MP 1,1-5,1 Manually Configured OpenCore on the Mac Pro

[h9826790]

Nothing here. You?
[automerge]1583361790[/automerge]
Diagnostics show:
Verifying Mac board-id... NOT OK. Board id length is not compatible. Expected 12 or 20 characters. Board id:Mac-7BA5B2D9E42DDD94
board-id

Is there a way to make hardware acceleration work without

<key>SMBIOS</key>
<dict>
<key>BoardProduct</key>
<string>Mac-7BA5B2D9E42DDD94</string>
</dict>

? This injection seems to be causing issues.

Please try cdf's config.plist in post #1 -> activate Handoff -> revert to my HWAccel enabled config.plist
[automerge]1583403636[/automerge]

On another note. I have my Rom Dump. I don’t know how to read a bin file. If this is the case I would like to see if windows EFI solves this, and also looking for a little more clarity

OpenCore on the Mac Pro

For those who don't want to install Brew / Binwalk, or not that familiar with terminal. You may also check the ROM image as simple as by using TextEdit (macOS build in apps). For corrupted NVRAM, if you search "Secure" (case sensitive) in TextEdit, you will see the actual certificate. e.g. This...

forums.macrumors.com

 

[Itconnects]

Please try cdf's config.plist in post #1 -> activate Handoff -> revert to my HWAccel enabled config.plist
[automerge]1583403636[/automerge]

OpenCore on the Mac Pro

For those who don't want to install Brew / Binwalk, or not that familiar with terminal. You may also check the ROM image as simple as by using TextEdit (macOS build in apps). For corrupted NVRAM, if you search "Secure" (case sensitive) in TextEdit, you will see the actual certificate. e.g. This...

forums.macrumors.com

Thank you. No certificates. Should I do the windows EFI install. Or are we not fully there yet as far as bugs..

 

[JeDiGM]

Please try cdf's config.plist in post #1 -> activate Handoff -> revert to my HWAccel enabled config.plist
[automerge]1583403636[/automerge]

Tried it last night. No joy. Do you have it working h9826790? How did you get it working?

 

[h9826790]

Thank you. No certificates. Should I do the windows EFI install. Or are we not fully there yet as far as bugs..

Install / boot EFI Windows on cMP is fine now with the OpenCore ProtectSecureBoot function
[automerge]1583418086[/automerge]

Tried it last night. No joy. Do you have it working h9826790? How did you get it working?

I have that working right now. But there are too many variables involved in my setup. e.g.

1) Not clean installation
2) too many different patches / mod since beta stage
3) I mixed used CAT and terminal commands may be a few times

So, I don't know the exact steps to make it work. What I am sure is that inject the board-id via OpenCore to activate HWAccel will not disable Handoff. At least this is true on my setup that has Handoff works before I install OpenCore.

 

[JeDiGM]

I have that working right now. But there are too many variables involved in my setup. e.g.

1) Not clean installation
2) too many different patches / mod since beta stage
3) I mixed used CAT and terminal commands may be a few times

So, I don't know the exact steps to make it work. What I am sure is that inject the board-id via OpenCore to activate HWAccel will not disable Handoff. At least this is true on my setup that has Handoff works before I install OpenCore.

Understood. Which terminal commands did you use? Do you remember?

Also, is your sidecar working? I was figuring the reason it's not working on mine is because of this as well? I can connect to an iPad but I can't see anything on the screen.

 

[h9826790]

Understood. Which terminal commands did you use? Do you remember?

Also, is your sidecar working? I was figuring the reason it's not working on mine is because of this as well? I can connect to an iPad but I can't see anything on the screen.

I don't have iPad to run Sidecar test

 

[roobarb!]

Also, is your sidecar working?

Sidecar doesn't work on my MacPro5,1. To the best of my knowledge it relies upon Intel Quick Sync Video, (edit, it doesn't, see below) despite there being plenty of horsepower and hardware acceleration with the Radeon VII. Handoff, continuity and unlock with Apple Watch all work fine for me without other tinkering once I have BoardProduct set to Mac-7BA5B2D9E42DDD94, which sidesteps the blacklist check.

As an aside, Sidecar also doesn't work properly with my MacBookPro12,1, which does have Quick Sync Video, but mustn't have the power to cope. On the MacPro5,1 I get a connection timeout and nothing else. On the MacBookPro12,1 I get an image which is laggy and distorted.

 

[h9826790]

Sidecar doesn't work on my MacPro5,1. To the best of my knowledge it relies upon Intel Quick Sync Video, despite there being plenty of horsepower and hardware acceleration with the Radeon VII. Handoff, continuity and unlock with Apple Watch all work fine for me without other tinkering once I have BoardProduct set to Mac-7BA5B2D9E42DDD94.

