MP 1,1-5,1 Mac Pro 5,1 kernal panic on sleep since 10.14.4, 10.14.5, 10.14.6 updates

[CrispyDog]

Any Westmere Xeon will solve the sleep and the audio problems of Mojave, people usually go for X5680/X5690 because it's the fastest supported Xeons on a MP5,1.


Hey, your Xeons were officially supported up to El Capitan, remember? It's very nice that it worked perfectly up to High Sierra.

That's a fair point. Though it seems apparent to me that Apple very likely intentionally removed CPU support out of AppleCPUPowerManagement extension. What exactly the issue may be, it's definitely unlikely to be at the hardware level. Mostly just a gripe from me . These old Nehalems don't hold up in a modern setting anyhow.

Not that it isn't easily addressable with new CPUs (and if you really wanted, 1333mhz memory, which i had already done because i had it lying around, new CPUs would unlock it to 1333 vs 1066 current)

 

[tsialex]

That's a fair point. Though it seems apparent to me that Apple very likely intentionally removed CPU support out of AppleCPUPowerManagement extension. What exactly the issue may be, it's definitely unlikely to be at the hardware level. Mostly just a gripe from me . These old Nehalems don't hold up in a modern setting anyhow.

Not that it isn't easily addressable with new CPUs (and if you really wanted, 1333mhz memory, which i had already done because i had it lying around, new CPUs would unlock it to 1333 vs 1066 current)

The problem it's not with power management, but with multiple CPU support.

A single CPU Nehalem Mac Pro have sleep and audio working perfectly with Mojave, but not a dual CPU one - some single CPU mid-2010 Mac Pros have Nehalem Xeons from factory, but all mid-2010 and mid-2012 dual CPU Mac Pros are Westmere.

Like I said before, dual CPU Nehalem Xeons are not a officially supported configuration since ElCapitan while dual CPU Westmere Xeons are still supported officially up to Mojave. Catalina have some problems with the now unsupported multiple socket CPU configs (AppleIntelMCEReporter.kext needs to be disabled for a dual CPU Mac Pro/hackintosh with dual socket to work with Catalina) and BigSur removed support for Westmere, or wording better, BigSur now require CPU instructions that are not present on Westmere Xeons (more specifically, XSAVE instructions are now a requirement for AppleHypervisor support from beta6 on).

 

[149TM]

Having same issue here on a Mac Pro 5,1 running Mac OS Majove, 10.14.16


UUID: BA925359-3715-4C66-8145-88B36F369528

Stackshot Reason: Sleep transition timed out after 180 seconds while entering darkwake on way to sleep. Suspected bundle: com.apple.driver.usb.AppleUSBXHCI. Thread 0x21ac9.
Failure code:: 0x5f838579 00000031

Using an external monitor on HDMI, nothing connected to USB ports, 4 internal drives only.

I'm also getting a lot of freeze on Safari and Google Chrome.

Any suggestions on what to do to fix this problem.

Kind regards

 

[InfinityVortex]

It seems my cMP 5,1 is having a similar problem.... I am not sure how to read the panic reports. I just did the open core upgrade last week with the open core solution from "Create Pro"....It seems to crash when I have left it on for a while like 6-8 hours. I have tried putting it to sleep and waking it with no crashing issues. This is not a 2009 upgraded cMP it is an actual mid 2010. I mainly use this machine as my daily driver so to speak and do some audio and video editing......I also did the Titan Ridge thunderbolt flashing and such last week when I was doing the open core....My boot drive is on a M.2 NvMe in PCIe slot 2.


Thank you for your time and consideration,

EM

 

[Sean__Wong]

OK
so now I'm here too, bummer. I was so happy with my cMP5,1. BTW for noobs to Northbridge mods I've screwed my heat sink down and built a venturi funnel to cool it, its not nice but it is ghetto tech.

Rightio' back to the problem. now this is not my professional opinion but becasue I've be in COVID lockdown I've been re-watching that HOUSE TV series. yep...the thoughts follow.

Symptoms
seemingly random Kernal Panics, and reboots.
Sometimes reboots don't ... reboot.
pulling your cats' hair out of the components doesn't help.
neither does pulling disks / GPU / PCI cards.

