Help Installing Big Sur on Xeon W-3375 and Supemicro X12SPA-TF

[yapan4]

In macos you can try Hibernation type "3"(S4 Hibernation or suspend to disk without power off) as alternative to S3(type "0" or suspend to RAM)

Edited February 5, 2022 by yapan4

[Balamut]

On 2/5/2022 at 1:08 AM, yapan4 said:

In macos you can try Hibernation type "3"(S4 Hibernation or suspend to disk without power off) as alternative to S3(type "0" or suspend to RAM)

 

Got this from Supermicro BIOS team:

Unfortunately, S3 & S4 support is related to the hardware/firmware design also.  So, there is no way to support those power mode.

So, ether deal with no sleep or use it until MAYBE Asus/Gigabyte will come up with WS board for W-33xx.

 

On the side note, Monterey is still no GO.

 

[Balamut]

Monterey Progress!!!!!

 

After re-flashing and resetting the bios, I've got Progress!!!!!

 

 

Bluetooth is a no go for Fenvi.

 

Run the benches multiple times so far so good, no issues.

 

 

 

 

 

So far seems like everything works, will install 12.2 later on tonight to test it, then will work on AQC113 to get it working without freezes. Will post full BIOS settings later on. 

Seems like everything is solid except sleep, will have more tests as in Photoshop/Lightroom/Final Cut in a few days to see if it is sold and can be a daily driver.

 

 

 

Found this error in the logs:

 

ACPI: Executed 13 blocks of module-level executable AML code
ACPI Error: Could not map memory at 0x0000200FFFF69004, size 12 (20160930/exregion-297)
ACPI Exception: AE_NO_MEMORY, Returned by Handler for [SystemMemory] (20160930/evregion-388)
[HWAK] @0000A #0070:  HRD = 0x01

No Local Variables are initialized for method [HWAK]

No Arguments are initialized for method [HWAK]

ACPI Error: Method parse/execution failed [\_SB.PC00.HEC2.HWAK] (Node ffffff8e63c8bf00), AE_NO_MEMORY (20160930/psparse-632)
ACPI Error: Method parse/execution failed [\_SB.PC00.HEC2._INI] (Node ffffff8e63c8b0f0), AE_NO_MEMORY (20160930/psparse-632)
ACPI Exception: AE_NO_MEMORY, during \_SB.PC00.HEC2._INI execution (20160930/nsinit-770)
ACPI: sleep states S5
ACPI Error: [DTGP] Namespace lookup failure, AE_NOT_FOUND (20160930/psargs-463)
[_DSM] @00066 #002D:  

Initialized Local Variables for method [_DSM]:
    Local0: ffffff932fb49640 <Obj>             Package ffffff932fb49640

Initialized Arguments for Method [_DSM]:  (0 arguments defined for method invocation)
    Arg0:   ffffff932fb49910 <Obj>             Buffer(16) C6 B7 B5 A0 18 13 1C 44
    Arg1:   ffffff932fb118c0 <Obj>             Integer 0000000000000001
    Arg2:   ffffff932fb11f00 <Obj>             Integer 0000000000000000
    Arg3:   ffffff932fb49000 <Obj>             Package ffffff932fb49000

ACPI Error: Method parse/execution failed [\_SB.PC00.RP09.XGBE._DSM] (Node ffffff8e63c9c570), AE_NOT_FOUND (20160930/psparse-632)
ACPI: breakpoint Executed AML Breakpoint opcode

 

 

 

 

 

Edited February 8, 2022 by Balamut

[yapan4]

22 hours ago, Balamut said:

Got this from Supermicro BIOS team:

Unfortunately, S3 & S4 support is related to the hardware/firmware design also.  So, there is no way to support those power mode.

Sadly, I hope Supermicro and/or Asrock will change this situation in the future.

Edited February 8, 2022 by yapan4

[MaLd0n]

Some time ago mobile X1 carbon gen 6 have similar problem and with some manual procedure s3 start work good 

idk if work in this case but u can try

https://delta-xi.net/blog/#056

 

btw this machine is insane. congratz

[Balamut]

47 minutes ago, MaLd0n said:

Some time ago mobile X1 carbon gen 6 have similar problem and with some manual procedure s3 start work good 

idk if work in this case but u can try

https://delta-xi.net/blog/#056

 

btw this machine is insane. congratz

Thanks man, will look into it, looks very interesting.

