path: root/wiki/Chat_log/20220310-mm-chatlog

10:01 < pinchartl> welcome to the multimedia team meeting
10:01 < pinchartl> Topic 1. Status Check for the Multimedia Tasks
10:01 < pinchartl> * Jacopo:
10:01 < pinchartl> Since last meeting:
10:01 < pinchartl> - ADV748x Virtual channel support
10:01 < pinchartl> - Review/discussion with Niklas of R-Car CSI-2/VIN routing
10:01 < pinchartl> Until next meeting:
10:01 < pinchartl> - New version of adv748x based on discussion with Niklas
10:01 < pinchartl> Issues and blockers: None
10:01 < pinchartl> jmondi: any comment ?
10:01 < wsa> geertu: yes, we have
10:01 < jmondi> not really, happy niklas' series has landed so I can rebase on it
10:02 < moriperi> jmondi: I think nothing happen about periject index.html :)
10:02 < pinchartl> * Kieran:
10:02 < pinchartl> Since last meeting:
10:02 < pinchartl> - V3U ModeTest issue tested
10:02 < pinchartl>   - Can't replicate the issue from BSP team.
10:02 < pinchartl>     Asked for clarification on the branch they are using.
10:02 < pinchartl> - V3U SN65DSI86 Hotplug DP implemented
10:02 < pinchartl>   - A few more hurdles than expected along the way, but it works now.
10:02 < pinchartl>   - v2 posted, but now have custodianship of other patches which need
10:02 < pinchartl>     more work.
10:02 < pinchartl> - Reviews and GMSL support
10:02 < pinchartl> Until next meeting:
10:02 < pinchartl> - Push through SN65DSI86 review comments for Optional Connectors
10:02 < pinchartl> Issues and Blockers: None
10:02 < pinchartl> kbingham: any comment ?
10:02 < jmondi> moriperi: my understanding from the conversation was that the patch was not required ?
10:03 < geertu> moriperi: Have you upgrade your browser? I can push the fix I'm carrying locally.
10:03 < kbingham> Only EDID support review question blocking upstreaming of the V3U display now I think. Not long to go.
10:04 < moriperi> geertu: thanks. please push it. I don't have the issue so far, but will get :)
10:04 < pinchartl> kbingham: we're seeing the end of the tunnel, nice
10:04 < pinchartl> * Laurent:
10:04 < pinchartl> Since last meeting:
10:04 < pinchartl> - V4L2 multiplexed streams development
10:04 < pinchartl> Continued working with Tomi Valkeinen. v11 on the list, still work in progress.
10:04 < pinchartl> - Reviewed patches from BSP team
10:04 < pinchartl> Partial fix sent upstream for DU flicker. Offset calculation fix for multi-planar format cropping in VSP is ready but missed the v5.18 merge window.
10:04 < pinchartl> - Passed bricked ULCB H3 ES2.0 to Wolfram
10:04 < pinchartl> Wolfram needed access to a Kingfisher setup with a ULCB. He now has the hardware, which came with the need to unbrick the ULCB :-)
10:04 < pinchartl> Until next meeting:
10:04 < pinchartl> - V4L2 multiplexed streams development
10:04 < pinchartl> Issues and blockers:
10:04 < pinchartl> - No V4H documentation
10:04 < pinchartl> One upport request from the BSP team is related to support of a new pixel format for the VSP in V4H. Without access to hardware and without documentation, this is difficult to upstream.
10:05 < pinchartl> any question ?
10:05 < pinchartl> we've already discussed the documentation question, hopefully that can be resolved soon, and I'll then upstream the new V4L2 formats
10:05 < moriperi> pinchartl: thank you for helping BSP team
10:06 < pinchartl> you're welcome
10:06 < pinchartl> there's still one DU flicker issue
10:06 < pinchartl> but I'm not sure we can solve that one. it's caused by the hardware design
10:06 < moriperi> And thank you for upstream support without HW, Doc :)
10:07 < pinchartl> :-)
10:07 < pinchartl> * Marek
10:07 < pinchartl> Since last meeting: None
10:07 < pinchartl> - GPU - Investigated PowerVR GX6250 driver was posted recently
10:07 < pinchartl> https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/15243
10:07 < pinchartl> Until next meeting:
10:07 < pinchartl> - Keep digging in the PowerVR driver !