Conjecture
Could there be something common to all the machines from page 1>7 above that causes a KP that is not 'swap-able' to allow a differential diagnosis?
Inverse Corollary, What would casue a KP on every machine but always be different?

Hypothesis
Suspect = RAM
When sleep is engaged the machine flushes its memory to disk. While this is happening ( as we are all addicted to big RAM configs ) the PMLock faults. A KP follows.

Observations

DIMM 7 Riser is Hot, it been that way for a long time.
I don't worry about it but I do swap my DIMM's around so that heat stress doesn't accumulate, until I get bigger Dimms, 2>4>8 etc. I'm currently running 8x4Gb Dimms and waiting for some reasonably priced 8Gb ones to arrive.

Test
When the 8Gb DIMMs arrive I'll load half of the capacity to test, I get to keep 32Gb of RAM config'd and we'll see IF a KP results.

Invitation
@tsialex etal please shred my thoughts and I'll have another go


Here's my KP log from just a few minutes ago.

Mon Apr 26 21:34:50 2021

*** Panic Report ***
panic(cpu 0 caller 0xffffff7f8ea03fe5): pmLock: waited too long, held by 9@0xffffff800b2dcb69
Backtrace (CPU 0), Frame : Return Address
0xffffffa3c0a0bbb0 : 0xffffff800b1ad5cd
0xffffffa3c0a0bc00 : 0xffffff800b2e9215
0xffffffa3c0a0bc40 : 0xffffff800b2da94a
0xffffffa3c0a0bcb0 : 0xffffff800b15a9d0
0xffffffa3c0a0bcd0 : 0xffffff800b1acfe7
0xffffffa3c0a0bdf0 : 0xffffff800b1ace33
0xffffffa3c0a0be60 : 0xffffff7f8ea03fe5
0xffffffa3c0a0bf10 : 0xffffff800b2dc97c
0xffffffa3c0a0bf30 : 0xffffff800b1cf207
0xffffffa3c0a0bf80 : 0xffffff800b1cf3f0
0xffffffa3c0a0bfa0 : 0xffffff800b15a0ce
Kernel Extensions in backtrace:
com.apple.driver.AppleIntelCPUPowerManagement(220.0)[B9A8B230-8403-37A2-BEEB-CE015F6BDAA2]@0xffffff7f8ea03000->0xffffff7f8ea2bfff

BSD process name corresponding to current thread: kernel_task

Mac OS version:
18G8022

Kernel version:
Darwin Kernel Version 18.7.0: Tue Jan 12 22:04:47 PST 2021; root:xnu-4903.278.56~1/RELEASE_X86_64
Kernel UUID: 33FBDE00-1755-34DE-AF5B-2CAD0DE8F412
Kernel slide: 0x000000000ae00000
Kernel text base: 0xffffff800b000000
__HIB text base: 0xffffff800af00000
System model name: MacPro5,1 (Mac-F221BEC8)

 

[Sean__Wong]

I'm having a clicking noise and a fan starting up and stopping over and over again during sleep mode, anyone run into this issue?

try checking your GPU fans, Ihave a Sapphire 580 and it's fans spin up at odd times and when they do they rise-up ever so slightly, certain temp + Atmospheric conditions ( low Pressure ) amazingly seems to cause the fans to clip on the outside edges but only during spin up. Its maddeningly annoying, so I stuck thin bit (3mm of 3M 25mm velcro ) on each side of the GPU to provide a little extra space. No more blade clicking!

 

[tsialex]

OK
so now I'm here too, bummer. I was so happy with my cMP5,1. BTW for noobs to Northbridge mods I've screwed my heat sink down and built a venturi funnel to cool it, its not nice but it is ghetto tech.

Rightio' back to the problem. now this is not my professional opinion but becasue I've be in COVID lockdown I've been re-watching that HOUSE TV series. yep...the thoughts follow.

Symptoms
seemingly random Kernal Panics, and reboots.
Sometimes reboots don't ... reboot.
pulling your cats' hair out of the components doesn't help.
neither does pulling disks / GPU / PCI cards.

Conjecture
Could there be something common to all the machines from page 1>7 above that causes a KP that is not 'swap-able' to allow a differential diagnosis?
Inverse Corollary, What would casue a KP on every machine but always be different?

