Sudden reboots on Asus TUF FA617NS

KatTheFox · 2024-02-08 08:28:37

I've been having an issue for a while where semi-frequently, usually (but not exclusively) when playing Minecraft, my laptop will suddenly reboot. Normally I can figure stuff like this out on my own, but honestly I'm stumped.

There are no messages in the journal (checked with sudo journalctl -x -b 1) for a few minutes before the reboot; often the last message is 10 or 20 minutes prior to the crash, and is usually something along the lines of:

wpa_supplicant[693]: wlp6s0: CTRL-EVENT-SIGNAL-CHANGE above=0 signal=-76 noise=9999 txrate=288200

Similarly, when checking dmesg (with sudo journalctl -o short-precise -k -b -1) there are no messages within a few minutes of the crash, with the last ones being related to wlp6s0, which is my wireless card.

Before the crash I will sometimes experience general system slowness, despite (according to top) low system resources being used.

Things I've tried:
- Switching the gpu kernel module to amd_pstate_epp and back to default. Made no difference.
- Installing the linux-g14 custom kernel and userspace utilities from the Asus Linux toolset. Helpful in other ways, but did not prevent the crash.
- Installing the rtw89-dkms driver. No difference.
- Searching the forums, the wiki, and the internet as a whole to try and find similar issues. Since the problem is so undiagnosable, it's hard to narrow results.
- I did find a known issue with AMD Ryzen 5000 series cpus on the wiki involving random reboots, but I did not have the noted messages in my dmesg logs. I also was unable to apply the fix listed, as Asus's laptop uefi does not allow me to adjust the power curves (it's quite sparse on options, to my annoyance).
- Attempted to replicate the fix in userspace with cpupower. No effect.
- Disabling power-saving on my wireless card.
- Disabling c-state 6.
- Adding nomwait kernel parameter
- Switching to amd_pstate guided mode. Massively decreased performance, but did not fix anything.
- Lowering the maximum frequency of all cpu cores to about 3/4 of maximum.

The only common thread I've noticed is that it happens much more often when I'm playing Minecraft, which may be something to do with increased network loads? I haven't observed it happening when I'm playing something like Baldur's Gate 3, which puts a much more intense load on the cpu and gpu. It does also tend to run fine for a while, and then once it starts happening it keeps happening frequently, across reboots, until I stop using the laptop for a few hours (usually to go to sleep). It doesn't get very hot, so I don't think it's thermal-related; BG3 can get my system temp up to 90C (which is also something I need to fix) and it doesn't happen then.

It didn't happen when I first got this laptop, but I've made so many changes to the software environment since then that it's impossible to narrow down whether it's the fault of a specific package or an update to something, or a hardware issue.

Sorry for the lack of info, I'm not sure what to add here since I haven't been able to narrow down the cause at all.

The laptop is a stock Asus Tuf A16 Advantage Edition 2023 FA617NS, with an AMD Ryzen 7 7735HS cpu, AMD ATI Radeon RX 7600S gpu, and RTL8852BE wireless card. I'm running the linux-g14 kernel, although it happened when I was using the stock linux kernel as well, and all my packages are up to date.

I've attached the journal and dmesg from the most recent crash, in case there's something I'm not seeing. If there's any more information that would help, please let me know.

Last edited by KatTheFox (2024-02-09 23:42:48)

KatTheFox · 2024-02-09 04:15:42

It just happened again- after noting the time of the crash and checking the journal, the wlp6s0 message repeats many times immediately prior to the crash in the journal, although there's nothing for 8 minutes prior in the dmesg

seth · 2024-02-09 08:29:51

The network signal cycles between excellent and "it's dead, jim" but that does typically not cause hard reboots - the latter points towards the CPU (MCE or not), temperature (you asserted is not an issue) or PSU issues. That being said, wifi chips can draw some serious power and if the system runs on the edge, it might be the straw to break the camels back.

