Multimedia-chat-meeting-2017-03-28 Hi sorry for my late hi Morimoto-san [16:02] no worries (I will post my list soon by email) it's my fault, I should have thought about DST when we scheduled this meeting [16:03] so let's get started and hope that Jacopo will show up topics for today are [16:04] Topic 1. Status check for the multimedia tasks Topic 2. Multimedia plan (related to VIN development) Topic 3. Meeting around LinuxCon Japan [16:05] Topic 4. Next meeting anything else ? pinchartl: yes I am... got confused with time zone switch [16:06] no worries Topic 1. Status check for the multimedia tasks [16:07] let's go in reverse alphabetical order to give Jacopo a few minutes to wake up [16:08] uli___: you're first * kbingham back so i have a max9286/9271/ov10635 driver combo that has a chance of actually doing something not yet checked if it actually does and the mt8137 chromebook embedded controller is connected to the soc via spi [16:09] not trivial to set up without a kernel that's a); nothing to report for b) c) d) [16:10] you don't plan to do any work for the next two weeks ? :-) well, tbd no concrete plans yet i might test the driver works, i guess :) [16:11] we sort of depend on the outcome of the max/ov combo prototype... a bit more than *sort of* actually =) it's very high priority so could you please elaborate on what has been done and what is left to be done ? [16:12] and why the code couldn't be tested yet ? like i wrote in the e-mail, it probes, media-ctl says everything is fine it's not tested yet because i finished it yesterday uli___: i think the power controller bits should be sorted since some time ago [16:13] can you access devices over I2C? the power controller bits ? apparently a second power adapter was needed for the board [16:14] it was not provided with the board from the beginning the devices are there * morimoto Renesas inside talk present devices is an improvement i think dammsan: do you mean the MAX9286 board ? i don't even know the name of the board the hardware i received from morimoto-san [16:15] the daughter board for salvator-x with tons of camera connections yes so now you have hooked up a power supply to that board ? since some time ago maybe about 2 weeks ago the hardware should be fine now at least [16:16] (what i can tell) so if the I2C devices are present then there is no need for the r8a7792 blanche backup plan? [16:17] I see power connectors in the schematics indeed dammsan: not for my point of view. but what exactly is the r8a7792 backup plan? i hooked up an unknown camera to the r8a7792 blanche board some time ago [16:18] * morimoto back i kind of wanted to see if some devices showed up on I2C there too uli___: you mentioned in your e-mail that you ported two drivers, one from Cogent, one from the BSP [16:19] yes and that our hardware is quite different from what the first driver support how so ? our hardware is connected like so: soc <=i2c= max9286 <=i2c= max9271 <=i2c= ov10635 [16:20] they support various boards, but they all have more than one chip connected directly to the soc some have an i2c multiplexer [16:21] but none are in a straight line like ours they don't support that hardware setup ? what kind of hardware setup do they support ? i think somewhere in their pile of patches there is the code from the renesas bsp as well [16:22] they support boards called "kingfisher" and "view", and an unnamed board i think so they have no direct support for our hardware platform, and while it might not be too difficult to fix that, you decided to use the BSP code instead ? [16:23] i was a bit pressed for time, and the bsp code is written for this platform... [16:24] where do they have the i2c multiplexer in ther setup, between the SoC and MAX9286 chip(s)? I'm not complaining :) I'm just trying to make sure I understand the situation correctly neg: i don't remember, i would have to look up the DT [16:25] Me too, I think it's awesome you got that code working with the VIN dirver. When I looked at that code I felt sick :-) uli___: OK np I was just curious how it differd it's a pca9somethingsomething, and it's connected to the soc i guess they packed everything in one driver because a proper implementation would need two additional i2c host drivers... [16:26] yes uli___: so, when do you think you'll be able to test the code ? and, more importantly, get it to work :-) [16:27] test, today or tomorrow i guess. get it to work, hmmm... :) I would really appreciate if you could provide the first test results today [16:28] I'll sleep better (or worse...) tonight i'll see what it does later [16:29] thank you you can send a quick e-mail to update us ok if everything works I'll be very happy if that's not the case, please try to describe what the problem(s) are with as