Hypothesis
Suspect = RAM
When sleep is engaged the machine flushes its memory to disk. While this is happening ( as we are all addicted to big RAM configs ) the PMLock faults. A KP follows.

Observations

DIMM 7 Riser is Hot, it been that way for a long time.
I don't worry about it but I do swap my DIMM's around so that heat stress doesn't accumulate, until I get bigger Dimms, 2>4>8 etc. I'm currently running 8x4Gb Dimms and waiting for some reasonably priced 8Gb ones to arrive.

Test
When the 8Gb DIMMs arrive I'll load half of the capacity to test, I get to keep 32Gb of RAM config'd and we'll see IF a KP results.

Invitation
@tsialex etal please shred my thoughts and I'll have another go


Here's my KP log from just a few minutes ago.

Mon Apr 26 21:34:50 2021

*** Panic Report ***
panic(cpu 0 caller 0xffffff7f8ea03fe5): pmLock: waited too long, held by 9@0xffffff800b2dcb69
Backtrace (CPU 0), Frame : Return Address
0xffffffa3c0a0bbb0 : 0xffffff800b1ad5cd
0xffffffa3c0a0bc00 : 0xffffff800b2e9215
0xffffffa3c0a0bc40 : 0xffffff800b2da94a
0xffffffa3c0a0bcb0 : 0xffffff800b15a9d0
0xffffffa3c0a0bcd0 : 0xffffff800b1acfe7
0xffffffa3c0a0bdf0 : 0xffffff800b1ace33
0xffffffa3c0a0be60 : 0xffffff7f8ea03fe5
0xffffffa3c0a0bf10 : 0xffffff800b2dc97c
0xffffffa3c0a0bf30 : 0xffffff800b1cf207
0xffffffa3c0a0bf80 : 0xffffff800b1cf3f0
0xffffffa3c0a0bfa0 : 0xffffff800b15a0ce
Kernel Extensions in backtrace:
com.apple.driver.AppleIntelCPUPowerManagement(220.0)[B9A8B230-8403-37A2-BEEB-CE015F6BDAA2]@0xffffff7f8ea03000->0xffffff7f8ea2bfff

BSD process name corresponding to current thread: kernel_task

Mac OS version:
18G8022

Kernel version:
Darwin Kernel Version 18.7.0: Tue Jan 12 22:04:47 PST 2021; root:xnu-4903.278.56~1/RELEASE_X86_64
Kernel UUID: 33FBDE00-1755-34DE-AF5B-2CAD0DE8F412
Kernel slide: 0x000000000ae00000
Kernel text base: 0xffffff800b000000
__HIB text base: 0xffffff800af00000
System model name: MacPro5,1 (Mac-F221BEC8)

You wrote a lot and said nothing. Please don't clip crash reports, always post it in full.

 

[tsialex]

One thing that people frequently overlook and one the most common cause of pmlocks after sleep is the timesinceboot counter of the RTC being stuck because a low volt RTC battery.

It's the first thing to check, if the BR2032 battery is anything below 3.0V, you need to replace it.

 

[bsbeamer]

Especially if you've NEVER replaced the battery, it's a cheap and easy fix to start troubleshooting. These machines are 10+ years old for many people. Battery does not last that long even in the best situations.

 

[tsialex]

Especially if you've NEVER replaced the battery, it's a cheap and easy fix to start troubleshooting. These machines are 10+ years old for many people. Battery does not last that long even in the best situations.

Other thing, even if it was replaced recently, you need to check if it was replaced with a real BR2032. While a CR2032 also works, the useful life is severely diminished when used as a RTC battery of a Mac Pro. The CR2032 chemistry is incompatible with the internal temperatures near the GPU.

 

[Sean__Wong]

You wrote a lot and said nothing. Please don't clip crash reports, always post it in full.

 

[tsialex]

@Sean__Wong Your Mac Pro have a Westmere Xeon, so the most common cause of the thread is not what is causing your crashes. You have several 3rd party kexts installed, this complicates a lot. Check the battery and do a clean install on another disk to see if you still have the same problem.

 

[bbb555]

Other thing, even if it was replaced recently, you need to check if it was replaced with a real BR2032. While a CR2032 also works, the useful life is severely diminished when used as a RTC battery of a Mac Pro. The CR2032 chemistry is incompatible with the internal temperatures near the GPU.

thanks @tsialex for this advice
my Mac Pro would panic so I permanently disabled sleep for a while now.
I replaced the RTC battery (only had a CR2032 handy for testing) and it no longer panics after sleep. Such a simple fix.
I've ordered a replacement BR2032 for a permanent fix.
thanks again.

