#aarch64-laptops on 2025-02-18 — irc logs at oftc.irclog.whitequark.org

2024-07-16 04:52 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

00:02 zdykstra has quit [Quit: WeeChat 4.5.1]

00:03 zdykstra has joined #aarch64-laptops

00:14 chrisl has quit [Ping timeout: 480 seconds]

00:22 chrisl has joined #aarch64-laptops

00:25 weirdtreething has quit [Read error: Connection reset by peer]

00:25 weirdtreething has joined #aarch64-laptops

00:30 chrisl has quit [Ping timeout: 480 seconds]

01:14 weirdtreething_ has joined #aarch64-laptops

01:18 weirdtreething has quit [Ping timeout: 480 seconds]

01:22 weirdtreething_ is now known as weirdtreething

01:30 chrisl has joined #aarch64-laptops

01:38 chrisl has quit [Ping timeout: 480 seconds]

02:36 chrisl has joined #aarch64-laptops

02:44 chrisl has quit [Ping timeout: 480 seconds]

03:09 tobhe has joined #aarch64-laptops

03:16 tobhe_ has quit [Ping timeout: 480 seconds]

03:56 hexdump01 has joined #aarch64-laptops

03:58 hexdump0815 has quit [Ping timeout: 480 seconds]

04:12 pabs has quit [Remote host closed the connection]

04:14 pabs has joined #aarch64-laptops

04:17 chrisl has joined #aarch64-laptops

04:25 chrisl has quit [Ping timeout: 480 seconds]

04:54 chrisl has joined #aarch64-laptops

04:56 ungeskriptet has quit [Remote host closed the connection]

04:57 ungeskriptet has joined #aarch64-laptops

05:02 chrisl has quit [Ping timeout: 480 seconds]

05:42 ungeskriptet_ has joined #aarch64-laptops

05:47 ungeskriptet has quit [Ping timeout: 480 seconds]

05:50 ungeskriptet_ is now known as ungeskriptet

06:00 chrisl has joined #aarch64-laptops

06:08 chrisl has quit [Ping timeout: 480 seconds]

06:32 landwork has quit [Ping timeout: 480 seconds]

06:42 landwork has joined #aarch64-laptops

07:03 jhovold has joined #aarch64-laptops

07:09 chrisl has joined #aarch64-laptops

07:20 chrisl has quit [Ping timeout: 480 seconds]

07:47 chrisl has joined #aarch64-laptops

07:55 chrisl has quit [Ping timeout: 480 seconds]

08:30 <jhovold> anthony25: yes, dp altmode works fine with my rc3 wip branch on both the crd and t14s

08:32 <jhovold> robclark: ok, but good that you seem to have a reliable reproducer for the pcie probe failure

08:33 <jhovold> it appears to be a race of sorts so you're probably just unlucky with the timings

08:33 srinik has joined #aarch64-laptops

08:38 <jhovold> robclark: regarding the audio probe failure you mentioned, don't you have the yoga slim7x?

08:39 <jhovold> that one doesn't have the wcd codec that was failing described in dt, so not sure what's going on there

08:55 chrisl has joined #aarch64-laptops

09:03 chrisl has quit [Ping timeout: 480 seconds]

09:08 <JensGlathe[m]> It has the codec in my definition. I guess it may not be complete or erroneous

09:27 kettenis has quit [Remote host closed the connection]

09:37 <jhovold> JensGlathe[m]: ok, thanks for explaining

10:41 chrisl has joined #aarch64-laptops

10:49 chrisl has quit [Ping timeout: 480 seconds]

11:00 ungeskriptet has quit [Remote host closed the connection]

11:00 ungeskriptet has joined #aarch64-laptops

12:12 landwork has quit [Read error: No route to host]

12:13 landwork has joined #aarch64-laptops

12:18 landwork has quit [Quit: Leaving.]

