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
dcavalca has joined #aarch64-laptops
anarchron has joined #aarch64-laptops
harvests[m] has joined #aarch64-laptops
malvi[m]1 has joined #aarch64-laptops
akawolf[m] has joined #aarch64-laptops
hlr[m] has joined #aarch64-laptops
pz[m] has joined #aarch64-laptops
sally[m]123 has joined #aarch64-laptops
matthew[m]12 has joined #aarch64-laptops
DigitalPirate[m] has joined #aarch64-laptops
harvestz[m] has joined #aarch64-laptops
EnigmaCurry[m] has joined #aarch64-laptops
cenunix[m] has joined #aarch64-laptops
AlexMarty[m] has joined #aarch64-laptops
Las[m] has joined #aarch64-laptops
Nick[m]123456 has joined #aarch64-laptops
emily[m]1 has joined #aarch64-laptops
Sobek[m] has joined #aarch64-laptops
lun[m] has joined #aarch64-laptops
sz3m3k[m] has joined #aarch64-laptops
wiizzard has joined #aarch64-laptops
patzek[m] has joined #aarch64-laptops
baspar[m] has joined #aarch64-laptops
averyfreeman[m] has joined #aarch64-laptops
szclsya[m] has joined #aarch64-laptops
DocGalaxyBlock[m] has joined #aarch64-laptops
danielt has joined #aarch64-laptops
LikeNeosMatrix[m] has joined #aarch64-laptops
Nei[m] has joined #aarch64-laptops
owc[m] has joined #aarch64-laptops
sporos11[m] has joined #aarch64-laptops
Nao[m] has joined #aarch64-laptops
nscnt[m] has joined #aarch64-laptops
ArtyomK[m] has joined #aarch64-laptops
shjim[m] has joined #aarch64-laptops
NomadNaomie[m] has joined #aarch64-laptops
<DanaG>
Two mac addresses in the mmc boot partition, that is.
<DanaG>
I'll have to file a proper bug later, but here's a spew of dmesg: https://termbin.com/oo6u
<DanaG>
There are a bunch of errors like: hdmi-audio-codec hdmi-audio-codec.19.auto: ASoC: error at snd_soc_dai_hw_params on i2s-hifi
<craftyguy>
is it possible to control the red thinkpad LED on the x13s under Linux? IIRC there was some THINKPAD_ACPI(?) kconfig thing but that doesn't seem to be an option on aarch64
clover[m] has joined #aarch64-laptops
<clover[m]>
No one has been able to figure it out yet
<craftyguy>
ahh ok
<craftyguy>
I thought it would be fun to use it somehow as an indication that the system is in s2idle
<robclark>
HdkR: LLC should be the point of coherency
<HdkR>
robclark: That's what I was thinking as well
<robclark>
there are slices allocate to gpu/etc.. and I guess memory access patterns could result in thrashing.. or we could just have some setting wrong
<HdkR>
Without LRCPC usage I was getting ~14% LLC miss, with LRCPC usage I was getting ~53% LLC miss. Could be indicative as to why LRCPC instructions are horrible
<robclark>
IIRC on pixel things they were trying to play some fancy games with dynamically changing slice (or whatever they call their equiv thing) allocation.. and then realized not using write-alloc but just read-alloc was better... not sure how LRCPC plays into that
<DanaG>
I tweaked the dtb, but it didn't fix the audio. And now that I think about it, I wonder if I need the 6.2 kernel?
DanaG has quit [Remote host closed the connection]
DanaG has joined #aarch64-laptops
<travmurav[m]>
<DanaG> "Looks like tweaking the DTB..." <- this smells like one should fix the ucm instead, afaict this swaps two subdev numbers on the card
<travmurav[m]>
uh did they never upstream the ucm :/
<travmurav[m]>
jenneron: sc7180-adau7002-max98357a.conf ucm doesn't exist I guess?
<DanaG>
I thought UCM told it how to set the controls... but with the kernel I was booting, there are absolutely zero controls in the first place.
<DanaG>
Even if UCM is broken, `amixer` should give some sort of output.
<travmurav[m]>
DanaG: what about `amixer -c 0` - explicitly setting a card?
<travmurav[m]>
I think the default card is usually virtual PA card instead of the hw
<travmurav[m]>
DanaG: my impression right now is that no one upstreamed coachz ucm and swapping the subdevs makes the speakers - the most trivial to drive (need pretty much no controls to set) is then autoselected
<travmurav[m]>
craftyguy: if it still glows in s2idle, my guess would be that it would stop when the device actually reaches low power state in s2idle. At least the EC on aspire1 gets the led signal from the pmic which I imagine is tied to some core power rail, and it glows unless all processors in the soc went to sleep...