 

[Sean__Wong]

OK
so now I'm here too, bummer. I was so happy with my cMP5,1. BTW for noobs to Northbridge mods I've screwed my heat sink down and built a venturi funnel to cool it, its not nice but it is ghetto tech.

Rightio' back to the problem. now this is not my professional opinion but becasue I've be in COVID lockdown I've been re-watching that HOUSE TV series. yep...the thoughts follow.

Symptoms
seemingly random Kernal Panics, and reboots.
Sometimes reboots don't ... reboot.
pulling your cats' hair out of the components doesn't help.
neither does pulling disks / GPU / PCI cards.

Conjecture
Could there be something common to all the machines from page 1>7 above that causes a KP that is not 'swap-able' to allow a differential diagnosis?
Inverse Corollary, What would casue a KP on every machine but always be different?

Hypothesis
Suspect = RAM
When sleep is engaged the machine flushes its memory to disk. While this is happening ( as we are all addicted to big RAM configs ) the PMLock faults. A KP follows.

Observations

DIMM 7 Riser is Hot, it been that way for a long time.
I don't worry about it but I do swap my DIMM's around so that heat stress doesn't accumulate, until I get bigger Dimms, 2>4>8 etc. I'm currently running 8x4Gb Dimms and waiting for some reasonably priced 8Gb ones to arrive.

Test
When the 8Gb DIMMs arrive I'll load half of the capacity to test, I get to keep 32Gb of RAM config'd and we'll see IF a KP results.

Invitation
@tsialex etal please shred my thoughts and I'll have another go


Here's my KP log from just a few minutes ago.

Mon Apr 26 21:34:50 2021

*** Panic Report ***
panic(cpu 0 caller 0xffffff7f8ea03fe5): pmLock: waited too long, held by 9@0xffffff800b2dcb69
Backtrace (CPU 0), Frame : Return Address
0xffffffa3c0a0bbb0 : 0xffffff800b1ad5cd
0xffffffa3c0a0bc00 : 0xffffff800b2e9215
0xffffffa3c0a0bc40 : 0xffffff800b2da94a
0xffffffa3c0a0bcb0 : 0xffffff800b15a9d0
0xffffffa3c0a0bcd0 : 0xffffff800b1acfe7
0xffffffa3c0a0bdf0 : 0xffffff800b1ace33
0xffffffa3c0a0be60 : 0xffffff7f8ea03fe5
0xffffffa3c0a0bf10 : 0xffffff800b2dc97c
0xffffffa3c0a0bf30 : 0xffffff800b1cf207
0xffffffa3c0a0bf80 : 0xffffff800b1cf3f0
0xffffffa3c0a0bfa0 : 0xffffff800b15a0ce
Kernel Extensions in backtrace:
com.apple.driver.AppleIntelCPUPowerManagement(220.0)[B9A8B230-8403-37A2-BEEB-CE015F6BDAA2]@0xffffff7f8ea03000->0xffffff7f8ea2bfff

BSD process name corresponding to current thread: kernel_task

Mac OS version:
18G8022

Kernel version:
Darwin Kernel Version 18.7.0: Tue Jan 12 22:04:47 PST 2021; root:xnu-4903.278.56~1/RELEASE_X86_64
Kernel UUID: 33FBDE00-1755-34DE-AF5B-2CAD0DE8F412
Kernel slide: 0x000000000ae00000
Kernel text base: 0xffffff800b000000
__HIB text base: 0xffffff800af00000
System model name: MacPro5,1 (Mac-F221BEC8)

OK
This worked for me

I diagnosed My RAM
I found that Slots 5-6-7-8 needed to be empty, Slot 7 reported higher temps regardless of the load.
Didnot seem to matter which DIMMS from my set of 8 x 2 ECC Buffered HYNIX, swapping them around and rebooting was a dead end.

