path: root/wiki/Chat_log/20180109-mm-chatlog

Multimedia-chat-meeting-2018-01-09

10:04 < pinchartl> hello everybody
10:04 < pinchartl> it looks like we have a full house today
10:04 < kbingham> \o/
10:04 < pinchartl> at least if Magnus is here
10:04 < pinchartl> but he doesn't seem to be
10:04 < pinchartl> first topic, status report
10:05 < pinchartl> * Kieran
10:05 < pinchartl> Since last meeting:
10:05 < pinchartl> - More GMSL/V4L async work (before Xmas)
10:05 < pinchartl> - Christmas / NY Holidays
10:05 < pinchartl> - HDMI Interlaced support query
10:05 < pinchartl> Until next meeting:
10:05 < pinchartl> - Additional tasks still to be confirmed, but those.
10:05 < pinchartl> - Resubmit kmstest verification based on review comments
10:05 < pinchartl> Issues and Blockers: None
10:05 < pinchartl> kbingham: anything to add ?
10:05 < kbingham> That's it from me I believe currently
10:05 < pinchartl> thank you
10:05 < pinchartl> * Morimoto-san
10:05 < pinchartl> Since last meeting:
10:05 < pinchartl> - Pinged OmniVision / IMI, no reply yet
10:05 < pinchartl> - More ALSA SoC framework cleanup
10:05 < pinchartl> - M3N Salvator-XS board export paper work (?)
10:05 < pinchartl> Until next meeting:
10:05 < pinchartl> - Keep pushing Omnivision and IMI
10:05 < pinchartl> - More ALSA SoC framework cleanup
10:05 < pinchartl> Issues and Blockers: None
10:05 < pinchartl> morimoto: anything to add ?
10:06  * morimoto Renesas Japan is very slow now
10:07 < morimoto> I have +1 for C)
10:07 < morimoto> I can't find DU jitter patch on Linux-Next > pinchartl?
10:07 < morimoto> [1/2] drm: rcar-du: use 1000 to avoid misunderstanding in rcar_du_dpll_divider()
10:07 < morimoto> [2/2] drm: rcar-du: calculate DPLLCR to be more small jitter
10:07 < pinchartl> it's in my tree but I'm afraid I missed the merge window with the xmas holidays. I'll send a pull request for the next kernel version
10:08 < morimoto> OK, thans
10:08 < morimoto> s/thans/thanks
10:08 < pinchartl> you're welcome
10:08 < pinchartl> * Niklas
10:08 < pinchartl> Since last meeting:
10:08 < pinchartl> - [RFC 0/2] v4l2-ctl: add ROUTING get and set options
10:08 < pinchartl> - [PATCH 0/4] Update the max9286 to the new multiplexed pad implementation
10:08 < pinchartl> - [PATCH/RFC v2 00/15] Add multiplexed pad streaming support
10:08 < pinchartl> - [HACK/RFT] v4l: max9286: take route configuration into account when configuring
10:08 < pinchartl> - Working on review comments from Laurent on rcar-vin.
10:08 < pinchartl> - Multiplexed streams support without a pad+stream-aware s_stream()
10:08 < pinchartl> There could be a way to add multiplexed stream support without the need for a pad+stream aware s_stream() implementation. More tests are needed but if it turns out OK adding multiplexed stream support upstream should be a lot easier.
10:08 < pinchartl> Until next meeting:
10:08 < pinchartl> - Post next version of VIN and CSI-2
10:08 < pinchartl> - Keep working on the multiplexed pad series
10:08 < pinchartl> Issues and Blockers: None
10:08 < pinchartl> neg: anything to add ?
10:09 < pinchartl> should we try to schedule some time for Lager CEC tests ?
