path: root/wiki/Chat_log/20170720-core-chatlog

Core-chat-meeting-2017-07-20

11:01 < geertu> Welcome to today's Core Group meeting!
11:01 < geertu> Agenda:
11:01 < geertu> 1. Status updates
11:01 < geertu> 2. Discussion Topics
11:01 < geertu> Topic 1. Status updates
11:01 < geertu> A) What have I done since last time
11:02 < geertu> B) What I plan to do till next time
11:02 < geertu> C) Problems I have currently
11:02 < geertu> First is Morimoto-san
11:02 [Users #periperi]
11:02 [ dammsan] [ kbingham   ] [ marex-cloud] [ neg      ] 
11:02 [ geertu ] [ kbingham[m]] [ morimoto   ] [ pinchartl] 
11:02 [ jmondi ] [ Marex      ] [ mturquette ] [ uli___   ] 
11:02 -!- Irssi: #periperi: Total of 12 nicks [0 ops, 0 halfops, 0 voices, 12 normal]
11:03 < morimoto> OK
11:03 < morimoto> A)=B)=C)= NULL, sir
11:03 < geertu> Thank you, Morimoto-san!
11:03 < geertu> Next is Niklas
11:04 < neg> A)
11:04 < neg>     - No Core tasks.
11:04 < neg> B)
11:04 < neg>     - No Core tasks planed.
11:04 < neg> C)
11:04 < neg>     - No Core problems.
11:04 < neg> --EOT--
11:04 < geertu> Thank you, Niklas!
11:04 < geertu> Next is Geert
11:04 < geertu> A)
11:05 < geertu>   - V2 of CNTVOFF initialization and SMP for R-Car E2
11:05 < geertu>   - Playing with r8a77995/draak
11:05 < geertu> B)
11:05 < geertu>   - Publish CPG/MSSR for R-Car D3 after checking against datasheet rev. 0.55E
11:05 < geertu>   - Holidays
11:05 < geertu>   - Publich v2 of CPG/MSSR DT for R-Car Gen2
11:05 < geertu>   - Suspend/resume for PFC
11:05 < geertu>   - Mark periupport priority < H commits that are in linux-next
11:05 < geertu> C)
11:05 < geertu>   - None
11:05 < geertu> D)
11:05 < geertu>   - [PATCH] ARM: shmobile: rcar-gen2: Fix deadlock in regulator quirk
11:05 < geertu>     Needed to boot v4.13-rc1 on cold lager, koelsch, and gose!!!
11:06 < geertu> --EOT--
11:06 < geertu> Next is Magnus
11:07 < dammsan> as per email
11:08 < geertu> A)
11:08 < geertu>     - Flashed and hooked up D3 Draak for remote access
11:08 < geertu>     - Updated and posted IPMMU cleanup series
11:08 < geertu>       [PATCH v2 00/05] iommu/ipmmu-vmsa: 32-bit ARM update V2
11:08 < geertu> B)
11:08 < geertu>     - Update IPMMU Gen3 patches
11:08 < geertu>     - Consider sysfs interface for IPMMU driver
11:08 < geertu> C)
11:08 < geertu>     - None
11:08 < geertu> (FTR)
11:08 < geertu> Thank you, Magnus!
11:08 < dammsan> thx
11:08 < geertu> Next is Jacopo
11:09 < pinchartl> quick question, what is the sysfs interface for IPMMU about ?
11:09 < geertu> Ah, jmondi has left?
11:09 < morimoto> (borrowing morimoto keyboard)
11:09 < geertu> Next is Laurent
11:10 < pinchartl> #!/usr/bin/python
11:10 < pinchartl> for section in ('A', 'B', 'C'):
11:10 < pinchartl>         print('%s) None' % section)
11:10 < morimoto> we need a way to assign devices to IPMMU iova spaces
11:10 < geertu> python: command not found
11:11 < morimoto> (says magnus)
11:11 < geertu> Thank you, Laurent!
11:11 < geertu> Next is Marek
11:11 < pinchartl> morimoto: by iova spaces, do you mean the IPMMU TLBs ?