I changed My Battery
I bough a new 4 pack of EverReady CR2032,

Nothing seemed to be directly related to the KP

I changed My Power Supply,
This mac is active 24x7 for the last 8 years so seemed fair with 4 x HDD and a full house of gear that any degradation in capacitors may have impacted. boots and sleep wake

I upgraded to 8 x 8 GB DIMMS ECC Buffered SAMSUNG.

Outcome
Now I have a stable CMP 5,1 again. and 2 x the RAM.

Just like cleaning you car it seems to run better.

My RAM Temps are Lower despite having 2 x the capacity.

 

[M-L-R]

The problem it's not with power management, but with multiple CPU support.

A single CPU Nehalem Mac Pro have sleep and audio working perfectly with Mojave, but not a dual CPU one - some single CPU mid-2010 Mac Pros have Nehalem Xeons from factory, but all mid-2010 and mid-2012 dual CPU Mac Pros are Westmere.

Like I said before, dual CPU Nehalem Xeons are not a officially supported configuration since ElCapitan while dual CPU Westmere Xeons are still supported officially up to Mojave. Catalina have some problems with the now unsupported multiple socket CPU configs (AppleIntelMCEReporter.kext needs to be disabled for a dual CPU Mac Pro/hackintosh with dual socket to work with Catalina) and BigSur removed support for Westmere, or wording better, BigSur now require CPU instructions that are not present on Westmere Xeons (more specifically, XSAVE instructions are now a requirement for AppleHypervisor support from beta6 on).

The problem it's not with power management, but with multiple CPU support.

A single CPU Nehalem Mac Pro have sleep and audio working perfectly with Mojave, but not a dual CPU one - some single CPU mid-2010 Mac Pros have Nehalem Xeons from factory, but all mid-2010 and mid-2012 dual CPU Mac Pros are Westmere.

Like I said before, dual CPU Nehalem Xeons are not a officially supported configuration since ElCapitan while dual CPU Westmere Xeons are still supported officially up to Mojave. Catalina have some problems with the now unsupported multiple socket CPU configs (AppleIntelMCEReporter.kext needs to be disabled for a dual CPU Mac Pro/hackintosh with dual socket to work with Catalina) and BigSur removed support for Westmere, or wording better, BigSur now require CPU instructions that are not present on Westmere Xeons (more specifically, XSAVE instructions are now a requirement for AppleHypervisor support from beta6 on).

The problem it's not with power management, but with multiple CPU support.

A single CPU Nehalem Mac Pro have sleep and audio working perfectly with Mojave, but not a dual CPU one - some single CPU mid-2010 Mac Pros have Nehalem Xeons from factory, but all mid-2010 and mid-2012 dual CPU Mac Pros are Westmere.

Like I said before, dual CPU Nehalem Xeons are not a officially supported configuration since ElCapitan while dual CPU Westmere Xeons are still supported officially up to Mojave. Catalina have some problems with the now unsupported multiple socket CPU configs (AppleIntelMCEReporter.kext needs to be disabled for a dual CPU Mac Pro/hackintosh with dual socket to work with Catalina) and BigSur removed support for Westmere, or wording better, BigSur now require CPU instructions that are not present on Westmere Xeons (more specifically, XSAVE instructions are now a requirement for AppleHypervisor support from beta6 on).

Hi tsialex, You stated; "AppleIntelMCEReporter.kext needs to be disabled...". I have a question (if it's ok). When I boot into MacOS 10.9.x my audio is good. When I boot into MacOS 10.14.x (Mojave) > MacOS 11.x (Big Sur) my audio flutters. Some apps are worse than others. Garageband constantly throws this error (see my attached image). I have a Hac Pro 4,1 (early 2009) flashed to 5,1 .... 2 x 2.93 GHz Quad-Core Intel Xeon processor, 16GB Memory, Radeon RX 580 8GB GPU (flashed for Mac), 1TB 3Gb/s SSD hard-drives in my SATA bays, Boot ROM Version: 144.0.0.0.0. Do you think your suggestion will help my audio issue? If yes; please provide instructions. I have limited knowledge. I Googled a lot, with no success. Do you have other (audio fix) suggestions? Keeping in mind my limited procedural knowledge. This is my first post. Sorry if I did something improper. Thank-you very much

 

[tsialex]

