2023-12-31 01:32
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
00:37
dellas has quit [Remote host closed the connection]
00:53
dellas has joined #ffmpeg-devel
01:08
thilo has quit [Ping timeout: 260 seconds]
01:09
thilo has joined #ffmpeg-devel
01:09
thilo has quit [Changing host]
01:09
thilo has joined #ffmpeg-devel
01:52
<
Lynne >
what's the overhead of larger? a function call that writes 36 bytes to an array, or a memcpy that writes a non-constant number of bytes (that's usually 36 bytes)?
01:58
lexano has quit [Ping timeout: 264 seconds]
01:59
<
BtbN >
Unless it's a really huge memcpy, I don't expect there to be a notable difference
01:59
<
Lynne >
the destination memory may be slow (non-RAM, like the NIC buffer), and the function does writes a few bytes at a time
02:00
<
Lynne >
yeah, but it avoids passing a megabyte of just mostly unused array
02:01
<
BtbN >
If it's some magic buffer, like gpu uswc memory or what it's called... It might be faster to prepare a local buffer, and then copy it over in one go with a specialized routine
02:01
<
Lynne >
linux finally supports direct NIC ring buffer writes btw, useful for gigabits of uncompressed video, you don't even need root access
02:02
<
BtbN >
Realistically, it will need benchmarked
02:02
<
BtbN >
The kernel might be doing some magic as well, to only copy on syscall
02:04
dellas has quit [Remote host closed the connection]
02:32
navi has quit [Quit: WeeChat 4.0.4]
02:33
<
Lynne >
I think I found the only legitimate use of c1x's auto keyword, in for loops
03:01
lemourin has quit [Read error: Connection reset by peer]
03:02
lemourin has joined #ffmpeg-devel
03:05
jamrial has quit []
03:27
BradleyS has quit [Quit: quit]
03:39
BradleyS has joined #ffmpeg-devel
03:51
BradleyS has quit [Quit: quit]
04:01
cone-785 has joined #ffmpeg-devel
04:01
<
cone-785 >
ffmpeg Wenbin Chen master:c695de56b5ba: libavfilter/dnn_bakcend_openvino: Add automatic input/output detection
04:01
<
cone-785 >
ffmpeg Wenbin Chen master:d52e8ed11430: libavfilter/vf_dnn_detect: Use class confidence to filt boxes
04:01
<
cone-785 >
ffmpeg Wenbin Chen master:3de38b9da5c2: libavfilter/dnn_interface: use dims to represent shapes
04:01
BradleyS has joined #ffmpeg-devel
04:17
Martchus has joined #ffmpeg-devel
04:18
Martchus_ has quit [Ping timeout: 252 seconds]
04:36
BradleyS has quit [Quit: quit]
04:38
BradleyS has joined #ffmpeg-devel
04:52
BradleyS has quit [Quit: quit]
04:56
BradleyS has joined #ffmpeg-devel
05:20
epony has quit [Remote host closed the connection]
05:28
epony has joined #ffmpeg-devel
06:53
HarshK23 has joined #ffmpeg-devel
06:57
AbleBacon has quit [Read error: Connection reset by peer]
07:01
cone-785 has quit [Quit: transmission timeout]
08:19
jarthur has quit [Quit: jarthur]
08:42
mkver has joined #ffmpeg-devel
09:43
blb has quit [Ping timeout: 260 seconds]
09:43
blb has joined #ffmpeg-devel
09:49
kurosu has joined #ffmpeg-devel
10:07
akapustin has quit [Remote host closed the connection]
10:46
kepstin has quit [Remote host closed the connection]
10:47
kepstin has joined #ffmpeg-devel
11:06
ccawley2011 has joined #ffmpeg-devel
11:22
cone-140 has joined #ffmpeg-devel
11:22
<
cone-140 >
ffmpeg Andreas Rheinhardt master:8f791304f2e7: avcodec/dxvenc, hap(dec|enc): Move TextureDSPContext to stack
11:22
<
cone-140 >
ffmpeg Andreas Rheinhardt master:e1d1304b4b30: avcodec/texturedspenc: Remove unused rgtc1_u_alpha encoding func
11:22
<
cone-140 >
ffmpeg Andreas Rheinhardt master:916f01674178: avcodec/dxvenc: Fix data races with slice threading
11:22
<
cone-140 >
ffmpeg Andreas Rheinhardt master:555879ca7cf5: avcodec/dxvenc: Don't cast const away
11:22
<
cone-140 >
ffmpeg Andreas Rheinhardt master:a3fc9fb9fbc0: avcodec/texturedsp: Add separate TextureDSPEncContext
11:22
<
cone-140 >
ffmpeg Andreas Rheinhardt master:341d0419e129: avcodec/texturedsp: Factor common code out
11:22
<
cone-140 >
ffmpeg Andreas Rheinhardt master:4f58372c10df: avcodec/hap: Avoid unnecessary opt.h inclusion
11:41
dellas has joined #ffmpeg-devel
11:58
kurosu has quit [Quit: Connection closed for inactivity]
12:05
ccawley2011 has quit [Ping timeout: 246 seconds]
12:12
ccawley2011 has joined #ffmpeg-devel
12:32
navi has joined #ffmpeg-devel
12:34
<
elenril >
mkver: do you feel particularly strongly about -Ilibavcodec for bsfs?
