PC Drivers & Software

[mr_neverdies]

Can be deleted.

[dxlc0]

I don't think the RMA will work for you. It didn't work for me. ASUS sent me a brand new TUF B450M Plus Gaming mobo and I thought it was working because I had wiped the SSD and reinstalled Windows 10 from scratch without updating the AMD Chipset Drivers. As soon as I did the update, I started getting the AMD USB 3.10 EHC crashing (Code 24).

After that, thanks to Windows Update, the old, stable drivers were no longer available to me, and I was not able to get back to perfectly stable. It would crash the USB 3.10 EHC maybe once a week, more often if I tried upgrading the AMD Chipset Drivers again each time they released a new version.

However, right now I am on day 10 of testing the fourth WiFi NIC (Gigabyte GC-WBAX200) and it has not crashed the AMD USB 3.10 EHC yet. If it reaches day 30, I'll claim it is fixed. Maybe I'll even put back the third WiFi NIC I tried to see if the problem comes back.

[bnshy]

How does it look?

Is there any fixx out there?

Have the same issue related since yesterday.

[dxlc0]

I've gone 16 days without a single crash of the AMD USB 3.10 EHC. It started on 4/9/2020.

This was after I tried the fourth WiFi NIC, BUT in the past 16 days, there have been some Windows 10 Updates.

4/9/2020 = Intel Corp. Bluetooth 21.60.0.4 (this is a grandchild of the AMD USB 3.10 EHC, so it may be related to the crashes).

4/15/2020 = KB4541335 (probably didn't cause the system to become stable, but it didn't break it either)

4/16/2020 = KB4549951 (also probably didn't cause the system to become stable).

My AMD Chipset Drivers are still at the versions posted a while ago:

AMD PCI: 1.0.0.75 (12/18/2019)

AMD GPIO Controller: 2.2.0.126 (1/14/2020)

AMD GPIO Controller: 2.0.1.0 (4/2/2019)

AMD PSP 3.0 Device: 4.4.0.0 (6/2/2017)

AMD SMBus: 5.12.0.31 (1/18/2015)

I just noticed something interesting, though. The driver for the AMD USB 3.10 eXtensible Host Controller - 1.10 (Microsoft) and also the AMD USB 3.0 eXtensible Host Controller - 1.0 (Microsoft) is newer than I remember:

It is dated 2/20/2020 and is version 10.0.18362.693.

I do not know which Windows Update installed this, but it is definitely newer than all the horrible crashes I had since July of 2019. Maybe Microsoft fixed the crashing by updating the driver for the AMD USB 3.10 EHC. HOWEVER, I had some Code 24 crashes of the AMD USB 3.10 EHC on the morning of April 9, which is after 2/20/2020.

Whether the new Microsoft driver fixes it depends on when it was installed. It could have been installed at any time between 2/20/2020 and today, when I noticed it. Let me see if I can figure out exactly when it was installed....

UPDATE:

There is no mention in my Update History > Driver Updates of the Microsoft driver 10.0.18362.693 for the AMD USB 3.10 EHC. I do not know exactly which update it came in, but my guess is that it came in one of the Windows Quality Updates. Unfortunately, my history does not show that any Quality update was installed on 4/9/2020, when I started this streak of zero crashes by installing the Gigabyte WiFi NIC. The only driver updated on 4/9/2020 was the Intel Bluetooth driver (and the Intel WiFi driver version 21.80.2.1 (dated 2/25/2020) was installed--not updated--on the same day).

So I do not know if the Microsoft driver 10.0.18362.693 for the AMD USB 3.10 EHC made any difference or not.

At this time, I'm leaning towards the Intel Wifi driver making the difference, since I did not experience much crashing when no WiFi NIC was installed in the system. I don't think it was the Intel Bluetooth driver because the second WiFi NIC I tested did not have any Bluetooth at all, yet it experienced crashes of the AMD USB 3.10 EHC.

Anyway, hope that helps. I can't pin anything down through process of elimination, but maybe you guys can try updating to the driver versions I have above and get a stable system.

[bnshy]

Thank you for your detailed reply.

