CPU/Mobo Frequent BSOD with 4 x CM2X1024-6400 - Abit IP35-E

Status
Not open for further replies.
Roxtin

Roxtin

Inactive
Galvanizer
Hi I am facing frequent BSOD while I am using 4 sticks of CM2X1024-6400 on my motherboard Abit IP35-E.

The version number of both kits are Kit 1 v5.2 and Kit 2 v7.3 (Don't know if it helps :()

I tried running system with Kit 1 and Kit 2 separately (in both dual channel slots) and the BSOD issue was not faced i.e. I placed 2 sticks of same version RAM in slot 1 and 3 (Blue color) and tested the system and later switched it to slot 2 and 4 (Black color) and no BSOD issue was noticed.

Also to check the other pair Kit 2 v7.3 I used the same method and results were same no BSOD.

But as soon as I put both the kits (4 sticks of 1 GB RAM) together the system gives me BSOD. Tested it both with Windows XP SP3 32 bit and Windows 7 Ultimate 64 bit.

I am really confused what is causing this issue :(

Complete System Config:

Motherboard - Abit IP35-E
Processor - Intel Dual Core E2140
Memory - XMS2-6400
Video Card - Galaxy Geforce 9500 GT 512 MB DDR2
Hard Drive 1 - WD Green 500GB
Hard Drive 2 - Segate 1TB 7200.11
Case - Lancool K56
Power Supply - TAGAN TG600-U37
Operating System - Windows 7 Ultimate X64

I can upload the crash dump of BSOD if anyone requires. Any help would be great.
 
Oh crap I thought that you might get a BSOD at least in one of the testing scenarios. Damit this is such a creepy thing that it BSOD's only when all the 4 RAM's are used.
 
Roxtin said:
I can upload the crash dump of BSOD if anyone requires. Any help would be great.
Click to expand...

Yes, please do so.

Also, is there any OC on. What voltages, and timings are the DIMMs running on..?
 
Okay headsup.

There is no oc, we tried but it wasnt stable

2ndly the bsod is only when 4 sticks are in and none of the mem slots is bad.

I also asked him to reset the bios using jumper and also try and flash the bios to a newr version.

Also roxting i have ip35e lying around if you wanna check if the mobo is causing the prob. Lemme know.
 
No OC settings I used the BIOS utility and loaded the optimized default settings and this is happening only with all 4 RAM inserted together.

@asingh here are the files.

BIOS version is the latest "BIOS ID:18" so didn't flash.
 
Ok,

Will only be able to check those out, today evening. Once I am back home from office.

Also like I asked you, what are the RAM timings, FSB : DRAM ratio..?
 
Ok....

Was able to analyze the BSOD dumps. For reference they will be called as:

Dump_for_032110-20421-01 - D1
Dump_for_032210-16828-01 - D2
Dump_for_032210-25734-01 - D3
D1, D2 clearly point to a driver issue. It is highly probable that it is your Forceware driver which is doing this. When did you last update the Forceware.

Please read these two links. Link1 and Link2. They have mentioned some logic, and fixes which you can attempt. I do apologize, that I cannot give you an exact answer here, but analyzing memory dumps is not exact science.

Now to D3:
This is what the debugger utility threw:

Mini Kernel Dump File: Only registers and stack trace are available

Symbol search path is: SRV*c:\symbols*Symbol information
Executable search path is:
Windows 7 Kernel Version 7600 MP (2 procs) Free x64
Product: WinNt, suite: TerminalServer SingleUserTS
Built by: 7600.16385.amd64fre.win7_rtm.090713-1255
Machine Name:
Kernel base = 0xfffff800`02851000 PsLoadedModuleList = 0xfffff800`02a8ee50
Debug session time: Mon Mar 22 00:50:40.568 2010 (GMT+5)
System Uptime: 0 days 0:04:09.802
Loading Kernel Symbols
...............................................................
................................................................
............
Loading User Symbols
Loading unloaded module list
.....
*******************************************************************************
* *
* Bugcheck Analysis *
* *
*******************************************************************************

Use !analyze -v to get detailed debugging information.

