[04:55] *** sr105 has quit IRC (Ping timeout: 244 seconds)
[06:50] *** narmstrong_ has left 
[08:07] <kbingham> Somewhat off-topic, if it werent' that I'm tyring to filter the linux-tv patchwork - but does anyone know how I filter by delegate from pwclient? : pwclient list -s "New" -a no -d "laurent.pinchart" isn't returning anything except laurent's three e-mail addresses, and -d pinchartl says no delegate found !?
[08:14] <hverkuil> mchehab: sailus: re: https://patchwork.linuxtv.org/patch/51259/
[08:15] <hverkuil> Can this be merged asap? It is causing warnings in the daily build and it is a trivial patch.
[08:53] <larsc> But is this the right fix?
[08:55] <hverkuil> larsc: yes, it is.
[09:43] <sailus> hverkuil: Thanks for the ping.
[09:43] <sailus> I'll handle it.
[09:43] <hverkuil> sailus: much appreciated
[09:51] <sailus> jmondi: I presume you wanted to handle the V4L2 sub-device uAPI dependency with the patch changing #ifdef rather than the one adding it as a Kconfig dependency?
[12:02] <sailus> mchehab: Any news on the Media summit?
[12:03] <mchehab> still talking with LF with regards to the room
[12:03] <sailus> Ack.
[12:03] <mchehab> I'm trying to do it together with ELCE
[12:03] <mchehab> I'll let you know after we get something
[12:03] <sailus> But after the actual conference?
[12:04] <sailus> Thanks.
[12:04] <mchehab> I'm checking both ways
[12:04] <mchehab> during - in order to share coffee breaks - or after - without any coffee breaks
[12:05] <mchehab> I also asking for the costs of coffee breaks if we decide to do it after the conference
[12:05] <mchehab> (how we'll pay for coffee breaks is a separate matter)
[12:06] <mchehab> let me first gather all info... then we decide :-D
[12:07] <mchehab> being at a conference center increases a lot the costs
[12:09] <sailus> Indeed.
[12:51] *** harrow has quit IRC (Quit: Leaving)
[12:51] *** juvenal has quit IRC (Ping timeout: 244 seconds)
[12:51] *** [LOGGER] has quit IRC (Ping timeout: 244 seconds)
[12:52] *** aballier has quit IRC (Changing host)
[12:53] *** ttomov has quit IRC (Ping timeout: 248 seconds)
[13:35] *** ao2 has quit IRC (Ping timeout: 244 seconds)
[13:51] *** YuGiOhJCJ has quit IRC (Quit: YuGiOhJCJ)
[14:33] *** Robert_Zenz has quit IRC (Remote host closed the connection)
[14:55] *** dmt has quit IRC (Remote host closed the connection)
[15:19] *** benjiG has left 
[15:49] *** eduardas_m has quit IRC (Quit: Konversation terminated!)
[16:13] <mjourdan> ndufresne: hi (changed my nick from Ely to mjourdan :>), in gst v4l2, what's the behaviour if a video packet is too big to fill in an OUTPUT buffer ? Do you drop it, or split it in multiple buffers with the same timestamp ?
[16:25] <mjourdan> in FFmpeg, the packet is cropped to fill the buffer, and the rest of the data seems to be lost
[16:28] *** pH5 has quit IRC (Quit: bye)
[17:15] *** Eloy has quit IRC (Ping timeout: 240 seconds)
[18:45] <ndufresne> mjourdan, I think currently it fails the entire pipleine
[18:46] <ndufresne> mjourdan, https://bugzilla.gnome.org/show_bug.cgi?id=749016
[18:47] <ndufresne> it's difficult problem, as many hardware expect specific frameing
[18:48] <ndufresne> so for the v4l2sink case describe in that bug, each v4l2buffers are 188 bytes, bad driver design imho, but let's ignore that, it's fixed size packet
[18:49] <ndufresne> so blindly chopping the buffer into 188bytes will work
[18:49] <ndufresne> but for dynamic size packet, if you split it up in multiple v4l2buffer but at random places, I'm not sure it will work
[18:54] <mjourdan> Thanks
[18:55] <mjourdan> Yeah I'm pretty sure many drivers will fail with split packets in multiple buffers
[18:55] <mjourdan> Do you know what driver is being referenced in the bug reports ? Can't seem to find it
[18:59] <ndufresne> mjourdan, I never new, does not seem to be upstream, but it seems something like a qam or qpsk encoder to me
[18:59] <ndufresne> * knew
[19:22] <mjourdan> alright