09:29 < geertu> Welcome to today's Core Group Chat
09:30 < geertu> Agenda:
09:30 < geertu> 1. Status Updates
09:30 < geertu> 2. Discussion Topics
09:30 < geertu> Topic 1. Status updates
09:30 < geertu> A) What have we done since last time:
09:30 < geertu> Kieran is attending Linaro Connect in Bangkok.
09:30 < geertu> Marek worked on ATF (D3 Draak and V3M Eagle upstreaming, and BSP v2.0.3
09:30 < geertu> upporting), U-Boot (E2 Alt and M3N ULCB), Linux (64bit MSI and PCIe
09:30 < geertu> cleanup), and OpenOCD (SH QSPI prototype for automated testing).
09:30 < geertu> Morimoto-san fixed sound clock issus in v5.1-rcX, fixed R-Car datasheet
09:30 < geertu> disclosure, welcomed Laurent and Marek to Hanami, and did Renesas work.
09:30 < geertu> Shimoda-san supported the BSP in CPG driver review.
09:30 < geertu> Simon enabled S2RAM on Ebisu and reposted E3 Z/Z2 clock patches.
09:30 < geertu> Ulrich reviewed pcie-rcar patches, posted I2C PFC patchs, and performend
09:30 < geertu> APE6EVM boot loader archeology.
09:30 < geertu> Geert assisted Jiada with the clk/genpd deadlock fix, posted PFC
09:30 < geertu> validation and compile-testing v3 and IPMMU suspend/resume v2, enabled
09:30 < geertu> CFI NOR FLASH on APE6EVM, upported clock and pin control errata fixes,
09:30 < geertu> completed the peri{upport,pelist}=>periject conversion for Core, and
09:30 < geertu> programmed all CP2102 USB-to-UART bridges to have unique names.
09:31 < geertu> (I hope I didn't miss anything when merging two sets of email reports)
09:32 < geertu> B) What we plan to do till next time:
09:32 < geertu> Kieran will finish Linaro Connect, and has booked holidays from 15th to
09:32 < geertu> 19th April.
09:32 < geertu> Marek plans to test and submit the BSP v2.0.3 ATF upport, to continue on
09:32 < geertu> the SH QSPI driver.
09:32 < geertu> Geert will send clock and pin control pull requests for v5.2, and will
09:32 < geertu> escape on holidays afterwards (until April 22th).
09:33 < geertu> C) Problems we have currently:
09:33 < geertu> Simon abandoned work on Gen3 ZG clocks as this is too complex to
09:33 < geertu> consider upstreaming without an upstream driver to exercise the code.
09:33 < geertu> Geert reports that the APE6EVM test application does not work when
09:33 < geertu> uploaded by OpenOCD to 0xe8200000.
09:33 < Marex> geertu: update: SH QSPI driver is in reasonable shape already , I can update U-Boot with it on E2 Alt at least ;-)
09:33 < Marex> geertu: from OpenOCD that is
09:34 < geertu> Marex: BTW, what's "SH QSPI"?
09:34 < Marex> geertu: the QSPI controller in Gen2 SoCs , used to talk to the boot SPI NOR
09:34 < Marex> geertu: it's likely in older SoCs too
09:34 < geertu> OK, so RSPI/QSPI
09:35 < geertu> Nice!
09:35 < geertu> Anything I missed?
09:37 < geertu> Topic 2. Discussion Topics
09:37 < geertu> Anything to discuss after 4 weeks of no chat meetings?
09:37 < wsa> general upporting status:
09:38 < wsa> looks like core is also mostly done with bsp392?
09:38 < geertu> Yes, there's not much left
09:39 < geertu> Most leftovers are due to missing hardware access (new ES dependencies), missing upstream drivers (3DG), or unclear policies (OPTEE)
09:41 < Marex> geertu: what's unclear about optee ?
09:41 < geertu> Marex: What does it do?
09:43 < damm> slices and dices?
09:43 < Marex> geertu: optee OS does secure tasks
09:43 < Marex> geertu: the linux side only communicates with the optee OS via SVCs
09:44 < Marex> geertu: by secure tasks, it looks like e.g. RPC HF update in a controlled manner
09:44 < Marex> geertu: you can use it to rewrite the HF even if it's locked in ATF, since Optee OS has privileges to do so
09:45 < geertu> Marex: Aha? But that needs to be enabled in firmware, too?
