2025-03-03 01:04
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
00:12
mkver has joined #ffmpeg-devel
00:20
BradleyS has quit [Read error: Connection reset by peer]
00:21
BradleyS has joined #ffmpeg-devel
00:22
BradleyS has quit [Client Quit]
00:24
BradleyS has joined #ffmpeg-devel
00:59
iive has quit [Quit: They came for me...]
01:10
___nick___ has joined #ffmpeg-devel
01:10
___nick___ has quit [Client Quit]
01:12
___nick___ has joined #ffmpeg-devel
01:25
sepro3 has joined #ffmpeg-devel
01:27
sepro has quit [Ping timeout: 244 seconds]
01:27
sepro3 is now known as sepro
01:46
sepro9 has joined #ffmpeg-devel
01:48
sepro has quit [Ping timeout: 252 seconds]
01:48
sepro9 is now known as sepro
01:55
System_Error has joined #ffmpeg-devel
02:01
sepro1 has joined #ffmpeg-devel
02:04
sepro has quit [Ping timeout: 276 seconds]
02:04
sepro1 is now known as sepro
02:43
mkver has quit [Remote host closed the connection]
02:44
mkver has joined #ffmpeg-devel
03:05
Anthony_ZO has joined #ffmpeg-devel
03:20
jamrial has quit []
03:22
Martchus has joined #ffmpeg-devel
03:23
Martchus_ has quit [Ping timeout: 252 seconds]
03:41
HarshK23 has joined #ffmpeg-devel
04:23
cone-758 has joined #ffmpeg-devel
04:23
<
cone-758 >
ffmpeg Jack Lau master:177b92df2b05: avformat/tls_openssl: fix warnings when openssl is lower version
04:23
<
cone-758 >
ffmpeg Jack Lau master:214248e2d6c6: avformat/whip: mark as experimental
04:52
_whitelogger has joined #ffmpeg-devel
04:53
philipl has joined #ffmpeg-devel
04:58
<
cone-758 >
ffmpeg Jack Lau master:8aa6df663e76: avformat/whip: replace AV_OPT_FLAG_DECODING_PARAM to ENCODING
05:00
System_Error has quit [Remote host closed the connection]
05:06
System_Error has joined #ffmpeg-devel
05:44
kurosu has joined #ffmpeg-devel
05:44
Richardcavell_ has joined #ffmpeg-devel
06:06
_whitelogger has joined #ffmpeg-devel
06:18
Richardcavell_ has quit [Ping timeout: 260 seconds]
07:19
TheVibeCoder has joined #ffmpeg-devel
07:58
cone-758 has quit [Quit: transmission timeout]
08:02
mkver has quit [Remote host closed the connection]
08:03
mkver has joined #ffmpeg-devel
08:04
kurosu has quit [Quit: Connection closed for inactivity]
08:19
Raz- has quit [Remote host closed the connection]
09:07
<
beastd >
wbs: BTW I wanted to say I love your commit messages! They are among the best I have ever read. Also I see you often add comments in the code as well; which is often left out by people that write good commit messages but it's really needed because not everyone will check the commit message history on every change of a line.
09:17
<
TheVibeCoder >
best commit messages are: fix bug
09:21
<
beastd >
by far biggest problem with commit messages is "missing why"
09:48
<
TheVibeCoder >
what happened to risc guy?
09:54
<
BtbN >
At work people only know two types of commits: "Fix", "Fix", "Another Fix", "Typo", "Fix", "qewauikjfdhbnsaeiufbwsaerfg", "Ahhh"
09:54
<
BtbN >
That then gets squash-merged into one giant -5000+5000 line commit that has all those helpfull messages in its commit message
09:55
<
BtbN >
Or they write such a giant commit themselves already, and give it the most minimal message possible.
