michaelni changed the topic of #ffmpeg-devel to: Welcome to the FFmpeg development channel | Questions about using FFmpeg or developing with libav* libs should be asked in #ffmpeg | This channel is publicly logged | FFmpeg 7.1.1 has been released! | Please read ffmpeg.org/developer.html#Code-of-conduct
<mkver> Does swscale have any alignment requirements for the strides and pointers?
marcj has joined #ffmpeg-devel
cone-979 has joined #ffmpeg-devel
<cone-979> ffmpeg Andreas Rheinhardt master:0401ca714a27: avcodec/asvenc: Don't waste bits encoding non-visible part
<cone-979> ffmpeg Andreas Rheinhardt master:05a5e2b5761f: avcodec/asvenc: Combine writing bits
<cone-979> ffmpeg Andreas Rheinhardt master:a5ff6ea32d9a: avcodec/asvenc: Simplify writing extradata
<cone-979> ffmpeg Andreas Rheinhardt master:ef95a8d7b5be: avcodec/mpegvideo_enc: Use av_unreachable() for unreachable code
<cone-979> ffmpeg Andreas Rheinhardt master:ded08d0b1bb4: avcodec/msmpeg4dec: Use av_unreachable() for unreachable code
<cone-979> ffmpeg Andreas Rheinhardt master:0d73ce4bc489: avcodec/h263dec: Use av_unreachable() for unreachable code
<cone-979> ffmpeg Andreas Rheinhardt master:7ae20272b79e: avcodec/pcm: Use av_unreachable() for unreachable code
_whitelogger has joined #ffmpeg-devel
<mkver> wbs: Apparently my commit 0401ca714a2 caused a bus error here: https://fate.ffmpeg.org/report.cgi?slot=armv7-linux-clang-trunk&time=20250526024313
<mkver> Is it the AV_COPY128? It should actually be properly aligned.
<cone-979> ffmpeg Andreas Rheinhardt master:99187419676d: avcodec/mpegvideo_enc: Set gob_index once during init
<cone-979> ffmpeg Andreas Rheinhardt master:a0a136e57b9d: avcodec/h263dec: Move calculating gob_index to {intel,itu}h263dec.c
<cone-979> ffmpeg Andreas Rheinhardt master:f887a2b00615: avcodec/mpeg4videodec: Don't initialize unused parts of RLTables
<cone-979> ffmpeg Andreas Rheinhardt master:295abb1fdc06: avcodec/mpegvideo_enc: Reduce stack usage
<cone-979> ffmpeg Andreas Rheinhardt master:d75a398d00c2: avcodec/mpeg4videoenc: Simplify creating LUT
<cone-979> ffmpeg Andreas Rheinhardt master:2a59cb0c8248: avcodec/mpeg4video: Nuke ff_mpeg4_init_rl_intra()
<cone-979> ffmpeg Andreas Rheinhardt master:f7bd2a5af87b: avcodec/mpeg4videoenc: Remove no-output code
<cone-979> ffmpeg Andreas Rheinhardt master:7267e4dab82a: avcodec/mpeg4videoenc: Split writing blocks into intra and inter
<cone-979> ffmpeg Andreas Rheinhardt master:6af1bcb18034: avcodec/mpeg4videoenc: Use smaller scope for variable
<cone-979> ffmpeg Andreas Rheinhardt master:eb5d900f87fa: avcodec/mpeg4videodec: Remove unnecessary ff_xvid_idct_init()
<cone-979> ffmpeg Andreas Rheinhardt master:6349a3324dd8: avcodec/xvididct: Remove always-true checks
<cone-979> ffmpeg Andreas Rheinhardt master:2e8a3e664504: avcodec/idctdsp: Only try to initialize xvid idct if it is used
<cone-979> ffmpeg Andreas Rheinhardt master:f2c04fd5870a: avcodec/mpegvideo_dec: Document which codecs take which codepath
<cone-979> ffmpeg Andreas Rheinhardt master:778c0a48fa46: avcodec/mpegvideo_dec: Simplify check for unquantizing inter blocks
<cone-979> ffmpeg Andreas Rheinhardt master:ed73675832ed: avcodec/mpegvideo: Move mpeg_quant to {Mpeg4Dec,MPVEnc}Context
<cone-979> ffmpeg Andreas Rheinhardt master:8987c293908b: avcodec/mpegvideo: Check h263_aic, not h263_plus for allocating dc_val
<cone-979> ffmpeg Andreas Rheinhardt master:da5c2a419b6d: avcodec/ituh263enc: Don't use h263_plus field
