[ChiefZeke]

Installed Everest Ultimate 5.30.1913 in XP Pro. Created shortcut to Everest in XP Pro x64, Win7 32- and 64-bit.

Build 1931 starts ok in all versions except for Win7 32-bit. System just froze; had to do hard reboot. (Also happened in 1911 - downloaded and installed 1913 before I had a chance to report on 1911.) (Did not have any problems with betas prior to 1911.)

When in Win7 32-bit and clicking on shortcut the Everest logo icon appears and main window then appears and freeze happens. Have no idea why and, obviously, no report can be created.

Anyone else having a similar problem?

Anything I can do to create any kind of error report?

[Fiery]

Let us know more about your hardware, please. Motherboard model, chipset type, CPU type, video card(s) model, audio controller type, any USB or FireWire devices connected, etc.

BTW, have you tried to delete EVEREST.INI file before starting EVEREST? That way you can reset all settings to their default.


Thanks,
Fiery

[ChiefZeke]

I will try deleting the .INI first.

Since it works in XP Pro, XP Pro x64, and Win7 64-bit is there a partial report or similar I can copy or extract from within that operating system that might help when I run Everest?

[ChiefZeke]

The hardware is:
ABIT IX48 GT3 w/BIOS 11 (no other versions available)
Intel Core 2 Quad Q9650 cpu
Kingston KVR1333D3N9K2/4G memory - total of 8Gb installed
Western Digital 1Tb SATA 11 32Mb cache drives (two each)
Plextor PX-712Sa and PX-716SA Internal SATA drives
Diamond Radeon HD 4870 Video card 1Gb GDDR5 RAM, PCI Express (two each) connected in CrossfireX mode
Creative SoundBlaster PCI Express X-Fi Titanium Fatal1ty audio card
PC Power & Cooling Turbo Cool 860 Power Supply
Lian-Li PC-A77 Full tower case
Samsung SyncMaster225bw (two each) one connected to DVI one to analog

Will test later tonight or tomorrow with .INI deleted. Results then.

I will be leaving Southern California for Florida early Sunday morning, the 18th, and I do not expect to return to California till 02 November - so if I miss things tonight or tomorrow then all will be on hold till I return. (Going to my 50th high school reunion.)

[ChiefZeke]

Deletion of Everest.ini made no difference. Program was started, main windows appeared and as soon as any item was selected the freeze happened. Hard reboot required.

Booted into XP Pro and Win7 64-bit and Everest worked just fine.

[ChiefZeke]

More info about Win7 and Everest 5.30.1913.

Tried in Win7 64-bit again - just fine.

Tried in Win7 32-bit again with previous .INI file - locked up after main window appeared and never did see the OSD which appears as a string of boxes with numbers in them as part of the taskbar.

Moved .INI to protect it; tried again - started ok but no OSD and no boxes with numbers. Tried a couple of selections in main window; appeared to be ok until I selected Computer > Sensor - then freeze occured again. Hard reboot and verified when freeze occurs.

Hard reboot then restarted from Win7 to XP Pro where I am now. Using original .INI started Everest build 1913 and it's working just fine; no freezes and OSD boxes with numbers appeared in taskbar.

So freeze appears to have something to do with reading the sensors under Win7 32-bit.

My trip to Florida starts early tomorrow, California time, so further questions/answers will be delayed for about 17 days unless something appears soon (my time).

[Canis-X]

I am having the same issue...I think....I have error logs reported in the system event viewer log to add.

Here is my setup info:

PHENOM II X4 955 BE at stock speeds manually entered stock voltages for CPU, DRAM, NB & SB
ASUS CROSSHAIR III FORMULA ( R.O.G. )[BIOS vr 1003] | (2) SAPPHIRE 4870 X2's (X-FIRED)[beta 9.10 RC drivers, also tried v9.9]
8GB (4 X 2GB) MUSHKIN XP ASCENT DDR3 @1600 [7-7-6-18]{tried 4 GB 2x2GB @1333)
AUZENTECH X-Fi PRELUDE 7.1
(2) 300GB VELOCORAPTOR (AHCI) + (3) 1TB WD CAVIAR GREEN
KINGWIN Mach 1 1220W PSU | DUAL OS: Win7 Ultimate RTM x32 & Win7 Ultimate RTM x64

I can open Everest and run the mem benchmark. I cannot run the stability test or select the sensors applet. Doing so locks the system up and it reboots.

