path: root/wiki/Chat_log/20190620-core-chatlog

09:31 < geertu> Welcome to Today's Core Group Chat!
09:31 < geertu> Agenda:
09:31 < geertu> 1. Status Updates
09:31 < geertu> 2. Discussion Topics
09:31 < geertu> Topic 1. Status updates
09:31 < geertu> A) What have we done since last time:
09:31 < geertu> Marek cleaned up the QoS and PFC drivers in ATF, and moved them out of
09:31 < geertu> staging.
09:31 < geertu> Morimoto-san gained one year of wisdom.
09:31 < geertu> Shimoda-san worked on exporting BSDL files to Marek.
09:31 < geertu> Simon posted CMT DT binding patches for R-Car H3/M3-N/E3.
09:31 < geertu> Geert sent clk-renesas and sh-pfc pull requests for v5.3, posted an RFC
09:31 < geertu> to create DMA slave links in sysfs, and implemented the
09:31 < geertu> .determine_rate() callback for Z and SD clocks on R-Car Gen2/Gen3.
09:32 < geertu> B) What we plan to do till next time:
09:32 < geertu> Marek may look into DDR support in ATF.
09:32 < geertu> Shimoda plans to perform export control for BSDL files.
09:32 < geertu> Geert plans to send a second clk-renesas pull request for v5.3,enable
09:32 < geertu> SDRAM on RZA2MEVB, and start queuing Renesas arm/arm64 patches for v5.4.
09:33 < geertu> C) Problems we have currently:
09:33 < geertu> Simon wonders if
09:33 < geertu> Documentation/devicetree/.../renesas-memory-controllers.txt should be
09:33 < geertu> renamed.
09:33 < geertu> EOT
09:33 < geertu> Anything I missed?
09:34 < geertu> horms: Personally, I don't like the proposed renesas-memory,controllers.txt
09:34 < geertu> What about renesas,dbsc.text (.yaml ;-)?
09:35 < geertu> As of R-Mobile A1, all SoCs have a DDR DRAM controller, hence DBSC
09:35 < geertu> Up to SH-Mobile AG5, it was SBSC (SDRAM ...)
09:36 < horms> Sure, that is fine
09:36 < horms> this is exactly the kind of suggestion I was hoping for
09:37 < geertu> Good ;-)
09:38 < geertu> Topic 2. Discussion Topics
09:39 < geertu> morimoto: Now the access to Magnus' server has been restored, do you have documentation updates?
09:41 < morimoto> No, I guess Renesas closed port 22
09:41 < morimoto> So, I have no chance to update datasheet so far
09:41 < geertu> It's port 10000
09:42 < damm> geertu: i am still waiting for an email with a summary of the problem from morimoto-san
09:43 < damm> it probably takes me 5 minutes to fix
09:43 < geertu> damm: my access was restored 2 weeks ago
09:43 < damm> yeah i restarted the vm
09:43 < damm> but i asked morimoto-san to send me an email at that point too
09:44 < damm> no response though
09:44 < damm> since long i added a workaround for renesas network
09:44 < damm> and it might need some update
09:44 < geertu> PEBCAC?
09:45 < damm> it doesn't matter how much you or morimoto-san look into it
09:45 < damm> the handle is on this side
09:45 < morimoto> I sent it via google hangout
09:45 < damm> i asked for email
09:45 < damm> how difficult can that be?
09:45 < damm> i asked for it 3 times
09:45 < damm> just send it
09:47 < morimoto> I sent it now
09:48 < geertu> Thx
09:48 < damm> thank you, does it work better now?
09:49 < kbingham> damm, Aha - I can confirm my update.sh script is now operational. :)
09:49 < Marex> same here
09:50 < kbingham> but now the speed has dropped to 4kB/s :D
09:50 < damm> i restarted the port forwarding for renesas
09:50 < kbingham> (but it was fine)
09:50 < morimoto> damm: wow !! It start works
09:50 < damm> good
09:50 < damm> thank you for sending the email
09:50 < morimoto> thanks
09:51 < geertu> 1.8 MB/s for R-Car-Gen3-rev1.50-20190605.zip
09:51 < kbingham> geertu, You're hogging all the bandwidth :)
09:51  * Marex furiously downloads
09:51 < geertu> kbingham: done
09:51 < morimoto> I will skip to send announce mail to PeriPeri-ML this time :)
09:51 < wsa> Yay, let's download all at the same time :D
09:51 < kbingham> Oh and I'm back up to 1MB/s
09:52 < kbingham> #periperi hammers damns' server :)
09:52 < Marex> kbingham: stop hogging the bandwidth, I'm at 500 kiB ;-)
09:52 < kbingham> hehe
09:52 < Marex> 19 kiB now
09:52 < damm> should be a gigabit link
09:52  * kbingham synced complete
09:52 < geertu> As you may have noticed from the report, the plan is for me to take over Renesas arm/arm64 maintenance from Simon as of v5.4
09:52 < Marex> 5.12 kiB/s :-D
09:52 < kbingham> damm, I guess it's the wet pieces of string in the middle :)
09:52 < geertu> Simon will continue to take care of fixes for v5.3
09:54 < horms> Yes, I have now closed my tree for v5.3. And will send pull-requests, handle any changes requested for them, and handle any fixes for v5.3.
