Jump to content

Open CORE Kernel & Kext patch for X99/X299 motherboard


Message added by fantomas

The contents of these post are outdated, all the XCPM patches for X99 and similar chipsets can be enabled within Kernel → Quirks section

400 posts in this topic

Recommended Posts

On 1/14/2020 at 12:02 AM, Mike Ranger said:

 

Going forward with OpenCore 0.54, none of these patches are needed anymore.... all covered by the built-in quirks.

 

 

Hi Mike Ranger, how I use quirks instead patches? Thanks...

@lucke enabling ones your cpu/motherboard need

flag quirks you need in proper opencore config.plist section

 

try also to read from page 22 of configuration.pdf (you can find easily in 0.5.4 release zip doc folder)

Edited by Guest
typo
4 hours ago, fabiosun said:

@lucke enabling ones your cpu/motherboard need

flag quirks you need in proper opencore config.plist section

 

try also to read from page 22 of configuration.pdf (you can find easily in 0.5.4 release zip doc folder)

Thanks.

  • 4 weeks later...

XCPM Patches 10.15.3  Modify By N.Mano

 

#1

Base: _cpuid_set_info
Comment: FakeCPUID - Model 0xE (0x5E,0x3F SKL,Haswell), 
Count: 0
Enabled: YES
Find:   89C1C0E9 04FEC1
Identifier: kernel
Limit: 0
Mask: (LEAVE THIS EMPTY)
MatchKernel: 
Replace: 
ReplaceMask: (LEAVE THIS EMPTY)
Skip: 0

#2

Base: _cpuid_set_info
Comment: FakeCPUID - Ext 0x5 (0x5E,0x3F SKL,Haswell), 
Count: 0
Enabled: YES
Find:   89C1C1E9 1089
Identifier: kernel
Limit: 0
Mask: (LEAVE THIS EMPTY)
MatchKernel: 
Replace: 
ReplaceMask: (LEAVE THIS EMPTY)
Skip: 0

#3

Base: xcpm_core_scope_msrs
Comment: xcpm_core_scope_msrs © Pike R. Alpha
Count: 1
Enabled: YES
Find:    31D2E857 D2FFFF
Identifier: kernel
Limit: 0
Mask: (LEAVE THIS EMPTY)
MatchKernel: 
Replace: 31D29090 909090
ReplaceMask: (LEAVE THIS EMPTY)
Skip: 0


#4

Base: xcpm_pkg_scope_msrs
Comment: xcpm_pkg_scope_msrs © Pike R. Alpha
Count: 0
Enabled: YES
Find:    31D2E8CE 88FFFF
Identifier: kernel
Limit: 0
Mask: (LEAVE THIS EMPTY)
MatchKernel: 
Replace: 31D29090 909090
ReplaceMask: (LEAVE THIS EMPTY)
Skip: 0

#5

Base: _xcpm_SMT_scope_msrs #1
Comment: _xcpm_SMT_scope_msrs #1 (c) Pike R. Alpha
Count: 0
Enabled: YES
Find:    BE060000 005DE908 000000
Identifier: kernel
Limit: 0
Mask: (LEAVE THIS EMPTY)
MatchKernel: 
Replace: BE060000 005DC390 909090
ReplaceMask: (LEAVE THIS EMPTY)
Skip: 0


#6

Base: _xcpm_SMT_scope_msrs_2
Comment: _xcpm_SMT_scope_msrs_2 (c) Pike R. Alpha
Count: 0
Enabled: YES
Find:   31D2E83E 9AFFFF
Identifier: kernel
Limit: 0
Mask: (LEAVE THIS EMPTY)
MatchKernel: 
Replace: 31D29090 909090
ReplaceMask: (LEAVE THIS EMPTY)
Skip: 0

#7

Base: _xcpm_ performance
Comment: _xcpm_ performance_patch © Pike R. Alpha
Count: 0
Enabled: YES
Find: 41C1E608 4963D689 D048C1EA 20
Identifier: kernel
Limit: 0
Mask: (LEAVE THIS EMPTY)
MatchKernel: 
Replace: 41C1E608 B800FF00 0031D290 90
ReplaceMask: (LEAVE THIS EMPTY)
Skip: 0