much information as possible [16:30] I'm sure Niklas can assist you with any VIN issue sure can neg: you're next :-) [16:32] a) Posted two VIN series '[PATCH 00/16] rcar-vin: fix issues with format and capturing' and '[PATCH v3 00/27] rcar-vin: Add Gen3 with media controller support' b) Act on reviews on the posted series and try to take a stab at extending the of graph framework to be able to iterate over all DT nodes in a graph connected by endpoints. c) None d) None if someone got review time please look at the first series as it should be simpler to review and fixes real issues and the second one depends on it. Also it would feel good to finaly get some VIN work merged :-) [16:33] ooh also in b) I will fixupo a few things for the CSI-2 driver and repost it [16:34] --EOT-- thank you do you feel you're on track schedule-wise ? Yes I hope so, there are still some work to be done in the are to be able to find, and then use the async framework to create all media controller links for arbitrary long device chains which needs to be sorted [16:36] ok [16:37] problem is as we talked shortly about your holiday is how to best relay information about pads from DT to MC but I imagine some work is needed there in the task to fix ADV7482 multiple subdevices [16:38] yes that's one of the nasty missing pieces have you discussed that with anyone while I was away ? so I currently focus on exteding of graph to provied helpers to find all DT nodes since this will be needed however the port->pad thing is solved and it would also be used later by the device allocator API [16:39] no I have not presude it further as the discussion with Russel King about MC in general I feelt safer to let it cool of a bit before trying to attack that issue I'll add the port->pad mapping to your list of issues [16:40] thanks morimoto: you're next [16:42] I know you must have a huge mail backlog but maybe sometime next week you and I (and anyone else interessed ofc) can have a short chat about it, I feel a bit lost on how to best think about the problem OK neg: sure. early next week would be best for me There is rumour I will be looking at some ADV7482 work based on the chat log from the last meeting - so let me know when you have the chat :) thanks, I can ping you and kbingham monday then and see when would be convinient, thanks [16:43] * morimoto let me know if I can start ;P morimoto: please go ahead :-) Hehe :O s/:O/:)/ A) o I posted OF-graph patchset again, no response [16:44] o Hacking for Sound driver bug fix o Hacking for bind/unbind issue for ALSA SoC o I could contact to OV guys, and discuss about OV camera datasheet. I sent you that you can receive datasheet from OmniVision, but you can call me liar. Final conclusion is that I need to export it to you. and it will takes not short term. o I asked you guys about camera board power cable, on Subject: Re: CSI2 multiple subdevices development Date: Wed, 22 Mar 2017 09:18:44 +0900 B) [16:45] o DT maintainer response is too slow o I'm very x10 sad (;_;), because no response from MultiMedia group leader (= holiday guy) for camera board cable ;P C) o Continue hacking for sound driver/framework :-) o OV camera export paper work --EOF-- thank you [16:46] so what's the situation with the OV datasheet ? do you have it ? Now I have it But I need to export paper work (In secret, Maybe I can put it on Redmine) [16:47] https://osdr.renesas.com/projects/linux-kernel-development/wiki/Salvator-X_MAXIM_camera_board (not yet) regardless of how we transfer the datasheet, it will be covered by our NDA, so that part is fine Yes [16:48] regarding the camera board cable [16:49] what's the issue ? :-) issue is that whether you can use existing power cable or not [16:51] if not, I need to export power cable, but it needs more paper work let me quickly check You can use Lager/Koelsh/Salvator power cable is that a regular 12V power supply with a barrel connector ? [16:52] Do you mean Lager/Koelsh/Salvator power cable ? then yes (2.1/5.5mm center positive) [16:53] that's easy to source no need to send them from Japan Nice !! we can order power supply pretty much anywhere if needed, that will be much faster [16:54] You saved me !! In the email you also state two power cables could be used right? One to Salvator-X and one to the Camera board and as long as you already have two Gen2/Gen3 power cables you are ok neg: Yes 1 note is that camera board doesn't have reset button [16:55] You need to plug out/in for reset camera board :P I think above url has such explanation do you know if the salvator-x and camera boards can be powered up in any order, or is a specific order required to avoid damaging the hardware ? [16:56] Yes, it has special order. you can check it on above URL see. Power ON order [16:57] thanks