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
enyalios has joined #aarch64-laptops
enyalios_ has quit [Ping timeout: 480 seconds]
kuter has quit [Quit: Leaving]
tobhe_ has quit [Quit: leaving]
tobhe has joined #aarch64-laptops
tobhe has quit []
tobhe has joined #aarch64-laptops
tsoni has joined #aarch64-laptops
hexdump01 has joined #aarch64-laptops
hexdump0815 has quit [Ping timeout: 480 seconds]
tsoni has quit [Ping timeout: 480 seconds]
rfs613 has quit [Ping timeout: 480 seconds]
alfredo has joined #aarch64-laptops
<steev>
yep, that's the ticket, sadly
<JensGlathe[m]>
newer gcc again?
possiblemeatball has quit [Remote host closed the connection]
<JensGlathe[m]>
We also have to choose an older gcc from snapshot.d.o because recent gcc-13 requires recent binutils. This makes the kernel work again.
alfredo1 has joined #aarch64-laptops
iivanov has joined #aarch64-laptops
alfredo has quit [Ping timeout: 480 seconds]
alfredo1 is now known as alfredo
iivanov has quit []
iivanov has joined #aarch64-laptops
tsoni has joined #aarch64-laptops
<steev>
we don't have the newer gcc (13.2.1 iirc) yet in kali, but we do have the affected binutils 2.42.50.20240618
<steev>
the gcc compile i'm doing is to punt the cve patchset
rfs613 has joined #aarch64-laptops
iivanov has quit [Remote host closed the connection]
alfredo1 has joined #aarch64-laptops
alfredo has quit [Ping timeout: 480 seconds]
alfredo1 is now known as alfredo
iivanov has joined #aarch64-laptops
iivanov has quit [Remote host closed the connection]
alfredo has quit [Ping timeout: 480 seconds]
iivanov has joined #aarch64-laptops
iivanov has quit [Quit: Leaving...]
jhovold has joined #aarch64-laptops
tsoni has quit [Ping timeout: 480 seconds]
iivanov has joined #aarch64-laptops
iivanov has quit []
iivanov has joined #aarch64-laptops
Lucanis has joined #aarch64-laptops
smpl has joined #aarch64-laptops
kuter has joined #aarch64-laptops
Caterpillar has joined #aarch64-laptops
Caterpillar has quit [Quit: Konversation terminated!]
Caterpillar has joined #aarch64-laptops
<abcdw>
Hey aarch64-laptops! Any x13s users here? Do you have sleep states working properly? (not draining battery quickly, while sleeping) If yes, do you need a custom kernel or vanilla one is ok?
<jhovold>
abcdw: no, you won't be able to hit the deepest low power state yet (with any kernel)
<jhovold>
you get about 30 h of suspend curretly
<abcdw>
jhovold: Thank you for reply. Do you know the hypotetical way to implement it? BTW, thank you for the work on the kernel!
iivanov has quit [Remote host closed the connection]
<abcdw>
Also, I heard there are problems with wifi and audio. Do speakers work (loud enough), do jack work? Any problems like coil whine or cracks? How often wifi crashes?
hightower2 has joined #aarch64-laptops
<jhovold>
there are some resources (e.g. related to PCIe) that are still left on and prevents the SoC from entering the deeper states
enyalios has quit [Read error: Connection reset by peer]
enyalios has joined #aarch64-laptops
<Jasper[m]>
<Eighth_Doctor> "why can't we use UEFI+ACPI on X1..." <- Qualcomm implements most of the functionality of ACPI in their Windows drivers. Everything that is present in the tables themselves aren't very useful with no driver implementation available on the Linux side of things
<Jasper[m]>
Getting it to feature parity would mean an insane amount of reverse engineering and debugging of something that already works fairly well while using Device Trees
<abcdw>
BTW, do qemu-kvm works fine on X13s?
<Jasper[m]>
Only when booted in el2 using slbounce, but you will lose functionality then.