11:11 < Marex> A) Linux: ROHM PMIC core is in, VC6 6901 is in clk-next
11:11 < Marex> A) U-Boot
11:11 < Marex> - Import DTS from Linux 4.12
11:11 < Marex> - clock: Implement driver for R8A7795/R8A7796
11:11 < Marex> - Serial-SH: - switch to use clock framework
11:11 < Marex>              - probe from DT
11:11 < Marex> - RAVB: - Fix PHY detection
11:11 < Marex>         - Switch to use clock framework
11:11 < Marex>         - Probe from DT
11:11 < dammsan> yep
11:11 < Marex> - SH-SDHI: - Fix ACMD handling
11:11 < Marex>            - Add clock framework support
11:11 < Marex>            - Add support for probing from DT
11:11 < Marex> - Uniphier-SD: - Make it working on RCar Gen3
11:11 < Marex>                  (this driver is much better)
11:11 < Marex> - USB: Zap ehci-rcar-gen3 in favor of ehci-generic
11:12 < Marex> B) U-Boot
11:12 < Marex> - pincontrol driver for Gen3
11:12 < Marex> - rename uniphier-sd to matsushita-sd (?)
11:12 < Marex> - post the patches
11:12 < Marex> C) U-Boot
11:12 < Marex> - Lots of work to be done still :-)
11:12 < Marex> I also have questions, ie. D)
11:12 < Marex> D) U-Boot
11:12 < Marex> - Can I post the patches ?
11:12 < Marex> - What about RPC driver, can I start working on it ?
11:12 < Marex>   (What about Linux RPC driver?)
11:12 < Marex> - EHCI/xHCI DT nodes for R8A7796 for Linux ?
11:12 < Marex> - Where is the Salvator-XS stuck, it didn't arrive yet
11:12 < Marex> -- EOF --
11:12 < pinchartl> dammsan: ok, thanks. yes, that's an issue that we have to solve. I'm not sure what the best option is
11:12 < pinchartl> Marex: what's the RPC driver ?
11:13 < Marex> pinchartl: QSPI/Hyperflash driver , it would be useful for me with patched BL2 to reprogram U-Boot from U-Boot
11:13 < Marex> pinchartl: there's even some example code for RZ/A1, which I suspect has the same or very similar block
11:13 < pinchartl> ok, thanks
11:13 < Marex> (and -- putting on my MTD co-maint hat on -- I'd be interested to learn about Hyperflash and how we can add that to Linux =) )
11:14 < geertu> Marex: Isn't it just QSPI / paired QSPI?
11:14 < Marex> geertu: I think the command set is different and the bus is a bit different too, but I didn't look too deeply
11:14 < Marex> geertu: we had a brief discussion about it on #mtd, but it didn't get far since there's no hardware in the wild
11:14 < Marex> (unless you bake a core into an FPGA of course :-) )
11:14 < geertu> Marex: until now?
11:15 < Marex> geertu: kindof, the RPC is locked in the BL2 by default, so you cannot access the regs, you have to patch the BL2 mapping tables to allow access to those regs
11:15  * Marex has a patch of course and runs a patched BL2 :-)
11:16 < geertu> Shimoda-san would be greatful if you added EHCI/xHCI DT nodes for R8A7796
11:16 < geertu> I believe the clocks are already there, courtesy Kaneko-san
11:16 < Marex> geertu: it works for me in U-Boot , so I probably should just test it in Linux too and submit the patch ; it's almost a copy of the R8A7795 with a bit of crosschecking the datasheet
11:16 < geertu> clocks -> in the driver
11:17 < geertu> The binding update needs to be resent, though (it dates from summer 2016?)
11:17 < Marex> geertu: the 7795 seems to have 4 EHCI, while the 7796 has two, I don't think on the EHCI side, there's any difference besides that
11:17 < Marex> geertu: jupp
11:17 < Marex> the HSUSB block is the same too, so should be easy
11:17 < geertu> dammsan: What about "Can I post the patches ?"
11:18 < dammsan> go ahead
11:18 < Marex> dammsan: thanks !
11:18 < dammsan> thx
11:18 < geertu> Marex: The XS you no longer need, as the VC6 support for XS is already in Simon's tree ;-)
11:19 < Marex> dammsan: the stack I sent you is a bit outdated, but I think it's OK if I have a bit more free reign about sending the R8A779x patches ?
11:19 < Marex> dammsan: I added some more patches on the SD and USB side , no xhci yet, just driver improvements
11:20 < geertu> Marex: BTW, is there a CNTVOFF register to initialize, like on CA7/CA15?
11:20 < Marex> geertu: where ?
11:20 < Marex> geertu: re XS, doesn't that have the R8A7795 on it ? :)
11:20 < Marex> dammsan: re XS, I should notify Anders that I no longer expect it, yes ?
