intermittent BSOD (common bug check string: bugcode_usb_driver)

billiamian · Apr 30, 2014

sys build is a few of years old.
non OEM W7 SP1 X64
MB & CPU are 4 years old
No OS reinstall

MB - ASUS M4A78T-E
CPU - AMD Phenom II X4 955 Quad Core Processor 3.20GHz
RAM - 2 x 4GB Patroit PC3-10600 PX34G1600C11
GPU - Galaxy GEForce GT-620 2GB
Pws Sup - coolermaster 500W

Drivers updated diligently (no significant updates for years, none at all recently...).
In Feb I added RAM, SSD, GPU, DVD, kbd/mouse.
- added 8g to 4g RAM which caused 45 second boot delay at "Initializing USB..."
Removed old 4 G & 8G seemed to be fine, though intermittent slight "Initializing USB..." pause persists.
- maybe it was always there...(?)

March BSOD began - once ever 10 days or so.
Nirsoft Bluescreen confirmed that for more than half, bug check string = bugcode_usb_driver
Yesterday a dozen times, with reboot problems (i.e. no boot, nada, zip, zilch).
Removed all USB - save WNIC, and disabled a few other seldomly used devices.
- I'm typing this on a 6 lb. 1998 Compaq Keyboard...

Thanks in advance,

Wm./

x BlueRobot · Apr 30, 2014

Code:

BugCheck FE, {5, fffffa8007f201a0, 10024396, fffffa8008a31258}

Probably caused by : Unknown_Image ( ANALYSIS_INCONCLUSIVE )

The Unknown_Image ( ANALYSIS_INCONCLUSIVE ) string is most likely due to WinDbg attempts to display a data structure which requires access to Microsoft's Private Symbol server, which is usually only accessible by Microsoft employees.

The third parameter contains the Vendor ID and Device ID strings which can be searched into a online PCI database.

I would suggest checking your motherboard support page for a chipset update.

The exact problem seems to be the USB Port driver attempting to access physical memory which hasn't been allocated to that driver. It may explain the Access Violation exceptions in some later bugchecks.

The last parameter contains the pointer to the Endpoint data structure, which I believe is the _USB_ENDPOINT_DESCRIPTOR data structure. Again, this data structure and symbol information required seem to be private. You can check the MSDN documentation for more details.

USB in a NutShell - Chapter 5 - USB Descriptors is a good source of information for USB documentation.

billiamian · May 1, 2014

Thanks for the quick response.
FYI - No new updates available from ASUS - none for years now, everything's current.
Removed Samsung USB for phones (had read complaints, and now I'm trying everything)
USB in a nutshell - I'm already in the deep end, and I live in Italy so my brain has low saturation thresholds... Not sure I can get my head around it!

Thought to reinstall chipset drivers, which is done via the onboard graphics driver package (ATI Catalyst install manager). Option "uninstall manager" indicated none to uninstall...(?)
- installing

Had configured driver verifier after posting. Might this yield any insights?
Also, would verifier cause sluggish overall behavior I'm now experiencing?
WmiPrvSE.exe has used an hour of CPU time in the past 24 - is this driver verifier related?
Can I disable verifier?
What to do with results?

...

Blue screen again during chipset install. Multiple reboot attempts with all the right noises, but no boot screen, no POST. Let sit for 2 hours & booted fine. Completed chipset install...

Next step I'm thinking to swap in old 2 x 2G Corsair RAM which worked fine until Feb...
Your thoughts?
Any chance the MB is bad?

Thanks again, Wm./

x BlueRobot · May 2, 2014

Driver Verifier will cause sluggish performance, it's normal for it to do so. For the WmiPrvSE.exe process, then you may wish to read this blog post - Is WMIprvse a real villain? - Windows Management Infrastructure Blog - Site Home - MSDN Blogs

Could you also please upload the new dump files? I'm just curious which part of Italy are you from? I know some of my family came from Southern Italy, I can't remember the exact name of the place.

billiamian · May 2, 2014

disabled driver verifier & it sped up.

