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
zane has joined #ffmpeg-devel
jarthur_ is now known as jarthur
jarthur has quit [Quit: jarthur]
x0r has joined #ffmpeg-devel
x0r has quit [Ping timeout: 250 seconds]
<fjlogger> [FFmpeg/FFmpeg] New comment on pull request #20279 forgejo/workflows: also run 32bit fate tests (https://code.ffmpeg.org/FFmpeg/FFmpeg/pulls/20279#issuecomment-3857) by B⁠tbN
iive has quit [Ping timeout: 260 seconds]
zane has quit [Ping timeout: 256 seconds]
zane has joined #ffmpeg-devel
fjlogger has quit [Server closed connection]
fjlogger has joined #ffmpeg-devel
jamrial has quit []
<fjlogger> [FFmpeg/FFmpeg] New comment on pull request #20279 forgejo/workflows: also run 32bit fate tests (https://code.ffmpeg.org/FFmpeg/FFmpeg/pulls/20279#issuecomment-3858) by k⁠asper93
DVedaa3 has quit [Ping timeout: 255 seconds]
zane has quit [Ping timeout: 252 seconds]
zane has joined #ffmpeg-devel
Arsen has quit [Server closed connection]
Arsen has joined #ffmpeg-devel
Kwiboo has quit [Server closed connection]
Kwiboo has joined #ffmpeg-devel
derpydoo has joined #ffmpeg-devel
DVedaa3 has joined #ffmpeg-devel
<fjlogger> [FFmpeg/FFmpeg] New comment on pull request #20253 avcodec/vima: support IMA4 (ADPCM QT) variant (https://code.ffmpeg.org/FFmpeg/FFmpeg/pulls/20253#issuecomment-3860) by p⁠ross
psykose has quit [Server closed connection]
psykose has joined #ffmpeg-devel
mateo`_ has quit [Read error: Connection reset by peer]
mateo` has joined #ffmpeg-devel
iive has joined #ffmpeg-devel
<fjlogger> [FFmpeg/FFmpeg] New comment on pull request #20279 forgejo/workflows: also run 32bit fate tests (https://code.ffmpeg.org/FFmpeg/FFmpeg/pulls/20279#issuecomment-3861) by m⁠ichaelni
microchip_ is now known as fartioso
mkver has joined #ffmpeg-devel
grillo_0 has quit [Quit: Ping timeout (120 seconds)]
grillo_0 has joined #ffmpeg-devel
derpydoo has quit [Quit: derpydoo]
<fjlogger> [FFmpeg/FFmpeg] Pull request #20290 opened: avcodec/hevc: Fix invalid null pointer check (https://code.ffmpeg.org/FFmpeg/FFmpeg/pulls/20290) by q⁠uink
<fjlogger> [FFmpeg/FFmpeg] New comment on pull request #20253 avcodec/vima: support IMA4 (ADPCM QT) variant (https://code.ffmpeg.org/FFmpeg/FFmpeg/pulls/20253#issuecomment-3867) by m⁠lauss2
<fjlogger> [FFmpeg/FFmpeg] New comment on pull request #20279 forgejo/workflows: also run 32bit fate tests (https://code.ffmpeg.org/FFmpeg/FFmpeg/pulls/20279#issuecomment-3868) by m⁠storsjo
<fjlogger> [FFmpeg/FFmpeg] Pull request #20253 merged: avcodec/vima: support IMA4 (ADPCM QT) variant (https://code.ffmpeg.org/FFmpeg/FFmpeg/pulls/20253) by m⁠lauss2
GewoonLeon has joined #ffmpeg-devel
Compnn has joined #ffmpeg-devel
Compn has quit [Ping timeout: 255 seconds]
CoreX- has joined #ffmpeg-devel
CoreX has quit [Quit: Computer Say's Goodbye]
<haasn> Lynne: in bwdif_vulkan you set temporal_diff[1] twice
<haasn> temp_diff I mean
CoreX- is now known as CoreX
jamrial has joined #ffmpeg-devel
<fjlogger> [FFmpeg/FFmpeg] Pull request #20290 merged: avcodec/hevc: Fix invalid null pointer check (https://code.ffmpeg.org/FFmpeg/FFmpeg/pulls/20290) by j⁠amrial
elvis_a_presley has joined #ffmpeg-devel
BradleyS has quit [Remote host closed the connection]
BradleyS has joined #ffmpeg-devel
mkver has quit [Ping timeout: 256 seconds]
mkver has joined #ffmpeg-devel
<fjlogger> [FFmpeg/FFmpeg] New comment on pull request #20279 forgejo/workflows: also run 32bit fate tests (https://code.ffmpeg.org/FFmpeg/FFmpeg/pulls/20279#issuecomment-3875) by B⁠tbN
<BtbN> wtf happened on nut-devel over night oO
<BtbN> "5024 NUT-devel moderator request(s) waiting"
<jamrial> after moving to mailman 3?