11:22 < dammsan> pls email anders
11:22 < Marex> dammsan: ACK
11:22 < geertu> Thank you, Marek
11:22 < geertu> Next is Ulrich
11:23 < uli___> no updates from me
11:23 < geertu> Thank you, Ulrich
11:24 < geertu> As Shimoda-san is not here, I'll copy his updates
11:24 < geertu> A)
11:24 < geertu>    - Make USB2.0 clock selector driver as a CCF driver.
11:24 < geertu>      But I have to update it for some feedbacks.
11:24 < geertu>    - About R-Car D3's CPG things, unfortunately I don't get any information
11:24 < geertu>      from BSP team yet.
11:24 < geertu>      (BSP team intends to ask HW team but BSP team seems busy for now?)
11:24 < geertu> B)
11:24 < geertu>     - Continue to improve USB2.0 clock selector driver.
11:24 < geertu>     - Need reply about R-Car D3's CPG things to Geert-san.
11:24 < geertu>     - Add USB3.0 clock to r8a7796-cpg-mssr.c for v4.14.
11:24 < geertu>     - Add hsusb, [oex]hci[01], usb2_phy1 nodes on r8a7796.dtsi instead of each
11:24 < geertu>       "placeholder" for v4.14.
11:24 < geertu> C)
11:24 < geertu>     - If someone can handle adding usb3.0 clock and usb related device nodes
11:24 < geertu>       for r8a7796, I'm very happy :)
11:24 < geertu> ---EOT---
11:24 < geertu> Topic 2. Discussion Topics
11:25 < geertu> Do we have any?
11:25 < Marex> dammsan: email is out
11:26 < Marex> geertu: RPC, can I start working on it ?
11:26 < neg> geertu: Noting from me
11:26 < morimoto> useful defconfig for Gen3 board, somehow
11:27 < Marex> morimoto: make savedefconfig is flaky again ?
11:27 < geertu> This is a recurring topic.
11:27  * Marex had problems with that before, sigh
11:27 < morimoto> not a big deal though
11:28 < geertu> Personally, I think Simon should have arch/arm64/configs/renesas_defconfig in his devel branch, not for merge upstream
11:28 < geertu> Then the individual developers can send an update when they enable hardware support
11:28 < geertu> update = for renesas_defconfig and plain (arm64) defconfig
11:28 < morimoto> it sounds good to me
11:28 < neg> I like that idea
11:32 < geertu> Marex: About RPC, that's not my call, although I like it.
11:32 < Marex> geertu: I like it too :-)
11:33 < Marex> kbingham: it's 7.30 hrs from PRG to San Sebastian for me, heh
11:33 < geertu> I think we're done with core?
11:34 < pinchartl> when does everybody plan to arrive in San Sebastian ? I was thinking about Friday the 1st, but I'm also contemplating Thursday the 31st
11:35 < neg> I was looking at flights for the 1st but I'm flexible
11:36 < geertu> Monday, if possible
11:36 < morimoto> Japanese 3 member are planning it now.
11:36 < morimoto> We are thinking that we will arrive at 2nd or 3rd
11:36 < neg> also should we try to book a euro-periperi airbnb house? If so how many beds do we need?
11:36 < morimoto> but should we at 1st ?
11:37 < pinchartl> morimoto: the meetings will start on the 4th. it depends how long you need to get adjusted to jetlag ;-)
11:37 < morimoto> Hehe :)
11:37 < pinchartl> neg: I would like to
11:38 < morimoto> If we can have extra meeting before 4th, it can be good reason to escape from office for us :)
11:38 < Marex> heh
11:39 < pinchartl> I'm sure we could organize a meeting before the 4th if needed :-)
11:40 < neg> pinchartl: so that is you, kbingham, uli___, geertu, jmondi, wsa, Marex, simon  and me => 9 ppl or am I forgeting someone?
11:41  * Marex is invited to the megaparty ? wow, I'm flattered
11:41 < pinchartl> I think that's it. however, some of us might prefer staying at a hotel, you would have to ask
11:41 < geertu> I don't mind a house
11:41 < geertu> I don't mind a hotel
11:41 < neg> ofc just looking for the upper limit to be able to browse the options
11:42 < pinchartl> geertu: there might be no pig this time
11:43 < neg> :-(
11:44 < geertu> pinchartl: As long as there's a pig in the main house, Morimoto-san will be happy
11:44 < pinchartl> neg: when do you plan to fly back ?
11:45 < morimoto> no pig this time :)
11:45 < neg> pinchartl: flexible, but I intend to stay atleast to the 9th to be able to join all meetings
11:46 < pinchartl> neg: I need to be in Belgium for the 10th, so I think I'll fly back on the 9th
11:46 < pinchartl> 1st to 9th then
11:46 < neg> I tentativly booked 4-10 in my calendar but as it now looks like ppl will show up the 1st I might as well make it 1-10
11:47 < neg> 1-9 also works, should we set that as a target when looking for airbnb options?
