Jump to content

Clover General discussion

ErmaC

By ErmaC,
in Clover

 Share
30960 posts in this topic

Recommended Posts

vector sigma
vector sigma
Posted
Posted
28 minutes ago, Sherlocks said:

great. thank you so much for support new smc3 and update both clover and fakesmc.:thumbsup_anim:

What about: 

// SetupDataForOSX
/// Sets the DataHub data used by OS X
VOID EFIAPI
SetupDataForOSX(BOOLEAN Hibernate)
{
  EFI_STATUS Status;

  UINT32     DevPathSupportedVal;
  UINT64     FrontSideBus;
  UINT64     CpuSpeed;
  UINT64     TscFrequency;
  UINT64     ARTFrequency;
  CHAR16*    ProductName;
  CHAR16*    SerialNumber;
  UINTN      Revision;
  UINT16     Zero = 0;
  BOOLEAN    isRevLess = (gSettings.REV[0] == 0 &&
                          gSettings.REV[1] == 0 &&
                          gSettings.REV[2] == 0 &&
                          gSettings.REV[3] == 0 &&
                          gSettings.REV[4] == 0 &&
                          gSettings.REV[5] == 0);
  
  if (AsciiStrStr(gSettings.BootArgs, "-withREV") != NULL) {
    isRevLess = false;
  } else if (AsciiStrStr(gSettings.BootArgs, "-noREV") != NULL) {
    isRevLess = true;
  }

and use the same arguments used by FakeSMC??

Sherlocks
Sherlocks
Posted
Posted (edited)
14 minutes ago, vector sigma said:

What about: 


// SetupDataForOSX
/// Sets the DataHub data used by OS X
VOID EFIAPI
SetupDataForOSX(BOOLEAN Hibernate)
{
  EFI_STATUS Status;

  UINT32     DevPathSupportedVal;
  UINT64     FrontSideBus;
  UINT64     CpuSpeed;
  UINT64     TscFrequency;
  UINT64     ARTFrequency;
  CHAR16*    ProductName;
  CHAR16*    SerialNumber;
  UINTN      Revision;
  UINT16     Zero = 0;
  BOOLEAN    isRevLess = (gSettings.REV[0] == 0 &&
                          gSettings.REV[1] == 0 &&
                          gSettings.REV[2] == 0 &&
                          gSettings.REV[3] == 0 &&
                          gSettings.REV[4] == 0 &&
                          gSettings.REV[5] == 0);
  
  if (AsciiStrStr(gSettings.BootArgs, "-withREV") != NULL) {
    isRevLess = false;
  } else if (AsciiStrStr(gSettings.BootArgs, "-noREV") != NULL) {
    isRevLess = true;
  }

and use the same arguments used by FakeSMC??

 

good boot-arg. can we consider that no need update fakesmc that is not relate new model without boot-arg?

from check clover's any values, does fakesmc have determination to add isRevLess as free?

the recent models of apple has t2 and smc3 except imac19.x.
we just think imac19.x have t2 and smc3. but it's not correct.
in future, there are possibility that not include smc3 version in macbook11,1 or macbookpro16,1 or etc.

therefore, if we add smbios data in clover's platformdata of the future model, can we consider freely more from smc3 according to clover in fakesmc?

ofc, default value has all(rev/pbr/epci)

it's just my think. sorry for my bad english.

Edited by Sherlocks
  • Like 1
vector sigma
vector sigma
Posted
Posted
1 minute ago, Sherlocks said:

good boot-arg. can we consider that no need update fakesmc that is not relate new model without boot-arg?

from check clover's any values, does fakesmc have determination to add isRevLess as free?

the recent models of apple has t2 and smc3 except imac19.x.
we just think imac19.x have t2 and smc3. but it's not correct.
in future, there are possibility that not include smc3 version in macbook11,1 or macbookpro16,1 or etc.

therefore, if we add smbios data in clover's platformdata of the future model, can we consider freely more from smc3 according to clover in fakesmc?

it's just my think. sorry for my bad english.