BugCheck 50, {fffff6fcc504d568, 0, fffff800028d6ae2, 2}
Could not read faulting driver name
Probably caused by : memory_corruption ( nt!MiAgeWorkingSet+1c2 )

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

1: kd> !analyze -ve
*******************************************************************************
* *
* 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: fffff6fcc504d568, memory referenced.
Arg2: 0000000000000000, value 0 = read operation, 1 = write operation.
Arg3: fffff800028d6ae2, If non-zero, the instruction address which referenced the bad memory
address.
Arg4: 0000000000000002, (reserved)

Debugging Details:
------------------
Could not read faulting driver name

READ_ADDRESS: GetPointerFromAddress: unable to read from fffff80002af90e0
fffff6fcc504d568

FAULTING_IP:
nt!MiAgeWorkingSet+1c2
fffff800`028d6ae2 488b19 mov rbx,qword ptr [rcx]

MM_INTERNAL_CODE: 2

CUSTOMER_CRASH_COUNT: 1

DEFAULT_BUCKET_ID: VISTA_DRIVER_FAULT

BUGCHECK_STR: 0x50

PROCESS_NAME: System

CURRENT_IRQL: 0

TRAP_FRAME: fffff880031397a0 -- (.trap 0xfffff880031397a0)
NOTE: The trap frame does not contain all registers.
Some register values may be zeroed or incorrect.
rax=0000007ffffffff8 rbx=0000000000000000 rcx=fffff6fcc504d568
rdx=0000000000000001 rsi=0000000000000000 rdi=0000000000000000
rip=fffff800028d6ae2 rsp=fffff88003139930 rbp=007ffffcc504d568
r8=0000000000000001 r9=fffff80002a4de00 r10=0000000000000005
r11=0000000000000000 r12=0000000000000000 r13=0000000000000000
r14=0000000000000000 r15=0000000000000000
iopl=0 nv up ei ng nz na pe cy
nt!MiAgeWorkingSet+0x1c2:
fffff800`028d6ae2 488b19 mov rbx,qword ptr [rcx] ds:08f0:fffff6fc`c504d568=????????????????
Resetting default scope

LAST_CONTROL_TRANSFER: from fffff80002940bc2 to fffff800028c2f00

STACK_TEXT:
fffff880`03139638 fffff800`02940bc2 : 00000000`00000050 fffff6fc`c504d568 00000000`00000000 fffff880`031397a0 : nt!KeBugCheckEx
fffff880`03139640 fffff800`028c0fee : 00000000`00000000 00000980`00000000 00000000`00000000 fffff800`029f5bfe : nt! ??

::FNODOBFM::`string'+0x40f90
fffff880`031397a0 fffff800`028d6ae2 : 00000000`0005e000 04f00000`c083b921 00000000`00000000 00000000`0000104f : nt!KiPageFault

+0x16e
fffff880`03139930 fffff800`02943a0e : fffff800`02a4de00 fffff880`00000001 00000000`00000001 fffff880`03139bb0 : nt!

MiAgeWorkingSet+0x1c2
fffff880`03139ae0 fffff800`028d76e2 : 00000000`000000fa 00000000`00000000 fffffa80`00000000 00000000`00000001 : nt! ??

::FNODOBFM::`string'+0x49926
fffff880`03139b80 fffff800`028d796f : 00000000`00000008 fffff880`03139c10 00000000`00000001 fffffa80`00000000 : nt!

MmWorkingSetManager+0x6e
fffff880`03139bd0 fffff800`02b66166 : fffffa80`039ed040 00000000`00000080 fffffa80`039cd040 00000000`00000001 : nt!

KeBalanceSetManager+0x1c3
fffff880`03139d40 fffff800`028a1486 : fffff800`02a3be80 fffffa80`039ed040 fffff800`02a49c40 00000000`00000000 : nt!