12:18 landwork has joined #aarch64-laptops

12:19 landwork has quit []

12:19 landwork has joined #aarch64-laptops

12:21 <albsen[m]> just installed the ubuntu 6.14.0.5 kernel it's booting and appears to be working. haven't encountered the random restarts/crashes yet that I've been seeing with the previous kernels. thx.

12:26 <albsen[m]> maybe should mentioned, this is on t14s 64gb (limited to 32gb)

12:33 <tobhe> great :) thanks for the feedback

12:41 landwork has quit [Quit: Leaving.]

13:10 hightower2 has quit [Remote host closed the connection]

13:13 hightower2 has joined #aarch64-laptops

13:18 <albsen[m]> tobhe: first thing I noticed, bluetooth is not working anymore

13:19 <tobhe> I'll have a look at that

13:24 <albsen[m]> I'll link the dmesg here in a second, also I can see a lot "qcom-iris aa00000.video-codec: core init failed"

13:25 <robclark> JensGlathe[m], jhovold: oh.. it is possible that JensGlathe[m]'s patch to add dp/retimer dts bits introduced that codec and broke audio.. I guess I can try to back out the audio related bits. (Would be nice to get proper external display support in place for 7x)

13:27 <JensGlathe[m]> See, I was nervous when it compiled on first try

13:27 <albsen[m]> https://pastebin.com/amRgJaUJ

13:36 <tobhe> albsen[m]: I can reproduce it on mine too

13:36 ungeskriptet has quit [Remote host closed the connection]

13:37 ungeskriptet has joined #aarch64-laptops

13:37 <albsen[m]> great, let me know if I can do anything else to help

13:41 pabs has quit [Remote host closed the connection]

13:52 pabs has joined #aarch64-laptops

14:12 chrisl has joined #aarch64-laptops

14:20 chrisl has quit [Ping timeout: 480 seconds]

14:46 <robclark> jhovold: do we know anything about what is the actual codec used for external display on 7x? I'm guestimating that JensGlathe[m] patch (which does work for the display side of things) is just copying the setup on CRD?

14:48 <sibis> jhovold: I'll make sure I post the cpufreq quirks for the perf protocol over the next 2 days so that I get feedback from Cristian

14:53 <jhovold> sibis: sounds good, thanks!

14:54 <jhovold> robclark: that was my impression as well, I'll see if I can dig something out

14:54 chrisl has joined #aarch64-laptops

15:14 <robclark> yeah, reverting the dp retimer patch gets wifi and audio back.. idk what the connection is to wifi, but I'll try to trim the patch down to just the retimer stuff in a bit.. hopefully I can have external display + sound + wifi at the same time ;-)

15:16 <jhovold> so it was also involved in triggering the pci probe issue? make sure to save that branch in case it was just due to (un)lucky timing of events

15:25 <robclark> well, either that or really lucky random odds that ath12k hadn't been working across a number of reboots and started working on first reboot after reverting patch ;-)

15:26 <robclark> jhovold: patch in question: https://gitlab.freedesktop.org/robclark/msm/-/commit/243fe5091abc850d64a0305b2749f0bce27de645

15:26 jglathe_ has joined #aarch64-laptops

15:32 <JensGlathe[m]> This hints to the pwrseq stuff...

15:32 <jhovold> thanks, but we're gonna need your help to track it down since you have a yoga :)

15:32 <anthony25> the wifi works for me with JensGlathe[m]'s patch

15:32 <robclark> stripping out the codec stuff from that gets audio working... now lets see if I can figure out which part breaks ath12k

15:33 <robclark> hmm, wcn7850-pmu doesn't seem display related

15:33 <jhovold> that's bluetooth

15:33 <jhovold> and wifi

15:34 <JensGlathe[m]> and WiFi

15:34 <robclark> hmm, never tried bt before or after

15:34 jglathe_angrybox has quit [Ping timeout: 480 seconds]

15:34 <anthony25> let me try bluetooth again with this patch

15:34 <anthony25> it used to work with 6.14-rc2

15:37 <anthony25> wifi and bluetooth work for me

15:37 <anthony25> I had to adapt the patch as I had some merge conflicts, let me see if there's something different with yours

