Thanks. I use W11 from the USB-C expansion card twice a year for clients who have this wonderful piece of engineering which is Barco ClickShare
For this usage it works well, but OK, got it, it won’t help with the bios upgrade.
I’m quite used with various hardware bios/FW updates, however I rarely had so much doubts about the process… The Framework is my professional computer and it’s not helping with taking “risks” on it.
For example, I don’t have a clue about what’s retimer firmware, how can you see that it is not up to date eventually after bios upgrade?
This is not really about bios update, but about bios/firmware functionality. I’m on bios 3.08.
I’m currently trying to get intel_lpmd working as a measure to save battery power. lpmd will deactivate most cores under light load conditions, thus saving more power than having those cores activated, but doing nothing.
My Intel 1260P should support HFI hints, used for lpmd to decide on activating/deactivating cores. Unfortunately, there seem none such hints to be delivered to the OS (see HfiLpmEnable set to 1 prevents from entering the low power mode on i7-1260P · Issue #40 · intel/intel-lpmd · GitHub) and the developer of lpmd seems to suspect that the firmware is simply not delivering HFI hints.
@ all users: Can someone approve or falsify my observation, that the firmware is not delivering HFI hints? @Kieran_Levin As a question to the tech team: is the firmware of 12th gen frameworks correctly delivering HFI hints?
What am I doing wrong? Any ideas what else I can try?
I am very frustrated. I bought a thunderbolt 4 docking station and an external NVMe case with my laptop. Unfortunately, I can’t use either of them
Today I’ve updated BIOS on my 13" Intel Gen12 using EFI method from 3.06 to 3.08 and upgraded the battery to the 61 Wh version.
I haven’t updated retimers yet, so it was only the automatic part.
However, now I am having issues with suspend & resume. Suspend works as usual, but upon resume:
Screen backlight continously switches on and off after several seconds
Sometimes KDE’s screen lock image is displayed, but it is not responding to key presses.
Sometimes image is visible when I move the mouse cursor (cursor is updating correctly without lags) and immediately disappears (goes completely black) once I stop moving the cursor (PSR issue?)
Sometimes image is broken into misaligned vertical stripes
Processor is heating, since after a while cooler starts blowing really strong.
I am able to reboot the system only by holding down the power button for several seconds
I am pretty sure it all broke after I updated the BIOS. Before that everything was working flawlessly (including hibernation to a LUKS encrypted swap partition).
CPU 12th Gen Intel(R) Core™ i7-1260P
BIOS 03.08
Kernel 5.15.0-97-generic
Ubuntu 22.04.4 LTS
During my investigation I’ve tested the latest Kubuntu 24.04 LTS from live USB (Framework’s 1TB memory), and… it worked!
I mean, I was able to successfully put the laptop to sleep and then bring it back with no apparent issues.
So, I believe it has something to do with the kernel version. Newer kernel 6.8 works fine. Still this is a bit unfortunate since, Ubuntu 22.04.4 LTS is still used by many.
P.S.: I also checked the hypothesis that the stalls could be caused by SSD. So I booted from USB, mounted my SSD partitions and then did a sleep/resume cycle. It worked too.
P.P.S: Also it still could have something to do with i915 and panel self refresh.
Indeed, I’ve just hacked-in kernel 6.8 on my existing Kubuntu 22.04.4 LTS installation and now it’s able to resume from the sleep.
So we can be pretty sure that the issue is due to outdated kernel version.
Update 1:
Still, there is one regression. Once in a while laptop fails to wake from sleep, but in a different way.
I am pressing the keyboard, pressing the power button and nothing happens, as if the keyboard module is completely powered off or detached from the mainboard. Power button just keeps blinking its LED.
Please note that I am using s2Idle sleep, not a deep sleep.
Update 2:
For now, solved the resume issue by installing the 6.5.0-1025-oem kernel.
@nf117@DenverCoder9@Brendan@Kieran_Levin Are any of you having boot issues since the unsuccessful update? Since I last posted (and failed), I have had two boot loops/failures. Both times, the LED, power button, and fans turned on, but the screen stayed blank, and then it would power off and repeat this state. I fixed it the first time by resetting the mainboard. I do not have time to do that again, and it’s not really a solution.
I will try to report the error indicators more completely. I don’t know if this has anything to do with the failed firmware update. IIRC, I managed to update the USB-PD firmware but nothing else? Did any of you manage the firmware update since then? Has anyone been able to go from 3.04 to 3.08 on Linux?
I haven’t had any unsuccessful booting. And, I haven’t tried upgrading again. I’m still on 3.04 and it’s stable, and works well with Fedora 40. I’ll just stick to this. I’m not planning on the battery upgrade, or others for now. So, I’ll just stick with 3.04.
