summaryrefslogtreecommitdiff
path: root/wiki/Chat_log
diff options
context:
space:
mode:
authorGeert Uytterhoeven <geert+renesas@glider.be>2021-05-20 10:16:24 +0200
committerGeert Uytterhoeven <geert+renesas@glider.be>2021-05-20 10:17:03 +0200
commitf226ba27b80fb13bcc2b28211a64edd79376db34 (patch)
treeef64960cb9c400f3d7eebbb67b16acff57df5aff /wiki/Chat_log
parentdf192330df2881250671d0fec07d0a8295dec2c0 (diff)
wiki: Add Core chatlog for 2021-05-20
Signed-off-by: Geert Uytterhoeven <geert+renesas@glider.be>
Diffstat (limited to 'wiki/Chat_log')
-rw-r--r--wiki/Chat_log/20210520-core-chatlog129
1 files changed, 129 insertions, 0 deletions
diff --git a/wiki/Chat_log/20210520-core-chatlog b/wiki/Chat_log/20210520-core-chatlog
new file mode 100644
index 0000000..681de80
--- /dev/null
+++ b/wiki/Chat_log/20210520-core-chatlog
@@ -0,0 +1,129 @@
+09:33 < geertu> Welcome to today's Core Group Chat Meeting!
+09:33 < geertu> Agenda:
+09:33 < geertu> 1. Status Updates
+09:33 < geertu> 2. Discussion Topics
+09:33 < geertu> Topic 1. Status updates
+09:33 < geertu> A) What have we done since last time:
+09:33 < geertu> Marek updated ATF patches, fixed MISRA C issues in ATF, and upported
+09:33 < geertu> R-Car V3U U-Boot support.
+09:33 < geertu> Shimoda-san reported about OSAL, asked the hardware team about R-Car D3
+09:33 < geertu> PFC documentation bugs, got information about R-Car Gen3e, and inspected
+09:33 < geertu> the R-Car V4H spec.
+09:33 < geertu> Geert did periject bsp-41x triage, upported BSP opp-table patches,
+09:33 < geertu> converted remaning core DT bindings to json-schema, added pin bias
+09:33 < geertu> support for R-Car SoCs that still lacked them, and added missing GPIO
+09:33 < geertu> keys pull-ups for R-Car Gen2 boards.
+09:34 < geertu> B) What we plan to do till next time:
+09:34 < geertu> Marek plans to work on OpenOCD support and U-Boot upstreaming for R-Car
+09:34 < geertu> V3U.
+09:34 < geertu> Shimoda-san plans to follow up R-Car V3U IPMMU work and R-Car D3 PFC
+09:34 < geertu> issues, and continue to inspect the R-Car V4H spec.
+09:34 < geertu> Geert plans to review the RZ/G2L patches, and post more pinctrl
+09:34 < geertu> validation code.
+09:34 < geertu> C) Problems we have currently:
+09:34 < geertu> Geert wants to kill "renesas,shdma-mux", as (un)used on R-Mobile APE6.
+09:34 < geertu> ---EOT---
+09:34 < geertu> Anything I missed?
+09:36 < geertu> If anyone with hardware access could test the GPIO key pull-ups, that would be great
+09:37 < wsa> geertu: just testing if they work?
+09:37 < geertu> wsa: yes, regression testing, and optionally measuring the pin voltage with bis enabled and disabled
+09:38 < geertu> s/bis/bias/
+09:38 < geertu> Topic 2. Discussion Topics
+09:39 < geertu> Are there plans to make R-Car Gen3 available to us, e.g. with remote access?
+09:39 < wsa> I can surely test the keys on Lager. I'll try to check the bias for H2/E2, too...
+09:39 < geertu> wsa: thx!
+09:40 < wsa> "Gen3e" probably?
+09:40 < geertu> oops, yes
+09:43 < wsa> shimoday: are there already plans for "Gen3e" availability
+09:43 < wsa> ?
+09:44 -!- damm [~damm@68.183.84.28] has quit [Ping timeout: 245 seconds]
+09:44 -!- damm [~damm@68.183.84.28] has joined #periperi
+09:44 < shimoday> wsa: i'd like to support Gen3e in this year somehow
+09:45 < wsa> geertu: I sent a patch to periperi moving optee-patches to non-target. are you okay with it?
+09:45 < shimoday> we can use "Gen3" boards as "Gen3e" even if 2G mode
+09:45 < wsa> i think optee is core, so i didn't want to push it right away
+09:45 < geertu> wsa: I'm happy to get rid of any optee patches ;-)
+09:45 < shimoday> because the SoCs are completely the same
+09:45 < geertu> marex: Are the BSP optee patches any good?
+09:46 < geertu> shimoday: sure, but one day we'll have to create DTS files for them, right?
+09:46 < marex> geertu: which ones ?
+09:47 < wsa> I found a message saying that the drives quite diverged, and it seems the patches are fixes to local code...
+09:47 < wsa> Quoting bsp392_tee_driver.yaml:
+09:47 < wsa> "It's difficult to refactor this driver for upstream. So, mark Abandoned"
+09:47 < shimoday> geertu: yes, we have to create DTS files somehow
+09:47 < geertu> shimoday: For Gen3e I agree, but won't we overclock our non-2G parts?