10:12
TheVibeCoder has quit [Changing host]
10:12
TheVibeCoder has joined #ffmpeg-devel
10:17
mkver has quit [Ping timeout: 265 seconds]
10:18
mkver has joined #ffmpeg-devel
10:25
mkver has quit [Ping timeout: 244 seconds]
10:25
Anthony_ZO has quit [Ping timeout: 260 seconds]
10:25
<
wbs >
beastd: thanks for noticing :-)
10:39
Richardcavell_ has joined #ffmpeg-devel
10:53
<
kasper93 >
I don't like squash&merge which takes "PR/MR" description as a base
10:53
<
kasper93 >
destroying all commit history
10:53
<
kasper93 >
works fir "fix", "fix", "fix" crowd
10:53
<
kasper93 >
but if you care about commits, you get punished
10:54
<
kasper93 >
Speaking of, my top commit message is (dot)
10:58
bsFFFFFF has joined #ffmpeg-devel
11:03
<
kasper93 >
blender is hardcoding avcodec ids in thier headers, that's crazy
11:13
usagi_mimi has quit [Quit: WeeChat 4.6.3]
11:14
mkver has joined #ffmpeg-devel
11:24
usagi_mimi has joined #ffmpeg-devel
11:26
mkver has quit [Ping timeout: 276 seconds]
11:27
mkver has joined #ffmpeg-devel
11:28
<
BtbN >
Yeah, they also merge my PRs that way, even though I took great care to make atomic commits with sensible messages
11:28
<
BtbN >
they then all get squahsed into one commit...
11:32
mkver has quit [Ping timeout: 265 seconds]
12:17
minimal has joined #ffmpeg-devel
12:26
Richardcavell_ has quit [Ping timeout: 248 seconds]
13:26
<
TheVibeCoder >
BtbN: at work? how so that company still alive?
13:31
Richardcavell_ has joined #ffmpeg-devel
13:32
bsFFFFFF has quit [Ping timeout: 272 seconds]
13:38
bsFFFFFF has joined #ffmpeg-devel
13:48
<
BtbN >
It's Science-Stuff
13:50
<
aaabbb >
haasn: i was just about to ask "bumblebee, is that the project that deleted everyone's /usr?" and then i actually checked the commit xD
13:51
<
BtbN >
Steam managed that as well at one point
13:51
<
BtbN >
Was / even, not "just" /usr
13:51
<
BtbN >
They assumed some variable was set, but it wasn't in all environment
13:51
<
BtbN >
and then ran rm -rf /$SOMEVAR
13:52
<
aaabbb >
that must be old, since at least gnu coreutils will refuse to act unless --no-preserve-root is passed
13:52
<
BtbN >
It somehow got around it
13:52
<
BtbN >
it also didn't run as root
13:52
<
BtbN >
so the only thing it ended up actually deleting was the users homedir
13:53
<
BtbN >
And whatever else that user had the permission to delete
13:53
<
aaabbb >
i always set -u when i have rm -r's that act on variables, just in case
13:53
<
BtbN >
that option is not documented?
13:54
<
BtbN >
oh, literally "set -u", not rm -u
13:54
<
aaabbb >
like in teh shell
13:55
<
aaabbb >
yeah the thing that bails if an unset var is referenced
13:55
<
BtbN >
Yeah, that should probably be default in any script anyway
14:00
microlappy has joined #ffmpeg-devel
14:01
microlappy has quit [Client Quit]
14:08
Kei_N_ has joined #ffmpeg-devel
14:11
Kei_N has quit [Ping timeout: 260 seconds]
15:06
kasper93 has quit [Ping timeout: 260 seconds]
15:07
kasper93 has joined #ffmpeg-devel
15:16
kurosu has joined #ffmpeg-devel
15:31
cone-182 has joined #ffmpeg-devel
15:31
<
cone-182 >
ffmpeg Niklas Haas master:2679e687ec07: avfilter/f_ebur128: use structs for biquad weights