 

Got the 12.2 Upgraded, worked like a charm except a little confusion with the disk names with OC, now only thing left is Bluetooth, ACQ113 and this stupid error:

 

ACPI: Executed 13 blocks of module-level executable AML code
ACPI Error: Could not map memory at 0x0000200FFFF69004, size 12 (20160930/exregion-297)
ACPI Exception: AE_NO_MEMORY, Returned by Handler for [SystemMemory] (20160930/evregion-388)
[HWAK] @0000A #0070:  HRD = 0x01

No Local Variables are initialized for method [HWAK]

No Arguments are initialized for method [HWAK]

ACPI Error: Method parse/execution failed [\_SB.PC00.HEC2.HWAK] (Node ffffff9315325150), AE_NO_MEMORY (20160930/psparse-632)
ACPI Error: Method parse/execution failed [\_SB.PC00.HEC2._INI] (Node ffffff93153251e0), AE_NO_MEMORY (20160930/psparse-632)
ACPI Exception: AE_NO_MEMORY, during \_SB.PC00.HEC2._INI execution (20160930/nsinit-770)
ACPI: sleep states S5
ACPI Error: [DTGP] Namespace lookup failure, AE_NOT_FOUND (20160930/psargs-463)
[_DSM] @00066 #002D:  

Initialized Local Variables for method [_DSM]:
    Local0: ffffff97e2e2a5a0 <Obj>             Package ffffff97e2e2a5a0

Initialized Arguments for Method [_DSM]:  (0 arguments defined for method invocation)
    Arg0:   ffffff97e2e2a3c0 <Obj>             Buffer(16) C6 B7 B5 A0 18 13 1C 44
    Arg1:   ffffff97e2e45a00 <Obj>             Integer 0000000000000001
    Arg2:   ffffff97e2e45b40 <Obj>             Integer 0000000000000000
    Arg3:   ffffff97e2e2ab40 <Obj>             Package ffffff97e2e2ab40

ACPI Error: Method parse/execution failed [\_SB.PC00.RP09.XGBE._DSM] (Node ffffff93152f2600), AE_NOT_FOUND (20160930/psparse-632)
ACPI: breakpoint Executed AML Breakpoint opcode

 

 

Look what I found in DSDT:

 Name (_S0, Package (0x04)  // _S0_: S0 System State
    {
        0x00, 
        0x00, 
        0x00, 
        0x00
    })
    Name (_S3, Package (0x04)  // _S3_: S3 System State
    {
        0x05, 
        0x00, 
        0x00, 
        0x00
    })
    Name (_S4, Package (0x04)  // _S4_: S4 System State
    {
        0x06, 
        0x00, 
        0x00, 
        0x00
    })
    Name (_S5, Package (0x04)  // _S5_: S5 System State
    {
        0x07, 
        0x00, 
        0x00, 
        0x00
    })

 

 

 

 

[Balamut]

Small update from the Supermicro people:

 

According to our BIOS team’s update, AMI has provided the solution of “Enabled/Disable Resize bar for the GPUs”.   However, we’re still waiting for the information regarding to “CFG Lock of the CPU” feature. 

 

[yapan4]

On 2/9/2022 at 12:06 AM, Balamut said:

and this stupid error:

 

ACPI: Executed 13 blocks of module-level executable AML code
ACPI Error: Could not map memory at 0x0000200FFFF69004, size 12 (20160930/exregion-297)
ACPI Exception: AE_NO_MEMORY, Returned by Handler for [SystemMemory] (20160930/evregion-388)
[HWAK] @0000A #0070:  HRD = 0x01

No Local Variables are initialized for method [HWAK]

No Arguments are initialized for method [HWAK]

ACPI Error: Method parse/execution failed [\_SB.PC00.HEC2.HWAK] (Node ffffff9315325150), AE_NO_MEMORY (20160930/psparse-632)
ACPI Error: Method parse/execution failed [\_SB.PC00.HEC2._INI] (Node ffffff93153251e0), AE_NO_MEMORY (20160930/psparse-632)
ACPI Exception: AE_NO_MEMORY, during \_SB.PC00.HEC2._INI execution (20160930/nsinit-770)
ACPI: sleep states S5
ACPI Error: [DTGP] Namespace lookup failure, AE_NOT_FOUND (20160930/psargs-463)
[_DSM] @00066 #002D:  

