I’d like a fix for the issue where, under vaguely moderate load, my laptop will throttle all CPU cores down to 400MHz and hold them there for 5-15 minutes, even though core temperatures are well below the danger zone within seconds of throttling. Since this happens on multiple OSes, I’ve seen it on two different mainboards, and others have experienced it as well, this is most likely a firmware issue (possibly the EC, though, and maybe not the UEFI bits).
If 3.08 (well, 3.06) had gotten the attention it needed over a year ago, I would hope that it would have been released long ago, and Framework could have been working on serious bugs like mine, rather than… whatever it is they’ve been doing.
My 12th gen had the issue where all cores would throttle down to 400MHz. Ultimately, I had to open the laptop and reapply the thermal paste on the cpu. Not ideal, but my CPU hasn’t been locked at 400 MHz since then.
Tried it, twice, on two different boards, didn’t help. (I’ve been having an intermittent back and forth with Framework support for quite a long time now… believe me when I say we’ve tried a lot of things.)
May I ask what thermal paste brand/model you used? Maybe that’s significant…
Edit: er, after posting I realize this is super OT for this thread. Sorry, y’all. @BigBoat, feel free to DM me or reply over on my thread about this.
My partner uses my old 12th gen i5 board. Just installed the the 3.08 update via the tool on Framework’s site. Completed without a hitch, and the system reports the bios as now being 3.08, it had come from the 3.05 beta previously.
EDIT: To clarify, my partner is running Windows 11.
I am sorry but I am going to join the chorus of people here and say that I’m deeply saddened by this state of affairs.
I’m one of the early adopters of the 3.06 beta BIOS update. I installed it, from Linux, and it worked, it was great. I was expecting this to come out of beta months ago. Now there’s a 3.08b available, and not only is it not available on LVFS/Linux, even the EFI option has been pulled.
Now, it seems the framework team has moved on to other things, and I’m left holding the bag, with a basically backdoored BIOS.
This is, quite plainly, unacceptable. I’ve been recommending Framework left and right, thinking you folks had our backs, but now it seems we (the free software community) are left holding the bag and will have to reverse engineer your stuff to get updates delivered properly.
For me, this clearly shows that Linux is a second-class citizen for Framework. This would never happen to the Windows or Chrome sides of the business. Can you imagine Framework coming out “oh yeah, poor windows users sorry you’ll have to boot a Linux OS to install that update?” Nevermind than that would actually be possible for windows users: Live linux distributions are actually a thing, and quite popular tools even for windows users to recover their mess.
In fact, why isn’t this the way updates are delivered? Linux is easy to embed in a thumb drive to deliver updates…
Anyways, having some word about a timeline on Linux support for the BIOS from the framework team would be much appreciated here. Maybe @Matt_Hartley knows?
This is the MSI installer only being “updated”. supposedly the firmware itself did not.
the EFI updater is still on the server. I can only agree on the bios should become release at some point. (and when to expect this “stable” efi variant, altho the efi version worked fine for me)
@Ray519 At this point with 3.08b released on the update page, the only major tripping point (aside from the lack of information during updates, no detailed info for aborted updates, etc.) is the ReTimer updates, correct? And that can be resolved by disconnecting the battery in BIOS for a few minutes and powering back on, right? So at this time we should be ok to run the update, assuming we’re ok with potentially needing to disconnect the battery in BIOS? Or was there something bad about the GOP driver that’s still worth holding off for?
From a stability point of view of the update once installed, I am aware of nothing. I concluded as much in February when I installed the beta on my mothers Framework assuming that Framework would be dragging their feet again (and after skipping the 3.06 beta, because that had issues with the USB legacy adapters).
The GOP thing and many other things I have commented since April are just about the way they describe or not describe things (like listing 2 different versions as same in the release notes, and in 3 months and for a release not having noticed this contradiction. So nothing directly about the quality of the update itself.
My main concern has long more been the way it has been handled. How if Framework had been getting private reports or I would have missed a user reporting another issue of the update here, they might have ignored them just the same as the other reports here. And what that says about FWs processes and ability to deliver updates and us being able to trust their PR-driven “opinion” that the update is ready. Especially if they intend to just ramp up speed without showing that they are making vast changes to prevent releasing the WRONG version ever again (and not noticing for months).
I think 3.08 itself, is probably of the same or higher quality as the factory version. And there is no solid testing, so there are probably still rare issues in there. But I don’t think the chance of that is higher than such issues being in the factory firmware from the same people or you hitting some of the known issues that are fixed with 3.08. And I do not see the quality and reliableness of the updates improving in the short term, so think 3.08 is still the better way to go.
Excellent, I appreciate the info! It’s funny, across here, Reddit, and the Ars Technica article comments, you’ve been the one keeping level-headed and providing good, solid info. I really appreciate that.
I try to keep the anger down when writing things and remain fair and also acknowledge positive developments etc. In the hopes of getting Framework to change their stance and rethink things. And getting me the products I would like to have. Although I have been getting a bit more snarky over time.
Maybe we get lucky and the same or other authors actually pick up on the 3.08b release situation and how FW is again trying to keep that quiet.
Maybe that gets us even more transparency and hope for the future.
However, now with Linux, entering “deep” suspend (in the /sys/power/mem_sleep sense) consistently soft bricks the device. Once in this state, a mainboard reset is required.