Quick Sync can't be the requirement, otherwise both iMac Pro and Mac Pro 7,1 can't do Sidecar, which obviously wrong. The question is just how to activate it, and what the real requirements are.

 

[roobarb!]

Quick Sync can't be the requirement, otherwise both iMac Pro and Mac Pro 7,1 can't do Sidecar, which obviously wrong. The question is just how to activate it, and what the real requirements are.

Ooh, interesting. I hadn't realised Quick Sync was absent on those; I shall report back to another forum their erroneous information!

 

[lxmeta]

Maybe stating the very obvious:
This guide also works perfectly fine with an existing Mojave installation, making it "Catalina ready". Did this now on different machines. Just follow the guide to copy and config the stuff, reboot, and you can update to Catalina. No need to install from a fresh disk etc...

 

[MIKX]

Which of our smart OpenCore buffs is gonna make an 'OpenCore Installer' . . .? ?

Joking of course, coz learning how to enriches & getting it done by oneself is 1/2 the fun.

 

[Cass67]

Ooh, interesting. I hadn't realised Quick Sync was absent on those; I shall report back to another forum their erroneous information!

T2 chip is what's key on iMacPro and 7,1. No igpu no party on other Mac's...

 

[eksu]

I can't even get sidecar to work on a 2017 MBP and an iPad Pro, nothing but connection timeouts. Connection timeouts might not mean that there is missing hardware, it could just be buggy.

 

[JM-Prod]

I followed the instructions, but my Mac seems to boot without OpenCore on disk b. Not able to run the "nvram 4D1FDA02-38C7-4A6A-9CC6-4BCCA8B30102pencore-version" command.

Don't see VMM after writing and running "sysctl machdep.cpu.features".

And not able to install Catalina.


Any idea what could be wrong?

 

[startergo]

Not able to run the "nvram 4D1FDA02-38C7-4A6A-9CC6-4BCCA8B30102:Opencore-version" command

Make sure:

<key>ExposeSensitiveData</key>
            <integer>15</integer>

In the security key. The sum of all is F in Hex=15 in Decimal.

Description: Sensitive data exposure bitmask (sum) to operating system.
• 0x01 — Expose printable booter path as an UEFI variable.
• 0x02 — Expose OpenCore version as an UEFI variable.
• 0x04 — Expose OpenCore version in boot picker menu title.
• 0x08 — Expose OEM information as a set of UEFI variables.

nvram 4D1FDA02-38C7-4A6A-9CC6-4BCCA8B30102:opencore-version
4D1FDA02-38C7-4A6A-9CC6-4BCCA8B30102:opencore-version    REL-057-2020-03-06
 nvram 4D1FDA02-38C7-4A6A-9CC6-4BCCA8B30102:oem-product
4D1FDA02-38C7-4A6A-9CC6-4BCCA8B30102:oem-product    MacPro5,1
nvram 4D1FDA02-38C7-4A6A-9CC6-4BCCA8B30102:oem-vendor
4D1FDA02-38C7-4A6A-9CC6-4BCCA8B30102:oem-vendor    Apple Inc.
nvram 4D1FDA02-38C7-4A6A-9CC6-4BCCA8B30102:oem-board
4D1FDA02-38C7-4A6A-9CC6-4BCCA8B30102:oem-board    Mac-F221BEC8

 

[Ludacrisvp]

Anyone else have the mac panic when attempting to bless?
This is on the xserve3,1 on 0.5.5 debug.
Was attempting to update a fresh 10.14.0 install to 10.14.6 but it seems it is always panic if requestbootvarrouting is enabled.
Same issue if i was just trying to change between already installed OS, I had to choose at the boot picker, if i tried in the OS that failed.
kernel trap page fault panic, mentioning the EFIRuntime and EFINVRAM kexts in backtrace

 

[jameslmoser]

Post time is longer if you have multiple MemoryConfig entries inside the NVRAM, but even with a totally clean NVRAM, POST/boot time with a MP5,1 is very long when compared with recent Macs/PCs with UEFI fast boot.

Talking about clean NVRAM, is it true Alex that if you get a Secure Boot Cert put in your NVRAM there is no way to remove it? or other things (memory configs?) you may no longer want.

On a post in a different thread you were talking about updating the boot loader of the 4,1>5,1 machines using the netkas method, and you have a method of fixing the boot loaders so you can use the firmware restore CDs. Does using one of those CDs (assuming your boot loader is fixed if you have a 4,1>5,1) reset/clean your NVRAM? If thats true I might look into your method, or seek your services for three machines I have that are 4,1>5,1 machines. =)
[automerge]1583558181[/automerge]
Also, has anyone had the Cert problem with the newer versions of the classic mac pro firmware?