Initialized Local Variables for method [_DSM]:
    Local0: ffffff97e2e2a5a0 <Obj>             Package ffffff97e2e2a5a0

Initialized Arguments for Method [_DSM]:  (0 arguments defined for method invocation)
    Arg0:   ffffff97e2e2a3c0 <Obj>             Buffer(16) C6 B7 B5 A0 18 13 1C 44
    Arg1:   ffffff97e2e45a00 <Obj>             Integer 0000000000000001
    Arg2:   ffffff97e2e45b40 <Obj>             Integer 0000000000000000
    Arg3:   ffffff97e2e2ab40 <Obj>             Package ffffff97e2e2ab40

ACPI Error: Method parse/execution failed [\_SB.PC00.RP09.XGBE._DSM] (Node ffffff93152f2600), AE_NOT_FOUND (20160930/psparse-632)
ACPI: breakpoint Executed AML Breakpoint opcode

 

My ASUS hackintosh also has some ACPI errors. I don't know what it affects and how to deal with it and I just ignore them because my hackintosh generally works well

Spoiler

2022-02-12 11:14:40.747514+0200  localhost kernel[0]: (AppleACPIPlatform) <AppleACPIPlatform`AcpiOsVprintf> ACPI Error:
2022-02-12 11:14:40.747517+0200  localhost kernel[0]: (AppleACPIPlatform) <AppleACPIPlatform`AcpiOsVprintf> ACPI Error:
2022-02-12 11:14:40.747521+0200  localhost kernel[0]: (AppleACPIPlatform) <AppleACPIPlatform`AcpiOsVprintf> [TSDB]
2022-02-12 11:14:40.747523+0200  localhost kernel[0]: (AppleACPIPlatform) <AppleACPIPlatform`AcpiOsVprintf> [TSDB]
2022-02-12 11:14:40.747527+0200  localhost kernel[0]: (AppleACPIPlatform) <AppleACPIPlatform`AcpiOsVprintf>  Namespace lookup failure, AE_NOT_FOUND
2022-02-12 11:14:40.747529+0200  localhost kernel[0]: (AppleACPIPlatform) <AppleACPIPlatform`AcpiOsVprintf>  Namespace lookup failure, AE_NOT_FOUND
2022-02-12 11:14:40.747534+0200  localhost kernel[0]: (AppleACPIPlatform) <AppleACPIPlatform`AcpiOsVprintf>  (20160930/psargs-463)
2022-02-12 11:14:40.747537+0200  localhost kernel[0]: (AppleACPIPlatform) <AppleACPIPlatform`AcpiOsVprintf>  (20160930/psargs-463)
2022-02-12 11:14:40.747545+0200  localhost kernel[0]: (AppleACPIPlatform) <AppleACPIPlatform`AcpiOsVprintf> [TSDL] @00003 #002D:
2022-02-12 11:14:40.747547+0200  localhost kernel[0]: (AppleACPIPlatform) <AppleACPIPlatform`AcpiOsVprintf> [TSDL] @00003 #002D:
2022-02-12 11:14:40.747551+0200  localhost kernel[0]: (AppleACPIPlatform) <AppleACPIPlatform`AcpiOsVprintf> 
2022-02-12 11:14:40.747554+0200  localhost kernel[0]: (AppleACPIPlatform) <AppleACPIPlatform`AcpiOsVprintf> 
2022-02-12 11:14:40.747558+0200  localhost kernel[0]: (AppleACPIPlatform) <AppleACPIPlatform`AcpiOsVprintf> No Local Variables are initialized for method [TSDL]
2022-02-12 11:14:40.747560+0200  localhost kernel[0]: (AppleACPIPlatform) <AppleACPIPlatform`AcpiOsVprintf> No Local Variables are initialized for method [TSDL]
2022-02-12 11:14:40.747564+0200  localhost kernel[0]: (AppleACPIPlatform) <AppleACPIPlatform`AcpiOsVprintf> 
2022-02-12 11:14:40.747566+0200  localhost kernel[0]: (AppleACPIPlatform) <AppleACPIPlatform`AcpiOsVprintf> 