10:07 < pinchartl> Issues and blockers:
10:07 < pinchartl> - PowerVR firmware blob compatibility between different DDK versions.
10:07 < pinchartl> marex: welcome to the MM group :-D
10:07 < pinchartl> and nice to see interest in GPU support
10:07 < pinchartl> hopefully translated into progress soon
10:07 < marex> pinchartl: not unless there is suitable firmware blob
10:08 < marex> because powervr kernel driver depends on specific ABI mutation of the firmware blob and the blob provided by renesas is not compatible
10:08 < marex> so we now have hard dependency between kernel driver and powervr firmware ... lovely
10:08 < pinchartl> is there any hope of getting a suitable blob, or of supporting the blob we have in the kernel driver ?
10:08 < geertu> moriperi: jmondi: pushed
10:08 < marex> I'll see if I can get suitable powervr firmware for r8a77960 out of the various renesas releases, maybe there is a compatible one somewhere there
10:09 < pinchartl> ok
10:09 < jmondi> geertu: thanks ;)
10:09 < moriperi> geertu: jmondi: thanks
10:09 < marex> pinchartl: I _suspect_ if I use blob built with DDK 1.14 , that _might_ be compatible
10:10 < marex> pinchartl: the latest greatest blobest blob is built with DDK 1.15, while the acer r13 blob provided alongside the MR is built with DDK 1.14
10:10 < pinchartl> ok, sounds like more investigation is needed
10:10 < pinchartl> keep us posted :-)
10:10 < marex> sounds like stable ABI is needed
10:10 < marex> or blob sources ;-)
10:10 < pinchartl> even better :-)
10:11 < pinchartl> * Morimoto-san
10:11 < pinchartl> Since last meeting: None
10:11 < pinchartl> Until next meeting: None
10:11 < pinchartl> Issues and blockers: None
10:11 < pinchartl> I'm not implying that Morimoto-san stopped doing anything :-)
10:11 < pinchartl> just busy on the new internal super-secret project
10:11 < pinchartl> any comment ?
10:11 < moriperi> no. thanks
10:12 < pinchartl> * Niklas:
10:12 < pinchartl> Since last meeting:
10:12 < pinchartl> - The VIN routing patches have recently been picked up, clearing my pending patch backlog for the first time :-)
10:12 < pinchartl> - Discussed the proposed V4L2 error event with other interested parties and started work to restart that series.
10:12 < pinchartl> Until next meeting:
10:12 < pinchartl> - Rebase and repost the UDS work that is now unblocked
10:12 < pinchartl> Issues and blockers: None
10:12 < pinchartl> neg: any comment ?
10:12 < neg> No comment from me, thanks.
10:12 < pinchartl> any discussion topic for today ?
10:13 < moriperi> geertu: oops? I though the solution is using hard link instead of symbolic link
10:14 < pinchartl> I'll take that as a no
10:14 < pinchartl> regarding the meeting on the 31st
10:14 < uli> marex: i have found rgx.fw.4.43.6.62 and rgx.fw.4.46.6.62 (for h3) and rgx.fw.4.45.2.58 (for m3) in my archives. would those be of any use to you?
10:14 < pinchartl> Europe will have switched to summer time
10:15 < pinchartl> so will it be 09:00 CEST, 16:00 JST, or 10:00 CEST, 17:00 JST ?
10:15 < marex> uli: run strings on it and see if there is _1.14 or _1.15 in each blob
10:15 < wsa> oh, I need to fix it in my announcement then
10:15 < wsa> 09 CEST and 16 JST
10:15 < marex> uli: I downloaded all those rgx.rw packages from renesas website already, still need to analyze them
10:15 < pinchartl> ok, so one hour earlier for Japan
10:16 < marex> uli: if those blobs are from renesas website, I likely already have them
10:16 < geertu> moriperi: oops. is it broken for you now?
10:17 < pinchartl> if that's all for today, I propose adjourning the meeting. does anyone second ?
10:17 < uli> marex: i don't actually remember where those are from, but no version numbers to be found with strings
10:18 < jmondi> seconded :)
10:18 < geertu> pinchartl: oh right, completely forgot about DST
10:18 < jmondi> so we'll meet again on the 31st as our last meeting ?
10:18 < pinchartl> meeting adjourned, thank you all for attending