 

[lxmeta]

I followed the instructions, but my Mac seems to boot without OpenCore on disk b. Not able to run the "nvram 4D1FDA02-38C7-4A6A-9CC6-4BCCA8B30102pencore-version" command.

Don't see VMM after writing and running "sysctl machdep.cpu.features".

And not able to install Catalina.


Any idea what could be wrong?

Are you sure the directory structure is exactly as described? Can you share your directory tree? Did you try the bless command with specifying the boot file?
The only case where I had a machine booting error free without enabling opencore was when I had copied the EFI folder into the wrong folder (one too many EFI) by accident...

 

[Dayo]

Anyone else have the mac panic when attempting to bless?
This is on the xserve3,1 on 0.5.5 debug.

I believe 0.5.6 dealt with quite a few issues. Have you tried this?

 

[Dayo]

@Ludacrisvp,

I have found an issue with "RequestBootVarFallback" on my cMP 3,1 (RX 580 Card) with OC 0.5.6. I don't get an issue when this is enabled as such until I try to change the startup disk ... at which point it a reboot will be triggered.

This is with "RequestBootVarRouting" enabled as well.

I.E., when both "RequestBootVarFallback" and "RequestBootVarRouting" are enabled, a change of the startup disk triggers an immediate reboot back into OpenCore. Sometimes the reboot would stop halfway through and restart.

With only "RequestBootVarRouting" enabled, there are no issues.

This was a problem as CarbonCopyCloner seems to change the startup disk to the current disk when there isn't any set (the case after blessing OpenCore).

So I was getting these reboots when my scheduled backups ran until I managed to isolate what I had changed around when they started happening and it was enabling "RequestBootVarFallback" in addition to "RequestBootVarRouting" which had been there.

I have now disabled it and all seems fine.

 

[Dayo]

In the security key. The sum of all is F in Hex=15 in Decimal.

Description: Sensitive data exposure bitmask (sum) to operating system.
• 0x01 — Expose printable booter path as an UEFI variable.
• 0x02 — Expose OpenCore version as an UEFI variable.
• 0x04 — Expose OpenCore version in boot picker menu title.
• 0x08 — Expose OEM information as a set of UEFI variables.

Went with all except "Expose OpenCore version in boot picker menu title".
This gives 0xB in Hex which is 11 in Decimal:

<key>ExposeSensitiveData</key>
            <integer>11</integer>

Terminal output equivalent to yours except does not show version string in boot picker:

$ nvram 4D1FDA02-38C7-4A6A-9CC6-4BCCA8B30102:opencore-version
4D1FDA02-38C7-4A6A-9CC6-4BCCA8B30102:opencore-version    REL-056-2020-03-02
$ nvram 4D1FDA02-38C7-4A6A-9CC6-4BCCA8B30102:oem-product
4D1FDA02-38C7-4A6A-9CC6-4BCCA8B30102:oem-product    MacPro3,1
$ nvram 4D1FDA02-38C7-4A6A-9CC6-4BCCA8B30102:oem-vendor
4D1FDA02-38C7-4A6A-9CC6-4BCCA8B30102:oem-vendor    Apple Inc.
$ nvram 4D1FDA02-38C7-4A6A-9CC6-4BCCA8B30102:oem-board
4D1FDA02-38C7-4A6A-9CC6-4BCCA8B30102:oem-board    Mac-F42C88C8

 

[Ludacrisvp]

So I was getting these reboots when my scheduled backups ran until I managed to isolate what I had changed around when they started happening and it was enabling "RequestBootVarFallback" in addition to "RequestBootVarRouting" which had been there.

I have now disabled it and all seems fine.

Nice, disabled the fallback option and now bless doesn't cause panic. +1

 

[Teletoby]

This error I get when I try to set the EFI partition, or isn’t this a error?

 

[jameslmoser]

Nice, disabled the fallback option and now bless doesn't cause panic. +1

If you install windows in UEFI mode, and don't have any software thats trying to change the startup disk, its probably still safer to leave this enabled, yes? It will prevent windows from making changes that will boot directly into the UEFI mode without the NVRAM protection provided by OpenCore? Just always use the opencore boot picker to change/manage your boot partition and you shouldn't have any problems?

Hopefully this is something they can fix soon enough.

 

[Ludacrisvp]

If you install windows in UEFI mode, and don't have any software thats trying to change the startup disk, its probably still safer to leave this enabled, yes? It will prevent windows from making changes that will boot directly into the UEFI mode without the NVRAM protection provided by OpenCore? Just always use the opencore boot picker to change/manage your boot partition and you shouldn't have any problems?

Hopefully this is something they can fix soon enough.

Don’t misunderstand this.
we had a setting enabled that based on reading the doc again we probably shouldn’t have enabled. The normal request boot var routing function that allows you to change OS using the prefpane works fine. It was just breaking due to the fallback one being enabled.