<BtbN> it's the one list that has been on there for a while already
<BtbN> And now it has over 5000 pending subscription requests over night
<BtbN> all from questionable addresses
<jamrial> did it start being crawled?
<BtbN> These came in via mail
<BtbN> Also, I've attached public-inbox to mailman3, which has much more workable archives imo: https://lists.mplayerhq.hu/lore
<BtbN> (Just don't look at the HTML it generates)
<BtbN> hm, most of those E-Mails and requests are ancient. And don't show up on the webif
<BtbN> very weird, they must have been migrated over, and only partially marked as dealt with
mkver has quit [Remote host closed the connection]
mkver has joined #ffmpeg-devel
DauntlessOne4985 has quit [Read error: Connection reset by peer]
DauntlessOne4985 has joined #ffmpeg-devel
<Lynne> haasn: thanks
<JEEB> michaelni, BBB - wasn't it our previous position that we would not upstream wrappers for hardware that is not generally available?
<JEEB> IIRC we nope'd at certain patch sets for certain cloud providers' wrappers due to that
<BBB> that was my response to the earlier mailinglist discussion, but I'm not sure everyone agrees with me
<BBB> I think with free access they're trying to fix the "not generally available" portion, as that is not very well defined (what it means etc.)
<JEEB> yea
<JEEB> although if it's free access that you specifically need to request and wait for etc etc...
<fjlogger> [FFmpeg/FFmpeg] New comment on issue #20145 filter_complex using two input files fails with error "Could not open encoder before EOF" (https://code.ffmpeg.org/FFmpeg/FFmpeg/issues/20145#issuecomment-3878) by a⁠ndrew.morgan
<fjlogger> [FFmpeg/FFmpeg] Issue #20291 opened: several cover-{art,image} tests fail on big-endian configurations (https://code.ffmpeg.org/FFmpeg/FFmpeg/issues/20291) by s⁠ean_mcg
<michaelni> BtbN, would be nice if "https://lists.mplayerhq.hu/lore/nut-devel/" would display names of authors of messages not just subjects
<BtbN> I don't think I have any influence on that
<BtbN> It's straight up the software that's running on https://lore.kernel.org/
<BtbN> I think the idea is that it's showing threads there, and it's not 100% clear which name it'd show. For one. Original author? Latest message?
<BtbN> In any case, I don't think it's a feature
<michaelni> ok :(
<BtbN> I agree it'd be nice.
<BtbN> But it's all written in perl
<BtbN> so you can kinda pick your favourite, they both work in parallel
<michaelni> i guess someone just needs to patch the pearl with 1 line to show author names
<BtbN> I've also just plain installed it from debian packages
<BtbN> so patching it would be somewhat involved
<michaelni> you dont like re building debian packages ? ;)
<BtbN> that, and doing it in a consistent and sustainable way is hard
<michaelni> maintaining a debian package repo with secuirty fixes for a 1 line fix is hard ;)
<BtbN> It's also complete overkill
<JEEB> yea, pbuilder is not too hard to figure out at the end of the day, but yes - a full repo...
