Please help... BSOD in Windows 7 (0x0000124)

[PrTv]

Hi all,

I've just finished my new build and the spec is as follows:

i7 4790K + Noctua U14s with 2 Fans
Gigabyte G1 Sniper Z97
Corsair Vengeance 8*2
Gigabyte GTX970 G1 Gaming
4-HDD
1-SSD
1 DVD
Windows 7 64bit SP1

The vga is connected to 2 Dell 27" monitors.

I have just installed the windows a week ago and not much is going on (just some games / photoshop / etc.

Recently, I had BSOD with 0x0000124 code when playing Witcher 3 and it happened 3 times already.

For your information:

My motherboard somehow automatically O/C my CPU to 4.4Ghz (all four cores) ---by the spec, 4790K will only reach 4.4Ghz when one core is loaded and will reach 4.2Ghz Max when 4 cores are loaded. However, I know that this would be OK as a lot of people can get even higher O/C, so I left it there. Then I tried to lower the V-Core as, in auto, the MB seemed to use too much voltage (1.265) and that generated quite a lot of heat. I managed to get it to 1.132 at 4.4Ghz.

I used this setting for a week without any issue. I used it for:

- playing games (mostly Witcher 3) 4-5 hours straight
- heavy Photoshop work (3-4Gb PSB file which eats up almost all my ram)
- leaving it idle with/without youtube playing

All that without any issue.

Then issues started to surface when:

- I tried to O/C my VGA and to test the stability, I left EVGA Precision-X open on one screen and play game on the other (main screen). That was when I got my first BSOD (it was about memory_management BSOD though). I used Windows Memory Diagnostic to check my ram afterward and the result was OK.
- When I got that BSOD, I decided... OK, f**k this and gave up my idea of O/C-ing. I then restarted and make sure that my GPU (and Mem) clock speed was back to normal.

I then tried the game again, and this time with MSI Afterburner loaded, to monitor my GPU/CPU usage and temp. I came up with this 0x0000124 errors on two occasions. First time when a huge scene was loaded (the game loaded the scene and I was about to walk and then BSOD). Another time when I tried to save, it BSODed 0x0000124 when I opened the save dialog. The game is Witcher 3, of course.

------------------------
I know that, when I said O/C or lowering VCore, it'd be like a red flag as people might think that that's the cause. However, I would believe that, that may be unlikely, as the machine run normally and I could use it heavily without a hitch for almost a week.

I would like to hear your though, please kindly share what you think about this.

At first, I thought it's MSI Afterburner that caused this issue but my VGA was not O/C in anyway when this happened, and I just use the program to monitor my hardware. When looking at the DMP file, I was surprised that the file contained so little information (it is a full dump of around 1gb in size).

Here is the extract of the file:

************* Symbol Path validation summary **************
Response                         Time (ms)     Location
Deferred                                       SRV*C:\Windows\symbol_cache*http://msdl.microsoft.com/download/symbols
Symbol search path is: SRV*C:\Windows\symbol_cache*http://msdl.microsoft.com/download/symbols
Executable search path is: 
Windows 7 Kernel Version 7601 (Service Pack 1) MP (8 procs) Free x64
Product: WinNt, suite: TerminalServer SingleUserTS
Built by: 7601.17514.amd64fre.win7sp1_rtm.101119-1850
Machine Name:
Kernel base = 0xfffff800`01a08000 PsLoadedModuleList = 0xfffff800`01c4de90
Debug session time: Fri Jul 10 01:27:43.304 2015 (UTC + 7:00)
System Uptime: 0 days 6:03:35.804
Loading Kernel Symbols
...............................................................
................................................................
........................................
Loading User Symbols
PEB is paged out (Peb.Ldr = 000007ff`fffd7018).  Type ".hh dbgerr001" for details
Loading unloaded module list
.........
The context is partially valid. Only x86 user-mode context is available.
The wow64exts extension must be loaded to access 32-bit state.
.load wow64exts will do this if you haven't loaded it already.
*******************************************************************************
*                                                                             *
*                        Bugcheck Analysis                                    *
*                                                                             *
*******************************************************************************

Use !analyze -v to get detailed debugging information.

BugCheck 124, {0, fffffa800daff028, bf800000, 124}

Probably caused by : GenuineIntel

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

There is no more information regarding any file/service eg .ini .drv, etc, that caused this. Don't know if "Probably caused by : GenuineIntel" is the indication that it's my CPU that caused this.

Do you guys think it's because I lowered the VCore too much? (despite the fact that it was stable 100% doing the same task a week ago, and that a lot of people could do 1.100v with that CPU --mine's at 1.132). I've been into computer for more than 10 years. Not to say that I'm an expert in anyway, but I find it hard to believe that this would be the cause. However, please feel free to let me know what you think, and any comment would be greatly appreciated. If you guys think it's really the VCore, I will give it a try tonight (now I'm at work).