15:31
<
cone-182 >
ffmpeg Niklas Haas master:c5f3033a8638: avfilter/f_ebur128: use transformed direct form II
15:31
<
cone-182 >
ffmpeg Niklas Haas master:1da0a70b098b: avfilter/f_ebur128: use a single packed array for the integrator cache
15:31
<
cone-182 >
ffmpeg Niklas Haas master:2d9435d76f91: avfilter/f_ebur128: simplify sample cache array
15:31
<
cone-182 >
ffmpeg Niklas Haas master:deab15e76a3e: avfilter/f_ebur128: split off C implementation to separate function
15:31
<
cone-182 >
ffmpeg Niklas Haas master:c7d2b0a7a1dc: avfilter/f_ebur128: move weights and cache to EBUR128DSPContext
15:31
<
cone-182 >
ffmpeg Niklas Haas master:53e03ec8afa8: avfilter/x86/f_ebur128: add x86 AVX implementation
15:31
<
cone-182 >
ffmpeg Niklas Haas master:a96175e76f2c: avfilter/f_ebur128: remove pointless macro
15:31
<
cone-182 >
ffmpeg Niklas Haas master:229393d8dc8f: avfilter/f_ebur128: move true peak calculation out of main loop
15:31
<
cone-182 >
ffmpeg Niklas Haas master:f362bacd27c8: avfilter/f_ebur128: lift sample peak calculation out of main loop
15:31
<
cone-182 >
ffmpeg Niklas Haas master:4c046517e7a7: avfilter/f_ebur128: move variable declarations to usage site
15:31
<
cone-182 >
ffmpeg Niklas Haas master:3b26b782eede: avfilter/f_ebur128: move peak detection to reusable DSP function
15:31
<
cone-182 >
ffmpeg Niklas Haas master:daef34857489: avfilter/x86/f_ebur128: implement AVX peak calculation
15:43
<
kasper93 >
regarding g.v.o/ffmpeg repo, servers seems to be not able to index whole repo. git clone fails
15:43
<
kasper93 >
--depth 1 works
15:44
<
kasper93 >
not sure if it's just slow, or need git gc
15:44
usagi_mimi has quit [Ping timeout: 252 seconds]
15:45
usagi_mimi has joined #ffmpeg-devel
15:48
<
kasper93 >
is anyone here with access there? I can ask on #videolan (probably low priority, but maybe nice to fix anyway)
15:59
<
JEEB >
I think that repo is maintained by videolan, so it'd probably be correct to poke there
16:02
elvis_a_presley has quit [Quit: smoke-bomb ; grapple-hook]
16:02
elvis_a_presley has joined #ffmpeg-devel
16:06
<
nevcairiel >
thresh is usually the person to ask, but he doesnt hang around this channel
16:57
<
beastd >
nevcairiel,BtbN: Can you by chance confirm my patch fixes your fate setups? Want to push it in in a few minutes
17:01
<
BtbN >
It's usually because AI scrappers are having a field day again, kasper93
17:01
<
BtbN >
just clone from github or git.ffmpeg.org and then change remote
17:06
bsFFFFFF has quit [Ping timeout: 260 seconds]
17:06
bsFFFFFF has joined #ffmpeg-devel
17:19
<
beastd >
Will wait with the configure patch until tomorrow. Anyone can push earlier if it is confirmed it fixes the configure hangs
17:30
System_Error has quit [Remote host closed the connection]
17:37
System_Error has joined #ffmpeg-devel
17:39
putacho has quit [Remote host closed the connection]
17:40
microchip_ has joined #ffmpeg-devel
17:41
System_Error has quit [Remote host closed the connection]
17:44
<
BtbN >
It's surprisingly annoying to do that, beastd since the scripts all will fetch and reset
17:47
System_Error has joined #ffmpeg-devel