#8 This patch not for X299

Base: xcpm_bootstrap
Comment: xcpm_bootstrap_HASwell_E © Pike R. Alpha 10.13,10.14,10.15
Count: 0
Enabled: YES
Find:    8D43C43C 227722
Identifier: kernel
Limit: 0
Mask: (LEAVE THIS EMPTY)
MatchKernel: 
Replace: 8D43C13C 227722
ReplaceMask: (LEAVE THIS EMPTY)
Skip: 0

#9

Base: xcpm MSR Patch 4206
Comment:xcpm MSR Patch 4206 © Pike R. Alpha
Count: 0
Enabled: YES
Find:    42060000 DC330000 00000000 00000000 1F000000 00000000 18000000 00000000 00000000 00000000 00000000 00000000 
Identifier: kernel
Limit: 0
Mask: (LEAVE THIS EMPTY)
MatchKernel: 
Replace: 42060000 00000000 00000000 00000000 1F000000 00000000 18000000 00000000 00000000 00000000 00000000 00000000 
ReplaceMask: (LEAVE THIS EMPTY)
Skip: 0

#10

Base: xcpm E2 Patch 
Comment:C6/C7_If=E2-Enabled=0x7E000403 
Count: 0
Enabled: YES
Find:    E2000000 4C000000 00000000 00000000 0F040000 00000000 0500001E 
Identifier: kernel
Limit: 0
Mask: (LEAVE THIS EMPTY)
MatchKernel: 
Replace: E2000000 4C000000 00000000 00000000 0F040000 00000000 0304007E 
ReplaceMask: (LEAVE THIS EMPTY)
Skip: 0

#11

Base: xcpm 3A06 Patch 
Comment:SammlerG Sierra xcpm MSR Patch 3 / xcpm support Pike R. Alpha/0x82D390/MSR_PP0_POLICY 0x63a
Count: 0
Enabled: YES
Find:    3A060000 DC330000 00000000 00000000 1F 
Identifier: kernel
Limit: 0
Mask: (LEAVE THIS EMPTY)
MatchKernel: 
Replace: 3A060000 00000000 00000000 00000000 00 
ReplaceMask: (LEAVE THIS EMPTY)
Skip: 0

 

Edited by nmano
  • Like 1
2 hours ago, nmano said:

XCPM Patches 10.15.3  Modify By N.Mano

 


#1

Base: _cpuid_set_info
Comment: FakeCPUID - Model 0xE (0x5E,0x3F SKL,Haswell), 
Count: 0
Enabled: YES
Find:   89C1C0E9 04FEC1
Identifier: kernel
Limit: 0
Mask: (LEAVE THIS EMPTY)
MatchKernel: 
Replace: 
ReplaceMask: (LEAVE THIS EMPTY)
Skip: 0

#2

Base: _cpuid_set_info
Comment: FakeCPUID - Ext 0x5 (0x5E,0x3F SKL,Haswell), 
Count: 0
Enabled: YES
Find:   89C1C1E9 1089
Identifier: kernel
Limit: 0
Mask: (LEAVE THIS EMPTY)
MatchKernel: 
Replace: 
ReplaceMask: (LEAVE THIS EMPTY)
Skip: 0

#3

Base: xcpm_core_scope_msrs
Comment: xcpm_core_scope_msrs © Pike R. Alpha
Count: 1
Enabled: YES
Find:    31D2E857 D2FFFF
Identifier: kernel
Limit: 0
Mask: (LEAVE THIS EMPTY)
MatchKernel: 
Replace: 31D29090 909090
ReplaceMask: (LEAVE THIS EMPTY)
Skip: 0


#4

Base: xcpm_pkg_scope_msrs
Comment: xcpm_pkg_scope_msrs © Pike R. Alpha
Count: 0
Enabled: YES
Find:    31D2E8CE 88FFFF
Identifier: kernel
Limit: 0
Mask: (LEAVE THIS EMPTY)
MatchKernel: 
Replace: 31D29090 909090
ReplaceMask: (LEAVE THIS EMPTY)
Skip: 0

#5

Base: _xcpm_SMT_scope_msrs #1
Comment: _xcpm_SMT_scope_msrs #1 (c) Pike R. Alpha
Count: 0
Enabled: YES
Find:    BE060000 005DE908 000000
Identifier: kernel
Limit: 0
Mask: (LEAVE THIS EMPTY)
MatchKernel: 
Replace: BE060000 005DC390 909090
ReplaceMask: (LEAVE THIS EMPTY)
Skip: 0


