ffmpeg: Upgrade 9.0 -> 9.0.1
PATCH 1/2 ffmpeg: Upgrade 9.0 -> 9.0.1
+version 9.0.1:
+ avformat/webp_anim_dec: use ffio_read_size for ICCP and EXIF chunks
+ avformat/rtpenc_av1: do not narrow the OBU size to (long)
+ avformat/rtpenc_av1: Check num_lebs
+ avformat/rtpenc_av1: bound OBU size in the keyframe search loop
+ avformat/gopher: Fix CRLF injection
+ avformat/rtpenc_vc2hq: reject data units larger than the RTP payload buffer
+ avformat/dashdec: reject a negative fragment index
+ avformat/hevc: reject hvcC NAL arrays that overflow the 16-bit count
+ avformat/mpegenc: reject stream counts that overflow the system header
+ avformat/mpegenc: pass buffer size into put_system_header()
+ avformat/librist: honor the caller buffer size in librist_read
+ avformat/scd: reject zero-channel tracks
+ avcodec/dolby_e: Add error recovery when parse_mantissas run out of bits
+ avfilter/af_pan: check the id of named input channels before use
+ swscale: avoid overflow in fast bilinear edge handling
+ avformat/shared: Use correct printf specifier
+ Revert "lavfi/bwdif: fix heap-buffer-overflow with small height videos"
+ avfilter/vf_bwdif: fix line boundary checks for >8 bits content
+ avformat/rawutils: reject raw RGB frames that do not fit an AVPacket
+ avcodec/nvenc: write AV1 timecode metadata in AV1 syntax
+ avcodec/utils: add ff_alloc_timecode_metadata_av1()
+ avcodec/utils: factor the timecode fields out of ff_alloc_timecode_sei
+ vulkan_encode_av1: set primary_ref_frame to a reference name, not a slot
+ avformat/mov: reject a trun sample count the input cannot hold
+ RELEASE_NOTES: Based on the version from 8.0
+ avformat/mp3enc: fix underflow of the LAME encoder delay
+ avformat/os_support: fix return value of win32_rename
+ vulkan_encode: fix leak and swallowed errors in init_base_units()
+ avfilter/dnn: fix async teardown race condition in all backends
+ avcodec/tiff: reject inflate output shorter than the strip
+ avformat/dashdec: check NULL pointer before use str_end_offset
+ avformat/dashdec: check NULL pointer of av_strtok value before use it
+ avfilter/vf_scale_cuda: fix non-scaling format conversion
+ fftools/opt_common: fix format string bug in print_program_info
+ avformat/dashdec: fix integer truncation in calc_max_seg_no()
+ swscale/x86: fix SIGILL in u8 SCALE on SSE4-only CPUs
+ avdevice/android_camera: fix OOB read in metadata parsing
+ avformat/dashdec: don't stop at the first input EOF
+ avformat/iamf_parse: check that num_sub_mixes and num_audio_elements in Mix Presentations are not zero
+ avformat/iamf_parse: bound the output mix gain duration by the audio elements
+ avcodec/cbs_h266: size vps_direct_ref_layer_flag for the full layer range
+ avcodec/vulkan/ffv1_dec_setup: act on the slice header rejection
+ avcodec/ffv1dec: reject a remap that produces zero entries
+ avcodec/ffv1dec: mark the slice damaged when its remap fails
+ avcodec/vulkan/ffv1_dec_setup: bound the fltmap write
+ avcodec/vulkan/ffv1_dec_setup: test mul_count as unsigned
+ avcodec/vulkan/ffv1_dec_setup: reject a remap that produces zero entries
+ avcodec/pgssubdec: always give an output rect a palette
+ scale_d3d11: Fix hw_frame_ctx reference leak
+ avcodec/cfhd: reject transform-2 output wider than the plane
+ avformat/rtsp: clear authentication on cross-origin redirects
+ avcodec/dovi_rpuenc: normalize vdr_dm_metadata_present to 0/1
+ avcodec/dovi_rpuenc: validate vdr_rpu_id from the input metadata
+ avcodec/bsf/dovi_rpu: handle update_rpu() returning no RPU
+ avcodec/dovi_rpudec: bound num_x/y_partitions
+ avcodec/dovi_rpuenc: validate the data mapping before generation
+ avfilter/af_arnndn: pad the DCT input buffers to the read length
+ doc/nut.texi: point at the latest spec in the git repository
+ avformat/mov: bound sgpd sync entry_count by the atom size
+ avcodec/rscc: do not leave uninitilized data when the input is too short
