Thunderbolt 3 Modified Firmware Custom SSDT Discussion

[EVILGENIUS]

@scottkendall Remove the marked SMC Resistor and sleep will work like it should...

 

[scottkendall]

Sleep works great as long as I jumper 3 and 5 if I plug in the THB C header cable then sleep goes bye bye... I am totally fine with the 3-5 pin jumper. Also I am not going to modify the card in someway that I can not undo.

 

[NorthAmTransAm]

Does anybody have an unmodified GC-AR card? Looking to do a test.

 

[PTuser]

Hi I'm running a titan ridge Rev 1 with nvm23 firmware in a non TB header OC Catalina build. Apart from a few boots to activate it, its ok except for one third party driver that is not faulty as it works perfectly in a real mac. It will load eventually but a part of the driver (which does three things) is not functioning. Heres the panic:

panic(cpu 0 caller 0xffffff8007c449ea): Kernel trap at 0xffffff7f8c75b78f, type 14=page fault, registers:
CR0: 0x000000008001003b, CR2: 0x0000000000000514, CR3: 0x00000000200b3000, CR4: 0x00000000000626e0
RAX: 0x0000000000000000, RBX: 0xffffff80a91c8000, RCX: 0xffffff7f89ba2201, RDX: 0x000000005f5e6580
RSP: 0xffffff8754b63d70, RBP: 0xffffff8754b63da0, RSI: 0xffffff80a8dba000, RDI: 0x0000000000000000
R8: 0x0000000000000000, R9: 0x0000000000000000, R10: 0x0000000000000001, R11: 0x0000000000000005
R12: 0x0000000000000514, R13: 0xffffff7f8c764290, R14: 0x0000000000000514, R15: 0xffffff80a91c8000
RFL: 0x0000000000010246, RIP: 0xffffff7f8c75b78f, CS: 0x0000000000000008, SS: 0x0000000000000010
Fault CR2: 0x0000000000000514, Error code: 0x0000000000000000, Fault CPU: 0x0, PL: 1, VF: 5

Backtrace (CPU 0), Frame : Return Address
0xffffff8754b637d0 : 0xffffff8007b1868d mach_kernel : _handle_debugger_trap + 0x49d
0xffffff8754b63820 : 0xffffff8007c52ab5 mach_kernel : _kdp_i386_trap + 0x155
0xffffff8754b63860 : 0xffffff8007c4463e mach_kernel : _kernel_trap + 0x4ee
0xffffff8754b638b0 : 0xffffff8007abea40 mach_kernel : _return_from_trap + 0xe0
0xffffff8754b638d0 : 0xffffff8007b17d57 mach_kernel : _DebuggerTrapWithState + 0x17
0xffffff8754b639d0 : 0xffffff8007b18147 mach_kernel : _panic_trap_to_debugger + 0x227
0xffffff8754b63a20 : 0xffffff80082bf2bc mach_kernel : _panic + 0x54
0xffffff8754b63a90 : 0xffffff8007c449ea mach_kernel : _sync_iss_to_iks + 0x2aa
0xffffff8754b63c10 : 0xffffff8007c446e8 mach_kernel : _kernel_trap + 0x598
0xffffff8754b63c60 : 0xffffff8007abea40 mach_kernel : _return_from_trap + 0xe0
0xffffff8754b63c80 : 0xffffff7f8c75b78f com.focusrite.driver.FocusritePCIe : __ZN33com_focusrite_FocusritePCIeDevice14readRegister32Ej + 0x9b
0xffffff8754b63da0 : 0xffffff7f8c75a312 com.focusrite.driver.FocusritePCIe : __ZN33com_focusrite_FocusritePCIeDevice5startEP9IOService + 0x494
0xffffff8754b63de0 : 0xffffff8008204dc6 mach_kernel : __ZN9IOService14startCandidateEPS_ + 0xf6
0xffffff8754b63e40 : 0xffffff8008204901 mach_kernel : __ZN9IOService15probeCandidatesEP12OSOrderedSet + 0xad1
0xffffff8754b63ef0 : 0xffffff8008203c9e mach_kernel : __ZN9IOService14doServiceMatchEj + 0x2de
0xffffff8754b63f50 : 0xffffff8008206456 mach_kernel : __ZN15_IOConfigThread4mainEPvi + 0x186
0xffffff8754b63fa0 : 0xffffff8007abe13e mach_kernel : _call_continuation + 0x2e
Kernel Extensions in backtrace:
com.focusrite.driver.FocusritePCIe(4.6.2)[5AACDD8E-0451-347E-A672-BBD1EF7A84F8]@0xffffff7f8c758000->0xffffff7f8c775fff
dependency: com.apple.iokit.IOPCIFamily(2.9)[2F37AE58-E6B9-3B18-9092-3B80D34C334B]@0xffffff7f888b4000

