Chat Report 2016-07-06 We had a multimedia group meeting on 2016-07-06. Here's a summary of the discussions. Please correct any mistake you would notice. Attendees: - Kieran - Laurent - Morimoto-san - Niklas - Ulrich Magnus was excused. Topic 1. Status check for the multimedia tasks ---------------------------------------------- - FDP,v4.8,prototype,kieran,Develop and upstream driver Patches have been posted for public review, DT bindings have been acked. Development itself is complete, the task will be split in two with only the upstreaming part remaining. The upstreaming target is v4.9. - AUDIO-DMAC,?,plan,magnus,IPMMU integration on Gen2 - AUDIO-DMAC,?,plan,magnus,IPMMU integration on Gen3 - AUDIO-DMAC,?,plan,magnus,IPMMU support on Gen2 - AUDIO-DMAC,?,plan,magnus,IPMMU support on Gen3 - VIN,?,prototype,magnus,IPMMU integration on Gen2 - VIN,?,prototype,magnus,IPMMU integration on Gen3 - VIN,?,prototype,magnus,IPMMU support on Gen2 - VIN,?,prototype,magnus,IPMMU support on Gen3 No progress. - RSND,v4.10,public,morimoto,DT bindings for graph sound - RSND,v4.10,public,morimoto,dw-hdmi-i2s-audio prototype on Gen3 - RSND,v4.10,prototype,morimoto,HDMI SSI prototype on Gen3 - RSND,v4.10,plan,morimoto,HDMI sound Upstream support without hotplug on Gen2 - RSND,2016-09-30,plan,morimoto,Hotplug support upstream on Gen3 The ALSA + HDMI patches are organized as follows. 1. Cleanup 2. OF graph 3. OF graph-based new sound card 4. HDMI sound driver Patches for 1. and 2. have been posted. Cleanup is progressing slowly. OF graph DT bindings (2) have passed maintainer's review, but the implementation (3) is required before the patches can be merged. DT bindings will also be discussed during PeriPeriCon. - ADV7482,v4.7,public,niklas,Prototype on Gen3 - VIN,v4.7,public,niklas,CSI2 prototype (Gen3) The tasks are complete. - ADV7482,v4.9,plan,niklas,Gen3 support upstream - ADV7482,v4.9,plan,niklas,Interlace support upstream - VIN,v4.9,plan,niklas,CSI2 interlace support upstream (Gen3) - VIN,v4.9,plan,niklas,CSI2 support upstream (Gen3) - VIN,v4.9,plan,niklas,Gen3 support upstream (without CSI-2) No progress. - VIN,v4.8,public,niklas,Gen3 support No progress. Upstreaming requires deciding how to handle the ADV7482, moving the task to v4.9 as a result. - VIN,v4.8,plan,niklas,Scaler support (on Gen3) No progress. A prototype will hopefully be implemented before PeriPeriCon. Moving to v4.9 for upstreaming. - DU,v4.7,plan,ulrich,Atomic API test program - DU,v4.7,prototype,ulrich,Test setup with HDMI output to HDMI input loopback (without EDID) The test application and setup are public, marking as complete. - DU,v4.7,public,ulrich,HDMI output on Gen3 prototype The prototype is public, marking as complete. - DU,v4.7,public,ulrich,EDID generation support for the HDMI loopback test setup Patches will be posted today. - DU,v4.9,plan,ulrich,HDMI output on Gen3 upstream No progress. - VIN,v4.8,public,ulrich,Add DV timings support to rcar-vin The patches have been merged to the linux-media tree, they should end up in v4.8. Marking as complete. - DU,?,plan,laurent,IPMMU integration on Gen3 - DU,?,plan,laurent,IPMMU support on Gen3 (through VSPD+FCP) - DU,v4.9,plan,laurent,Fix 3 planes formats - VSP,?,plan,laurent,Fixed alpha support (VI6_DPR_*_ROUTE.FXA) No progress. - VSP,?,plan,laurent,UDS regression fix UDS on Gen3 can scale images wider than 128 or 256 pixels. It requires slicing (partitioning) input images in software and running the VSP pipeline for each slice. Renaming the task to "Image partitioning support". This will require a significant amount of work. There's no completion estimate as the task hasn't been scheduled yet. - VSP,v4.8,plan,laurent,Fix suspend/resume crash No progress. The task isn't scheduled, removing the target version. - VSP,v4.8,public,laurent,HGO operation mode selection - VSP,v4.8,public,laurent,HGO support upstream on Gen3 - VSP,v4.8,public,laurent,HGO test application No progress. A pull request will hopefully be sent this week, but this will be too late for v4.8. Moving to v4.9 as a result. - VSP,v4.8,prototype,laurent,Rotation support Rotation itself is implemented, but requires image partitioning support for images wider than 256 pixels. Removing the target version due to the dependency. - VSP,v4.8,public,laurent,V4L2 request API usable prototype