#6

Base: _xcpm_SMT_scope_msrs_2
Comment: _xcpm_SMT_scope_msrs_2 (c) Pike R. Alpha
Count: 0
Enabled: YES
Find:   31D2E83E 9AFFFF
Identifier: kernel
Limit: 0
Mask: (LEAVE THIS EMPTY)
MatchKernel: 
Replace: 31D29090 909090
ReplaceMask: (LEAVE THIS EMPTY)
Skip: 0

#7

Base: _xcpm_ performance
Comment: _xcpm_ performance_patch © Pike R. Alpha
Count: 0
Enabled: YES
Find: 41C1E608 4963D689 D048C1EA 20
Identifier: kernel
Limit: 0
Mask: (LEAVE THIS EMPTY)
MatchKernel: 
Replace: 41C1E608 B800FF00 0031D290 90
ReplaceMask: (LEAVE THIS EMPTY)
Skip: 0

#8 This patch not for X299

Base: xcpm_bootstrap
Comment: xcpm_bootstrap_HASwell_E © Pike R. Alpha 10.13,10.14,10.15
Count: 0
Enabled: YES
Find:    8D43C43C 227722
Identifier: kernel
Limit: 0
Mask: (LEAVE THIS EMPTY)
MatchKernel: 
Replace: 8D43C13C 227722
ReplaceMask: (LEAVE THIS EMPTY)
Skip: 0

#9

Base: xcpm MSR Patch 4206
Comment:xcpm MSR Patch 4206 © Pike R. Alpha
Count: 0
Enabled: YES
Find:    42060000 DC330000 00000000 00000000 1F000000 00000000 18000000 00000000 00000000 00000000 00000000 00000000 
Identifier: kernel
Limit: 0
Mask: (LEAVE THIS EMPTY)
MatchKernel: 
Replace: 42060000 00000000 00000000 00000000 1F000000 00000000 18000000 00000000 00000000 00000000 00000000 00000000 
ReplaceMask: (LEAVE THIS EMPTY)
Skip: 0

#10

Base: xcpm E2 Patch 
Comment:C6/C7_If=E2-Enabled=0x7E000403 
Count: 0
Enabled: YES
Find:    E2000000 4C000000 00000000 00000000 0F040000 00000000 0500001E 
Identifier: kernel
Limit: 0
Mask: (LEAVE THIS EMPTY)
MatchKernel: 
Replace: E2000000 4C000000 00000000 00000000 0F040000 00000000 0304007E 
ReplaceMask: (LEAVE THIS EMPTY)
Skip: 0

#11

Base: xcpm 3A06 Patch 
Comment:SammlerG Sierra xcpm MSR Patch 3 / xcpm support Pike R. Alpha/0x82D390/MSR_PP0_POLICY 0x63a
Count: 0
Enabled: YES
Find:    3A060000 DC330000 00000000 00000000 1F 
Identifier: kernel
Limit: 0
Mask: (LEAVE THIS EMPTY)
MatchKernel: 
Replace: 3A060000 00000000 00000000 00000000 00 
ReplaceMask: (LEAVE THIS EMPTY)
Skip: 0

 

 

Base: _cpuid_set_info

Replase: -----     should be empty?

Edited by pitrysha
  • Like 1
12 hours ago, pitrysha said:

 

Base: _cpuid_set_info

Replase: -----     should be empty?

 
I am using now

#Kernel->Emulate->

 

#FakeCPUID         0x050654  Skylake-X/Skylake-H 0x0406E3

 

Cpuid1Data   <->  E3060400 00000000 00000000 00000000

Cpuid1Mask  <->   FFFFFFFF 00000000 00000000 00000000

 

#FakeCPUID         0x0306F2   Haswell-E/Haswell 0x0306C3

 

Cpuid1Data   <->  C3060300 00000000 00000000 00000000

Cpuid1Mask  <->   FFFFFFFF 00000000 00000000 00000000 

 

#FakeCPUID         0x040674   Broadwell-E/Broadwell 0x0306D4

 

Cpuid1Data   <-> D4060300 00000000 00000000 00000000

Cpuid1Mask  <->  FFFFFFFF 00000000 00000000 00000000

 