Reran the script - attached

I'm in Sardinia, closer to Tunisia than Italy!
not a native, but an American émigré (I married a local girl)
LMK if you ever come down. I rent vacation homes here.

another BSDO this AM. For some reason I touched the heat sync & almost burned myself, so now watching temps...
(were no signs of excessive heat, though it's been a minor issue in the past)

x BlueRobot · May 4, 2014

Code:

BugCheck FE, {5, fffffa8007fc71a0, 10024396, fffffa80090c4308}

Probably caused by : usbehci.sys ( usbehci!EHCI_sMode_PollEndpointSlot+67 )

It's the same bugcheck as before, Polling means that a driver or device is checking the I/O status and waiting until the I/O operation is completed.

Code:

3: kd> lmvm nvlddmkm

start             end                 module name
fffff880`07207000 fffff880`07e00000   nvlddmkm T (no symbols)           
    Loaded symbol image file: nvlddmkm.sys
    Image path: nvlddmkm.sys
    Image name: nvlddmkm.sys
    Timestamp:        Sat Feb 08 15:48:31 2014 (52F651CF)
    CheckSum:         00BC7EE8
    ImageSize:        00BF9000
    Translations:     0000.04b0 0000.04e4 0409.04b0 0409.04e4

The raw stack revealed the nVidia graphics card driver, you may want to check for any possible updates from their support page. I would keep a watch on those temperatures too.

billiamian · May 4, 2014

Found 2 driver updates to the video card - 2 in 3 months!
(+ a beta)
downloading the update it bluescreened.
- removed & disabled it.
Read about issues with mem -
"...[FONT=Verdana, Tahoma, Nimbus Sans L, arial, sans-serif]this motherboard, possibly owing to an AMD CPU bug or incompatibility, can freeze with dual-channel memory at 1333 or higher"[/FONT]
[FONT=Verdana, Tahoma, Nimbus Sans L, arial, sans-serif]so I wrote to Patriot asking about system compatibility.[/FONT]
[FONT=Verdana, Tahoma, Nimbus Sans L, arial, sans-serif]Also disabled CnQ in BIOS, as speeds have seemed forcibly reduced for some time, with temps constantly between upper 40s / mid 50s...

Did Driver Verifier leave behind some archive worth reviewing?

Also, if you give me a surname I'll ask my wife for a location. Many can be localized this way, and she's pretty good at it.

LMK, Cheers, Wm./[/FONT]

x BlueRobot · May 5, 2014

I've been told before that surnames are usually based around region, I've haven't got a Italian surname since my Grandad was Scottish (Miller) but I know my Nan's maiden name was Buffone.

Driver Verifier didn't cause any crashes, it will only cause a crash if the any of settings are violated, the common settings are listed here - Using Driver Verifier to identify issues with Windows drivers for advanced users

Have you got a Kernel Memory Dump?

You can find the file here:

Code:

C:\Windows\MEMORY.DMP

If there is file present, then please place it into a zipped folder, and then upload to a free file sharing site such as Dropbox or OneDrive (SkyDrive), and then post the URL to the file.

billiamian · May 16, 2014

I asked the Mrs. about the surname Buffone, and it was a rare occasion that she was stumped. Grammatically it's synomous with "giullare" - a court jester...

I initially sat to write due to a recent spate, in rapid succession shortly after booting. Then it's fine for days, sleeping overnight & waking up without issue.
Then issues...
When I saw my previous post re: RAM (Patriot never replied), I decided to reboot and lower it to 1033 before writing you and, viola! It wouldn't boot at all, no bios, nada, even after a couple of hard resets.

2 hours later, got to BIOS & manually lowered RAM to 1033.
I attach another zip of the results of your Sysnative jcgriff2 BSOD script results.

I would greatly appreciate any insights you might offer...

Thanks again for your time & attention,

Wm./

Search

Search

intermittent BSOD (common bug check string: bugcode_usb_driver)

billiamian

Member

Attachments

x BlueRobot

Administrator

billiamian

Member

x BlueRobot

Administrator

billiamian

Member

Attachments

x BlueRobot

Administrator

billiamian

Member

x BlueRobot

Administrator

billiamian

Member

Attachments