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 6.1.1 has been released! | Please read ffmpeg.org/developer.html#Code-of-conduct
rvalue has quit [Ping timeout: 264 seconds]
rvalue has joined #ffmpeg-devel
wyatt8740 has joined #ffmpeg-devel
iive has quit [Quit: They came for me...]
cone-341 has joined #ffmpeg-devel
<cone-341>
ffmpeg Andreas Rheinhardt master:b6b33f7edd94: avutil/common: Move includes to the beginning of the file
<cone-341>
ffmpeg Andreas Rheinhardt master:27cd7011bc2e: avutil/tests/pixelutils: Remove dead code
<cone-341>
ffmpeg Andreas Rheinhardt master:1d66a122df9f: avcodec/avcodec: Deprecate AV_INPUT_BUFFER_MIN_SIZE
<cone-341>
ffmpeg Andreas Rheinhardt master:1eafbc27e2d3: avcodec/vvc/vvc_ps: Check before access
<cone-341>
ffmpeg Andreas Rheinhardt master:ef04737e189e: avcodec/vvc/vvc_ps: Use union for luts to avoid unaligned accesses
<cone-341>
ffmpeg Andreas Rheinhardt master:48612de63c2a: avcodec/vvc/vvcdsp: Remove pointless wrappers
<cone-341>
ffmpeg Andreas Rheinhardt master:39b4b5aad7f5: avcodec: Remove superfluous ';' outside of functions
<cone-341>
ffmpeg Andreas Rheinhardt master:cb9f1f59a1db: avutil/version: Remove outdated checks
<jdarnley>
lmao what how is h264 "not currently supported in container" for mp4?
<mkver>
?
<jdarnley>
beats me, for now
<jdarnley>
[mp4 @ 0x559232f3f740] Could not find tag for codec h264 in stream #0, codec not currently supported in container
<jdarnley>
mov_find_codec_tag() returns some erorr
<jdarnley>
Ah "disposition"
<jdarnley>
There we go much better
<jdarnley>
And that is why I hate embedded covers
thilo has quit [Ping timeout: 255 seconds]
thilo has joined #ffmpeg-devel
<compn>
h264 isnt the right tag is it ?
<compn>
er no ignore me
* compn
avoids looking at isom.c
<mkver>
Great, since version 16 clang now needs fno-sanitize-address-use-odr-indicator with asan on non-windows because otherwise configure thinks that symbol names need to be mangled (via extern_prefix).
qeed has joined #ffmpeg-devel
qeed has quit [Remote host closed the connection]
qeed has joined #ffmpeg-devel
qeed has quit [Quit: Leaving]
mkver has quit [Ping timeout: 264 seconds]
derpydoo has quit [Ping timeout: 264 seconds]
jamrial has quit []
cone-341 has quit [Quit: transmission timeout]
elastic_dog has quit [Ping timeout: 240 seconds]
AbleBacon has quit [Read error: Connection reset by peer]
elastic_dog has joined #ffmpeg-devel
Martchus_ has joined #ffmpeg-devel
Martchus has quit [Ping timeout: 264 seconds]
HarshK23 has joined #ffmpeg-devel
rooisnoek has joined #ffmpeg-devel
rooisnoek has quit [Client Quit]
kurosu has joined #ffmpeg-devel
lexano has quit [Ping timeout: 268 seconds]
lexano has joined #ffmpeg-devel
Krowl has joined #ffmpeg-devel
quietvoid has quit [Quit: No Ping reply in 180 seconds.]
