04:47 <steev> bamse: https://github.com/steev/linux/commits/linux-v5.19-rc3-tests is my 5.19 with the c630 stuff and your flex 5g stuff and "a few" other misc patches - haven't pulled in the ones qzed asked me to test yet, but i wanted to push it so i can build my cd here

04:48 <steev> https://www.gsmarena.com/the_xiaomibook_s_124_is_a_snapdragon_8cx_gen_2_powered_windows_2in1_convertible-news-54751.php

04:53 <steev> Added bonus is that it includes the patch that you asked me to submit a long time ago :/ I’ll try to get to that this week

04:56 <bamse> steev: little bit odd that they are shipping a gen2 soon...but looks like a nice device

04:57 <bamse> hmm, seems i can buy it on the german site...althoguh my german is a little bit rusty

05:00 <bamse> steev: invited mani_s here as well, he's been poking at the debian installer to get it up to snuff for the x13s

05:01 <bamse> and now it's tomorrow already, time to call it a yesterday...

05:02 <steev> yarp, and cool, i look forward to him showing up )

05:03 <mani_s> steev, I just got the laptop last week and been trying to get the installer work with devicetree (not using acpi tables)

05:03 <steev> oh, interesting

05:04 <steev> i think part of not using devicetree was so that it wasn't device specific, using acpi tables allows us to use the 1 installer across all the devices (in theory)

05:05 <mani_s> steev, right but we still need to use devicetree anyway for the debian kernel. So I thought why not for the installer also

05:05 <mani_s> plus bamse and me are thinking that it will allow us to have more functionalities compared to acpi

05:06 <steev> if the installer were to use grub, it would be possible, while debian doesn't have the patch that ubuntu has, afaik, grub does support the devicetree directive, so you could modify the entries to point to each different device tree - but i don't believe it does

05:06 <steev> it does use grub, i mean

05:06 <mani_s> steev, I tried the ubuntu installer and yes, it has the grub support to boot using devicetree

05:07 <steev> right, but they use... ubiquity? i think? xnox can correct me if i'm wrong

05:08 <steev> debian's grub supports device tree too, it just doesn't have the patch afaik yet

05:08 <mani_s> yeah they were

05:08 <steev> the patch that adds that entry

05:08 <mani_s> hmm. Now I'm building the debian-installer. Maybe we need to add that patch in aarch64-laptops

05:08 <mani_s> for sure it is a nice value add

05:09 <steev> https://git.launchpad.net/~ubuntu-core-dev/grub/+git/ubuntu/commit/?id=fdb8c44103af4ed21e9b7d3b2f55360932b84834

05:09 <steev> that's a part of it

05:11 <mani_s> that;s just debug output, isn't it?

05:11 <steev> right - that just, afaik, makes it spit out - i'm trying to find the patch, but it's definitely older

05:14 <mani_s> https://git.launchpad.net/~ubuntu-core-dev/grub/+git/ubuntu/tree/debian/patches/ubuntu-add-devicetree-command-support.patch?id=fdb8c44103af4ed21e9b7d3b2f55360932b84834

05:14 <steev> let me clone it, i really hate the launchpad gitweb ui

05:14 <mani_s> maybe this?

05:14 <steev> ah yep that's the one, xnox is the author)

05:15 <mani_s> cool!

05:15 <mani_s> yeah, this needs to be included

05:17 <steev> i don't recall what upstream's reasoning is for denying it

05:17 <steev> i *believe* it was sent upstream though

05:18 <mani_s> hmm. I also don't see any obvious reason to deny

05:18 <mani_s> it's really convenient to pass devicetree

05:20 <steev> perhaps they didn't want to have a list of... every possible device tree option

05:20 <steev> i don't know, but yeah, that patch and i think there are a couple others (e.g. the one that forbids it if secureboot is enabled)

05:21 <mani_s> steev, thanks for pointing them out. I'll try to include them

05:23 <steev> i know shawnguo also tried to get some of the patches applied upstream

05:23 <steev> debian upstream, i mean

05:23 <steev> i haven't checked in on the status of those recently though

05:24 <mani_s> steev, ndec just shared this: https://www.gnu.org/software/grub/manual/grub/html_node/devicetree.html

05:24 <mani_s> looks like time to rebase ;)

05:25 <steev> right, it's supported, but afaik, there is no way to add the option like that patch does automatically

05:25 <steev> xnox's patch, i mean

05:25 <mani_s> hmm

05:32 <steev> like, you can just edit the grub config at boot time and add it, and it'll use it (though we use dtbloader to use the chid and then we have the python3 script to install the dtbs to /boot/efi/dtbs or whatever it is (and i wrote a hook but i don't believe it's in the repo)

05:32 <steev> it's also not a great hook because it copies them every time you do anything with kernels

05:33 <steev> https://paste.debian.net/1244816/

05:35 <mani_s> steev, yeah, that's for not hardcoding the devicetree entry in grub I believe

05:36 <mani_s> eventually that logic should also be part of the installer

05:36 <steev> there is/was another reason, but i can't recall - i blame robclark, since he wrote the dtbloader thingie

05:37 <steev> but maybe it was just not hardcoding

05:39 <mani_s> well leif wrote it and robclark improved it ;)

05:40 <steev> all above my paygrade, i could never figure that edk2 stuff out at all

05:42 <mani_s> :)

05:42 <steev> i'm just smart enough to be dangerous

19:51 <steev> alright, got the mail, the x13s should be on its way/here soon

20:33 <bamse> nice!

23:21 <Dylanger> <hexdump0815> "Dylanger: thx for the config" <- Is there anything I can try to get jump start this?

23:40 <qzed> is `cpu cpu4: Voltage update failed freq=3072000` a known issue for the sc8180x?

23:40 <qzed> also there's one CPU core that seems to have a constant high usage

23:42 <hail_eris> Dylanger: which kernel did you use for the dts? 5.15 doesn't have tomato, it's in the 5.10 branch

23:44 <Dylanger> Currently trying with 5.18

23:44 <Dylanger> Yeah I'm currently trying to add cherry + tomato DTS myself

23:45 <Dylanger> Into mainline, oh wait, tomato is already in 5.10 mainline?

23:45 <hail_eris> I mean it's in Chromium's fork

23:52 <hail_eris> Try with the latest kernel (5.19-rc3), i'm able to at least build the dtbs by copying over chromium 5.10's arch/arm64/boot/dts/mediatek if I copy over mt6359.dtsi mt8192* mt8195*, then modify Makefile to include only those files. I doubt it'll work without a lot of hacking around, though. May be more fruitful to first boot chromium's fork before trying.