Unreliable/slow boot, non-functional ports, incomplete shutdown,

Community Support
, ,
1

I already contacted support about this but I haven’t heard back so I figured I’d make a post here, I don’t remember the exact series of events but essentially:

Laptop failed to resume from S2RAM after closing the lid;
a couple days later laptop failed to resume from S4 hibernate and attempting to boot led to a black screen with zero indication of boot process;
multiple attempted boots later (holding the power button to force shut down between each) the side LEDs indicated all hardware checks passed (12 green blinks, My Framework Laptop (Intel 12th Gen Intel® Core™) is not powering on ) and only 4 blinks of the final byte flashed before the framework logo appeared and systemd-boot started, so I don’t know what the actual POST code was;
after briefly using it and letting it hibernate, it failed to resume again and when I manually powered it off and booted to similar effect as before the battery was nearly dead, presumably because it never successfully hibernated;
shutting it down to troubleshoot it failed to shutdown (showing a shutdown filesystem unmount error and usb device enumeration errors) before falling to a black, powered screen, with a lit power button;
after disabling hibernate and booting again I got the same 12 green blinks and the code 11111110 0xFE which according to Framework Laptop port80 codes · GitHub corresponds to “POST_BDS_JUMP_BOOT_SECTOR 0xFE // Try to Boot with INT 19”;
with no idea what this meant I proceeded to reset the mainboard following the official guide (I guess new users can only have 2 links in a post >-> );
booting again then gave 12 green lights and the code (if I recall correctly) 10000010 0x82 which from the same gist corresponds to “PEI_SWITCH_STACK 0x82 // Start to use Memory”, unfortunately I don’t have any other SODIMMs to test with the eliminate the possibility of a memory issue, and the system failed to boot;
after another force shut down and boot the framework logo flashed and systemd-boot started at ~7 of 12 green lights and the system continues to boot like so before any POST codes are displayed;
at some point during this process all 4 ports became entirely non-functional aside from charging;
shutdown still fails with a powered screen and lit power button before forcing the system to power off.

The only thing I can imagine causing an issue would have been occasionally running short (~30m) blender renders over the past few days. They had the CPU pinned at 100% but I imagine it should have throttled before it did any damage to the system. :confused:

At this point I imagine I’m going to end up buying a new mainboard as they’re cheaper than a new laptop, but if anyone has troubleshooting advice to try beforehand, or any idea what could have happened I would more than open to hearing it.

System info:
Laptop 13 Intel 12th gen i7-1260p
16GB 2x8 RAM
NixOS 25.11 (kernel 6.12.44) / rarely used Windows 11 dual boot
BIOS 3.5 (I tried to update a few months ago but that produced an error I don’t remember)

2

Support will be the best resource for sure.

Though, one thing to remember is when the board needs to be reset (completely battery drain, EC lockup, etc.) sometimes the NVRAM in the EC does not reset to a good state.

After clearing the mainboard state, go into the BIOS and load the defaults. Reboot, then power down. Disconnect the power and wait for 60 seconds for the EC to fully shutdown.

You might look on the left side of the mainboard to see if there are any spots on the left that have cooked themselves. These are just some things that are sticking out in my head right now as you are describing what is going on.

Someone else posted recently that their 12th Gen mainboard managed to cook some parts of itself and basically cooked itself. The owner said they took it to a repair shop and they said there was too much damage to different areas of the board that it would not be worth fixing.

On the brighter side of things the newest AMD AI boards are incredibly fast, even the lower spec board is considerably faster than the i7 board in yours not to mention even more power efficient.

3

I had seen that thread as well yeah, when I have the time I’ll pop it open again and take a look. Curious about the BIOS needing to reload the defaults, I’ll try that first

That’s good to hear about the mainboards though, I was looking at the R5/7 boards but if the newer ones are better for the same price I might consider those. And I’m definitely going AMD, the power consumption and battery life on this board has easily been one of the worst parts.

4

@kk-slider When did you write in?

5

Sep. 8th 7:30PM PDT

6

Cool. If you don’t see a response today / tomorrow, shoot me a modmail with the email address you used and I’ll make sure we received it correctly.