so it looks like its to do with IOkit.IOPCIFamily. I looked in the info.plist of the kext but if anyone can understand this and give me a pointer on what this is driver is looking for that it cant find that would be great!

 

[NorthAmTransAm]

Hi I'm running a titan ridge Rev 1 with nvm23 firmware in a non TB header OC Catalina build. Apart from a few boots to activate it, its ok except for one third party driver that is not faulty as it works perfectly in a real mac. It will load eventually but a part of the driver (which does three things) is not functioning. Heres the panic:

panic(cpu 0 caller 0xffffff8007c449ea): Kernel trap at 0xffffff7f8c75b78f, type 14=page fault, registers:
CR0: 0x000000008001003b, CR2: 0x0000000000000514, CR3: 0x00000000200b3000, CR4: 0x00000000000626e0
RAX: 0x0000000000000000, RBX: 0xffffff80a91c8000, RCX: 0xffffff7f89ba2201, RDX: 0x000000005f5e6580
RSP: 0xffffff8754b63d70, RBP: 0xffffff8754b63da0, RSI: 0xffffff80a8dba000, RDI: 0x0000000000000000
R8: 0x0000000000000000, R9: 0x0000000000000000, R10: 0x0000000000000001, R11: 0x0000000000000005
R12: 0x0000000000000514, R13: 0xffffff7f8c764290, R14: 0x0000000000000514, R15: 0xffffff80a91c8000
RFL: 0x0000000000010246, RIP: 0xffffff7f8c75b78f, CS: 0x0000000000000008, SS: 0x0000000000000010
Fault CR2: 0x0000000000000514, Error code: 0x0000000000000000, Fault CPU: 0x0, PL: 1, VF: 5