FakeSMC is not used only by Clover, so it has to know which models needs REV or not. In fact REV (and a lot of keys) can be injected by it's Info.plist. 

Only users that will have older Clover or FakeSMC (the actual revisions, from now on) can decide to override this behaviour with boot-args, with or without having an updated platformdata for newer models.

  • Like 1
Sherlocks
Sherlocks
Posted
Posted
3 minutes ago, vector sigma said:

FakeSMC is not used only by Clover, so it has to know which models needs REV or not. In fact REV (and a lot of keys) can be injected by it's Info.plist. 

Only users that will have older Clover or FakeSMC (the actual revisions, from now on) can decide to override this behaviour with boot-args, with or without having an updated platformdata for newer models.

 

FakeSMC is not used only by Clover

- i forgot it. so seems it's best now.

  • Like 1
vector sigma
vector sigma
Posted
Posted (edited)

Anyway why Apple done that? I've received notification from the System when the revision was bad.... something tells me that this can have some evolution.. who knows..

or was something else?:unsure:

Edited by vector sigma
  • Like 1
Badruzeus
Badruzeus
Posted
Posted

Thay's bit weird that since r4956 I got random AppleSMC kernel panic when mach halt (reboot / shutdown). Not sure why mine is affected bcoz I'm sure my SMBios (MBP10,1) is not on those updated platformdata.c list. A similar issue with 10.3.6 upto 10.15 Beta using FakeSMC (Rehabman), not trying VirtualSMC yet. (Sorry, can not upload full crash logs. I'm currently on mobile).

Sherlocks
Sherlocks
Posted
Posted
1 hour ago, Badruzeus said:

Thay's bit weird that since r4956 I got random AppleSMC kernel panic when mach halt (reboot / shutdown). Not sure why mine is affected bcoz I'm sure my SMBios (MBP10,1) is not on those updated platformdata.c list. A similar issue with 10.3.6 upto 10.15 Beta using FakeSMC (Rehabman), not trying VirtualSMC yet. (Sorry, can not upload full crash logs. I'm currently on mobile).

 

strange. did you try to cleanup nvram?

  • Thanks 1
Slice
Slice
Posted
Posted
2 hours ago, Badruzeus said:

Thay's bit weird that since r4956 I got random AppleSMC kernel panic when mach halt (reboot / shutdown). Not sure why mine is affected bcoz I'm sure my SMBios (MBP10,1) is not on those updated platformdata.c list. A similar issue with 10.3.6 upto 10.15 Beta using FakeSMC (Rehabman), not trying VirtualSMC yet. (Sorry, can not upload full crash logs. I'm currently on mobile).

Is it happen with FakeSMC-6 and never with FakeSMC-3?

  • Like 1
PMheart
PMheart
Posted
Posted

Hi,

 

I propose this new patch for kext injection of 10.15: 

Index: rEFIt_UEFI/Platform/kext_inject.c
===================================================================
--- rEFIt_UEFI/Platform/kext_inject.c	(revision 4959)
+++ rEFIt_UEFI/Platform/kext_inject.c	(working copy)
@@ -681,8 +681,8 @@
 // Catalina (SIP)
 // PMheart: checked
 // Need to pair with KBEMojaCata*EXT
-UINT8   KBECataSearchSIP[]            = { 0x00, 0x85, 0xC0, 0x0F, 0x84, 0x87, 0x00, 0x00, 0x00, 0x49 };
-UINT8   KBECataReplaceSIP[]           = { 0x00, 0x85, 0xC0, 0x90, 0x90, 0x90, 0x90, 0x90, 0x90, 0x49 };
+UINT8   KBECataSearchSIP[]            = { 0xE8, 0xB0, 0xEE, 0x0C, 0x00, 0x85, 0xC0, 0x0F, 0x84 };
+UINT8   KBECataReplaceSIP[]           = { 0xE8, 0xB0, 0xEE, 0x0C, 0x00, 0x31, 0xC0, 0x0F, 0x84 };
 
 // Mojave 10.14 - 10.14.3
 // Avoid race condition in OSKext::removeKextBootstrap when using booter kexts without keepsyms=1.