<cone-979> ffmpeg Andreas Rheinhardt master:ac1e6d21718e: avcodec/mpegvideo: Remove h263_plus
<cone-979> ffmpeg Andreas Rheinhardt master:43a69886b216: avcodec/mpeg4videodec: Use union to save space
jamrial has quit []
mkver has quit [Ping timeout: 245 seconds]
Martchus_ has joined #ffmpeg-devel
Martchus has quit [Ping timeout: 276 seconds]
BradleyS has quit [Read error: Connection reset by peer]
BradleyS has joined #ffmpeg-devel
BradleyS has quit [Client Quit]
BradleyS has joined #ffmpeg-devel
BradleyS has quit [Read error: Connection reset by peer]
BradleyS has joined #ffmpeg-devel
BradleyS has quit [Quit: quit]
Anthony_ZO has joined #ffmpeg-devel
System_Error has quit [Remote host closed the connection]
<fflogger> [newticket] Andrew-R: Ticket #11609 ([avcodec] Vulkan ffv1 hw decode broken at least for 1920x1080 test file) created https://trac.ffmpeg.org/ticket/11609
BradleyS has joined #ffmpeg-devel
System_Error has joined #ffmpeg-devel
cone-979 has quit [Quit: transmission timeout]
<fflogger> [editedticket] Lynne: Ticket #11609 ([avcodec] Vulkan ffv1 hw decode broken at least for 1920x1080 test file) updated https://trac.ffmpeg.org/ticket/11609#comment:1
<fflogger> [editedticket] Ferdinand_Scholten: Ticket #11607 ([ffplay] Linking ffplay fails) updated https://trac.ffmpeg.org/ticket/11607#comment:2
<fflogger> [editedticket] Ferdinand_Scholten: Ticket #11608 ([avfilter] libavfilter undefined symbol) updated https://trac.ffmpeg.org/ticket/11608#comment:3
IndecisiveTurtle has joined #ffmpeg-devel
Faeez has joined #ffmpeg-devel
ngaullier has joined #ffmpeg-devel
Kei_N_ has joined #ffmpeg-devel
Kei_N has quit [Ping timeout: 252 seconds]
putacho has joined #ffmpeg-devel
microchip_ has quit [Ping timeout: 276 seconds]
IndecisiveTurtle has quit [Ping timeout: 272 seconds]
tufei_ has joined #ffmpeg-devel
tufei__ has quit [Ping timeout: 264 seconds]
microlappy has joined #ffmpeg-devel
omegatron has joined #ffmpeg-devel
microlappy has quit [Quit: Konversation terminated!]
microlappy has joined #ffmpeg-devel
microlappy has quit [Quit: Konversation terminated!]
cone-818 has joined #ffmpeg-devel
<cone-818> ffmpeg Coia Prant master:df967d095ae6: configure: Use MSYSTEM_CARCH for default arch on msys2
mkver has joined #ffmpeg-devel
Kei_N_ has quit [Read error: Connection reset by peer]
Kei_N has joined #ffmpeg-devel
Anthony_ZO has quit [Quit: Anthony_ZO]
Anthony_ZO has joined #ffmpeg-devel
jamrial has joined #ffmpeg-devel
<ePirat> haasn, do you know how I use query_formats in a filter to set pixel formats? seems there is no ff_set_common_pixfmts
<haasn> ePirat: ff_set_common_formats
<ePirat> haasn, if I understand that correctly it would not work for multimedia filter
<ePirat> as it uses AVMEDIA_TYPE_UNKNOWN internally, applying pixelformats to any link, even audio
<ePirat> so my audio links would end up with some garbage format in them…
<haasn> ah, true
<haasn> then just ff_formats_ref it onto the relevant input/outputs
<haasn> ff_set_common_pixfmt is just a wrapper for doing that
<ePirat> maybe we should add ff_set_common_pixfmts that just applies to video, intended to set pixel formats?
<ePirat> ff_set_common_formats seems like a nice footgun for multimedia filters…
<ePirat> as it reads like it would do what you want but doesnt
NotWarcop has quit [Remote host closed the connection]
wyatt8740 has quit [Quit: ZNC got killed or something else has gone wrong, probably.]
