Question Kernel Power 41 event consistently while laptop is asleep ?

[SBazu]

Recently I noticed that when I went to wake my laptop from sleep, it was fully powered off. This happens fairly consistently, although not all the time (sometimes it will sleep properly, as indicated by the pulsing power light. Other times it shuts off.)

Checking event viewer, this is always a Kernel Power 41 event. It is listed with either BugCheckCode 0, 307, or 335. I am unsure if there's any consistency in events leading up to this.

I have collected 6 minidump files generated on each critical error. They are in a .zip here:
https://www.mediafire.com/file/kas8o4v0vbbuvjd/minidumps.zip/file

I have no experience in reading these files, I'm hoping someone could spot any consistent issue in them.

Otherwise, I have updated my BIOS and many drivers to the latest version, I have changed sleep wait time to 0 minutes (something recommended for similar issues elsewhere), I have updated Windows and so on, but the problem persists. I have absolutely no issues with this system, crashes or otherwise, besides this.

Any help or insight is much appreciated.
Thanks.

 

[johnbl]

something is taking too long to respond, it is most likely a sound device. I would suspect
OMENAudio.sys or maybe the usbaudio driver.
looks like you have 3 amd audio services also running.
AMDAfdAudioService v1.15(0000) 0xffffaa0df5e95d90
AMDAcpBtAudioService v1.15(0000) 0xffffaa0df5e44280
AMDAcpUsbAudioService v1.15(0000) 0xffffaa0df5fe3dd0

you have to find the bad driver. All you can really do is start disabling drivers. I would disable all audio devices that do not have a speaker connected to them.
(disable hardware in bios if you can, then in windows control panel device manager.

looks like you may have attempted to reinstall the nvidia driver incorrectly. you will want to remove the entire nvidia driver and reinstall, including the nvidia sound driver. after the install you can disable the nvidia sound driver if you want.
this tool can help:
https://github.com/lostindark/DriverStoreExplorer/releases/tag/v0.12.82

you can remove drivers from the driverstore to prevent plug and play reinstalling them a few seconds after you uninstall them.
OR start cmd.exe as an admin then run
net.exe stop "plug and play"
this will block plug and play attempting a reinstall until you restart the plug and play service with.
net.exe start "plug and play"


----------------------
\SystemRoot\system32\drivers\nvhda64v.sys Mon Feb 19 02:44:23 2024

SystemRoot\System32\DriverStore\FileRepository\nvhmi.inf_amd64_86f8a0c7adbcfe28\nvlddmkm.sys Fri Apr 11 22:59:06 2025

audio drivers installed:
AtihdWT6.sys Tue Apr 30 07:51:17 2024
\SystemRoot\System32\drivers\HDAudBus.sys
nvhda64v.sys Mon Feb 19 02:44:23 2024

OMENAudio.sys Thu Feb 23 01:25:38 2023

RTKVHD64.sys Fri Dec 27 01:16:18 2024

usbaudio.sys







why is this installed:
pwdrvio.sys Mon Jun 15 18:43:45 2009
================

power state failure coming out your audio ports.

looks like the nvidia gpu driver called directx, then directx called the nvidia gpu driver. then to some kernel routines and back to
nvhda64v.sys then called
portcls.sys which handles audio drivers

then bugchecked:
A device object has been blocking an IRP for too long a time.
the blocked irp was owned by
\Driver\nvlddmkm

 

[johnbl]

thread NVDisplay.Container.exe
Contention Count = 211402

looks like this helps these programs:

• NVIDIA Control Panel
• NVIDIA ShadowPlay
• NVIDIA GeForce Experience
• NVIDIA GameStream
• NVIDIA Ansel

you might turn off these functions.
there are 2596 locks currently held

you have something called cheatengine75.exe and
cheatengine76.exe running
just fyi.

couple of games running.
this is running:
IkinariH.exe
(did not show up on google, so find out what it is)

qbittorrent.exe running
you might disable if not using it.

Dalamud.Inject running
part of FFXIV_TexTools ??

​

XivAlexanderLo running

atieah32.exe running (make sure it is not a trojan

sharex.exe running. looks like it captures video/sound output.
(could be causing the delay)

nvngx_update.exe running.
does nvidia ai powered optimizations.

