Has anyone been able to get kernel crash dumps working on the T14s? Most else is at the point where I am pretty happy with it, but I do occasionally experience kernel panics that I am unable to trace back to their cause, so I’ve tried getting kernel crash dumps working but haven’t had any success.
I am using the 64GB variant of the machine and the symptoms, when I trigger a panic with a crash kernel and reserved memory configured, are very similar to running the machine without the memory cut to 32GB: the screen goes entirely blue for a few seconds and the machine restarts. The restart does not boot into the crash kernel, but restarts as usual.
Has anyone else tried to make this work? Or has there been much progress in determining the cause of the “blue screen” on the 64GB variant? My first thought is that I am reserving memory that is somehow intended for the firmware but am not sure really where to start tracing this down.
Hmm. AFAIK the cause has been pinpointed as an smmu mapping bug in Gunyah in the UEFI BIOS. There is rumoured to be a fix in UEFI, never seen it (but I am far away).
I also have a T14s 64GB, actually specifically bought it for tests. I can operate on EL1 with ~30GB with the use of cutmem, and on EL2 currently yet without sound, but 64GB fully usable and /dev/kvm. Newest developments like qebspil allow the loading of the firmware before booting, and it will give full adsp functionality, including sound. Have to test this yet.
As for BSOD crashes - none with cutmem, none with EL2.
Has anyone gotten sound to work on a recent build with a T14s Gen 6? I’ve gone to the extreme of upgrading to 26.04 Resolute Racoon in hopes of getting working (internal) sound. It seems to be nigh impossible however. No matter what I try it, I get the error below in dmesg and a “dummy output” for my sound device.
[ 7.910457] MultiMedia1 Playback: ASoC: no backend DAIs enabled for MultiMedia1 Playback, possibly missing ALSA mixer-based routing or UCM profile
Unrelated, the flipped webcam is a tough nut to crack as well.
Oh well. There is an issue with alsa-ucm-conf and a newer T14s-hifi.conf, it relies on also having the newer audioreach topology file. Took me a while to bisect and check, still working on it. Support for HDMI/DP breaks sound with the old topology file.
Thank you for your answers, I have finally been able to try with the the IdeaPad Slim5x device tree (sorry it took so long!).
At the “Try Ubuntu” step, I have no keyboard (touchpad is ok), no wifi, “0%” battery and the fan is constantly running. And lots of “failed” in the logs :
Logs
21:50:07 systemd-udevd: /usr/lib/udev/rules.d/90-alsa-restore.rules:20 GOTO="alsa_restore_std" has no matching label, ignoring.
21:48:18 systemd: Failed to start app-gnome-user\x2ddirs\x2dupdate\x2dgtk-3482.scope - Application launched by gnome-session-binary.
21:47:59 gdm-session-wor: gkr-pam: couldn't unlock the login keyring.
21:47:57 systemd-udevd: /usr/lib/udev/rules.d/90-alsa-restore.rules:20 GOTO="alsa_restore_std" has no matching label, ignoring.
21:47:55 kernel: Bluetooth: hci0: QCA Failed to download patch (-2)
21:47:53 systemd: Failed to start pd-mapper.service - Qualcomm PD mapper service.
21:47:53 systemd-udevd: /usr/lib/udev/rules.d/90-alsa-restore.rules:20 GOTO="alsa_restore_std" has no matching label, ignoring.
21:47:39 kernel: ath11k_pci 0004:01:00.0: failed to fetch board.bin from WCN6855/hw2.1
21:47:38 kernel: snd-x1e80100 sound: probe with driver snd-x1e80100 failed with error -22
21:47:38 kernel: Bluetooth: hci0: QCA Failed to download patch (-2)
21:47:37 systemd-udevd: /usr/lib/udev/rules.d/90-alsa-restore.rules:20 GOTO="alsa_restore_std" has no matching label, ignoring.
