ChanServ 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
chrisl has quit [Ping timeout: 480 seconds]
nashpa has joined #aarch64-laptops
dliviu has quit [Ping timeout: 480 seconds]
hexa- has quit [Quit: WeeChat 4.6.3]
hexa- has joined #aarch64-laptops
indeo has quit [Ping timeout: 480 seconds]
patrickm has quit [Read error: Network is unreachable]
patrickm has joined #aarch64-laptops
chrisl has joined #aarch64-laptops
chrisl has quit [Ping timeout: 480 seconds]
tobhe has joined #aarch64-laptops
hexdump0815 has joined #aarch64-laptops
hexdump02 has quit [Ping timeout: 480 seconds]
tobhe_ has quit [Ping timeout: 480 seconds]
chrisl has joined #aarch64-laptops
indeo has joined #aarch64-laptops
chrisl has quit [Ping timeout: 480 seconds]
indeo has quit [Ping timeout: 480 seconds]
chrisl has joined #aarch64-laptops
chrisl has quit [Ping timeout: 480 seconds]
<steev> if you're on trixie, can you make sure you've run full-upgrade and not just upgrade?
<steev> i'm using kali on my x13s, and we're based on debian testing, and i've never seen the issue here
<steev> oh, if you don't have non-free-firmware enabled, then yeah, you'll be missing some i think, but the qcvss file isn't in a release in debian yet
pabs has quit [Ping timeout: 480 seconds]
chrisl has joined #aarch64-laptops
chrisl has quit [Ping timeout: 480 seconds]
pabs has joined #aarch64-laptops
chrisl has joined #aarch64-laptops
chrisl has quit [Ping timeout: 480 seconds]
<valpackett> tfw you decide to build&install a new kernel with more debug opts and reverted nvme quirk etc. to test stuff out and "reboot to it upon next freeze/crash" and it's been 4 days of uptime with plenty of bluetooth audio usage etc. and there's no freeze nor crash xD
chrisl has joined #aarch64-laptops
chrisl has quit [Ping timeout: 480 seconds]
jglathe_volterra has quit [Remote host closed the connection]
artur has joined #aarch64-laptops
artur is now known as madart
madart has left #aarch64-laptops [#aarch64-laptops]
madart has joined #aarch64-laptops
madart is now known as zaktur
pbrobinson has quit [Ping timeout: 480 seconds]
chrisl has joined #aarch64-laptops
chrisl has quit [Ping timeout: 480 seconds]
pbrobinson has joined #aarch64-laptops
zaktur_ has joined #aarch64-laptops
zaktur has quit [Ping timeout: 480 seconds]
zaktur_ is now known as zaktur
fantom has joined #aarch64-laptops
chrisl has joined #aarch64-laptops
chrisl has quit [Ping timeout: 480 seconds]
zaktur has quit [Quit: zaktur]
nashpa is now known as dliviu
Rayyan has quit [Remote host closed the connection]
Rayyan has joined #aarch64-laptops
chrisl has joined #aarch64-laptops
pbrobinson has quit [Ping timeout: 480 seconds]
pbrobinson has joined #aarch64-laptops
chrisl has quit [Ping timeout: 480 seconds]
sudeepholla__ has joined #aarch64-laptops
sudeepholla has quit [Read error: Connection reset by peer]
bryanodonoghue has joined #aarch64-laptops
pbrobinson has quit [Remote host closed the connection]
pbrobinson has joined #aarch64-laptops
ungeskriptet has quit [Remote host closed the connection]
ungeskriptet has joined #aarch64-laptops
ungeskriptet has quit [Remote host closed the connection]
ungeskriptet has joined #aarch64-laptops
ungeskriptet has quit [Remote host closed the connection]
ungeskriptet has joined #aarch64-laptops
ungeskriptet has quit [Remote host closed the connection]
ungeskriptet has joined #aarch64-laptops
chrisl has joined #aarch64-laptops
chrisl has quit [Ping timeout: 480 seconds]
krei-se has quit [Quit: ZNC 1.9.1 - https://znc.in]
krei-se has joined #aarch64-laptops
<steev> it does be like that sometimes :D
chrisl has joined #aarch64-laptops
chrisl has quit [Ping timeout: 480 seconds]
jhugo has joined #aarch64-laptops
albsen[m] has joined #aarch64-laptops
<albsen[m]> steev: on trixie I have to run `sudo apt dist-upgrade` from time to time to ensure that all packages come across.
