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 has been released! | Please read ffmpeg.org/developer.html#Code-of-conduct
System_Error has quit [Ping timeout: 260 seconds]
System_Error has joined #ffmpeg-devel
iive has quit [Quit: They came for me...]
thilo has quit [Ping timeout: 256 seconds]
thilo has joined #ffmpeg-devel
thilo has quit [Changing host]
thilo has joined #ffmpeg-devel
haihao has quit [Ping timeout: 256 seconds]
haihao has joined #ffmpeg-devel
System_Error has quit [Remote host closed the connection]
IndecisiveTurtle has quit [Ping timeout: 256 seconds]
haihao has quit [Ping timeout: 246 seconds]
haihao has joined #ffmpeg-devel
AbleBacon has quit [Read error: Connection reset by peer]
jamrial has quit []
Martchus_ has joined #ffmpeg-devel
Martchus has quit [Ping timeout: 268 seconds]
<Lynne>
updated my usac branch with proper functional preroll parsing
<Lynne>
only thing left is the stereo prediction, but I'll get some sleep first
HarshK23 has joined #ffmpeg-devel
deer3 has quit [Ping timeout: 240 seconds]
deer3 has joined #ffmpeg-devel
chipmunk_snack has joined #ffmpeg-devel
haihao has quit [Ping timeout: 246 seconds]
haihao has joined #ffmpeg-devel
microchip__ has joined #ffmpeg-devel
microchip_ has quit [Ping timeout: 256 seconds]
microchip__ is now known as microchip_
microchip_ has quit [Quit: There is no spoon!]
microchip_ has joined #ffmpeg-devel
chipmunk_snack has quit [Quit: Client closed]
Krowl has joined #ffmpeg-devel
dkozinski has joined #ffmpeg-devel
dkozinski is now known as wellsakus
chainik1 has quit [Quit: (╯°□°）╯︵ ┻━┻]
Krowl has quit [Read error: Connection reset by peer]
Krowl has joined #ffmpeg-devel
marcj- has joined #ffmpeg-devel
marcj has quit [Ping timeout: 252 seconds]
microchip_ has quit [Quit: There is no spoon!]
microchip_ has joined #ffmpeg-devel
mkver has joined #ffmpeg-devel
ngaullier has joined #ffmpeg-devel
Teukka has quit [Read error: Connection reset by peer]
Teukka has joined #ffmpeg-devel
Teukka has quit [Changing host]
Teukka has joined #ffmpeg-devel
Krowl has quit [Read error: Connection reset by peer]
dionisis has joined #ffmpeg-devel
Krowl has joined #ffmpeg-devel
chenjy has joined #ffmpeg-devel
<chenjy>
hi
chenjy has quit [Quit: Client closed]
Sean_McG has quit [Remote host closed the connection]
chenjy has joined #ffmpeg-devel
<chenjy>
if i want to add a new architecture branch into ffmpeg,and make community support it,what should i do?
chenjy has quit [Quit: Client closed]
chenjy has joined #ffmpeg-devel
paulk has quit [Ping timeout: 268 seconds]
<sfan5>
whoever documented that command on https://ffmpeg.org/developer.html that allows sending patches via a normal email client saved my day
paulk has joined #ffmpeg-devel
paulk has joined #ffmpeg-devel
paulk has quit [Changing host]
chenjy has quit [Client Quit]
chenjy has joined #ffmpeg-devel
Krowl has quit [Read error: Connection reset by peer]
paulk has quit [Ping timeout: 256 seconds]
paulk has joined #ffmpeg-devel
paulk has joined #ffmpeg-devel
paulk has quit [Changing host]
<mkver>
sfan5: SoftWorks in 89ba6e9a04d285
chenjy has quit [Quit: Client closed]
chenjy has joined #ffmpeg-devel
chenjy has quit [Client Quit]
Krowl has joined #ffmpeg-devel
chainik1 has joined #ffmpeg-devel
mkver has quit [Ping timeout: 255 seconds]
mkver has joined #ffmpeg-devel
pheo has quit [Ping timeout: 252 seconds]
pheo has joined #ffmpeg-devel
ccawley2011 has joined #ffmpeg-devel
Krowl has quit [Read error: Connection reset by peer]
System_Error has quit [Quit: Leaving]
System_Error has joined #ffmpeg-devel
cone-393 has joined #ffmpeg-devel
<cone-393>
ffmpeg Andreas Rheinhardt master:99135a2f8bb9: tests/checkasm/llviddsp: Use correct function pointer type