21:47:36 kernel: power_supply qcom-battmgr-wls: uevent: failed to send synthetic uevent: -11
21:47:36 udevadm: qcom-battmgr-wls: Failed to write 'add' to '/sys/devices/platform/pmic-glink/pmic_glink.power-supply.0/power_supply/qcom-battmgr-wls/uevent': Resource temporarily unavailable
21:47:36 kernel: power_supply qcom-battmgr-bat: uevent: failed to send synthetic uevent: -11
21:47:36 udevadm: qcom-battmgr-ac: Failed to write 'add' to '/sys/devices/platform/pmic-glink/pmic_glink.power-supply.0/power_supply/qcom-battmgr-ac/uevent': Resource temporarily unavailable
21:47:36 kernel: power_supply qcom-battmgr-ac: uevent: failed to send synthetic uevent: -11
21:47:36 kernel: qcom_pmic_glink pmic-glink: Failed to create device link (0x180) with supplier a800000.usb for /pmic-glink/connector@1
21:47:36 kernel: msm_dpu ae01000.display-controller: no GPU device was found
21:47:36 kernel: qcom-pcie 1bd0000.pcie: probe with driver qcom-pcie failed with error -61
21:47:36 kernel: phy phy-1bd4000.phy.6: phy poweron failed --> -61
21:47:36 kernel: qcom-qmp-pcie-phy 1bd4000.phy: Init sequence not available
21:47:36 kernel: remoteproc remoteproc1: request_firmware failed: -2
21:47:36 kernel: qcom_pmic_glink pmic-glink: Failed to create device link (0x180) with supplier fda000.phy for /pmic-glink/connector@1
21:47:36 kernel: remoteproc remoteproc0: request_firmware failed: -2
21:47:36 kernel: qcom_pmic_glink pmic-glink: Failed to create device link (0x180) with supplier fd5000.phy for /pmic-glink/connector@0
21:47:36 kernel: qcom-ice 1d88000.crypto: probe with driver qcom-ice failed with error -95
21:47:36 kernel: qcom_pmic_glink pmic-glink: Failed to create device link (0x180) with supplier usb-1-ss1-sbu-mux for /pmic-glink/connector@1
For the moment, I have not tried to install given these issues.
Hmm hey man, it fscking boots, that’s something It is enough to get some data. But one of the error messages is fun: qcom-pcie 1bd0000 and phy 1bd4000 errors are a sign that PCIe3 isn’t there. On the Ideapad Slim5x there is the pcie-connected sdcard reader. But since you’re trying to get the Vivobook 14 up, another candidate to try may be the VivoBook S15 x1p42100.
Keyboard not there: yes possible. Maybe different i2c bus, maybe shared IRQ line with the touchpad. Things that can be rectified with the ACPI dump and a dedicated device tree, the shared IRQ will be harder (need to write a mux driver). But as long as USB ports work you could further explore with an external keyboard / mouse if needed. Overall good to know that it boots at all
I was wondering if there is any progress on having USB-C work reliably (screen+PD+usb-hub). Obviously not expecting anything and am really grateful for the amount of work unpaid people have invested into this platform.
But the issues around USB-C are driving me nuts. Every time I boot the system, there is an odyssey of unplugging/replugging until the screen works, sometimes it takes 20 tries. Sometimes even that’s not enough, and I reboot, which seems to slightly raise the chances of it working. Every time the monitor goes into power-save, I need to re-do this.
And these problems are made a lot worse by the other (unrelated) problems, e.g. spending 10min messing with the cable until the screen finally works, just to realise: “oops, this was one of the boots where wifi doesn’t work” → reboot and cable mess again. Or needing to reboot, because of running OOM because only half of the memory works. Or having BSOD resulting in reboot. Or not being able to use standby, and therefore having to turn off and later on again…
Of course no one here is to blame (on the contrary), but after one year of trying to use this laptop productively, I am thinking of returning it to my employer and getting an AMD laptop again. Unless this is likely to get fixed in the near future?
In general, it would be very helpful to me, if the smart and helpful people in this thread (@glathe@tobhe), could give a vague statement regarding the future; something like “At the current rate of progress, most things on T14s will work a year from now”, or “This platform will always be buggy”, or “USB-C is unlikely to ever work reliably”, or “Standby will never do powersaving”…
I could send someone a 100€ note if that raises the chance of getting USB-C working in the short-term (although I know the amount of time required to fix this is more than the money compensates).
At least for my use case using the jg-1 kernel I haven’t had that many issues with USB-C displays. What kills the usability for me is the suspend battery drain.
I just did a battery drain test last night in suspend. I put the laptop in suspend last night with 84% battery, nothing plugged in, running just an xfce with i3 overtop and a firefox instance open. When I woke up the laptop drained 40% battery. Looking at the logs, the laptop went into suspend.
Hmm try 6.17.0-jg-2. For me USB-C to HDMI dongle works fine, for quite some time and on all devices. Hub… is another matter, but I have one SSK hub (whip) that works quite well. However, with 6.17.0-jg-2 HDMI out should work as expected. Actually did a few tests with it in the last days trying to also get soundwire working via dp/hdmi, but no success yet. Just display, no sound - no issue here.
For me, the USB-C-DP feels like it has regressed with 6.17 (both yours and official): I now rarely get a picture out of it. And now the usb-hub functionality often also does not work. This had never been a problem before for me.