PspSystemThreadStartup+0x5a
fffff880`03139d80 00000000`00000000 : 00000000`00000000 00000000`00000000 00000000`00000000 00000000`00000000 : nt!

KxStartSystemThread+0x16
STACK_COMMAND: kb

FOLLOWUP_IP:
nt!MiAgeWorkingSet+1c2
fffff800`028d6ae2 488b19 mov rbx,qword ptr [rcx]

SYMBOL_STACK_INDEX: 3

SYMBOL_NAME: nt!MiAgeWorkingSet+1c2

FOLLOWUP_NAME: MachineOwner

MODULE_NAME: nt

DEBUG_FLR_IMAGE_TIMESTAMP: 4a5bc600

IMAGE_NAME: memory_corruption

FAILURE_BUCKET_ID: X64_0x50_nt!MiAgeWorkingSet+1c2

BUCKET_ID: X64_0x50_nt!MiAgeWorkingSet+1c2

Followup: MachineOwner
Click to expand...
This last dump, is showing an error due to memory. This could be an issue too. But you are saying that in singular mode, the DIMMs are fine. So we could rule this out. Though if you get more BSODs we can cross check them again, to see if they point to memory issues. For now I would suggest the following:

1. Update all your chipset drivers. Even if on latest, just overlay them again.
2. Read those links. They suggest some Forceware changes.
3. Run each DIMM alone, and run a full MEMTEST on it + run a LinX test, and select 'maximum free memory'. Better to rule out bad memory issues.

Let me know if this helps. Best of luck. By the way, the dump analysis files are attached, appended with the dump name in the file name.
 

Attachments

@asingh thanks for the detailed reply. I am doing a clean install of all the drivers and would run the MEMTEST and LinX test over weekend. Thanks for the help would keep you posted :)
 
Very good analysis by asingh looks like.

I looked at the stuff and I would guess its simply you need the latest x64 (64 bit) win7 drivers because there is possible poor handling of memory sizes beyond 2gb (integer overflow etc) because the MinWorkingSet/MaxWorkingSet APIs return pointer sized data and they probably convert 64bit number into 32bit causing an overflow exception and kaboom :)

edit: For example searching for "GetPointerFromAddress" gives good info like: Junfeng Zhang's Windows Programming Notes : Understand why your computer restarted unexpectedly

Roxtin said:
I am doing a clean install of all the drivers and would run the MEMTEST and LinX test over weekend.
Click to expand...

Just ensure you don't install the same drivers get the latest ones downloaded...

update: ok then can you clarify that you already "downloaded and installed latest drivers" and its still crashing.... where did you download from ? abit site or direct from nvidia/ms/whatever?
 
The crash issue started on Windows XP 32 bit after putting all the 4 RAM sticks, GFX drivers in use at that time was "191.07". Later I moved over to Windows 7 Ultimate x64 and GFX driver in use was "196.21".

Still after asingh recommended I manually removed the gfx drivers and again installed the latest 196.21 Geforce drivers. Regarding the Chipset drivers I am confused as the one available on Abit site is pretty old 2008 release which I think would be already installed on my system. Is there any other way I can check for latest chipset drivers?

Also after this update I am running the system only on 2 RAM sticks and have not faced a crash so far once I reach home late night today will try putting all the sticks back.

I have also posted the same issue on Corsair forums and some mod over there "RAMGUY" gave the following replay
Code: 
Most likely it is a limitation of the memory controller on the motherboard. You would either need to underclock the memory to 667MHz or try giving the nb memory controller more voltage.

Now I am noob at setting the voltage and the OC stuff :ashamed: AFAIK the value should never increase 1.4v so now what is the voltage which I have to set it to?
 
Roxtin:
Did the Corsair analyst look at the memory dumps, and give that recommendation. Also for the latest chipset drivers, probably only the motherboard site would host them. Else you could browse some forums specific to your board. The two links I gave you initially, the first one, mentioned a roll-back Forceware driver, did you try that. At times the latest is not always stable.