Hi tsialex, You stated; "AppleIntelMCEReporter.kext needs to be disabled...". I have a question (if it's ok). When I boot into MacOS 10.9.x my audio is good. When I boot into MacOS 10.14.x (Mojave) > MacOS 11.x (Big Sur) my audio flutters. Some apps are worse than others. Garageband constantly throws this error (see my attached image). I have a Hac Pro 4,1 (early 2009) flashed to 5,1 .... 2 x 2.93 GHz Quad-Core Intel Xeon processor, 16GB Memory, Radeon RX 580 8GB GPU (flashed for Mac), 1TB 3Gb/s SSD hard-drives in my SATA bays, Boot ROM Version: 144.0.0.0.0. Do you think your suggestion will help my audio issue? If yes; please provide instructions. I have limited knowledge. I Googled a lot, with no success. Do you have other (audio fix) suggestions? Keeping in mind my limited procedural knowledge. This is my first post. Sorry if I did something improper. Thank-you very much

No. Nehalem Xeons are not supported anymore, you need to upgrade to Westmere Xeons. There are a full thread about this issue:

Strange Audio Issue on MP 4,1>5,1 Mojave 10.14.4

Ok, so this is my first post here since Ive been working with these machines for quite some time and Im fairly knowledgeable with these kinds of things, but this time Ive hit a brick wall so any help would be great. Recently got a Dual Quad 2.26 MP 4.1 with the intentions of maxing it on the...

forums.macrumors.com

Dual CPU Nehalem Xeons are officially supported up to El Capitan, deprecated with Sierra, audio problems start with 10.14.3.

 

[M-L-R]

Hi tsialex, I will look into Westmere Xeons. I was hoping to find a "no cost" fix. It would be a drag to find "it's not the issue (the Westmere fix)" I assume my Mac Pro requires two Westmeres. It's strange the audio in "QuickTime X" & my browsers is good. Most other app audio sputters (VLC, iTunes, ... in Mojave+). If I install the Westmeres, I will let you know my result. I saw your referenced post about 6 months ago. I don't have the "sleep" issue mentioned (in a different post). Thank-you for your help. M-L-R

 

[tsialex]

Hi tsialex, I will look into Westmere Xeons. I was hoping to find a "no cost" fix. It would be a drag to find "it's not the issue (the Westmere fix)" I assume my Mac Pro requires two Westmeres. It's strange the audio in "QuickTime X" & my browsers is good. Most other app audio sputters (VLC, iTunes, ... in Mojave+). If I install the Westmeres, I will let you know my result. I saw your referenced post about 6 months ago. I don't have the "sleep" issue mentioned (in a different post). Thank-you for your help. M-L-R

There isn't.

It's over two years now and no workaround was found, not even a way to minimize it. This issue is only solvable by upgrading early-2009 dual CPU Mac Pros from Nehalem Xeons to Westmeres. Don't forget that your CPU tray requires de-lided Xeons.

 

[M-L-R]

Hi tsialex, I (just now) purchased two Westmere Xeons ($99.00 each refurbished) > 3.46GHz. If this forum allows, I can post a link to the product. I talked over the phone to their tech assistant. I mentioned: Mac Pro 4,1 (early 2009) flashed to 5,1 ... Westmere (only) Xeons > delidded. Just in case, I watched a YouTube (generic) delidding video (below). I also watched installation videos. I will let you know the results (good &/or bad). Thank-you (again) for your help. M-L-R

 

[tsialex]

Hi tsialex, I (just now) purchased two Westmere Xeons ($99.00 each refurbished) > 3.46GHz. If this forum allows, I can post a link to the product. I talked over the phone to their tech assistant. I mentioned: Mac Pro 4,1 (early 2009) flashed to 5,1 ... Westmere (only) Xeons > delidded. Just in case, I watched a YouTube (generic) delidding video (below). I also watched installation videos. I will let you know the results (good &/or bad). Thank-you (again) for your help. M-L-R

This is now completely off-topic, but if you didn't bought it already de-lided, buy 3 or 4 $5 X55xx Xeons to destroy and learn with them.

Yes, you will destroy some de-liding until you learn how to do it correctly.