<DanaG>
I was able to swap the sections of the dtb, but it didn't change anything (let me check if it even took effect...)
<DanaG>
Checking the booted kernel's /sys/firmware/fdt, it didn't change compared to before. I don't know how to make a modification actually take effect in postmarketos.
<travmurav[m]>
so amixer -c0 is empty?
<travmurav[m]>
how did you modify the dts?
<DanaG>
I used dtc to decompile /boot/sc7180-trogdor-coachz-r3.dtb to dts, then edited the dts, then used dtc to convert it back to dtb. But apparently dtc of that adjusted dtb shows no changes.
<DanaG>
When I run `amixer -c0`, it has no output at all.
<travmurav[m]>
oh weird
<DanaG>
aplay -l lists two subdevices (one mentioning HiFi-1 and one mentioning i2s-hifi-2), but aplay -L lists only a null sink.
<travmurav[m]>
/boot/foo.dtb is owned by the mkinitfs tooling, you should make changes to /boot/dtbs/foo.dtb
<travmurav[m]>
that is, assuming pmOS
<travmurav[m]>
so you would need to change the second one then run mkinitfs
<travmurav[m]>
and the tooling will install stuff properly (afaiu it needs to flash the blob into a special place)
<travmurav[m]>
but I think the more important problem is that the sound card doesn't appear at all
rmsilva has quit [Ping timeout: 480 seconds]
<travmurav[m]>
the hdmi-audio-codec error is, I guess, for when the cable is not plugged in but the sound device is attempted to be opened - which happens when PA probes ucm profiles :S
<travmurav[m]>
but this kinda implies the device is there :S
iivanov has joined #aarch64-laptops
<travmurav[m]>
uhhh
<travmurav[m]>
DanaG: am I correct that coachz is "a tablet with speakers, microphone, and no headset jack"?
hexdump01 has joined #aarch64-laptops
<DanaG>
Huh, I tried modifying that other dtb, and it's not happy, now I get a black screen at boot. So I'll have to boot ChromeOS and mount and copy back the backup.
<DanaG>
Yes, the HP is a detachable tablet, no headphone jack.
hexdump0815 has quit [Ping timeout: 480 seconds]
<travmurav[m]>
DanaG: ah right, I'm not sure but sometimes decompile/recompile with dtc creates unbootable dt sadly
<travmurav[m]>
you can take the dts from source and compile that instead ig, but as I said, that's not your problem, and if there is no headset jack, the ucm would be trivial for it
<DanaG>
I wish Google had included u-boot as a prebuilt altfw option.
rmsilva has joined #aarch64-laptops
<DanaG>
I remember when I was at Google, there was some chromebook or other where one of the bootloader options was u-boot... on x86. U-boot on X86, what kind of nonsense is that?
<travmurav[m]>
when you get it back, could you check cat /proc/asound/cards
<DanaG>
Seems like I'll have to rewrite the image, because chrooting and undoing stuff didn't fix it. No big loss aside from a few minutes to redownload the image.
<travmurav[m]>
yeah, as I said, afaik it's more than just changing the file on the fs that the tooling does
<DanaG>
jeez, doing mkinitfs from a chroot trampled on chromeos's own boot partition.
<travmurav[m]>
DanaG: please create i.e. `ucm2/Qualcomm/sc7180/sc7180-adau7002-max98357a/HiFi.conf`, paste inner contents of SectionUseCase into it, except Comment, and replace with `File "Qualcomm/sc7180/sc7180-adau7002-max98357a/HiFi.conf"`
<travmurav[m]>
ah sorry, File "/Qualcomm/...."
<travmurav[m]>
leading slash
<DanaG>
Wait, so I move just SectionVerb and SectionDevice into the HiFi.conf, and then what goes in the old filename?
<DanaG>
Looks like the other file in the conf.d directory has SectionUseCase { File }
<travmurav[m]>
Conf.d file should just have usecase with comment and file
<travmurav[m]>
That points to hifi.conf with verb abs device(s)
<travmurav[m]>
*and
<travmurav[m]>
(I somehow assumed it's not file but include.file and was like "cool can share a single file instead of explaining where to put them")
<DanaG>
The other directory inside Qualcomm/sc7180 doesn't have a redundant "sc7180" in the name, does that matter?
<travmurav[m]>
Uhh
<travmurav[m]>
It accepts the profile at least :s
<DanaG>
Looks like the name didn't matter, but I changed it (and the referring include) for consistency.
<DanaG>
Oh, I can try setting a PlaybackPCM.