I have the exact same driver version for the AMD USB 3.10 Hostcontroller (Microsoft) and restored my pc yesterday (25th of April)

So far, I wasn't brave enough to install anything related to AMD or ASUS. Radeon Settings is the only thing on my computer (it was just there after the reset) and its version is 19.20 - the newest Version would be Adrenalin 20.2.2.

What I also noticed - all the "other" AMD-Drivers like PCI or GPIO are no longer visible (or installed).

Any advice to go on? Should I try to install the latest Adrenalin and see what happens? Sounds a little nerve-wracking to me....

[dxlc0]

Yeah, play it safe.

Eventually your "missing" AMD devices will show up. That happened to me in the most recent uninstall of the AMD Chipset Drivers (prior to April 9). Two of the devices (AMD PSP and AMD SMBus) were actually exclamation points in Device Manager. The others were rolled back to drivers provided by Windows Update. I left them like that for a while, but they auto-installed and auto-updated on their own several days later. At the time (prior to April 9), I was still getting the AMD USB 3.10 EHC Code 24 crashes, just far less often than when the AMD Chipset Driver was installed.

So it made no difference whether the two devices were exclamation points or not.

[bnshy]

OK. So I installed the latest drivers and now my system behaves like this:

If my wifi-adapter (USB) auto-connects to my 2.4GHz - everything is fine and stable, but - if I manually want to switch to another wifi (5GHz) it's crashing immediately!

Did you already received an answer form AMD since you opened a ticket?

This can't be true that this issue consists since Dec19 and there's no real stable fixx.

[dxlc0]

I didn't have to open the ticket with AMD since the 4th WiFi NIC I tried (Gigabyte's Intel AX200) miraculously worked.

Which WiFi NIC are you using? I had trouble with the 3rd WiFi NIC I tried crashing on 5GHz also. It was an Intel 7260-ac. It crashed one of these ways:

1) Connection to 5GHz router band would suddenly cut off, then when looking at available WiFi networks, I could only see the 2.4GHz band on the same router. After some time had passed, maybe 30 seconds to a minute, the 5GHz band would show up again.

2) Sudden BSODs, usually CRITICAL_PROCESS_DIED, and usually right in the middle of disconnecting or connecting to a VPN (Cisco AnyConnect). I am still using the same Cisco AnyConnect software with this 4th WiFi NIC without the same crashing, so it was the 3rd NIC's fault. If I recall correctly, the 2nd NIC I tested (Intel 5100) also crashed this way.

3) Occasional AMD USB 3.10 EHC crashes, but I cannot prove that this was due to the 3rd NIC. It could have been the AMD Chipset Driver(s) that I was testing at the time.

Anyway, the above three types of crashing all stopped when I stopped using the 3rd NIC. Here's the really strange part. I took that exact same 3rd NIC and put it in a second PC (totally different motherboard than the ASUS TUF B450M Plus Gaming) and ran it 24-7 for over a week (streaming videos/audio, never sleeping). It never crashed once, even on the 5GHz band. So it was only malfunctioning when inserted into the TUF B450M Plus Gaming mobo.

[bnshy]

I have the Realtek RTL8811AU Wireless LAN 802.11ac USB 2.0 Network Adapter.

Drivers are from Realtek (1030.38.712.2019).

I also have two of them (and another) and I switched them to check - no success.

I can't use an internal wifi-card cause there sits my M2.SSD.

So now, I switched to LAN - it's a little bit tricky with all those cables but it's stable
and safe now. I have no problem with other devices like a bluetooth-dongle.

If this won't bring it's success - I will definitely change my MoBo.

Thanks for helping so far (:

[dxlc0]

I actually tried a RealTek RTL8812AU chipset-based USB WiFi adapter with this motherboard long ago, before testing the 1st internal WiFi NIC (the Intel 9260).

It worked okay most of the time, but the driver wasn't great and it would BSOD crash once every 3 weeks when I was disconnecting/connecting the Cisco AnyConnect VPN. I used 5GHz.

I don't recall getting AMD USB 3.10 EHC crashes while using the RTL8812AU adapter.

If I uninstalled the Cisco AnyConnect software, the BSOD crashes stopped, but I need that for work, so I went to an internal Intel WiFi card thinking this would be more stable (boy was I wrong).

