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 7.0.1 has been released! | Please read ffmpeg.org/developer.html#Code-of-conduct
Daemon404 has quit [Ping timeout: 252 seconds]
IndecisiveTurtle has quit [Ping timeout: 256 seconds]
IndecisiveTurtle has quit [Ping timeout: 264 seconds]
jarthur has joined #ffmpeg-devel
Marth64 has joined #ffmpeg-devel
jamrial has quit []
Marth64[m] has quit [Ping timeout: 268 seconds]
Martchus_ has joined #ffmpeg-devel
Martchus has quit [Ping timeout: 268 seconds]
System_Error has quit [Remote host closed the connection]
System_Error has joined #ffmpeg-devel
AbleBacon has quit [Read error: Connection reset by peer]
rvalue has quit [Read error: Connection reset by peer]
rvalue has joined #ffmpeg-devel
ramiro has quit [Ping timeout: 264 seconds]
ramiro has joined #ffmpeg-devel
tufei has quit [Remote host closed the connection]
tufei has joined #ffmpeg-devel
<michaelni>
kasper93, maybe, i didnt see that
<nevcairiel>
pretty sure the official git docs have mentioned app passwords for a logn time now, its the only thing I have ever used and i've been seding patches for 15 years :D
<JEEB>
yeh, I was surprised normal auth worked at all :D I thought it all stopped working ages ago
Krowl has joined #ffmpeg-devel
ramiro has quit [Ping timeout: 240 seconds]
jarthur has quit [Quit: jarthur]
ramiro has joined #ffmpeg-devel
Krowl has quit [Read error: Connection reset by peer]
System_Error has quit [Ping timeout: 260 seconds]
System_Error has joined #ffmpeg-devel
ngaullier has joined #ffmpeg-devel
ramiro has quit [Ping timeout: 268 seconds]
ramiro has joined #ffmpeg-devel
Krowl has joined #ffmpeg-devel
ramiro has quit [Ping timeout: 264 seconds]
ramiro has joined #ffmpeg-devel
Daemon404 has quit [Changing host]
Daemon404 has joined #ffmpeg-devel
ccawley2011 has joined #ffmpeg-devel
j45_ has joined #ffmpeg-devel
j45 has quit [Ping timeout: 260 seconds]
j45_ is now known as j45
j45 has quit [Changing host]
j45 has joined #ffmpeg-devel
cone-402 has joined #ffmpeg-devel
<cone-402>
ffmpeg Anton Khirnov master:0060a368b175: lavc/hevcdec: always call hevc_frame_end() after successfully decoding an AU
ramiro has quit [Ping timeout: 264 seconds]
ramiro has joined #ffmpeg-devel
ramiro has quit [Ping timeout: 264 seconds]
ramiro has joined #ffmpeg-devel
ramiro has quit [Ping timeout: 256 seconds]
ramiro has joined #ffmpeg-devel
Krowl has quit [Read error: Connection reset by peer]
<Daemon404>
doea git stillnot support oauth?
cubicibo has joined #ffmpeg-devel
lexano has joined #ffmpeg-devel
jamrial has joined #ffmpeg-devel
ramiro has quit [Ping timeout: 264 seconds]
ramiro has joined #ffmpeg-devel
cubicibo has quit [Ping timeout: 250 seconds]
Krowl has joined #ffmpeg-devel
cone-402 has quit [Quit: transmission timeout]
Krowl has quit [Read error: Connection reset by peer]
<nevcairiel>
a neckbeard project supporting real-world tech? what do you think this is? :D
<nevcairiel>
they delegate any advanced authentication to the credential helper functionality, just have it call some tool to provide a password token
<michaelni>
when googling for gmail + git i did see some pages explaining how to use git send email with oauth* but they all looked more combersome
<kasper93>
nevcairiel: I'm surprised that those app password never expire or have to be renewed. I also have one for ages and it just works (TM)
Krowl has joined #ffmpeg-devel
SuperFashi_ has quit [Ping timeout: 272 seconds]
SuperFashi has joined #ffmpeg-devel
AbleBacon has joined #ffmpeg-devel
microchip_ has quit [Ping timeout: 256 seconds]
microchip_ has joined #ffmpeg-devel
<ramiro>
while adding yuv2rgb to checkasm, I noticed that the majority of the mmx/mmxext functions are slower than their c counterpart (both on 32- and 64-bit). could this be due to a measurement error, or the impact of emms, or could it be that really are just slower?