But I try Pikes

https://pikeralpha.wordpress.com/2016/07/26/xcpm-for-unsupported-processor/

kernel location 0x1fb451 (0xE9 in DP2/DP3/DP4, 0xF1 in DP1) _cpuid_set_info
//
// Broadwell E (0x4F)   -> Broadwell (0x47)  => 0x4F-0x47 = 0x08  -> change 0xE9 into 0xE1
// Haswell E (0x3F)     -> Haswell (0x3C)    => 0x3F-0x3C = 0x03  -> change 0xE9 into 0xE6
// Ivy Bridge E (0x3E)  -> Ivy Bridge (0x3A) => 0x3E-0x3A = 0x04  -> change 0xE9 into 0xE5
 
For Haswell 0xE9 into 0xE6
89C1C0E9 04FEC1
89C1C0E6 04FEC1
 
For Broadwell change 0xE9 into 0xE1
89C1C0E9 04FEC1
89C1C0E1 04FEC1
Edited by nmano
  • Like 1
40 minutes ago, nmano said:
 
I am using now

#Kernel->Emulate->

 

#FakeCPUID         0x050654  Skylake-X/Skylake-H 0x0406E3

 

Cpuid1Data   <->  E3060400 00000000 00000000 00000000

Cpuid1Mask  <->   FFFFFFFF 00000000 00000000 00000000

 

#FakeCPUID         0x0306F2   Haswell-E/Haswell 0x0306C3

 

Cpuid1Data   <->  C3060300 00000000 00000000 00000000

Cpuid1Mask  <->   FFFFFFFF 00000000 00000000 00000000 

 

#FakeCPUID         0x040674   Broadwell-E/Broadwell 0x0306D4

 

Cpuid1Data   <-> D4060300 00000000 00000000 00000000

Cpuid1Mask  <->  FFFFFFFF 00000000 00000000 00000000

 

But I try Pikes

https://pikeralpha.wordpress.com/2016/07/26/xcpm-for-unsupported-processor/

kernel location 0x1fb451 (0xE9 in DP2/DP3/DP4, 0xF1 in DP1) _cpuid_set_info
//
// Broadwell E (0x4F)   -> Broadwell (0x47)  => 0x4F-0x47 = 0x08  -> change 0xE9 into 0xE1
// Haswell E (0x3F)     -> Haswell (0x3C)    => 0x3F-0x3C = 0x03  -> change 0xE9 into 0xE6
// Ivy Bridge E (0x3E)  -> Ivy Bridge (0x3A) => 0x3E-0x3A = 0x04  -> change 0xE9 into 0xE5
 
For Haswell 0xE9 into 0xE6
89C1C0E9 04FEC1
89C1C0E6 04FEC1
 
For Broadwell change 0xE9 into 0xE1
89C1C0E9 04FEC1
89C1C0E1 04FEC1

Thanks.

I don’t understand about this part

 

Base: _cpuid_set_info

Comment: FakeCPUID - Model 0xE (0x5E,0x3F SKL,Haswell), 

Count: 0

Enabled: YES

Find:   89C1C0E9 04FEC1

Identifier: kernel

Limit: 0

Mask: (LEAVE THIS EMPTY)

MatchKernel: 

Replace: 

ReplaceMask: (LEAVE THIS EMPTY)

Skip: 0

 

#2

 

Base: _cpuid_set_info

Comment: FakeCPUID - Ext 0x5 (0x5E,0x3F SKL,Haswell), 

Count: 0

Enabled: YES

Find:   89C1C1E9 1089

Identifier: kernel

Limit: 0

Mask: (LEAVE THIS EMPTY)

MatchKernel: 

Replace: 

ReplaceMask: (LEAVE THIS EMPTY)

Skip: 0

  • Like 1
17 minutes ago, pitrysha said:

Thanks.