The DMP file was uploaded to my Dropbox and the link is here
https://www.dropbox.com/s/cuj0h30ul60qt14/MEMORY.rar?dl=0

Sorry for the file size, I tried to make it as small as possible but 168Mb is as small as I can get it. I guarantee that this file is safe and virus free, so please download without worries.

Also, sorry for the long post, as I tried to explain the problem as clear as possible. Please let me know if anything is missing.

EDIT:
When this happened, my CPU and GPU temps were fine as Noctua U14s is one of the best air-cooling out there and the CPU temp did not exceed 60c when this happened.

 

[Patrick]

The bottom line is all auto-overclocking, regardless of the BIOS/motherboard manufacturer, is terrible. To be honest, I'm not even sure why it exists. Even if you manage to get a stable overclock with it, it pumps beyond too much voltage more than you'd need to if you did it manually. Every CPU and piece of hardware is different, and therefore a voltage that for example runs an i7 2600k @ 4.5 GHz may not run another, because they're different although the exact same model.

Disable the auto-OC and monitor for further crashes. They shouldn't happen.

 

[PrTv]

The bottom line is all auto-overclocking, regardless of the BIOS/motherboard manufacturer, is terrible. To be honest, I'm not even sure why it exists. Even if you manage to get a stable overclock with it, it pumps beyond too much voltage more than you'd need to if you did it manually. Every CPU and piece of hardware is different, and therefore a voltage that for example runs an i7 2600k @ 4.5 GHz may not run another, because they're different although the exact same model.

Disable the auto-OC and monitor for further crashes. They shouldn't happen.

Thanks for your reply.

Unfortunately, there is no way to turn off that "auto OC" of my MB. Only way to fix that is to manually set my turbo boost to 44/43/42/42 in BIOS.

That said, can you please comment on the following:
- why my dump file doesn't contain information regarding services/drv that cause the BSOD? Is there anything I have to enable to have that full information?
- looking at my dump file, do you think the "Probably caused by : GenuineIntel" is the indication that it's the CPU that caused this? (and not MSI Afterburner).

Thing is, I got 2 BSODs when running the game with MSI-AB, but the game used to run fine before (with just FRAPS).

In any case, tonight, I will try to disable that extra OC and test some more.

 

[PrTv]

I did some search (as I'm at work and cannot test anything anyway) and found this thread
0x124 BSOD,suspected CPU Cache failure....(had a go at debugging) - Windows 7 Help Forums

This guy's issue is quite similar to mine and his error is

BugCheck 124, {0, fffffa800d070028, bf800000, 124}

Mine is

BugCheck 124, {0, fffffa800daff028, bf800000, 124}

Based on that thread, removing Afterburner seemed to do the trick. I will try removing my O/C and removing that program and will report back tonight.

 

[Jared]

That was originally said by Patrick, it's the auto overclocking that is most likely the problem.

Those crashes won't be the same, just because the parameters passed to KeBugCheckEx.
Parameters passed, depend on the type of crash, most 0x124s with a machine check exception pass almost identical information. The main thing that differs will be the addresses to the low and high bits of the MCi_STATUS structure, which will vary on different boots depending on where it will be loaded in memory. The MCi_STATUS record sends the hardware error information, via a machine check exception, to the operating system, which determines what to do, in this case, a bugcheck.

EDIT:
When this happened, my CPU and GPU temps were fine as Noctua U14s is one of the best air-cooling out there and the CPU temp did not exceed 60c when this happened.

It's not necessarily a temperature issue, it's more than likely a voltage problem, excess (probably this) or lack of (doubtful).

---

Then issues started to surface when:

- I tried to O/C my VGA and to test the stability, I left EVGA Precision-X open on one screen and play game on the other (main screen). That was when I got my first BSOD (it was about memory_management BSOD though). I used Windows Memory Diagnostic to check my ram afterward and the result was OK.
- When I got that BSOD, I decided... OK, f**k this and gave up my idea of O/C-ing. I then restarted and make sure that my GPU (and Mem) clock speed was back to normal.

I then tried the game again, and this time with MSI Afterburner loaded, to monitor my GPU/CPU usage and temp. I came up with this 0x0000124 errors on two occasions. First time when a huge scene was loaded (the game loaded the scene and I was about to walk and then BSOD). Another time when I tried to save, it BSODed 0x0000124 when I opened the save dialog. The game is Witcher 3, of course.