2022-02-12 11:14:40.747570+0200  localhost kernel[0]: (AppleACPIPlatform) <AppleACPIPlatform`AcpiOsVprintf> No Arguments are initialized for method [TSDL]
2022-02-12 11:14:40.747572+0200  localhost kernel[0]: (AppleACPIPlatform) <AppleACPIPlatform`AcpiOsVprintf> No Arguments are initialized for method [TSDL]
2022-02-12 11:14:40.747576+0200  localhost kernel[0]: (AppleACPIPlatform) <AppleACPIPlatform`AcpiOsVprintf> 
2022-02-12 11:14:40.747578+0200  localhost kernel[0]: (AppleACPIPlatform) <AppleACPIPlatform`AcpiOsVprintf> 
2022-02-12 11:14:40.747583+0200  localhost kernel[0]: (AppleACPIPlatform) <AppleACPIPlatform`AcpiOsVprintf> ACPI Error:
2022-02-12 11:14:40.747584+0200  localhost kernel[0]: (AppleACPIPlatform) <AppleACPIPlatform`AcpiOsVprintf> ACPI Error:
2022-02-12 11:14:40.747588+0200  localhost kernel[0]: (AppleACPIPlatform) <AppleACPIPlatform`AcpiOsVprintf> Method parse/execution failed
2022-02-12 11:14:40.747590+0200  localhost kernel[0]: (AppleACPIPlatform) <AppleACPIPlatform`AcpiOsVprintf> Method parse/execution failed
2022-02-12 11:14:40.747594+0200  localhost kernel[0]: (AppleACPIPlatform) <AppleACPIPlatform`AcpiOsVprintf> [\_SB.PTID.TSDL] (Node ffffff8bd42f1e70)
2022-02-12 11:14:40.747596+0200  localhost kernel[0]: (AppleACPIPlatform) <AppleACPIPlatform`AcpiOsVprintf> [\_SB.PTID.TSDL] (Node ffffff8bd42f1e70)
2022-02-12 11:14:40.747600+0200  localhost kernel[0]: (AppleACPIPlatform) <AppleACPIPlatform`AcpiOsVprintf> , AE_NOT_FOUND
2022-02-12 11:14:40.747601+0200  localhost kernel[0]: (AppleACPIPlatform) <AppleACPIPlatform`AcpiOsVprintf> , AE_NOT_FOUND
2022-02-12 11:14:40.747605+0200  localhost kernel[0]: (AppleACPIPlatform) <AppleACPIPlatform`AcpiOsVprintf>  (20160930/psparse-632)
2022-02-12 11:14:40.747608+0200  localhost kernel[0]: (AppleACPIPlatform) <AppleACPIPlatform`AcpiOsVprintf>  (20160930/psparse-632)
2022-02-12 11:14:40.747620+0200  localhost kernel[0]: (Contents\MacOS\CPUSensors) <compose failure [UUID]>
2022-02-12 11:14:40.747647+0200  localhost kernel[0]: (AppleACPIPlatform) <AppleACPIPlatform`AcpiOsVprintf> ACPI Error:
2022-02-12 11:14:40.747649+0200  localhost kernel[0]: (AppleACPIPlatform) <AppleACPIPlatform`AcpiOsVprintf> ACPI Error:
2022-02-12 11:14:40.747652+0200  localhost kernel[0]: (AppleACPIPlatform) <AppleACPIPlatform`AcpiOsVprintf> [TSDB]
2022-02-12 11:14:40.747654+0200  localhost kernel[0]: (AppleACPIPlatform) <AppleACPIPlatform`AcpiOsVprintf> [TSDB]
2022-02-12 11:14:40.747657+0200  localhost kernel[0]: (AppleACPIPlatform) <AppleACPIPlatform`AcpiOsVprintf>  Namespace lookup failure, AE_NOT_FOUND
2022-02-12 11:14:40.747659+0200  localhost kernel[0]: (AppleACPIPlatform) <AppleACPIPlatform`AcpiOsVprintf>  Namespace lookup failure, AE_NOT_FOUND
2022-02-12 11:14:40.747663+0200  localhost kernel[0]: (AppleACPIPlatform) <AppleACPIPlatform`AcpiOsVprintf>  (20160930/psargs-463)
2022-02-12 11:14:40.747665+0200  localhost kernel[0]: (AppleACPIPlatform) <AppleACPIPlatform`AcpiOsVprintf>  (20160930/psargs-463)
