
    } else {
@@ -88,11 +88,6 @@ static CodeBook unpack_codebook(GetBitContext* gb, unsigned depth,
    unsigned i, j;
    CodeBook cb = { 0 };

-    if (size >= INT_MAX / 34 || get_bits_left(gb) < size * 34)
-        return cb;
-
-    if (size >= INT_MAX / sizeof(MacroBlock))
-        return cb;
    cb.blocks = av_malloc(size ? size * sizeof(MacroBlock) : 1);
    if (!cb.blocks)
        return cb;
@@ -162,7 +157,7 @@ static MacroBlock decode_macroblock(Escape124Context* s, GetBitContext* gb,

    // This condition can occur with invalid bitstreams and
    // *codebook_index == 2
-    if (block_index >= s->codebooks[*codebook_index].size)
+    if (block_index >= s->codebooks[*codebook_index].size || !s->codebooks[*codebook_index].blocks)
        return (MacroBlock) { { 0 } };

    return s->codebooks[*codebook_index].blocks[block_index];
@@ -226,7 +221,7 @@ static int escape124_decode_frame(AVCodecContext *avctx,
    // represent a lower bound of the space needed for skipped superblocks. Non
    // skipped SBs need more space.
    if (get_bits_left(&gb) < 64 + s->num_superblocks * 23LL / 4320)
-        return -1;
+        return AVERROR_INVALIDDATA;

    frame_flags = get_bits_long(&gb, 32);
    frame_size  = get_bits_long(&gb, 32);
@@ -243,7 +238,7 @@ static int escape124_decode_frame(AVCodecContext *avctx,
        if ((ret = av_frame_ref(frame, s->frame)) < 0)
            return ret;

-        return frame_size;
+        return 0;
    }

    for (i = 0; i < 3; i++) {
@@ -277,9 +272,14 @@ static int escape124_decode_frame(AVCodecContext *avctx,
            }

            av_freep(&s->codebooks[i].blocks);
+            if (cb_size >= INT_MAX / 34 || get_bits_left(&gb) < (int)cb_size * 34)
+                return AVERROR_INVALIDDATA;
+
+            if (cb_size >= INT_MAX / sizeof(MacroBlock))
+                return AVERROR_INVALIDDATA;
            s->codebooks[i] = unpack_codebook(&gb, cb_depth, cb_size);
            if (!s->codebooks[i].blocks)
-                return -1;
+                return AVERROR(ENOMEM);
        }
    }

@@ -372,7 +372,7 @@ static int escape124_decode_frame(AVCodecContext *avctx,

    *got_frame = 1;

-    return frame_size;
+    return 0;
 }


@@ -1511,6 +1511,7 @@ static int decode_header(EXRContext *s, AVFrame *frame)
            ymax   = bytestream2_get_le32(&s->gb);

            if (xmin > xmax || ymin > ymax ||
+                ymax == INT_MAX || xmax == INT_MAX ||
                (unsigned)xmax - xmin >= INT_MAX ||
                (unsigned)ymax - ymin >= INT_MAX) {
                ret = AVERROR_INVALIDDATA;
@@ -1825,13 +1826,14 @@ static int decode_frame(AVCodecContext *avctx, void *data,
    avctx->execute2(avctx, decode_block, s->thread_data, NULL, nb_blocks);

    // Zero out the end if ymax+1 is not h
-    for (i = 0; i < planes; i++) {
-        ptr = picture->data[i] + ((s->ymax+1) * picture->linesize[i]);
-        for (y = s->ymax + 1; y < avctx->height; y++) {
-            memset(ptr, 0, out_line_size);
-            ptr += picture->linesize[i];
+    if ((s->ymax+1) < avctx->height)
+        for (i = 0; i < planes; i++) {
+            ptr = picture->data[i] + ((s->ymax+1) * picture->linesize[i]);
+            for (y = s->ymax + 1; y < avctx->height; y++) {
+                memset(ptr, 0, out_line_size);
+                ptr += picture->linesize[i];
+            }
        }
-    }