<BtbN> I'd rather just not use debian packages, and manually install it from git
<BtbN> then patching is easy
<michaelni> yes
<BtbN> But still someone needs to first be able to read all that perl
<JEEB> at least now debian seems to have git-based tooling as well
<michaelni> nicolas knows perl IIRC, maybe if we ask him nicely he can change it or maybe upstream would be open to such feature
<BtbN> I would not bother with debian packages at all then
<JEEB> but yea, that set of tools never was as clear for me as the stuff used for fedora. that was just "you use fedpkg"
<BtbN> https://repo.or.cz/public-inbox.git/tree the code of it is here
<BtbN> https://public-inbox.org/public-inbox-config.txt if it's already an option, it should be described here
Anthony_ZO has quit [Ping timeout: 245 seconds]
<jamrial> Traneptora: what's the reason you overwrite the Orientation tag in exif after converting it to displaymatrix side data?
ePirat has joined #ffmpeg-devel
<kasper93> Lynne: get_value(&gb, ac_codebook) can return negative value which causes oob because it indexes array by this value
<kasper93> also `31 - ff_clz(nb_blocks)` has to be replaced by ff_log2(), because clz(0) is invalid and ff_log2 takes care of that
<fjlogger> [FFmpeg/FFmpeg] Pull request #20072 merged: avutil/hwcontext_vulkan: add disable_host_mapping debug option (https://code.ffmpeg.org/FFmpeg/FFmpeg/pulls/20072) by h⁠aasn
zane has quit [Quit: WeeChat 4.7.0]
minimal has joined #ffmpeg-devel
<fjlogger> [FFmpeg/FFmpeg] New comment on pull request #20273 avcodec/d3d12va_encode: texture array support for HEVC (https://code.ffmpeg.org/FFmpeg/FFmpeg/pulls/20273#issuecomment-3924) by t⁠ong1wu
welder has quit [Server closed connection]
welder has joined #ffmpeg-devel
Martchus has joined #ffmpeg-devel
tmatth has quit [Server closed connection]
tmatth has joined #ffmpeg-devel
<philipl> BtbN: do you have an opinion on the nvidia changes to add more cuda pix fmts from early july?
<BtbN> I don't see the point in adding random new pix fmts if nothing uses them yet
<BtbN> If they come with a patch that actually makes use of them, it's a non-issue to just add them to the list
<philipl> BtbN: most (maybe not all?) of them are used in the scale_cuda patch
<BtbN> Yeah, then it's fine.
<philipl> K. I'll try and look over the scale patch this week
<fjlogger> [FFmpeg/FFmpeg] Pull request #20292 opened: swscale6_clean (https://code.ffmpeg.org/FFmpeg/FFmpeg/pulls/20292) by h⁠aasn
<fjlogger> [FFmpeg/FFmpeg] New comment on pull request #20292 swscale: new ops framework / backend (https://code.ffmpeg.org/FFmpeg/FFmpeg/pulls/20292#issuecomment-3938) by h⁠aasn
<kasper93> happening
scat117 has quit [Server closed connection]
scat117 has joined #ffmpeg-devel
rvalue has quit [Read error: Connection reset by peer]
rvalue has joined #ffmpeg-devel
DVedaa3 has quit [Read error: Connection reset by peer]
DVedaa3 has joined #ffmpeg-devel
rvalue has quit [Max SendQ exceeded]
rvalue has joined #ffmpeg-devel
<fjlogger> [FFmpeg/FFmpeg] Pull request #20293 opened: .forgejo/CODEOWNERS: add myself for vp8, vp9, wmavoice and csp. (https://code.ffmpeg.org/FFmpeg/FFmpeg/pulls/20293) by r⁠bultje
<BBB> hurray it works
realies9346 has quit [Quit: ~]
<BBB> I don't actually understand the syntax at all, everyone appears to use .* instead of *, does that mean it's regexp style, not shell-expansion style?
<GewoonLeon> BBB: yeah, it's regex (https://code.ffmpeg.org/FFmpeg/FFmpeg/pulls/20180)
<BBB> jamrial: so then vp8* would be wrong? That would mean vp88888888(etc)
<jamrial> ah, i see
RT|AO has quit [Server closed connection]
RT|AO has joined #ffmpeg-devel
sr55 has joined #ffmpeg-devel
<BtbN> It is regexps, not globs, yes
s55 has quit [Ping timeout: 260 seconds]
sr55 has quit [Client Quit]
s55 has joined #ffmpeg-devel
<fjlogger> [FFmpeg/FFmpeg] Pull request #20289 merged: forgejo/CODEOWNERS: add myself (https://code.ffmpeg.org/FFmpeg/FFmpeg/pulls/20289) by m⁠ichaelni
rvalue- has joined #ffmpeg-devel
rvalue has quit [Ping timeout: 252 seconds]
rvalue- is now known as rvalue
<Traneptora> <jamrial> Traneptora: what's the reason you overwrite the Orientation tag in exif after converting it to displaymatrix side data?