17:54
System_Error has quit [Remote host closed the connection]
18:00
System_Error has joined #ffmpeg-devel
18:11
<
TheVibeCoder >
haasn: with your ffmpeg latest changes, i get wrong true peak (-inf) results
18:13
<
TheVibeCoder >
if i end early (pressing q key), the results are ok
18:14
mkver has joined #ffmpeg-devel
18:26
<
kasper93 >
BtbN: if you set repo to file://<dir> it won't reset
18:28
<
kasper93 >
and yeah, I have g.v.o added as remote, but noticed that full clone wouldn't work, it doesn't affect me directly
18:28
<
kasper93 >
though, I think the timeout is very low, because it response 504 quite quickly
18:31
cone-182 has quit [Quit: transmission timeout]
18:32
___nick___ has quit [Ping timeout: 276 seconds]
18:35
___nick___ has joined #ffmpeg-devel
18:36
cone-960 has joined #ffmpeg-devel
18:36
<
cone-960 >
ffmpeg Marton Balint master:606efaa2cfb2: fftools/ffmpeg_filter: simplify control flow in read_frames
18:36
<
cone-960 >
ffmpeg Marton Balint master:28a7b9c86355: avfilter/split: consume all frames before forwarding inlink status
18:36
<
cone-960 >
ffmpeg Marton Balint master:a21429e134d4: avfilter/af_aresample: make aresample return FFERROR_NOT_READY when no progress can be made
18:36
<
cone-960 >
ffmpeg Marton Balint master:ffcdd2cdc1cf: avfilter/af_aresample: merge request_frame into activate function
18:36
<
cone-960 >
ffmpeg Marton Balint master:32153fac8468: avfilter/af_aresample: rework activate logic to follow the advised flow more strictly
18:48
System_Error has quit [Ping timeout: 244 seconds]
19:07
System_Error has joined #ffmpeg-devel
19:49
<
cone-960 >
ffmpeg Alexander Strasser master:204d03c803de: configure: Fix a regression when probing link.exe
19:59
bsFFFFFF has quit [Quit: bsFFFFFF]
20:03
<
BtbN >
lets re-run fate then
20:04
___nick___ has quit [Ping timeout: 244 seconds]
20:07
<
wbs >
thanks, and sorry about it!
20:08
usagi_mimi has quit [Ping timeout: 268 seconds]
20:12
<
BtbN >
We'll need to contact all the msvc fate hosts somehow, to unstuck their boxes
20:12
<
BtbN >
not sure how many there are
20:12
<
wbs >
you and nevcairiel probably?
20:12
<
nevcairiel >
Mine resets on its own
20:13
<
nevcairiel >
But i think msvc fate are all just mine on actual windows and was on wine
20:13
<
nevcairiel >
wbs on wine*
20:14
usagi_mimi has joined #ffmpeg-devel
20:16
<
nevcairiel >
ah no kasper93 also has some now, but he said his werent affected
20:17
minimal has quit [Quit: Leaving]
20:18
<
nevcairiel >
well i manually unblocked mine so it doesnt take hours or whatever i set the timeout to
20:21
<
cone-960 >
ffmpeg Andreas Rheinhardt master:8e0f064a1170: avcodec/mpeg4videodec: Don't zero blocks twice
20:21
<
cone-960 >
ffmpeg Andreas Rheinhardt master:f5fc31acecde: avcodec/rv10: Perform RV20 initialization during init
20:21
<
cone-960 >
ffmpeg Andreas Rheinhardt master:ea4b9a121d05: avcodec/rv10: Perform RV20 check only for RV20
20:21
<
cone-960 >
ffmpeg Andreas Rheinhardt master:5507da21e663: avcodec/ituh263dec: Don't process unnecessarily many coefficients
20:21
<
cone-960 >
ffmpeg Andreas Rheinhardt master:d95041dffe61: avcodec/ituh263dec: Remove redundant store