Unfortunately, I’m afflicted by the failing charging detection, so I do want to update. Considering just putting in a cheap nvme, installing windows, updating firmware, and swapping it back.
I’m still on 3.04. The Linux 3.08 updater always boot ubuntu directly on my laptop.
I’m having problems with Thunderbolt 4 devices. Docking station and external NVMe does not work. Still waiting for an answer from the team about what I can do
So far I haven’t had any loops/failures after trying the update.
Maybe only some parte of the updates worked. You can compare the versions with my original values
Ran the update (from 3.06 to 3.08) as indicated in the top post. Unfortunately, I am assuming it failed, each reboot reports 3.06 as the current firmware but when I attempt to run subsequent updates, gives me an error “Full FW Update using the same version is not allowed” (probably because it thinks it’s 3.08 even though everywhere else is reading 3.06).
Update ran fine first run, no issues. Ran retimers update after, all ran through fine without issue. Tried using backing out with the esc key and running startup.nsh with the -allowsv string and it fails. Not sure where to go from here.
Couple of days after updating to 3.08 I’ve caught the following glitch:
System was in a s2idle sleep state, power button was “breathing” normally.
I woke the system up by opening the lid.
System woke up almost normally, except I’ve noticed that the power button LED is still breathing as if it was still sleeping.
Then I noticed that Fn key is not working. I wasn’t able to adjust volume or screen brightness as if the button itself was broken
Finally, I’ve noticed that the ACPI1 metric is reporting temperature of 175°C which was nonsense, since the cooler was almost silent and the laptop was suspended for more than an hour before that.
After hibernating the system (to disk) and resuming, everything went normal again.
All combined makes me think that my EC was stuck in some weird state, and the power cycle fixed it somehow.
Are you sure, Framework is affected?
I have not found anything relating this CVE to Insyde, also it is not listed in their security advisories: Insyde's Security Pledge | Insyde Software
I’m pretty confident Framework isn’t affected. The article is regarding Phoenix built UEFI and not Insyde. The point of sharing was to illustrate how important firmware updates are. And how tardy Framework can be in addressing in these issues at times.
Yes, i did use the earlier efi update first, and when noticing csme not being updated (and apparently the retimer for 1 side) i ran the newer update, the efi still.many many reboots, and thinkering with the scripts myself to figure out whats wrong. Eventually it seems a normal updater on the left side usb, with a writable and large efi boot on my nvme was able to update. Mind you, it still had my windows and arch install on there. Im not using grub and allowed whatever boot edits it wanted to make. No secure boot.
I might have triggered the updates by hand without noticing succes at some times. But eventually all parts where updated.
Then I noticed that Fn key is not working. I wasn’t able to adjust volume or screen brightness as if the button itself was broken
Finally, I’ve noticed that the ACPI1 metric is reporting temperature of 175°C which was nonsense, since the cooler was almost silent and the laptop was suspended for more than an hour before that.
After hibernating the system (to disk) and resuming, everything went normal again.
This has been happening to me too—I’m not sure what causes it but occasionally the function keys are messed up with no way to fix, and I get the same wildly inaccurate temperature reading. Reboot seems to fix for me too.
Are any of you having boot issues since the unsuccessful update? Since I last posted (and failed), I have had two boot loops/failures. Both times, the LED, power button, and fans turned on, but the screen stayed blank, and then it would power off and repeat this state. I fixed it the first time by resetting the mainboard. I do not have time to do that again, and it’s not really a solution.
Yes, I am having the same boot loop issue. For me, it happens if I close the laptop lid and it goes to sleep—then it becomes impossible for me to hard power off, reboot, or anything, without opening up the machine and resetting the mainboard state. I have started carrying around a screwdriver with me, the first time this happened was at a conference and I had to take a taxi to a hardware store to get a screwdriver, so that I could have a working laptop at the conference!
I noticed it also happened once when I didn’t close the laptop lid. something about going to sleep causes it but I’m not sure what.
Moral of the story is I seem to have a pretty broken system now, and I don’t even have a way to fix it, because every attempt at the updater scripts I have tried has failed, and I don’t have windows on my machine. This is a pretty frustrating situation to say the least! Frequently unplugging and replugging the battery connector to reset the mainboard state is making me really nervous, because the guide says there is a high risk of bending the pins.
Has anyone been able to go from 3.04 to 3.08 on Linux?
Has anyone been able to go from 3.04 to 3.08 on Linux?
It seems like the answer is no?
Well, apparently it had happened… But all these last message for the past week are surely preventing me to do anything before I have more information (and yet, this is the answer #562 on this topic )
)