<Traneptora> it was recommended to me on the ML after I submitted the first patch, lemme find the relevant mail link
<jamrial> IMO, it should be either left as is, or removed. Otherwise there will be conflicting values for the same property
<Traneptora> you think it should be removed entirely rather than zeroed? that makes sense
<jamrial> yes
<jamrial> Orientation == 1 means no rotation, right? that's still a conflicting value to whatever is in display_matrix
<jamrial> no Orientation tag at all is undefined, so display_matrix is the only thing defining rotation
<Traneptora> yea, I can patch that
<jamrial> i already did, will make a PR later
<Traneptora> aight
<Traneptora> I'll also send a patch for a few of those CIDs probably either today or tomorrow
<Traneptora> been busy with $dayjob
<Traneptora> a few are false positives, but at least one is legit
Guest76 has joined #ffmpeg-devel
Guest76 has quit [Client Quit]
<fflogger> [editedticket] mlauss: Ticket #5406 ([undetermined] adpcm_vima: problem with stereo files) updated https://trac.ffmpeg.org/ticket/5406#comment:8
mateo` has quit [Quit: WeeChat 4.7.0]
mateo` has joined #ffmpeg-devel
<BBB> so if it's approved, I can merge it, right?
<BBB> or am I missing some custom?
<jamrial> yes, with one approval you can click the merge/rebase button
<BBB> it says "rebase then fast-forward", that one?
<jamrial> yeah
<BBB> and that merges?
<jamrial> you need to click it twice for some reason, to confirm your choice
<fjlogger> [FFmpeg/FFmpeg] Pull request #20293 merged: .forgejo/CODEOWNERS: add myself for vp8, vp9, wmavoice and csp. (https://code.ffmpeg.org/FFmpeg/FFmpeg/pulls/20293) by r⁠bultje
<BBB> hey it worked
<BBB> cool
<jamrial> was the branch up to date, or did it rebase?
<jamrial> I'd expect a rebase to first run through CI before merging
<BBB> it said rebase but it did not re-run ci, I think
<beastd> BBB: Sorry for being late. Why did you remove wbs there? libavfilter/aarch64/.* @mstorsjo
<BBB> I moved him
<BBB> libavcodec has simd at the end, avfilter had it inline with the c code
<BBB> so I moved simd to the end for libavfilter also, including him
<beastd> ah ok, thought we did alphabetic. but that also makes sense
<BBB> right, it was inconsistent and I agree either way would've been fine
realies9346 has joined #ffmpeg-devel
<beastd> np. sorry for not noticing the reordering. fine for me
<fjlogger> [FFmpeg/FFmpeg] Issue #20294 opened: DWAA compressed EXR images cannot be read (decode_block() error / black video output) (https://code.ffmpeg.org/FFmpeg/FFmpeg/issues/20294) by l⁠eoflint
mkver has quit [Ping timeout: 244 seconds]
mkver has joined #ffmpeg-devel
<fjlogger> [FFmpeg/FFmpeg] Pull request #20295 opened: avutil/hwcontext_vulkan: always enable extra usage flags (https://code.ffmpeg.org/FFmpeg/FFmpeg/pulls/20295) by h⁠aasn
<fjlogger> [FFmpeg/FFmpeg] New comment on pull request #20295 avutil/hwcontext_vulkan: always enable extra usage flags (https://code.ffmpeg.org/FFmpeg/FFmpeg/pulls/20295#issuecomment-3957) by h⁠aasn
HarshK23 has quit [Quit: Connection closed for inactivity]
fartioso has quit [Quit: Do coders dream of sheep() ?]
microchip_ has joined #ffmpeg-devel
Kimapr has quit [Remote host closed the connection]
Kimapr has joined #ffmpeg-devel
Kei_N_ has quit [Server closed connection]
Kei_N has joined #ffmpeg-devel
dacid44 has joined #ffmpeg-devel
dacid44 has quit [Client Quit]
dacid44 has joined #ffmpeg-devel
Anthony_ZO has joined #ffmpeg-devel
desmond-netint has joined #ffmpeg-devel
witchymary has quit [Remote host closed the connection]
dacid44 has quit [Quit: Client closed]