I don’t understand about this part

 

Base: _cpuid_set_info

Comment: FakeCPUID - Model 0xE (0x5E,0x3F SKL,Haswell), 

Count: 0

Enabled: YES

Find:   89C1C0E9 04FEC1

Identifier: kernel

Limit: 0

Mask: (LEAVE THIS EMPTY)

MatchKernel: 

Replace: 

ReplaceMask: (LEAVE THIS EMPTY)

Skip: 0

 

#2

 

Base: _cpuid_set_info

Comment: FakeCPUID - Ext 0x5 (0x5E,0x3F SKL,Haswell), 

Count: 0

Enabled: YES

Find:   89C1C1E9 1089

Identifier: kernel

Limit: 0

Mask: (LEAVE THIS EMPTY)

MatchKernel: 

Replace: 

ReplaceMask: (LEAVE THIS EMPTY)

Skip: 0

Base: _cpuid_set_info

Comment: FakeCPUID - Model 0xE (0x5E,0x3F SKL,Haswell), 

Count: 0

Enabled:No

Find:  LEAVE THIS EMPTY Still testing.

Identifier: kernel

Limit: 0

Mask: (LEAVE THIS EMPTY)

MatchKernel: 

Replace: 

ReplaceMask: (LEAVE THIS EMPTY)

Skip: 0

 

#2

 

Base: _cpuid_set_info

Comment: FakeCPUID - Ext 0x5 (0x5E,0x3F SKL,Haswell), 

Count: 0

Enabled:No

Find:   LEAVE THIS EMPTY Still testing.

Identifier: kernel

Limit: 0

Mask: (LEAVE THIS EMPTY)

MatchKernel: 

Replace: 

ReplaceMask: (LEAVE THIS EMPTY)

Skip: 0

Edited by nmano
  • Like 1

For the record, my Asus X99 Deluxe dies with a kernel panic early during boot

* Asus X99 Deluxe (original, i.e. not II)

* i7-5820k

* OpenCore 0.5.5

* Settings according to https://github.com/khronokernel/Opencore-Vanilla-Desktop-Guide/blob/master/config-HEDT/haswell-e.md

* Catalina installer and Hight Sierra both panic

The last non-panic log line is "VirtualSMC efend: @ no extra rtc memory is present for key erase", this seems normal. The panic backtrace is exceedingly uninformative...

IMG_20200223_130443.jpg

config.plist

		<key>Quirks</key>
		<dict>
			<key>AppleCpuPmCfgLock</key>
			<false/>
			<key>AppleXcpmCfgLock</key>
			<true/>
			<key>AppleXcpmExtraMsrs</key>
			<true/>
			<key>AppleXcpmForceBoost</key>
			<false/>
			<key>CustomSMBIOSGuid</key>
			<false/>
			<key>DisableIoMapper</key>
			<true/>
			<key>DummyPowerManagement</key>
			<false/>
			<key>ExternalDiskIcons</key>
			<false/>
			<key>IncreasePciBarSize</key>
			<true/>
			<key>LapicKernelPanic</key>
			<false/>
			<key>PanicNoKextDump</key>
			<false/>
			<key>PowerTimeoutKernelPanic</key>
			<false/>
			<key>ThirdPartyDrives</key>
			<false/>
			<key>XhciPortLimit</key>
			<false/>
		</dict>

These are the Kernel Quirks I used.

As I said, no manual kernel patches needed anymore.

 

If everything else is setup right, system will boot.

Thanks, I tried that but same panic. I also tried FakeSMC instead of VirtualSMC, with the same outcome (except that the last log line before the panic is now from FakeSMC). 

 

I also tried nmano's EFI from 3 comments up and I can boot with that. And boy, its chock-full of custom patches ;-) Not sure if its one of those or some OC 0.5.6 fix, will probably wait for the release and then try again.

Edited by vbraun
41 minutes ago, vbraun said:

Thanks, I tried that but same panic. I also tried FakeSMC instead of VirtualSMC, with the same outcome (except that the last log line before the panic is now from FakeSMC). 

 

I also tried nmano's EFI from 3 comments up and I can boot with that. And boy, its chock-full of custom patches ;-) Not sure if its one of those or some OC 0.5.6 fix, will probably wait for the release and then try again.

Add VoodooTSCSync.kext     

Add npci=0x2000

Edited by pitrysha
  • Like 1

I tried all combinations of npci=0x2000 and the "Above 4G decoding" bios setting to no avail.

 

I also tried the kexts from nmano's  EFI (which include VoodooTSCSync.kext) on my OC 0.5.5 config and that didn't change anything with the panic either.

Hello All.