<mkver>
ramiro: Maybe CPU manufacturers deprioritized MMX. It may be that these functions are slower than C on older processors.
<mkver>
Anyway, feel free to remove such MMX code.
<ramiro>
I tested with ffmpeg itself and they are indeed slower.
<courmisch>
do you maybe have autovec enabled?
<ramiro>
courmisch: most likely, yes. but I assume the vast majority of x86 builds do as well.
<courmisch>
ramiro: you're comparing SSE2 with MMX then. It's not as dramatic as if scalar was faster than Mmx
<kepstin>
I guess the comparison that was interesting at the time was mmx vs. compiling for 32bit x86 with -mfpmath=387
___nick___ has joined #ffmpeg-devel
Krowl has quit [Read error: Connection reset by peer]
<ramiro>
the c code uses LUTs with precalculated values, the simd code recalculates the values
kasper93 has quit [Remote host closed the connection]
IndecisiveTurtle has joined #ffmpeg-devel
IndecisiveTurtle has quit [Remote host closed the connection]
Krowl has joined #ffmpeg-devel
<Lynne>
mkver: did you lose your mind working on mpegvideoenc?
<mkver>
Lynne: Definitely.
<Lynne>
condolences
<Lynne>
I'm sure it made sense at some point to unify everything mpeg into a single piece of code
<Lynne>
20 years ago when ffmpeg didn't have anything else
<haasn>
opinions on adding decoded_side_data to AVCodecParameters, so that bitstream filters can look at it as well?
<haasn>
horrible, leaky violation of abstraction layers?
<elenril>
away, satan
* courmisch
summons an S-tier wild elenril
<elenril>
why should bitstream filters have anything to do with decoded side data
<haasn>
I was just playing with the idea of moving dolby vision synthesis out of encoders and into a bitstream filter (so it can be used with _any_ encoder)
<haasn>
and in particular, so we can add support for features which requires prior knowledge about how individual frames are actually encoded (specifically, the frame type)
<elenril>
then have lavc map decoded side data into coded side data
<haasn>
but currently the configuration record generation depends on decoded_side_data because it looks at the first frame's attached side data to see if it has dolby vision metadata or not
<haasn>
elenril: type mismatch, coded_side_data is packet data, decoded_side_data is frame data
<elenril>
bitstream filters only work with packets, so that seems correct
<haasn>
decoded_side_data is frame-level side data that conceptually applies to the whole stream
<haasn>
or is at least informative of the entire stream
<haasn>
indeed, both coded_side_data and decoded_side_data are "wrong" inside AVCodecParameters from this PoV, because the latter also applies to the whole stream
<elenril>
I don't see how that follows
<elenril>
AVCodecParameters in general applies to the whole stream
<haasn>
yes, so putting AVPacketSideData in there only works with the implicit promise that the data really applies to the whole stream
<haasn>
same deal with AVFrameSideData
<haasn>
AVPacketSideData seems confused about whether it should be for per-stream data, per-packet data, or per-frame data
<haasn>
and some side data types even seem to be confused about whether they should be in AVPacketSideData or AVFrameSideData (e.g. ICC profiles)
<elenril>
how about both
System_Error has quit [Remote host closed the connection]
Livio has joined #ffmpeg-devel
System_Error has joined #ffmpeg-devel
<haasn>
clearly we need AVStreamSideData, AVPacketSideData, AVFrameSideData and a clear separation of all three, with duplication where necessary
<haasn>
(or just AVSideData and none of this nonsense)
<haasn>
(plus a tag telling us if the source of the AVSideData is the container, the codec header, the frame, etc.)
rvalue has quit [Remote host closed the connection]
System_Error has quit [Remote host closed the connection]
System_Error has joined #ffmpeg-devel
Livio has joined #ffmpeg-devel
Krowl has joined #ffmpeg-devel
mkver has quit [Ping timeout: 256 seconds]
<courmisch>
I wonder how I was able to bench code that does not compile
<courmisch>
oh it compiles, that explains
Krowl has quit [Read error: Connection reset by peer]
HarshK23 has quit [Quit: Connection closed for inactivity]
<cone-966>
ffmpeg Michael Niedermayer release/5.1:32af5cb10ac8: avfilter/signature_lookup: Fix 2 differences to the refernce SW
<cone-966>
ffmpeg Michael Niedermayer release/5.1:601e4fb1f067: avfilter/signature_lookup: Dont copy uninitialized stuff around
<cone-966>
ffmpeg Michael Niedermayer release/5.1:d9ccbd71f482: avfilter/vf_thumbnail_cuda: Set ret before checking it
<cone-966>
ffmpeg Michael Niedermayer release/5.1:b9867cade36c: doc/examples/vaapi_transcode: Simplify loop
<cone-966>
ffmpeg Michael Niedermayer release/5.1:2f0919b26a3d: avcodec/hevcdec: Check ref frame
<cone-966>
ffmpeg Michael Niedermayer release/5.1:dc9d2eb69d55: avcodec/pngdec: Check last AVFrame before deref
<cone-966>
ffmpeg Michael Niedermayer release/5.1:06cc1197e500: avcodec/ac3_parser: Check init_get_bits8() for failure
<cone-966>
ffmpeg Michael Niedermayer release/5.1:a2e4f3e5f4b4: avformat/kvag: Check sample_rate
<cone-966>
ffmpeg Michael Niedermayer release/5.1:a17885a73d37: avformat/mxfdec: Check body_offset
<cone-966>
ffmpeg Michael Niedermayer release/5.1:4408336d9eb7: avcodec/avs2_parser: Assert init_get_bits8() success with const size 15
<cone-966>
ffmpeg Michael Niedermayer release/5.1:d71036b4559c: avcodec/avs3_parser: assert the return value of init_get_bits()
<cone-966>
ffmpeg Michael Niedermayer release/5.1:91df1b66c717: swscale/output: Fix integer overflow in yuv2rgba64_1_c_template
<cone-966>
ffmpeg Michael Niedermayer release/5.1:3ca9febc6e4c: swscale/output: Fix integer overflow in yuv2rgba64_full_1_c_template()
<cone-966>
ffmpeg Michael Niedermayer release/5.1:7618da9b2ddf: avcodec/mpegvideo_enc: Fix 1 line and one column images
<cone-966>
ffmpeg Michael Niedermayer release/5.1:c379893ebe8a: avformat/concatdec: Check file
<cone-966>
ffmpeg Michael Niedermayer release/5.1:c3471d55c595: doc/examples/demux_decode: Simplify loop
<cone-966>
ffmpeg Michael Niedermayer release/5.1:b0754513f481: tools/opt_common: Check for malloc failure
<cone-966>
ffmpeg Michael Niedermayer release/5.1:b76f24c7087c: fftools/ffplay: Check return of swr_alloc_set_opts2()
<cone-966>
ffmpeg Michael Niedermayer release/5.1:ed6b5d616439: avcodec/cbs_av1: Avoid shift overflow
<cone-966>
ffmpeg Michael Niedermayer release/5.1:f7fcdeb6a0e9: libavutil/base64: Try not to write over the array end
<cone-966>
ffmpeg Michael Niedermayer release/5.1:353cc497cc36: avutil/tests/base64: Check with too short output array
<cone-966>
ffmpeg Michael Niedermayer release/5.1:0a27f1d3318c: avcodec/lpc: copy levenson coeffs only when they have been computed
<cone-966>
ffmpeg Michael Niedermayer release/5.1:148a3ed506f7: avcodec/h264_slice: Remove dead sps check
<cone-966>
ffmpeg Michael Niedermayer release/5.1:21249e8cbef1: avcodec/fmvc: remove dead assignment
<cone-966>
ffmpeg Michael Niedermayer release/5.1:cb03082cfa9e: avcodec/exr: Fix preview overflow