15:39 <robclark> ok, after backing out those changes, I have wifi working again as well

15:40 <anthony25> bluetooth still has some disconnections with my earplugs, however (that I don't have on my desktop PC)

15:41 <robclark> ok, here is slimmed down patch that has external display working without breaking audio/wifi:

15:41 <robclark> https://www.irccloud.com/pastebin/hQav5GDa/

15:41 <robclark> I've not gone thru to cross check that all the regulators/etc are required/used

15:42 <robclark> oh, and I guess there is a 3rd retimer on 7x.. the two left usb-c ports work, the one port on the right has no display (but usb seems to work)

15:43 <JensGlathe[m]> There are three retimers configured

15:43 <anthony25> yeah I saw that last time, displayport was detected but I was hitting another error

15:44 <JensGlathe[m]> The port on the right doesn't work on Windows, either (?)

15:44 <JensGlathe[m]> for dp

15:45 <anthony25> I need to look at the logs of our conversation about it last time, but I think it didn't work (even on 1080p 60Hz)

15:45 <robclark> hmm.. so abelvesa posted dp support for 7x early on (before the retimer driver)... and pretty sure the situation at the time was the right port and one of the two left ports worked

15:45 <JensGlathe[m]> that was what I remember

15:51 <anthony25> this is what I'm using: https://bpa.st/7REA

15:51 <anthony25> I don't know if it changes anything, but I have the EC enabled

15:53 <anthony25> full dts: https://bpa.st/WLSA

15:56 <anthony25> ha, actually, DP altmode works

15:56 <anthony25> I don't know why it stopped working on my 4K@60Hz, but it works on my 1080p screen

15:57 <anthony25> the right port still fails with a "failed DP link training", but the 2 left ports work

15:57 <robclark> at a minimum, the bt and wifi should be separate patches.. I'm not entirely sure what the implications of the wifi changes since it was working before (but maybe some regulators were not getting shut down when wifi suspends?

15:58 <robclark> oh, I am getting spammed with:

15:58 <robclark> [ +0.959334] ath12k_pci 0004:01:00.0: failed to transmit frame -12

15:59 <robclark> hmm, so maybe wifi is not in fact working (but I do at least see the ath12k device).. I have ethernet connected via external display which was masking the issue

15:59 <anthony25> I'm talking about the EC because if I understand correctly, it can shutdown the wifi and bluetooth (through the airplane mode key)

16:00 <robclark> and some splats:

16:00 <robclark> https://www.irccloud.com/pastebin/QmvMbFGN/

16:01 <anthony25> JensGlathe[m]: on windows, the 3 ports work with display port

16:01 <robclark> oh, neat, fn-f8 (airplane key) locks things up solid

16:03 <JensGlathe[m]> Okay. Sounds like there is another oddity with regulators and/or gpio

16:03 <anthony25> just windows can only push 4K@120 on one port (2nd port on the left), the rest are failing

16:52 <robclark> trimmed a few more thigns out and ath12k is working again

16:52 <robclark> https://www.irccloud.com/pastebin/kLNwbeqR/

17:06 <anthony25> what did you change?

17:08 <robclark> delta from original patch:

17:08 <robclark> https://www.irccloud.com/pastebin/W3B1FiUO/

17:08 <robclark> basically just removing everything that was not actually related to retimer/dp

17:24 <anthony25> ha so you have no bluetooth now, no?