2022-02-12 11:14:40.747671+0200  localhost kernel[0]: (AppleACPIPlatform) <AppleACPIPlatform`AcpiOsVprintf> [OSDL] @00003 #002D:
2022-02-12 11:14:40.747672+0200  localhost kernel[0]: (AppleACPIPlatform) <AppleACPIPlatform`AcpiOsVprintf> [OSDL] @00003 #002D:
2022-02-12 11:14:40.747675+0200  localhost kernel[0]: (AppleACPIPlatform) <AppleACPIPlatform`AcpiOsVprintf> 
2022-02-12 11:14:40.747677+0200  localhost kernel[0]: (AppleACPIPlatform) <AppleACPIPlatform`AcpiOsVprintf> 
2022-02-12 11:14:40.747681+0200  localhost kernel[0]: (AppleACPIPlatform) <AppleACPIPlatform`AcpiOsVprintf> No Local Variables are initialized for method [OSDL]
2022-02-12 11:14:40.747684+0200  localhost kernel[0]: (AppleACPIPlatform) <AppleACPIPlatform`AcpiOsVprintf> No Local Variables are initialized for method [OSDL]
2022-02-12 11:14:40.747688+0200  localhost kernel[0]: (AppleACPIPlatform) <AppleACPIPlatform`AcpiOsVprintf> 
2022-02-12 11:14:40.747690+0200  localhost kernel[0]: (AppleACPIPlatform) <AppleACPIPlatform`AcpiOsVprintf> 
2022-02-12 11:14:40.747694+0200  localhost kernel[0]: (AppleACPIPlatform) <AppleACPIPlatform`AcpiOsVprintf> No Arguments are initialized for method [OSDL]
2022-02-12 11:14:40.747696+0200  localhost kernel[0]: (AppleACPIPlatform) <AppleACPIPlatform`AcpiOsVprintf> No Arguments are initialized for method [OSDL]
2022-02-12 11:14:40.747700+0200  localhost kernel[0]: (AppleACPIPlatform) <AppleACPIPlatform`AcpiOsVprintf> 
2022-02-12 11:14:40.747702+0200  localhost kernel[0]: (AppleACPIPlatform) <AppleACPIPlatform`AcpiOsVprintf> 
2022-02-12 11:14:40.747706+0200  localhost kernel[0]: (AppleACPIPlatform) <AppleACPIPlatform`AcpiOsVprintf> ACPI Error:
2022-02-12 11:14:40.747707+0200  localhost kernel[0]: (AppleACPIPlatform) <AppleACPIPlatform`AcpiOsVprintf> ACPI Error:
2022-02-12 11:14:40.747710+0200  localhost kernel[0]: (AppleACPIPlatform) <AppleACPIPlatform`AcpiOsVprintf> Method parse/execution failed
2022-02-12 11:14:40.747712+0200  localhost kernel[0]: (AppleACPIPlatform) <AppleACPIPlatform`AcpiOsVprintf> Method parse/execution failed
2022-02-12 11:14:40.747717+0200  localhost kernel[0]: (AppleACPIPlatform) <AppleACPIPlatform`AcpiOsVprintf> [\_SB.PTID.OSDL] (Node ffffff8bd42f13f0)
2022-02-12 11:14:40.747719+0200  localhost kernel[0]: (AppleACPIPlatform) <AppleACPIPlatform`AcpiOsVprintf> [\_SB.PTID.OSDL] (Node ffffff8bd42f13f0)
2022-02-12 11:14:40.747722+0200  localhost kernel[0]: (AppleACPIPlatform) <AppleACPIPlatform`AcpiOsVprintf> , AE_NOT_FOUND
2022-02-12 11:14:40.747724+0200  localhost kernel[0]: (AppleACPIPlatform) <AppleACPIPlatform`AcpiOsVprintf> , AE_NOT_FOUND
2022-02-12 11:14:40.747727+0200  localhost kernel[0]: (AppleACPIPlatform) <AppleACPIPlatform`AcpiOsVprintf>  (20160930/psparse-632)
2022-02-12 11:14:40.747729+0200  localhost kernel[0]: (AppleACPIPlatform) <AppleACPIPlatform`AcpiOsVprintf>  (20160930/pspars