<cone-966>
ffmpeg Michael Niedermayer release/5.1:52af506fdb34: avcodec/qsvdec: Check av_image_get_buffer_size() for failure
<cone-966>
ffmpeg Michael Niedermayer release/5.1:21fd9fb7db8d: avcodec/jpeg2000dec: remove ST=3 case
<cone-966>
ffmpeg Michael Niedermayer release/5.1:5bc913c2351c: avcodec/vp3: Replace check by assert
<cone-966>
ffmpeg Michael Niedermayer release/5.1:61332f89a62c: avcodec/vble: Check av_image_get_buffer_size() for failure
<cone-966>
ffmpeg Michael Niedermayer release/5.1:701ca03b0063: avcodec/mpeg12dec: Use 64bit in bit computation
<cone-966>
ffmpeg Michael Niedermayer release/5.1:ed40d4850b88: avcodec/mpeg4videodec: assert impossible wrap points
<cone-966>
ffmpeg Michael Niedermayer release/5.1:cb4e45231cdb: avcodec/mpegvideo_enc: Fix potential overflow in RD
<cone-966>
ffmpeg Michael Niedermayer release/5.1:21fe5ce6937e: avcodec/mscc & mwsc: Check loop counts before use
<cone-966>
ffmpeg Michael Niedermayer release/5.1:84953f5c232e: avformat/demux: resurrect dead stores
<cone-966>
ffmpeg Michael Niedermayer release/5.1:2f9a4353bb38: avdevice/dshow: fix badly indented line
<cone-966>
ffmpeg Michael Niedermayer release/5.1:3ab1855847ec: avutil/tests/dict: Check av_dict_set() before get for failure
<cone-966>
ffmpeg Michael Niedermayer release/5.1:b2864f699e96: avutil/tests/opt: Check av_set_options_string() for failure
<cone-966>
ffmpeg Michael Niedermayer release/5.1:b8dfe07c5099: swscale/x86/swscale: use a clearer name for INPUT_PLANER_RGB_A_FUNC_CASE
<cone-966>
ffmpeg Michael Niedermayer release/5.1:3b94cf8b20b6: swscale/yuv2rgb: Use 64bit for brightness computation
<cone-966>
ffmpeg Michael Niedermayer release/5.1:f3424f04704f: avcodec/tiff: Assert init_get_bits8() success in horizontal_fill()
<cone-966>
ffmpeg Michael Niedermayer release/5.1:bc5b55a4caad: avcodec/tiff: Assert init_get_bits8() success in unpack_gray()
<cone-966>
ffmpeg Michael Niedermayer release/5.1:b7efe0f70489: avdevice/pulse_audio_enc: Use av_rescale() to avoid integer overflow
<cone-966>
ffmpeg Michael Niedermayer release/5.1:1763189d0e93: avformat/ape: Use 64bit for final frame size
<cone-966>
ffmpeg Michael Niedermayer release/5.1:66e306b8e3a6: avformat/argo_asf: Use 64bit in offset intermediate
<cone-966>
ffmpeg Michael Niedermayer release/5.1:8b19671c78fe: avformat/asfdec_f: Use 64bit for preroll computation
<cone-966>
ffmpeg Michael Niedermayer release/5.1:a0f0e6d81379: avcodec/sga: Make it clear that the return is intentionally not checked
<cone-966>
ffmpeg Michael Niedermayer release/5.1:c48387a3f532: avformat/fwse: Remove always false expression
<cone-966>
ffmpeg Michael Niedermayer release/5.1:f85527cd8b6d: avcodec/tests/jpeg2000dwt: Use 64bit in err2 computation
<cone-966>
ffmpeg Michael Niedermayer release/5.1:2b7260fbb793: avcodec/tests/jpeg2000dwt: Use 64bit in comparission
<cone-966>
ffmpeg Michael Niedermayer release/5.1:896af4bd3e3d: avcodec/wavpackenc: Use unsigned for potential 31bit shift
<cone-966>
ffmpeg Michael Niedermayer release/5.1:442d18371ecd: avcodec/rv34: assert that size is not 0 in rv34_gen_vlc_ext()
<cone-966>
ffmpeg Michael Niedermayer release/5.1:4d9530afdfbd: avcodec/scpr3: Check add_dec() for failure
<cone-966>
ffmpeg Michael Niedermayer release/5.1:9e46b7043630: avcodec/tests/dct: Use 64bit in intermediate for error computation