<albsen[m]> is it normal that the camera has a green tint on x13s, don't really use it anyway, just asking out of curiosity, maybe I have a missing config somewhere
exeat has joined #aarch64-laptops
chrisl has joined #aarch64-laptops
agl_ has joined #aarch64-laptops
agl has quit [Ping timeout: 480 seconds]
chrisl has quit [Ping timeout: 480 seconds]
<jdb[m]> steev: maybe the firmware are well integrated for the x13s on Debian or kali
<jdb[m]> but the mechanism doesn't seem totally generic for other models
<jdb[m]> At least, that's what I've discovered when doing the bring up of the Surface Pro 9 5G
<jdb[m]> And it seems to be the same for the Surface Pro 11
<steev> albsen[m]: yeah that's "normal" for the green tint - and trixie is still a moving target (though it should slow down now that they're in freeze for release) which is why dist-upgrade is sometimes needed
<steev> jdb[m]: si, the X13s is decently supported on debian (testing+), and that's very nice. it would be nice if MS or whomever is the OEM of the surface models would help with the bringup on linux, but also slightly understandable that they don't (but then again, they have Azure linux, and whatever their IR team uses (don't know the name but i know they have a specialized linux for IR... so maybe linux support would be good there)
<gwolf> steev: Thanks! I am opening it now...
<gwolf> steev: I started recently seeing some messages regarding I/O errors, so I'm worrying my SSD might be close to malfunctioning :-(
<gwolf> OK, I understand this patch will allow both ports to be able to recognize things equally -- but anyway, given bamse's patch for HDMI support was rejected back in 5.19, I don't think it will be very useful :-(
<steev> gwolf: i'm afraid that might be what happened to mine too since it doesn't even get to bios
<gwolf> (I'm still booting 5.19 whenever I need to use external vide)
<gwolf> Oh... and given they are roughly the same age, I might be looking at something worrying... :-(
<steev> mmm, i'm not sure re: hdmi on the usb on newer kernels and i can't check, but i am watching ebay, because i so love the c630
<steev> there is 1 but they want 800 for it still (new)
<gwolf> anyway, I'll keep using it... at least while it's alive :-|
<gwolf> ouch!
<gwolf> It is a very nice machine. One of the best laptops I've had.
<steev> si, it's also ridiculously light
life00[m] has joined #aarch64-laptops
<life00[m]> hiiii
<life00[m]> do you guys think it would be a mistake to buy a snapdragon thinkpad t14s specifically to run linux on it?
<life00[m]> or maybe a regular amd thinkpad would work better? if I am just aiming for battery life
<steev> linux on the t14s is still pretty rough, but getting better
<JensGlathe[m]> @life00:tchncs.de: If you're aiming for battery life purely, ideapad 5 or Thinkbook 16 might be an option
<freekurt[m]> I think battery life is good except for during hibernation, so if you are cool turning the computer off instead of using hibernate when you aren't plugged in, it should be fine.
<JensGlathe[m]> Running a ThinkBook 16 (without touch), got an ideapad 5 withtouch to run (almost the same mainboard)
<JensGlathe[m]> oh and GPU support for these x1p42100 boxes is around the corner
chrisl has joined #aarch64-laptops
chrisl has quit [Ping timeout: 480 seconds]
<life00[m]> <JensGlathe[m]> "@life00:tchncs.de: If you're..." <- well, I also need camera, microphone, and usb-c support, also preferably hdmi as well
<life00[m]> looking through the chat log it seems that thinkpad snapdragon X laptops are still not really usable
<life00[m]> I am currently daily driving macbook M1 with asahi and it's flawless. I aim for a similar experience except more storage, RAM, and a working HDMI, microphone and camera
<life00[m]> plus also I want to be able to dualboot windows
<JensGlathe[m]> The fun part is: On the ThinkBook 16, the only pending non-exotic peripherals are HDMI and GPU acceleration. EC support and fn keys also not, but volume and brightness keys work. 84Wh battery, 32GB RAM, x1p42100.
<JensGlathe[m]> All other peripherals work.
<JensGlathe[m]> FP reader and camera work, too.
<life00[m]> GPU acceleration is kinda important for me. Without it, emulation just kills the battery
<life00[m]> or maybe I am mistaken
<JensGlathe[m]> The patches for X1-45 gpu arrived 2 days ago (?), pending firmware files it will work soon-ish. Maybe faster than a laptop ahips
<life00[m]> thinkbook 16 is better supported than a thinkpad 14s, really?
<life00[m]> I thought thinkpads have the best support
<JensGlathe[m]> what can I say 🤷‍♂️
<life00[m]> what about the regular x86 thinkpads? do they also have similar issues?
<life00[m]> or it's just qualcomm messing everything up
<JensGlathe[m]> All vendors are good at messing up
<JensGlathe[m]> The ThinkBook / Ideapads are cheaper-ish tech, like a normal uvc usb camera (which works ootb once you have the right bus and power supply guessed out). Same for fp reader.