 

Edited February 12, 2022 by yapan4

[yapan4]

For review all BIOS settings use AMIbcp windows app

Archve.zip

[Balamut]

@yapan4 Thank you for the app, still trying to figure out how to properly use it but initial observation:

 

BIOS does contain S4/Hibernate sleep state, at least its in the BIOS, it shows Show: Yes but I can't seem to see it in the bios.

 

 

Didn't know the Xeon Ws can be overclocked??

 

 

 

 

 

And HEC1-3 not showing int he BIOS, the entire Workstation ME is not showing.

 

 

 

Do you guys know how to enable them to show in the BIOS? Red the manual but it doesn't make sense. For example, AMiBCP says Show: Yes
 

 

But the BIOS is missing a lot of the items.

 

 

Is it because of the Access/Use is set to default? Should it be changed to Supervisor?

 

Thanks fellas.

 

 

 

 

 

 

 

Edited February 23, 2022 by Balamut

[Anonym323]

On 2/23/2022 at 9:36 PM, Balamut said:

@yapan4 

 

Is it because of the Access/Use is set to default? Should it be changed to Supervisor?

 

Thanks fellas.

 

@BalamutSet it to User and safe the firmware…

 

How do you flashed the firmware ?

Started a new project also on this platform but still in the early stage.

Using an X12DPL-NT6

Will probably order an debugger today for it…

Edited May 15, 2022 by Anonym323

[metacollin]

Don't waste your time editing the BIOS menus. Supermicro typically has an option to unlock the chipset (which normally will unlock the 0E2h MSR in most other motherboard vendors' BIOSes) hidden away that you can reveal with the AMI bios editor, but the option doesn't do anything with Supermicro.  They just didn't bother to remove it, but it is just a dummy placeholder and turning it to Enabled or Disabled has no effect.

 

But don't worry, I got you fam

 

Let me see.... oh yep, there's the little varmit!  Setting the lock bit, which is bit 15 (0x8000 in hex) of the 0E2h MSR upon system initialization:

 

 

 

We can't have that, can we?  Better just have it do exactly what it was doing, only without bit 15 set.  In this case, just a somewhat pointless but liberating write of 0x0000 to 0E2h:

 

 

 

I'll put together a pull request for UEFIPatch if you test it and it works.  It should work, but sometimes there is another spot where they will check if the lock bit is set, and if not, lock it.  I typically see that kind of bs from Asus  but never from Supermicro.  So hopefully this is all you'll need.

 

And just to be clear, there is no way this can cause any problems or brick your motherboard (and even if it did, you can just flash the BIOS from IPMI so it is about as safe as you can get).

 

My other patch still works for unlocking the 1AAh MSR for your BIOS.  So grab UEFIPatch if you don't have it already, and replace the patches.txt file in the UEFIPatch folder with the new one I've attached.

 

Then just run UEFIPatch from the terminal on the .bin bios image, you should see 2 successful patches applied.   Both your 1AAh and 0E2h MSRs should be unlocked and you can boot macOS without any power management patches turned on. 

 

 

patches.txt

Edited June 13, 2022 by metacollin

[Balamut]

On 6/13/2022 at 1:02 PM, metacollin said:

Don't waste your time editing the BIOS menus. Supermicro typically has an option to unlock the chipset (which normally will unlock the 0E2h MSR in most other motherboard vendors' BIOSes) hidden away that you can reveal with the AMI bios editor, but the option doesn't do anything with Supermicro.  They just didn't bother to remove it, but it is just a dummy placeholder and turning it to Enabled or Disabled has no effect.

 

But don't worry, I got you fam

 

Let me see.... oh yep, there's the little varmit!  Setting the lock bit, which is bit 15 (0x8000 in hex) of the 0E2h MSR upon system initialization:

patches.txt 2.95 kB · 12 downloads

 

Thank you brother, will test it in a few days will get back to you. Sorry work keeps me pretty busy lately.

 

 

 

[khe91]

Hi Balamut,

 

Could you please share your EFI for the X12SPA-TF ?

Got a X12SPi-TF (X12SPA-TF is not available here).

 

Thanks,

 

khe91

[etorix]

Hi @khe91. A December 2021 EFI for the X12SPA-TF was posted above. Hopefully, this can get you started with the X12SPi-TF.