Warcop has joined #ffmpeg-devel
wyatt8740 has joined #ffmpeg-devel
<cone-818> ffmpeg Henrik Gramner master:eda0ac7e5f09: avcodec/x86/vp9: Add AVX-512ICL for 16x16 and 32x32 10bpc inverse transforms
Anthony_ZO has quit [Ping timeout: 276 seconds]
System_Error has quit [Remote host closed the connection]
thardin has quit [Ping timeout: 268 seconds]
System_Error has joined #ffmpeg-devel
minimal has joined #ffmpeg-devel
ccawley2011 has joined #ffmpeg-devel
ccawley2011_ has joined #ffmpeg-devel
ccawley2011__ has joined #ffmpeg-devel
ccawley2011 has quit [Ping timeout: 244 seconds]
ccawley2011 has joined #ffmpeg-devel
ccawley2011_ has quit [Ping timeout: 244 seconds]
ccawley2011_ has joined #ffmpeg-devel
ccawley2011__ has quit [Ping timeout: 244 seconds]
ccawley2011__ has joined #ffmpeg-devel
ccawley2011 has quit [Ping timeout: 244 seconds]
ccawley2011_ has quit [Ping timeout: 265 seconds]
ccawley2011__ has quit [Ping timeout: 244 seconds]
TheVibeCoder has joined #ffmpeg-devel
<TheVibeCoder> Do not trust single word from your "Leader" - he contacted me via e-mail only once long, long ago.
<jamrial> TheVibeCoder: sorry for this whole situation
<ePirat> Ah great my noop filter still has cases where its broken…
<ePirat> if I chain two, it explodes again…
<TheVibeCoder> in EOF?
<ePirat> yeah
<ePirat> time to bring back all the debug logging again…
<TheVibeCoder> :)
<ePirat> how many engineers does it take to write one activate function…
<another|> How many engineers does it take to write one noop filter /s
tufei_ has quit [Ping timeout: 264 seconds]
thardin has joined #ffmpeg-devel
<jamrial> haasn: can you address the ff_xyz12Torgb48 comment thread?
<JEEB> haasn: responded to The Thread, sorry that it took this long. been quite a bit of time, and somehow in my head the overhead of replying to an email is often overly large (as opposed to giving "just quick notes" on IRC or so)
Faeez has quit [Ping timeout: 240 seconds]
tufei has joined #ffmpeg-devel
TheVibeCoder has quit [Quit: Client closed]
cone-818 has quit [Quit: transmission timeout]
tufei has quit [Remote host closed the connection]
ngaullier has quit [Remote host closed the connection]
<JEEB> nice, 2023 ed of H.222 (MPEG-PS/TS) is now free https://www.itu.int/rec/T-REC-H.222.0-202308-S/en
DodoGTA has quit [Quit: DodoGTA]
Faeez has joined #ffmpeg-devel
DodoGTA has joined #ffmpeg-devel
cone-542 has joined #ffmpeg-devel
<cone-542> ffmpeg Gyan Doshi master:c79bce92b840: avformat/movenc: disallow buggy flags combination
<ramiro> BtbN: why do we run test_exec to enable zlib_gzip in configure?
<BtbN> How else would you test if gzip is there?
<ramiro> BtbN: isn't the "[ -x "$(command -v gzip)" ] && enable gzip" check enough to see that gzip works for the build?
<BtbN> what?
<BtbN> zlib_gzip tests if _zlib_ has gzip support enabled, i.e. can extract what the gzip command produces
ccawley2011 has joined #ffmpeg-devel
<ramiro> BtbN: sure, but why do we care at compile time? :)
<ramiro> isn't that something that the user should care about at runtime?
<BtbN> Cause it enables compression of embedded resources
<BtbN> So it needs to be decided at compile time
ccawley2011 has quit [Read error: Connection reset by peer]
<ramiro> BtbN: yes, but the build won't necessarily run at the machine that built it
<BtbN> there is nothing that can be done about that
<BtbN> no reason to not at least check anyway
<BtbN> chances are if your build env lacks it, so does your runtime env
<ramiro> I think I misphrased myself. the build machine is not necessarily the machine where the build will be run.