At this time I am unable to blame the RTL8812AU + TUF B450M Plus Gaming combination for the BSOD crashes. It could have simply been a crappy RealTek driver, and Cisco AnyConnect's virtual NIC driver is pretty unreliable, spewing errors into the Event Viewer even when it appears to be working fine.

However, I can't rule out the combination of RTL8812AU + TUF B450M Plus Gaming mobo either, because I did not test the RTL8812AU on another PC to make sure it was rock-solid.

[dxlc0]

BTW, you have another option. Use a WiFi to Ethernet bridge. I used a little pocket travel router with OpenWRT to do this for a few months way back in 2019. It worked fine (no crashes b/c I was using the integrated RealTek GbE NIC on the mobo).

I only stopped using it because the pocket router didn't have a big enough antenna. This was before I started trying the internal WiFi NICs, which do have large antennae.

[dxlc0]

Final follow-up.

Between 4/9/2020 and today, 5/12/2020, I have not had a single crash of the AMD USB 3.10 eXtensible Host Controller.

The evening of April 9 was when I switched from an Intel 7260-ac to a Gigabyte GC-WBAX200 WiFi NIC in the bottom PCI-e slot.

Current driver versions:

AMD PCI: 1.0.0.75 (12/18/2019)

AMD GPIO Controller: 2.2.0.126 (1/14/2020)

AMD GPIO Controller: 2.0.1.0 (4/2/2019)

AMD PSP 3.0 Device: 4.4.0.0 (6/2/2017)

AMD SMBus: 5.12.0.31 (1/18/2015)

AMD USB 3.10 eXtensible Host Controller - 1.10 (Microsoft): 10.0.18362.693 (2/20/2020)

AMD USB 3.0 eXtensible Host Controller - 1.0 (Microsoft): 10.0.18362.693 (2/20/2020)

I did not install the Ryzen Power Plan.

These are not the latest versions of the drivers that you can get! You can get newer ones from AMD's website. The above are just the latest versions that Windows Update provided.

None of the above has changed since 4/9/2020.

There have been several Windows Updates between 4/9/2020 and 5/12/2020, but the change that appeared to fix my problem was switching the WiFi NIC out. I can't prove it, but that's when the system became stable.

Good luck to everyone else! I hope this helps you. All in all, I wasted dozens of hours diagnosing this and $15 shipping the not-defective ASUS TUF B450M Plus Gaming mobo back to ASUS for a new one. But my PC is solid again and I'm happy.

[wjfr23]

I have been experiencing a similar problem with my PC. In April 2019 I set up a setup with Mobo Asus Prime B450-Gaming / BR, Ryzen 7 2700x, Gigabyte RTX 2070 Gaming OC, 2x16GB DDR4 2400mhz HyperX Fury.

As soon as I set up the computer, I ran stress tests and everything worked perfectly. The problem is that when connecting a wireless usb adapter, the computer immediately froze and after a few seconds came the dreaded BSOD.

I tried several adapters from several different brands, and they all caused BSOD a few minutes after plugging in.

Even in the first week, after a BSOD, the computer when restarting no longer showed an image on the monitor, requiring a hard reset in the BIOS.

This behavior started to be repeated. Whenever a BSOD occurred it was necessary to reset the BIOS for the boot PC again. This issue was resolved by updating the BIOS from version 0806 to 1201 in mid-May.

However the BSOD triggered by the USB wireless adapters continued to occur. Sometimes several times a day or once or twice in an entire month. All of these occurrences without an apparent pattern. In November, the occurrences became so frequent that it was impossible to use the PC.

In addition I was also having problems with sleep and hibernation, since upon waking up the computer would lock up and then restart instead of resuming Windows from where it left off.

I tried everything you can imagine: BIOS Upgrades and Downgrades, chipset drivers from the Asus page, the AMD page, native Windows drivers, and many other things. I got to the point of formatting the PC three times in the same day.

These BSODs always had the following codes:

* Critical_Process_Died (related to the interruption of the WLAN Auto Config, due to the disconnection / connection of the adapter, or the deliberate interruption of the process.

* Whea_Uncorrectable_Error (This error is related to the hardware failure, which I discarded after several tests on the components)

* "Unexpected_Store_Exception" (I even thought it was a problem in the SSD M.2, but that hypothesis was soon discarded after the execution of CHKDSK.

