From a736d5ac6f0a907191fd029ec1ca347ba0b0578a Mon Sep 17 00:00:00 2001 From: Geert Uytterhoeven Date: Thu, 2 Jul 2020 10:23:25 +0200 Subject: wiki: Add Core chatlog for 2020-07-02 Signed-off-by: Geert Uytterhoeven --- wiki/Chat_log/20200702-core-chatlog | 107 ++++++++++++++++++++++++++++++++++++ 1 file changed, 107 insertions(+) create mode 100644 wiki/Chat_log/20200702-core-chatlog diff --git a/wiki/Chat_log/20200702-core-chatlog b/wiki/Chat_log/20200702-core-chatlog new file mode 100644 index 0000000..ede979a --- /dev/null +++ b/wiki/Chat_log/20200702-core-chatlog @@ -0,0 +1,107 @@ +09:41 < geertu> Welcome to today's Core Group Chat Meeting! +09:41 < geertu> Agenda: +09:41 < geertu> 1. Status Updates +09:41 < geertu> 2. Discussion Topics +09:41 < geertu> Topic 1. Status updates +09:41 < geertu> A) What have we done since last time: +09:41 < geertu> Marked updated the U-Boot dc2114x driver in preparation for DM +09:41 < geertu> conversion, and researched BootROM/IPL A/B copy switching for +09:41 < geertu> reboot-into-recovery-ATF. +09:41 < geertu> Niklas udated his Gen3 boards with upstream firmware. +09:41 < geertu> Shimoda-san submitted patches to set the usb-dmac tx_result +09:41 < geertu> parameters, and add R-Car M3-W+ IPMMU support. +09:41 < geertu> Ulricht sent v3 of the RWDT critical clock series. +09:41 < geertu> Geert did more DT binding doc conversions (rza2-pinctrl), refactored +09:41 < geertu> GPIO Aggregator parsing to use bitmap_parselist(), investigated s2ram +09:41 < geertu> wake-up regression in v5.8-rc1, reviewed RZ/G2 patches, attended ELC-NA, +09:41 < geertu> and researched QEMU gpiodev support. +09:42 < geertu> B) What we plan to do till next time: +09:42 < geertu> Marek plans to continue working on IPL A/B copy switching. +09:42 < geertu> Shimoda-san plans to convert the usb2-clock doc to json-schema. +09:42 < geertu> Geert plans to do more DT binding doc conversions, and continue QEMU +09:42 < geertu> GPIO virtualization. +09:42 < geertu> C) Problems we have currently: +09:42 < geertu> Geert is struggling with describing multi-level sh-pfc subnodes in +09:42 < geertu> json-schema, and untangling the QEMU OO framework. +09:42 < geertu> ---E--- +09:42 < geertu> Anything I missed? +09:42 < kbingham> geertu, DTB location issues ? +09:43 < wsa> what is QEMU OO? +09:43 < wsa> object oriented +09:43 < wsa> ? +09:43 < geertu> yep +09:43 < geertu> kbingham: 0x50000000/0x58000000 works for me +09:43 < geertu> i.e.: +09:43 < geertu> tftpboot 0x50000000 h3-salvator-xs/Image +09:43 < kbingham> geertu, Did you work out /why/ the DTB location is no longer valid? or just find new addresses that work. +09:43 < geertu> tftpboot 0x58000000 h3-salvator-xs/r8a77951-salvator-xs.dtb +09:43 < geertu> booti 0x50000000 - 0x58000000 +09:44 < kbingham> I have a new address that works currently (though not the same as those ones) +09:44 < geertu> kbingham: Haven't looked into the U-Boot sources +09:44 < geertu> I had hoped The Expert(MV^WTM) would chime in +09:44 < wsa> according to neg, 0x50000000 and 0x58000000 is approved by Marek, right? +09:44 < geertu> Niklas confirmed the DTB overwrite happens with latest U-Boot, too +09:44 < kbingham> Ok, I think I've seen in a few bootlogs, that the 'failing' address was quite commonly shared, so it's likely everyone was using the same location. +09:45 < kbingham> So can we take these new addresses as the new 'official' locations? +09:45 < geertu> kbingham: Yeah, I documented that address on the elinux wiki, so everyone uses it ;-) +09:45 < kbingham> geertu, Aha thanks, Sorry I haven't seen the elinux update. +09:45 < wsa> geertu: thanks! I was about to suggest a wiki page :) +09:46 < wsa> shimoda: I have a question about sharing new binary firmware... +09:46 < geertu> Still, would be interesting to know why the DTB is overwritten. Might be an off-by-something factor bug in the copy parameters +09:46 < wsa> you said that you are not allowed to send the binaries but