<ramiro> all that is needed for the build is that gzip exists to compress the data. if the user's zlib doesn't handle gzip, that's their problem
omegatron has quit [Quit: Power is a curious thing. It can be contained, hidden, locked away, and yet it always breaks free.]
ocrete has quit [Quit: The Lounge - https://thelounge.chat]
<BtbN> yes? I don't see what that changes about my answer
<BtbN> still seems like a good idea to check
<ramiro> if I understood the check correctly, it tests whether the user will be able to decompress the data at runtime.
putacho has quit [Quit: There is no spoon!]
ocrete has joined #ffmpeg-devel
<ramiro> but I could change my system's zlib.so after the build
<ramiro> honestly, it seems like a bad idea to me. an extra check that is not necessary, and may limit the build
<BtbN> again: Chances are if the build envs zlib lacks it, so will the runtime
<sfan5> why would zlib not be able to read gzip data anyway? third-party implementations?
microchip_ has joined #ffmpeg-devel
<BtbN> old versions of it don't, and I think it can in theory be turned off
<BtbN> So the check seems important when building for older distros
<sfan5> or even if this were the case you can IIRC just skip the gzip header bytes and the rest is a raw deflate stream
<BtbN> It's not a problem in reality anyway, and protects some people building for ancient stuff from a confusing surprise
<sfan5> I guess
<fflogger> [editedticket] j7n: Ticket #11578 ([ffmpeg] Waveform discontinuity in decoded E-AC-3) updated https://trac.ffmpeg.org/ticket/11578#comment:7
rvalue- has joined #ffmpeg-devel
rvalue has quit [Ping timeout: 252 seconds]
rvalue- is now known as rvalue
<cone-542> ffmpeg Michael Niedermayer master:9d229440ace1: Makefile: Split ALLFFLIBS
microchip_ has quit [Ping timeout: 244 seconds]
<ePirat> How does an AVFilterLink determine if a frame is wanted?
<ePirat> ok I fixed my issue by adding the cargo culted ff_filter_set_ready back, why its necessary I have no clue but whatever
<ePirat> the docs mention explicitly that this is done automatically on status change but well
<ePirat> guess the docs are a lie?
<jamrial> ePirat: behavior may have been altered at some point, and not reflected in the docs
<jamrial> wouldn't be the first time it happens
<ePirat> well I checked a bit the source and there it mentions the notable detail
<ePirat> "changes in the input or output status of an input link"
<ePirat> so when I set the outlink status, it doesnt mark my filter as ready again
<ePirat> which also means FF_FILTER_FORWARD_STATUS is a bit of a footgun...
<fflogger> [editedticket] malvinas2: Ticket #7337 ([avformat] FFmpeg not recognizing WebVTT subtitle stream from HLS playlist) updated https://trac.ffmpeg.org/ticket/7337#comment:12
microchip_ has joined #ffmpeg-devel
kasper93 has quit [Ping timeout: 252 seconds]
kasper93 has joined #ffmpeg-devel
<fflogger> [newticket] oldshev: Ticket #11610 ([swscale] scale with reference prevents video to be scaled with a second scale filter instance) created https://trac.ffmpeg.org/ticket/11610
Traneptora has quit [Quit: Quit]
Kwiboo has quit [Ping timeout: 245 seconds]
<cone-542> ffmpeg Lynne master:c395ad7c2cdc: vulkan_ffv1: small cleanup for golomb
<cone-542> ffmpeg Lynne master:3cbe3418b2b7: vulkan_ffv1: fix golomb coding for non-RGB streams
<fflogger> [editedticket] Lynne: Ticket #11609 ([avcodec] Vulkan ffv1 hw decode broken at least for 1920x1080 test file) updated https://trac.ffmpeg.org/ticket/11609#comment:2
Kwiboo has joined #ffmpeg-devel
uau_ has joined #ffmpeg-devel
uau is now known as Guest2296
Guest2296 has quit [Killed (iridium.libera.chat (Nickname regained by services))]
uau_ is now known as uau
secondcreek has quit [Quit: secondcreek]
<fflogger> [editedticket] ua0lnj: Ticket #11514 ([avcodec] Cuvid decoder problem with deinterlacing) updated https://trac.ffmpeg.org/ticket/11514#comment:6
kasper93_ has joined #ffmpeg-devel
kasper93 has quit [Killed (tungsten.libera.chat (Nickname regained by services))]
kasper93_ is now known as kasper93
_whitelogger has joined #ffmpeg-devel