 

[johnbl]

most of your system is waiting for a wake up signal.
3 devices need to respond
nvlddmkm
HDAudBus
NVHDA

something is going on with your usb 3 subsystem. it took 15.4 seconds for it to wake up there were a huge number of log entries in the log.
34.4 million.
3 devices connected ,
HP True Vision FHD Camera
G502 HERO Gaming Mouse
USB\VID_0E8D&PID_223C&REV_0100
0e8d =MediaTek Inc. could not find out what device id 223c is for.
looks like you have two media tek drivers installed:
mtkbtfilterx.sys Tue Jun 11 04:13:28 2024 and
mtkwl6ex.sys Mon Feb 17 22:14:34 2025
(looks like they are from different builds)

all the root hubs were suspended. I would see if i could update firmware for each of these devices.
or not suspend the usb 3 hubs so they do not make the devices sleep. ie go to windows control panel, device manager, right mouse click on the usb 3 hubs and bring up the properties, go to the power management and set it so that windows does not turn off power to save energy.
I suspect one of these does not wake up correctly.
most likely the media tek device. I think the video device was on a second usb 3 hub.

note 34.5 million usb messages over 20 hours come out to 28000 messages per minute. 479 log entries per second.

 

[SBazu]

When you say "running" do you mean all of these were running at some point, or were still actively running at the time of the crash? A number of these, like dalamud.inject and cheatengine should *not* have still been active when the crash happened but were things I had active at some point during the session before the crash.

pwdrvio.sys is apparently installed alongside MiniTool Partition Wizard, which is a tool I installed to check my drives for issues *after* the crashes were already a problem.

Much appreciated for all the help. I agree that I've just got to start disabling things and go down the list at this point.

 

[SBazu]

Additionally, I've noticed that when I plug or unplug my headphones into this laptop, there's a small audio freeze that sometimes causes things like videos to require a reload. I hadn't thought much of it, but it might be related to all the audio issues.

 

[johnbl]

debugger thinks a lot of your windows drivers have been modified,
guard_check_icall_fptr:

looks like parts are paged out to the pagefile.sys

 

[johnbl]

Additionally, I've noticed that when I plug or unplug my headphones into this laptop, there's a small audio freeze that sometimes causes things like videos to require a reload. I hadn't thought much of it, but it might be related to all the audio issues.

is there any firmware update to your headphones?
the usb logs had some strange traffic.
some failures because something was using the wrong buffer sizes.
some restarts that took 15 seconds (long time)
I could not figure out the device. I did not see any headphones listed in the usb connections. is it using bluetooth?

there were failures on the usb indicating that the device did not implement the usb command that it was requested to run.

I then tried to check to see if your usb drivers were standard, debugger complained they were modified. I checked other drivers and they also reported that they were modified. Some of the info was paged out.

normally when I see this, I have people delete the pagefile.sys and run cmd.exe as an admin and run
dism.exe /online /cleanup-image /restorehealth
(to remove any system hacks)


or reinstall windows from a clean install image.

there is some kind of problem in your usb subsystem.
logs complaining about wrong buffer sizes being used, (too small)
and functions that are not implemented or not supported when called.

(make sure to do any firmware updates to usb devices)

delete the pagefile.sys by turning off virtual memory and turn it back on. this can dump malware hiding in the pagefile.sys.

are you using any usb connection from your GPU backplane connection?

 

[johnbl]

When you say "running" do you mean all of these were running at some point, or were still actively running at the time of the crash? A number of these, like dalamud.inject and cheatengine should *not* have still been active when the crash happened but were things I had active at some point during the session before the crash.

pwdrvio.sys is apparently installed alongside MiniTool Partition Wizard, which is a tool I installed to check my drives for issues *after* the crashes were already a problem.

Much appreciated for all the help. I agree that I've just got to start disabling things and go down the list at this point.

all were running at the time of the crash

 

[johnbl]

note: headphones now have the ability to wake themself up.
they have to tell the chain of the usb hubs that they are on to wake up.

be sure to update headphone firmware. lots of problems with this, your usb hubs were powered down at the time of the crash.

you could try and tell windows not to power down you usb hubs
(device manager->right mouse click for properties then find power management tab to disable)
(this would be a work around for the 15 second delay)

 

[SBazu]

all were running at the time of the crash

That's strange, I had ended several of those programs hours before the crash, and they shouldn't have persisted.

are you using any usb connection from your GPU backplane connection?

I don't know what this would mean. I don't believe so, but I don't know how to check if I am.

is there any firmware update to your headphones?
the usb logs had some strange traffic.
some failures because something was using the wrong buffer sizes.
some restarts that took 15 seconds (long time)
I could not figure out the device. I did not see any headphones listed in the usb connections. is it using bluetooth?

there is some kind of problem in your usb subsystem.
logs complaining about wrong buffer sizes being used, (too small)
and functions that are not implemented or not supported when called.

(make sure to do any firmware updates to usb devices)

The headphones are just connected via the audio jack, nothing installed. They're operating via a Realtek Audio device which, according to device manager, has no associated drivers. But they're still considered a separate device to the speakers. This is a new implementation to me, I'm used to the audio jack and the speakers being the same device in older systems I've owned.

As for the firmware for my USB devices, the only USB devices I use are a microphone (occassionally) and the G502 Hero mouse (always). The firmware for all my USB devices is up to date. The only one I can't be sure on is this internal mediatek thing, but I've updated that via HP's drivers page and proprietary driver installer.