we should build them ourselves with yocto +09:46 < geertu> kbingham: Haven't updated the wiki for the new addresses yetr +09:46 < Marex> geertu: isnt the kernel using the 2 MiB below it for something ? +09:46 < kbingham> geertu, Yes, and still very odd that it started happening due to some difference in the kernel update... +09:46 < Marex> geertu: stack for decompressor maybe ? +09:47 < kbingham> geertu, OH ... I see... sorry - the OLD address is documented at the wiki ;-) +09:47 < geertu> Marex: arm64 doesn't decompress, AFAIK +09:47 < wsa> shimoda: can we put the "ourselves built" firmware into some internal repository? do we still have osdg? +09:48 < shimoda> wsa: yes. this is my side problem. pre-built binaries are stored renesas internal storage server. But, since i'm working from home, I could not access the server. +09:49 < wsa> shimoda: ah, so it is a technical problem and not a legal problem? +09:49 < shimoda> wsa: we can use osdg server until end of this year... +09:49 < shimoda> wsa: yes, it's technical problem. +09:49 < wsa> geertu: the init section exploded with 5.8-rc1, around 1.5 meg larger +09:50 < wsa> shimoda: ok, this makes it a bit easier then :) +09:50 < morimoto> shimoda: do you mean Redmine ? +09:51 < shimoda> morimoto: both osdr Redmine and osdg Gitlab (will be closed) +09:51 < wsa> so, we need to move the periject repo +09:51 < morimoto> shimoda: if so, Redmine part is now under git +09:52 < shimoda> morimoto: yes, under git of osdg :) +09:52 < shimoda> so, we need to move the osdg to somewhere +09:52 < morimoto> I don't remember, I created private project under Github ? +09:52 < morimoto> OK, thanks +09:53 < Marex> geertu: then maybe its the bss section as wsa mentioned ? or stack ... +09:54 < geertu> Marex: does "booti" have subcommands like "bootm"? The former doesn't seem to be documented in the U-Boot documentation on denx.de +09:54 < Marex> geertu: subcommands for what ? +09:54 < wsa> shimoda, morimoto: do you work from home all the time now? +09:54 < geertu> Marex: booti subcommands +09:55 < geertu> Marex: You have bootm start / loados / ... +09:55 < morimoto> wsa: yes +09:55 < Marex> geertu: iirc booti calls the same code as bootm, except for a bit of aarch64 setup, so that should work +09:55 < geertu> Marex: Have you read my email with my overwriting findings ("[periperi] ERROR: Did not find a cmdline Flattened Device Tree")? +09:56 < shimoda> wsa: yes. also, renesas employees can choose office or home from this Aug. +09:56 < geertu> ok, will investigate again +09:56 < geertu> to find out in which part it is overwritten +09:56 < wsa> shimoda: ah, from August +09:57 < morimoto> wsa: because of that, I need to work by small windows note PC... +09:57 < Marex> geertu: OK +09:57 < geertu> Marex: Note that an arm64 Image does not include bss +09:57 < wsa> morimoto: windows?? +09:57 < geertu> The error is from U-Boot, i.e. before it attempts to start the kernel +09:57 < morimoto> wsa: yes, modern windows10 :) +09:58 < wsa> I heard git and windows is a troublesome combination +09:58 < damm> i've been asked to use microsoft teams too +09:58 < damm> for conferencing purpose +09:58 < damm> very exciting +09:59 < morimoto> wsa: I'm using WSL(= windows subsystem Linux) and/or SSH to Renesas machine +09:59 < wsa> damm: i had to install this here, too. For remote schooling the kids +09:59 < neg> morimoto: small windows, small problems? +09:59 < wsa> the linux client worked surprisingly well +09:59 < shimoda> wsa: same here (use windows 10). but i also use a server which installed Linux at the office via ssh. +09:59 < morimoto> neg: small windows, big problems which I can't kill :) +09:59 < wsa> poor JapaPERI! +10:00 < geertu> wsa: Yep, Teams and Office 365 in the browser for school +10:00 < wsa> they turned to jitsi later +10:00 < geertu> Anything else to discuss? +10:00 < morimoto> wsa: yes we are (ToT) +10:01 < geertu> Thanks for joining, and have a nice continued day! -- cgit v1.2.3