11:48 < pinchartl> sounds good to me
11:48 < pinchartl> Kieran will likely leave on the 8th
11:48 < geertu> pinchartl: me too
11:49 < dammsan> me too
11:49 < neg> So since ppl intend to show up erlier should we aim to move the IO meeting from the 8th to some other day?
11:49 < pinchartl> geertu: and you will arrive on Monday the 4th, right ?
11:50 < pinchartl> neg: I'd rather not move the social day to Friday though
11:50 < dammsan> same here
11:50 < geertu> pinchartl: Correct (still to book flights)
11:50 < Marex> geertu: damned iberia is oneworld :/
11:51 < dammsan> can we host i/o meeting in other airbnb place?
11:51 < pinchartl> dammsan: I think we should be able to
11:51 < neg> If we managed to find one I'm sure that is possible
11:52 < dammsan> my guess is that 9 ppl needs two more airbnb probably
11:52 < dammsan> unless you want to share a single shower
11:53 < dammsan> the map browsing feature of airbnb is pretty good btw
11:53 < dammsan> i recommend to check cancellation policy, the strict one is pretty strict imo
11:53 < neg> a qucik search suggest there are ~2 options near by that have >= 9 beds but yes there might be other constrains such as showers (but the sea is so close...)
11:54 < pinchartl> neg: maybe we should aim for 2 places
11:54 < dammsan> usually number of baths are kind of limited. also many double beds
11:54 < dammsan> think 12ppl place turns into 4ppl
11:55  * neg never used airbnb so good point about checking the cancellation policy
11:55 < dammsan> if you are tight on space then there most likely is an unused twin bed in my room
11:56 < dammsan> but we are a tad low on bath options so i prefer not to push it more than necessary
11:58 < neg> cheking the old mail thread it looks like everyone is attending but Marex who provided no prefrences on dates and Simon who did not prefere the dates but not sure if he made it work or not since, anyone know?
11:59 < dammsan> i think simon is fine
11:59 < geertu> Yeah, I talked to him last week, and he is planning to attend
12:04 < dammsan> Marex: how difficult is it to work on the RPC? can you do it w/o further docs?
12:08 < Marex> dammsan: I'd have to check that, but probably yes
12:09 < Marex> dammsan: I would need a few hours to familiarize myself with the hyperflash part
12:09 < Marex> dammsan: the RPC itself is documented and there is example code too
12:09 < dammsan> i think you should go ahead and scratch that itch
12:09 < Marex> dammsan: OK, thank you
12:09 < dammsan> worst case if renesas prefers not to use it then we can keep it disabled
12:10 < Marex> dammsan: right
12:17 < dammsan> pinchartl: i booked the airbnb meeting place from the 2nd. was thinking of exploring bilbao before that.
12:18 < dammsan> but if more people arrive in san sebastian earlier then i'm happy to be there instead
12:20 < pinchartl> dammsan: I'll fly to San Sebastian, but I could then move to Bilbao or somewhere else
12:20 < dammsan> its just around the corner
12:21 < dammsan> Bilbao is supposed to have a bit more of museums if i understand it correctly
12:22 < dammsan> according to my guide books busses to san sebastian tend to go via bilbao by default, but it may vary with airport and timing
12:23 < pinchartl> the question is, where are the good restaurants ? :-)
12:23 < dammsan> i got a michelin guide book but it is in spanish =)
12:24 < dammsan> still easier than japanese
12:24 < geertu> spanish is a (far) descendant of greek?
12:25 < dammsan> according to "lonely planet pocket bilbao and san sebastian" you should book months ahead
12:26 < pinchartl> indeed, there's a risk of that
12:26 < dammsan> geertu: not sure, but many words seem easy to understand to me
12:26 < pinchartl> there's clearly no way we'll fit the whole crowd around the same table anyway
12:26  * pinchartl is off for lunch, I'll be back later
12:26 < dammsan> probably so
12:28 < dammsan> we should be able to eat a lot of pintxos at least!
12:28 < geertu> dammsan: french, spanish, italian, latin, ... all the same ;-)
12:28 < dammsan> txoko seems interesting too
12:28 < dammsan> indeed =)
12:34  * geertu lunchtime
12:50  * morimoto will go back to lovely home
12:50 < morimoto> I hope someone can put this chat log to Wiki page somehow