Over time I discovered that moments before a BSOD the controller "AMD USB 3.10 eXtensible Host Controller - 1.10 (Microsoft)" stopped working.

On 04/18/2020 I formatted the PC again and decided not to install any third party drivers, not even the video card driver. I left only the drivers that are automatically installed by Windows itself. Then the BOSD miraculously disappeared, along with the hibernation and suspension problems.Today it is 05/16/2020 and in that almost a month my computer has been working perfectly as never before.

Then, investigating the device manager, I saw that the controller "AMD USB 3.10 eXtensible Host Controller - 1.10 (Microsoft)" is using the driver 10.0.18362.693 dated 02/20/2020.

I can say that this Driver 10.0.18362.693 solved the problem.

However, there is a caveat: I decided to try the 2004 version of Windows 10 Insider Preview and there was a BSOD, since the USB driver is replaced with each new version of Windows. So I downgraded to version 1909 and will stay there until there is a more stable driver for version 2004 or I find a way to install Driver 10.0.18362.693 on all next versions of Windows, since it is the only one that worked satisfactorily .

[dxlc0]

Thanks for reporting on your trials and success!

I kinda wish I had kept the USB WiFi adapters I tried (from middle of 2019) so that I could test them now to see if they are reliable with the 10.0.18362.693 USB driver.

What USB WiFi adapter are you using right now? It might help others to know.

[wjfr23]

The adapter I am using is the TP-Link TL-WN722N V3. His signal strength is excellent.

As for the USB driver 10.0.18362.693, I will try to see if there is a possibility to backup it to reinstall it in future versions of Windows. Otherwise I will have to be stuck on Windows 10 version 1909.

[warrenuk]

Hi all.

I am having the same issues with an FM2+ MOBO.  Everything was stable for 6 months and then I had a Windows update on the 14 April and straight away my PC is randomly turning off.  Sometimes once a week and sometimes once a day.  

AMD 3.00 eXtensible Host Controller crashes a lot and sometimes my USB sockets do not work and require multiple reboots in order to start up again.  

People have told me it's a hardware issue as the board is 5 years old but to me that is not old.  I dual boot with Linux and I have never had a shutdown episode, so that points to a Windows issue.

Funnily enough I am running a TP-Link TL-WN722N from the usb port, but prior to the windows update the system was stable even with this wifi stick.

This is definitely a USB driver problem that AMD needs to look at.

[dxlc0]

Can you roll back the USB driver(s), or did the Windows Update make that impossible?

What happens if you run without the TP-Link TL-WN722N and only use whatever minimal USB devices you have to have (keyboard+mouse and that's it)? I noticed with my setup that if I removed the WiFi NIC from the 4th slot and used the built-in Gigabit Ethernet NIC the board was stable (only had keyboard+mouse attached).

If the system is stable without the TP-Link TL-WN722N, maybe it's that particular USB NIC that doesn't play well with the AMD USB buses.

[wjfr23]

Hello warrenuk

Apparently Microsoft has not yet been able to fix AMD USB Buses issues.

It is interesting to note that both for me and dxlc0 the fixes were delivered with April 2020 Tuesday Patch, while for you this Patch triggered the issues.

I want to point out that the incompatibility is not limited to the model TL-WN722N. I can state this properly because I tested several USB wireless adapters from different brands and models on my MOBO Asus Prime B450-Gaming/BR and all of them, without exception, caused crashes and blue screens of death.

It is also good to clarify that the driver that controls the USB Buses is provided by microsoft itself, and its version is numbered according to the windows version number. So the only way to change the version of this driver is upgrading or downgrading the entire system.

So I would like to suggest some procedures:

First: Check if it is possible to downgrade Windows to the last version before you get troubles. Try to uninstall the update manually via the Control Panel, or then try System Restore. If is impossible to downgrade, try to install the latest update, provided on Tuesday, 06/09/2020. It may be that this subsequent patch has fixed this issue.

Second: Try plugging the wireless adapter in one of the USB ports that are controlled directly by the CPU instead the controlled by chipset, wich controls those ports throught  "AMD 3.00 eXtensible Host Controller".