Don't ask me if it works, of course it does - just read the asm code. ;)

 

PMheart

  • Like 2
Badruzeus
Badruzeus
Posted
Posted
1 hour ago, Slice said:

Is it happen with FakeSMC-6 and never with FakeSMC-3?

I think bcoz of Nvram, no longer having issue after Clover's F11 key for full clean, thanks @Sherlocks for suggestion. Oh ya, could be affected by VirtualSMC either.. I remember previously boot with it using OpenCore, then switch to Clover (FakeSMC). I will retry with your FakeSMC v3.5.1.

Sherlocks
Sherlocks
Posted
Posted (edited)
8 minutes ago, Badruzeus said:

I think bcoz of Nvram, no longer having issue after Clover's F11 key for full clean, thanks @Sherlocks for suggestion. Oh ya, could be affected by VirtualSMC either.. I remember previously boot with it using OpenCore, then switch to Clover (FakeSMC). I will retry with your FakeSMC v3.5.1.

 

probably. recent commit for smc3 is no problem for rehabman and slice fakesmc version.

anyways f11 can clean opencore nvram keys.

 

now i keep rehabman fakesmc. his kext suppport all macos(10.6~10.15). its all in one. but his kext need to update for smc3 like vector's update of hwsensor3.

Edited by Sherlocks
  • Like 2
Badruzeus
Badruzeus
Posted
Posted (edited)

When building Clover r4961 with `./ebuild.sh --ext-co -fr` I got this errors (seem related to OcSupportPkg): 

Building ... /Users/badruzeus/src/EXT_PACKAGES/OcSupportPkg/Library/OcStringLib/OcStringLib.inf [X64]
[CC] OcAsciiLib
[CC] BitOverflow
[CC] DataPatcher
[CC] FileProtocol
/Users/badruzeus/src/EXT_PACKAGES/OcSupportPkg/Library/OcStringLib/OcUnicodeLib.c:67:23: error: implicit declaration of function 'CharToUpper' is invalid in C99 [-Werror,-Wimplicit-function-declaration]
  UpperFirstString  = CharToUpper (*FirstString);
                      ^
/Users/badruzeus/src/EXT_PACKAGES/OcSupportPkg/Library/OcStringLib/OcUnicodeLib.c:143:23: error: implicit declaration of function 'CharToUpper' is invalid in C99 [-Werror,-Wimplicit-function-declaration]
  UpperFirstString  = CharToUpper (*FirstString);
                      ^
2 errors generated.
make: *** [/Users/badruzeus/src/UDK2018/Build/AptioFixPkg/RELEASE_XCODE8/X64/OcSupportPkg/Library/OcStringLib/OcStringLib/OUTPUT/OcUnicodeLib.obj] Error 1


build.py...
 : error 7000: Failed to execute command
	make tbuild [/Users/badruzeus/src/UDK2018/Build/AptioFixPkg/RELEASE_XCODE8/X64/OcSupportPkg/Library/OcStringLib/OcStringLib]


build.py...
 : error F002: Failed to build module
	/Users/badruzeus/src/EXT_PACKAGES/OcSupportPkg/Library/OcStringLib/OcStringLib.inf [X64, XCODE8, RELEASE]

- Failed -
Build end time: 09:07:33, Jun.10 2019
Build total time: 00:00:10

MacBookPro:Clover badruzeus$

Thanks.

Edited by Badruzeus
Slice
Slice
Posted
Posted
1 hour ago, Badruzeus said:

When building Clover r4961 with `./ebuild.sh --ext-co -fr` I got this errors (seem related to OcSupportPkg): 