<life00[m]> oh it's also substantially cheaper
<JensGlathe[m]> like one third or so, yeah
<life00[m]> weird
<life00[m]> the price is actually much nicer. maybe I will consider it instead
<life00[m]> are you an owner of such a device or where do you know about the support?
<JensGlathe[m]> If I would chose again I would probably consider a blue ideapad 5 with 32GB and 1TB drive. I like the form factor of the tb16, though
<JensGlathe[m]> If you need screaming power, though... x1e78100 it is, with thermal issues around the corner.
<life00[m]> oh I see, so you own this thinkbook. how's the experience?
<JensGlathe[m]> I like it. Keyboard on these cheaper ones is different, but actually better than on the hp x14. Display is cool but nothing special, no touchscreen skus available. Performance is fairly good (except for graphics, not seen accelerated on Linux), noise level is pretty good. Sound is as good as it can get now with the odd qcom support, mic not tested, camera the usual. Thermal management is great. Very powerful fan, keeping the
<JensGlathe[m]> thing cool on afterburner. No thermal issues seen. Touchpad is meh but usable (I don't like them anyway, mouse guy). Battery life is "not a concern" level good.
<robclark> yoga 7x is pretty well supported, I think prettymuch everything works if you collect up some patches (although I've not gotten around to try camera myself).. for gpu on x1p42 you'll need to backport a patch or wait for mesa 25.2
<JensGlathe[m]> Performance vs the x1 Elite is nearly proportional scaling, compile time 55m vs ~40m on x1e (8 cores vs. 12 cores).
<valpackett> "All vendors are good at messing up" << afaik only Lenovo managed to make ExitBootServices not work reliably lolol
<life00[m]> <robclark> "yoga 7x is pretty well supported..." <- looks nice, but no hdmi and fans I suppose
<valpackett> life00[m]: do you absolutely require a physical onboard hdmi port? usb-c dongle to hdmi works everywhere
<robclark> fwiw, fans aren't too loud
<life00[m]> robclark: there are fans? I meant I thought there are no fans
<robclark> it does have 3x usb-c, I guess you could use usb-c -> hdmi cable
<robclark> 7x does have fans.. you'd have to be in a reasonably quite environment to hear them, they aren't like hairdryers or anything
<life00[m]> wait it doesn't have an audio jack port?
<life00[m]> :'(
<robclark> hmm, I guess not.. it does have bt
chrisl has joined #aarch64-laptops
<travmurav[m]> who needs sound anyways /s
<valpackett> funnily enough the fan issue i had on the latitude is the fan spinning up *due to the lid being closed*.. including in s2idle, lol (i *think* it doesn't happen with deep suspend)
<travmurav[m]> and speaking of not having sound, I've updated slbounce to use the zap shader heuristic by default so it should be a bit easier to "dual boot" el2 now with -el2.dtb newly added in mainline
Treibholz has quit [Remote host closed the connection]
Treibholz has joined #aarch64-laptops
<life00[m]> Jens Glathe: ideapad looks really nice actually.
<life00[m]> it is even more expensive
<life00[m]> around 1k euro with max specs
<life00[m]> *I meant less expensive
<JensGlathe[m]> no idea on the fan, don't have one, but looks nice and 14"
<life00[m]> JensGlathe[m]: yes. I am so used to my 13 inch macbook that I can't really imagine using 16 inch
<life00[m]> how do you know about linux support of this ideapad though? any tips on how to figure this out before buying?
<JensGlathe[m]> I brought it up with a tester who has one 🙂
<life00[m]> nice
<JensGlathe[m]> but it was this weekend only
chrisl has quit [Ping timeout: 480 seconds]
<JensGlathe[m]> hwinfo64 log says its same mainboard as the tb16, different rev
<JensGlathe[m]> sooo the dt should mostly match
<JensGlathe[m]> different address for the touchscreen (confirmed working), a touchpad issue (fixed). WCN7850 wifi / bt works. fp reader and camera different, both USB devices on exactly the same buses. You get the idea.
<life00[m]> is there anything special regarding installation process?
<life00[m]> other than the fact that you use an aarch64 ISO from a distro which supports it?
<valpackett> oh actually maybe the fan still spins up in deep sleep and that's why it's still discharging noticeably. gotta try the ec driver (https://github.com/jglathe/linux_ms_dev_kit/commit/94db1c7456a32bc49e10310f0dd86189b373532d)
<valpackett> life00[m]: just make sure the distro has a recent enough kernel and has the dtb for your device.. and the dtb gets loaded (see https://github.com/TravMurav/dtbloader)
<steev> valpackett: the number of acpi fixups suggests that its not lenovo only (but yeah that exitbootservices....)