+ avcodec/dvbsub_parser: avoid signed overflow in the capacity check
+ avformat/tls_openssl: bind peer identity for numeric-IP verify
+ avformat/codec2: avoid integer overflow in packet size and duration
+ avfilter/vf_xpsnr: avoid a zero block size on small frames
+ avcodec/cbs_av1: pad the ITU-T T.35 payload buffer
+ avcodec/screenpresso: reject deflate output shorter than the frame
+ avformat/hls: Enforce protocol checks when opening child playlists
+ avformat/hlsenc: Fix heap buffer overflow in parse_playlist()
+ avformat/hlsenc: Handle extensionless URIs in extract_segment_number()
+ avfilter/vf_lut3d: do not compute size*size before the size is validated
+ avfilter/vf_hqdn3d: support dynamic frame sizes
+ avfilter/vf_hqdn3d: reject unsupported frame parameter changes
+ avcodec/bsf/truehd_core: clear profile value on init()
+ avcodec/bsf/eac3_core: clear profile value on init()
+ avformat/iamf_parse: fix inverted subblock duration validation
+ avcodec/get_buffer: use frame pixel format instead of context
+ avformat/lcevc: add a log context parameter to all functions
---
.../ffmpeg/{ffmpeg_9.0.bb => ffmpeg_9.0.1.bb} | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
rename meta/recipes-multimedia/ffmpeg/{ffmpeg_9.0.bb => ffmpeg_9.0.1.bb} (99%)meta/recipes-multimedia/ffmpeg/ffmpeg_9.0.1.bb renamed from meta/recipes-multimedia/ffmpeg/ffmpeg_9.0.bb +1 -1
@@ -26,7 +26,7 @@ SRC_URI = "https://www.ffmpeg.org/releases/${BP}.tar.xz \ file://0002-ffbuild-common.mak-ensure-target-directories-are-cre.patch \ " -SRC_URI[sha256sum] = "7f607a00dd0d28a729d5a4811205812eef01cf6ef6155025febb6f36a9062d52" +SRC_URI[sha256sum] = "cf38e0e28c7e5605942c4a77755349b0145804a397af37eb1fb4c77cb237f635" # Build fails when thumb is enabled: https://bugzilla.yoctoproject.org/show_bug.cgi?id=7717 ARM_INSTRUCTION_SET:armv4 = "arm"
PATCH 2/2 ffmpeg: Fix makefile directory creation race
The recent upgrade exposed another make race in ffmpeg. Add a patch which resolves it, submitted upstream (pending moderator approval).
--- .../ffmpeg/ffmpeg/makerace.patch | 29 +++++++++++++++++++ .../recipes-multimedia/ffmpeg/ffmpeg_9.0.1.bb | 1 + 2 files changed, 30 insertions(+) create mode 100644 meta/recipes-multimedia/ffmpeg/ffmpeg/makerace.patch
meta/recipes-multimedia/ffmpeg/ffmpeg/makerace.patch new file +29 -0
@@ -0,0 +1,29 @@ +An easy to reproduce make race: + +$ make libswscale/x86/uops_macros.gen.asm +HOSTCC libswscale/x86/uops_macros.gen.asm +cc1: fatal error: opening output file libswscale/x86/uops_macros.gen.asm: No such file or directory + +It needs the missing directory creation dependency adding. + +Upstream-Status: Submitted [Sent to https://lists.ffmpeg.org/archives/list/[email protected]/ 2026/9/11] +Signed-off-by: Richard Purdie <[email protected]> + +Index: ffmpeg-9.0.1/libswscale/x86/Makefile +=================================================================== +--- ffmpeg-9.0.1.orig/libswscale/x86/Makefile ++++ ffmpeg-9.0.1/libswscale/x86/Makefile +@@ -1,3 +1,5 @@ ++OUTDIRS += $(SUBDIR)x86 ++ + OBJS += x86/rgb2rgb.o \ + x86/swscale.o \ + x86/yuv2rgb.o \ +@@ -31,6 +33,6 @@ $(SUBDIR)x86/ops_common.o: $(SUBDIR)x86/ + $(SUBDIR)x86/ops_int.o: $(SUBDIR)x86/uops_macros.gen.asm + $(SUBDIR)x86/ops_float.o: $(SUBDIR)x86/uops_macros.gen.asm + $(SUBDIR)x86/uops_macros.gen.asm: $(SRC_PATH)/libswscale/x86/uops_macros.asm.h \ +- $(SRC_PATH)/libswscale/uops_macros.h ++ $(SRC_PATH)/libswscale/uops_macros.h | $(SUBDIR)x86 + $(HOSTCC) $(HOSTCC_E) $(CPPFLAGS) $< + endif
meta/recipes-multimedia/ffmpeg/ffmpeg_9.0.1.bb +1 -0
@@ -24,6 +24,7 @@ LIC_FILES_CHKSUM = "file://COPYING.GPLv2;md5=b234ee4d69f5fce4486a80fdaf4a4263 \ SRC_URI = "https://www.ffmpeg.org/releases/${BP}.tar.xz \ file://0002-ffbuild-common.mak-ensure-target-directories-are-cre.patch \ + file://makerace.patch \ " SRC_URI[sha256sum] = "cf38e0e28c7e5605942c4a77755349b0145804a397af37eb1fb4c77cb237f635"