intel-ucode 20220207-1 , freeze whole system ( very often ) on idle #58
Comments
|
I am also experiencing this exact issue. CPU : i7-12700KF |
|
Currently testing one of the boot parameters ( intel_idle.max_cstate=1 ) . On the Linux kernel where the problem is very common. However, after using this parameter, I have not yet seen an error. It tests relatively shortly. So too early for conclusions. Edit : After using this parameter, I have not yet experienced a freeze. |
|
Check your dmesg log for the word "microcode" and see if you are actually updating the microcode update. It is more likely the version that is being supplied by your BIOS. A microcode update for 06-97-02 is not yet included here. |
If Intel is not interested in providing updated microcode for Alder Lake processors to fix this, what are the next steps? Should those of us impacted by this defect just purchase something else if we want to run Linux? |
|
Thanks for the info - your dmesg info indicates that the microcode update that you have came from BIOS. When it needs to update, the log contains additional information. There were no ADL updates ready for posting at the last update (Feb 8). The update that you have is fairly recent. I have been looking at the BERT record posted above and I will be reviewing changes in more recent microcode updates to see if there is a change that may help. Do you have any messages similar to the above BERT dump that you could post? That would help me determine if it is the same issue, or something different. Also, I recommend looking at the board OEM to see if they have posted any BIOS updates. Issues like this could be from anywhere, at this point - BIOS, microcode update, OS, etc. It's always possible the OEM has found and fixed something as well. |
Ok, that's good to know.
Here is the relevant output from
My board is an ASUS PRIME Z690-A, and I check their support site for firmware updates daily. I'm currently on version 1304 from 2022/03/09 which at the time of writing is their latest release to my knowledge. |
|
Tom, There seems to be some issue with the log above. Part of the data seems to be missing. For what data is here, I don't see any error logged. |
|
Well that is all that is shown in This happened again yesterday so I can post it again: |
|
Tom, Thanks. I'll look at this dump and see if it's any different. It's always possible the failure that is happening is corrupting some data in crashlog. I'll take a closer look at the data. |
|
A firmware update was released for the ASUS Z690-A, version 1403, and I have applied it. This did update the microcode, but it's too soon to tell if it has resolved the problem. |
|
Thanks for the feedback. Let's hope that resolves the issue. |
|
I have update to from gigabyte. But there is nothing special on changelog.
so i don't think that, this resolve my problem. |
|
Just had another freeze after several days without a problem, so this still isn't fixed with microcode revision |
|
I too have this issue in an 11th gen CPU, I can open a new issue if its considered a different issue but I have the same record identifier in dmesg error. The machine will randomly lockup and I have to hold the power button. I also noticed a lockup for the first time in the BIOS menu. I usually encounter the lock up when not doing much like watching a youtube video or web browsing. I have seen this happen in windows a few times also. Can happen daily. Sometimes it only happens once per week. boot line: |
|
I've had this problem twice now on Windows. Same "symptoms". All components have been replaced except the board and processor. I'm returning the hardware and not going back to Intel. I paid money and time and feel cheated. |
|
@hestela, @tom-seewald : Each of these are slightly different. I recommend talking with the board vendors as there may be an issue with these boards. One other possibility is to make sure there is no contamination on the pads of the processor that could lead to contact issues. |
|
@hestela, @tom-seewald : have you resolved this issue? Just to check, since you are using K parts, are you overclocking? |
I am still experiencing total system hangs and I am not overclocking. I have a replacement 12700KF on order, to see if that resolves it. Overall this system has been the worst tech purchase I have ever made. Regarding reaching out to board vendors, there is no clear way for an individual to do that beyond talking to people following a script who are not able to help with advanced issues, especially when running non-Windows operating systems. |
|
same issue, same motherboard. Hardware : i7-172700k
|
|
Maybe it would be worth it to try to disable hyper-threading? It is a shot in the dark, but hey, it is one that is far easier than hardware replacement... If it fixes things, report back, and Intel will have some work to do to find out what's wrong... (BTW: I am not an Intel employee, and I do not speak for Intel) |
|
Same issue. 8-Core 11th Gen Intel Core i9-11900F |
|
updated to the latest bios for my motherboard (z690 gaming x bios version F8b) a few days ago and the issue has gotten worse. before it was random and would seem to occur every couple days. now it freezes everyday multiple times. I had to hard reset my computer about 3 times yesterday and twice today.