Event viewer logs:

****************************************************************************************************************************************************************************************************
Log Name: System
Source: Microsoft-Windows-WHEA-Logger
Date: 10/19/2009 7:22:48 PM
Event ID: 20
Task Category: None
Level: Error
Keywords:
User: LOCAL SERVICE
Computer: XXXXXXXXXXXXXXXXXXXXX
Description:
A fatal hardware error has occurred.

Component: AMD Northbridge
Error Source: Machine Check Exception
Error Type: HyperTransport Watchdog Timeout Error
Processor ID: 0

The details view of this entry contains further information.
Event Xml:
<Event xmlns="http://schemas.microsoft.com/win/2004/08/events/event">
<System>
<Provider Name="Microsoft-Windows-WHEA-Logger" Guid="{C26C4F3C-3F66-4E99-8F8A-39405CFED220}" />
<EventID>20</EventID>
<Version>0</Version>
<Level>2</Level>
<Task>0</Task>
<Opcode>0</Opcode>
<Keywords>0x8000000000000000</Keywords>
<TimeCreated SystemTime="2009-10-20T00:22:48.784079400Z" />
<EventRecordID>4073</EventRecordID>
<Correlation ActivityID="{4815021E-C245-4C3E-B5E3-3056A5805891}" />
<Execution ProcessID="1700" ThreadID="2112" />
<Channel>System</Channel>
<Computer>XXXXXXXXXXXXXXXXXXXXX</Computer>
<Security UserID="S-1-5-19" />
</System>
<EventData>
<Data Name="ErrorSource">3</Data>
<Data Name="ApicId">0</Data>
<Data Name="MCABank">4</Data>
<Data Name="MciStat">0xf200000000070f0f</Data>
<Data Name="MciAddr">0x0</Data>
<Data Name="MciMisc">0x0</Data>
<Data Name="ErrorType">7</Data>
<Data Name="Length">928</Data>
<Data Name="RawData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ata>
</EventData>
</Event>
****************************************************************************************************************************************************************************************************



****************************************************************************************************************************************************************************************************
Log Name: System
Source: Microsoft-Windows-WER-SystemErrorReporting
Date: 10/19/2009 7:22:45 PM
Event ID: 1001
Task Category: None
Level: Error
Keywords: Classic
User: N/A
Computer: XXXXXXXXXXXXXXXX
Description:
The computer has rebooted from a bugcheck. The bugcheck was: 0x00000124 (0x00000000, 0x8667d63c, 0x00000000, 0x00000000). A dump was saved in: C:\Windows\Minidump\101909-38859-01.dmp. Report Id: 101909-38859-01.
Event Xml:
<Event xmlns="http://schemas.microsoft.com/win/2004/08/events/event">
<System>
<Provider Name="Microsoft-Windows-WER-SystemErrorReporting" Guid="{ABCE23E7-DE45-4366-8631-84FA6C525952}" EventSourceName="BugCheck" />
<EventID Qualifiers="16384">1001</EventID>
<Version>0</Version>
<Level>2</Level>
<Task>0</Task>
<Opcode>0</Opcode>
<Keywords>0x80000000000000</Keywords>
<TimeCreated SystemTime="2009-10-20T00:22:45.000000000Z" />
<EventRecordID>4016</EventRecordID>
<Correlation />
<Execution ProcessID="0" ThreadID="0" />
<Channel>System</Channel>
<Computer>XXXXXXXXXXXXXXXXXXXXX</Computer>
<Security />
</System>
<EventData>
<Data Name="param1">0x00000124 (0x00000000, 0x8667d63c, 0x00000000, 0x00000000)</Data>
<Data Name="param2">C:\Windows\Minidump\101909-38859-01.dmp</Data>
<Data Name="param3">101909-38859-01</Data>
</EventData>
</Event>
****************************************************************************************************************************************************************************************************
dmp file attached

I have replaced the MoBo, RAM, and CPU so far. I know that everything worked fine in XP PRO x86 and Vista Ultimate SP2 x64.