<albsen[m]>
<jhovold> "wifi works fine for most people,..." <- this is my workaround, maybe we can add that to your wiki: `sudo modprobe ath11k_pci -r; sudo modprobe -r ath11k; sudo modprobe ath11k_pci; sudo modprobe ath11k;` just run that a few times until wifi is back online.
<albsen[m]>
and this only happens on the 32gb model I have (without modem), the 16gb model with modem doesn't have that issue.
<Jasper[m]>
albsen[m]: Have you updated your firmware to .41?
<JensGlathe[m]>
I had it with .41, too, but maybe once a week or less
<albsen[m]>
whats actually more annoying is that my debian crashes on wayland whenever I connect an external screen and I'm forced to switch to x11. but that's likely some debian shinanigans
<Jasper[m]>
<albsen[m]> "Screenshot from 2024-06-24 15-50..." <- Oh yeah that's probably good too
<Jasper[m]>
I just copied WCN6855 related things
iivanov has quit [Remote host closed the connection]
iivanov has joined #aarch64-laptops
kuter has quit [Quit: Leaving]
<kalebris>
yeah, probably debian, I am using wayland with external screen and not much issue apart from the dont disconnect while in suspend thing
kuter has joined #aarch64-laptops
<agl>
steev: I see at the us-page of Lenovo that the T14s Gen 6 costs $1889,30. Do you want to order it or do you wait until it is configurable for example to order an better Snapdragon X elite than X1E-78-100 Processor (it have no Turbo boost).
<agl>
?
<agl>
steev: The fix configuration which can bee ordered is the smales of Snapdragon X Elite.
kuter has quit [Quit: Leaving]
<agl>
*smalest
<steev>
agl: i will still wait. even at 1900, my wallet is tight. i have defcon, best friend's wedding and need to get my car fixed. the x13s is my daily driver and works great for me for now. plus, this way, things don't get forgotten on the x13s
kuter has joined #aarch64-laptops
<agl>
steev: I will wait also until the T14s is configurable an I can select the Snapdragon X Elite X1E-84-100 (with Turbo Boost). I also don't have the money at the moment and I have a major dental issue coming up.
<robclark>
seems my yoga slim is arriving earlier than expected.. on Wed :-)
<agl>
steev: The x13s runs also very good for me. I found it is relativ fast. It's my third fastest machine.
<steev>
robclark: nice!
<steev>
agl: nice :) my fastest is my m1, then the x13s, then (potentially) flex5g, and then c630, and then all the various other arm devices... though my radxa zero 3w should be arriving today
<HdkR>
Is the T14s even going to allow 84-100?
<agl>
HdkR: I will wait if it is configurable. In the fix configuration which you can order ist the 78-100 without Turbo Boost, the smalest X Elite.
<HdkR>
Yea, and that's why I'm curious if they announced they will be using any other SoC
<HdkR>
Since it isn't in psref that they will support more than 78-100
<agl>
HdkR: I found only a Laptop of Samsung which have the 84-100.
kuter has quit [Quit: Leaving]
<HdkR>
Indeed, they're the only one currently
<agl>
Samsung don't support Linux!
<agl>
I have asked.
<steev>
none of them "support linux", that's why the community does it not the company
<steev>
it would be nice to get to a point where they do
<agl>
Qualcomm itself supports Linux
<agl>
I don't have the Web-Page where I read this.
<albsen[m]>
tuxedo said they're planning to release an x elite based system (I've read on a random blog post)
<agl>
albsen[m]: Yes, the Company is 50 km from me. I have send an E-Mail to it. They write it will be nex year that this Linux-Notebook will be on the market.
<agl>
*next
<agl>
The price will be ca. 2200 € (EURO)
<agl>
TUXEDO costs always ca. 200-500 € more than other Notebook seller/developer.
<agl>
albsen[m]: TUXEDO is a German Linux-Notebook developer.
possiblemeatball has quit [Remote host closed the connection]
possiblemeatball has joined #aarch64-laptops
alfredo has quit [Quit: alfredo]
<steev>
yeah, my issue was that newer binutils are busted in debian testing/unstable currently
jglathe_sdbox2 has joined #aarch64-laptops
<steev>
but i'm still building gcc, so as soon as that's done, i'll be pushing my 6.9.6 and 6.10-rc5
<steev>
which, i also tested that fix from rafael, and it seems to work good here - in fact, it seems more snappy because it snaps back faster
<steev>
"that fix" being what he submitted to the mailing list/cc'd me on
<jglathe_sdbox2>
yes,its good when its in... comes back fast when skin temp is below 55°C
<jglathe_sdbox2>
matrix is unreachable. I thought I killed my network with a new kernel in the VM
<agl>
jglathe_sdbox2: No Matrix have problems, i try to connect with the revolt-client - there givs an error
<Libre___>
Hello :)
<agl>
*gives
<Libre___>
Is https://github.com/jglathe/linux_ms_dev_kit the repo I should be using as a starting point if I want to get linux booting on the ms devkit/volterra/blackrock(?) today?
<Libre___>
seems to have dtb etc
<jglathe_sdbox2>
yes it is IMO
<jglathe_sdbox2>
there are also pre-installed desktop images
<jglathe_sdbox2>
look into discussions/announcements
jglathe_sdbox2 has quit []
<Libre___>
I figured you'd say that, cheers :)
<Libre___>
aw
jglathe_sdbox2 has joined #aarch64-laptops
<Libre___>
19:45 < Libre___> I figured you'd say that, cheers :)
<Libre___>
My plan is to make a fedora live usb stick for it I think
<JensGlathe[m]>
My recommendation is to try an existing image first, to get a feeling. Also, its good if the Windows partition is de-bitlockered
<JensGlathe[m]>
Bootup via USB type-c is quite hairy. Therefore the images for a test drive
jglathe_sdbox2 has quit []
<Libre___>
Right yeah, I have the recovery key somewhere if I feel like reverting
<Libre___>
current plan is setting it up as a htpc/home server :)
<Libre___>
and arm64 hacking box
<Libre___>
thanks for all the hard work :)
<JensGlathe[m]>
I have one running with EL2 (dev/kvm available)
<Libre___>
O_o
<Libre___>
that is neat
smpl has quit [Ping timeout: 480 seconds]
cyrinux30 has quit []
<steev>
not a laptop, but the radxa zero 3w showed up today
<Jasper[m]>
steev: ooo I'd like one of those
<Jasper[m]>
Especially with a beepberry
<steev>
it's not too shabby so far. but i've only been playing with it a wee bit
<steev>
seems like it would be possible to do something like the uconsole with it, they already have a rockchip version in the a06, but it's out of stock
cyrinux30 has joined #aarch64-laptops
<agl>
steev: Are the binutils at this Moment not runable from Debian testing?
<steev>
they run
<agl>
steev: When I make "apt update" it talks 487 packages for upgrading.
<steev>
that's a lot of updates. if you're on testing, you need to update more often because things roll in and out fairly quickly
<agl>
steev: bevor you mean the binutils where busted
<agl>
steev: I have for a longer time the x13s switched off.
<steev>
i'm not sure what you're asking, sorry. if you have 2.42.50.20241618-1 (i think it is) installed, and build a kernel against it, that kernel will not boot, it will just sit at a black screen
<steev>
i do not know about the newer than 20241618, and as i said, i'm building gcc so i can't do any sort of testing on it. it's https://bugs.debian.org/1074111 - i do not have the time or bandwidth right now
<agl>
ok
possiblemeatball has quit [Remote host closed the connection]
possiblemeatball has joined #aarch64-laptops
<agl>
steev: I have the kernel 6.9.5-1 bevore my upgrade. After the upgrade I produce 6.9.5-2. When I install this the new vmlinux is smaller than vmlinux-6.9.2. When I install again 6.9.5-1 vmlinux is bigger than vmlinux-6.9.2.