10:09 < neg> Yes that is the only thing I forgot to mention
10:09 < neg> Basicly whenever you have time and feel like it
10:09 < neg> Othre then that nothing to add
10:09 < pinchartl> this afternoon should work for me
10:10 < neg> sounds good
10:10 < pinchartl> * Laurent
10:10 < pinchartl> Since last meeting:
10:10 < pinchartl> - Submitted display color keying support for Gen3
10:10 < pinchartl> - More patch review (including the VIN driver)
10:10 < pinchartl> Until next meeting:
10:10 < pinchartl> - Prepare the FOSDEM code camp
10:10 < pinchartl> I will work with the team to make sure all dependencies are sorted out, that 
10:10 < pinchartl> necessary hardware will be available, and that support channels will be in 
10:10 < pinchartl> place.
10:10 < pinchartl> - Virtualization investigation
10:10 < pinchartl> - Experiment dim usage for V4L2 maintenance
10:10 < pinchartl> The dim tool is available at https://01.org/linuxgraphics/gfx-docs/maintainer-tools/dim.html.
10:10 < pinchartl> Issues and blockers: None
10:11 < pinchartl> * Ulrich
10:11 < pinchartl> Since last meeting:
10:11 < pinchartl> - Sent ZG clock support for M3-W
10:11 < pinchartl> - Sent patches to get out-of-tree Rogue GPU driver to work on M3-W mainline 
10:11 < pinchartl> - Working out with Jinso how to test the GPU support
10:11 < pinchartl> Until next meeting:
10:11 < pinchartl> Issues and Blockers: None
10:11 < pinchartl> uli___: you mentioned last time that you would mix and match the various versions of GPU user space libraries to get OpenGL working
10:11 < pinchartl> what's the status of that ?
10:12 < uli___> no success yet. the jinso tester is trying it with yocto, but that hasn't succeeded yet
10:12 < uli___> we're working on it still
10:13 < pinchartl> ok
10:13 < pinchartl> should I add it to the tasks you will work on during the next two weeks then ?
10:13 < uli___> please do so
10:13 < pinchartl> thanks
10:14 < pinchartl> and regarding the patches to get the out-of-tree GPU driver to work on M3-W mainline, I think you mentioned it as done during the last meeting already. is it a wrong copy&paste ?
10:14 < pinchartl> or has there been new work in that area ?
10:15 < uli___> i think i mixed up the dates because i sent it out after the meeting, but it was done before
10:15 < uli___> or something like that
10:16 < pinchartl> no worries
10:16 < pinchartl> that's it for the status reports, unless there is any question ?
10:16 < jmondi> pinchartl: where's mine status? :)
10:17 < jmondi> s/mine/my
10:17 < pinchartl> oops :-)
10:17 < pinchartl> * Jacopo
10:17 < pinchartl> Since last meeting:
10:17 < pinchartl> - Submitted CEU v2 and v3
10:17 < pinchartl> - Submitted OV7670 v1 and v2 (DT parsing and PLL calculation)
10:17 < pinchartl> - Enable single camera capture with GMSL on Salvator-X v4.15-rc4
10:17 < pinchartl> - GMSL patch reviews and discussions
10:17 < pinchartl> - Attempted to build Cogent GMSL Yocto layer (in progress)
10:17 < pinchartl> Until next meeting:
10:17 < pinchartl> - Submit CEU v4 (should be the last one)
10:17 < pinchartl> If the driver gets merged, start compile porting other SH boards that use old CEU driver.
10:17 < pinchartl> - Build and test the Cogent GMSL Yocto layer before the code camp
10:17 < pinchartl> - Keep pushing Maxim for support
10:17 < pinchartl> - Re-send v4l2-async debug patches
10:17 < pinchartl> There seems however to be little interest in the topic upstream given that use cases currently merged mainline are quite trivial.
10:17 < pinchartl> Issues and Blockers: None
10:17 < pinchartl> here :-)
10:17 < pinchartl> sorry
10:17 < pinchartl> any comment ?
10:18 < jmondi> if nothing breaks I should be able to compile Cogent's yocto images thanks to Vladimir provided link
10:19 < jmondi> I mean, it's running righ now.. apart from that, no
10:19 < pinchartl> let's see if something breaks then :)
10:19 < neg> \o/
10:19 < jmondi> of course it will
10:19 < pinchartl> Morimoto-san, is there any question from the BSP team this time ?