<travmurav[m]>
Weird, not sure why it doesn't want to do playback
<travmurav[m]>
DanaG: but the speaker bit should be the same as the other ucm config there, I guess you could try to compare them... I'm afk for some time
<DanaG>
Oh, I have a stray closing curly bracket when I copied the CapturePCM to PlaybackPCM. With it ending in `,1` instead of `,0`, it works, and has volume control. Thanks for the help!
<travmurav[m]>
DanaG: cool! Could you please document that i.e. @ pmos issue tracker, later someone should probably clean this up and upstream to ucm
<travmurav[m]>
(cc jenneron)
<DanaG>
Do you happen to have a direct link handy to the issue tracker?
<DanaG>
Another issue I have is that the touchpad is sluggish (it doesn't respond sometimes), and two-finger scrolling doesn't work even though the libinput test tool shows multiple fingers.
<jenneron[m]>
<DanaG> "I tweaked the dtb, but it didn't..." <- we use lts kernel, 6.6 may be the next one we upgrade to
<jenneron[m]>
<travmurav[m]> "jenneron: sc7180-adau7002-max983..." <- possible
<jenneron[m]>
<DanaG> "I used dtc to decompile /boot/sc..." <- the problem is that that dtb is copied from /boot/dtbs when making initramfs && creating depthcarge image
<jenneron[m]>
i don't recommend replacing it there, it is provided with package
<jenneron[m]>
oh, that's solved now
<jenneron[m]>
but yeah, open an issue, we can make it properly implemented
jhovold has joined #aarch64-laptops
abelvesa has quit [Server closed connection]
abelvesa has joined #aarch64-laptops
HdkR has quit [Server closed connection]
HdkR has joined #aarch64-laptops
javierm has quit [Server closed connection]
javierm has joined #aarch64-laptops
wookey has quit [Server closed connection]
wookey has joined #aarch64-laptops
agraf has quit [Server closed connection]
agraf has joined #aarch64-laptops
alexeymin has quit [Server closed connection]
alexeymin has joined #aarch64-laptops
enyalios has joined #aarch64-laptops
enyalios_ has quit [Ping timeout: 480 seconds]
<bamse>
gwolf: no, but i will try to get it (battery and display) settled for 6.8...and come up with something for the suspend crash
<gwolf>
bamse: Right, that's what I understood as well. Well, I'd love to be useful in any way it's possible, now that I resolved my failure to build new kernels. So please tell me if you need testers for anything! (I cannot be of much help code-wise at kernel level, sadly)
<gwolf>
My friend did a dti walkthrough at the Uruguay MiniDebConf, and I'd be willing to take a stab at that... but even so, it's a long shot to pretend to be useful
<gwolf>
(but at least I got 6.6 happy and running!)
<bamse>
nice
<bamse>
at plumbers this week, but once i get through the backlog that builds up i'll respin the patches
<gwolf>
<3 !
svarbanov has quit [Ping timeout: 480 seconds]
svarbanov has joined #aarch64-laptops
Mathew has joined #aarch64-laptops
mcbridematt has quit [Remote host closed the connection]
Caterpillar has quit [Quit: Konversation terminated!]
svarbanov has quit [Remote host closed the connection]
svarbanov has joined #aarch64-laptops
neobrain has joined #aarch64-laptops
svarbanov has quit [Remote host closed the connection]
haise01_ has quit [Server closed connection]
svarbanov has joined #aarch64-laptops
haise01 has joined #aarch64-laptops
<neobrain>
What's the preferred method of installing Ubuntu on a Lenovo X13s these days? Do the official 23.10 installers just work or does one still use these "Lunar Lobster" builds that haven't been updated since March?
<neobrain>
(in case it matters, the device already has a Linux installation... it just happens to be broken and for a number of reasons I'd rather start from scratch than try to fix it)
ema has quit [Server closed connection]
ema has joined #aarch64-laptops
<steev>
iirc, xnox1 said that you can just use the 23.10 installer
<neobrain>
cool, will give that a shot then, thanks :)
<steev>
the c630 battery time truly is breathtaking
Caterpillar has joined #aarch64-laptops
<neobrain>
I'm guessing a hang at "EFI stub: Exiting boot services..." can mean anything?