<agl>
So with today's upgrade, the error still exists!
<agl>
steev: Du you know how I can downgrade the gcc and the binutils? (I have nver done this.
<agl>
*never
<bamse>
install clang/llvm and do "LLVM=1 make"
<agl>
bamse: ag@Debian-x13s:/media/ag/DATA/steev-kernel/linux-lenovo-x13s-linux-6.9.y$ LLVM=1 make -j 4 bindeb-pkg
<agl>
SYNC include/config/auto.conf.cmd
<agl>
scripts/Kconfig.include:40: linker 'ld.lld' not found
<bamse>
agl: apt install lld
<steev>
i warned!
<steev>
no one ever learns from my mistakes :(
<steev>
until the bug is closed, the error will exist
<tobhe>
is there a technical reason why the drivers don't declare their firmware using MODULE_FIRMWARE() for the x13s? requiring a special initramfs hook seems a little unusual
<bamse>
tobhe: which drivers?
<steev>
none of the firmware is required to be in the initramfs i don't think? unless everything is built in
<agl>
bamse: ag@Debian-x13s:/media/ag/DATA/steev-kernel/linux-lenovo-x13s-linux-6.9.y$ LLVM=1 make -j 4 bindeb-pkg
<bamse>
agl: i know, but i don't see the cause for the fehler in those lines of output...perhaps i'm just reading it wrong
<steev>
tobhe: it will depend on how they have the kernel configured, worst case, it's sitting in the initramfs and unused
<tobhe>
I guess I'll have to take a closer look
<agl>
bamse: OK, I will wait until the gcc and binutils are updated ... that will be in some days. My x13s runs with the 6.9.5-1 Kernel.
jhovold has quit [Ping timeout: 480 seconds]
<bamse>
tobhe: we reuse the same drivers across 300 qualcomm-based devices, each one with vendor-specific signed firmware...so in a perfect world the size of your ramdisk would approach infinity
<steev>
agl: fwiw, if they're still around (and they may not be but you can get them from snapshot.d.o) gcc-13.2.0-25 and binutils-2.42-4 work if you want to try to downgrade them all, but i believe you'll need to downgrade everything in lock step because the newer gcc requires the newer binutils; just make sure you get all of the packages (something like dpkg -l | grep 2.42.50 should spit out all the binutils packages, you'd need to check
<steev>
gcc's version to get it)
<agl>
steev: Thx
<tobhe>
yeah makes sense. thx for the explanation. I wasn't considering signed per-device firmware
<agl>
steev: Sorry that I have to ask this: How can i get the old packages from snapshot.d.o --> https://snapshot.d.o does not work!
<steev>
ah yeah, d.o is shorthand for debian.org sorry
<steev>
like if you see in a package changelog d/whatever: it means something in the debian directory
tsoni has quit [Ping timeout: 480 seconds]
tsoni has joined #aarch64-laptops
tsoni has quit [Read error: Connection reset by peer]
<agl>
steev & bamse: I have it. I can compile again steev's kernel.
<steev>
lucky you
<steev>
i backed out a bunch of patches, went back to previous revisions, luckily i hadn't sent any emails saying "hey, this latest revision breaks boot"
djakov has quit [Ping timeout: 480 seconds]
<agl>
steev: It needs some time, beause I don't found libstdc++-13 (13.2.0.25). At least I found it ober the search.
<agl>
s/ober/over/
djakov has joined #aarch64-laptops
<steev>
should have been on the same page (if you search for the source package it will show the versions and from there every package for each version)
<steev>
or maybe not
<steev>
i have 14 here but i also haven't updated since finding out about binutils and also ringing the alarm in kali so we don't build a kernel currently