Attached File(s)

•  101909-38859-01.zip (10.37K)
  Number of downloads: 0

This post has been edited by Canis-X: 20 October 2009 - 06:01 PM

[Fiery]

The only thing I can find common in the two configs you have is that both uses Radeon HD 4870 generation video cards in CrossFire mode. If it's not too much trouble, can you please try to remove one of the video cards from the system to physically disable CrossFire mode? Let me know if that fixes the lock up.

Meanwhile we'll get two identical HD 4870 cards to run the same test in our lab as well.

[Canis-X]

I can, but my cards have 2 GPU's on one board, so technically it is still cross fired internal on a single board...will this matter?

[Fiery]

Canis-X, on 20 October 2009 - 09:47 PM, said:

I can, but my cards have 2 GPU's on one board, so technically it is still cross fired internal on a single board...will this matter?

Maybe you're right, I'm not sure. But quite frankly, other than blaming Catalyst Win7 32-bit driver, I cannot think of a possible explanation on this issue. We'll know more once we can get our CrossFire system up'n'running. We do have an older CrossFire system, based on X1600 cards, but it's not Win7-ready

[Canis-X]

Fiery, on 20 October 2009 - 03:08 PM, said:

Maybe you're right, I'm not sure. But quite frankly, other than blaming Catalyst Win7 32-bit driver, I cannot think of a possible explanation on this issue. We'll know more once we can get our CrossFire system up'n'running. We do have an older CrossFire system, based on X1600 cards, but it's not Win7-ready

Hello, the issue still remains. I tested each of the two cards independantly, but I'm not sure how to disable the internal to the board Crossfire. I did run SiSandra last night for 4 hours straight with no reported errors. Do you have any results from your end since your last post? Any further troubleshooting steps that you would like me to try?

[Fiery]

We could reproduce the issue under Win7 32-bit using 2 identical Radeon HD 4850 cards in CrossFireX mode (Catalyst 9.9). It looks suspiciously like a driver bug, but we'll do more test runs (e.g. with HD 4870 X2), and try other software (e.g. GPU-Z, Rivatuner) as well. I'll report back in this topic when we know more.

[Canis-X]

Fiery, on 21 October 2009 - 04:01 PM, said:

We could reproduce the issue under Win7 32-bit using 2 identical Radeon HD 4850 cards in CrossFireX mode (Catalyst 9.9). It looks suspiciously like a driver bug, but we'll do more test runs (e.g. with HD 4870 X2), and try other software (e.g. GPU-Z, Rivatuner) as well. I'll report back in this topic when we know more.

You guys rock!!! I have posted this same issue in numerous places, Asus, AMD processor forums...etc....and you are the first ones to actually pick this up and doing anything with it at all!!! Grade A in my book, and I also work a support desk for my company, so I speak from experience!!!

Thanks for at least acknowledging my issue and then taking it an extra step farther!

[Canis-X]

If you don't mind me asking....what CPU / RAM / MoBo were you using....AMD (AM2+ / AM3), Intel / Corsair, Muskin, GSkill (DDR2 / DDR3 @ ?Mhz) / Asus, MSI, ASRock (chipset = 7xx series, 6xx series)....??? I'm trying to find any commonalities that I can bring to my other threads that I have opened for this matter.

This post has been edited by Canis-X: 21 October 2009 - 09:48 PM

[Canis-X]

Any news?

[Fiery]

Canis-X, on 26 October 2009 - 06:59 PM, said:

Any news?

Yes, we could finally implement a workaround today to fix this issue. Please upgrade to the following new EVEREST beta:

http://www.lavalys.c...wlkpyk3xq0r.zip

After upgrading to this new version, make sure to restart Windows to finalize the upgrade.