<cone-393>
ffmpeg Andreas Rheinhardt master:895dd370a262: tests/checkasm/hevc_*: Fix funtion pointer types
<cone-393>
ffmpeg Andreas Rheinhardt master:dff0a1557eab: tests/checkasm/llauddsp: Avoid UB integer overflow
<cone-393>
ffmpeg Andreas Rheinhardt master:631636e58216: tests/checkasm/huffyuvdsp: Use correct function pointer type
<cone-393>
ffmpeg Andreas Rheinhardt master:6c0994864ef6: tests/checkasm/vp8dsp: Use correct function pointer type
<cone-393>
ffmpeg Andreas Rheinhardt master:dcbdcc3bf55a: tests/checkasm/motion: Use correct function pointer type
<cone-393>
ffmpeg Andreas Rheinhardt master:e3de22e30771: tests/checkasm/vvc_mc: Use correct function pointer type
<cone-393>
ffmpeg Andreas Rheinhardt master:4c57cbda6796: tests/checkasm/vf_colorspace: Use correct function pointer type
<cone-393>
ffmpeg Andreas Rheinhardt master:caec57eb3ab7: tests/checkasm/vf_bwdif: Use correct function pointer type
<cone-393>
ffmpeg Andreas Rheinhardt master:9126705e6e4a: tests/checkasm/vf_blend: Update function type
<cone-393>
ffmpeg Andreas Rheinhardt master:784672b83351: tests/checkasm/sw_gbrp: Use correct function types for calls
rvalue- has joined #ffmpeg-devel
rvalue has quit [Ping timeout: 240 seconds]
jamrial has joined #ffmpeg-devel
lexano has joined #ffmpeg-devel
rvalue- is now known as rvalue
Krowl has joined #ffmpeg-devel
ngaullier has quit [Ping timeout: 268 seconds]
Krowl has quit [Read error: Connection reset by peer]
haihao has quit [Ping timeout: 246 seconds]
haihao has joined #ffmpeg-devel
Krowl has joined #ffmpeg-devel
haihao has quit [Ping timeout: 252 seconds]
haihao has joined #ffmpeg-devel
System_Error has quit [Read error: Connection reset by peer]
cone-393 has quit [Quit: transmission timeout]
feiwan1 has quit [Ping timeout: 260 seconds]
feiwan1 has joined #ffmpeg-devel
<courmisch>
seems my BPI-F3 board got lost in Superalko
<courmisch>
no news since leaving Kaunas 48h ago
<courmisch>
it should take 11h plus ferry waiting time
cone-049 has joined #ffmpeg-devel
<cone-049>
ffmpeg Rémi Denis-Courmont master:88d973a5d658: lavc/flacdsp: R-V V flac_wasted33
<cone-049>
ffmpeg sunyuechi master:d521b7280c2a: lavc/vp9dsp: R-V V ipred tm
MisterMinister has joined #ffmpeg-devel
Krowl has quit [Read error: Connection reset by peer]
<BtbN>
Again someone who claims the schannel backend is buggy and wants to use openssl instead on Windows.
<BtbN>
That's the third of fourth time someone said that, I asked all of them to please provide insight into what's buggy and how to reproduce it. But nobody ever did so far.
<BtbN>
And I could never find an issue with it
Sean_McG has joined #ffmpeg-devel
<Lynne>
figured out the noise synth issue, now getting compliant mono output
<Lynne>
should have stereo working tonight
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
j45_ has joined #ffmpeg-devel
j45 has quit [Ping timeout: 268 seconds]
j45_ is now known as j45
j45 has quit [Changing host]
j45 has joined #ffmpeg-devel
Sean_McG has quit [Quit: leaving]
j45_ has joined #ffmpeg-devel
j45 has quit [Ping timeout: 240 seconds]
j45_ is now known as j45
j45 has quit [Changing host]
j45 has joined #ffmpeg-devel
j45_ has joined #ffmpeg-devel
j45 has quit [Ping timeout: 240 seconds]
j45_ is now known as j45
j45 has quit [Changing host]
j45 has joined #ffmpeg-devel
NotWarcop has quit [Remote host closed the connection]
iive has joined #ffmpeg-devel
AbleBacon has joined #ffmpeg-devel
<courmisch>
clearly all the STF monies should go to writing checkasm
kurosu has joined #ffmpeg-devel
cone-049 has quit [Quit: transmission timeout]
<jamrial>
courmisch: btw, how does wasted33 fail for some seeds on riscv? are you not doing sign extend 32 > 64 then shift?