Not only is overclocking software garbage, that barely works properly. The drivers themselves are crap, another reason to avoid it.

hal!_MCI_STATUS
   +0x000 McaErrorCode     : 0x5043 // 0101 0000 [COLOR="#FF0000"]0100[/COLOR] [COLOR="#008000"]00[/COLOR][COLOR="#0000FF"]11[/COLOR]
   +0x002 ModelErrorCode   : 0x5245
   +0x004 OtherInformation : 0y11111110000001000010000 (0x7f0210)
   +0x004 ActionRequired   : 0y1
   +0x004 Signalling       : 0y1
   +0x004 ContextCorrupt   : 0y1
   +0x004 AddressValid     : 0y1
   +0x004 MiscValid        : 0y1
   +0x004 ErrorEnabled     : 0y1
   +0x004 UncorrectedError : 0y1
   +0x004 StatusOverFlow   : 0y1
   +0x004 Valid            : 0y1
   +0x000 QuadPart         : 0xffff0210`52455043

A generic data write instruction error. It's difficult to say whether this data structure is correct because it's undocumented, and doesn't seem to work half of the time, so I maybe wrong, thus it might be better to look at the WHEA error structure instead.
Anyway, it seems with the context corrupt, the error could be different, slightly. This doesn't mention the level 0 cache, but rather a generic cache error.
If what I'm seeing is correct, it looks like the instruction itself is wrong.
The context is only partially valid, only x86 is available. The thread has a typing mismatch, making searching through previous executions almost impossible.
The stack seems to be zeroed, but that might be because of the context corruption.

Just see how it goes with no overclocking, either manual or auto.
Then report back when you feel necessary.

 

[PrTv]

Thanks for your detailed analysis.

However, please forgive my ignorance, but I couldn't understand half of what you said.

Anyway, I tried uninstalling MSI Afterburner and still got that dreaded BSOD, so it seems Afterburner may not be the cause. But I will not reinstall it until the matter is resolved.

Strangely enough, I could not stop my MB from auto O/Cing my CPU even though I set the number manually in the BIOS, the processor still runs at 4.4 Ghz at all cores full loaded.

Last thing I can try is manually "under clock" the CPU at 3.5 Ghz and use Vcore at 1.200. Doing this, I should be able to eliminate the CPU factor (unless my 4790K is faulty).

BTW, can you please explain what I quoted below in layman terms? As at present, I only encountered this issue when playing game (Witcher 3 to be specific), so I would like to also know if it is the game itself that causes this issue.

A generic data write instruction error. It's difficult to say whether this data structure is correct because it's undocumented, and doesn't seem to work half of the time, so I maybe wrong, thus it might be better to look at the WHEA error structure instead.
Anyway, it seems with the context corrupt, the error could be different, slightly. This doesn't mention the level 0 cache, but rather a generic cache error.
If what I'm seeing is correct, it looks like the instruction itself is wrong.
The context is only partially valid, only x86 is available. The thread has a typing mismatch, making searching through previous executions almost impossible.
The stack seems to be zeroed, but that might be because of the context corruption.

I actually did some search but could not find anybody playing this game reporting this problem. Is it possible that my CPU or MB is faulty (that needs to be RMA)? Based on the error code that points to the CPU (if I understand correctly), is it possible that this error may be caused by VGA driver or bad sector on HDD?

Again, thanks a lot for your help Patrick and Jared. Really appreciate it.

That was originally said by Patrick, it's the auto overclocking that is most likely the problem.

Those crashes won't be the same, just because the parameters passed to KeBugCheckEx.
Parameters passed, depend on the type of crash, most 0x124s with a machine check exception pass almost identical information. The main thing that differs will be the addresses to the low and high bits of the MCi_STATUS structure, which will vary on different boots depending on where it will be loaded in memory. The MCi_STATUS record sends the hardware error information, via a machine check exception, to the operating system, which determines what to do, in this case, a bugcheck.

EDIT:
When this happened, my CPU and GPU temps were fine as Noctua U14s is one of the best air-cooling out there and the CPU temp did not exceed 60c when this happened.

It's not necessarily a temperature issue, it's more than likely a voltage problem, excess (probably this) or lack of (doubtful).

---

Then issues started to surface when:

- I tried to O/C my VGA and to test the stability, I left EVGA Precision-X open on one screen and play game on the other (main screen). That was when I got my first BSOD (it was about memory_management BSOD though). I used Windows Memory Diagnostic to check my ram afterward and the result was OK.
- When I got that BSOD, I decided... OK, f**k this and gave up my idea of O/C-ing. I then restarted and make sure that my GPU (and Mem) clock speed was back to normal.