10:20 < morimoto> I'm sending small question email to periperi ML
10:20 < morimoto> So no more topic from BSP team
10:20 < pinchartl> ok, thank you
10:21 < pinchartl> next topic, additional tasks
10:22 < pinchartl> I haven't heard back from Magnus since December the 22nd
10:22 < pinchartl> Morimoto-san, do you know if he's still alive ?
10:23 < neg> Too much Chirstmas food, still in food coma?
10:24 < morimoto> pinchartl: I don't know...
10:25 < pinchartl> :-/
10:26 < pinchartl> I'm afraid all we can do is wait and see. the last thing I heard was that Magnus wanted a light Q1/1 and a heavier Q1/2. that means 5 days of additional multimedia tasks each for Niklas and Jacopo, and 10 days for Kieran and me
10:26 < pinchartl> for Q1/1
10:26 < pinchartl> with the leftover in Q1/2
10:27 < pinchartl> this means that more of the base contract should be consumed in Q1/1 (there will thus be very little left for Q1/2)
10:27 < pinchartl> and we of course need to agree on tasks for Q1/2 earlier than mid-February, in order to start working on them soon enough
10:27 < jmondi> pinchartl: you do have a list of possible tasks?
10:27 < morimoto> Hmm... I will contact to him tomorrow
10:28 < pinchartl> end of January would be my personal deadline for that, and I'd prefer earlier
10:28 < neg> pinchartl: I was offerd a IO additonal task for the first half of Q1 by wsa_, had you wished for me to also do a 5 day MM task during the first part of Q1 ?
10:28 < pinchartl> neg: Magnus would like you to have 5 days for multimedia, yes
10:28 < pinchartl> jmondi: the list was posted in the last meeting report
10:28 < pinchartl> and from that the following tasks were pre-selected:
10:29  * jmondi digs his emails
10:29 < pinchartl> - Multimedia PFC development (D3 + M3-N) 
10:29 < pinchartl> - BRS/BRU bug fix (on your task list) 
10:29 < pinchartl> - V3M Multimedia integration upstreaming 
10:29 < pinchartl> - D3 ADV chip slave address conflict workaround 
10:29 < pinchartl> - Multimedia Virtualization Investigation
10:30 < pinchartl> but I'm not entirely happy with that, I voiced my concerns to Magnus and I'm waiting for his feedback
10:32 < jmondi> ok, let's wait and see
10:32 < jmondi> there's plenty of GMSL related development, now that I see the list you shared after last meeting
10:32 < pinchartl> morimoto: thanks for offering to contact Magnus
10:33 < pinchartl> jmondi: Magnus wanted to focus on tasks that can produce upstream patches in Q1/1
10:33 < pinchartl> so there will likely be no GMSL additional task scheduled
10:33 < jmondi> makes sense..
10:33 < pinchartl> the GMSL code camp thus has to be covered by the base contract I'm afraid
10:34 < pinchartl> any other question related to additional tasks ?
10:34 < jmondi> not from here
10:35 < pinchartl> next topic then, the FOSDEM meeting
10:36 < pinchartl> first of all
10:36 < pinchartl> uli___: have you decided whether you will come to Brussels ?
10:36 < uli___> probably yes
10:37 < pinchartl> good news :)
10:37 < pinchartl> when would you arrive ?
10:37 < uli___> i basically made that decision the minute you asked, so i haven't worked out the details yet :)
10:38 < pinchartl> :-)
10:38 < pinchartl> well, let us know
10:38 < uli___> ok
10:38 < pinchartl> as everybody should be aware by now, there will be a GMSL code camp from Monday to Thursday
10:38 < pinchartl> we will have a multimedia meeting on Thursday afternoon
10:38 < pinchartl> and the core and I/O meetings will be on Friday
10:39 < pinchartl> the code camp and multimedia meeting will be held at Rue Antoine Dansaert 24, 
10:39 < pinchartl> 1000 Brussels
10:39 < wsa_> pinchartl: can you add that to the internal wiki?
10:39 < pinchartl> wsa_: sure
10:39 < wsa_> thanks!