Third: Create a new HD partition and do in it a clean installation of Windows in dualboot. Do not install any of the drivers provided by AMD neither by the manufacturer of MOBO. Let Windows Update install all the drivers, including the VGA driver. Do not install antivirus or any other software so that you can ensure that the problems presented are not caused by third-party drivers or processes.

Fourth: Are you having BSOD? If so, what is the error code? Is your system generating Dump File?

[_daisy_]

Try this next time it fails. Open Device Manager, go down to the Universal Serial Bus controllers, click on it to drop down the controls and then right click on the "AMD 3.0 eXtensible Host Controller", uninstall the driver and then, restart your PC. The driver will be back and, your ports most likely will work again too. It worked like a charm for me, a couple minutes tops!  Good Luck...hope this helps!

[_daisy_]

Hi guys! My AMD USB 3.10 eXtensible Host Controller also failed on my Dell laptop after a Microsoft update on 7/10/20. Two of my USB plug-ins {on the left side} had stopped working. Well, I had remembered reading about another problem I had with one of my other drivers, and it said to open Device Manager, go to where the problem is, right click on the driver and click uninstall, restart my PC, log on, and the driver will reappear and may work properly again. So, I tried that with this problem, {except this time I shut down rather than restarted my laptop, I guess it doesn't matter.}, and, sure enough, when I logged back on, my problem was solved! All my USB plug-ins work fine! It's such a quick and easy fix. I hope this helps someone down the line. Cheers!    

[gpaganini]

Anyone has managed to fix this issue?

In my scenario, it seems that whenever I'm in some Zoom or Teams meeting, also doing some work on Photoshop or Lightroom, the AMD 3.10 eXtensible Host Controller Driver fails, bringing down Storage Controllers, Audio, Ethernet, also starts freezing until if finally stop responding at all. Only comes back after a hard reset with reset button.

I've tried everything i know: BIOS upgrade/downgrade, multiple chipset versions (newer/older), various AMD drivers versions, multiple combinations between this driver sets and nothing. Everytime I think I've fixed it (run some tests on zoom or teams meetings, open programs that causes the EHC to fail for several hours whithout a fail), it comes back to haunt me.

I'm seriously thinking about replacing this mobo whit an inferior model, like some Gigabyte with B350 Chipset, as seems to me that the real problem is this B450 Chipset.

My setup:
Asus Prime B450m-Gaming/BR
Ryzen 3200G

16GB RAM 2133MHz

M.2 SK Hynix 128GB SSD

Windows 10 2004

AMD Software v 20.4.2

AMD Chipset Drivers v2.07.14.327

Thank you all, anyway.

[dxlc0]

What's your version and release date for the AMD USB 3.10 eXtensible Host Controller 1.10 provided by Microsoft?

[gpaganini]

Driver Version: 10.0.19041.423

Driver Date: 07/27/2020

Also, the most stable BIOS version that I've found and I'm using is 1823. Not tested newest BIOS v2202 yet.

Thanks.

[dxlc0]

Hmm. 10.0.19041.423 is the version of the USB driver I am also at, and it's been rock-solid for me since the prior version release.

I don't have the exact same mobo, so I can't compare my BIOS version to yours. Maybe someone else will chime in.

Do you have any other hardware on the same PCI bus as the USB controller? If so, you might want to temporarily remove them for testing.

[rajeh]

Hi, I've been having a slightly different issue with the same cause on my PC. It's got a Ryzen 5 3500X on a Biostar B45M2 (B350!!!) with Windows 2004 installed (existed in 1909 too). I've tried getting rid of the Ryzen Power Plan, enabling/disabling USB selective suspend, using older chipset drivers, and reinstalling Windows. None of this has worked.

The issue only occurs when I disconnect a USB device from the rear I/O, it brings down every other device for a good 5 seconds. This can be annoying when I want to put my wireless mouse on charge in a Zoom meeting, because it disables my webcam, microphone and headset. I have no Wi-Fi adapters in my PC.

Now I'm testing whether it works properly after uninstalling the xHCI driver and letting Windows reinstall it.

I don't remember experiencing this issue until recently, so I'm thinking it may be a Windows update within 1909 which bricked it, because I also did have 1909 stable for a bit.

