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
<cone-795>
ffmpeg Andreas Rheinhardt master:cd63dab55cde: avcodec/mips/ac3dsp_mips: Add missing includes
<cone-795>
ffmpeg Andreas Rheinhardt master:347a70f101be: avcodec/pcm-bluray/dvd: Use correct pointer types on BE
<cone-795>
ffmpeg Andreas Rheinhardt master:72f4f1dafbdd: swscale/ppc/swscale_altivec: Fix build with -O0
<cone-795>
ffmpeg Andreas Rheinhardt master:b49e621c8320: swscale/ppc/swscale_altivec: Simplify macro
<cone-795>
ffmpeg Andreas Rheinhardt master:a265e8ca9266: avcodec, avfilter: Don't use "" for system headers
<cone-795>
ffmpeg Andreas Rheinhardt master:c8549d480f75: avcodec/msmpeg4: Don't include x86-specific header unconditionally
<cone-795>
ffmpeg Andreas Rheinhardt master:ad1cef04a9b2: swscale/swscale_internal: Hoist branch out of loop
<cone-795>
ffmpeg Andreas Rheinhardt master:2d38141ea610: swscale/swscale_internal: Don't export internal function
<cone-795>
ffmpeg Andreas Rheinhardt master:b616be1649f9: lib*/version: Use static_assert for static asserts
<cone-795>
ffmpeg Andreas Rheinhardt master:790f79384439: avutil/common: Don't auto-include mem.h
<cone-795>
ffmpeg Andreas Rheinhardt master:e4e6377afcd1: avcodec/arm/mpegvideo_arm: Use static_assert to check offsets
<cone-795>
ffmpeg Andreas Rheinhardt master:8041a91a32be: avutil/internal: Move libm inclusion to the beginning
<cone-795>
ffmpeg Andreas Rheinhardt master:0f78b26e9caa: avutil/internal: Move FF_MEMORY_POISON to its only user
<cone-795>
ffmpeg Andreas Rheinhardt master:888e2c0a85d9: avutil/hwcontext_vulkan: Include hwcontext.h
<cone-795>
ffmpeg Andreas Rheinhardt master:0e5f71230a16: avutil/internal: Move avpriv_set_systematic_pal2 decl to imgutils_internal.h
<cone-795>
ffmpeg Andreas Rheinhardt master:199c479b9ac0: avcodec/arm/vp8: Don't discard const
rvalue has quit [Ping timeout: 264 seconds]
rvalue has joined #ffmpeg-devel
thilo has quit [Ping timeout: 255 seconds]
thilo has joined #ffmpeg-devel
rvalue has quit [Read error: Connection reset by peer]
rvalue has joined #ffmpeg-devel
<Lynne>
I know the obvious solution is to use git worktrees
<Lynne>
but I wish git had a central database where all fetches were cached, so you could do global cherry-picking
iive has quit [Quit: They came for me...]
IndecisiveTurtle has quit [Ping timeout: 268 seconds]
<andrewrk>
curious, the volume filter reports ERANGE if you use av_opt_set_double, but it works fine with the same value passed as a string
<mkver>
andrewrk: The volume option supports expression and therefore an AV_OPT_TYPE_STRING.
<mkver>
The reason for ERANGE is that the min and max values are not explicitly set and therefore zeroed; otherwise you would get an AVERROR(EINVAL).
MisterMinister has quit [Ping timeout: 268 seconds]
Livio has quit [Ping timeout: 268 seconds]
Livio has joined #ffmpeg-devel
Livio has quit [Ping timeout: 260 seconds]
kepstin has quit [Remote host closed the connection]
kepstin has joined #ffmpeg-devel
andream has joined #ffmpeg-devel
Traneptora has quit [Quit: Quit]
<kierank>
So apparently gpac will be selling their wares on the FFmpeg booth
<kierank>
Great job!