 

[johnbl]

LOL, it occurred to me what is going on.
find and delete this process
qbittorrent.exe

it would explain the datagram socket streaming data to your hard drive.
it would explain the large number of usb packets
it would explain the modified files, the data hidden in the pagefile.sys
it would explain the random bugchecks via timeouts.


find and prevent qbittorrent.exe from running and your most likely your machine will not bugcheck.

 

[SBazu]

find and prevent qbittorrent.exe from running and your most likely your machine will not bugcheck.

I will certainly try this, but the crash-during-sleep events have occurred at times when I'm pretty sure qbittorrent wasn't running at all. It doesn't run in the background unless I tell it to, and I only have it running sometimes (but certainly it was running during at least a few of the crashes.)
It might be a confounding factor but I'm not sure it's the cause. I'll try anyway.

Out of curiosity, why would it explain the timeouts? The other three things I understand.

 

[johnbl]

I will certainly try this, but the crash-during-sleep events have occurred at times when I'm pretty sure qbittorrent wasn't running at all. It doesn't run in the background unless I tell it to, and I only have it running sometimes (but certainly it was running during at least a few of the crashes.)
It might be a confounding factor but I'm not sure it's the cause. I'll try anyway.

Out of curiosity, why would it explain the timeouts? The other three things I understand.

i think your system goes idle, qtorrent kicks up and is using your system to the max. then the usb ports go to sleep. when the system wakes up there is such a load that it takes over 15000 milliseconds for the usb to start back up. normally it would be under say 200 milliseconds.
I think the delay is just long enough for usb devices to timeout. must be something special about usb audio, the time out value was pretty large.
(over 7000 clock tick) where the usual connected USB device timeout is 500 clock ticks. (was strange that you got timeouts on the slower devices but not the fast usb devices)

the streaming to your drive using a datagram socket rather than a connected socket only makes sense if a file sharing program was doing it.
any program that really cared about your data would use a connected socket not a datagram socket.

connected socket will retry and will make sure the data is saved, a datagram socket it like a radio broadcast, it just throws the data and anyone that want to listen can. Why would you do that from your usb bus to your hard drive if you wanted to really capture streaming data.

you did have spotify running, it could be using datagram packets but why would it stream to the harddrive unless you were saving the stream. (most likely not allowed)

 

[johnbl]

usb connection from the gpu backplane. the back of the gpu card can have a usb 3 connection it is a special connection to a usb 3.x bus inside the GPU. I did not know if you had something connected to it. I was looking to the cause of a bug I saw in the internal usb log. It was a sequence of a controller and a client basically arguing on who is charge of the wakeup.
one said you do the wake up, the reply was what me? I do not know how to do that. (function was not implemented) I was trying to find out what the devices were, guess was that it was some audio device that needs its firmware updated. audio devices started implementing there own sleep functions, they go to sleep by themself. Then the usb bus sees no traffic and it starts a sleep countdown timer and later goes to sleep. the audio device now can wake it self up but it also should wake up the USB port. you might have to have a firmware update to get it to wake up the usb port.
(this can happen for usb mice, microphones and headphones, check for firmware updates)

there were also issues of bad buffer sizes being used by a usb client. the microsoft driver caught the bad buffer and tried to work around the mistake.

ie client using a zero buffer when 12 bytes was needed.
client using a 70 byte buffer when 212 bytes was needed.

these were logged and would be bugs in the client firmware.
i could see the errors but could not figure out what device caused the error from the log.

 

[johnbl]

just for kicks, roll back your
MediaTek Wi-Fi 6E MT7922 (RZ616) 160MHz PCIe Adapter
your version is 5 days newer than the one in the microsoft catalog
https://www.catalog.update.microsof...Fi 6E MT7922 (RZ616) 160MHz Wireless LAN Card

I was thinking maybe microsoft pushed out a bad driver and they rolled it back after you already picked it up.

your driver date
mtkwl6ex.sys Mon Feb 17 22:14:34 2025

microsoft catalog date feb 12 2025, driver version
MediaTek, Inc. - Net - 3.4.0.1193

it is very unusual, do you know where you got it? roll back might make sense. All your network adapters knew the media was disconnect
but the media tek driver did not. I think it was also the power master but did not know what to do.

 

[SBazu]

Setting my USB Root Hubs to not go into power saving mode *seems* to have stopped the crashes but I understand this is a band-aid over the problem more than a fix.

it is very unusual, do you know where you got it? roll back might make sense. All your network adapters knew the media was disconnect
but the media tek driver did not. I think it was also the power master but did not know what to do.

The mediatek wifi device in my device manager is dated to Feb 13th.
mtkwl6ex.sys is also dated to Feb 13th in the driver store.

I'm not sure why you're seeing Feb 17th.