12:39
<
mkver >
It is wrong that some the compilation flag of a file will be affected by whether something in another subfolder needs it.
12:40
<
elenril >
not sure what you mean by this
12:41
<
elenril >
I'm modifying CPPFLAGS for bsfs only
12:41
<
cone-140 >
ffmpeg Anton Khirnov master:e0da916b8f5b: fftools/ffmpeg: optimize inter-thread queue sizes
12:43
<
mkver >
No, you are modifying it for all files that are needed by BSFs. Including e.g. h264_levels.o.
12:43
<
elenril >
oh, I didn't realize
12:43
<
elenril >
stupid of me
12:43
<
elenril >
would you be ok with doing it just for bsf/*
12:55
Krowl has joined #ffmpeg-devel
13:56
Krowl has quit [Read error: Connection reset by peer]
14:16
___nick___ has joined #ffmpeg-devel
14:17
___nick___ has quit [Client Quit]
14:18
haasn` has joined #ffmpeg-devel
14:18
tufei has quit [Remote host closed the connection]
14:19
tufei has joined #ffmpeg-devel
14:19
___nick___ has joined #ffmpeg-devel
14:37
Krowl has joined #ffmpeg-devel
15:20
tmm1_ has joined #ffmpeg-devel
15:23
tmm1 has quit [Ping timeout: 264 seconds]
15:41
cone-140 has quit [Quit: transmission timeout]
15:44
lexano has joined #ffmpeg-devel
15:59
Jhlearn has joined #ffmpeg-devel
16:07
Jhlearn has left #ffmpeg-devel [#ffmpeg-devel]
16:16
Traneptora has joined #ffmpeg-devel
16:59
tufei_ has joined #ffmpeg-devel
17:00
tufei has quit [Remote host closed the connection]
17:01
rvalue has quit [Ping timeout: 264 seconds]
17:06
jnbek has quit [Quit: kthx]
17:10
jnbek has joined #ffmpeg-devel
17:11
rvalue has joined #ffmpeg-devel
17:30
Krowl has quit [Read error: Connection reset by peer]
17:53
Krowl has joined #ffmpeg-devel
17:53
Krowl has quit [Read error: Connection reset by peer]
18:01
epony has quit [Remote host closed the connection]
18:08
dellas has quit [Remote host closed the connection]
18:30
zsoltiv_ has joined #ffmpeg-devel
18:34
epony has joined #ffmpeg-devel
18:47
Krowl has joined #ffmpeg-devel
18:56
dellas has joined #ffmpeg-devel
19:07
sudden has quit [Ping timeout: 268 seconds]
19:11
AbleBacon has joined #ffmpeg-devel
19:14
sudden has joined #ffmpeg-devel
19:56
HarshK23 has quit [Quit: Connection closed for inactivity]
20:58
Krowl has quit [Read error: Connection reset by peer]
21:04
___nick___ has quit [Ping timeout: 260 seconds]
21:21
mkver has quit [Ping timeout: 252 seconds]
21:32
jarthur has joined #ffmpeg-devel
21:39
<
kierank >
another beautiful day on ffmpeg-devel
21:45
<
elenril >
...17 mails from the last 5 hours
21:45
<
elenril >
must be another cursed thread
22:09
aljazmc has quit [Quit: Leaving]
22:34
ccawley2011 has quit [Read error: Connection reset by peer]
23:45
navi has quit [Quit: WeeChat 4.0.4]
23:45
jamrial has joined #ffmpeg-devel
23:59
sudden has quit [Ping timeout: 260 seconds]