quietvoid has joined #ffmpeg-devel
jarthur_ has quit [Quit: jarthur_]
Krowl has quit [Read error: Connection reset by peer]
meerpi has joined #ffmpeg-devel
Krowl has joined #ffmpeg-devel
deer3 has quit [Ping timeout: 272 seconds]
lemourin has quit [Read error: Connection reset by peer]
lemourin has joined #ffmpeg-devel
meerpi has quit [Quit: Client closed]
<cone-002>
ffmpeg Anton Khirnov master:8f592eb35f4b: fftools/ffmpeg_filter: compute input trim start/end in demuxer
<cone-002>
ffmpeg Anton Khirnov master:bd3c1c194bd2: fftools/ffmpeg_filter: accept a name from its upstream input
cone-002 has joined #ffmpeg-devel
<cone-002>
ffmpeg Anton Khirnov master:09438d652997: fftools/ffmpeg_filter: stop taking display matrix from global side data
<cone-002>
ffmpeg Anton Khirnov master:fef3052df316: fftools/ffmpeg_filter: pass autorotate/reinit flags through InputFilterOptions
<cone-002>
ffmpeg Anton Khirnov master:826cfd9997ff: fftools/ffmpeg_filter: pass framerate through InputFilterOptions
<cone-002>
ffmpeg Anton Khirnov master:6315f78e0c0d: fftools/ffmpeg_filter: pass sub2video canvas size through InputFilterOptions
<cone-002>
ffmpeg Anton Khirnov master:6b6815b1c875: fftools/ffmpeg: move subtitle helpers to ffmpeg_dec, their only user
<cone-002>
ffmpeg Anton Khirnov master:a3f69cdec73d: fftools/ffmpeg: cosmetics, vertically align structs
signalhunter has quit [Quit: Zzz]
signalhunter has joined #ffmpeg-devel
wyatt8740 has quit [Remote host closed the connection]
wyatt8740 has joined #ffmpeg-devel
wyatt8740 has quit [Ping timeout: 260 seconds]
wyatt8740 has joined #ffmpeg-devel
deer3 has joined #ffmpeg-devel
kasper93 has joined #ffmpeg-devel
Krowl has quit [Read error: Connection reset by peer]
kierank has joined #ffmpeg-devel
Krowl has joined #ffmpeg-devel
mkver has joined #ffmpeg-devel
jamrial has joined #ffmpeg-devel
cone-002 has quit [Quit: transmission timeout]
Krowl has quit [Quit: Krowl]
Krowl has joined #ffmpeg-devel
<jdek>
wbs: fwiw the luma deblock asm was written with having support for no_p/no_q in mind, and check_indent.sh is nice thanks
<jdek>
will setup something for checking build with all the compilers (only tested on linux/arm64 since i switched to asahi finally)
<JEEB>
asahi <3
<JEEB>
very good work done there
Kei_N_ has joined #ffmpeg-devel
Kei_N has quit [Ping timeout: 264 seconds]
<Traneptora>
is that the linux distro that runs on macbooks
<JEEB>
yea
<JEEB>
the aarch64 ones
<Traneptora>
part of it doesn't make much sense to me, like why would you buy apple hardware just to put linux on it
<JEEB>
because it's literally the only reall capable aarch64 hardware out there :D
<JEEB>
which you can easily utilize freely
<Traneptora>
oh, right, cause it's aarch64
<wbs>
windows/aarch64 laptops are available too; WSL works pretty nicely on them :P
<JEEB>
wbs: yea but IIRC the IPC is *much* better on the apple hardware
<wbs>
yeah they're much much beefier indeed
<Traneptora>
why would you specifically want an aarch64 laptop though for non-dev purposes
<JEEB>
battery perf mostly
<Traneptora>
the instruction set makes the cpu use less power?
<JEEB>
apple's aarch64 ARM implementation
<JEEB>
there are probably plenty of hardware that can have nice battery life, but apple's ARM design is "oh maaaan"
MisterMinister has joined #ffmpeg-devel
Krowl has quit [Read error: Connection reset by peer]
<JEEB>
so in a way I'd *really* hope other places would catch up to that :P
Taripe is now known as ePirat
AbleBacon has joined #ffmpeg-devel
Krowl has joined #ffmpeg-devel
Traneptora has quit [Quit: Quit]
<Lynne>
courmisch: does stuff like the opusdsp and float_dsp rvv code need something special to run?
<Lynne>
I got a shell on a c908 and checkasm doesn't show them, but shows stuff like vp8's idcts and autocorrelate
jarthur has joined #ffmpeg-devel
Krowl has quit [Read error: Connection reset by peer]
devinheitmuell-1 has joined #ffmpeg-devel
devinheitmueller has quit [Ping timeout: 260 seconds]
<cone-605>
ffmpeg Michael Niedermayer master:8caa84cec74a: tools: Add target_sws_fuzzer.c
<cone-605>
ffmpeg Michael Niedermayer master:18f26f8a2f8d: swscale/utils: Allocate more dithererror
<cone-605>
ffmpeg Michael Niedermayer master:64098d0cd8ab: swscale/swscale: Check srcSliceH for bayer
<cone-605>
ffmpeg Michael Niedermayer master:f9906911f0b7: Revert "swscale: fix sws_setColorspaceDetails after sws_init_context"
<cone-605>
ffmpeg Michael Niedermayer master:e9cc9e492f98: libswscale/utils: Fix bayer to yuvj
<courmisch>
Lynne: there's no runtime detection for Zba and Zbb as yet, and even if there was, it wouldn't work with vendor Linux 5.10 :(
<courmisch>
people are waiting for me to port 6.6 or upstream, and I'm waiting for the community to do it
<courmisch>
So I just force-enable Zba and Zbb in CFLAGS
<JEEB>
that sounds kind of like v4l2-requests forks of FFmpeg. saw another by yet another consultancy agency being mentioned in #ffmpeg yesterday. so now there's what, three or four v4l2-requests forks of FFmpeg? and end users keeps asking if there's any plans to upstream any of them
<Lynne>
courmisch: how is that related to the functions not running?