    picture->pict_type = AV_PICTURE_TYPE_I;
    *got_frame = 1;
@@ -141,6 +141,8 @@ static int decode_uncompressed(AVCodecContext *avctx, GetBitContext *gb,
                return AVERROR_INVALIDDATA;
            }
            cwi = 10 - av_log2(cwi);
+            if (get_bits_left(gb) < cwi + 1)
+                return AVERROR_INVALIDDATA;
            skip_bits(gb, cwi + 1);
            if (cwi > 5) {
                newmode = get_bits1(gb);
@@ -206,6 +208,8 @@ static int decode_group3_1d_line(AVCodecContext *avctx, GetBitContext *gb,
    unsigned int run = 0;
    unsigned int t;
    for (;;) {
+        if (get_bits_left(gb) <= 0)
+            return AVERROR_INVALIDDATA;
        t    = get_vlc2(gb, ccitt_vlc[mode].table, 9, 2);
        run += t;
        if (t < 64) {
@@ -224,7 +228,7 @@ static int decode_group3_1d_line(AVCodecContext *avctx, GetBitContext *gb,
            run       = 0;
            mode      = !mode;
        } else if ((int)t == -1) {
-            if (show_bits(gb, 12) == 15) {
+            if (get_bits_left(gb) > 12 && show_bits(gb, 12) == 15) {
                int ret;
                skip_bits(gb, 12);
                ret = decode_uncompressed(avctx, gb, &pix_left, &runs, runend, &mode);
@@ -251,7 +255,10 @@ static int decode_group3_2d_line(AVCodecContext *avctx, GetBitContext *gb,
    unsigned int offs = 0, run = 0;

    while (offs < width) {
-        int cmode = get_vlc2(gb, ccitt_group3_2d_vlc.table, 9, 1);
+        int cmode;
+        if (get_bits_left(gb) <= 0)
+            return AVERROR_INVALIDDATA;
+        cmode = get_vlc2(gb, ccitt_group3_2d_vlc.table, 9, 1);
        if (cmode == -1) {
            av_log(avctx, AV_LOG_ERROR, "Incorrect mode VLC\n");
            return AVERROR_INVALIDDATA;
@@ -273,6 +280,8 @@ static int decode_group3_2d_line(AVCodecContext *avctx, GetBitContext *gb,
            for (k = 0; k < 2; k++) {
                run = 0;
                for (;;) {
+                    if (get_bits_left(gb) <= 0)
+                        return AVERROR_INVALIDDATA;
                    t = get_vlc2(gb, ccitt_vlc[mode].table, 9, 2);
                    if (t == -1) {
                        av_log(avctx, AV_LOG_ERROR, "Incorrect code\n");
@@ -296,7 +305,10 @@ static int decode_group3_2d_line(AVCodecContext *avctx, GetBitContext *gb,
                mode = !mode;
            }
        } else if (cmode == 9 || cmode == 10) {
-            int xxx = get_bits(gb, 3);
+            int xxx;
+            if (get_bits_left(gb) < 3)
+                return AVERROR_INVALIDDATA;
+            xxx = get_bits(gb, 3);
            if (cmode == 9 && xxx == 7) {
                int ret;
                int pix_left = width - offs;
@@ -51,6 +51,8 @@
 * @file
 * definitions and initialization of LUT table for FFT
 */
+#include "libavutil/thread.h"
+
 #include "libavcodec/fft_table.h"

 const int32_t ff_w_tab_sr[MAX_FFT_SIZE/(4*16)] = {
@@ -314,15 +316,29 @@ const int32_t ff_w_tab_sr[MAX_FFT_SIZE/(4*16)] = {

 uint16_t ff_fft_offsets_lut[21845];

-void ff_fft_lut_init(uint16_t *table, int off, int size, int *index)
+static void fft_lut_init(uint16_t *table, int off, int size, int *index)
 {
    if (size < 16) {
        table[*index] = off >> 2;
        (*index)++;
    }
    else {
-        ff_fft_lut_init(table, off, size>>1, index);
-        ff_fft_lut_init(table, off+(size>>1), size>>2, index);
-        ff_fft_lut_init(table, off+3*(size>>2), size>>2, index);
+        fft_lut_init(table, off,                   size >> 1, index);
+        fft_lut_init(table, off +     (size >> 1), size >> 2, index);
+        fft_lut_init(table, off + 3 * (size >> 2), size >> 2, index);
    }
 }
+
+static void fft_lut_init_start(void)
+{
+    int n = 0;
+
+    fft_lut_init(ff_fft_offsets_lut, 0, 1 << 17, &n);
+}
+
+void ff_fft_lut_init(void)
+{
+    static AVOnce init_once = AV_ONCE_INIT;
+
+    ff_thread_once(&init_once, fft_lut_init_start);
+}
@@ -61,6 +61,6 @@

 extern const int32_t ff_w_tab_sr[];
 extern uint16_t ff_fft_offsets_lut[];
-void ff_fft_lut_init(uint16_t *table, int off, int size, int *index);
+void ff_fft_lut_init(void);

 #endif /* AVCODEC_FFT_TABLE_H */
--- a/Show More
+++ b/Show More
@@ -1 +1 @@
 .3.2
 .3.9