09:45 < geertu> And how does access control work in that case?
09:46 < Marex> geertu: "that" ?
09:46 < Marex> geertu: Linux is running in non-secure world and optee OS is the access controller
09:47 < Marex> that's my understanding of the bare bones situation
09:48 < geertu> Marex: Can it distinguish between my Linux or your Linux?
09:48 < Marex> geertu: there is some setting where optee os boots the next stage (u-boot), so it can authenticate what it starts
09:49 < Marex> geertu: and it can block your linux from running at all
09:50 < Marex> geertu: whether it can authenticate requests from random stuff in the non-secure world, I didn't check ; presumably yes
09:52 < geertu> ok
09:53 < Marex> geertu: optee OS upporting is on my list, when I get to it, I'll know more
09:53 < geertu> Marex: Great!
09:53 < Marex> geertu: ... is what you think, until you see the code :)
09:54 < Marex> morimoto: off-topic, when shall we meet tomorrow ?
09:55 < morimoto> Marex: 13:00 at Kokubuji ?
09:56 < geertu> Next Core Group Chat date?
09:56 < geertu> I won't be available in two weeks.
09:57 < marex-cloud> morimoto: Roger!
09:58 < pinchartl> geertu: I'm afraid I won't be available in 3 weeks
09:58 < shimoda> geertu: how about 18th Apr?
09:58 < geertu> shimoda: That's in two weeks ;-)
09:58 < morimoto> pinchartl: because Sky :) ?
09:58 < pinchartl> morimoto: I wish. because work :-)
09:59 < morimoto> pinchartl: ohh... orz
09:59 < horms> I will be on a break on the 18th. But I can send in a report via email.
09:59 < geertu> May 2nd?
09:59 < morimoto> geertu: it is GW (= Golden week) at Japan
09:59 < morimoto> GW = long Holiday
10:00 < pinchartl> geertu: or maybe you could appoint a deputy on the 18th ? :-)
10:01 < shimoda> geertu: oops. i missed your comment ;) In Japan side, from 29th Apr and 6th May is a GW. so 9th May?
10:01 < geertu> pinchartl: Are you volunteering? ;-)
10:01 < pinchartl> may the 2nd is in 4 weeks. if we can afford to always wait for four weeks between meetings they we should make them monthly meetings, not bi-weekly meetings
10:01 < kbingham> I'm away on the 18th :)
10:01 < pinchartl> geertu: not at all :-)
10:01 < pinchartl> 17th ?
10:01 < geertu> Marex: U R deputy?
10:02 < geertu> May 9th is an option for me
10:02 < morimoto> pinchartl: April 17th ? or May 17th ?
10:03 < geertu> As long a it's not May 32 (which is a Brexit joke in the newspaper ;-)
10:03 < pinchartl> morimoto: April
10:03 < morimoto> pinchartl: IC
10:04 < damm> pinchartl: need to leave early to prepare for another appointment, will check the report =)
10:04 -!- damm [~damm@s214090.ppp.asahi-net.or.jp] has quit [Quit: leaving]
10:05 < morimoto> 17th April is available if meeting can be finished until JSP 17:30 ;)
10:05 < morimoto> It is Renesas special-go-back-day
10:05 < pinchartl> geertu: jokes aside, there's no issue if you have to skip one of the meetings, you can appoint a deputy
10:05 < pinchartl> or we can move it to another day of the same week
10:05 < geertu> I'm not available before April 22
10:06 < geertu> I won't have anything to report anyway
10:06 < geertu> Without Core, you can probably finish by JST 1730 on April 17th?
10:07 < wsa> how about keeping the 18th and have core reports email only?
10:07 < wsa> I can host core discussions in my IO slot if needed
10:08 < geertu> wsa: thx!
10:09 < wsa> and probaby we should schedule the next meeting after that to may, 9th because of GW?
10:09 < geertu> wsa: OK
10:09 < morimoto> 18th April, and 9th May are OK for me
10:09 < wsa> actually, that would fit my schedule very well. I am busy on May, 2nd as well
10:10 < wsa> pinchartl: ok?
10:11 < pinchartl> works for me
10:12 < wsa> cool
10:12 < pinchartl> can we start with MM ?
10:12 < geertu> Thanks for joining, and have a nice continued day!