10:39 < pinchartl> we have rented an apartment
10:40 < pinchartl> and Jacopo, Kieran, Niklas and me will stay there
10:40 < pinchartl> the code camp will focus on GMSL development, and we need to plan for that
10:40 < pinchartl> on Thursday evening I have booked a table for dinner at Bonsoir Clara, Rue Antoine Dansaert 22, 1000 Brussels
10:40 < pinchartl> as you can notice it won't take long to get there :-)
10:41 < kbingham> pinchartl: Is Niklas in the apartment too? I thought he was staying with WSA?
10:41 < pinchartl> currently Jacopo, Kieran (+ Keri), Niklas, Simon and me have confirmed attendence
10:41 < geertu> and Hugo?
10:42 < kbingham> Anywhere Keri goes - Hugo goes :D
10:42 < pinchartl> and Hugo, yes
10:42 < pinchartl> I have thus booked a table for 6 adults. if anyone else wants to join, please let me know ASAP
10:42 < pinchartl> neg: if I'm not mistaken you will stay with us for the first half of the week and then move in with Wolfram, right ?
10:43 < neg> pinchartl: yes
10:43 < pinchartl> neg: what are the dates ?
10:44 < neg> pinchartl: I will stay with wsa_ from Friday to Monday
10:45 < pinchartl> ok
10:46 < pinchartl> any question regarding the logistics ? we'll then get to the topic of GMSL code camp planning
10:47 < neg> And so I don't mess up the apartment is avliable from the afternoon-ish on Sun the 28th right?
10:47 < jmondi> I will probably arrive earlier than everybody else, so I might want to sync with Kieran to enter the apartment
10:47 < kbingham> I arrive in Brussels at 19:45 on the sunday.
10:48 < pinchartl> neg: correct
10:48 < kbingham> I assumed pinchartl would be the first to arrive, and have thus added his name to the booking so he can check in.
10:48 < pinchartl> I'll arrive on Sunday as well but I'm not sure when yet
10:48 < kbingham> (/side note - I've since seen whomever checks in gets the priviledge of paying)
10:48 < pinchartl> jmondi: at what time will you arrive ?
10:48 < jmondi> let me check again
10:48 < jmondi> I guess it's 7pm
10:48 < jmondi> so not that earlier as I first thought
10:49 < jmondi> 19:35
10:49 < jmondi> I guess I will meet Kieran&family at the airport
10:49 < pinchartl> I'll likely arrive earlier
10:49 < kbingham> jmondi: We can probably share a taxi or something in that case :)
10:49 < jmondi> kbingham: sure thing
10:50  * kbingham needs to move from the sofa to the office to plug laptop in ...
10:50 < pinchartl> could you please all fill your travel dates (and times if possible) in https://osdr.renesas.com/projects/linux-kernel-development/wiki/Periperi-2018-02 ?
10:51 < neg> I have yet to book my flight but will update wiki once I have done so
10:51 < pinchartl> kbingham: taxi is an option, otherwise you can take the train to Brussels North, and from there metro 3 or 4 to La Bourse - Beurs. the metro stop is a 5 minutes walk to the apartment (one block and a half)
10:53 < pinchartl> regarding the GMSL code camp
10:53 < kbingham> pinchartl: That might be better than us bringing a car seat actually.
10:53 < pinchartl> kbingham: it's pretty easy yes
10:53 < pinchartl> about 15-20 minutes by train from the airport to Brussels North
10:53 < pinchartl> you can even buy the train ticket while waiting for your luggage at the airport
10:54 < kbingham> :D
10:54 < pinchartl> we need to plan for hardware availability
10:54 < pinchartl> I believe the following would be useful
10:54 < pinchartl> H3 + expansion
10:54 < pinchartl> V3M
10:54 < pinchartl> H3 + kingfisher
10:54 < pinchartl> and of course cameras
10:54 < pinchartl> I'll bring the kingfisher-based setup
10:55 < pinchartl> Kieran and Niklas, you're the only ones to have cameras, please bring them
10:55 < jmondi> easy for me: I have nothing
10:55 < pinchartl> we'll focus on the RDACM20
10:55 < pinchartl> but it would be useful to have some RDACM21 just in case
10:55 < pinchartl> 8 should be enough, there's no need to have 16
10:55 < kbingham> I'll bring hte RDACM21's too ... they won't take up much room.