[Balamut]

@metacollin Just wanted to let you know that I'm still working on trying to get the BIOS to update, apparently Supermicro got BIOS verification security on and can't seem to figure out how to remove that security check.

 

On 8/15/2022 at 4:50 AM, khe91 said:

Could you please share your EFI for the X12SPA-TF ?

Got a X12SPi-TF (X12SPA-TF is not available here).

 

Here.. this is the latest one, It's OC 8.3 Release version but config file got debug on.

OC.zip

 

 

Guys does anyone experienced with Clover can take a look at mine? It shutdown the system right on the #####################, CPUID is there and seems like the XCPM patches are on too.

 

Main reason I'm want to try clover is because I'm having a HUGE issue with the second GPU, no matter what PCIE SLOT (except PCIE 1, I get PCI Conf Begin freeze on it) the second GPU is just shows as display even thought individually they work perfectly fine. This is the second C62X system that does that and both are with the OC.

 

Clover.zip

 

 

[khe91]

Thanks @Balamut for the files.

 

Your OC hangs on  17:551 00:019 AAPL: #[EB|LOG:EXITBS:START] 2022-08-20T07:52:51

Clover switches the PC off

 

Guess I'm have some wrong BIOS settings.

I changed:

 

---

Advanced - PCIe/PCI/PnP Configuration

Above 4G Decoding Enabled -> Disabled

SR-IOV Enabled -> Disabled

Consistent Device Name Support Disabled -> Enabled

 

Advanced - ACPI Settings

NUMA Enabled -> Disabled

 

Security

Secure Boot

CSM Enabled -> Disabled

---

 

Which BIOS version do you use ?

Do you remember your changed BIOS settings ?

Edited August 20, 2022 by khe91

[Balamut]

On 8/20/2022 at 1:04 AM, khe91 said:

Your OC hangs on  17:551 00:019 AAPL: #[EB|LOG:EXITBS:START] 2022-08-20T07:52:51

Clover switches the PC off

 

I've got the same issue with Clover, on the OC looks like the CFG Lock issue, are you sure the XCPM are checked in the Quirks in the OC?

 

 

 

On 8/20/2022 at 1:04 AM, khe91 said:

---

Advanced - PCIe/PCI/PnP Configuration

Above 4G Decoding Enabled -> Disabled

SR-IOV Enabled -> Disabled

Consistent Device Name Support Disabled -> Enabled

 

Advanced - ACPI Settings

NUMA Enabled -> Disabled

 

Security

Secure Boot

CSM Enabled -> Disabled

---

 

Mine: 

Advanced - PCIe/PCI/PnP Configuration

Above 4G Decoding - Enabled

SR-IOV - Disabled

Consistent Device Name Support - Enabled

 

Advanced - ACPI Settings

NUMA -  Enabled 

 

Security

Secure Boot

CSM - Disabled

 

On 8/20/2022 at 1:04 AM, khe91 said:

Which BIOS version do you use ?

 

It's 1.1 

BIOS_X12SPA-1B57_20210621_1.1

Although the boards might seem similar the bios will be completely  different. 

 

On 8/20/2022 at 1:04 AM, khe91 said:

Do you remember your changed BIOS settings ?

To be honest I'm not sure, I was tweaking and playing with setting for almost 3 month until I've managed to get it working, waiting on the new SSD to arrive to get some test running, the SN750 SE(I didn't noticed the SE part when I was ordering it) that I've got is oddly slow on the boot.

 

 

Also on the side note I've never heard of Xeon Gold 5315Y (Whats Y ?), I doubt it got anything to do with the freeze, but gurus here might be able to help you more with that.

 

 

Edited August 22, 2022 by Balamut

[khe91]

Yes looks like CFG lock, the Quirks are activated.

 

I will try to get the 1.1 BIOS version for my board.

 

For your information about the Xeon (Y is low power version):

https://ark.intel.com/content/www/us/en/ark/products/215286/intel-xeon-gold-5315y-processor-12m-cache-3-20-ghz.html

https://www.intel.com/content/www/us/en/processors/processor-numbers.html

[etorix]

17 hours ago, Balamut said:

Also on the side note I've never heard of Xeon Gold 5315Y (Whats Y ?), I doubt it got anything to do with the freeze, but gurus here might be able to help you more with that.

Intel's product stack is designed to confuse, and that's the part Intel does best these days.