20:21
<
cone-960 >
ffmpeg Andreas Rheinhardt master:a2272aff7699: avcodec/ituh263enc: Don't use array unnecessarily
20:21
<
cone-960 >
ffmpeg Andreas Rheinhardt master:069329ec0257: avcodec/ituh263enc: Fix indentation
20:21
<
cone-960 >
ffmpeg Andreas Rheinhardt master:c2e5f7466db0: avcodec/mpeg12dec: Don't store block_last_index unnecessarily
20:21
<
cone-960 >
ffmpeg Andreas Rheinhardt master:01d6235faefb: avcodec/rv10: Avoid indirection when reading VLC codes
20:21
<
cone-960 >
ffmpeg Andreas Rheinhardt master:00926669f775: avcodec/mpeg_er: Mark ff_mpeg_er_init() as av_cold
20:21
<
cone-960 >
ffmpeg Andreas Rheinhardt master:5b1e8056539b: avcodec/mpegvideo_enc: Remove always-true branch
20:21
<
cone-960 >
ffmpeg Andreas Rheinhardt master:109bf099d529: avcodec/motion_est: Fix indentation
20:21
<
cone-960 >
ffmpeg Andreas Rheinhardt master:1fe08d629a27: avcodec/svq1enc: Don't free scratchbuf upon error
20:21
<
cone-960 >
ffmpeg Andreas Rheinhardt master:fbb8e605ed8a: avcodec/svq1enc: Allocate motion_val{8,16} during init
20:21
<
cone-960 >
ffmpeg Andreas Rheinhardt master:58a222f561f3: avcodec/svq1enc: Stop calling ff_mpv_common_init()
20:21
<
cone-960 >
ffmpeg Andreas Rheinhardt master:4b7f1595ccb5: avcodec/svq1enc: Remove write-only c_block_{width,height}
20:21
<
cone-960 >
ffmpeg Andreas Rheinhardt master:83ddf99267c0: avcodec/svq1enc: Set MpegEncContext.avctx only once
20:21
<
cone-960 >
ffmpeg Andreas Rheinhardt master:2583c7abcf2c: avcodec/svq1enc: Don't initialize unneeded block_index
20:22
<
cone-960 >
ffmpeg Andreas Rheinhardt master:9b409ea1e615: configure: Factor mpegvideoencdsp out of mpegvideoenc
20:22
<
cone-960 >
ffmpeg Andreas Rheinhardt master:8764b2104816: configure: Relax svq1enc->mpegvideoenc dependency
20:22
<
cone-960 >
ffmpeg Andreas Rheinhardt master:da22ee281edc: configure: Relax diracdec->mpegvideoenc dependency
20:22
<
cone-960 >
ffmpeg Andreas Rheinhardt master:6f84160b82a4: configure: Relax snowenc->mpegvideoenc dependency
20:22
<
cone-960 >
ffmpeg Andreas Rheinhardt master:b4e713598659: avcodec/mpegvideo: Don't reset thread_context ptrs unnecessarily
20:22
<
cone-960 >
ffmpeg Andreas Rheinhardt master:85692e5da1ea: avcodec/mpegvideo: Don't zero unnecessarily
20:22
<
cone-960 >
ffmpeg Andreas Rheinhardt master:8f92885b1fd7: avcodec/mpegvideo: Defer init of enc slice ctxs in ff_mpv_common_init()
20:22
<
cone-960 >
ffmpeg Andreas Rheinhardt master:822be72f0a9d: avcodec/mpegvideo_enc: Simplify allocating non-slice buffers
20:22
<
cone-960 >
ffmpeg Andreas Rheinhardt master:950137a7c82a: avcodec/ituh263dec: Simplify AIC handling
20:22
<
cone-960 >
ffmpeg Andreas Rheinhardt master:7643269ec20e: avcodec/ituh263enc: Simplify AIC handling
20:22
<
cone-960 >
ffmpeg Andreas Rheinhardt master:61d34c2b8eac: avcodec/flvenc: Combine writing bits
20:22
<
cone-960 >
ffmpeg Andreas Rheinhardt master:514e5ea0db84: avcodec/flvenc: Move ff_flv2_encode_ac_esc() to ituh263enc.c
20:22
<
cone-960 >
ffmpeg Andreas Rheinhardt master:13527b39bb1c: avcodec/mpegvideo: Redo resetting intra table entry
20:22
<
cone-960 >
ffmpeg Andreas Rheinhardt master:32cbadcd74a3: avcodec/mpegvideo: Avoid {a,d}c_val[{1,2}]