There're two BSSIDs for the same "corbettjlc" SSID, one's 2.4GHz and the other 5GHz
The 5GHz one isn't too god and the 2.4GHz has these drastic swings:

Feb 07 23:41:59 FoxTop3 wpa_supplicant[693]: wlp6s0: CTRL-EVENT-SIGNAL-CHANGE above=1 signal=-26 noise=9999 txrate=206500
Feb 07 23:42:04 FoxTop3 wpa_supplicant[693]: wlp6s0: CTRL-EVENT-SIGNAL-CHANGE above=1 signal=-26 noise=9999 txrate=206500
Feb 07 23:42:04 FoxTop3 wpa_supplicant[693]: wlp6s0: CTRL-EVENT-SIGNAL-CHANGE above=1 signal=-23 noise=9999 txrate=206500
Feb 07 23:42:12 FoxTop3 wpa_supplicant[693]: wlp6s0: CTRL-EVENT-SIGNAL-CHANGE above=1 signal=-25 noise=9999 txrate=206500
Feb 07 23:42:38 FoxTop3 wpa_supplicant[693]: wlp6s0: CTRL-EVENT-SIGNAL-CHANGE above=0 signal=-67 noise=9999 txrate=172000
Feb 07 23:42:41 FoxTop3 wpa_supplicant[693]: wlp6s0: CTRL-EVENT-SIGNAL-CHANGE above=1 signal=-38 noise=9999 txrate=154800
Feb 07 23:42:43 FoxTop3 wpa_supplicant[693]: wlp6s0: CTRL-EVENT-SIGNAL-CHANGE above=0 signal=-95 noise=9999 txrate=154800
Feb 07 23:42:45 FoxTop3 wpa_supplicant[693]: wlp6s0: CTRL-EVENT-SIGNAL-CHANGE above=1 signal=-42 noise=9999 txrate=154800
Feb 07 23:42:48 FoxTop3 wpa_supplicant[693]: wlp6s0: CTRL-EVENT-SIGNAL-CHANGE above=0 signal=-67 noise=9999 txrate=129900
Feb 07 23:42:51 FoxTop3 wpa_supplicant[693]: wlp6s0: CTRL-EVENT-SIGNAL-CHANGE above=1 signal=-47 noise=9999 txrate=129900

(This is all within a minute, I guess you're not running around that fast?)
Bluetooth interference? Leaking microwave?

There's an RJ45 NIC, can you reliably play minecraft on a wired connection?

KatTheFox · 2024-02-09 08:38:54

The laptop is stationary, far away from the kitchen and the only nearby bluetooth-enabled devices are my headset and phone. I don't have access to a wired connection, but I'll test if I can get access to one. It's also plugged in to the wall with the adapter that came with the laptop (20v 12a 240w output) so hopefully power shouldn't be an issue? At least, the battery isn't draining from what I can see.

seth · 2024-02-09 08:44:43

It's not that you're running out of battery but that the system momentarily demands more current/voltage than the PSU can supply.

rfkill the bluetooth
wrt "I don't have access to a wired connection":
yes, you do ("nearby … phone")

KatTheFox · 2024-02-09 08:47:37

Ah, fair- I forgot about tethering, let me see if I can set it up. I don't think I'll be running minecraft very well on my mobile data, though. Interesting that the system can potentially run out of power like that, I'd assume the laptop would be engineered to prevent that. Guess I shouldn't have assumed.

seth · 2024-02-09 08:52:29

Can't you just connect the phone to the WLAN?

KatTheFox · 2024-02-09 08:55:13

Yes, that works. I don't really use tethering so I wasn't aware it could forward the connection like that.

KatTheFox · 2024-02-09 09:10:39

Well after blocking Bluetooth with rfkill and connecting with tethering, I haven't experienced any crashes- although it has only been 10 minutes. It looks like I'll have to test more tomorrow as it's well past midnight and the wifi just went down (phone won't let me tether with data at all, actually, thanks carrier).

KatTheFox · 2024-02-09 10:33:29

Upon further testing, it was stable for about an hour before crashing again. The last lines in the journal were 4 minutes before the crash:

Feb 09 02:26:11 FoxTop3 NetworkManager[632]: <info>  [1707474371.4834] device (wlp6s0): set-hw-addr: set MAC address to 8E:3C:8D:01:BE:A7 (scanning)
Feb 09 02:26:11 FoxTop3 NetworkManager[632]: <info>  [1707474371.6055] device (wlp6s0): supplicant interface state: disconnected -> interface_disabled
Feb 09 02:26:11 FoxTop3 NetworkManager[632]: <info>  [1707474371.6055] device (p2p-dev-wlp6s0): supplicant management interface state: disconnected -> interface_disabled
Feb 09 02:26:11 FoxTop3 NetworkManager[632]: <info>  [1707474371.6058] device (wlp6s0): supplicant interface state: interface_disabled -> disconnected
Feb 09 02:26:11 FoxTop3 NetworkManager[632]: <info>  [1707474371.6058] device (p2p-dev-wlp6s0): supplicant management interface state: interface_disabled -> disconnected