Let me know how it works.

---

As for some background info on this issue... We've been contacted by AMD back in April 2009:

"I'm from AMD's developer Relations team and need to talk to some about our new project and we think that your application may come to a halt once we release this feature. So we would like to talk to an engineer about sleep mode detection on our graphics cards."

So they actually warned us about the effects of their new Catalyst feature, but after we've sent a reply asking for more details about the possible issues, they didn't provide any technical info that we could use to avoid this pitfall. And in one of the recent Catalyst releases they've implemented the sleep mode feature (that they call ULPS), but apparently they only enable ULPS under Windows 7 with 32-bit kernel. This time they forgot to notify us about the implementation of ULPS and the lockup that it causes in EVEREST and other diagnostic software (e.g. GPU-Z).

As it turns out, ULPS is used to put the non-primary GPU(s) in a CrossFireX setup to sleep when the computer is not running a 3D application or a game. It's actually implemented quite smart, since even though the secondary GPU(s) are shut down, you can still read their relevant PCI registers and memory-mapped GPU registers without any problems. However, once you try to read the flash BIOS image out of the secondary GPU(s), the whole system locks up.

What we've done in the above beta is detect the status of ULPS via Registry (scan for "EnableUlps", and check if its value is non-zero), and when it is enabled and CrossFireX is enabled, then we don't read flash BIOS image from the secondary GPU(s). That seems to do the job until we can get a proper reply from AMD about this matter -- or maybe they can avoid such lockups by fixing their drivers.

However, by disabling flash BIOS readout for the secondary GPU(s), EVEREST cannot detect the factory defaults clocks and other BIOS details for the secondary GPU(s), hence the information you can see on the GPU page is only fully accurate for the primary GPU in ULPS-enabled CrossFireX configs.

I hope as a short term fix this would do the job, but in the long run we'd like to get a reply from AMD and hopefully we can work together with them to fix this issue without using any weird workarounds...

BTW, GPU-Z implemented the same workaround in its latest release.

[Canis-X]

Very good! I've downloaded it and I'll give it a try tonight when I get home and get some free time. Is there anything special that I should do before installing this workaround (i.e. uninstall the old version first, with/without a reboot....run a reg clean...etc)?

Again, thanks so much for taking the time to look into this and actaully acknowledging my request for help here...all the other forums I've posted in I have not one response...tons of views, but no responses!

Top notch!

[Fiery]

Canis-X, on 26 October 2009 - 08:53 PM, said:

Very good! I've downloaded it and I'll give it a try tonight when I get home and get some free time. Is there anything special that I should do before installing this workaround (i.e. uninstall the old version first, with/without a reboot....run a reg clean...etc)?

Just leave the old version installed, close it (if it's running), and download the new ZIP package, extract all its contents into the existing installation folder of EVEREST. Let it overwrite any existing files, then restart Windows to finalize the upgrade. That's it You can also do this form of upgrade with any future EVEREST beta releases.

Quote

Again, thanks so much for taking the time to look into this and actaully acknowledging my request for help here...all the other forums I've posted in I have not one response...tons of views, but no responses! Top notch!

We're happy to help, but this time it took quite a long time to deliver the fix, since we wanted to investigate this under 5 different Windows variants (XP 32-bit, Vista 32-bit, Vista 64-bit, Win7 32-bit, Win7 64-bit) with both a two-card HD4850 CrossFireX setup and a HD4870X2 card -- and also we had to contact a couple of experts to dig out some info on how ULPS and CrossFireX setups work, etc.

[Canis-X]

Ok, sorry it took me so long. Here is my status:

I can now view the sensor area now without issue....I can start the stability test without the system crashing with a BSOD, but I do get the following error.



There is no corresponding event log entry for me to reference and post here.

This post has been edited by Canis-X: 28 October 2009 - 03:18 PM

[Canis-X]

Continued....

With the above error the app hangs and I have to close it out in task manager. After that my system continues to behave poorly until I reboot it back into x86.

Side note:

Win7 x64 bit appears to be working like a charm now!