7

Unfortunate updates: support has not helped and only provided a novel suggestion after multiple back and forth emails providing information which was in my initial message; resetting the board multiple times has not helped, including removing most components and completely power-draining the board as support suggested; there were no obvious hardware failures when I looked at the board; doing a BIOS update with LVFS succeeded, but did not fix any issues and made the boot times significantly worse (potentially because some part of the update failed, but I couldn’t tell because it froze multiple times); POST codes are still giving inscrutable errors, including ones that don’t exist in the list I linked earlier; and USB devices are still not readable.

When I try to shut down I get systemd errors usb usb3-port10: unable to enumerate device(also occurs with port9) which lead me to believe there was some kind of USB hardware failure alongside whatever else occurred. Though unsure what to make of this. After a minute or two of systemd attempting to shutdown the screen turns off but the power button remains lit and the fans spin up.

Wild speculation, but I looked at the block diagram for the 12th intel boards and it looks like USB is handled directly by the CPU, which then leads me to believe something failed in the CPU. Maybe it overheated? There were a few things like loading certain websites which would make the CPU thermal throttle so maybe the cooler wasn’t adequately seated? The blender renders I had been doing the days before all these issues occurred may have just pushed it too far >->

Either way I just ended up buying a Ryzen 5 board that I’m gonna install tonight, and I guess I’ll watch the thermals closer. Though ultimately I have not had a very nice experience with this laptop.

8

For anyone coming behind, I had the exact same experience.

i7-1260p, suddenly dead ports, failing to resume from sleep, very long (~10 minute) boot times, all good on LED hardware checks. Initially USB was still working fine.

Mine, though, started when I tried multi booting to play around with GUIX (Primary: NixOS). Suddenly I had metadata corruptions on my BTRFS home partition and upon recovering from backups and removing the GUIX install I started to have significant problems like in the original post.

Support was generally unhelpful, and following their instructions potentially caused further damage to my board. Namely:

—
For troubleshooting, could you please try to do the following:

  • You mentioned that you removed all expansion cards. Does this happen after the laptop successfully boots? Could you please boot the laptop with all expansion cards removed? Once the laptop is stable, can you insert one at a time? This allows us to identify which expansion cards may be causing the issue.

  • We just want some clarification. Did the laptop boots up when the RAM is installed in channel 1?

Could you please try resetting the mainboard?
Turn off the laptop
Remove all expansion cards and the Input Cover
Then, disconnect or remove only the RAM, the SSD, and the battery cable (no need to remove the entire battery).
Leave the removed components for 15 minutes.
Reinstall or reinsert the RAM, the SSD, and the battery cable.
Plug the laptop into the charger.
Press the chassis open switch in the center of the mainboard 10 times. The case switch is usually located near the upper left memory channel.

alt image
alt image1795×827 416 KB

NOTE: Press slowly, holding for 2 seconds, then release and wait for the red blink on the mainboard LEDs.You should see flashing red LEDs on both sides when the chassis open switch is not pressed.

While the input cover is removed, peel off the black stickers (marked in the provided photo) and take clear, close-up pictures (with sufficient background lighting) of the components underneath them. Also, take a clear photo showing the entire mainboard with and without protective stickers. We want to check for anything outside the standard or out of spec. We provided a picture below for your reference.

alt image
alt image1040×774 318 KB

  • Reinstall the input cover and press the power button to boot the system.

Let us know how it goes
— </support email snippet>

After following these steps my USB ports no longer worked. Upon my notifying them that my usb ports no longer work they suggested that I try flashing the latest BIOS again with a thumb drive. :person_facepalming: This is after covering that I had already updated to the latest firmware via fwupdmgr per the official process. It’s totally possible that they’ve seen this failure mode before and assume that my USB ports were about to go anyway but it happening in the middle of this troubleshooting session stings a bit.

They basically told me “wow that’s too bad, it’s out of warranty though so good luck.” Allegedly someone from engineering is supposed to be reaching out. I’m not holding my breath but hoping aggregating some experience with this issue will help others in the future.

In the meantime I ordered the cheapest mainboard they had on hand that would work with my wifi card. Hopefully that holds out through the rest of my grad school term.