Seems like NetworkManager is still doing stuff with the wireless card, maybe if I block it with rfkill it'll be different?

seth · 2024-02-09 13:56:39

Scanning 4 minutes before the crash won't have caused the crash. Look into the https://wiki.archlinux.org/title/Ryzen#Random_reboots situation.

KatTheFox · 2024-02-09 20:16:27

I did see that, but I wasn't able to replicate the fix thanks to the horribly limited bios options my laptop has - Asus apparently decided laptops don't need customization.

KatTheFox · 2024-02-09 20:20:27

To be clear, there are exactly 0 options related to CBP, PBO, DRAM Voltage, or anything adjacent. The only options I have are to toggle secure boot, virtualization support, some various usb/sata/nvme settings, and hardware diagnostics. The only fix listed there was the kernel parameter idle=nomwait, which I am testing now.

Last edited by KatTheFox (2024-02-09 20:23:28)

seth · 2024-02-09 22:02:33

That, like the c-state control is to prevent soft-locks when the CPU powers down to save energy.
The symptom for that is typically not a hard reboot.

You might have luck constraining the maximum CPU frequency, https://wiki.archlinux.org/title/CPU_fr … requencies
Edit: also https://wiki.archlinux.org/title/Underv … CPU#amdctl

Feb 07 23:39:48 FoxTop3 kernel: DMI: ASUSTeK COMPUTER INC. ASUS TUF Gaming A16 FA617NS_FA617NS/FA617NS, BIOS FA617NS.414 09/27/2023

Is there a newer BIOS available for the system?

Last edited by seth (2024-02-09 22:03:41)

KatTheFox · 2024-02-09 22:17:30

I just did a bios update recently- not sure why it says 9/27/2023. I'll look into it. I have no idea what to set the maximum cpu frequency to, but I'll do some research.

KatTheFox · 2024-02-09 22:18:09

The most recent bios available is version 414, which I have

KatTheFox · 2024-02-09 22:26:55

After setting the max frequency for all cores to 3/4 of maximum, it happened again

KatTheFox · 2024-02-11 05:25:06

Just verified it's not a hardware issue- after installing windows (ugh) on a secondary partition and running minecraft for hours, no reboots occurred

seth · 2024-02-11 08:59:25

As the wiki points out

After investigating and reading reports on the Internet, It seems that out of the box, Windows seems to run the CPUs at higher voltage and lower peak frequencies, compared to the stock linux kernel, which depending on your draw from the silicon lottery could cause a host of random application crashes or hardware errors that lead to reboots.

Did you use amdctl to downclock the CPU?
You could also try to lock the GPU in a lower power profile, https://wiki.archlinux.org/title/AMDGPU#Power_profiles
(Though typically the GPU would not result in a reboot, you're only indirectly leaving more current/voltage to the CPU this way)

reddit btw. has a truckload of windows and linux users complaining about spontanous reboots w/ TUF systems

(what kind of "gaming" notebook doesn't allow you to control the CPU and RAM performance settings? What's the "gaming" part there? Fancy plastic case? / rantaboutstupidmarketing)

KatTheFox · 2024-02-11 09:03:05

seth wrote:

what kind of "gaming" notebook doesn't allow you to control the CPU and RAM performance settings?

A very good question! I would not have gotten this model if I'd known that beforehand. I have not tried downclocking with amdctl, will do as soon as I next can.

KatTheFox · 2024-02-13 21:05:08

I've just found potentially another clue- while doing nothing overly stressful, my wifi connection died and my computer massively slowed down. All processes that were using the internet (discord, zoom [I was in a meeting], and firefox) apparently crashed, and some time after, the computer went back to normal. I wonder if that's what's happening, causing minecraft to crash in a way that reboots the whole computer.