Patches should go out today. - VSP,?,prototype,laurent,V4L2 request API upstream No progress. Topic 2. Additional '50%' tasks (June) -------------------------------------- The additional tasks are - VIN CSI-2 Support Prototype - ADV7482 Driver Prototype - VSP CLU Improvements - VSP Rotation Support - HDMI Input/Output Loopback for R-Car Gen2 - FDP Support on Gen3 All tasks are complete, with a caveat for rotation support that requires image partitioning for images wider than 256 pixels. Rotation will automatically work for wider images once image partitioning is implemented. Topic 3. Additional '50%' tasks (Q3) ------------------------------------ Due to unforeseen and sad circumstances that required Magnus to travel to Sweden negotiation of the tasks got delayed. The two tasks currently planned for Ulrich are - VIN HDMI input EDID - VIN integration on Gen2 For Niklas we have initially discussed the UDS regression fix, but as it turns out it will require more work than expected it wasn't a good match. The current plan is to implement support for one of the currently unimplemented VSP modules. The SHP (sharpener) was mentioned but the hardware isn't fully documented and the available source code just programs registers with parameters passed by upper software layers that are not available to us. We will thus likely switch to a different module. Topic 4. PeriPeriCon & RenesasCon preparation ---------------------------09:32 < geertu> Welcome to today's Core Group Chat Meeting! 09:32 < geertu> Agenda: 09:32 < geertu> 1. Status Updates 09:32 < geertu> 2. Discussion Topics 09:32 < geertu> A) What have we done since last time: 09:32 < geertu> Marek worked on U-Boot (driver model conversion for Lager/Koelsch/Gose, 09:32 < geertu> and experiments with OpenOCD/USB), ATF (passing DT to OpTee OS), and 09:32 < geertu> VLAB VM OE prototyping. 09:32 < geertu> Morimoto-san posted SH warning cleanup patches, and worked on the 09:32 < geertu> Renesas Wiki transfer project. 09:32 < geertu> Niklas tested the re-enablement of debugfs. 09:32 < geertu> Geert updated periject/core for v5.5-rc1, added more M3-W+ support, 09:32 < geertu> investigated merge window issues, sent various fixes for DTS and PFC 09:32 < geertu> validation, and split/renamed R-Car H3 and ULCB support. 09:33 < geertu> B) What we plan to do till next time: 09:33 < geertu> Marek plans to continue working on OpTee and JTAG. 09:33 < geertu> Morimoto-san plans to go snowboarding, and let presents and blessings 09:33 < geertu> rain down on all the people. 09:33 < geertu> Shimoda-san plans to convert rcar-dmac and ipmmu DT bindings to 09:33 < geertu> json-schema, and will continue using SMMU-v3. 09:33 < geertu> Geert plans to have holidays, post new sh-pfc runtime checks, and resume 09:33 < geertu> DMAC sysfs work. 09:34 < geertu> C) Problems we have currently: 09:34 < geertu> Marek is still suffering from the upstream ATF (non)process. 09:34 < geertu> Morimoto-san needs a well-deserved power boost. 09:34 < geertu> Shimoda-san has difficulty finding time for upstream activities. 09:34 < geertu> Geert wonders if the sh73a0 DTSes should be renamed to r8a73a0. 09:35 < geertu> ---EOT--- 09:35 < geertu> Anything I missed? 09:35 < geertu> Marex: I assume VLAB VM OE prototyping is for Magnus' lab? 09:40 < Marex> geertu: yes 09:41 < Marex> geertu: so we can run OpenOCD easily for example 09:41 < Marex> geertu: which reminds me, I might want to add gitlab-runner into the VM, so we can implement CI workers easily 09:41 < geertu> Marex: OK, that's what I thought it was 09:41 < Marex> geertu: which, yes, would then allow for easy CI of the boards :-) 09:41 < Marex> so much possibility for improving the process here 09:42 < geertu> Topic 2. Discussion Topics 09:42 < geertu> A) Renesas Wiki 09:42 < geertu> But we already discussed that in I/O 09:43 < geertu> Anyone with an opinion about sh73a0=>r8a73a0? 09:43 < morimoto> Oops which repository we will use ? Gitlab ? 09:43 < geertu> The latter name is already mentioned in several files 09:43 < Marex> morimoto: I would prefer gitlab, because of the wiki and CI runner support 09:43 < geertu> morimoto: Hasn't been decided yet 09:43 < Marex> morimoto: then you can CI boards on push :) 09:44 < morimoto> OK :) 09:44 -!