<neobrain>
(couldn't try Ubuntu yet, but that's what I seem to get on both openSUSE+Fedora... sadly I'm also stuck with firmware 1.48, which may be part of the issue)
JoshuaAshton has quit [Server closed connection]
JoshuaAshton has joined #aarch64-laptops
<steev>
you probably need to add
<steev>
arm64.nopauth
<steev>
to the kernel command line for them
<neobrain>
Just to make sure, that's the line starting with "linux", right? I've tried that, didn't seem to change anything sadly
<neobrain>
the opensuse wiki also suggested iommu.passthrough=0, but no dice
<neobrain>
it actually resets/reboots the device instead of hanging sometimes. curious
Jasper[m] has joined #aarch64-laptops
<Jasper[m]>
<neobrain> "(couldn't try Ubuntu yet, but..." <- Do you have a different pc with linux running?
<danielt>
@neobrain: There are lots of reasons to get stuck at exiting boot services but one of them is not loading a devicetree (loading a devicetree is one of the feature added in the newer firmware).
<neobrain>
Jasper[m]: ah, I'd seen that repo but I'd assumed they ran on-device and hence didn't look too far. Will try with that, hopefully that'll get me unstuck
<neobrain>
danielt: makes sense. let's see if I can get somewhere with updated fw then
<Jasper[m]>
neobrain: It is ran on device, but off a bootable usb stick. You can create the stick anywhere else :)
<danielt>
neobrain: I've never tried it myself but from the looks of it the Ubuntu 23.10 installer "cheat" (by making custom install media specific to the X13s) in order to avoid making the user to anything with the device tree before they get started...
<neobrain>
Alright, looks like that went through fine and I'm on 1.58 now :)
<neobrain>
Sadly, openSUSE is still not getting past that EFI stub message. Will try Fedora in a moment
Amber_Harmonia has quit [Server closed connection]
Amber_Harmonia has joined #aarch64-laptops
shoragan has quit [Server closed connection]
<neobrain>
nope, doesn't work either. Are there any odds that Ubuntu is going to magically work in this case, which would justify spending a couple of bucks to buy a new USB boot device?
shoragan has joined #aarch64-laptops
<neobrain>
(my silly problem is this one only has 2 GB but the Ubuntu iso is 2.2 GB... so yeah)
<Jasper[m]>
Well I know there's at least 1 image source that'll work
<Jasper[m]>
Mine is currently installed from that until Fedora starts shipping something that boots (or if there are more direct instructions)
<neobrain>
Which one is that? Is it smaller than 2GB? :D
alpernebbi has quit [Server closed connection]
alpernebbi has joined #aarch64-laptops
<Jasper[m]>
Not sure, I'll try to find it in a sec
<neobrain>
Oh wow, the debian installer at least manages to get past the EFI stub message now. Prints a screen worth of console messages that instantly disappear, then the laptop resets
<HdkR>
Holy shit, the firmware update resolves the n-key rollover issues on X13s
<HdkR>
What a twist
<neobrain>
Meanwhile, I managed to get the Ubuntu iso down to 1.9 by just throwing out the nvidia driver (heh), same issue. Starting to think this usb drive is just borked
<neobrain>
*1.9 GB
<Jasper[m]>
<neobrain> "Oh wow, the debian installer..." <- Did you turn on the "Linux" option in UEFI?
<enyalios>
no other installers that i tried were able to boot other than the ubuntu concept one and this custom arch image https://github.com/ironrobin/archiso-x13s
<enyalios>
i think everything else was missing the devicetree required to boot
<neobrain>
Ah interesting, that's reassuring
<Jasper[m]>
There was some automated ci thing I downloaded my iso from but I can't find it rn
<neobrain>
A problem is that it's arch, but at least it's literally anything that works
<neobrain>
lol
<neobrain>
So with Arch running, I finaly managed to fix my previous Linux installation... because all I had to do was rename /boot/grub.cfg.old to /boot/grub.cfg
<neobrain>
(still planning a full reinstall, but lesson learned that it can be surprisingly difficult to revert simple changes :D)
jhovold has quit [Quit: WeeChat 4.0.4]
<enyalios>
yay, glad you fixed things
<neobrain>
Well, thanks to you all for the help :)
<neobrain>
I also managed to get openSUSE booting now. Their wiki mentions needing to manually copy over the dtb file to /boot/efi, which I hadn't done before since the wiki only mentions it for a different image that I wasn't using (oh, also I couldn't copy over the file without a running system :p)
Mathew has quit [Quit: Leaving]
mcbridematt has joined #aarch64-laptops
<neobrain>
Install complete, looks like openSUSE works fine!
mkrawczuk has quit [Quit: No Ping reply in 180 seconds.]
mkrawczuk has joined #aarch64-laptops
<steev>
fwiw, you can force it to load from from your rootfs (if not using an encrypted rootfs, obviously), in grub by passing the path to the device tree; sadly, only ubuntu has the patch that automagically adds it
<robclark>
if dtb is in esp you can also load it via kernel cmdline param