Gigabyte BIOS update won't stick; thoughts?

[UsandThem]

Alright, about 9 hours to run Memtest86 and it's a Pass. No errors detected. That's with XMP Profile1 and no other modifications.

Good news: I've got F5 installed! Thanks, Edrick. I'm going to run this for a few weeks with the XMP Profile and see how it goes.

UsandThem, there's a new option in the F5 BIOS which I didn't see before. Under the "BIOS Features" tab there's an option to enable something called "TBT USB3.1 Force Power." Unfortunately there's no documentation and the summary in BIOS only says "Enable / Disable TBT USB3.1 Force Power"... gee thanks, Gigabyte :sneaky:

No clue what "Force Power" means but maybe it's related to the issue you're having (?)

Thanks for the update. I'm going to get a USB 3.1 cable first and try it. If it works, then the issue would be on the external enclosure.

If I still have problems, then the issue will be on the onboard controller. Then I guess I will have to move up to F5 to see if it fixes it.

 

[FlippedBit]

I'm not having good luck with stability. Fired up the desktop tonight and got a crash with error code 0xa (IRQL_NOT_LESS_OR_EQUAL). I've seen this one before.

After restart the system did an automatic disk check on my C: partition.

Starting to wonder if my SSD is not liked by this motherboard. I've got a Crucial MX100 on SATA port 0, in AHCI mode. Intel RST drivers installed, but apparently so does UsandThem (without issues).

 

[UsandThem]

I'm not having good luck with stability. Fired up the desktop tonight and got a crash with error code 0xa (IRQL_NOT_LESS_OR_EQUAL). I've seen this one before.

After restart the system did an automatic disk check on my C: partition.

Starting to wonder if my SSD is not liked by this motherboard. I've got a Crucial MX100 on SATA port 0, in AHCI mode. Intel RST drivers installed, but apparently so does UsandThem (without issues).

I had that same type of issue on my son's HP. The system came with a Sandisk X300 SSD, and it kept randomly giving those errors.

How old is the MX100 drive you have? Try running the Crucial Executive utility on it and see if it reports anything odd: http://www.crucial.com/usa/en/support-storage-executive.

When I ran the Sandisk utility on the X300, it gave several warnings about the drive's health even though it was brand new.

I did encounter another odd issue with my UD5 today though: I bought another Noctua fan to go dual fan on my heatsink, and when I hooked both of them up to my CPU header using the supplied Noctua Y-cable, my system would turn on and go to the splash screen and hang. After several attempts, I ended up having to plug one into my CPU_Fan header and the other one into my CPU_OPT header.

Most people who have dual fans can just use the Y-adapter and be fine......but no go for me on this board for some reason.

 

[FlippedBit]

I had that same type of issue on my son's HP. The system came with a Sandisk X300 SSD, and it kept randomly giving those errors.

How old is the MX100 drive you have?

I don't remember exactly but I've had it at least a couple years. Before I upgraded to Windows 10 recently I was running the Crucial Storage Executive app, and the drive was in good health. It's at 98% wear.

I'm running the Intel RST app now and it's also not showing any errors. CrystalDiskInfo also reports "Good."

The thing it has in common with that Sandisk X300 is the Marvell 88SS918* series controller. That's interesting.

I haven't heard of this issue yet with the Y power cable for dual fans. I'm running a single fan myself, the one that came with my Cryorig H7.

 

[UsandThem]

I don't remember exactly but I've had it at least a couple years. Before I upgraded to Windows 10 recently I was running the Crucial Storage Executive app, and the drive was in good health. It's at 98% wear.

I'm running the Intel RST app now and it's also not showing any errors. CrystalDiskInfo also reports "Good."

The thing it has in common with that Sandisk X300 is the Marvell 88SS918* series controller. That's interesting.

I haven't heard of this issue yet with the Y power cable for dual fans. I'm running a single fan myself, the one that came with my Cryorig H7.

I found this thread that might help you figure out what is causing your problems:

http://www.tomshardware.com/answers/id-2745512/irql-equal-error-windows.html

http://www.pcworld.com/article/2999...ity-monitor-to-fix-and-fine-tune-your-pc.html

 

[vailr]

I've got a Crucial MX100

Check for a possible firmware update:
http://www.crucial.com/usa/en/support-ssd-firmware

 

[FlippedBit]

@UsandThem Thanks. I'm familiar with the reliability monitor but I need to spend more time digging into the core dumps.