09:55 < horms> Geert, could I ask you to hold off until the pull requests are sent until you accept anything for v5.4. i.e. until next week?
09:55 < Marex> shimoda: btw. do you think we could release that ULCB CPLD USB control tool in public ?
09:55 < geertu> horms: Sure, I had planned to wait to Jul 1
09:55 < geertu> s/to/until/
09:56 < horms> That is up to you :)
09:56 < geertu> horms: Thanks for taking care during all these years!
09:56 < horms> Thanks, its been a pleasure.
09:56 < shimoda> Marex: it's depend on Renesas Europe side policy. I heard from Goda-san that Renesas Europe made the tool
09:57 < Marex> shimoda: but I made my own tool, which implements the same protocol, and I would like to release my own tool
09:57 < pinchartl> horms: do I understand correctly that our ways will part (at least on the Renesas side) after v5.3 ? should I assume this applies to Kaneko-san too ?
09:57 < Marex> shimoda: the renesas europe windoze-only tool is not interesting to me
09:57 < horms> I am handing over the maintenance role now, as Geert described.
09:57 < Marex> shimoda: but maybe I can ask Mr. Dege from Renesas Europe for approval and if he says yes, I can release my own tool ?
09:57 < horms> Kaneko-san and I plan to part ways with Renesas at the end of next quarter, i.e. end of Sept.
09:58 < Marex> horms: :-(
09:58 < pinchartl> horms: I'm sure we will still hear from you, one way or another :-)
09:58 < horms> Yes, I don't plan to stop existing alltogether
09:59 < geertu> horms: Will you join Microsoft? ;-)
09:59 < pinchartl> that is very good news, although not really a surprise :-)
09:59 < morimoto> Marex: windoze :)
09:59 < shimoda> Marex: oh, you made the tool. Yeah, i think if Mr Dege accepts you can release in public
09:59 < pinchartl> horms: thank you for the information. so I can expect more MM patches from you and Kaneko-san in Q3 \o/
09:59 < horms> geertu: anything is possible!
09:59 < wsa> horms: thanks for all your maintenance work!
09:59 < jmondi> horms: I'm sorry to hear that, I hope you'll get around here anyhow after Q3
10:00 < horms> patersonc: yes. I think most of the patching is complete and I owe you a ticket update. I'll try and get that to you soon so we can clafify what is/isn't outstanding
10:00 < patersonc> horms: Thank you! It's a shame you're stopping!
10:00 < Marex> shimoda: yep, I analyzed the protocol and made my own tool, which would be convenient for CI / testing
10:00 < patersonc> horms: Sure, no worries
10:00 < horms> patersonc: sorry, i meant that response for pinchartl. But thanks for yuor good wishes
10:01 < horms> patersonc: I can also send you a list of patches if you like, but it won't be very different to what is visible on patchwork
10:01 < patersonc> horms: :) no worries
10:01 < patersonc> horms: The system seems to work
10:01 < Marex> shimoda: i didn't want to circumvent you, hence I asked you about releasing the tool first, but if it's renesas europe who's in charge, I can ask for approval there
10:03 < geertu> Anything else to discuss?
10:03 < shimoda> Marex: thank you for asking first :)
10:03 < patersonc> Marex: Asking Dege is probably a good idea.
10:03 < Marex> shimoda: you're welcome
10:04 < patersonc> horms:  What are your plans for the future?
10:04 < pinchartl> geertu: next meeting ?
10:04 < neg> flipping burgers at McDonalds ?
10:05 < pinchartl> now that many team members have seen their time allocation significantly reduced, I wonder if a bi-weekly meeting isn't too frequent
10:05 < kbingham> neg, That's automated now by IoT machines connected to the McNet :)
10:06 < morimoto> McNet :)
10:06 < wsa> pinchartl: i wondered about this, too, starting from Q4. But maybe Q3 is apropriate also?
10:07 < pinchartl> all additional tasks have been dropped, so we're down to 5 days a month on average for some of us
10:07 < pinchartl> and spending two mornings per months in meetings would be relative large in comparison
10:08 < geertu> so jul 4 is too early?
10:08 < kbingham> pinchartl, Indeed 4 hours of meetings per month is rather a large quantity of my renesas time budget
10:08 < geertu> jul 18 several people are in JP
10:09 < pinchartl> I think that on the multimedia side there will be very little to report by July the 4th
10:11 < geertu> For Core too, I think
10:11 < pinchartl> and the 18th will be during OSSJ, so not an option
10:11 < wsa> would that make July, 11th a candidate? I won't be available but we can handle IO stuff by mail, I'd think
10:12 < geertu> Jul 11 is fine for me
10:12 < morimoto> in 3 weeks pattern, 11th July -> 1st Aug -> 22th Aug...
10:13 < morimoto> or 4 weeks ?
10:13 < shimoda> or once per month? anyway Jul 11 is fine for me.
10:13 < pinchartl> we can try the 11th, but MM may need to be handled by e-mail too
10:14 < geertu> I don't want to be an exception ;-)
10:14 < geertu> we'll see
10:15 < geertu> Thanks for joining, and have a nice continued day!