Building ... /Users/badruzeus/src/EXT_PACKAGES/OcSupportPkg/Library/OcStringLib/OcStringLib.inf [X64]
[CC] OcAsciiLib
[CC] BitOverflow
[CC] DataPatcher
[CC] FileProtocol
/Users/badruzeus/src/EXT_PACKAGES/OcSupportPkg/Library/OcStringLib/OcUnicodeLib.c:67:23: error: implicit declaration of function 'CharToUpper' is invalid in C99 [-Werror,-Wimplicit-function-declaration]
  UpperFirstString  = CharToUpper (*FirstString);
                      ^
/Users/badruzeus/src/EXT_PACKAGES/OcSupportPkg/Library/OcStringLib/OcUnicodeLib.c:143:23: error: implicit declaration of function 'CharToUpper' is invalid in C99 [-Werror,-Wimplicit-function-declaration]
  UpperFirstString  = CharToUpper (*FirstString);
                      ^
2 errors generated.
make: *** [/Users/badruzeus/src/UDK2018/Build/AptioFixPkg/RELEASE_XCODE8/X64/OcSupportPkg/Library/OcStringLib/OcStringLib/OUTPUT/OcUnicodeLib.obj] Error 1


build.py...
 : error 7000: Failed to execute command
	make tbuild [/Users/badruzeus/src/UDK2018/Build/AptioFixPkg/RELEASE_XCODE8/X64/OcSupportPkg/Library/OcStringLib/OcStringLib]


build.py...
 : error F002: Failed to build module
	/Users/badruzeus/src/EXT_PACKAGES/OcSupportPkg/Library/OcStringLib/OcStringLib.inf [X64, XCODE8, RELEASE]

- Failed -
Build end time: 09:07:33, Jun.10 2019
Build total time: 00:00:10

MacBookPro:Clover badruzeus$

Thanks.

It should be reported to Acidantera.

  • Thanks 1
Slice
Slice
Posted
Posted
22 hours ago, PMheart said:

Hi,

 

I propose this new patch for kext injection of 10.15: 


Index: rEFIt_UEFI/Platform/kext_inject.c
===================================================================
--- rEFIt_UEFI/Platform/kext_inject.c	(revision 4959)
+++ rEFIt_UEFI/Platform/kext_inject.c	(working copy)
@@ -681,8 +681,8 @@
 // Catalina (SIP)
 // PMheart: checked
 // Need to pair with KBEMojaCata*EXT
-UINT8   KBECataSearchSIP[]            = { 0x00, 0x85, 0xC0, 0x0F, 0x84, 0x87, 0x00, 0x00, 0x00, 0x49 };
-UINT8   KBECataReplaceSIP[]           = { 0x00, 0x85, 0xC0, 0x90, 0x90, 0x90, 0x90, 0x90, 0x90, 0x49 };
+UINT8   KBECataSearchSIP[]            = { 0xE8, 0xB0, 0xEE, 0x0C, 0x00, 0x85, 0xC0, 0x0F, 0x84 };
+UINT8   KBECataReplaceSIP[]           = { 0xE8, 0xB0, 0xEE, 0x0C, 0x00, 0x31, 0xC0, 0x0F, 0x84 };
 
 // Mojave 10.14 - 10.14.3
 // Avoid race condition in OSKext::removeKextBootstrap when using booter kexts without keepsyms=1.

Don't ask me if it works, of course it does - just read the asm code. ;)

 

PMheart

Sorry for commiting too late.

  • Like 2
telepati
telepati
Posted
Posted (edited)
7 hours ago, SavageAUS said:

FakeSMC jump from 3 to 6? Where is new version please?

 

Here it is;

 

cloverefiboot-code-r4961-CloverPackage-CloverV2-ThirdParty-kexts-Other.zip

 

I download a snapshot from here but I am not sure did I right.

 

https://sourceforge.net/p/cloverefiboot/code/4961/tree//CloverPackage/CloverV2/ThirdParty/kexts/Other/

Edited by telepati
arsradu
arsradu
Posted
Posted (edited)

Looks like ext_packages have yet another issue when building Clover.

 

Just letting you guys know in case you come across the same thing.

 

OCSupportPkg should be for OpenCore only. Not sure why is it trying to compile it for Clover... But anyway. Maybe there's no condition put in place to detect what you're trying to compile and only add that support pkg where it's needed.

 

Anyway, I just thought you guys might wanna know about this. 

