ChanServ changed the topic of #aarch64-laptops to: Linux support for AArch64 Laptops (Asus NovaGo TP370QL - HP Envy x2 - Lenovo Mixx 630 - Lenovo Yoga C630)
Evaia6 has quit []
derzahl has quit [Remote host closed the connection]
derzahl has joined #aarch64-laptops
Evaia6 has joined #aarch64-laptops
Evaia6 has quit []
Evaia6 has joined #aarch64-laptops
WoC has joined #aarch64-laptops
WoC has quit [Remote host closed the connection]
WoC has joined #aarch64-laptops
<steev>
bamse: https://easyupload.io/wa27a4 this *might* be the ec firmware - it's in the uefi binary as "ecb.bin"
<derzahl>
well, i figured out my problem
<derzahl>
basically me, being a idiot
<derzahl>
some time ago i mustve updated grub, and it still has that bug where it cuts off the zpool name when it generates the config
<derzahl>
i fixed it originally, but completely missed that it was happening again for the last 2 weeks....and tried just about every else but what I needed to fix the stupid problem
iivanov has joined #aarch64-laptops
<derzahl>
does anyone use anything but grub for a bootloader? would be nice to replace it
<derzahl>
woohoo, i can probably even run the latest kernel now that i solved the problem of myself being an oblivious retard
<derzahl>
steev: are you gonna maintain an LTS branch with 5.15 now? I know you mentioned it awhile ago
<derzahl>
or is there enough compelling reasons to only focus on 5.16+?
<maz>
steev: interesting. the system doesn't have LPI support, and yet we try and handle the allocation. I'll look into it.
<maz>
steev: I have a c630 here. if you could send a config file my way, I can verify the potential fix on the actual HW.
<maz>
steev: ah, it looks like the other way around. there is no ITS, but I'm not sure the GIC has LPI support disable. which is consistent with QC's usual hiding of the ITS behind the hypervisor (OMG the HW is scary).
<derzahl>
5.15 boots now. albeit, to a black screen
<derzahl>
but I can still enter my unlock code and it goes
<derzahl>
i actually kind of like that. a little bit of balls out security by obscurity
<steev>
it also looks like the webcam is at the bottom of the screen?
<derzahl>
not nearly as thick as the c630
<derzahl>
its got over an inch of black 'non-screen' at the bottoom
<derzahl>
are there any other more powerful aarch64 laptops out that run linux?
<derzahl>
or likely could without too much work?
<derzahl>
i thought i heard linux was running pretty well on some of the newish galaxybooks
<derzahl>
the ones with the 8cx maybe
<steev>
chromebooks
<steev>
M1
<gwolf>
steev: oh,but derzahl mentioned "more powerful aarch64 laptops ... that run linux ... or likely could without too much work"
<steev>
both of those fit that bill
<gwolf>
AIUI, chromebooks cannot fulfill the first part...
<gwolf>
M1 cannot the last
<steev>
chromeos is just a custom gentoo install
<steev>
gentoo is, last i checked, linux
<broonie>
steev: I suspect the "more powerful" bit is an issue there.
<steev>
m1 can absolutely do it without too much work
<steev>
well he said 8cx, and i thought there was 8cx chromebooks
<broonie>
M1 can already run Linux reasonably well if you're OK with a VM rather than metal.
<steev>
i suppose, if one could get their hands on the flex 5g
<gwolf>
steev: I guess your definition of "too much" is not the same as mine. Last I heard, it was still not possible to use a M1 for doing real Linux work
<gwolf>
broonie: right, that might be the trick...
<steev>
broonie: gwolf: alyssa is already using linux on hers while working on the graphics driver. in fact, she gave the presentation while using it
<broonie>
steev: Oh, I'm thinking regular user rather than someone willing to engage with development stuff.
<gwolf>
steev: But I guess alyssa is beyond what was requested, "too much work"...
<gwolf>
anyway, I won't fight over technicalities (-:
<broonie>
I don't know why she's slacking off working on hardware she can physically access :P
<steev>
broonie: well, derzahl uses zfs for his rootfs, so i think he'd be willing to engage with development stuff :P
<steev>
i believe at this point, marcan has a script that automates a majority of it
<steev>
tbh though, i'm more excited about finding the schematic for the c630, and it showing that we do in fact have an ene kb9022gd
<steev>
but also the lsm6dsm
<steev>
though, no idea how to figure out what spi info it needs
<broonie>
For SPI it should mostly just be the chip select and figuring out what controller it's connected to surely?
<steev>
you vastly over estimate my skills
<steev>
it's the latter part, that i've no idea
<broonie>
I guess if you can see the pins it's connected to on the SoC it should be possible to cross-reference with the pinctrl driver to see which SPI controller can be brought out to those pins?
<steev>
that's true... i just only have the first page of it, i didn't buy the thing because i don't know what the... status of that is?
<broonie>
Ah, that won't help :(
<steev>
it's less than 10 bucks to buy, i just don't wanna get in trouble for using it?