<JEEB>
Lynne: nice stuff :)
<Lynne>
JEEB: could you take a look at usac_ch_pos_to_av?
<JEEB>
I can attempt to, sure :)
Livio has joined #ffmpeg-devel
<Lynne>
thanks, you just need to decipher what the other 24 channels map to in ffmpeg land
<courmisch>
jamrial: widening multiply doesn't work if the shift is 32+, which depends on RNG
<courmisch>
jamrial: since the scalar factor overflows to 0
<courmisch>
jamrial: so for shift up to 31, we do 32x32 widening multiply (faster). otherwise now fallback to 64-bit extension then 64-bit shift
HarshK23 has quit [Quit: Connection closed for inactivity]
Livio has quit [Ping timeout: 260 seconds]
cone-481 has joined #ffmpeg-devel
<cone-481>
ffmpeg James Almer master:727a603158b4: checkasm/h264dsp: use int64_t scale values
<BtbN>
Why does flvenc do this horrible thing of calculating the flags size ahead of time, instead of doing it while writing it? This duplicates so much logic.
<BtbN>
flv_write_codec_header even calculates the size. Or is that slow, cause it seeks around?
<nevcairiel>
for streaming in particular you dont really want to go back and update the size, in the best case you add latency, in the worst case you cant update it :D
<BtbN>
don't even need to go back and update it from the looks of flv_write_packet()
<BtbN>
flags_size is never written or anything from what I can tell
<BtbN>
I don't immediately see why it couldn't just calculate it using the cur_offset it already saves right after writing them
kurosu has quit [Quit: Connection closed for inactivity]
<BtbN>
hm, no. It does write flags_size before actually writing them flags
<BtbN>
This will be a mess, cause the size of the flags is dynamic now
<BtbN>
It doesn't look like a valid flv tag to me either
<mkver>
BtbN: FLV_CODECID_VP6A == 5
<BtbN>
ah, right. Those aren't shifted.
<BtbN>
But what on earth is it doing.
<mkver>
I don't know.
<mkver>
I don't know whether FLV_CODECID_VP6A was intended in that line or something else.
<BtbN>
It's setting some magic "reserved" field to 5. Which corresponds to the StreamID field according to the flv spec, which is always supposed to be 0
<mkver>
michaelni: Did I send a patch for something that you already claimed?
<michaelni>
mkver, i think not but i dont know what to look at next because i was going over file by file and was getting towards the end of libavcodec and thats exactly where you seem to have worked on now
<michaelni>
also i think you need to contact thilo & Pierre-Anthony if you want to claim STF funds for the work you do/did
AbleBacon has quit [Read error: Connection reset by peer]
<BtbN>
Interesting, the enhanced flv stuff does not seem to signal sample rate and stuff at all anymore, outside of the onMetaData, which ffmpeg ignores unless told not to.
<BtbN>
But they did add rather advanced channel layout stuff for audio
ccawley2011 has quit [Read error: Connection reset by peer]
<BtbN>
Weird, muxing opus works fine, muxing flac also works fine, but fails at the very end, with "Empty audio Packet", cause flv_write_packet() got an AVPacket with !size.
<BtbN>
Is that something that just happens for flac? Using the native encoder.
<BtbN>
Why is it bad specifically for audio? Removing the check does indeed make it run into an assert.
<BtbN>
Removing both the assert and the check also makes stuff work just fine. I see no immediate reason why a size of 0 would be dangerous in any way. It'd just write an empty package.
<BtbN>
michaelni: do you by any chance still remember anything about that? It originates back to a 20 year old svn commit.
<BtbN>
Just removing the assert works just fine with empty packets, so I'm inclined to just go that route. movenc muxes that packet without complaint as well
IndecisiveTurtle has joined #ffmpeg-devel
<Lynne>
zero-length packets are valid in audio land
<Lynne>
iirc opus assumes a packet was lost and activates recovery, that's why it has an internal silence field, with an extremely low probability so it wastes as few bytes as possible
<iive>
BtbN, do we even have to write a zero length packet? can't we just ignore it?
<BtbN>
No idea, all I know is flacenc finishes the stream with one
<BtbN>
imo the muxer should just do as told there, it has no business knowing what the codec needs or not
<iive>
BtbN, e.g. audio filters sometimes accumulate samples (lag) so at the end of playback a zero length is pushed to flush the remaining samples.