- pinchartl [~laurent@85-76-163-249-nat.elisa-mobile.fi] has joined #periperi 09:44 < geertu> Marex: gitlab@gitlab or gitlab@elsewhere? 09:44 < Marex> could save a lot of manual work with boards 09:44 < Marex> geertu: private instance of course 09:44 < Marex> geertu: gitlab@magnus :) 09:44 < geertu> Marex: I believe you can still host a private instance @gitlab, too? 09:44 < pinchartl> (gandi has to fail me right during the meeting :-S sorry about that) 09:45 < wsa> I gotta run for today 09:45 < Marex> geertu: I don't want to ? 09:45 < Marex> geertu: esp. with sensitive stuff like renesas topics 09:46 < wsa> please let me know if there are new conclusions about next meeting or new repos 09:46 < wsa> have a nice day! 09:46 -!- wsa [~wsa@p54B33138.dip0.t-ipconnect.de] has quit [Quit: ...] 09:49 < geertu> Marex: OK, so how to add gitlab to your VM? ;-) 09:50 < geertu> Anything else to discuss? 09:50 < Marex> geertu: no 09:51 < Marex> geertu: you only need the runner in the VM 09:51 < Marex> geertu: the gitlab instance can run somewhere else, maybe in another VM :) 09:52 < geertu> Marex: <ironic>And the latter has no need to be set up?</ironic> 09:53 < Marex> geertu: that's a separate topic 09:54 < Marex> geertu: so, do you have another suggestion for a git server (with wiki and CI) ? I use gitlab to satisfaction this way for years :) 09:54 < geertu> Marex: No more suggestions from my side 09:55 < Marex> :( 09:55 < Marex> geertu: but do you at least agree that the CI would be quite helpful ? 09:55 < geertu> I there are no more topics to discussion, I'll pass the mic to pinchartl 09:55 < geertu> Marex: Yes, a CI is nice 09:55 < Marex> good 09:55 < geertu> But do we need a CI for the wiki? 09:56 < Marex> geertu: errrr, what ? 09:56 < Marex> geertu: wiki could be a replacement for the redmine we use now 09:56 < Marex> geertu: CI would CI the U-Boot/Linux patches on the boards to keep the quality high, without manual frobbing 09:56 -!- pinchartl [~laurent@85-76-163-249-nat.elisa-mobile.fi] has quit [Quit: leaving] 09:56 < geertu> Currently the git repo just has periject 09:57 < geertu> Do we need a CI for periject? 09:57 -!- pinchartl [~laurent@perceval.ideasonboard.com] has joined #periperi 09:58 < Marex> geertu: you can add u-boot/linux/atf/opteeos repos there and trigger CI on push 09:59 < geertu> Marex: All of that is public work, so no need to host it locally at Magnus 09:59 < geertu> Anyway, time's up 09:59 < Marex> geertu: does korg support CI workers now ? 09:59 < geertu> Thanks for joining, and have a nice continued day 09:59 < geertu> Marex: it relies on e.g. kernelci 10:00 < geertu> pinchartl: ready? 10:00 < Marex> geertu: which is only for kernel 10:00 < pinchartl> geertu: yes, thank you 10:00 < Marex> geertu: and can you register custom workers ?
generated by cgit v1.2.3 (git 2.25.1) at 2025-01-10 03:14:13 +0000
ic,ulrich,HDMI output on Gen3 prototype DU,v4.7,prototype,ulrich,Test setup with HDMI output to HDMI input loopback (without EDID) DU,v4.7,public,ulrich,EDID generation support for the HDMI loopback test setup DU,v4.9,plan,ulrich,HDMI output on Gen3 upstream the test setup is up on elinux/github [17:28] edid generation will go out today that's it do you mean the HDMI loopback test setup ? [17:30] yes how about DU,v4.7,plan,ulrich,Atomic API test program ? i moved the wiki pages that's what was missing, iirc thanks [17:31] and DU,v4.7,public,ulrich,HDMI output on Gen3 prototype [17:32] is that considered as complete ? let's consider it as such. [17:33] there was also [17:34] VIN,v4.8,public,ulrich,Add DV timings support to rcar-vin that has been picked up recently nice [17:35] by Hans ? i just saw the auto-generated e-mail for the media tree then they have been applied by Mauro, perfect next, me, messing up the alphabetical order again [17:37] DU,?,plan,laurent,IPMMU integration on Gen3 DU,?,plan,laurent,IPMMU support on Gen3 (through VSPD+FCP) DU,v4.9,plan,laurent,Fix 3 planes formats no progress VSP,?,plan,laurent,Fixed alpha support (VI6_DPR_*_ROUTE.FXA) no progress VSP,?,plan,laurent,UDS regression fix [17:38] it turns out that this requires implementing image partitioning support in the VSP driver as, unlike the VSP1 in Gen2, the VSP2 in Gen3 can't scale images wider than 128 or 256 pixels (I have to check for the exact number) so the images must be sliced by software, and the VSP run for each slice [17:39] I'll thus rename the task to "Image partitioning support" [17:40] UDS will work out of the box then pinchartl: about UDS regression, it was VSP driver side issue do you mean ? [17:41] it wasn't a regression [17:42] or rather it's a hardware regression Ahh, OK that requires significant additional work in the VSP driver VSP,v4.8,plan,laurent,Fix suspend/resume crash [17:43] this isn't scheduled, I'll remove the target version VSP,v4.8,public,laurent,HGO operation mode selection [17:44] VSP,v4.8,public,laurent,HGO support upstream on Gen3 VSP,v4.8,public,laurent,HGO test application no progress I hope to send a pull request this week but it might be too late for v4.8 I'll move it to v4.9 [17:45] VSP,v4.8,prototype,laurent,Rotation support rotation itself is implemented but it also requires image partitioning support for images wider than 256 pixels [17:46] this is a dependency that I wasn't aware of I'll thus remove the target version due to the dependency on image partitioning that hasn't been scheduled yet [17:47] VSP,v4.8,public,laurent,V4L2 request API usable prototype [17:48] I was hoping to send the latest version Monday last week but got delayed, partly by the preparation of additional tasks for the multimedia team for Q3 [17:49] and then by being unavailable from Friday to Tuesday I'll resume now, the patches should go out today BSP team want to use latest verion on renesas-driver. [17:50] as soon as the patches are out I'll of course send a pull request to Geert [17:51] sound gr8 that's it for the tasks [17:53] Topic 2. Additional '50%' tasks (June) pinchartl: what is current status of CLU/LUT ? merged nice ! - VIN CSI-2 Support Prototype [17:54] - ADV7482 Driver Prototype - VSP CLU Improvements - VSP Rotation Support - HDMI Input/Output Loopback for R-Car Gen2 - FDP Support on Gen3 I believe everything is complete except for VSP rotation support that is complete for images not wider than 256 pixels Topic 3. Additional '50%' tasks (Q3) [17:55] uli___: what is latest status of HDMI out for ES1.1 issue ? (go ahead, we'll move to topic 3 after) [17:56] pinchartl: sorry still todo... OK. creation of the 50% tasks for Q3 has been delayed [17:57] I've submitted the tasks for Ulrich, all others are still under discussion with Magnus due to unforeseen and sad circumstances it didn't go as expected uli___: have you received an SoW from Jinzai ? not yet [17:58] for the 50% tasks, that is ok the two tasks currently planned for Ulrich are [17:59] VIN HDMI input EDID VIN integration on Gen2 for Niklas we have initially discussed the UDS regression fix [18:00] but as it turns out it will require more work than expected it wasn't a good match the current plan is to implement support for one of the currently unimplemented VSP modules the SHP (sharpener) was mentioned [18:01] but the hardware isn't fully documented and the available source code just programs registers with parameters passed by upper software layers that are not available to us so we'll likely switch to a different module for Kieran and me the situation isn't clear either, I expected feedback from Magnus today [18:02] we'll try to fix this asap [18:03] Understandable if magnus is away. Topic 4. PeriPeriCon & RenesasCon preparation [18:04] unless there are questions about the additional tasks for Q3 ? I'm good for now :-) [18:05] for, for peripericon [18:06] s/for/so/ we'll discuss multimedia topics on Monday afternoon there's no fixed agenda yet now is the time to propose topics :-) I would like to talk about OF graph [18:07] extensions to v4l2 to support subdevices with more then one output pipeline anything else ? uli___ and kbingham I know you won't be there, but are there topics you would like us to discuss ? [18:09] can't think of anything [18:10] I'd like to discuss the following topics - Second batch of additional tasks for Q3 - What are we doing right or wrong, and how can we improve - Long-term roadmap hum I don't know if this is on the roadmap but if there is plans to support audio in or CEC from VIN HDMI how that would work. But maybe that is covered in morimoto topic about OF graph? [18:11] the current m2m design is limited and seems to prevent pipelined usage of hw ... but I think that's possibly a topic for linux-media . But that's the only thing I can think of on my future