robclark changed the topic of #aarch64-laptops to: Linux support for AArch64 Laptops (Chrome OS Trogdor Devices - Asus NovaGo TP370QL - HP Envy x2 - Lenovo Mixx 630 - Lenovo Yoga C630 - Lenovo ThinkPad X13s - and various other snapdragon laptops) - https://oftc.irclog.whitequark.org/aarch64-laptops
<steev>
i don't have bootctl, but i also don't see that message here
<steev>
are you sure you have the right dtb booting?
<lun[m]>
You need to run bootctl status to cause the error and it's intermittent, failing either with that error or with another or resetting the laptop immediately
<lun[m]>
dtb is dtbs/qcom/sc8280xp-lenovo-thinkpad-x13s.dtb in your 6.6.0-rc4 branch
matthias_bgg has quit [Ping timeout: 480 seconds]
<steev>
what would the equivalent of that be though, as i don't use systemd-boot, is what i'm saying. i'm assuming just trying to read the efivars?
<steev>
efibootmgr -v sometimes shows `Skipping unreadable variable "Boot0012": Input/output error` but, never causes a reboot
<steev>
and it doesn't cause the -22
<lun[m]>
Ran efibootmgr a few times. Didn't notice errors the first two times, then the laptop reset on the third.
<lun[m]>
* Ran efibootmgr -v a few
<steev>
i don't think it'll change anything, but i just pushed rc5
<steev>
clover[m]: also just pushed 6.5.7, nothing super major different, picked up some minor fixes that shouldn't affect anything, but i DID drop the make it hotter patch
<clover[m]>
OK cool
matalama has joined #aarch64-laptops
matalama has quit [Remote host closed the connection]
iivanov has joined #aarch64-laptops
<qzed>
lun: is there anything in the kernel log when you get that error?
hexdump0815 has joined #aarch64-laptops
<_[m]1>
if secure boot works, you don't need /boot inside your fde, correct?
martiert has quit [Quit: WeeChat 4.0.5]
martiert has joined #aarch64-laptops
<xnox>
steev: ubuntu installs sometimes fail with grub-multi-install failing to read things or write things; and then causing reset. ditto i.e. just doing `mokutil --dbx`
hightower3 has joined #aarch64-laptops
hightower2 has quit [Ping timeout: 480 seconds]
<_[m]1>
<Jasper[m]> "It does work fine on Linux if..." <- did you manage? I have up last time
<jhovold>
Besides providing a status update on where we stand with the x13s support, it may also answer some questions that people have raised about why me, bamse and the linaro folks have spent so much time on this machine in the first place.
<jhovold>
I've also set up a wiki page where I intend to track the status, known issues and workarounds going forward (to complement my posts in this channel):
<jhovold>
Feature wise things look pretty good already with the 6.5 release, and the only thing that a distro wanting to support the X13s needs to provide is really the UCM volume fixes that have not yet been merged.
maz has quit [Remote host closed the connection]
_kbingham has joined #aarch64-laptops
ungeskriptet has quit [Quit: Ping timeout (120 seconds)]
kbingham has quit [Remote host closed the connection]
krzk has quit [Remote host closed the connection]
calebccff has quit [Remote host closed the connection]
calebccff has joined #aarch64-laptops
krzk has joined #aarch64-laptops
_kbingham is now known as kbingham
<steev>
i shall watch it now for my lunch break
<xnox>
jhovold: ucm volume fixes? i can't remember if we did that, but like sound works perfect for me. and on installer boot there is this jingle to let you know that desktop booted, and one can turn on screenreader etc.
<xnox>
jhovold: also fingerprint reader just works for me
<xnox>
(at least gnome shell lets me do fingerprint login)
<xnox>
ooooh i wonder if usb throughput is why our installer takes it sweet time to boot
<xnox>
jhovold: about mac addresses for wifi & bluetooth -> can we really not use qrtr stuff to access those?