 

https://www.servethehome.com/3rd-gen-intel-xeon-scalable-ice-lake-sku-list-and-value-analysis/

 

Quote

With this generation, not only do we have the “Ice Lake” designation, but Intel has made two-thirds of its SKUs from Ark special by adding additional letters at the end of the four digits.

• Y – Supports Intel Speeds Select Technology Performance Profile 2.0 (SST-PP)
• N – Networking Optimized
• P – IaaS Cloud Specialized Processor
• V – SaaS Cloud Specialized Processor
• Q – Liquid Cooling Optimized
• M – Media Processing Optimized
• T – Long-life and Thermal friendly
• S – SGX 512GB secure enclave (Some 512GB parts)
• U – Usually means single socket only or “Uni-processor” unless there is a single socket N SKU that will instead use N

When two-thirds of items are reportedly "special", I'd say that the CPUs without an added letter are now the "special exceptions". But what do I know about marketing?

[khe91]

On 8/17/2022 at 5:04 AM, Balamut said:

 

 

Main reason I'm want to try clover is because I'm having a HUGE issue with the second GPU, no matter what PCIE SLOT (except PCIE 1, I get PCI Conf Begin freeze on it) the second GPU is just shows as display even thought individually they work perfectly fine. This is the second C62X system that does that and both are with the OC.

 

@Balamut Since its not booting native, I'm playing with Proxmox and there is something working OOB:

 

[Balamut]

Play with the settings, quirks, what I did was use com port and capture the log from clover and macOS, it’ll give you the general idea of what’s wrong.  
 

I see both GPUs are detected, PCI Passtrought? 
 

I’m getting convinced that C62x boards and their PCIE implementation are breaking something, could be the way they did PCIE/NVME sharing and bifurcation. Doesn’t explain though why on Supermicro board regardless what slot you use the second GPU won’t get fully recognized.  

On the side note I’ve been reading some rumors about the new Mac Pro, seems like it is on the way of getting the M2 which I suspect will be the end of the new  x86 hardware, but I’ve been wrong before so who knows. 

 

[khe91]

8 hours ago, Balamut said:

Play with the settings, quirks, what I did was use com port and capture the log from clover and macOS, it’ll give you the general idea of what’s wrong.  
 

I see both GPUs are detected, PCI Passtrought? 
 

I’m getting convinced that C62x boards and their PCIE implementation are breaking something, could be the way they did PCIE/NVME sharing and bifurcation. Doesn’t explain though why on Supermicro board regardless what slot you use the second GPU won’t get fully recognized.  

On the side note I’ve been reading some rumors about the new Mac Pro, seems like it is on the way of getting the M2 which I suspect will be the end of the new  x86 hardware, but I’ve been wrong before so who knows. 

 

 

Will try the com port, just ordered a W-3323, maybe the Xeon Gold 5315Y is the problem, since it has different P and E cores like the latest i5/i7.

 

Both GPUs detected and working well with PCIe-Passtrough under Proxmox.

 

No problem with the C62x boards, know some people with working macOS systems and two graphic cards, they just don't like to share their configs.

It's all in the SSDT, secret are the graphic card PCIe bridge definitions.

Under Proxmox I'm hiding the bridge and passthrough direct the GPU.

Edited August 24, 2022 by khe91

[Balamut]

11 hours ago, khe91 said:

Will try the com port, just ordered a W-3323, maybe the Xeon Gold 5315Y is the problem, since it has different P and E cores like the latest i5/i7.

Could be, the 3375 works great, haven't seen any kernel panic yet, except I had to cripple the cores to 32. 

 

 

11 hours ago, khe91 said:

Both GPUs detected and working well with PCIe-Passtrough under Proxmox.

 

No problem with the C62x boards, know some people with working macOS systems and two graphic cards, they just don't like to share their configs.

It's all in the SSDT, secret are the graphic card PCIe bridge definitions.

Under Proxmox I'm hiding the bridge and passthrough direct the GPU.

I'd even appreciate any direction or hints from them.

 

Somewhere something is preventing the second GPU from being seen, even changing from MacPro to iMacPro didn't do anything. Going to try Clover see if there is anything different in the system.

 

[Anonym323]

Sounds to me like you use a ES Chip and not an Retail one… I know this behavior from ES Chips only @Balamut