HDMI is a painful workaround. It reliably produces a picture, but the picture quality is bad (something about the contrasts is off, I can send you photos if you want). This increases eye-strain. In addition, I really need the usb hub functionality because the T14s doesn’t have enough USB-A for mouse+kbd+webcam. This results in the screen being plugged in and then sometimes being detected as HDMI and in USB-C-PD mode (albeit non-working), resulting in the OS thinking there are two screens attached.
I also regularly work in a floating desk office. If I can’t use USB-C reliably, I need to unplug the keyboard+mouse+webcam, bring my own hub, bring my own HDMI, and bring my own power cable and then crawl under tables to find power outlets
I have a Lenovo Snapdragon X Elite. I am just getting up to speed with the forum here, but it has been essential to getting it running, so thank you for posting. I am particularly interested in audio. What I have learned so far is that the audio is running through the AI accellerator, and that is why it is so strange-problematic. The kernel module is snd_soc_x1e80100. It is intentionally disabled in the kernel because it lacks audio-protection, and you can blow out your speakers.
One neceesary-but-not-suffificient step is to add a kernel module flag in grub: snd_soc_x1e80100.i_accept_the_danger Which I find quite entertaining…however, once I add that, I still get an error in dmesg saying that “Parent card not yet available, widget binding deferred”
This is still using 6.14-rc7. I am willing to move to a later Kernel. I am building my own. Is there a repo I can pull from to get both the upstream changes and the tweaks needed for Ubuntu booting? I am on Ubuntu 25.10
It’s in my kernel and extended image I reference here so often. It’s a bit raw, though. It apparently has shared extended interrupt GPIO on the keyboard and touchpad, so only one of them can work at the moment. There is some effort underway to write a mux driver to resolve this, but a bit stalled at the moment. If have further questions (or data) feel free to open a discussion / issue on my repo.
@glathe Thanks to your efforts thus far I was able to get Audio working as well. I am a few Kernel releases behind, but will try to catch up to the current version here over the weekend. While Video and Camera would be nice, it was audio I really needed.
Is there anything I could do to help out? I have a decent amount of kernel knowledge, but minimal on the audio and video tracks, and I would be very happy to learn more.
There is a grub option that switches the screen and shows you more of the bring-up messages. The will likely show you the last pre-freeze message, which might be relevant info.
There have been a few new useful features added for X1e on NixOS. I was wondering if these could be made available for us humans on Ubuntu!
Gnome Night Light (Red Shift) and Color Management now working - so no more oversaturated desktop if your laptop has an OLED P3 display, such as the Slim 7x
Battery Charge Limits now work (see commits in above Git)
Also, the Power Profiles (Balanced, Power Saver) do nothing on my Slim 7x. They seem to be just placeholders. Does anyone know if any work is being done on this?
Lastly - Kernel 6.18 released today enables Inter Frame Power Collapse for the Adreno GPU which may help a lot with battery life. So would be good to have the X1e version of that on Ubuntu.
I admit up front that this is probably a dumb newbie question but I’m going to ask it any way. I installed the Ubuntu Development branch Resolute Racoon 26.04 and it’s working remarkably well out of the box so to speak with my T14s. Sound is working straight off of the install with no additional work. The only things I’ve noticed that don’t work so far is the webcam and Thunderbolt.
At any rate I tried to install the 6.18 mainline kernel to see if it added any functionality. However, the machine reboots on loading initramfs. In fact ANY version of the mainline aarch64 kernels do this. So my question is simply why is this? Is there something missing vs the Ubuntu kernels?
I was messing with running vanilla kernels on this laptop under Fedora and they worked fine vs the stock Fedora kernels. Unfortunately, while I prefer Fedora they have a LONG way to go to match the level of functionality that the Ubuntu community has achieved on the Snapdragon X Elite.
Edit-Answering my own question. From the GitHub of the Mainline tool:"Note that these kernel packages are missing quite a bit that would be needed on most systems, and many dkms modules and other tools won’t work with them (NVIDIA drivers, VMware modules, etc.).
These packages also will not install nor boot on ARM64 (RPi4 for example), despite being spun for those architectures because they lack DTBs and correctly aligned headers."
It is enough to get some data. But one of the error messages is fun: qcom-pcie 1bd0000 and phy 1bd4000 errors are a sign that PCIe3 isn’t there. On the Ideapad Slim5x there is the pcie-connected sdcard reader. But since you’re trying to get the Vivobook 14 up, another candidate to try may be the VivoBook S15 x1p42100.