[dxlc0]

That could be a bad USB cable or device that is shorting out the USB port and triggering a current overload condition.

1) Do you have multiple USB buses in your PC? Does the problem occur on the other USB buses? In my case, my AMD USB 3.0 eXtensible Host Controller was stable while my AMD USB 3.1 eXtensible Host Controller was having problems. Try moving as many USB devices as possible to the other USB buses.

2) What happens if you use a powered USB hub to take some of the current draw off of the motherboard?

3) Check with the mobo mfg. for a BIOS update, and you might have to submit a problem ticket with them to get a fix.

4) Download and run the freeware USBTreeView tool and watch the USB buses as you connect and disconnect devices.

[rajeh]

1) I believe so, 2x AMD USB 3.10 xHCI controllers. And I think disabling one of them fixed it, but that's a workaround and I don't like those especially when I have 5 USB devices, taking up almost the entire I/O. 

2) I don't have one, but FWIW one of my Vbus pins on the front USB header is broken. However I doubt that's the issue because everything is brought down, and that's without even touching the front I/O.

3) I'm on AGESA 1.0.0.4B, Biostar have had plenty of time to fix it. I tried AGESA 1.0.0.6 but it made overclocking very weird, like my SOC voltage was at 900mV to get FCLK stable.

4) Will do.

Edit: What should I be looking for? It seems that all the software will do is state the obvious, that everything disconnects.

Also I can't seem to reproduce it after uninstalling the xHCI driver and letting Windows install it again. I don't know if this is bad or good, because on one hand my issue may be fixed, but on the other it'll get me when I least expect it.

Edit 2: Just after the previous update I got it to do it again, it never disconnected my webcam which is the only device on the other xHCI controller (PCI Bus 13) but it disconnected everything on the other other xHCI controller (PCI Bus 1). It said the DEVICE_QUALIFIER_DESCRIPTOR's error was ERROR_GEN_FAILURE.

[dxlc0]

ERROR_GEN_FAILURE is sadly too unspecific.

But you did learn from the webcam that you're only bringing down one USB bus at a time. That's good.

You should be able to balance the devices across the buses so that you minimize the harm when one bus goes down. I had to do this for several months.

USBTreeView can show more specific error codes (Problem Code) or power problems. You might also see that it starts to connect a device and then fails at a specific step in the process.

Regarding charging your wireless mouse, I highly recommend that you use a totally separate smartphone charger (one of those small wall warts) to charge it.

Other than that, you need more specific error codes, so keep trying to trigger the problem and figure out if it's one specific USB device at fault or one specific USB bus that always fails.

Also, if your mobo is under warranty, perhaps you could get it replaced. Though it turned out to be software in my case, it might be hardware in your case.

[rajeh]

Yeah I just moved my microphone to the other bus. Unfortunately it'd be impossible to claim warranty because:

1) I broke a Vbus pin when I was reconnecting the USB 3.0 header.

2) It's a prebuilt, and when I tried to claim warranty on a GPU they just said "oH No It ISn'T FaUlTy". Like is 110c junction ever ok???

I have some dump files, I'll reinstall Windbg and post them here. I believe it was consistently something to do with LIVEDUMP and CONTEXT. A reliability monitor error appears too, LiveKernelEvent 0x144 and Parameter 1 is 1020.

Update:

   Minidump: https://pastebin.com/yJUEDPm1

Update 2: I had like a million updates but I deleted them because they were all useless. I disabled IOMMU, and I was pretty happy until it came back again. I guess I'll just live with it until a new windows update comes along.

Update 3: Yay a new bugcheck for me to look into! 0x144 0x100b and 0x144 0x100d. Looks like I found someone on the internet with the same issue, and the conclusion they came to with someone who was helping them was that it was a 3rd party crashing XHCI or the 3rd party crashing and making it look like it was XHCI's fault. The error was quite dramatic too, permanently taking down XHCI on PCI Bbus 1 until I reinstalled the driver. I uninstalled Logitech Gaming Software, the first thing which comes to mind, and I'm testing now.