<steev> the big thing is that most of x86 has abstracted the issues away (or no one cares to fix)
<steev> https://x0rw3ll.com/re/acpi/acpi.html buddy of mine did a dive to fix... something on his machine and wrote it all up (ASUS tuf something) (also fun fact, 6.16 will have his first kernel commit)
<valpackett> lol yes, the quirks on x86 are just "not new"
<life00[m]> this makes me even more anxious
<life00[m]> uhhh ok I will think about it
<life00[m]> is there any issues with dualbooting windows generally? I heard it sometimes wipes out efi partition after updates
<JensGlathe[m]> No. Very few.
<life00[m]> and the installation process is as usual? just plug in an aarch64 ISO of a linux installer and that's it?
<life00[m]> I will probably be using fedora, so it should be fairly up to date
<life00[m]> the kernel I mean
<life00[m]> but is there any separate patches you applied?
<JensGlathe[m]> Have a look at my repo to get an impression. Its mostly dt stuff, but also newer patches on top of @jhovold‘s tree.
chrisl has joined #aarch64-laptops
chrisl has quit [Ping timeout: 480 seconds]
jglathe_volterra has joined #aarch64-laptops
indeo has joined #aarch64-laptops
<valpackett> hm, the EC FW is substantially different on dell vs. slim7x/magicbook/vivobook, but also the DSDT does not have the notify-EC-on-sleep thing anyway
chrisl has joined #aarch64-laptops
<valpackett> btw trying to run cdsp stuff with https://github.com/quic/fastrpc and i get '32300000.remoteproc:glink-edge.fastrpcglink-apps-dsp.-1.-1: No context ID matches response'
chrisl has quit [Ping timeout: 480 seconds]
<valpackett> big facepalm with the ath12k cpu% investigation: i was wondering why a lot of symbols were missing in perf report, turns out because i was intentionally viewing the report as a regular user but non-root users don't get the addresses from /proc/kallsyms
<valpackett> no new info tho, it's the same spinlock i already knew about
<JensGlathe[m]> But you noticed ✅️
<valpackett> wait but WTF is actually..?? ath12k_mac_flush uses wait_event_timeout which is supposed to be task sleep?? not a spinlock?
indeo has quit [Ping timeout: 480 seconds]
<valpackett> there literally can't be any contention on these things, this wait is the only wait in the whole codebase, so it's just.. waiting for the wakeup using the spinlock somehow instead of sleeping and waiting??
<JensGlathe[m]> it isn't event-driven?
<valpackett> it's supposed to be waiting for an event but something's busted
<valpackett> https://bugzilla.kernel.org/show_bug.cgi?id=220182 huh, other thingy (debugfs) but very related looking
<bamse> valpackett: i agree with your question...glanced at the perf output and it didn't make sense to me either
<bamse> valpackett: perhaps there's some funky inlining or something tricking us, but i didn't find the time to dig deeper
<valpackett> in that debugfs case, the fix mentions that it's the condition not happening and hitting the timeout
<valpackett> but here we have a warning message for the timeout.. could a warn be silenced by default?
<valpackett> oh there they were actually using spinlocks directly in the debug thing heh
chrisl has joined #aarch64-laptops
<valpackett> the perf annotated assembly of ath12k_mac_flush somehow mentions ath12k_core_suspend all the time 0.o
<bamse> valpackett: that "avoid burning" is an open-coded busy wait...so i'd expect that to show up rather clearly in the perf report
<valpackett> bamse: yeah ^^^ 'oh there they were actually using spinlocks directly in the debug thing heh'
chrisl has quit [Ping timeout: 480 seconds]
<bamse> valpackett: as long as contention is rare, that should be okay...but i guess that commit message tells us this isn't the case
<valpackett> here there *can't* be contention
<valpackett> something's really busted
<valpackett> https://0x0.st/8Es6.annotati what even ???
<valpackett> (well, that may just be a perf thing idk)
chrisl has joined #aarch64-laptops
chrisl has quit [Ping timeout: 480 seconds]
indeo has joined #aarch64-laptops
chrisl has joined #aarch64-laptops
chrisl has quit [Ping timeout: 480 seconds]
shawnguo has joined #aarch64-laptops
<bamse> valpackett: i can't tell you why...but perf is lying to us
<bamse> valpackett: it's the same bug as in ath11k...and something has changed so we end up in that loop every time userspace asks for the wifi status
<bamse> valpackett: try bpftrace -e 'kprobe:_raw_spin_unlock_bh { @[kstack] = count(); }'
mbuhl has quit [Remote host closed the connection]
mbuhl has joined #aarch64-laptops
<bamse> valpackett: do you have a sense of when this problem was introduced? (i mean the broken loop code has always been there, but why do we hit it now?)
pabs has quit [Ping timeout: 480 seconds]
chrisl has joined #aarch64-laptops