10:56 < pinchartl> the V3M is with Kieran if I'm not mistaken, Niklas you don't have one, right ?
10:56 < kbingham> I have V3M yes.
10:57 < kbingham> https://osdr.renesas.com/projects/linux-kernel-development/wiki/Hardware says neg has one too!
10:57 < pinchartl> I'm not sure we will need the V3M though
10:58 < neg> Yes I have a V3M
10:58 < pinchartl> Kieran, you've worked on GMSL for V3M, would you focus on H3 + expansion or on V3M ?
10:58 < pinchartl> maybe Niklas could bring the V3M and Kieran the H3 + expansion ?
10:59 < neg> And I will bring all 16 cameras in case there should be some trading, maybe it would be useful to return some to pinchartl as he now has a kingfisher?
10:59 < pinchartl> neg: but then it would be useful if you could test Kieran's GMSL V3M patches beforehand, to make sure your hardware is functional
10:59 < pinchartl> neg: good point
10:59 < neg> pinchartl: I have tested my V3M and I can capture from it
10:59 < pinchartl> neg: perfect :)
11:00 < pinchartl> so that would be one H3 + expansion (Kieran), one V3M (Niklas) and one Kingfisher (Laurent)
11:00 < pinchartl> I think that should be enough
11:01 < pinchartl> jmondi: would you be able to bring your scope ?
11:01 < jmondi> I could, yes
11:02 < pinchartl> I'll bring a multimeter, a second one would be useful if someone could bring one
11:02 < jmondi> I have no boards, after all :)
11:02 < pinchartl> jmondi: I'd advise packing the scope in carry-on luggage and the probes in checked luggage
11:02 < jmondi> pinchartl: why that?
11:02 < pinchartl> because the scope is fragile and the probes are pointy
11:03 < pinchartl> you might have trouble going through security with the probes
11:03 < jmondi> indeed
11:03 < jmondi> I hope I'll be able to carry the scope in my carry-on
11:03 < pinchartl> and I would be worried packing a scope in checked luggage, as there are risks of losing it (or getting it stolen)
11:04 < jmondi> yes, I was thinking the same
11:04 < jmondi> I should be able to bring it, no worries
11:05 < geertu> FWIW, I had no issues with my scope probes when going to San Sebastian
11:05 < jmondi> geertu: better safe than sorry
11:06 < pinchartl> I'll bring a multimeter, a second one would be useful if someone could bring one
11:06 < jmondi> or, better safe than inspected
11:06 < pinchartl> oops, sorry, already mentioned that :)
11:06 < pinchartl> who has a multimeter he can bring ?
11:06 < geertu> I did suffer from my racial profiling in Biarritz, I think.
11:06 < neg> kbingham: I can bring my logic probe, is it possible to prep the break-out camera with i2c probe points? Or do the camera still break if probes are attached?
11:06 < jmondi> pinchartl:  I do have a very basic one
11:06 < kbingham> neg: I'll bring hte breakout yes. The points are still attached and glued.
11:06 < neg> kbingham: greate!
11:07 < jmondi> geertu: being belgian increases the chances of being randomly selected for inspections in France?
11:07 < pinchartl> neg: good idea, thanks
11:07 < geertu> When I had issues understanding their "French", they asked where I was from, and immediately stopped bothering me.
11:07 < pinchartl> jmondi: please bring the multimeter if you can, thanks
11:08 < pinchartl> please all bring cables (network, USB and display) and power supplies for all the boards you will bring
11:08 < geertu> So I guess their French was the local Spanish from just across the border
11:08 < pinchartl> we will need power strips. in the worst case we can buy them in Belgium, but if you have space in your luggage please think about bringing a power strip
11:09 < kbingham> And a network switch
11:09 < kbingham> I can bring one - but it will be powered by a UK plug - an EU plug version would be useful.