<cone-966>
ffmpeg Michael Niedermayer release/5.1:85242a8aafd8: avcodec/notchlc: Check init_get_bits8() for failure
<cone-966>
ffmpeg Michael Niedermayer release/5.1:206816e91546: avcodec/pcm-dvdenc: 64bit pkt-size
<cone-966>
ffmpeg Michael Niedermayer release/5.1:417723acccd4: avcodec/vp8: Check mutex init
<cone-966>
ffmpeg Michael Niedermayer release/5.1:188fe46e92cc: avcodec/vp8: Check cond init
<cone-966>
ffmpeg Michael Niedermayer release/5.1:5d891ffbe5c7: avcodec/ilbcdec: Remove dead code
<cone-966>
ffmpeg Michael Niedermayer release/5.1:f46626516995: avformat/sdp: Check before appending ","
<cone-966>
ffmpeg Michael Niedermayer release/5.1:040db2b44747: fftools/cmdutils: Add protective () to FLAGS
<cone-966>
ffmpeg Michael Niedermayer release/5.1:4446e6b55fe4: avdevice/xcbgrab: Check sscanf() return
<cone-966>
ffmpeg Michael Niedermayer release/5.1:5f4036b66bb1: avformat/img2dec: Move DQT after unrelated if()
<cone-966>
ffmpeg Michael Niedermayer release/5.1:393005575b32: avformat/img2dec: Little JFIF / Exif cleanup
<cone-966>
ffmpeg Michael Niedermayer release/5.1:8dc091daee45: avformat/libzmq: Check av_strstart()
<cone-966>
ffmpeg Michael Niedermayer release/5.1:e0e1a3893d1f: avformat/matroskadec: Assert that num_levels is non negative
<cone-966>
ffmpeg Michael Niedermayer release/5.1:14aff02a1040: avformat/mov: Use 64bit in intermediate for current_dts
<cone-966>
ffmpeg Michael Niedermayer release/5.1:74aeed90eb93: avformat/mov: Use int64_t in intermediate for corrected_dts
<cone-966>
ffmpeg Michael Niedermayer release/5.1:293b368b81b7: avfilter/vf_rotate: Check ff_draw_init2() return value
<cone-966>
ffmpeg Michael Niedermayer release/5.1:12edfe7004ce: doc/developer: Provide information about git send-email and gmail
<cone-966>
ffmpeg Michael Niedermayer release/5.1:0dfbf1dffcd6: Changelog: more backported commits
<thardin>
much commit
thardin has quit [Ping timeout: 268 seconds]
kasper93_ has joined #ffmpeg-devel
kasper93 has quit [Ping timeout: 264 seconds]
kasper93_ is now known as kasper93
<jamrial>
courmisch: fwiw, i fixed the x86 asm for dct_unquantize. no overwrites anymore
<cone-966>
ffmpeg Sean McGovern master:cdd139d76068: libavcodec/alpha: remove DSP & support code
rvalue- has joined #ffmpeg-devel
rvalue has quit [Ping timeout: 268 seconds]
rvalue- is now known as rvalue
AbleBacon has joined #ffmpeg-devel
ccawley2011 has quit [Read error: Connection reset by peer]
ramiro has quit [Ping timeout: 268 seconds]
ramiro has joined #ffmpeg-devel
Livio has quit [Ping timeout: 240 seconds]
rmmh has joined #ffmpeg-devel
<rmmh>
I'd like to revert b1a213ab5d8 -- filtergraphs without outputs *can* have useful side-effects (through filters writing to files, logging debug messages, etc), and this used to work. My precise use case is doing a stream copy while also analyzing the stream.
<BtbN>
That makes no sense, in copy-mode, there is nothing to filter
<rmmh>
ex: `ffmpeg -i aud.mp4 -filter_complex "[0:a]ebur128,anullsink" -map 0:a:0 -c copy aud.m4a` (this writes to stdout, the full command writes to a file)
<BtbN>
Yes, that command makes no sense. There won't be a decoder in copy mode, so nothing to filter.
<BtbN>
If you really want that, make a second output, -f null or something, and decode and filter there