path: root/wiki/Chat_log/20180809-mm-chatlog

Multimedia-chat-meeting-2018-08-09

10:36 < pinchartl> welcome everybody to the multimedia meeting
10:37 < pinchartl> Topic 1. Status Check for the Multimedia Tasks
10:37 < pinchartl> * Jacopo
10:37 < pinchartl> Since last meeting:
10:37 < pinchartl> - DU input clock selection improvement
10:37 < pinchartl> [PATCH 0/3] drm: rcar-du: Rework clock configuration
10:37 < pinchartl> - LVDS reset handling
10:37 < pinchartl> [PATCH] drm: rcar-du: lvds: Handle LVDS interface reset
10:37 < pinchartl> - MT9V11 driver merged: a few follow ups:
10:37 < pinchartl> [PATCH] media: i2c: mt9v111: Fix v4l2-ctrl error handling
10:37 < pinchartl> [PATCH] media: i2c: mt9v111: Add VIDEO_V4L2_SUBDEV_API dependency
10:37 < pinchartl> - Tested and reviewed new v4l2 notifier helpers on CEU and VIN
10:37 < pinchartl> [PATCH v6 00/17] media: imx: Switch to subdev notifiers
10:37 < pinchartl> - Misc reviews:
10:37 < pinchartl> [PATCH v6 0/3] drm/atmel-hlcdc: bus-width override support
10:37 < pinchartl> [PATCH 00/14] Add minimal DTS support for M3-N Starter Kit
10:37 < pinchartl> - periupport patches:
10:37 < pinchartl> [PATCH v2] v4.14: LVDS: Mark LVDS panel patches as optional du clock sources 
10:37 < pinchartl> patches
10:37 < pinchartl> Until next meeting:
10:37 < pinchartl> - Follow DU clock rework
10:37 < pinchartl> - Ebisu VIN upstreaming
10:37 < pinchartl> - Start looking again to GMSL?
10:37 < pinchartl> - Sakari's v4l2-fwnode properties rework
10:37 < pinchartl> - Keep looking into display upstreaming (vga still to be clarified)
10:37 < pinchartl> Issues and Blockers:
10:37 < pinchartl> - Planning for September/October meetings conference still pending
10:37 < pinchartl> jmondi: any comment ?
10:38 < jmondi> not really
10:38 < jmondi> just catching up with backlog curently
10:38 < pinchartl> regarding conference planning, let's discuss that towards the end of the meeting
10:39 < pinchartl> * Kieran
10:39 < pinchartl> Since last meeting:
10:39 < pinchartl> - Reviewed/supported "[PATCH] v4l: vsp1: Fix deadlock in VSPDL DRM pipelines"
10:39 < pinchartl> - D3 DU/Interlaced investigations
10:39 < pinchartl> - Chasing GMSL. Need to rebase Sakari's VC work.
10:39 < pinchartl> - VSP1 1x1 pixel issue
10:39 < pinchartl> - DU/interlaced v6. Accepted by Maintainer at last \o/
10:39 < pinchartl> - GMSL for-renesas driver pull request (finally)
10:39 < pinchartl> - GMSL v2 public posting (sans VC)
10:39 < pinchartl> - Documented ADV748x secondary addresses
10:39 < pinchartl> Until next meeting:
10:39 < pinchartl> - 4K support
10:39 < pinchartl> - DU/D3 interlaced restrictions
10:39 < pinchartl> Issues and blockers: None
10:39 < pinchartl> kbingham: you also mentioned two weeks ago that you would look at reposting the VSP pending patches series (partition algo & writeback)
10:39 < pinchartl> is that still planned ?
10:39 < kbingham> The need to rebase Sakari's VC work was wrong/removed. I posted GMSL without VC
10:40 < kbingham> Hrm ... I rebased writeback ...
10:40 < kbingham> I guess I haven't included it in my notes as I haven't posted it...
10:40 < pinchartl> should I add both to the todo list ?
10:41 < kbingham> Yup.
10:41 < pinchartl> thanks
10:41 < pinchartl> * Laurent
10:41 < pinchartl> Since last meeting:
10:41 < pinchartl> - Patch review
10:41 < pinchartl> - Got VSP interlaced support merged (DU will follow in v4.20)
10:41 < pinchartl> - Additional tasks preparation
10:41 < pinchartl> Until next meeting:
10:41 < pinchartl> - Patch review
10:41 < pinchartl> - D3/E3 DU LVDS PLL upstreaming
10:41 < pinchartl> - Get GMSL patches merged
10:41 < pinchartl> Issues and blockers: None
10:42 < pinchartl> any comment from anyone?
10:42 < jmondi> that's a big one, and I don't want to resume the process/periject discussion
10:43 < jmondi> but how would you prefer to receive periupport updates? patches are ok? do you handle that by yourself looking at patches sent to mailing list?
10:43 < pinchartl> jmondi: patches are best
10:44 < pinchartl> as that's easiest to handle for me
10:44 < jmondi> I sent a few updates to periupport not "for inclusion"  but as a matter of keeping track of which BSP patches we were more or less working with
10:44 < pinchartl> otherwise I do it manually (but not very often)
10:44 < jmondi> I think there needs to be some work from your side too to apply those patches, I'm wondering how to minimize that
10:45 < jmondi> I feel like bulk updates do not work that well
10:45 < pinchartl> ideally we would update it in real time as soon as we get something merged upstream
10:46 < pinchartl> but in reality I don't think this will ever happen
10:46 < jmondi> I'm thinking for each BSP whose status changed, a single patch to periupport you can then squash, with a referce to the series/task that upports/obsolates them
10:46  * kbingham goes to check periupport du/interlaced lines  :)
10:46 < pinchartl> same problem as filling timesheets really, how often is that done at the end of the week when we can't remember what we've done and have to dig up, knowing perfectly well it would be easier to do it in realtime ?
10:46 < jmondi> because we rarely have a 1 to 1 match
10:48 < pinchartl> I don't really see a good solution to this problem if we don't enforce updates through process and tooling
10:48 < pinchartl> so at the moment, bulk patches is likely the best I can hope for
10:49 < pinchartl> unless someone else has a better advice ?
10:50 < kbingham> chmod g+w periupport ?
10:50 < pinchartl> kbingham: we'll still need review, based on my experience with patches that Jacopo posted before
10:50 < jmondi> kbingham: I'm not sure we all should push to periupport
10:51 < jmondi> kbingham: in the last week I would have screwed up myself only a few times
10:51 < kbingham> Ok :)
10:52 < pinchartl> * Morimoto-san
10:52 < pinchartl> Since last meeting:
10:52 < pinchartl> - RSND multi DAI support
10:52 < pinchartl> Until next meeting:
10:52 < pinchartl> - Continue multi DAI support and post patches
10:52 < pinchartl> Issues and Blockers: None
10:52 < pinchartl> morimoto: any comment?
10:54 < pinchartl> seems not :)
10:54 < pinchartl> * Niklas
10:54 < pinchartl> Since last meeting:
10:54 < pinchartl> - [PATCH] rcar-csi2: update stream start for V3M
10:54 < pinchartl> - [PATCH] v4l2-ctl: add support for SUBDEV_{G_STD,S_STD,QUERYSTD}
10:54 < pinchartl> - Reviewed V3H  VIN and CSI-2 patches.
10:54 < pinchartl> Until next meeting:
10:54 < pinchartl> - Finish last VC patch for adv748x use-case.
10:54 < pinchartl> - Post VC series.
10:54 < pinchartl> - Post held back VIN format patches.
10:54 < pinchartl> Issues and blockers: None
10:54 < pinchartl> neg: any comment ? I know you might be lurking :-)
10:55 < neg> no comment :-)
10:55 < pinchartl> :-)
10:55 < pinchartl> * Simon (Kaneko-san)
10:55 < pinchartl> Since last meeting:
10:55 < pinchartl> - D3 audio support upport
10:55 < pinchartl> Until next meeting:
10:55 < pinchartl> - Continue D3 audio support upport
10:55 < pinchartl> - Testing M3-N audio support
10:55 < pinchartl> Issues and blockers: None
10:55 < pinchartl> horms: any comment?
10:56 < horms> hi pinchartl
10:56 < morimoto> pinchartl: sorry for delay: no comment
10:56 < horms> The D3 audio backport looks good. We have access to a machine in Magnus's lab. And it has looback audio present. So that part can proceed.
10:56 < pinchartl> morimoto: no issue
10:56 < horms> s/D2/M3-N/
10:57 < horms> I think upporting for M3-N can also proceed as I unblocked Kaneko-san regarding a BSP patch he couldn't locate
10:57 < horms> But going might be a bit slow due to holiday activities
10:57 < horms> No other comments :)
10:57 < pinchartl> no worries about that :)
10:57 < pinchartl> thanks
10:57 < pinchartl> * Ulrich
10:57 < pinchartl> Since last meeting:
10:57 < pinchartl> - Prototyping LVDS PLL support on D3
10:57 < pinchartl> Until next meeting:
10:57 < pinchartl> - Finish prototyping LVDS PLL support on D3
10:57 < pinchartl> Issues and blockers: None
10:57 < pinchartl> uli_: any comment ?
10:57 < uli_> nope
10:58 < pinchartl> I got a question though, do you think you will be able to post a functional prototype by the 15th ?
10:58  * geertu starts looking for all "remove before leaving on holidays" tags...
10:58 < uli_> i think so
10:58 < pinchartl> great, thanks
10:59 < pinchartl> Topic 2. Additional Multimedia Tasks for Q3
10:59 < pinchartl> additional tasks are out now, you should all have received them
10:59 < pinchartl> please let me know if there's any issue
10:59 < pinchartl> Topic 3. Periperi Meetings
11:00 < pinchartl> it looks like all of us will go to Edinburgh
11:00 < pinchartl> and that the best day for a global meeting will be Friday the 26th
11:00 < pinchartl> is that correct ?
11:01 < morimoto> no topics yet ?
11:01 < pinchartl> morimoto: do you mean for the meeting agenda ?
11:01 < morimoto> yes
11:02 < pinchartl> I would personally like to discuss the development process and its relationship with periject. by the time of the conference I certainly hope we will bee able to reach a consensus
11:03 < pinchartl> and if we complete that topic earlier, then we'll have more time to discuss other topics :-)
11:03 < morimoto> OK, F2F is nice for it :)
11:03 < pinchartl> we can lock the room until we reach an agreement :-D
11:03 < wsa> morimoto: I am very glad to hear that you will come to ELCE! :D
11:04 < morimoto> hehe :)
11:04 < morimoto> I'm looking forward to see you guys again !
11:04 < neg> wsa: are you OK with moving SDHI hacking day so it won't collide with the V4L2 meeting?
11:04 < wsa> someone else from JapaPERI coming?
11:05 < kbingham> I guess we aught to do a haggis and whisky evening while we're there :)
11:05 < Marex> neg: can I join the SDHI hacking day ?
11:05 < Marex> wsa: ^
11:05 < morimoto> wsa: I and Ito-san maybe. No Magnus, No shimoda-san
11:06 < Marex> no morimoto no life (TM) ?
11:06 < wsa> morimoto: I see. Then you have to take Magnus place when boozing with Niklas :D
11:06 < Marex> morimoto: looking forward to seeing you again in the UK :)
11:06 < wsa> Marex: in general, sure yes, very welcome
11:06 < morimoto> Marex: you are correct :)
11:07 < morimoto> wsa: :)
11:07 < Marex> wsa: thanks ... I seem to be doing something with SDHI in U-Boot on my own and it seems you and neg are solving similar problems in Linux ... I think it makes sense to sync up
11:07 < wsa> not decided yet, if we do a full hacking day on saturday (but we won't have the flat anymore then), or spread it over the days (doesn't sound likely)...
11:07 < morimoto> Marex: me to
11:07 < wsa> Marex: makes perfect sense
11:08 < Marex> morimoto: :))
11:08 < wsa> Marex: although some of the sync should happen before ELCE by mail ;)
11:08 < Marex> wsa: fine by me
11:08 < wsa> or we do the SDHI hacking in Dunbar at the coast :D
11:09 < wsa> at weekends
11:09 < pinchartl> do we have an agreement for the 26th then ?
11:09 < wsa> yes
11:09 < wsa> SDHI hacking involves less people, so let's move this around
11:09 < neg> works for me
11:09 < wsa> since everyone else has time on Friday for the "big" meeting
11:09 < horms> fewer people
11:10 < wsa> oh!
11:10 < pinchartl> on the multimedia side, I still want to organize a code camp
11:10 < pinchartl> KR and XDC have been mentioned before
11:10 < pinchartl> but I think LPC could be a better option
11:11 < Marex> pinchartl: who's going to LPC ? did you submit anything ?
11:12 < wsa> Marex: did you plan to stay the weekend after ELCE?
11:12 < pinchartl> I haven't submitted anything
11:12 < pinchartl> but I will go
11:12 < kbingham> I plan to go to LPC
11:13 < pinchartl> neg: so do you I think?
11:13 < pinchartl> jmondi: will you be there?
11:13 < pinchartl> uli_: how about you?
11:14 < uli_> not at lpc
11:15 < pinchartl> the plan would be to spend at least 3 days together
11:15 < pinchartl> so likely the week before or after the conference
11:16 < neg> If there is MM meeting at LPC I will go if not I stay home
11:17 < pinchartl> ok
11:17 < pinchartl> that's it for me for today. any other topic that we should discuss ?
11:18  * morimoto good timing for me. I need to go back to home soon
11:20 < morimoto> sorry, but I will quit
11:20 < morimoto> bye-bye, enjoy your summer vacation
11:21 < pinchartl> as there's no other topic let's adjourne this meeting
11:21 < pinchartl> thank you all for attending