Update 4:  Still unstable. I returned my memory and infinity fabric to stock et voila it stopped (well, I didn't test for more than 10 minutes ). I've thoroughly tested the memory part, so I'm certain it's the infinity fabric. I lowered VDDG (that helps because of localized heat) and I'm running Prime95 now and checking for WHEA errors. Then I'll keep trying to disconnect and reconnect my mouse and see if it bugs out.

[dxlc0]

https://www.reddit.com/r/techsupport/comments/iapwu9/am_i_reading_into_reliability_monitor_errors_to... 

"BUGCODE_USB3_DRIVER (144)

This bugcheck usually happens when the USB3 core stack detects an invalid

operation being performed by a USB client. This bugcheck may also occur

due to hardware failure on a USB Boot Device."

I'm leaning towards a faulty USB device or a flaky USB cable. I know you need your USB devices, but if you start with just the basics (USB keyboard and USB mouse only) and then add the devices back one at a time, you might find that one device seems to trigger the bus crash more than the others.

[rajeh]

That's my reddit post from when I didn't know what it was lol.

I want my PC to be bulletproof, I found that my infinity fabric is potentially the cause here. I returned to stock 1333MHz and poof it stopped crashing. I'm testing a lower VDDG now (lower is more stable for me, counter intuitive lol).

[dxlc0]

Very cool--I hope your changes stabilize your system so that it is completely bulletproof. Let us know how it goes.

[rajeh]

Lol you know what I meant! I want it to just work and not be so finnicky. I tried changing up my memory settings and even at stock it came back with the issue, but what's weird is that it also happened on my WinPE drive. I just checked another minidump and the stack text mentions "wdf01000" so I'm looking through it's dumps now.

[dxlc0]

WinPE--that shouldn't have the same driver for the AMD USB 3.10 eXtensible Host Controller.

Are you able to try out LINUX (such as using an Ubuntu LINUX Live CD) to see if the USB bus also misbehaves under a totally different OS?

If it crashes in the same way under LINUX as Windows, then you almost certainly have a hardware problem.

[rajeh]

I'm not sure it's WinPE, or at least unmodified. It's Sergei Strelec's boot CD. I'll try Linux when I can. If it is a stuffed motherboard I might be screwed because of that broken Vbus pin and the shitty company.

Btw, it came up with hidtelephony.dll being the culprit, failing the driver verifier. I'm not sure what to make of it though.

[dxlc0]

If you can narrow the problem down to the motherboard, swapping it out for something better is not expensive.

With the broken pins, even if the problem isn't the motherboard, it might still be worth swapping it out.

[rajeh]

The device HID-compliant headset (location (unknown)) is offline due to a user-mode driver crash.  Windows will attempt to restart the device 4 more times.  Please contact the device manufacturer for more information about this problem.

Found this in event viewer, I never checked critical because I thought errors was as bad as it could get ‍:facepalm:

It's consistently been the headset too.

My Cloud IIs aren't dying are they?

[dxlc0]

It's possible. Test the Cloud II headphones by attaching them to some other computer or game console. If the crashing on your PC stops, it might be the Cloud II. If the crashing follows the headphones, then it definitely is the Cloud II.

Could be a bad cable on the headphones. That probably can be swapped easily.

[rajeh]

Lmao I spent 8 hours yesterday trying to fix this (I wouldn't have if the issue didn't escalate into permanently crashing the system) and I woke up and fixed it in 10 minutes. It wasn't the headset. For some odd reason my keyboard and mouse really don't like being in the same row, they can even be in the same XHCI controller, just not the row.

Edit: Gah! It happened again in a Zoom meeting. Though the occurrence isn't as common. I moved the keyboard to the other XHCI controller and I think it's even better. If it isn't fixed I'm going to say it's a driver thing or a Windows thing. It's good enough now. Strange though because I used to have it working with no problems (or did I?) even with the mouse and keyboard on the same row.

[dxlc0]

Hey, a win is a win!

Check the keyboard and mouse for any damage to the USB cable or connector. If the pins inside the USB connector are oxidized, you can use DeOxit spray to clean them up.

[rajeh]

They look fine to me. It'd be pretty bad if my high end Logitech keyboard and mouse had cheapy connectors. I'm not sure if it's just this peripheral set or anything, maybe keyboard and mouse need to be on different XHCI controllers. Or it could be the way the keyboard handles N-key rollover tripping up the motherboard.