@vailr Thanks, I've already got the latest f/w (MU02)

 

[Ketchup]

When I first got my current board, I would get a similar blue screen every couple months. Not enough to make a person worry too much, but still annoying. After a couple new BIOSes (which I am not dissing now that I know you have an actual problem) and a new SATA cable, the problem hasn't come back. Now, I don't know which exactly fixed it. But I will say that Asus is much better at rolling out BIOS updates for issues like this, and not just rolling them out when they have a new feature or CPU they want to add support for (like Gigabyte).

 

[FlippedBit]

I get one about every few days, or once a week. What's annoying is that it's not always the same code. I don't remember having as many issues with my Asus P6T Deluxe, but at this point I'm not able to point at any one thing. Hard to say.

I just downloaded a tool called WhoCrashed and ran it on the last crashdump that I had. Here's what it said:

[FONT=Segoe UI, Arial]

On Sun 3/27/2016 12:17:54 AM GMT your computer crashed
crash dump file: C[FONT=Segoe UI, Arial]:\[/FONT]Windows\Minidump\032616-8125-01.dmp
This was probably caused by the following module: [FONT=Segoe UI, Arial]ntkrnlmp.exe[/FONT] (nt!KeBugCheckEx+0x0)
Bugcheck code: 0xA (0xFFFFD000AC06AA10, 0x2, 0x1, 0xFFFFF8010725BAC3)
Error: [FONT=Segoe UI, Arial]IRQL_NOT_LESS_OR_EQUAL[/FONT]
Bug check description: This indicates that Microsoft Windows or a kernel-mode driver accessed paged memory at DISPATCH_LEVEL or above.
This appears to be a typical software driver bug and is not likely to be caused by a hardware problem.
The crash took place in the Windows kernel. Possibly this problem is caused by another driver that cannot be identified at this time.

[FONT=Segoe UI, Arial]So [FONT=Segoe UI, Arial]as a next step, I followed the article here[FONT=Segoe UI, Arial] [FONT=Segoe UI, Arial]([/FONT][/FONT]http://www.reviversoft.com/blog/2013/09/using-driver-verifier-to-fix-a-blue-screen-of-death/[FONT=Segoe UI, Arial])[/FONT] [FONT=Segoe UI, Arial]to enable [FONT=Segoe UI, Arial]driver debug [FONT=Segoe UI, Arial](verifier). When i did that my comp[FONT=Segoe UI, Arial]u[FONT=Segoe UI, Arial]ter wouldn't boot[FONT=Segoe UI, Arial]. I[FONT=Segoe UI, Arial]t wo[FONT=Segoe UI, Arial]uld show the G[FONT=Segoe UI, Arial]igabyte spla[FONT=Segoe UI, Arial]sh screen a few [FONT=Segoe UI, Arial]times and then Windows [FONT=Segoe UI, Arial]recovery would come up. I had to go into Safe Mode and disable the driver veri[FONT=Segoe UI, Arial]fier. I'm going to try again, but only enable debug on [FONT=Segoe UI, Arial]a few drivers at a time.[/FONT]

[FONT=Segoe UI, Arial][FONT=Segoe UI, Arial][FONT=Segoe UI, Arial]Still not ruling out the [FONT=Segoe UI, Arial]possibility[/FONT] of a hardware issue[FONT=Segoe UI, Arial], but at least [FONT=Segoe UI, Arial]knowing which driver to loo[FONT=Segoe UI, Arial]k at would be a good [FONT=Segoe UI, Arial]start.
[/FONT][/FONT][/FONT][/FONT][/FONT][/FONT][/FONT][/FONT][/FONT][/FONT][/FONT][/FONT][/FONT][/FONT][/FONT][/FONT][/FONT][/FONT][/FONT][/FONT][/FONT][/FONT][/FONT]

 

[FlippedBit]

Another thing I noticed is that most of the crashes seem to happen on bootup, after the computer had been shut down the night before. I use Power->Shut down instead of Power->Sleep because I noticed that my computer never stays asleep for very long. The few times I've tried I always come back to find it powered on and sitting on the Windows log-in screen.