discussion lists. neg: I might be mistaken but I don't think we have hardware support for CEC [18:12] the target market for Renesas devices isn't exactly consumer electronics :-) [18:13] kbingham: yes, I think that's a topic for linux-media I'll write it down as a potential task for the future pinchartl: hum ofc CEC is terminated in the adv7482 so it might not be high on the agenda [18:14] ah, ADV7482 has CEC support ? good to know I'm sure Hans would be interested but I don't think Renesas would. we should of course ask same for audio input. Morimoto-san, do you know if there's an interest in HDMI audio input ? [18:15] Yeah, of course Oops, Do you mean BSP request ? BSP team doesn't have such request at this point. [18:16] either an existing BSP request, or something that would be of interest to them in the future same for HDMI CEC support, is there an interest ? but it seems they are starting about HDMI support for customer I can say BSP team is stating about "HDMI" only, they don't know detail of it :) [18:17] ok So yes, in the future, they will request such detail feature we can discuss that during RenesasCon [18:18] I'll send out the list of topics for PeriPeriCon in the meeting report, if anyone wants to add topics to the agenda it will still be possible until the end of the week Morimoto-san and Niklas, could you each prepare a very brief presentation for the topic you proposed for the meeting ? [18:19] it should be 5 minutes long at most just to get everybody on the same page to start discussions on email ? or here ? no, for PeriPeriCon [18:20] will do OK, will do thank you regarding RenesasCon I propose finalizing the preparation for it during PeriPeriCon Renesas is driving the agenda there [18:21] but it will be a good occasion to ask questions for instance regarding HDMI audio input and CEC I will also prepare a presentation of the multimedia group activities [18:22] neg: I've seen the slides you've sent, thank you morimoto: do you think we need a detailed presentation from each developer, or a single summary presentation of our activities ? [18:23] I hope they are on the right level for the event. Thanks morimoto for your comments! pinchartl: it up to you :) I can say it is free style [18:24] neg: no problem :P basically, 1 person has 10min. style is free [18:25] BTW, I created Wiki page for MiniPeriCon and RenesasCon [18:26] https://osdr.renesas.com/projects/linux-kernel-development/wiki#PeriPeriCon We can use it thank you I think that's it then [18:27] the next multimedia meeting will be during PeriPeriCon sounds nice :) and the one after that will be on Wednesday two weeks later, I'll send an invite * kbingham wishes he was going to Japan :) [18:28] kbingham: it's not too late to book a plane ticket ;-) I have a practical question for peripericon and renesascon, do anyone plan to bring a HDMI->VGA adapter incase we find an old beamer :-) kbingham: can we meet on Oct ELCE ? morimoto: Of course :) I'll be at ELCE pinchartl: It's not the ticket preventing me from going to japan next week :) [18:29] Oops, we need projector on MiniPeriCon ? I need to prepare morimoto: yes, that would be useful neg: I don't have one [18:30] me neither, so I will buy one then so we have if needed [18:31] I propose adjourning the meeting. does anyone second ? [18:32] seconded [18:33] thank you the meeting is adjourned, talk to you in Tōkyō for the ones who will be there [18:34] thanks all thanks Cheers and Enjoy! [18:35] pinchartl: when you arrive to Jinbotyo station A4 exit, you can email me by "kuninori.morimoto.gx@renesas.com" or "dr.mk.gemini@gmail.com" [18:36] morimoto: I don't think I'll have an internet connection Maybe it can be good lunch time :) Oops, from Hotel ? I can probably send you an e-mail when I leave the hotel [18:37] and call you when I get to the station I can also send a text message to Niklas or Geert or Wolfram OK, sound nice !! Have a nice trip you guys !! [18:38] thank you note that, unfortunately, Japan in rainy season now. another Q, can I fit my US power plug into a JP socket physicly? (yes I know the voltages are different) the internet gives me different awnsers [18:39] I always using Japan plug on US without exchanger [18:40] I'm always using Japanese plug in US without exchanger nice thx OK, thanks you guys, I will quit [18:42]