And special know-how for jumper pin, too thank you [16:58] dammsan: you're next nothing to report here really just sent out an email regarding V2H and serial multiplexing [16:59] not sure if it made it to periperi or not with the large size photo of schematic this is related to "plan b" in case ulrich can't get it going apart from that nothing to report [17:00] thank you [17:01] I'm next [17:02] I've been on holidays for two weeks, so not much to report my short term plan is catching up with my e-mail backlogs and with patch reviews kbingham: you're next [17:03] Ack! Nothing to report since last time, I've been working on other projects as my Renesas budget was empty. In the upcoming weeks, I expect that to replenish, and will be getting my existing series' rebased and following up on review comments. I have a small support task for jinzai which should be solving today, and then I will be following whatever the latest plan is on our multimedia task force. The notes from the last meeting suggest that I will be looking at the ADV7482 sometime, but I don't know any further details about this planned work yet. -- EOT -- [17:04] thank you [17:05] Any comments? Is the ADV7482 still my task in the plan? yes I propose discussing it with Niklas early next week [17:06] I was about to say the same :D Monday morning or Tuesday would work best for me Tuesday is better for me ... But I can do monday. neg: how about you? both work for me oh wait Actually I think Tuesday might be bad for me :) * kbingham checks I have to go out at 10am on Tuesday, so Monday is better :) [17:07] so standard meeting time monday morning? Can we make it 9am UK time :) [17:08] works for me sure great. I've sent you an invitation [17:09] Received. [17:10] jmondi: you're next (and last) thanks! [17:11] few things to report since last time I only spent 1 week of my time on multimedia, and I have ported the DRM/KMS test application to gen 3 not tested with VIN patches yet but I have closed pinchartl review comments on last version [17:12] I should send out an email with details probably... that would be nice apart from that not much, still waiting to have a more defined plan for next months, where I should be all about multimedia thank you [17:13] I guess the OV datasheet export issue has impact on me, as I was supposed to work on ov10(don't remember name) sensor, right? correct [17:14] morimoto: maybe the datasheet should appear magically on a server somewhere nobody will know where it came from [17:15] and nobody will ask any question for just a few minutes maybe? life will be good, everybody will be happy pinchartl: actually, it will takes long term for export [17:16] but, I can put it on Redmine if I can be guilty guy [17:17] nobody will tell :-) Hehe :) jmondi: when do you need it ? immediately ? [17:18] morimoto: the sooner the better OK... since we're both online, you can upload and delete in 5 minutes Please pray that no stone-head guys come here :) there's usually an application note that goes with the OV sensors datasheets if you have to too, it would be nice to get it It will be formal, but not yet now * kbingham grabbed a copy to see too :) [17:22] no problem I only have datasheets, not application note that closes the first topic Topic 2. Multimedia plan If you need it, I can ask to OmniVision guy dammsan: what's the status of base contracts and additional tasks for Q2 ? morimoto: it would be nice to get it, the datasheet usually lacks important information that is present in the application note pinchartl: OK, will do [17:23] neg: I will delete datasheet, are you OK ? yes [17:24] thanks, and what datasheet :-) pinchartl: the base tasks have been submitted, waiting for processing [17:25] additional tasks do you expect any issue or delay for the base tasks ? not really i think the only unknown is that camera prototype at this point * morimoto bye-bye datasheet [17:26] i expect base tasks to be ready this week ok how about the additional tasks ? [17:27] i think your idea with the shared more generic task description is good so i would like to proceed with that [17:28] however, i would like to know that we at least have _something_ that works first uli___: I think there's a bit of pressure on you :-) [17:29] from my side, we can be blocked on several other random issues before we get there but in any case we have to go in that direction (who knows, another power supply needed or perhaps some jumper is missing) which is why I want more generic tasks [17:30] to handle the issues as we progress sure i mainly want results myself =) have you submitted the generic task proposal internally to Renesas to get their feedback ? [17:31] nope we've been busy with base tasks Q2 starts next week, and I think everybody is getting nervous [17:32] this is why we have the base tasks i would have preferred to have overlapping ones myself but hey how would you like to proceed ? i want to get the updated result from the camera prototype [17:33] i propose that we revisit next week, hopefully some update is available then? that should have sorted out the base tasks too ok [17:34] by the way morimoto: what's the status about the hardware setup that should be sent to Europe ? [17:35] in the mean time, lets do ping-pong-email you and me in case some detail need to be adjusted in the additional task descripiton pinchartl: camera board ? [17:36] morimoto: yes [17:37] camera *board* I already have camera itself will be arrive at 5 5 = May at 5 ? 