Take the 2-minute tour ×
Super User is a question and answer site for computer enthusiasts and power users. It's 100% free, no registration required.

I have been getting randoms BSoDs recently and they are becoming more frequent. The error is always different but it has something to do with memory corruption. Bsods are totally random. Can happen while the computer is idle and while it's doing something CPU/Ram intensive. Can happen after 15 mins and after 8+ hours.

I did a RAM test and it returned no errors on both sticks.

I also ran the driver verifier tool and it caused Windows to crash on boot. It made no logs so I was unable to determine which driver was causing it.

OS: Windows 7 Home Premium 64bit
CPU: AMD A8-5500
RAM: Kingston HyperX 1600Mhz 4GB + 8GB
GPU: Radeon HD 7750

25 Oct - Added a 8GB RAM stick. I had one with 4GB before. This one is the same model just with 8GB. 29 Oct - Halved the amount of virtual memory.
30 Oct - Installed AMD overdrive and did some overclocking. After a couple of days, I restored everything and uninstalled it but bsods continued.
30 Oct - First Bsod.
5 Nov - Second Bsod.
5 Nov - RAM test. No errors.
5 Nov - Driver verifier. Failed to boot.
10 Nov - Third Bsod
15 Nov - 2 more bsods
16 Nov - 2 more bsods

The erros:
KMODE_EXCEPTION_NOT_HANDLED
UNEXPECTED_KERNEL_MODE_TRAP
ATTEMPTED_WRITE_TO_READONLY_MEMORY
CRITICAL_STRUCTURE_CORRUPTION x2
MEMORY_MANAGEMENT
PAGE_FAULT_IN_NONPAGED_AREA

Minidumps: http://www.mediafire.com/download/k63wmtd7tdg9sru/minidumps.zip

share|improve this question
    
I see you say that you "added" memory. First place to start is to remove the old memory & just run on the new, and see if you still get errors, then reverse; it's likely not coincidental that this started after you added memory. Also "Memory corruption" is a very vague term; look at this as finding the faulty component that is causing it, instead. I agree with below that the issue could also be your power supply, but check your system temps & fans (esp. CPU cooling) as well. And, as David says, use Memtest, run overnight. And why halve the virtual mem? -- let the system manage it. –  Debra Nov 17 '13 at 15:49

2 Answers 2

1) Check your memory dump with Windbg. (may not help in this case)
2) Check both sticks, one-by-one, not just two of them together.
3) Check the memory thing with Prime95. It can stress memory.

Your PSU may also play a role in this. Especially if you run into BSODs while you do intensive tasks.
Try getting a second one and test stabilty with that.

share|improve this answer
    
Prime95 is not a good memory stress test. Though it has the advantage to stress test the CPU as well, which is another possible cause of error. For memory tests use Memtest86+, for CPU stress tests use Linpack or it's derivative IntelBurnTest (Windows only). –  David Foerster Nov 17 '13 at 14:44
    
I know, but he complained about memory corruption during stress. Mem86 does not stress the computer at all. –  Shiki Nov 17 '13 at 16:05

According to the dump the Kernel is corrupted at 1 Bit:

*******************************************************************************
*                                                                             *
*                        Bugcheck Analysis                                    *
*                                                                             *
*******************************************************************************

PAGE_FAULT_IN_NONPAGED_AREA (50)
Invalid system memory was referenced.  This cannot be protected by try-except,
it must be protected by a Probe.  Typically the address is just plain bad or it
is pointing at freed memory.
Arguments:
Arg1: ffffffffffffffb8, memory referenced.
Arg2: 0000000000000001, value 0 = read operation, 1 = write operation.
Arg3: fffff80003fba634, If non-zero, the instruction address which referenced the bad memory
    address.
Arg4: 0000000000000000, (reserved)

Debugging Details:


Could not read faulting driver name

CUSTOMER_CRASH_COUNT:  1

DEFAULT_BUCKET_ID:  CODE_CORRUPTION

BUGCHECK_STR:  0x50

PROCESS_NAME:  csrss.exe

STACK_TEXT:  
nt!KeBugCheckEx
nt! ?? ::FNODOBFM::`string'
nt!KiPageFault
nt!NtReadFile
nt!KiSystemServiceCopyEnd
nt!KiServiceLinkage
win32k!StartDeviceRead
win32k!InputApc
nt!KiDeliverApc
nt!KiCommitThreadWait
nt!KeWaitForMultipleObjects
win32k!xxxMsgWaitForMultipleObjects
win32k!xxxDesktopThread
win32k!xxxCreateSystemThreads
win32k!NtUserCallNoParam
nt!KiSystemServiceCopyEnd
0x0


CHKIMG_EXTENSION: !chkimg -lo 50 -d !nt
    fffff80003fba630 - nt!NtReadFile+5ae
    [ b8:38 ]
1 error : !nt (fffff80003fba630)

MODULE_NAME: memory_corruption

Run sfc /scannow to check for damaged ntkrnlmp.exe.

share|improve this answer

Your Answer

 
discard

By posting your answer, you agree to the privacy policy and terms of service.

Not the answer you're looking for? Browse other questions tagged or ask your own question.