18:08 srinik has quit [Ping timeout: 480 seconds]

18:15 <robclark> anthony25: no bt.. I'd never used it before.. (but bt shouldn't be part of the same patch as dp/retimer ;-))

19:05 cyrinux has quit []

19:07 cyrinux has joined #aarch64-laptops

19:21 jglathe_ has quit [Ping timeout: 480 seconds]

19:56 rz_ has joined #aarch64-laptops

20:03 rz has quit [Ping timeout: 480 seconds]

20:03 rz has joined #aarch64-laptops

20:08 rz_ has quit [Ping timeout: 480 seconds]

20:21 rz has quit [Ping timeout: 480 seconds]

20:27 akaWolf has joined #aarch64-laptops

20:30 rz has joined #aarch64-laptops

20:45 rz has quit [Ping timeout: 480 seconds]

20:45 weirdtreething has quit [Ping timeout: 480 seconds]

20:46 weirdtreething has joined #aarch64-laptops

20:52 rz has joined #aarch64-laptops

21:06 rz has quit [Ping timeout: 480 seconds]

21:23 <d0pefish> Surface Pro 11 / Arch Linux ARM with 6.14-rc3: https://github.com/dwhinham/linux-surface-pro-11/releases/tag/2025-02-18

21:24 <JensGlathe[m]> Do you get touchscreen / keyboard/ touchpad to run?

21:25 <d0pefish> type cover keyboard with its touchpad mouse is working but still need to get touchscreen+pen working

21:26 <d0pefish> I have the kernel module for touchscreen as it's supposed to be spi-hid but I need to try and write some device tree bits for it

21:27 <JensGlathe[m]> are these behind an extra controller ore is it a matter of scanning the i2c buses? DSDT looks really weird

21:28 <d0pefish> Yeah, the detachable keyboard/trackpad comes via surface aggregator module

21:28 <d0pefish> the drivers are in-tree and at least partially work, some unhandled events are being logged but at least it's usable

21:34 jhovold has quit [Ping timeout: 480 seconds]

21:50 <JensGlathe[m]> so we need a device tree definition for them

22:38 <bamse> perhaps i've missed something, but i think we're missing support for geni spi-hid

22:38 <bamse> konradybcio: ^^

22:40 <d0pefish> I tried to enable an spi device with &spi1 { status = "okay"; }; in my device tree but I get a crash in spi_geni_probe()

22:45 <bamse> crash or it just wouldn't work?

22:48 icecream95 has joined #aarch64-laptops

23:14 <icecream95> Has anyone else been seeing occasional memory corruption with 6.14? (I'm using the Ubuntu kernel)

23:15 <d0pefish> bamse: a crash, here's a dmesg snippet https://gist.github.com/dwhinham/365d469f894af68e9d21e254008f7ca9

23:15 <d0pefish> No idea if it should work at all or not, I just tried to enable one of the SPI devices with "okay" in the DT

23:17 <icecream95> Unfortunately I haven't managed to find a way to reliably reproduce the corruption, but I've seen it twice over the past few days

23:26 <konradybcio> d0pefish: bamse we need some geni bits and some hid-spi bits.. hopefully it's not far away but as always, no promises

23:28 <d0pefish> thanks :)

23:35 <bamse> konradybcio: and that's still stuck in endianess purgatory?

23:37 <bamse> d0pefish: hmm, sync ext abort would imply some other issue...are you sure that spi1 is what you're looking for?

23:38 <bamse> i.e. it's some indication of access of unavailable resources... the protocol issue should be a cleaner reject by the driver

23:49 <konradybcio> bamse I haven't checked out the latest updates yet, been busy working on some other fun stuff

23:50 <konradybcio> at least on SL7, GENI talks pure QSPI and we still need to borrow a driver from surface duo's android kernel & clean it up to the point of upstreamability

23:50 <konradybcio> unfortunately though, it's not conformant with microsoft's pdf-described hidspi (it seems to be like a pre-release version)

23:52 rz has joined #aarch64-laptops

23:56 <d0pefish> bamse: I'm actually not sure :) I just found spi0 and 1 in the base x1e dtsi and tried to enable them. My goal was to try the surface duo spi-hid driver Konrad mentioned, but I can't do that until I can get spi to come up

23:57 <d0pefish> was hoping that once spi was successfully enabled I could poke around with some userspace tools e.g. on i2c you have i2cdetect and so on

23:58 <konradybcio> we should be able to do something about spi in the not very distant future..