<jhovold>
xnox: pretty sure you have the ucm volume fixes otherwise audio would not work
<jhovold>
xnox: and for the fingerprint reader, you are using patches from my wip branch as support is not upstream yet
<xnox>
probably, we did cherry-pick a tonne of patches from all over the place
mani_s is now known as Guest3136
<xnox>
we tried to take everything we could basically
<xnox>
jhovold: and i even have a plan with grub 2.12, sd-boot, zimg, to have secureboot signed bootable linux image
<jhovold>
we are looking into how to access the wifi and bluetooth addresses, we assume we need to talk to the secure world fw for that
<xnox>
and about tpm, i am getting valid looking measurements, so was hoping after secureboot to have tpm-fde enabled
<xnox>
jhovold: also i thought modem doesn't work, but supposedly it does, so pondering what i am missing
Guest3136 is now known as mani_s
Guest3137 has joined #aarch64-laptops
<jhovold>
tpm may not be that straight forward it's probably also managed by the secure world fw, but i haven't really looked at that much at all yet
Guest3137 is now known as mani_s_
<jhovold>
yeah, modem works, but you may need to set up symlink or two manually to use it
<steev>
good talk, i enjoyed
<clover[m]>
whats the tl;dr
<steev>
it works
<clover[m]>
lol
<enyalios>
jhovold: that status page is awesome, i have a couple of question about it though
<enyalios>
what does the arm64.nopauth kernel arg do?
<enyalios>
and im running linux-firmware from git, but im not seeing the qcvss8280.mbn firmware file mentioned
<konradybcio>
arm64.nopauth disables pointer authentication due to a firmware bug
<konradybcio>
qcvss8280.mbn is stuck in legal purgatory
<konradybcio>
you can pull it from the windows install, somewhere in C:\Windows
<jhovold>
thanks, steve
<enyalios>
ah thanks, what is that firmware file for?
<jhovold>
enyalios: video acceleration
<enyalios>
ah, that would be nice to have someday, but not critical for me
<steev>
you still want to add it though
<jhovold>
just make sure you have that fw if you run my wip branches (or steve's which are based on those) as otherwise you'll lose a couple of hours of battery time
<enyalios>
i blew away my windows install
<enyalios>
oh wow, i could be getting more battery life than this?!
<enyalios>
this laptop is amazing
<steev>
i believe you can download it from lenovo's x13s downloads area (don't recall which file it is) - the exe's can be extracted with inno-xtract
<jhovold>
i'm at 15h of idle time with PCIe ASPM
<enyalios>
neat, ill poke around when i have some time
<enyalios>
wow, thats awesome, ive been getting around 10-12h but thats while using it for web browsing and ssh
<jhovold>
i'm pretty sure we'll have that fw released soon too, so hopefully not an issue for much longer
<enyalios>
awesome, thanks
<exeat>
What do people use that works correctly with that v4l hwaccel interface?
<konradybcio>
gstreamer
<enyalios>
so that file goes in the qcom/sc8280xp/LENOVO/21BX/ directory?
<exeat>
Ah, so far I've only tried mpv (which segfaults)
<steev>
it's in the qualcom graphics driver download on their site - currently n3hdr19w.exe
<jhovold>
steev: that's two times s/steve/steev/ above, sorry...
<steev>
it's fine :)
<steev>
you got it right in the presentation ;)
<steev>
that tickled me :)
<Jasper[m]>
<steev> "good talk, i enjoyed" <- Same, also funny how aggressively Dutch the alwinner simplefb guy is
<steev>
haha
<steev>
i appreciate that johan wants a fix not a workaround
<exeat>
The video says I'm an early adopter but it hasn't felt that way for quite some time!
<exeat>
And the x13s is now the ideal fanless Debian laptop ;), thanks jhovold and everyone else who's worked to make that happen
<clover[m]>
no arch mention in distro support :'(
<steev>
but he did say he uses arch btw
<steev>
sadly, he left the btw off, but he did mention he uses arch on his
<clover[m]>
Hehe
<steev>
i am also slightly jealous that arch users get 3 minut kernel compiles and debian packages take longer :(
<steev>
actually, it's really not so bad with the debian ones either when you keep the cache around
<Jasper[m]>
<steev> "i am also slightly jealous..." <- To be fair, it's also johan_defconfig as he says in the talk
<Jasper[m]>
Debian has a whole host of other drivers enabled
<steev>
yeah we do
<steev>
he's kinda right though... it's not like 90% of the modules will ever be used by me on that system... kinda why i was a gentoo dev for so long
<lun[m]>
<qzed> "lun: is there anything in the..." <- The `qcom_scm firmware:scm: qseecom: scm call failed with error -22` error is from dmesg coinciding with `Failed to read EFI variable LoaderDevicePartUUID: Input/output error` from `bootctl status`