+09:47 < marex> geertu: upstream optee works without extra patches
+09:47 < wsa> marex: check "[PATCH periject] linux: bsp-41x: non-target: move optee changes" from periperi-ml
+09:49 < shimoday> geertu: i heard almost all Gen3 chips can overclock to 2G. If we are unlucky, the SoC just cannot work correctly without broken, I heard
+09:49 < geertu> shimoday: Ok (so the fans do serve a good purpose ;-)
+09:50 -!- damm [~damm@68.183.84.28] has quit [Ping timeout: 252 seconds]
+09:51 -!- damm [~damm@68.183.84.28] has joined #periperi
+09:51 < shimoday> geertu: yes :)
+09:51 < wsa> time to remount the fans then :)
+09:53 < marex> wsa: seems like a whole lot of it is related to the SMC "update" call, i.e. renesas has custom SMC service where you put in bootloader update and OpTee has enough permissions to write it into RPC HF
+09:53 < marex> wsa: that's what my initial look tells me anyway
+09:54 < geertu> marex: thx for checking! Hence non-target it'll be?
+09:54 < marex> geertu: non-target ?
+09:55 < wsa> non-target == no upporting
+09:55 < marex> wsa: for this custom SMC call, no, no upporting
+09:56 < wsa> good, thanks for looking
+09:56 < wsa> into it
+09:57 < marex> sure ...
+09:57 < geertu> Anything else to discuss?
+09:57 < marex> which however brings me to another question
+09:57 < marex> yes
+09:57 < marex> ATF upstream is starting to turn increasingly ... uh ... not great
+09:58 < marex> so I was thinking, it might make sense to stick to the U-Boot PSCI implementation for upstream V3U port
+09:58 < marex> (i.e. avoid ATF there altogether)
+09:58 < marex> I hope that would be OK with you ?
+09:59 < geertu> marex: Does it support SMP, suspend, ...?
+09:59 < marex> geertu: whatever is missing should be added ; I still don't have the ATF sources for V3U , so I cannot tell
+10:00 < marex> geertu: but the idea would be to have minimal PSCI support to toggle CPU cores on/off as mandated by the arm64 spec, and not any extra stuff, let Linux do its job
+10:01 < geertu> marex: fine for me, if upstream arm64 won't complain
+10:01 < marex> geertu: it won't, since it is compliant to the spec
+10:01 < marex> geertu: also, they didn't complain when intel used it on their aarch64, so why should they complain now
+10:02 < geertu> ok
+10:03 < geertu> Anything else to discuss?
+10:03 < geertu> Perhaps #periperi and freenode?
+10:03 < geertu> Or too early for that?
+10:03 < wsa> next meeting on June, 17th?
+10:03 < wsa> is something wrong with Freenode?
+10:04 < pinchartl> geertu: I think it's not too early to discuss a possible move, but too early for a conclusion. we need the dust to settle first, there's too much ongoing drama
+10:04 < geertu> https://lwn.net/Articles/856543/
+10:05 < pinchartl> what would be the options ?
+10:06 < geertu> perichat ;-)
+10:06 < geertu> From the knights of NIH...
+10:06 < pinchartl> - another public IRC server (libera.chat ?) - this would likely suffer from similar privacy concerns
+10:06 < kbingham> NIH
+10:06 < pinchartl> - a private IRC server ?
+10:06 < pinchartl> - matrix ?
+10:06 < damm> i can setup a local IRC server in a VM if needed
+10:07 < damm> or we can put it in the cloud to improve latency
+10:07 < neg> In case it's needed I registered #periperi at libera.chat
+10:08 < marex> neg: it is real nice
+10:09 < geertu> OK, let's wait and see
+10:09 < neg> damm: I love the zero maintenance of the could :-)
+10:09 < geertu> June 17th is fine for me
+10:09 < geertu> neg: cloud?
+10:09 < shimoday> renesas has MS Teams so if we all have Microsoft account I could invite you :) This is a voice meeting though.
+10:10 < geertu> shimoday: can;t you access Teams in a web browser, without account?
+10:10 < pinchartl> noooooooo not teams :-)
+10:10 < shimoday> geertu: I'm not sure about the detail. perhaps, Magnus-san knows
+10:11 < shimoday> pinchartl: :)
+10:11 < pinchartl> would everybody generally prefer to keep using IRC, or move to matrix ?
+10:11 < geertu> pinchartl: Google Hangouts? ;-)
+10:11 < pinchartl> (we could also consider a voice discussion instead of IRC discussion)
+10:12 < wsa> wen can meet in one of the C64 BBSes available via telnet ;)
+10:12 < damm> maybe geert has something compatible with 68k as well? =)
+10:12 < wsa> seriously, I agree with Laurent, let's wait a little until the dust settles
+10:13 < neg> Why not meet in person each month in Tokyo?
+10:13 < wsa> I'd personally like to stay with IRC, either Freenode or LibraChat
+10:13 < geertu> OK, let's wait and see
+10:13 < marex> wsa: yes
+10:13 < wsa> shimoday: moriperi: damm: 17th of June is good for you?
+10:14 < geertu> neg: plus two weeks in quarantaine before, and after
+10:14 < wsa> pinchartl: ^?
+10:14 < shimoday> wsa: yes, 17th of June is good to me
+10:14 < moriperi> wsa: yes, it is OK for me. thanks
+10:14 < geertu> 17th it'll be
+10:14 < geertu> Thanks for joining, and have a nice continued day!