|
|
According to GunDoom's report, the newest release (20220510), which has microcode 0x1f, is also causing problems on these motherboards. |
|
Any of u have X.M.P enabled? |
|
when I turn X.M.P on my computer doesn't POST |
|
so after test ( disable X.M.P ) i have still the same issue. now just writing for a refund for motherboard and processor. I have already tested all possible options, and intel support is a joke. I caution people against buying this hardware. disabling c-state on the motherboard may help to obtain system stability. I'm not saying that the problem will go away, but it will definitely be less noticeable. The problem occurs: with low resource consumption: editing text files, browsing the Internet, listening to music, leaving the system for several minutes after logging in. The problem never occurred when playing games that use the CPU as the primary video recording source. Best regards and have fun. |
|
I've added the apm=off boot parameter and the error hadn't occurred in the past week. You can also try with acpi=off if that works for you |
I had same problem with my Linux machines (3 - more precisely). Leaving machine(s) alone for few minutes resulted in system freeze (no response - black screen). I assumed that situation was related to GPU (integrated). I've found that page https://linuxreviews.org/Intel_graphics (Troubleshooting section) and applied kernel boot parameters and I was able to fix hangs. My kernel bootargs that fixed problems: |
|
So this is cpu or mbo issue ? And why c-state doesn't work properly even on Windows ? Or maybe the components are faulty at the factory , but which one ? Intel Diagnostic Tool still doesn't support 12th gen... Anyone using NVME? |
IMHO it's a combination of HW/SW bugs (just guess - I'm not an Intel engieneer just simple user).
Maybe because it's a 12th gen it doesn't work on Windows ;-) |
|
I already know where the problem is. In my case it is an NVME drive which is either a design problem or a problem when configured with this board or a 12th generation processor. The disk "bites" when there is a power saving process through a mode such as C-STATE. Therefore, the best solution is to use it only to REDUCE the frequency of the error. Or completely disable the feature in the NVME drive itself. Probably called APST. I haven't checked the second option yet, but it might be a much more interesting solution when it comes to the power saving process itself in the whole PC ecosystem. As a workaround, add the [kernel parameter] I haven't fully checked everything yet, so I need some time to confirm unanimously. If anyone has the time and could describe in more detail the parameters of your own PC it would be nice. Regards. Update : So, in my case it was an NVME drive or rather a combination of hardware. I didn't have to disable the whole power saving module from the NVME drive. I disabled only one mode : PST4. Update The problem appeared after 2 days of use. Unfortunately, the parameter when booting the system was set to one energy saving mode, now I will test it to complete off. |
|
I also tried disabling C-state in my BIOS and adding more to my kernel boot args. Worked for about a day, next morning my machine froze and had to reboot it. I have a RTX 3070 so the i915 options won't help me much. |
|
Add this para to boot:
this is no true, i915 is still active in background. I mean even if u don't using it , this module isn't omitted, disabled at all. |
|
i915 driver is not loaded in my machine but I'll give it a try. |
|
when I add |
can u show full boot parameters ? anyway, i was testing everything, and forget about most important thing PSU. I check that my PSU support only C6/C7 Intel states. But on my motherboard and newest CPU have more states like C8/C9/C10 . So i just disabled three of them. I know that for someone even disabled full C-STATE doesn't help at all , but in my case might be that thing ( i hope so.. ) Edit : PSU work correct. |
here are my full boot parameters. Other than |
|
For me looks fine , make sure u have loaded intel module : @GunDoom can you tell me your RAM specific ( name, serial ) ? it's important . |
|
my ram is 32gb 2x16 corsair CMK32GX4M2D3600C18 ram |
like mine, your ram memory is not on the list of compatible ram memory . ( gigabyte z690 gaming x ddr4 ) . But these memory support intel 600 platform ( z690 chipset ) . So there is a shadow of chance that your ram memory is unstable on this motherboard . As for X.M.P, I see you have 2 memory modules, so make sure your module is in slots 2 and 4. This is a requirement for X.M.P mode to work properly on this motherboard. Then in bios select X.M.P profile and check if it works. As for the problem itself, the problem still occurs, I have the impression that less often than before. But it is still significant. I tested the ram memory for about 10 H and no error or hangs. The only thing left is to return the hardware to the seller's store. |
try only with apm=off |
apm=off works, acpi=off causes lightdm to fail. thank you.
my ram is in slot 2 and 4. X.M.P has never worked, system won't post and I have to clear CMOS when I enable it. I guess I'll have to look into getting some compatibility memory. |
|
update. issue has resolved itself on my machine. I haven't changed anything or done anything it just stopped freezing up one day. been about 3 weeks and not a single freeze. |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Hello, I would like to report an issue that crashes my entire system without being able to do anything with the mouse or keyboard. Of course everything literally crashes, USB, Screen, Sound. Even if I reset the system using the "RESET" button the computer cannot boot the bios. I have had this from the very beginning. The problem cannot be due to a physical device because everything works fine on Windows. The problem occurred even when I had ArchLinux installer on USB. To turn off the PC I have to hold down the "POWER" button and then turn on the PC again. The hardware never crashed when I was doing some work or under heavy load, on the contrary only when it was doing nothing and not much was going on in the background. I was able to find similar behavior in the logs when it comes to total system crashes, and I hope this is a problem related to the microcode of this package.
Hardware : i5-12600k
OS : Archlinux
Kernel : 5.16.15 also 5.15.x also 5.17
Ram : Dual Channel 16gb 3600Mhz
MBO : Gigabyte Z690 Gaming x DDR4
Error :
The text was updated successfully, but these errors were encountered: