-
Notifications
You must be signed in to change notification settings - Fork 4
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
1.32 BIOS update for t14s possibly causes the laptop to be bricked #129
Comments
Can you try disconnecting all the batteries (there might be multiple internal/external ones) and waiting 30 minutes, then reconnecting the batteries? @mrhpearson might have some more ideas. So sorry to hear this, I'm really disappointed this happened. |
Sure, I'll give it a try, thanks. I disconnected the CMOS and the main internal battery (I don't see any others) |
Unfortunately nothing seems to have changed, both having the battery disconnected (and on AC power) and having it reconnected. |
Really sorry to hear that - and like Richard very disappointed too. Richard - there seem to be a lot of issues around this update for the AMD platforms. I'm seeing failed updates (albeit not bricks) raised on the forum. Should we pull this to prevent users hitting the issue while I work with the firmware team to understand what's going on? Note - I tested the update myself on the T14 AMD with no issues - but there's obviously something else going wrong. @espidev - please contact Lenovo support about your system in the meantime. |
I will, thanks! |
Do you have secure boot disabled? Thx. |
FWIW I've had a similar experience on my T14 AMD (see #74 (comment)), but thankfully without the "stretched text" and/or bricking part. 😅 I contacted Lenovo Premier support when the hang happened since I didn't want to interrupt a potentially still running upgrade, and they recommended to restart it in the following way:
Your situation might be entirely different and I'm sure Lenovo support can give you a better answer for your case, but perhaps something worth trying anyways? |
Not sure if you were replying to me, but I do have it disabled.
Thanks! Doesn't seem to work for me though :c I've filed a support ticket with Lenovo already, I guess I'll have to see if the depot can get it fixed |
@espidev: My question was meant for @mrhpearson :-) |
I'm having issues with this update, too. My P14s Gen 1 (AMD) is not bricked, but it's not updating either. In may case I saw:
But I don't dare to try again after reading this here. IMO it would be super helpful if:
(Take what you think is helpful and ignore the rest.) |
FWIW: I installed the update using a bootable stick (geteltorito workflow) and that worked with secure boot disabled (so at least for p14s it's not an incompatibility issue per se). The description inside the official image is somewhat more comprehensive (it also contains the line i quoted in my previous post; didn't realize it was from there) and explains that before resetting to defaults, the "Enable OS specific defaults" option must be enabled. I am not sure that I had that box checked before trying the fwupd update. So that might be the difference between people who successfully update and those who don't. Also: The screen stayed dark for quite some time for me as well; so long, that I got worried. but then the fans spun up and some sort of status screen with progress info appeared. The OP reported stretched words and waiting for ~30 min. I neither saw stretched text, not did it take 30 minutes before that progress screen appeared. more like two or three. whatever went wrong there, from what I saw during the update and from what the OP wrote, the problem likely occurred long before the forced reboot. |
This is interesting news, thank you, which is not obvious from the notes provided by Lenovo via LVFS. Incidentally, after 10 years with DELL, where I never had any problems with firmware updates on Linux machines, I am really disappointed with the crap that Lenovo's firmware team provides. Given the issues and complaints, e.g., discussed at https://forums.lenovo.com/t5/Other-Linux-Discussions/bd-p/Special_Interest_Linux, it seems that they are completely overwhelmed. |
I wish I had read this report before attempting the update. In the end I think I got lucky but I'm not entirely sure. Here's how the update worked out for me on my T14s AMD (secure boot and OS defaults disabled in BIOS):
Finally, the machine seems to work normal again. Windows has no non-functional devices anymore. With a Linux system on a USB stick I managed to restore the UEFI boot entry and Linux now runs again as well. All in all a quite frightening experience. I am also not entirely sure in what state the machine now is. Are all components properly updated? Did the BIOS backup actually succeed after dozens of attempts or will the next broken BIOS finally brick the device? I highly recommend pulling this update until those issues are resolved. |
Oh boy, I just got bit by this on a T14s G1 AMD (MT 20UH). Updated from 0.1.25 to 0.1.32 yesterday --- unusually, that update was entirely uneventful (modulo some stuffing around to get it to start --- fwupdtool helpfully claims EFULL on a path on the EFISP, if, say, efivarfs was full because pstore had filled it). All seemed okay until a suspend/resume cycle today caused my display panel to cycle through its self-test (red, green, blue, black, grey, white, 5x5 black/white checkerboard) --- the rest of the system was still running happily --- and this self-test didn't go away across suspend/resume, a hibernate/resume cycle, a full power cycle ... continued while the system was powered off (albeit not backlit) ... but went away while reseating the LCD connector until the system was powered on again. Very terrifying, especially five minutes before a meeting. @The-Compiler's reset-five-times trick seems to have brought my laptop back to life. I'm firmly in favour of yanking the update. I'll be annoying my Lenovo support rep later today too. (Update: I've spoken to a Lenovo support rep who claims that, because I have recovered from the issue and have already run diagnostics that don't show any issues, and because they haven't ever heard of any issues with this BIOS update, there's no issue and the BIOS update is perfectly fine.) |
You and some of the other posters with bricked or otherwise problematic updates could open a thread describing your symptoms here: https://forums.lenovo.com/t5/Other-Linux-Discussions/bd-p/Special_Interest_Linux I have seen Lenovo reps respond to problems voiced on that forum. This forum is also what lead me to the update in the first place (I was experiencing various problems with the machine that this update fixes). In any case, what you folks describe seems to be less of a fwupd problem and rather a Lenovo problem. There seems to be the additional problem that the fwupd process fails for some people (myself included), while working for others. |