 

[Ketchup]

If you still aren't getting anywhere, you can try ruling things out. For example, disable the Asmedia SATA controller on your board. See if that makes a difference.

 

[UsandThem]

Another thing I noticed is that most of the crashes seem to happen on bootup, after the computer had been shut down the night before. I use Power->Shut down instead of Power->Sleep because I noticed that my computer never stays asleep for very long. The few times I've tried I always come back to find it powered on and sitting on the Windows log-in screen.

I haven't had that issue very often after I disabled all the different components that would wake my PC up (LAN, mouse, etc). I have it where only my keyboard can take it out of sleep (other than the power button of course).

Now the only time it wakes up on its own is if Windows has downloaded an update that I didn't notice, and a restart is required.

I'm with you on trying to figure out random computer problems. Computer forums like this one used to have so many more people, and more often than not, you could find the answer or someone who had experienced the same issue. The 'For Sale' section here used to be the place to go to buy, sell, or trade. Now, it looks almost like a K-Mart location getting ready to go out of business.

I looked at the Gigabyte forums to see if anybody had a USB 3.1 issue or the dual-fans not working on the same header, but there is hardly any responses to the questions people ask.

 

[FlippedBit]

I looked at the Gigabyte forums to see if anybody had a USB 3.1 issue or the dual-fans not working on the same header, but there is hardly any responses to the questions people ask.

I didn't have much luck on the Gigabyte forums when I was asking about how to get the F5 BIOS loaded. One of the regulars there, who I felt was trying to be genuinely helpful, could only offer that I should submit a bug to Gigabyte. (which I tried and failed because the site wouldn't let me log in)

 

[FlippedBit]

from this cached google search, cached search

its probably best to keep it off as it relates to the apline ridgde/thunderbolt controller

Sorry, I meant to acknowledge this post earlier and I think I forgot. Thank you for finding that.

 

[FlippedBit]

I spent the last few hours messing around with the windows driver verifier tool and systematically disabling any that caused verification errors. Well- it's messy, but I learned some things.

First of all, I highly recommend these tools:
- "sfc /scannow" in Administrator console (to fix corrupt system files)
- "verifier" (to enable additional debug and stress on drivers)
- "WhoCrashed" (to read minidumps from crashes and tell you in plain English which driver caused it)

The first problem area was the AMD drivers, so I thoroughly uninstalled them and pulled the GPU. I'm now using the Intel iGPU to test (with the Windows basic display drivers).

After that, I started getting verification errors on a number of other 3rd party drivers, including Gigabyte software, Logitech software, etc. I uninstalled all the Gigabyte apps and the Logitech Gaming software, but left most of the others intact. I realized that the settings I was using in the driver verifier were too aggressive and they were picking up all sorts of arcane driver bugs which are probably not causing my issue.

Finally, I scanned for corrupt system files; found & fixed one (opencl.dll). I also dialed back the verifier debug settings to "Standard."

Now the system is running with standard level driver verification and no AMD GPU. Let's see how long until I get another blue screen. If I don't get any over the next week, then I guess I have some evidence to suspect AMD GPU and/or Catalyst drivers.

All of the Intel, Asmedia stuff checks out so far; so I'm probably OK on the storage front. Fingers crossed.

At the end of this little experiment I ended up with over a dozen crash dumps. Some of them are redundant because I was caught in a boot loop with the same crash happening 2-3 times before I could get into Safe Mode.

In all, these are all the drivers where I saw some sort of verification error:
[FONT=Segoe UI, Arial][FONT=Segoe UI, Arial]
[/FONT][/FONT]

rtkvhd64.sys (Realtek(r) High Definition Audio Function Driver, Realtek Semiconductor Corp.)
vboxnetlwf.sys (VirtualBox NDIS 6.0 Lightweight Filter Driver, Oracle Corporation)
vboxnetadp6.sys (VirtualBox NDIS 6.0 Host-Only Network Adapter Driver, Oracle Corporation)
spyder3.sys (Spyder3 USB Driver 1.0-1, )
lgvirhid.sys
atikmpag.sys
gdrv.sys

[FONT=Segoe UI, Arial][FONT=Segoe UI, Arial]

[/FONT][/FONT][FONT=Segoe UI, Arial]
[/FONT]

 

[FlippedBit]

I'm turning off the driver verifier because it's catching far too much and I'm not sure what's related to my issue and what's not.

When I booted up today I got this:

[FONT=Segoe UI, Arial]

On Mon 3/28/2016 10:20:17 PM GMT your computer crashed
crash dump file: C:\Windows\Minidump\032816-9828-01.dmp
This was probably caused by the following module: [FONT=Segoe UI, Arial]verifierext.sys[/FONT] (VerifierExt!ExFreePoolWithTag_wrapper+0x10)
Bugcheck code: 0xC1 (0xFFFFCF81D5646F90, 0xFFFFCF81D56461CC, 0x6DC070, 0x32)
Error: [FONT=Segoe UI, Arial]SPECIAL_POOL_DETECTED_MEMORY_CORRUPTION[/FONT]
file path: C:\Windows\system32\drivers\verifierext.sys
product: [FONT=Segoe UI, Arial]Microsoft® Windows® Operating System[/FONT]
company: [FONT=Segoe UI, Arial]Microsoft Corporation[/FONT]
description: Driver Verifier Extension
Bug check description: This indicates that the driver wrote to an invalid section of the special pool.
This appears to be a typical software driver bug and is not likely to be caused by a hardware problem.
The crash took place in a standard Microsoft module. Your system configuration may be incorrect. Possibly this problem is caused by another driver on your system that cannot be identified at this time.

[FONT=Segoe UI, Arial]Memory corruption?[FONT=Segoe UI, Arial] Bad RAM?
[FONT=Segoe UI, Arial]Or j[FONT=Segoe UI, Arial]ust some driver writi[FONT=Segoe UI, Arial]ng [FONT=Segoe UI, Arial]where it shouldn't?

:\
[/FONT][/FONT][/FONT][/FONT][/FONT][/FONT][/FONT]

 

[Ketchup]

Remove Intel RST and see if there are any changes.

 

[UsandThem]

I'm turning off the driver verifier because it's catching far too much and I'm not sure what's related to my issue and what's not.

When I booted up today I got this:

[FONT=Segoe UI, Arial]

[FONT=Segoe UI, Arial]Memory corruption?[FONT=Segoe UI, Arial] Bad RAM?
[FONT=Segoe UI, Arial]Or j[FONT=Segoe UI, Arial]ust some driver writi[FONT=Segoe UI, Arial]ng [FONT=Segoe UI, Arial]where it shouldn't?

:\
[/FONT][/FONT][/FONT][/FONT][/FONT][/FONT][/FONT]

I might have missed it in previous posts, but what brand of ram, how much, and what speed do you have it running at (XMP)? Also, what channel are you running it in (A or B)?

I have RST installed on my system, and I haven't had any issues with it.

Also, when you installed Windows 10, you downloaded all the drivers directly from Gigabyte's website, correct (i.e did not install any off of the CD)? Your SSD is connected to one of the Intel SATA3 connectors, and not the ASM1061 SATA3 connectors correct?

Oustside of that, the only other thing I would look at is to take a look at your voltages in the System Information Viewer (system alert tab) to make sure they look solid, and if they do, maybe try another SSD or hard drive if you have one.

Something is definitely not working good in your system, and it's tough to tell if the source is hardware or software based.

 

[FlippedBit]

Remove Intel RST and see if there are any changes.

Alright, Intel RST is gone. Back to default AHCI driver.

Turned driver verifier back on. Let's see If the crashes go away then I'll add the GPU back in.

By the way the performance difference between the two AHCI drivers seems pretty small. I might just keep the MS default.

Intel on left. Microsoft on right.

 

[FlippedBit]

I might have missed it in previous posts, but what brand of ram, how much, and what speed do you have it running at (XMP)? Also, what channel are you running it in (A or B)?

I have RST installed on my system, and I haven't had any issues with it.

Also, when you installed Windows 10, you downloaded all the drivers directly from Gigabyte's website, correct (i.e did not install any off of the CD)? Your SSD is connected to one of the Intel SATA3 connectors, and not the ASM1061 SATA3 connectors correct?