KatTheFox · 2024-02-13 21:10:18

Journal and dmesg after the incident. The slowdown started around 12:50ish- a bit beforehand probably. I'm seeing a network disconnect and reconnect, and NetworkManager is attempting to do... something?

seth · 2024-02-13 21:36:34

Please never use "-x", it just adds spam.

The 5GHz signal remains "poor" and the 2.4GHz signal "all over the place"
The only two reconnects are

Feb 13 10:30:45 FoxTop3 NetworkManager[639]: <info>  [1707849045.4734] manager: NetworkManager state is now CONNECTED_GLOBAL
Feb 13 13:04:41 FoxTop3 NetworkManager[639]: <info>  [1707858281.8575] manager: NetworkManager state is now CONNECTED_GLOBAL

the rest is NM grasping for radio.

Is your phone still lying next to the system? What if you move that somewhere else?

However, while a network loss causing some client to crash is plausible, that client then somehow triggering a hard reboot is not all that realistic and you had the reboot w/ the tethered connection as well, did you?

Try to add

pcie_aspm=off rtw89_core.disable_ps_mode=Y rtw89_pci.disable_aspm_l1ss=Y rtw89_pci.disable_aspm_l1=Y

https://wiki.archlinux.org/title/Kernel_parameters
and see whether that stabilizes the signal
You can also add " rtw89_pci.disable_clkreq=Y" but I'm not sure whether that would rather make it better or worse.

If the signal becomes stable, see whether it remains w/o pcie_aspm=off - if yes, try to strip away other parameters.

KatTheFox · 2024-02-13 21:43:39

Can confirm, moving the phone away doesn't fix the connection. I forgot about the tethered crash- this is really quite confusing... I'll test with those kernel parameters when I get a chance.

KatTheFox · 2024-02-14 08:34:57

Still getting the signal change messages with all the kernel parameters

Arch Linux

#1 2024-02-08 08:28:37

Sudden reboots on Asus TUF FA617NS

#2 2024-02-09 04:15:42

Re: Sudden reboots on Asus TUF FA617NS

#3 2024-02-09 08:29:51

Re: Sudden reboots on Asus TUF FA617NS

#4 2024-02-09 08:38:54

Re: Sudden reboots on Asus TUF FA617NS

#5 2024-02-09 08:44:43

Re: Sudden reboots on Asus TUF FA617NS

#6 2024-02-09 08:47:37

Re: Sudden reboots on Asus TUF FA617NS

#7 2024-02-09 08:52:29

Re: Sudden reboots on Asus TUF FA617NS

#8 2024-02-09 08:55:13

Re: Sudden reboots on Asus TUF FA617NS

#9 2024-02-09 09:10:39

Re: Sudden reboots on Asus TUF FA617NS

#10 2024-02-09 10:33:29

Re: Sudden reboots on Asus TUF FA617NS

#11 2024-02-09 13:56:39

Re: Sudden reboots on Asus TUF FA617NS

#12 2024-02-09 20:16:27

Re: Sudden reboots on Asus TUF FA617NS

#13 2024-02-09 20:20:27

Re: Sudden reboots on Asus TUF FA617NS

#14 2024-02-09 22:02:33

Re: Sudden reboots on Asus TUF FA617NS

#15 2024-02-09 22:17:30

Re: Sudden reboots on Asus TUF FA617NS

#16 2024-02-09 22:18:09

Re: Sudden reboots on Asus TUF FA617NS

#17 2024-02-09 22:26:55

Re: Sudden reboots on Asus TUF FA617NS

#18 2024-02-11 05:25:06

Re: Sudden reboots on Asus TUF FA617NS

#19 2024-02-11 08:59:25

Re: Sudden reboots on Asus TUF FA617NS

#20 2024-02-11 09:03:05

Re: Sudden reboots on Asus TUF FA617NS

#21 2024-02-13 21:05:08

Re: Sudden reboots on Asus TUF FA617NS

#22 2024-02-13 21:10:18

Re: Sudden reboots on Asus TUF FA617NS

#23 2024-02-13 21:36:34

Re: Sudden reboots on Asus TUF FA617NS

#24 2024-02-13 21:43:39

Re: Sudden reboots on Asus TUF FA617NS

#25 2024-02-14 08:34:57

Re: Sudden reboots on Asus TUF FA617NS

Board footer