11:09 < pinchartl> I only have a small 4-ports switch. can someone bring a larger one ? we'll need 8 ports
11:10 < pinchartl> kbingham: agreed
11:11 < pinchartl> I'll try to bring the power strip I bought in San Sebastian :)
11:11 < jmondi> I have a shiny new 8 ports switch
11:11 < neg> I also only have small 4port switch, but can bring if there is a need
11:11 < pinchartl> jmondi: could you bring it ?
11:11 < jmondi> still to be unpacked
11:12 < jmondi> I will
11:12 < pinchartl> thank you
11:12 < pinchartl> anything else we need to bring ?
11:12 < jmondi> I would like to write this list down... wiki page?
11:13 < pinchartl> I'll include it in the meeting report but we can add it to the wiki, yes, that's a good idea
11:13 < jmondi> doing it right now
11:13 < pinchartl> thank you
11:14 < pinchartl> that's it for me for the GMSL code camp topic
11:15 < pinchartl> the last topic for today is the next meeting, which will be held on 2018-01-25 (back on a Thursday) at the usual time
11:15 < pinchartl> is there anything else than anyone wants to discuss ?
11:16 < pinchartl> regarding the GMSL code camp I still want to synchronize with all of you and make sure we won't be blocked by any missing dependency but we can do that later
11:16 < neg> Anything else then GMSL preparations that should be prepared for the face-to-face meeting?
11:16 < pinchartl> this afternoon if everybody is available
11:17 < pinchartl> neg: good point. I don't think there's any need to prepare anything (well, I'll prepare a planning proposal), but if there's any topic you'd like to discuss, please submit it by e-mail
11:17 < wsa_> I'd like to discuss the time slots on 2018-01-25
11:17 < wsa_> but we can do that by mail
11:17 < jmondi> pinchartl: sorry, I didn't get it.. would you like to meet thsi afternoon?
11:18 < wsa_> if possible, I'd like if IO is not first; the later the better
11:18 < pinchartl> jmondi: yes, to discuss GMSL, if you're available
11:18 < pinchartl> wsa_: multimedia has a tendency to spread over the 30 minutes boundary so it's convenient if it's last, but I'm sure we can find a schedule that will suit everybody
11:19 < neg> this afternoon works for me, I will work as always until ~17:00 CET then a break until ~22:00 CET. But is avilable inbetween if scheduled time as I try to do something else then sit at my desk duing the break :-)
11:19 < pinchartl> any time from 14:00 to 17:00 CET is fine with me
11:19 < jmondi> any time before 7pm is fine with me
11:20 < morimoto> So, JaPERIese side will quick. Thanks, bye-bye
11:21 < pinchartl> morimoto: thank you for attending. have a nice evening
11:21 < morimoto> thanks. happy new year, you guys
11:21 < pinchartl> kbingham: any preference time-wise ?
11:21 < kbingham> pinchartl: I'll leave the office at 5.30 GMT today...
11:22 < pinchartl> should we go for 14:00 CET ? or a bit later to allow Kieran to have a proper lunch break ? 14:30 CET ?
11:22 < kbingham> 14.00 CET will be fine here.
11:22  * kbingham usually eats lunch 'al desko'
11:23 < pinchartl> jmondi: is 14:00 CET OK for you too ? I know you usually have lunch late
11:23 < jmondi> no, it's fine
11:23 < pinchartl> ok
11:23 < jmondi> I can have lunch while talking with you :)
11:24 < pinchartl> it's settled then
11:24 < pinchartl> any  other topic for today?
11:26 < neg> pinchartl: shall we try for the lager test after the GMSL meeting?
11:26 < pinchartl> neg: yes, or even at the same time :-)
11:26 < neg> pinchartl: sure :-)
11:26 < pinchartl> no other topic, I propose adjourning this meeting. does anyone second ?
11:27 < jmondi> yup! I'll see you later then
11:27 < pinchartl> meeting adjourned. thank you all for attending, and talk to you at 14:00 CET for GMSL