<JEEB>
funky
Martchus_ has quit [Quit: ZNC 1.6.3 - http://znc.in]
HarshK23 has joined #ffmpeg-devel
IndecisiveTurtle has joined #ffmpeg-devel
jamrial has joined #ffmpeg-devel
<thardin>
Lynne: we discussed this a bit in the local hackerspace channel, it's possible to do similar nasty stuff in make if you can get say CC=foo;bar into the build system somehow
<thardin>
to prevent this kind of stuff in the future some other build system with a rich type system is probably necessary
<thardin>
it might even be necessary to get it deep into the operating system itself
<Lynne>
does meson's way of making variables immutable help?
<thardin>
not familiar with meson so I can't say
<Lynne>
it doesn't allow you to set a variable to a (list of) strings, and then overwrite the variable with something of a different type (a number or dependency)
nitroxis has quit []
nitroxis has joined #ffmpeg-devel
<thardin>
part of the problem is that parsing and evaluation are mixed
<thardin>
taking bash as an example, there's no way to know how foo | $BAR will parse until execution actually gets there
<thardin>
no guarantees can be provided as to what commands will actually be run
<thardin>
m4 appears to have the same issue
<Gramner>
meson is definitely more restrictive in what you can do. while execution of random binaries is possible it requires a lot more verbosity and would be difficult to hide in the same way
<thardin>
oh also foo $BAR is not safe either, and such patterns proliferate
<Lynne>
what would have made m4 safer is if it wasn't read-only
<Lynne>
*write-only
Martchus has joined #ffmpeg-devel
<Daemon404>
they also wouldnt have had to roll their own tarballs if they didnt use autoconf
<Daemon404>
as they could have used github's tarball support for tags
Raz- has joined #ffmpeg-devel
<Lynne>
archives that github generates are only stable for a year
<BtbN>
They also don't contain submodules, and don't clean up git-noise
<BtbN>
It's very common and normal to upload your own release archives, specially if you sign them, you HAVE to
<jamrial>
i wish more projects signed the tar rather than the tar.{gz,xz,zst}
<Lynne>
.tar.gz isn't necessarily a tar file in a gzip file
<Daemon404>
Lynne, TIL. dunno why i expectedd them to be sane.
<aaabbb>
why wouldn't .tar.gz be a tar file in a gzip?
mkver has quit [Ping timeout: 256 seconds]
<j-b>
good morning
* another|
grumbles
<j-b>
why grumbl?
<another|>
very little sleep; laptop suspend broken
Marth64 has joined #ffmpeg-devel
<jamrial>
happy bunny day
<Marth64>
yesss
<j-b>
bunny and chocolate
<Marth64>
ffmpeg in las vegas real deal?
<another|>
spin the wheel to win a free codec!
<Marth64>
rofl
<Marth64>
the airport itself is a casino
stevenliu has quit [Remote host closed the connection]
stevenliu has joined #ffmpeg-devel
<Marth64>
maybe next yr will go
mkver has joined #ffmpeg-devel
kurosu has joined #ffmpeg-devel
rvalue has quit [Read error: Connection reset by peer]
rvalue has joined #ffmpeg-devel
MisterMinister has joined #ffmpeg-devel
Warcop has quit [Remote host closed the connection]
Marth64 has quit [Remote host closed the connection]
aljazmc has joined #ffmpeg-devel
IndecisiveTurtle has quit [Ping timeout: 268 seconds]
Mister_D has joined #ffmpeg-devel
Mister_D has quit [Remote host closed the connection]
MisterMinister has quit [Ping timeout: 255 seconds]
Livio has joined #ffmpeg-devel
rvalue has quit [Ping timeout: 264 seconds]
rvalue has joined #ffmpeg-devel
<BtbN>
Can AV_OPT_CONST be set up to reject unknown values?
<BtbN>
or rather, an INT opt to only allow those constant values
Warcop has joined #ffmpeg-devel
<jamrial>
BtbN: set the range to cover only those
<BtbN>
There are holes
<jamrial>
ah, then no
<jamrial>
other than a manual check
<BtbN>
I'm tempted to propose a new flag for that
<BtbN>
Cause Nvidia introduced a bunch of new options, which all have holes
<BtbN>
They just put the autoselect-value at the very end with a huge "invalid values" hole in front
AbleBacon has joined #ffmpeg-devel
<Lynne>
michaelni: you still want something fun for 1st of april to put on the web?