20:22
<
cone-960 >
ffmpeg Andreas Rheinhardt master:dcd7408c9200: avcodec/mpegvideo: Provide alignment hint to compiler
20:22
<
cone-960 >
ffmpeg Andreas Rheinhardt master:70ce68d0212f: avcodec/mpegvideo: Don't reset AC values of upper-left luma block
20:22
<
cone-960 >
ffmpeg Andreas Rheinhardt master:b2661569b129: avcodec/mpegvideo: Combine stores
20:22
<
cone-960 >
ffmpeg Andreas Rheinhardt master:4aca6b19ac84: avcodec/mpeg4video: Don't use ac_val[1], ac_val[2] when cleaning buffers
20:22
<
cone-960 >
ffmpeg Andreas Rheinhardt master:dba1061ba456: avcodec/mpegvideo: Only use a single ac_val pointer
20:22
<
cone-960 >
ffmpeg Andreas Rheinhardt master:c4ebe55f94ff: avcodec/mpegvideo: Allocate ac_val jointly
20:22
<
cone-960 >
ffmpeg Andreas Rheinhardt master:c4fd33a95159: avcodec/mpegvideo: Zero-init mbintra_table
20:22
<
cone-960 >
ffmpeg Andreas Rheinhardt master:81ab1106c81a: avcodec/mpegvideo: Only keep MpegEncContext.dc_val[0]
20:22
<
cone-960 >
ffmpeg Andreas Rheinhardt master:eb597d9a1e0e: avcodec/vc1: Only keep mb_type[0]
20:22
<
cone-960 >
ffmpeg Andreas Rheinhardt master:c84a03d75200: avcodec/mpegvideo: Allocate dc_val for each encoder slice
20:22
<
cone-960 >
ffmpeg Andreas Rheinhardt master:56fd45d2a938: avcodec/{ituh263,mpeg4video}enc: Simplify out-of-slice DC prediction
20:22
<
cone-960 >
ffmpeg Andreas Rheinhardt master:689cde75ee6f: avcodec/mpeg4video: Move ff_mpeg4_pred_dc() to decoder
20:22
<
cone-960 >
ffmpeg Andreas Rheinhardt master:f8c0ac9984e3: avcodec/ituh263enc: Simplify encoding umotion vectors
20:22
<
cone-960 >
ffmpeg Andreas Rheinhardt master:5cdce56a7951: avcodec/mpegvideo: Reduce stack usage when copying MpegEncContext
20:22
<
cone-960 >
ffmpeg Andreas Rheinhardt master:34953e195fe1: avcodec/mpeg4videoenc: Fix data race when using AC prediction
20:23
usagi_mimi has quit [Ping timeout: 268 seconds]
20:23
kasper93 has quit [Read error: Connection reset by peer]
20:37
rvalue has quit [Changing host]
20:37
rvalue has joined #ffmpeg-devel
20:46
rvalue has quit [Quit: reconnecting]
20:46
rvalue has joined #ffmpeg-devel
21:01
kasper93 has joined #ffmpeg-devel
21:17
Richardcavell_ has quit [Ping timeout: 248 seconds]
21:24
<
TheVibeCoder >
mkver: the dfpwmenc encoder may produce non-bitexact output, because it will consume uninitialized data
21:26
<
mkver >
When the number of input samples is not divisible by 8?
21:30
<
beastd >
wbs: thanks for pushing
22:11
RT|AO_ has joined #ffmpeg-devel
22:14
RT|AO has quit [Ping timeout: 276 seconds]
22:49
TheVibeCoder has quit [Ping timeout: 248 seconds]
22:57
mkver has quit [Remote host closed the connection]
23:00
kurosu has quit [Quit: Connection closed for inactivity]
23:25
_whitelogger has joined #ffmpeg-devel
23:28
cone-285 has joined #ffmpeg-devel
23:28
<
cone-285 >
ffmpeg Peter Ross master:08943c82c629: configure: cbs_apv depends on cbs
23:28
<
cone-285 >
ffmpeg Peter Ross master:b49c4d1034d0: configure: h264_vulkan_encoder depends on atsc_a53
23:28
<
cone-285 >
ffmpeg Peter Ross master:baad40bdd1e9: avcodec/mips: fix build failure when specific decoder/encoder enabled