Funny because I updated the bios last week to 3.06beta, and the issue you report is exactly what lead me to this thread. The poweroff is not working either.
I’ll contact support. My 2 rear USB ports are dead (except for charging) and this suspend issue are a blocker for me.
OK, I tried the update to 3.08, and it completely broke my motherboard. The 4 usb ports are not visible even in the BIOS (boot selector) anymore. Contacting support.
Thank you for all of your feedback and we are working on a stable Linux update method via EFI (among other improvements you all have noted in the thread for our next release) as quickly and as high of quality as we can. We also read all your feedback (I have personally read every post here and tried to reproduce many of the issues!), so keep it coming!
We know words are cheap, and you will only believe us once the work is completed and delivered to all of you.
In the meantime, some more context with the missing EFI updater on the current 11th and 12th Gen BIOS releases (3.19 and 3.08 respectively) we thought would be helpful.
The reason we haven’t released the EFI update is that the EFI updater is missing functionality to update the CSME region to the latest version which includes security updates from Intel. The EFI updater we currently use is based on work from Framework, which did not originally account for the possibility of CSME updates. We’re currently developing a new update script, which allows us to update this region going forward.
Again, we don’t want to make promises or empty words in forum posts, but we are working to improve the communication frequency and look forward to deliver more updates for all of you over time.
In the meantime, when you have an issue it will help us move significantly faster on our side to reproduce and work with our partners if you can use the bug report template (as much information as possible helps us set up repro cases to work towards a root cause and solution!)
We very much need the following to track down where things are failing OR succeeding (as much information as you can, especially the reproduction steps)
SYS CONFIG: (i7, i5, specific model details, get the details from the motherboard sticker, same location as above)
RAM: Brand and how much, 1 or 2 sticks
SSD: Brand/model and how large is the capacity. If removed, please indicate.
Wi-Fi: Which wifi card? If removed, please indicate.
External Devices/Other: Anything attached? If so, what and how? If not, please indicate.
EXPANSION CARD TYPES: What cards were inserted? Example: 2x Display Port, 2x USB-A
BIOS VERSION: Which BIOS version were on on before attempting the flash?
DRIVER PACKAGE VERSION: If known and if using Windows.
OS VERSION: if Windows, 10 or 11? If Linux, which distro and release version? For Arch, this would be Arch, fully updated or not, please indicate.
FAIL RATIO: Can you reliably reproduce this issue 100% of the time? Is it only 50% of the time? Occurs randomly?
STEP TO REPRODUCE: EFI Shell via USB drive, Windows package, and the steps you recall taking. Just do your best. I realize no one will remember all of this.
Step 1 -
Step 2 -
Step 3 -
Etc, Etc.
OBSERVED RESULT: How it failed, what you saw on the screen.
EXPECTED RESULT: Obviously this one, we expect it to flash successfully and be reflected in BIOS settings.
ISSUE RECOVERY METHOD: If you were able to recover, walk us through this process.
EXTERNAL DEVICE MODE or NAME: For USB flash drives, Generic or the brand.
Sorry to stumble upon the flowers on the carpet, as we say in french, but did you really mean the 12th gen 3.19 update? I thought that was 3.08?
Also, you seem to be saying we can test the update now, is there an actual way to do the 3.08 upgrade now on Linux? (And no, install windows is not a way.)
Also, it seems relevant to mention there was a whole blog post about this very topic on the website now:
I found it’s pretty thin on actual contents. The only new thing in there is:
for Linux we’re developing a new updater to handle the specific firmware regions involved.
which, quite frankly, is actually pretty disappointing. I did the 3.06b update on Debian Linux through LVFS and it Just Worked. I understand this update is trickier, but I find it regrettable that we’re building a whole new way to update firmware instead of leveraging what the community has been working really hard on for years at this point, and that has been adopted by many large, much more commercial/proprietary vendors than Framework, including Dell, HP, Lenovo…
Really, I would love to hear more about what the actual problem is with LVFS and whether or not the LVFS people have to say about this… This really sounds an awful lot like what Purism and System76 ended up with, which is that they’ve basically given up on LVFS. I didn’t buy it from them, and I don’t buy it from Framework either.
Sorry to stumble upon the flowers on the carpet, as we say in french, but did you really mean the 12th gen 3.19 update? I thought that was 3.08?
Hey @anarcat - good catch, that was a typo and I modified it in the above post.
you seem to be saying we can test the update now, is there an actual way to do the 3.08 upgrade now on Linux?
Unfortunately, the way I have worked around it is using the Windows BIOS installer.
Yes, installing Windows, running the BIOS update and then wiping the Windows partition afterwards…not an acceptable solution as you mentioned, but it got me there in the meantime.
I’m also working with @Kieran_Levin to see if we have another solution we can write a guide on (non-Windows) in the meantime, no promises, but I will keep everyone posted.
Edit: PS - I love this saying, have never heard it as an American, may be using it more in my own day to day
not to go far off topic here, but yeah, americans do love it, and seem to have rarely heard it. It’s pretty colloquial (in french!) where I’m from, but maybe not in France…
Same here. @pixelforest don’t reinvent the wheel, or have a case of “not designed here” let’s design something that in the long term is just one more thing we will support until we have to drop it…mir, upstart, unity same concept different projects.