Thanks. Mark RH Pearson is Lenovo's lead technical engineer for the linux PC team (https://forums.lenovo.com/user/viewprofilepage/user-id/1942528). According to his today's comment on bricked systems due to the update to FW 1.32 at https://forums.lenovo.com/t5/Other-Linux-Discussions/T14-AMD-battery-drain-in-standby-Linux/m-p/5037674?page=37, he states that "I have had a few other reports of the same and we're looking at those with urgency but so far we've been unable to reproduce." Maybe those of you who are hit by this issue should leave a comment there. |
Thanks for the note above @fkru. As a note, I'm personally happy to keep the firmware update issues here - the Lenovo forum tends to go into power management, camera, tsc, audio issues and can take a bit of wading through. Here I get to focus on just the one issue which helps. I'm trying to go through fwupd and Lenovo forums and make sense of the reports. Having secure boot enabled seems to be a key factor. I've seen lots of 'black screen need pin hole hard resets to recover ' cases when secure boot is enabled - do we have any of those with secure boot disabled? I'm not sure for the two bricked systems I'm aware of if SB was on or not (maybe it's three systems - but it does get confusing with all the cross-posting) This issue has been escalated to the firmware team, with a summary of all the cases above and those from ticket #79 - We're very aware of the issue. As noted in my above forum post though - we've not been able reproduce it ourselves. I'm going to give it another go shortly...and slightly nervously. Mark |
@mrhpearson: Thanks, Mark, for the speedy reply and the clarification. I appreciate it. |
In my case secure boot was disabled as far as I can remember. However, issued from Linux via fwupd I just got the "black screen, no update" issue. The hickups started when trying to update from Windows as described in #129 (comment) Adding an afterthought: It could of course be that the first attempt via fwupd already brought the machine into some weird state, causing the subsequent attempt from Windows to make things even worse. |
Failed sys update from 1.29 to 1.32 for Post-reboot report as follows:
After uploading the report it stated
a) Is this the case? b) Also note |
As an update the issue has apparently been fixed and you shouldn't see the same in the 1.34 update that has been released. It worked in my testing. Recommend disabling secure boot during updates until the secure boot issue that was introduced in shim is resolved Any objections if I close the ticket? |
Regarding the shim issue, is there any solution in sight? Thx. |
Fix for shim has been delivered: rhboot/shim@87fbf42 How it gets rolled out in your distro of choice may vary. If it's useful the Ubuntu notes are here: https://bugs.launchpad.net/ubuntu/+source/shim/+bug/1929471 Mark |
I am puzzled, this fix is part of shim 15.3 released on 23 March. I am with shim 15.4 and firmware update via lvfs stopped working some time ago when secure boot is enabled. So, the culprit is something else. |
As far as I can see, that commit was never merged and the corresponding pull request was closed. The patch that was actually merged seems to be this one: rhboot/shim@4d64389 That patch is pretty recent and not included in 15.3 or 15.4. A good place to track the current shim issues seems to be rhboot/shim-review#165. It also looks to me like that particular shim issue is not related to secure boot, so that would explain why the update also fails for people with secure boot disabled. Maybe it all just depends on what shim version people have installed. |
- replace with binaries rebuilt with multiple upstream fixes + address fwupd/firmware-lenovo#129
Hi Guys, I had this same issue with T14. I had ugraded the memory and done updates etc and noticed the screen would go black and only power light would stay on. I took out the installed memory and the unit powered on fine and BIOS update continued and was able to successfully install. I then put the memory back in and the unit would again not boot black screen with power light. |
So the T14 worked with the added memory and the old BIOS, but doesn't work with the added memory and the new BIOS? |
For the memory issue - I think that's likely unrelated to this original ticket. I have a ticket open with the FW team about an issue with the T14 G3 with user installed memory after BIOS version 1.05 (for my reference - LO-2111). We've been unable to reproduce in-house - it seems to be very memory independent. I've been working with a customer and the FW team testing trial BIOS images to try and figure out what change caused the problem. It's quite a slow task as it relies on time from both the customer to test and the FW team to generate trial BIOS images - we're narrowing down on the likely culprit but not there yet. If you think that's the issue you're seeing and you're willing to help debug let me know and we can communicate off thread. |
Thanks, I was wondering if the device would boot without memory. It took some time for me to remember that some of the RAM was soldered directly to the mainboard.
I can only answer for my P14s Gen 1. I took out the after market memory (Teamgroup, 32GB if it matters). Let the procedure finish, go to re BIOS setup screen and configure my desired state again. Shut down, reboot and not its working fine again. TPM-based FDE is broken and I need to install one of the OSes, but I was prepared for such cases. |
I updated my laptop yesterday (on Arch, through Discover's fwupd backend) and after a reboot it seems to be bricked.
On the first reboot, it showed a black screen for a long time, and then showed some slightly stretched words in white text on black (that I unfortunately can't remember). I do remember that it did not say anything along the lines of doing background upgrades, and it just froze like that for about half an hour. I then decided to force reboot, and now the screen is just always black, with the power button glowing on.
I've tried to reset power through the reset pin hole, and also disconnecting the CMOS battery, but nothing seems to have helped.
It's possible I may have done something wrong? The update has been out for 2 weeks and I haven't see any other complaints...
Description I remember seeing in Discover:
https://download.lenovo.com/pccbbs/mobiles/r1cuj63w.txt
The text was updated successfully, but these errors were encountered: