Firewire Win11 PFN LIST

om1ei · Post by **om1ei** » Tue Mar 28, 2023 6:48 am

I had a problem with firewire under windows 11, DPC latency and shared PCI IRQ (firewire and graphics card).
PowerSDR in Demo mode work ok. Powersdr with Flexradio occasionally threw an error 'PFN LIST CORRUPTED' and windows freze.
In my case, changing the PCI firewire card to another PCI slot did not help. The PCI IRQ remained the same.

My Solution:

1. With the LatencyMon program to check latency for realtime use.
2. If the drops, increase the "Processor performance core parking" value.
instructions on how to enable the item:
regedit
Find 'dec35c318583' only in 'Power Setings'
change 'Attributes' to 0

now accessible item:
Control Panel / Power Options / Procesor performance core parking (The original value was 4, I increased to 5)
I tried on several PCs and somewhere it was enough to increase the value to 5 and somewhere 40 somewhere 80.

Now latencymon no longer reports audio drops, and Powersdr does not freeze win11 even if the PCI IRQ remains shared with graphics card.

UPDATE (01.02.2024):
on w2jon's advice, I edited the Properties/Compatibility mode=Windows7 for file "FlexRadio.exe" and it hasn't frozen yet.

UPDATE [18.03.2024]:
After some other update Windows11 started freezing again.

YouTube · Post by **ke9ns** » Wed Mar 29, 2023 8:34 am

Thanks for finding that information.
I always keep my PC on "High performance" mode (and PowerSDR Setup->General->Options->Process Priority->High) but I forgot that Windows does not default to "High performance" mode. Good Catch.

I would say anyone running PowerSDR should set the Windows Control Panel->Power Options->High performance.

FYI
Found an explanation of what Core Parking is:

Core Parking is a sleep state (C6) supported by most newer x86 processors, and newer editions of Windows. Core Parking dynamically disables CPU cores in an effort to conserve power when idle. Disabled cores are re-enabled as the CPU load increases once again. This technology is very similar to frequency scaling, in that it seeks to throttle the CPU when idle.

The problem is that Window's default power profiles are configured far too aggressively when it comes to core parking, especially on workstations. Their interest was in conserving energy, even if this meant marginally decreasing performance. A number of complex parameters control when a core should be parked, and Microsoft tuned heavily towards power savings.

The core parking settings in Windows are implemented as parameters of power plans (aka power profiles). That means you can, for example, disable core parking for the High Performance power plan, but leave it enabled for other plans. And that is exactly the desired tweak for most users: disable parking only for high performance power plans.

The full string people should search for is:
[HKEY_LOCAL_MACHINE\SYSTEM\CurrentControlSet\Control\Power\PowerSettings\54533251-82be-4824-96c1-47b60b740d00\0cc5b647-c1df-4637-891a-dec35c318583] (Then set the attribute = 2)

Navigate to: Control Panel\All Control Panel Items\Power Options\Edit Plan Settings
Then click on "Change Advanced power settings"
Now under "Processor Power management" you will see an additional option of "Processor performance cor parking min cores"
Make sure to set your Power Plan to "High performance"

: SP32-20230329-084647.png (14.91 KiB) Viewed 22062 times

Darrin

w2jon · Post by **w2jon** » Tue Dec 19, 2023 8:56 pm

Analyzing the dump file after a PFN fault reveals the same offender for me each time.
Of course, this is all tripping up in the Flexradio.sys driver of which we have no source.
But for your reading pleasure please see the attached text excerpt from WinDbg. (process details are closer to the bottom)

Code: Select all

*******************************************************************************
*                                                                             *
*                        Bugcheck Analysis                                    *
*                                                                             *
*******************************************************************************

PFN_LIST_CORRUPT (4e)
Typically caused by drivers passing bad memory descriptor lists (ie: calling
MmUnlockPages twice with the same list, etc).  If a kernel debugger is
available get the stack trace.
Arguments:
Arg1: 000000000000009a, 
Arg2: 0000000000424f2e
Arg3: 0000000000000006
Arg4: 0000000000000002

Debugging Details:
------------------

*** WARNING: Check Image - Checksum mismatch - Dump: 0xd67b, File: 0xf8a0 - C:\ProgramData\Dbg\sym\hal.dll\1865382A6000\hal.dll

KEY_VALUES_STRING: 1

    Key  : Analysis.CPU.mSec
    Value: 4531

    Key  : Analysis.Elapsed.mSec
    Value: 4559

    Key  : Analysis.IO.Other.Mb
    Value: 0

    Key  : Analysis.IO.Read.Mb
    Value: 0

    Key  : Analysis.IO.Write.Mb
    Value: 0

    Key  : Analysis.Init.CPU.mSec
    Value: 437

    Key  : Analysis.Init.Elapsed.mSec
    Value: 55868

    Key  : Analysis.Memory.CommitPeak.Mb
    Value: 100

    Key  : Bugcheck.Code.KiBugCheckData
    Value: 0x4e

    Key  : Bugcheck.Code.LegacyAPI
    Value: 0x4e

    Key  : Dump.Attributes.AsUlong
    Value: 1800

    Key  : Dump.Attributes.DiagDataWrittenToHeader
    Value: 1

    Key  : Dump.Attributes.ErrorCode
    Value: 0

    Key  : Dump.Attributes.LastLine
    Value: Dump completed successfully.

    Key  : Dump.Attributes.ProgressPercentage
    Value: 100

    Key  : Failure.Bucket
    Value: 0x4E_9a_FlexRadio!unknown_function

    Key  : Failure.Hash
    Value: {dc24fd2d-7bb0-229c-be3e-43708b5f65a1}

    Key  : Hypervisor.Enlightenments.ValueHex
    Value: 1417cf94

    Key  : Hypervisor.Flags.AnyHypervisorPresent
    Value: 1

    Key  : Hypervisor.Flags.ApicEnlightened
    Value: 1

    Key  : Hypervisor.Flags.ApicVirtualizationAvailable
    Value: 0

    Key  : Hypervisor.Flags.AsyncMemoryHint
    Value: 0

    Key  : Hypervisor.Flags.CoreSchedulerRequested
    Value: 0

    Key  : Hypervisor.Flags.CpuManager
    Value: 1

    Key  : Hypervisor.Flags.DeprecateAutoEoi
    Value: 0

    Key  : Hypervisor.Flags.DynamicCpuDisabled
    Value: 1

    Key  : Hypervisor.Flags.Epf
    Value: 0

    Key  : Hypervisor.Flags.ExtendedProcessorMasks
    Value: 1

    Key  : Hypervisor.Flags.HardwareMbecAvailable
    Value: 1

    Key  : Hypervisor.Flags.MaxBankNumber
    Value: 0

    Key  : Hypervisor.Flags.MemoryZeroingControl
    Value: 0

    Key  : Hypervisor.Flags.NoExtendedRangeFlush
    Value: 0

    Key  : Hypervisor.Flags.NoNonArchCoreSharing
    Value: 1

    Key  : Hypervisor.Flags.Phase0InitDone
    Value: 1

    Key  : Hypervisor.Flags.PowerSchedulerQos
    Value: 0

    Key  : Hypervisor.Flags.RootScheduler
    Value: 0

    Key  : Hypervisor.Flags.SynicAvailable
    Value: 1

    Key  : Hypervisor.Flags.UseQpcBias
    Value: 0

    Key  : Hypervisor.Flags.Value
    Value: 4853999

    Key  : Hypervisor.Flags.ValueHex
    Value: 4a10ef

    Key  : Hypervisor.Flags.VpAssistPage
    Value: 1

    Key  : Hypervisor.Flags.VsmAvailable
    Value: 1

    Key  : Hypervisor.RootFlags.AccessStats
    Value: 1

    Key  : Hypervisor.RootFlags.CrashdumpEnlightened
    Value: 1

    Key  : Hypervisor.RootFlags.CreateVirtualProcessor
    Value: 1

    Key  : Hypervisor.RootFlags.DisableHyperthreading
    Value: 0

    Key  : Hypervisor.RootFlags.HostTimelineSync
    Value: 1

    Key  : Hypervisor.RootFlags.HypervisorDebuggingEnabled
    Value: 0

    Key  : Hypervisor.RootFlags.IsHyperV
    Value: 1

    Key  : Hypervisor.RootFlags.LivedumpEnlightened
    Value: 1

    Key  : Hypervisor.RootFlags.MapDeviceInterrupt
    Value: 1

    Key  : Hypervisor.RootFlags.MceEnlightened
    Value: 1

    Key  : Hypervisor.RootFlags.Nested
    Value: 0

    Key  : Hypervisor.RootFlags.StartLogicalProcessor
    Value: 1

    Key  : Hypervisor.RootFlags.Value
    Value: 1015

    Key  : Hypervisor.RootFlags.ValueHex
    Value: 3f7

    Key  : SecureKernel.HalpHvciEnabled
    Value: 0

    Key  : WER.OS.Branch
    Value: ni_release

    Key  : WER.OS.Version
    Value: 10.0.22621.1


BUGCHECK_CODE:  4e

BUGCHECK_P1: 9a

BUGCHECK_P2: 424f2e

BUGCHECK_P3: 6

BUGCHECK_P4: 2

FILE_IN_CAB:  MEMORY.DMP

DUMP_FILE_ATTRIBUTES: 0x1800

BLACKBOXBSD: 1 (!blackboxbsd)


BLACKBOXNTFS: 1 (!blackboxntfs)


BLACKBOXPNP: 1 (!blackboxpnp)


BLACKBOXWINLOGON: 1

PROCESS_NAME:  PowerSDR.exe

STACK_TEXT:  
ffffef86`1815ee78 fffff801`6844b1d6     : 00000000`0000004e 00000000`0000009a 00000000`00424f2e 00000000`00000006 : nt!KeBugCheckEx
ffffef86`1815ee80 fffff801`682aea1e     : ffffd300`0c9e7870 00000000`00000000 00000000`00000000 00000000`00000000 : nt!MiBadRefCount+0x42
ffffef86`1815eec0 fffff801`680ff6f4     : 00000000`00000000 00000000`00000000 ffff8303`00000002 0a000004`24f2ea63 : nt!MiDeleteNonPagedPoolPte+0x1af1fe
ffffef86`1815ef30 fffff801`680ff3ff     : 00000000`00000000 00000000`00007d00 ffff8302`70871000 fffff801`00000000 : nt!MiClearNonPagedPtes+0x16c
ffffef86`1815f170 fffff801`680fd6f6     : ffff8302`70871f80 fffff801`68a70dd8 00000000`00000001 00000000`00000002 : nt!MmFreePoolMemory+0xa3
ffffef86`1815f200 fffff801`680ff30d     : 00000000`00000000 ffffef86`1815f2f0 ffff8303`8fc00000 00000000`00000000 : nt!RtlpHpVaMgrCtxFree+0xaa
ffffef86`1815f250 fffff801`680fbd78     : 00000000`00000000 ffff8302`686c0e3c ffff8302`686c0e00 00000000`000071f8 : nt!RtlpHpFreeVA+0x7d
ffffef86`1815f2a0 fffff801`680fbbee     : 00000000`00100000 00000000`00000000 00000000`0000000a ffff8303`8fc00000 : nt!RtlpHpSegMgrRelease+0x134
ffffef86`1815f320 fffff801`680fb241     : ffffffff`ffffffff ffff8303`6ae00140 ffffffff`ffffffff 00000000`00000002 : nt!RtlpHpSegSegmentFree+0x7e
ffffef86`1815f350 fffff801`68121825     : 00000000`00000000 ffff8303`00000000 ffffef86`8fc00000 00000000`00000000 : nt!RtlpHpSegPageRangeShrink+0x3d1
ffffef86`1815f3d0 fffff801`688acce5     : ffff8303`656e6f4e fffff801`00000000 01000000`00000000 00000000`00004000 : nt!ExFreeHeapPool+0x4c5
ffffef86`1815f480 fffff801`b453520f     : 00000000`00000060 00000000`0000060c ffffdf81`d794f9b0 00000000`00000001 : nt!ExFreePool+0x25
ffffef86`1815f4c0 fffff801`b4535096     : ffff8303`8d4b5880 00000000`00000000 00000000`00000000 00000000`00000000 : FlexRadio+0x520f
ffffef86`1815f520 fffff801`b4545e97     : 00000000`00000000 ffff8303`76f19430 00000000`00222334 00000000`00000000 : FlexRadio+0x5096
ffffef86`1815f560 fffff801`b45323eb     : 00000000`00000000 ffff8303`76f19430 ffff8303`891c0000 00000000`00000000 : FlexRadio+0x15e97
ffffef86`1815f5b0 fffff801`b4531fc7     : ffff8303`76f19430 ffffef86`1815f6c1 ffff8303`76f19430 00000000`00000002 : FlexRadio+0x23eb
ffffef86`1815f5f0 fffff801`680ebef5     : ffff8303`42536f49 fffff801`68542417 00000000`00000000 00000000`00000000 : FlexRadio+0x1fc7
ffffef86`1815f620 fffff801`68540060     : ffff8303`76f19430 ffffef86`1815f6c1 ffff8303`8fb4be80 ffff8303`8fb4be80 : nt!IofCallDriver+0x55
ffffef86`1815f660 fffff801`68541a90     : ffff8303`76f19430 00000000`00000008 ffffef86`1815fa05 ffff8303`8fb4be80 : nt!IopSynchronousServiceTail+0x1d0
ffffef86`1815f710 fffff801`68541376     : 00000000`0000001b 00000000`00000001 00000000`00000000 00000000`00000000 : nt!IopXxxControlFile+0x700
ffffef86`1815f900 fffff801`6822bbe5     : ffffef86`1815f9f8 00000000`00000000 00000000`00000000 ffff8303`8ad11e90 : nt!NtDeviceIoControlFile+0x56
ffffef86`1815f970 00000000`76fd1cf3     : 00000000`00000000 00000000`00000000 00000000`00000000 00000000`00000000 : nt!KiSystemServiceCopyEnd+0x25
00000000`0178ec48 00000000`00000000     : 00000000`00000000 00000000`00000000 00000000`00000000 00000000`00000000 : 0x76fd1cf3


SYMBOL_NAME:  FlexRadio+520f

MODULE_NAME: FlexRadio

IMAGE_NAME:  FlexRadio.sys

STACK_COMMAND:  .cxr; .ecxr ; kb

BUCKET_ID_FUNC_OFFSET:  520f

FAILURE_BUCKET_ID:  0x4E_9a_FlexRadio!unknown_function

OS_VERSION:  10.0.22621.1

BUILDLAB_STR:  ni_release

OSPLATFORM_TYPE:  x64

OSNAME:  Windows 10

FAILURE_ID_HASH:  {dc24fd2d-7bb0-229c-be3e-43708b5f65a1}

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

YouTube · Post by **ke9ns** » Wed Dec 20, 2023 11:47 am

Yup.. Thanks

I seems like the PFN faults started to occur after Microsoft started to make "Security" updates to FireWire drivers. Putting limitations on to the original specs due to security concerns.
Newer FireWire chips seem to work better along with making sure you dont have an IRQ conflict
BUT... I was forced to add STOP/START commands into PowerSDR code, whenever you make changes related to DSP to help prevent the PFN crash.
I have periodically made updates to the software to shorten the STOP/START, but ended up going back

The current STOP/START software has cured the PFN for me, but not necessarily for everyone.

And its virtually impossible for me to get to the root cause since the computer crashes.

I would really like to find out if anyone using a USB-C Thunderbolt / FireWire adapter can still get a PFN crash or not?

Darrin

w2jon · Post by **w2jon** » Thu Dec 21, 2023 7:03 pm

I have the option and I was thinking about testing that myself.
But instead of switching to the Lenovo or Macbook Pro, actually what I think may be of more value is if I can get TB hooked up on my current machine so that I'll be testing with the same machine and software complexities as well.
I typically see these PFN errors whenever the DttSP stream is stopped and started. i.e. when the hourly DttSP reset is invoked or when enabling or disabling a VAC. But of course that's all the same things you've already researched.

The most annoying time is upon start-up when the radio completes initializing and begins to play.
I've tried to step into the process with the debugger, but because the issue is seemingly inconsistent it's hard to diagnose or pin down the bit where it faults.

Of course, I have tried disabling all other sound devices as well as resolved all IRQ sharing conflicts.
It just faults on some unknown function sent to flexradio.sys and the dump shows the same thing and memory location every time.
what isn't showing up are the symbols for debugging the stack.

one thing that does stand out that IDK is a red herring or not is that the instruction it kicks out on is :

Code: Select all

nt!KeBugCheckEx:
fffff805`7da16b70 48894c2408      mov     qword ptr [rsp+8],rcx ss:0018:ffff858a`5b436d80=000000000000004e

Usually, from what I have learned...the offset notation would have an 'h' after the stack frame memory location i.e. [rsp+8h] but this one instruction does not. if you step through the next instruction does have the hexadecimal notation. That being said, this is compiled code so I think the system knows what and where. I think the issue is like you said, there is some enhancement in the new OS that is glitching this out.
I'll do some more log digging...

IDK. this may be a fun problem to solve, (BTW I'm running Windows 11 Pro 23H2, but something told the debugger that it's 10)

: PFNDEbug1.png (215.91 KiB) Viewed 13438 times

w2jon · Post by **w2jon** » Thu Dec 21, 2023 8:39 pm

Yeah and since the drivers we need don't have support for DMA-Remapping you need to ensure that the Memory access protection feature in the core isolation area of Windows is disabled.. but the thing we see here is that even though it is disabled the PFN still happens from time to time.. I wonder if Defender is getting involved in some way. I'll check those logs too,

w2jon · Post by **w2jon** » Thu Dec 21, 2023 9:09 pm

With that Defender idea in my head, the one thing I'm giving a try is to run a custom exclusion list in the exploit protection features.
Long shot, but I wonder if the interrogation could be causing some sort of issue?

YouTube · Post by **ke9ns** » Fri Dec 22, 2023 1:53 pm

Someone also said to make sure control panel Power Options was set to High performance and to set "Core Parking. Meaning that we dont want the OS to shut down CPU's to save on power consumption while running PowerSDR. I believe I put that info on my webpage.
I know that some BIOS settings allow you to enable/disable other CPU power settings (more Core Parking stuff). I normally disable all the power reduction features in the BIOS. My Ryzen runs cool so I dont worry, and I dont overclock it.

w2jon · Post by **w2jon** » Fri Dec 22, 2023 1:58 pm

Okay.. even with the override exclusions we still enjoyed the pleasure of a PFN after a while.

Whats more disappointing is the test I just completed.
Lenovo P53s 8th Gen i7
Fresh build of Windows 11 Pro 22h2. No pending updates.
Installed .271 then updated to .286
TB to FW dongle enumerated properly.
Powered on 5k and Windows complained about flexradio.sys being unsupported.
Turned off core isolation and rebooted.
Started PowerSDR first startup PSDR grabbed the TRX and RX2 EEPROMs as expected
I maximized the window and hit start. No panadapter or waterfall activity nor audio on headphones. (no vac configured)
Hit stop and got a PFN
Rebooted and started PSDR as admin this time..DB init errors out as corrupt.
Started again with fresh DB, Same result on start.. no activity.. but this time I checked the Primary sample rate, and it defaulted at 9600.
Tried to change to 19200 but got PFN on applying. (of course)
Rebooted again, and started PSDR as admin again. this time I changed to 19200 first before starting DttSP.
No activity this time either but I was able to exit cleanly with a right click on the start button.
Now with 19200 as the setting I restart PSDR as admin
I can now start and stop DttSP at will without crashing. but still no activity or audio...
Kicked on compatibility mode for win7.. still DOA... hmm.

Interestingly, a fresh setup is giving me such a hard time. I need to identify the component that is interfering.

w2jon · Post by **w2jon** » Fri Dec 22, 2023 2:04 pm

ke9ns wrote: ↑
Fri Dec 22, 2023 1:53 pm
Someone also said to make sure control panel Power Options was set to High performance and to set "Core Parking. Meaning that we dont want the OS to shut down CPU's to save on power consumption while running PowerSDR. I believe I put that info on my webpage.
I know that some BIOS settings allow you to enable/disable other CPU power settings (more Core Parking stuff). I normally disable all the power reduction features in the BIOS. My Ryzen runs cool so I dont worry, and I dont overclock it.

These are good suggestions and I have all power conservation disabled on my machine as well.
Not so sure that's what we are up against here though. I think the idea of modifying the power profile came from the SmartSDR community due to network interfaces going to sleep mid-session.
I think you're correct in thinking that this is a side effect of security enrichment. Now to identify the update that is causing the fault...

w2jon · Post by **w2jon** » Fri Dec 22, 2023 2:21 pm

Hmm, I turned off the Microsoft Vulnerable driver blocklist, and then the next start I was able to get pan adapter and waterfall activity. Of course when I tried to alter any of the audio settings they faulted. I did install WINDbg so let's look at this dump file.
It's still faulting with a failure_bucket_id of FlexRadio!unknown_function .. grrr...

w2jon · Post by **w2jon** » Fri Dec 22, 2023 2:47 pm

Okay, I think I may have something.
KB5032007 released to update Nov, 14 2023
This update contains 2 remediations for .NET Framework 3.5 and 4.8.1
CVE-2023-36560 – .NET Framework Security Feature Bypass Vulnerability
CVE-2023-36049 – .NET Framework Elevation of Privilege Vulnerability

The bit that caught my eye is when I monitored the Security log when PSDR was started I saw LSA popping a special Logon event which includes the assertions: SeAssignPrimaryTokenPrivilege SeTcbPrivilege SeSecurityPrivilege SeTakeOwnershipPrivilege SeLoadDriverPrivilege SeBackupPrivilege SeRestorePrivilege SeDebugPrivilege SeAuditPrivilege SeSystemEnvironmentPrivilege SeImpersonatePrivilege SeDelegateSessionUserImpersonatePrivilege

So I took the leap and uninstalled KB5032007

After rebooting I fired up PSDR as a regular user and started banging around.
Changed bitrate and turned on and off VAC willy-nilly..
No issues with stability! no PFN at all where before I couldn't blow on the keyboard.
Ill replicate the setup on my desktop and see if stability returns.

YouTube · Post by **ke9ns** » Thu Dec 28, 2023 7:48 pm

I looked and I dont have that KB5032007 installed on my Win11 x64 Pro, but I am at ver 21H2 (not 22H2)
I will need to keep that KB in mind.

Thanks
Darrin

om1ei · Post by **om1ei** » Fri Dec 29, 2023 10:41 am

My Windows11 has KB5032007 from 15.11.2023, but I had the 'PFN LIST' problem since the beginning of 2023. Installing a newer driver for the Realtek(R) audio sound card did not help.

EDIT:
The old version of PowerSDR 2.7.2 also freezes when handling VAC.

w2jon · Post by **w2jon** » Fri Dec 29, 2023 7:20 pm

ke9ns wrote: ↑
Thu Dec 28, 2023 7:48 pm
I looked and I dont have that KB5032007 installed on my Win11 x64 Pro, but I am at ver 21H2 (not 22H2)
I will need to keep that KB in mind.

Thanks
Darrin

I'm running on 23H2 (Windows Insider build 22635.2915)
I'l do some more testing. The real issue is understanding what is happening here.
We know that Flexradio.sys is one of the files that is getting into trouble. But why is the real question?

w2jon · Post by **w2jon** » Wed Jan 10, 2024 5:25 pm

I haven't had a PFN at all since the full removal and Compatibility mode reinstall.

I was able to influence a PFN by fumbling around with the VAC settings and start/stop.
Honestly, it wasn't hard to get to fault before.

Now it functions fine no matter how hard I abuse it.
I even was able to switch the primary audio sample rate on the fly while the stream was running.
The very idea of trying to change the sample rate would fault the system before the reinstall.
It is very tolerant now. ..

In summary:
1. Export Database
2. Uninstall all PowerSDR and FlexRadio applications.
3. Reboot
4. Install KE9NS FULL Installer .289 or greater under Windows7 compatibility mode.
5. Import Saved Database
6. Upgrade to the current version .290 or greater.
7. Enjoy stability.

One interesting thing to note is that the first time I ran the new install it was set to run under compatibility for XP SP3.
I'm not sure how/why the shortcut made that choice, but the border around the application was an XP Window border. I switched the mode to Windows 7 and it blended nicely with the newer Metro-style windows frame. but runs the same under either mode.
Top

WA4SCA · Post by **WA4SCA** » Tue Jan 30, 2024 12:19 pm

Question: do you need to set compatibility mode for incremental updates as well, or just the base install?

YouTube · Post by **ke9ns** » Wed Jan 31, 2024 10:11 am

Just for the Full installer. Not even sure you can run a Win7 compatibility mode for the incremental, since its just file replacement.

When I compile PowerSDR code its on a Win11 x64 machine, but I dont work with the FireWire software or its drivers (They were developed under Win7)

Win 7 compatibility mode on the Full installer is really to help the FireWire software and drivers around all the Win10/11 security updates.

I dont really have data on whether this will really help PFN faults or not, but its worth a try, since it appears to have helped w2jon

Darrin

ND1T · Post by **ND1T** » Mon Feb 26, 2024 1:13 pm

w2jon, I want to thank you for the information you provided. It was something I had not thought of trying. I have tried 3 different motherboards, 5 different firewire cards, windows 10 and windows 11 along with numerous operating system settings, and have always had the PFN issue enough that I basically stopped using my Flex3000. I have gathered a whole bunch of windbg output as well, to see if I could tell what was causing it. Recently, I re-installed windows 7 on a Panasonic CF-52 with an embedded Ricoh firewire card, and I was using that without crashes.

Based on your post, I resurrected the original Dell Inspiron 570 I purchased when I bought the new flex3000, installed windows 10 22H2. installed PowerSDR 2.8.0.289 by explicitly setting "windows 7" compatibility mode in the properties of the installer. I have yet to experience the PFN issue. I didn't even bother checking for IRQ conflicts, as I really didn't think it would work.

I use the Virtual Audio Cables 4.70 drivers, K5FR vspMgr for serial, and a PCI-E FW800 card with a TI chipset, and Nvidia 210 graphics, along with the same general settings that are on KE9NS "general settings". I change the frame rate to 15 from 30 in the Display menu as this system is a bit too slow for 30. I also use 96000 sample size for the flex3000.

If it turns out that the PFN returns, I will repost, but so far it looks pretty good.

Jim (ND1T)

oz2pm · Post by **oz2pm** » Tue Feb 27, 2024 3:04 pm

Thanks for finding that information om1ei

It helped me a lot
I tried something a fresh setup in MSI util v3
and it got even better

I often had a dcp over 4000 and when I pressed a button in powersdr I got a PC crash

so far it's running fine but still testing

: MSI MODE.jpg (152.9 KiB) Viewed 1288 times

PowerSDR KE9NS v2.8

Firewire Win11 PFN LIST

Firewire Win11 PFN LIST

Re: Firewire Win11 PFN LIST

Re: Firewire Win11 PFN LIST

Re: Firewire Win11 PFN LIST

Re: Firewire Win11 PFN LIST

Re: Firewire Win11 PFN LIST

Re: Firewire Win11 PFN LIST

Re: Firewire Win11 PFN LIST

Re: Firewire Win11 PFN LIST

Re: Firewire Win11 PFN LIST

Re: Firewire Win11 PFN LIST

Re: Firewire Win11 PFN LIST

Re: Firewire Win11 PFN LIST

Re: Firewire Win11 PFN LIST

Re: Firewire Win11 PFN LIST

Re: Firewire Win11 PFN LIST

Re: Firewire Win11 PFN LIST

Re: Firewire Win11 PFN LIST

Re: Firewire Win11 PFN LIST

Re: Firewire Win11 PFN LIST