 

[M-L-R]

Hi tsialex, Thnx. It sounds like a fun hobby I'll reply ASAP. I "back-liked" you previous replies. Very much appreciated. M-L-R

 

[M-L-R]

Hi tsialex, I updated my two Nehalems to Westmeres. My audio now works properly. For good measures, I upgraded to 64GB 1333MHz RAM & SSD (previously mentioned). I use MacOS 11.5.2. I read another of your replies > "Mac Pro 2009 5,1 & Big Sur has many issues. Don't waste your time". I have many other (Big Sur) issues. For now, I will see if I can eventually find fixes or avoid trigging the issue. I use Parallels 17 > MacOS 10.9 (Mavericks) virtual machine. My GPU > RX 580 8GB (metal) is ok with that. I can boot up into (a different installation of) Mavericks to use my Firewire (when required). But my (incompatible) RX 580 only works well enough for simple tasks.
- - - - - - - - - -
Does Catalina work well with Mac Pro 2009 5,1 > worth back-dating from Big Sur?
Thank-you for your help. Fixing the audio was my main concern. M-L-R

 

[tsialex]

Does Catalina work well with Mac Pro 2009 5,1 > worth back-dating from Big Sur?

Yes, Catalina works fine - learn to use OpenCore:

OpenCore on the Mac Pro

This guide explains how to use the excellent OpenCore boot loader to install, run, and update the latest version of macOS on the MacPro5,1, resulting in a clean, unaltered operating system just like on a supported Mac.

forums.macrumors.com

Btw, you are beyond crazy to try to run Big Sur and all it's KPs with an early-2009 Mac Pro without a clean BootROM or a fall back plan like a MATT card.

 

[M-L-R]

Hi tsialex, Previously, I used an Dortania's OpenCore process to upgrade to Big Sur.
I read your linked to "OpenCore on the Mac Pro" article.
I went passed MacOS 11.2.x because the YouTube video I followed stated it's ok with SSD. Apparently not correct.
I'm guessing KPs means "known problems" (?).
Before upgrading to Big Sur, my Boot ROM Version showed: 144.0.0.0.0.
Please point to an article (or video) explaining a "clean BootROM".
I'll consider doing a clean install (downgrade) to Catalina.
I read a reply recommending your "recosntruction services". What is your price or inquiry process?
If I used your services, I'm guessing Big Sur is still not recommended??? I'd like to keep Big Sur.
Sorry if I used incorrect terms & descriptions (nomenclature). Thank-you very much.

 

[tsialex]

Hi tsialex, Previously, I used an Dortania's OpenCore process to upgrade to Big Sur.
I read your linked to "OpenCore on the Mac Pro" article.
I'm guessing KPs means "known problems" (?).

KPs = Kernel Panics

Before upgrading to Big Sur, my Boot ROM Version showed: 144.0.0.0.0.
Please point to an article (or video) explaining a "clean BootROM".

You have a 12 years+ old Mac. The moment you/original owner took it out of the box back in 2009, the NVRAM started to be populated with all sort of settings, some public, some private, some user accessible, some not user accessible.

Now you have 12+ years of cruft inside the NVRAM and to make matters even worse, a BootROM cross-flashed from MP4,1 to MP5,1 that is extremely prone to bricking since a MP4,1>5,1 is not one thing or the other, there are MP4,1 and MP5,1 components inside the BootROM image.

Before you ask, reseting the NVRAM just clear the public/user accessible settings, nothing else. The cruft accumulated over the years is never cleared with the NVRAM resets.

A BootROM reconstruction is the process I developed over the years to re-create a never booted BootROM image for your Mac Pro exactly like the OEM flashed back in the production line, but updated to the most recent release of the BootROM image components and EFI.

I'll consider doing a clean install (downgrade) to Catalina.
I read a reply recommending your "recosntruction services". What is your price or inquiry process?
If I used your services, I'm guessing Big Sur is still not recommended??? I'd like to keep Big Sur.
Sorry if I used incorrect terms & descriptions (nomenclature). Thank-you very much.

Things are not linked. Even a real mid-2012 with a pristine reconstructed BootROM still have KPs with any version of Big Sur newer than 11.2.3. You can just mitigate it using latebloom.

Latebloom - An experimental workaround for the 11.3+ race condition

UPDATE 27sep21: latebloom is now deprecated/unnecessary. Please see this thread for more information. I have created a kext that may provide a workaround for the Big Sur 11.3+ race condition. Alpha testing with a small group has provided encouraging results, so I'm now posting it publicly...

forums.macrumors.com