 

Been following this thread for the last couple of days.. Learning a lot but still confused here and there.   

My System as follows.

 

Gigabyte X99P-SLI

Intel 6800K

32GB Crucial Ram

Gigabyte Windforce R9 390 8GB

 

Had success before  10.15.3, but it was buggy.  Maybe some has a system close to mine can work together to get a fully stable system up.

 

 

6 hours ago, jabbaranks said:

Hello All.

 

Been following this thread for the last couple of days.. Learning a lot but still confused here and there.   

My System as follows.

 

Gigabyte X99P-SLI

Intel 6800K

32GB Crucial Ram

Gigabyte Windforce R9 390 8GB

 

Had success before  10.15.3, but it was buggy.  Maybe some has a system close to mine can work together to get a fully stable system up.

 

 

 

Interesting i read the thread this is the same motherboard and CPU that i use, I haven’t upgraded form 10.13.6 because my system is stable with Clover. I’ll keep and eye and ear out, maybe we can work together but also I haven’t got much time at the moment .

Hi! Thanx for great topic! My build is ASUS X99 Deluxe + Intel 5960x + GT 740. I used attached EFI from the first post. Everything seems to boot properly, but in the end I stuck on gIOScreenLockState 3... I know this is GPU related, but can anyone tell me please where to dig? Thanks!

p.s. CSM disabled in BIOS.

 

IMG_0737.png

Edited by tarkh
  • Like 1
4 hours ago, tarkh said:

Hi! Thanx for great topic! My build is ASUS X99 Deluxe + Intel 5960x + GT 740. I used attached EFI from the first post. Everything seems to boot properly, but in the end I stuck on gIOScreenLockState 3... I know this is GPU related, but can anyone tell me please where to dig? Thanks!

p.s. CSM disabled in BIOS.

 

IMG_0737.png

I thing Black screen Error .

Try

#WhateverGreen.kext

#Identifier com.apple.driver.AppleGraphicsDevicePolicy

Find        626F6172 642D6964 

Replace     626F6172 642D6978 

# CSM Enabled

Edited by nmano
5 hours ago, nmano said:

I thing Black screen Error .

Try

#WhateverGreen.kext

#Identifier com.apple.driver.AppleGraphicsDevicePolicy

Find        626F6172 642D6964 

Replace     626F6172 642D6978 

# CSM Enabled

 

nmano, thank you for replying! I've tried to boot with AppleGraphicsDevicePolicy patch and CSM enabled (UEFI+Legacy and just UEFI) - same gIOScreenLockState 3 error. Btw I've 2 monitors, one connected through HDMI, second - DVI, so the only changes was after enabling CSM is that POST messages and OC started on DVI monitor instead of HDMI.

Edited by tarkh

Okay, I finally booted.

What I did to run ASUS x99 Deluxe5960x + GT 740:

 

1. Compiled latest release of OC

2. Used SampleFull.plist as template

3. Added VBoxHfs.efi and ApfsDriverLoader.efi to Drivers folder (kept all defaults)

3. Added custom SSDT patches to ACPI

4. Added proper Kexts to Kexts directory

 

in config.plist:

5. Booter>DeviceProperties>Add - removed entry with ig-platform

6. Kernel>Patch - inserted custom 5960x patch and xcpm_bootstrap patch, thats all. + Enabled AppleXcpmExtraMsrs and DisableIoMapper (for VT-d) in Kernel>Quirks

7. NVRAM>Add> boot-args - added npci=0x2000, nvda_drv set to 31

 

So what I think the reason of black screen was nvda_drv, which is not 31 in config, presented in first post of this topic... But I'm not sure.

 

Any way now I've got fully working system with proper CPU PM. On next week I'll polish all settings and will post my EFI here.

Thank you!

Hello fellow Asus X99 Deluxe user ;-) 

 

For the record I got the same black screen with WiFi errors (AppleKeyStore: operation failed ...) with nmano's EFI. But when booting an older High Sierra install I can at that point ssh into the box over WiFi, so its just the GPU not turning on. But the OS is basically working, and the console messages are apparently not that important. I'm in the process of switching from HS+nVidia to Catalina+Radeon but haven't gotten any signal out of the Radeon card yet. Looking forward to trying your EFI ;-)

  • Allan unpinned this topic
×
×
  • Create New...