5pm today ? :-) ok 5pm in may when in May ? Hehe Don't know, just "May" [17:38] so potentially we won't have access to local hardware before end of May ? Unfortunately we can pick them at LCJ directly :/ [17:39] ok jmondi: Unfortunately, it is not allowed that's not very nice, but we have no choice how comes the camera will arrive so late ? is it because of the lead time from the manufacturer ? I'm not sure, I know is that it will came from Philippine [17:40] who's the manufacturer ? do you have a model number ? [17:41] Module name is "RDACM20-01" [17:42] thank you manufacturer seems... IMI Japan [17:43] I don't think it makes sense to really update the multimedia plan at this point before we have results from Ulrich's work [17:45] uli___: is there any way we can help to speed it up ? during this meeting i have received an e-mail from my tax guy, who has failed to do my taxes for 2015, which are at the absolute latest due tomorrow, until now [17:46] i need to look up some documents http://www.global-imi.com/about-imi/global-facilities/ does not list Japan. But it does list Czech Republic and Bulgaria i'm afraid that has 110% priority i'll still try to do the test today, though uli___: let's discuss the results tomorrow morning then [17:47] ok Topic 3. Meeting around LinuxCon Japan [17:48] we have agreed on dates for the trip [17:49] with a VIN code camp with local hardware access on the 29th and 30th of May so far no meeting with the BSP team is scheduled, as no need to have such a meeting has been expressed [17:50] morimoto: dammsan: does that still hold ? or do you think we should organize a meeting ? pinchartl: no plan with BSP team meeting at this point. [17:51] pincharl: we should go out for drinks together at least =) [17:52] dammsan: of course :-) I'd like to spend as much time coding as possible [17:53] so if meetings are needed, I'd like to know ASAP makes sense I would also like to book my trip soonish :-) * jmondi too ^ [17:54] Yes, flights have increased in price by 30% already here I think. (google keeps telling me when the price changes) Of course that could just be because the £ has decreased in value :) [17:55] I think I managed to clear my previous date troubles so I should be avaliable from/to both weekends before and after and be happy to stay both if other ppl are around for code or drinking sessions ;-) neg: \o/ neg: you're talking about the May 27th/28th weekend and June 3rd/4th, right? [17:56] morimoto: persmission to book flights then ? I can do before much easier than after :) [17:57] jmondi: yes pinchartl: sorry, what do you mean? morimoto: can we go ahead and book flights for the week in Japan ? and can we invoice them to Jinso ? [17:58] Ahh, OK. I will ask to our boss, please wait thank you [18:00] last topic, next meeting I propose two weeks from now [18:01] Tuesday the 11th of April works for me I guess at 8:00 GMT / 09:00 CEST / 10:00 EEST / 16:00 JST (usual time) sounds good Ack here. [18:02] good, I'll send the invite now pinchartl: how many people caming to Japan ? morimoto: Ulrich can't make it, so that would be Jacopo, Kieran, Niklas and me And Core/IO group member ? [18:03] date works for me pinchartl: You really want to conflict with all future Core Meetings? morimoto: I don't know about Gert or Wolfram's plans for LCJ geertu: good point should we make it Wednesday the 12th ? pinchartl: What happened with the MM meetings on Wednesdays? geertu: lack of sleep I think :-) [18:04] we've switched this week, but let's go back to the original schedule so Wednesday the 12th at 8:00 GMT / 09:00 CEST / 10:00 EEST / 16:00 JST gotcha [18:05] I propose closing this meeting does anyone second? pinchartl: Thank you! pinchartl: about trip budget, maybe OK geertu: you're welcome pinchartl: about 4/12 2nd [18:06] 4/12 2nd ? s/2nd/second/ I mean OK for 4/12 meeting [18:07] :-) 4th December is a bit late :D perfect thank you all for attending and have a nice day Cheers all thanks all kbingham: oops, 4/12 = 4th December in Europe, it is April, 12th in Japan ;P [18:08] cheers! see you thanks