I then tried the game again, and this time with MSI Afterburner loaded, to monitor my GPU/CPU usage and temp. I came up with this 0x0000124 errors on two occasions. First time when a huge scene was loaded (the game loaded the scene and I was about to walk and then BSOD). Another time when I tried to save, it BSODed 0x0000124 when I opened the save dialog. The game is Witcher 3, of course.

Not only is overclocking software garbage, that barely works properly. The drivers themselves are crap, another reason to avoid it.

hal!_MCI_STATUS
   +0x000 McaErrorCode     : 0x5043 // 0101 0000 [COLOR=#FF0000]0100[/COLOR] [COLOR=#008000]00[/COLOR][COLOR=#0000FF]11[/COLOR]
   +0x002 ModelErrorCode   : 0x5245
   +0x004 OtherInformation : 0y11111110000001000010000 (0x7f0210)
   +0x004 ActionRequired   : 0y1
   +0x004 Signalling       : 0y1
   +0x004 ContextCorrupt   : 0y1
   +0x004 AddressValid     : 0y1
   +0x004 MiscValid        : 0y1
   +0x004 ErrorEnabled     : 0y1
   +0x004 UncorrectedError : 0y1
   +0x004 StatusOverFlow   : 0y1
   +0x004 Valid            : 0y1
   +0x000 QuadPart         : 0xffff0210`52455043

A generic data write instruction error. It's difficult to say whether this data structure is correct because it's undocumented, and doesn't seem to work half of the time, so I maybe wrong, thus it might be better to look at the WHEA error structure instead.
Anyway, it seems with the context corrupt, the error could be different, slightly. This doesn't mention the level 0 cache, but rather a generic cache error.
If what I'm seeing is correct, it looks like the instruction itself is wrong.
The context is only partially valid, only x86 is available. The thread has a typing mismatch, making searching through previous executions almost impossible.
The stack seems to be zeroed, but that might be because of the context corruption.

Just see how it goes with no overclocking, either manual or auto.
Then report back when you feel necessary.

 

[Jared]

Thanks for your detailed analysis.

However, please forgive my ignorance, but I couldn't understand half of what you said.

Anyway, I tried uninstalling MSI Afterburner and still got that dreaded BSOD, so it seems Afterburner may not be the cause. But I will not reinstall it until the matter is resolved.

Strangely enough, I could not stop my MB from auto O/Cing my CPU even though I set the number manually in the BIOS, the processor still runs at 4.4 Ghz at all cores full loaded.

Last thing I can try is manually "under clock" the CPU at 3.5 Ghz and use Vcore at 1.200. Doing this, I should be able to eliminate the CPU factor (unless my 4790K is faulty).

It still goes to 4.4GHz because that is turbo boost, I have the same CPU, it's normal.

BTW, can you please explain what I quoted below in layman terms? As at present, I only encountered this issue when playing game (Witcher 3 to be specific), so I would like to also know if it is the game itself that causes this issue.

A generic data write instruction error. It's difficult to say whether this data structure is correct because it's undocumented, and doesn't seem to work half of the time, so I maybe wrong, thus it might be better to look at the WHEA error structure instead.
Anyway, it seems with the context corrupt, the error could be different, slightly. This doesn't mention the level 0 cache, but rather a generic cache error.
If what I'm seeing is correct, it looks like the instruction itself is wrong.
The context is only partially valid, only x86 is available. The thread has a typing mismatch, making searching through previous executions almost impossible.
The stack seems to be zeroed, but that might be because of the context corruption.

I actually did some search but could not find anybody playing this game reporting this problem. Is it possible that my CPU or MB is faulty (that needs to be RMA)? Based on the error code that points to the CPU (if I understand correctly), is it possible that this error may be caused by VGA driver or bad sector on HDD?

Again, thanks a lot for your help Patrick and Jared. Really appreciate it.

It's usually a CPU error when this occurs, upload the rest of the dump files to check for consistency.
But you can, very rarely, get software conflicts.
Is the CPU under warranty?

 

[PrTv]

I'm back to tell you guys that I believe I found the source of the issue.

Like you guys said, it's my CPU. As I said before, my M/B automatically O/Ced my CPU to 4.4Ghz (when all four cores are loaded), and at the same time, I lowered my VCore to try to maintain proper temperature. I got it to 1.132 and I though it was stable, as I could throw anything at it (game, transcoding, image edition, etc) for almost a week and I never crashed once.

Then the machine started to crash on its own a week later. This is why I initially said I wouldn't believe it was my CPU.

Anyway, I tried increasing the VCore to 1.200 and have been testing it with various things for more than a day, and I've encountered any crash yet (finger crossed).

So thanks Patrick and Jared for helping me out. I definitely have learned something here.