I would not suggest changing the Northbridge (MCH) voltage as of now, cause you are not on OC mode. Best you could do, is toggle the FSB : DRAM ratios, so your RAM is at 667Mhz. Change the RAM voltage to 1.9V. Leave the NB voltage at default for now.

Also update your signature please.

EDIT:
Last resort: change the NB voltage to 1.33V. Do this, if all is failing.
 
asingh said:
Roxtin:
Did the Corsair analyst look at the memory dumps, and give that recommendation.
Click to expand...

Nope

asingh said:
Best you could do, is toggle the FSB : DRAM ratios, so your RAM is at 667Mhz. Change the RAM voltage to 1.9V. Leave the NB voltage at default for now.
Click to expand...

The board supports upto 8GB 800MHz RAM so why do I need to underclock the RAM?

asingh said:
Also update your signature please.
Click to expand...

You don't like Man U :P
 
^^
Asking to underclock it, so we get a baseline stability, and rule out the issue of BSOD due to RAM being set for quick speeds. Make the ram latencies 5-5-5-18 @ 1.8V. (Previously I had mentioned 1.9V, disregard that. Sorry.)

Its not about liking/not liking Manchester United. Just that, I have to keep scrolling to the top, to 'remember' what system specification you have. Makes it easier for posters. That is about it.
 
Ok so today I tried running the system with 3GB ram and again it gave frequent BSODs. I am attaching the new dumps and screenshot of BSOD and my BIOS screen.

GFX drivers are latest & I did not changed any voltage or RAM settings as I am a bit scared of tampering with the BIOS as I have totally no idea about it.


 
So everything is updated? Drivers, Windows service packs/patches (windows update) etc?

Now try booting into safe mode (press F8 during boot I think) then see if you still get "frequent" crashes...

BTW when reinstalling, did you check (before installing any drivers) how the system stability was with plain windows and no sound/video/network etc drivers?

And by "frequent bsod" do you mean it happens randomly/anywhere/anytime or can you think of a common pattern in your usage that triggers this BSOD?

edit: (after you try safe mode windows)

your BIOS shows RAM voltage "auto" and cpuid text shows 1.8 volts setting while corsair forum shows 1.9 volts for your RAM model...

can you double check timings and manually set RAM voltage to 1.9 and try again? PROCEED WITH CAUTION though, double check the info online after googling...

some other links (after googling) suggest looking for other bios settings like "memory remapping over 4gb" - disable such options...
 
Have you run the MEMTEST on both sticks individually.

There is no issue with changing BIOS setting. We are not asking you to OC it. There should be a setting VRAM, or DRAM voltage, just change that to 1.8V. (Vishal, is it not 1.8V for the XMS2-6400 at 5-5-5-18. This page shows it 1.8V). Exit the BIOS, before that save/commit the changes.

Your latest dumps showed AVP.EXE causing an error(one of them), was that game running, when you got the error. Also please run the full memtest on each stick. We have to rule out the memory error.
 
I have not run MEMTEST yet will do it over weekend as currently I have no time in hand. I would try to see for the VRAM/DRAM voltage thing (Is DDR2 Voltage VRAM/DRAM?).

AVP.EXE is related to kaspersky and yes it was running.
 
Yea that "forceware driver" may have been a red herring... something is definitely unable to handle 4gb ram amount being returned by those APIs (could be kaspersky or driver or whatever)...

Roxtin, please only if you still facing too much hassle, just reinstall fresh/clean windows and dont install anything else in one shot... boot plain / empty windows and see if you get BSODs then install your drivers/apps ONE BY ONE and each time reboot and check for BSOD that way you can identify the culprit by brute force POE / process-of-elimination... yes its slow but better than banging head trying to figure out whats wrong if we're unable to identify it easily anyway...
 
Status
Not open for further replies.
Top Bottom