BSOD with hardware error on HP desktop i5-650

[KeMalm]

Hi,

I'm experiencing frequent BSODs since a while. It's been sneaking up on me, at first very infrequent (1 per month) but then increasing rapidly to unusable (every boot and no later than 20 min after).

Edit: The crashes seem random. Can come at login, before or just after without any other activity. Could come when I start browser, or later when using it. Just now I managed to install the latest updates from Win update - but crashes quite soon after the reboot to finish the update. No real pattern...

Se profile and dump files from the forum's tool for details on the i5-650 based HP system. Out of warranty, naturally.

I have been through the following steps already:
1) Updates of Win7 x64 SP1 and drivers for GFX, WiFi, LAN.
2) Memory tests - both Win and Memtest on Ultimate Boot CD. Also swapped places of the memory modules as well as running them one and one.
3) Disconnected and reconnected all connectors on motherboard. Including processor, Gfx & memory modules.
4) Removed ATI Gfx card
5) HD tests from Win and Ultimate Boot CD
6) To exclude Win problems I also installed Ubuntu 13.04 - works fine, but usually no longer than Win before crash / "Kernel panic"

What might it be?

 

[Vir Gnarus]

They're reporting an issue with CPU internal L1 data cache during an evict of data inside the cache. Basically your CPU is bad. If you want to try and confirm this, run Prime95 for a few hours on Torture Test on Large FFTs. If system BSODs during it send us crashdump, otherwise if the app itself crashes or fails, then it definitely adds credence to the CPU being the cause. Even so, I have no doubts your CPU is broken.

Use !analyze -v to get detailed debugging information.

BugCheck 124, {0, [COLOR=Red]fffffa80062d1028[/COLOR], b2000000, 175}

TRIAGER: Could not open triage file : C:\Program Files (x86)\Windows Kits\8.0\Debuggers\x64\triage\modclass.ini, error 2
Probably caused by : GenuineIntel

Followup: MachineOwner
---------

2: kd>[COLOR=Blue] !errrec [/COLOR][COLOR=Red]fffffa80062d1028[/COLOR]
===============================================================================
Common Platform Error Record @ fffffa80062d1028
-------------------------------------------------------------------------------
Record Id     : 01ce56202563c205
Severity      : Fatal (1)
Length        : 928
Creator       : Microsoft
Notify Type   : Machine Check Exception
Timestamp     : 5/21/2013 12:39:19 (UTC)
Flags         : 0x00000000

===============================================================================
Section 0     : Processor Generic
-------------------------------------------------------------------------------
Descriptor    @ fffffa80062d10a8
Section       @ fffffa80062d1180
Offset        : 344
Length        : 192
Flags         : 0x00000001 Primary
Severity      : Fatal

Proc. Type    : x86/x64
Instr. Set    : x64
[COLOR=Red]Error Type    : Cache error[/COLOR]
Operation     : Generic
Flags         : 0x00
Level         : 1
CPU Version   : 0x0000000000020655
Processor ID  : 0x0000000000000004

===============================================================================
Section 1     : x86/x64 Processor Specific
-------------------------------------------------------------------------------
Descriptor    @ fffffa80062d10f0
Section       @ fffffa80062d1240
Offset        : 536
Length        : 128
Flags         : 0x00000000
Severity      : Fatal

Local APIC Id : 0x0000000000000004
CPU Id        : 55 06 02 00 00 08 10 04 - ff e3 9a 02 ff fb eb bf
                00 00 00 00 00 00 00 00 - 00 00 00 00 00 00 00 00
                00 00 00 00 00 00 00 00 - 00 00 00 00 00 00 00 00

Proc. Info 0  @ fffffa80062d1240

===============================================================================
Section 2     : x86/x64 MCA
-------------------------------------------------------------------------------
Descriptor    @ fffffa80062d1138
Section       @ fffffa80062d12c0
Offset        : 664
Length        : 264
Flags         : 0x00000000
Severity      : Fatal

Error         [COLOR=Red]: DCACHEL1_EVICT_ERR (Proc 2 Bank 3)[/COLOR]
  Status      : 0xb200000000000175

 

[KeMalm]

I see, fun times Started the test with Prime95, didn't immediately crash at least. I will let it run and get back with results.

When you say that the CPU is bad, does that mean that "only" swapping the CPU it will work? Or could I be looking at a motherboard swap too?

I was considering finding a newer generation of the i5, but then I'm looking at new board plus Windows license too...minimum.

 

[Vir Gnarus]

Almost positive that it's just the CPU that needs replacing. It's an internal cache problem, so motherboard shouldn't make much difference (unless of course the mobo has some problems with the CPU you replace it with).

 

[KeMalm]

Crashed again after stress test, results are in.

Here we go, the test lasted a few hours before crashing. Logs attached. It may not be the last recorded crash in the logs as it crashed again when I rebooted to get the dumps, but it should be in there.

The diagnosis stands?

 

[Vir Gnarus]

Yeah, exact same error occurred in the Prime95 running thread at the time. Time to get a replacement CPU!