<nevcairiel>
please dont, its a thing that needs to stop
Raz- has quit [Ping timeout: 252 seconds]
<jamrial>
yeah, lets not
<michaelni>
Lynne, depends on what the community prefers
<Lynne>
yeah, let's not, I've never liked pranks on the 1st
<Gramner>
agree with the "no" crowd, keep such things to silly jokes between friends and family if jou want to do it. when done by companies/projects/organizations it's just annoying, unprofessional, and a source of confusion when some quote or snippet shows up out-of-context 6 months down the line
SystemError has quit [Ping timeout: 260 seconds]
qeed_ has joined #ffmpeg-devel
qeed has quit [Ping timeout: 260 seconds]
<JEEB>
there, added my vote into the ongoing vote
aljazmc has quit [Remote host closed the connection]
aljazmc has joined #ffmpeg-devel
aljazmc has quit [Remote host closed the connection]
aljazmc has joined #ffmpeg-devel
System_Error has joined #ffmpeg-devel
<Lynne>
thanks for the reminder
Livio has quit [Ping timeout: 256 seconds]
kurosu has quit [Quit: Connection closed for inactivity]
cone-857 has joined #ffmpeg-devel
<cone-857>
ffmpeg Timo Rothenpieler master:43b417d516b0: avcodec/nvenc: stop using long deprecated format specifiers
<cone-857>
ffmpeg Timo Rothenpieler master:06c2a2c425f2: avcodec/nvenc: support SDK 12.2 bit depth API
<cone-857>
ffmpeg Timo Rothenpieler master:64e3fc906971: avcodec/nvenc: add support for HEVC temporal filtering
<cone-857>
ffmpeg Timo Rothenpieler master:21b470896941: avcodec/nvenc: add support for unidirectional b-frames
<cone-857>
ffmpeg Timo Rothenpieler master:77d23bcb1b91: avcodec/nvenc: add support for lookahead_level
<cone-857>
ffmpeg Timo Rothenpieler master:a27096611835: avcodec/nvenc: add support for uhq tune
<cone-789>
ffmpeg Timo Rothenpieler master:271a0a55bcc5: avcodec/nvenc: provide nvenc with monotonic frame index
<cone-789>
ffmpeg Timo Rothenpieler release/7.0:5ff5a431c758: avcodec/nvenc: stop using long deprecated format specifiers
<cone-789>
ffmpeg Timo Rothenpieler release/7.0:4c5a809388f4: avcodec/nvenc: support SDK 12.2 bit depth API
<BtbN>
What's the status with the 7.0 branch? Is it still fine to backport new features, or already fixed-only?
<jamrial>
only bug fixes
<jamrial>
we do a minor bump dance on branches, so feature freeze
IndecisiveTurtle has quit [Remote host closed the connection]
<cone-789>
ffmpeg Timo Rothenpieler release/6.1:3481f8d99fc4: avcodec/nvenc: stop using long deprecated format specifiers
<cone-789>
ffmpeg Timo Rothenpieler release/6.1:38346298973e: avcodec/nvenc: support SDK 12.2 bit depth API
AbleBacon_ has joined #ffmpeg-devel
AbleBacon has quit [Ping timeout: 264 seconds]
HarshK23 has quit [Quit: Connection closed for inactivity]
IndecisiveTurtle has joined #ffmpeg-devel
<IndecisiveTurtle>
Lynne: Added simple deinterleave shader in the repo. Also any update regarding the other student? If the situation is still pending, I would like to work on vc2 encoder and prepare the proposal, as there are about 48 hours left before the deadline. If it has been decided they will work on that, I can do jpeg2000 instead, but I need to know what to write about.
<Lynne>
IndecisiveTurtle: no news, go ahead with your plan