Oustside of that, the only other thing I would look at is to take a look at your voltages in the System Information Viewer (system alert tab) to make sure they look solid, and if they do, maybe try another SSD or hard drive if you have one.

Something is definitely not working good in your system, and it's tough to tell if the source is hardware or software based.

This is the kit: http://www.newegg.com/Product/Product.aspx?Item=N82E16820233752

CORSAIR Vengeance LPX 16GB (2 x 8GB) 288-Pin DDR4 SDRAM DDR4 2400 (PC4 19200) Memory Kit Model CMK16GX4M2A2400C14

It's running in the one and only XMP profile it has, XMP-2400. Timings & SPD info below:



It shows up as Slots #2 and #4 in CPUID, but it's actually installed in slots DDR4_1 and DDR4_2 on the motherboard. It seems like there's one stick each on Channel A and Channel B, but I followed the chart in the manual. That's what it says to do with just two sticks.

All the drivers are from Gigabyte's site, except I think I updated Intel RST from Intel's website at some point.

Yep, the SSD shows up on SATA port 0 in BIOS.

I don't have the Gigabyte tools installed anymore... will work on getting some voltages.

 

[FlippedBit]

I has the volts.

Sorry for the crap pic.

 

[Ketchup]

...By the way the performance difference between the two AHCI drivers seems pretty small. I might just keep the MS default....

I found the same. Intel RST is of little benefit on a single drive. It does make for an extra service and startup program, so there are good reasons to leave it off.

 

[UsandThem]

I has the volts.

Sorry for the crap pic.

Man, I guess I am out of ideas on what's causing your crashes. I just haven't had anything like that with my system.

What you are dealing with these random crashes reminds me of when I first joined Anandtech trying to figure out all the stability issues I was having with an Abit BP6 motherboard.

I never could figure out why it sucked so bad, so I trashed it out of frustration and replaced it with an Asus board which was great.

I guess the only thing you can do at this point is to keep checking the drivers and software in your system (I assume you have already ran MSCONFIG or task manager to check start up programs and processes).

After that, you could try resetting Windows 10 to see if the issue goes away. Install only needed drivers and see if the issue appears again. If it does, then comes the fun part of isolating which part could be flaky.

I know it sucks because I did the same thing with that Abit board.

If you do figure it, please post your solution here to help others out.

Lastly, here are a couple utilities that might help you in your troubleshooting:

http://www.howtogeek.com/222730/how-to-find-out-why-your-windows-pc-crashed-or-froze/

 

[vailr]

That "verifier" program itself is likely the cause of BSODs.
I'd advise disabling that software, and not using it for any purpose.
With it running, got a BSOD while simply updating the Realtek audio driver.

 

[FlippedBit]

I might have a clue about where my problem area is. Here's my thought process...

Last night I turned the driver verifier off, and this morning I got another BSOD at startup. As before, it's the IRQL_NOT_LESS_OR_EQUAL message; the one I'm hunting. Great!

Comparing it with the earlier IRQL_xxx message, I see that it's in the same memory space (0xFFFFF....). Something is accessing this memory and crashing.

So I looked at the memory mappings in the Windows System Information viewer, and the space 0xFF000000 - 0xFFFFFFFF belongs to a device called "Legacy device." Right away I'm thinking "south bridge / chipset."

So I go into Device Manager, set it to "Show Resources by Type" and expand the Memory tree. I find the device called "Legacy device" and righlt-click->Properties. I check and see that the Parent Device is "PCI."

Well, I don't have any legacy PCI devices (just the 3.0 x16 card for the GPU which is not on the SB/chipset bus on Skylake, anyway).

So now I'm thinking, either:

1. Chipset driver bug
2. Bad USB device plugged in somewhere (I've got a bunch of old USB2 devices on a hub)
3. Bad USB device driver
4. Bad chipset or motherboard peripheral (serial, PS/2 port.. who knows)

I'm hoping for #2 or 3. When I get some time I'm going to update the chipset driver from Intel and as a second test, if needed, I'll pull my USB2 hub and all the stuff I don't need and go from there.

I really do hope I'm on the right track here.