<steev>
when i was at genesi, we were trying to figure out how to get the schematics to people in the community without requiring an NDA
<steev>
alas, we never did before i left... and the only other guy who had them also left
<broonie>
I mean, technically it's easy - you can just publish them if you own them. Otherwise you need to convince whoever owns them that it's OK to publish.
<steev>
right, it was the latter
<steev>
quanta had already reversed it anyway, so i didn't see why we needed to keep it so secret
WoC has quit [Read error: Connection reset by peer]
<derzahl>
obviously got the m1 on my radar. last I checked it still didnt have X11 or wayland running
<derzahl>
chromOS is gentoo based? I was thinking it was android
<derzahl>
still, unless it can be rooted and replaced I dont count that as "linux":) same with android
<derzahl>
i heard that I might have the option of an m1 at my new job
<derzahl>
so ill need to look into exactly what the status of asahi is now
<macc24>
derzahl: idk i heard there was someone doing linux on arm chromebooks
<derzahl>
macc24: that would be cool
<macc24>
derzahl: yea it would
<derzahl>
steev: what kernel options are no longer required now out of the 3: clk_ignore_unused, pd_ignore_unused, and efi=novamap ?
waifro has joined #aarch64-laptops
<derzahl>
any other options that work best with the display?
<steev>
derzahl: clk_ignore_unused is required, the others can be dropped
<derzahl>
i tried console=efifb. it scrolls super slow for a bit after switching modes then still goes blac
<steev>
it should only go back when it's switching over from efifb to msmfb
<steev>
black*
<derzahl>
what are you boot options?
<steev>
derzahl: this is 5.15 right? not 5.16? if 5.16, make sure you've enabled DRM_PANEL_EDP in your config
<derzahl>
can you specify msmfb for console?
<derzahl>
5.15
<steev>
no idea, but probably not if it's a module
<derzahl>
are you able to do graphical boot stuff. like plymouth?
<steev>
linux /boot/vmlinuz-5.15.18 root=UUID=83d75c37-0ba4-4a8e-8e13-996d96bf39b1 ro clk_ignore_unused quiet splash
<steev>
yes, that's what my line is
<macc24>
imagine needing plymouth to have a splash screen
<derzahl>
ah ok, pretty standard
<steev>
don't need it, but i have it because it's pulled in by our metapackages
<derzahl>
yeh i was thinking about trying the plymouth-zfs module to do the unlocking
<macc24>
i like 'fbcon=logo-pos:center,logo-count:1' cmdline options
<derzahl>
i think my problem may be that im just doing a text boot
<macc24>
whether you have fbcon or graphical whatever running is unrelated if display output doesn't wokr
<macc24>
work*
<derzahl>
i should probably just embrace the gooey
<derzahl>
macc24: well it works for the grub menu and for a few lines of booting the kernel. but then it tries to switch into another video mode and goes black
<steev>
it's not really a gooey, it's just a splash screen, you hit esc and it goes away
<derzahl>
if every boots correctly then gdm pops up shortly and all i fine
<macc24>
derzahl: 'switch into another video mode' did you mean 'switch display output from efifb to msmfb'?
<derzahl>
but if there is any problem it just stays black so is really hard to troubleshoot
<macc24>
blind-type 'dmesg > log' and recover log file
<steev>
derzahl: i do also have set gfxmode=1920x1080,1280x720,1280x800,auto (aka GRUB_GFXMODE=1920x1080 in /etc/default/grub )
<macc24>
i do that when i don't want to look for my suzyqable
<derzahl>
macc24: not sure about msmfb
<derzahl>
how would dmesg > log do anything during boot?
<derzahl>
ok cool, ill try that
<macc24>
it would dump kernel log if it booted up without display
<derzahl>
i remember someone else in this channel posting more options that increased verbosity
<derzahl>
i think it was actually" verbose=something or other"
<macc24>
loglevel=7?
<javierm>
derzahl: what I use is: earlycon efi=debug debug initcall_debug log_buf_len=16M ignore_loglevel boot_delay=100
<javierm>
I think ignore_loglevel is what you are asking for
<derzahl>
javierm: ahh i think thats it
<derzahl>
thanks
<derzahl>
macc24: "dump > log" doesnt need a shell or to even be logged in? the kernel listens for that key sequence?
<macc24>
derzahl: `dmesg > log` it's a shell command
<derzahl>
macc24: right. so i would first need the system to boot and then login:)
<steev>
i'm gonna blame zfs
<derzahl>
eh. id be surprised if it was zfs. i would guess something else in arch is causing the difference
<derzahl>
does debian use systemd boot? i forget
<steev>
yes
<steev>
wait
<steev>
you mean systemd-boot? or systemd as the init system
<steev>
the former is a no
<derzahl>
sorry, i mean systemd init
<derzahl>
vs busybox
<derzahl>
for the initrd
derzahl has quit [Remote host closed the connection]
derzahl has joined #aarch64-laptops
<derzahl>
was there any progress make on the hwclock for the c630?
<derzahl>
or some workaround to not have the clock reset every boot
<steev>
iirc there isn't anything we can do? i just let systemd-timesyncd do its thing
Esmil has joined #aarch64-laptops
javierm has joined #aarch64-laptops
iivanov has quit [Remote host closed the connection]