[SLINK] BaseIoLibIntrinsic
Building ... /Users/jimmy/src/EXT_PACKAGES/OcSupportPkg/Library/OcDataHubLib/OcDataHubLib.inf [X64]
/Users/jimmy/src/EXT_PACKAGES/OcSupportPkg/Library/OcStringLib/OcUnicodeLib.c:67:23: error: implicit declaration of function 'CharToUpper' is invalid in C99 [-Werror,-Wimplicit-function-declaration]
  UpperFirstString  = CharToUpper (*FirstString);
                      ^
/Users/jimmy/src/EXT_PACKAGES/OcSupportPkg/Library/OcStringLib/OcUnicodeLib.c:143:23: error: implicit declaration of function 'CharToUpper' is invalid in C99 [-Werror,-Wimplicit-function-declaration]
  UpperFirstString  = CharToUpper (*FirstString);
                      ^
2 errors generated.
make: *** [/Users/jimmy/src/UDK2018/Build/AptioFixPkg/RELEASE_XCODE8/X64/OcSupportPkg/Library/OcStringLib/OcStringLib/OUTPUT/OcUnicodeLib.obj] Error 1


build.py...
 : error 7000: Failed to execute command
	make tbuild [/Users/jimmy/src/UDK2018/Build/AptioFixPkg/RELEASE_XCODE8/X64/OcSupportPkg/Library/OcStringLib/OcStringLib]


build.py...
 : error F002: Failed to build module
	/Users/jimmy/src/EXT_PACKAGES/OcSupportPkg/Library/OcStringLib/OcStringLib.inf [X64, XCODE8, RELEASE]

- Failed -
Build end time: 14:35:16, Jun.10 2019
Build total time: 00:00:04

 

Edited by arsradu
Badruzeus
Badruzeus
Posted
Posted
8 minutes ago, arsradu said:

Looks like ext_packages have yet another issue when building Clover.

 

Just letting you guys know in case you come across the same thing.

 

OCSupportPkg should be for OpenCore only. Not sure why is it trying to compile it for Clover... But anyway. Maybe there's no condition put in place to detect what you're trying to compile and only add that support pkg where it's needed.

 

Anyway, I just thought you guys might wanna know about this. 


[SLINK] BaseIoLibIntrinsic
Building ... /Users/jimmy/src/EXT_PACKAGES/OcSupportPkg/Library/OcDataHubLib/OcDataHubLib.inf [X64]
/Users/jimmy/src/EXT_PACKAGES/OcSupportPkg/Library/OcStringLib/OcUnicodeLib.c:67:23: error: implicit declaration of function 'CharToUpper' is invalid in C99 [-Werror,-Wimplicit-function-declaration]
  UpperFirstString  = CharToUpper (*FirstString);
                      ^
/Users/jimmy/src/EXT_PACKAGES/OcSupportPkg/Library/OcStringLib/OcUnicodeLib.c:143:23: error: implicit declaration of function 'CharToUpper' is invalid in C99 [-Werror,-Wimplicit-function-declaration]
  UpperFirstString  = CharToUpper (*FirstString);
                      ^
2 errors generated.
make: *** [/Users/jimmy/src/UDK2018/Build/AptioFixPkg/RELEASE_XCODE8/X64/OcSupportPkg/Library/OcStringLib/OcStringLib/OUTPUT/OcUnicodeLib.obj] Error 1


build.py...
 : error 7000: Failed to execute command
	make tbuild [/Users/jimmy/src/UDK2018/Build/AptioFixPkg/RELEASE_XCODE8/X64/OcSupportPkg/Library/OcStringLib/OcStringLib]


build.py...
 : error F002: Failed to build module
	/Users/jimmy/src/EXT_PACKAGES/OcSupportPkg/Library/OcStringLib/OcStringLib.inf [X64, XCODE8, RELEASE]

- Failed -
Build end time: 14:35:16, Jun.10 2019
Build total time: 00:00:04

 

vit9696 told me on OpenCore thread it needs AUDK (Acidenthera) instead of UDK.

 Share
Go to topic listing
×
×
  • Create New...