<courmisch>
riscv_hwprobe() is added in 6.5, so even if FFmpeg supported it, it wouldn't work on vendor kernel which is based on 5.10 LTS
<courmisch>
the community seems mostly focused on VF2 and other JH7110-based boards, not caring a whole lot about C908, so upstreaming is not really happening
<courmisch>
to be fair, VF2 is much better hardware in general. But it has no vectors
Traneptora has joined #ffmpeg-devel
<Lynne>
oh, I see
<Lynne>
but you mentioned you force enable the flags
<Lynne>
does that not turn on AV_CPU_FLAG_RVB_ADDR e.g?
<courmisch>
yes it does
<courmisch>
so at least I don't use custom patches
microchip_ has quit [Quit: There is no spoon!]
microchip_ has joined #ffmpeg-devel
<cone-605>
ffmpeg sunyuechi master:b41e115ddec4: lavc/me_cmp: R-V V pix_abs
<cone-605>
ffmpeg sunyuechi master:f1ec475f6685: lavc/me_cmp: R-V V pix_abs_x2
<cone-605>
ffmpeg sunyuechi master:37463d7979ed: lavc/me_cmp: R-V V pix_abs_y2
<cone-605>
ffmpeg sunyuechi master:9cb8f262f2bc: lavc/me_cmp: R-V V sse
<thardin>
fate-mov-write-amve is failing
<courmisch>
eh! don't say that right after I push, it scares me
<JEEB>
:D
<mkver>
thardin: It works fine here, but I don't have the commits by courmisch in my tree yet.
<mkver>
What is the error?
<mkver>
ubitux: What's the matter with your fate boxes?
<mkver>
Wait, ubitux isn't here at all. So: What's the matter with ubitux?
<thardin>
this.. commit. it scares me.
<thardin>
will pull and re-run FATE
<courmisch>
did the FATE server fail at night?
<courmisch>
only 1 of my 3 batches were uploaded succesfully
<cone-605>
ffmpeg Andreas Rheinhardt master:e2f56f2bac5d: fate/image: Enable fate-webp-rgb-lossless-palette-predictor test
<cone-605>
ffmpeg Andreas Rheinhardt master:4d86e64d33af: avcodec/cbs_vp8: Remove empty flush callback
<Traneptora>
fate passes on my end
rvalue has quit [Ping timeout: 264 seconds]
<thardin>
strange. I'm running ./configure and fate-rsync, default config
<thardin>
oh wait I was testing to mxf patches. let me revert them
<thardin>
also TIL yt-dlp cuts out after 6 hours
<thardin>
when downloading streams off of youtube
<mkver>
Shocking!
<thardin>
it is actually
<thardin>
it doesn't limit how much you can upload. I have a test stream that's been running for 8½ hours
rvalue has joined #ffmpeg-devel
<thardin>
nope, still broken
<Traneptora>
I configured with --disable-stripping --disable-doc --assert-level=2 but otherwise no options
<Traneptora>
this is on x86_64, possibly relevant?
<thardin>
same here
<thardin>
it says threads=1
<thardin>
and also Codec AVOption dct (DCT algorithm) has not been used for any stream. The most likely reason is either wrong type (e.g. a video optio
<thardin>
n with no video streams) or that it is a private option of some encoder which was not actually used for any stream.
<thardin>
the actual error is Ambient viewing environment side data missing
<mkver>
This is a transcode test: It remuxes a file and then reads the generated file again (once with ffmpeg and once with ffprobe). At what stage does the missing side data happen?
<mkver>
The other warnings you mentioned can be ignored.
<thardin>
second stage I think
<thardin>
there's two ffmpeg sessions in the error log