Backtrace (CPU 0), Frame : Return Address
0xffffff8754b637d0 : 0xffffff8007b1868d mach_kernel : _handle_debugger_trap + 0x49d
0xffffff8754b63820 : 0xffffff8007c52ab5 mach_kernel : _kdp_i386_trap + 0x155
0xffffff8754b63860 : 0xffffff8007c4463e mach_kernel : _kernel_trap + 0x4ee
0xffffff8754b638b0 : 0xffffff8007abea40 mach_kernel : _return_from_trap + 0xe0
0xffffff8754b638d0 : 0xffffff8007b17d57 mach_kernel : _DebuggerTrapWithState + 0x17
0xffffff8754b639d0 : 0xffffff8007b18147 mach_kernel : _panic_trap_to_debugger + 0x227
0xffffff8754b63a20 : 0xffffff80082bf2bc mach_kernel : _panic + 0x54
0xffffff8754b63a90 : 0xffffff8007c449ea mach_kernel : _sync_iss_to_iks + 0x2aa
0xffffff8754b63c10 : 0xffffff8007c446e8 mach_kernel : _kernel_trap + 0x598
0xffffff8754b63c60 : 0xffffff8007abea40 mach_kernel : _return_from_trap + 0xe0
0xffffff8754b63c80 : 0xffffff7f8c75b78f com.focusrite.driver.FocusritePCIe : __ZN33com_focusrite_FocusritePCIeDevice14readRegister32Ej + 0x9b
0xffffff8754b63da0 : 0xffffff7f8c75a312 com.focusrite.driver.FocusritePCIe : __ZN33com_focusrite_FocusritePCIeDevice5startEP9IOService + 0x494
0xffffff8754b63de0 : 0xffffff8008204dc6 mach_kernel : __ZN9IOService14startCandidateEPS_ + 0xf6
0xffffff8754b63e40 : 0xffffff8008204901 mach_kernel : __ZN9IOService15probeCandidatesEP12OSOrderedSet + 0xad1
0xffffff8754b63ef0 : 0xffffff8008203c9e mach_kernel : __ZN9IOService14doServiceMatchEj + 0x2de
0xffffff8754b63f50 : 0xffffff8008206456 mach_kernel : __ZN15_IOConfigThread4mainEPvi + 0x186
0xffffff8754b63fa0 : 0xffffff8007abe13e mach_kernel : _call_continuation + 0x2e
Kernel Extensions in backtrace:
com.focusrite.driver.FocusritePCIe(4.6.2)[5AACDD8E-0451-347E-A672-BBD1EF7A84F8]@0xffffff7f8c758000->0xffffff7f8c775fff
dependency: com.apple.iokit.IOPCIFamily(2.9)[2F37AE58-E6B9-3B18-9092-3B80D34C334B]@0xffffff7f888b4000

so it looks like its to do with IOkit.IOPCIFamily. I looked in the info.plist of the kext but if anyone can understand this and give me a pointer on what this is driver is looking for that it cant find that would be great!

Looks like its more to do with your Focusrite interface. Check for an update from them first?

 

[PTuser]

Looks like its more to do with your Focusrite interface. Check for an update from them first?

its the latest driver which as ive explained works fine an a real mac so ive established my build is causing this...what Im asking for is an understanding of the detail in the log

 

[NorthAmTransAm]

its the latest driver which as ive explained works fine an a real mac so ive established my build is causing this...what Im asking for is an understanding of the detail in the log

Right, I see that.

Question, did a previous version of this driver, on your hack, work fine?

You're in a bit of uncharted territory with the flashed card on an non-TB board as I'm sure you're aware so it will be about finding the right combination for things.

What SSDT are you using for your for TB setup?

As for understanding what's in the log it looks like com.focusrite.driver.FocusritePCIee crashed/panicked and it was tied to com.apple.iokit.IOPCIFamily. So the source issue is the first one.

 

[PTuser]

Hi @NorthAmTransAm
The previous driver was the same. I have updated SMBIOS since which has improved general tb performance. Im left with just needing a few boots to get it going, then in order to avoid the kernal panic, I hot plug the device in question, so generally its only this one driver Im having issues with, which is not bad considering. Its not working with core audio. Its other two functions are working so its not completely dead. In IO reg its seen as loaded. I have a number of thunderbolt devices. This is the only one thats giving me any trouble.

I was hopeful that all the backtrace info could be deciphered as to what its not seeing....so a fix or a kext edit can be attempted

SSDT enclosed....

 

[scottkendall]

Hi @NorthAmTransAm
The previous driver was the same. I have updated SMBIOS since which has improved general tb performance. Im left with just needing a few boots to get it going, then in order to avoid the kernal panic, I hot plug the device in question, so generally its only this one driver Im having issues with, which is not bad considering. Its not working with core audio. Its other two functions are working so its not completely dead. In IO reg its seen as loaded. I have a number of thunderbolt devices. This is the only one thats giving me any trouble.

I was hopeful that all the backtrace info could be deciphered as to what its not seeing....so a fix or a kext edit can be attempted

SSDT enclosed....

Thunderbolt on a hack is not the same as thunderbolt on a real Mac so saying that the driver is fine on a real Mac does not make the driver not the problem. Second problem is you do not have a board that is supporting thunderbolt, I have a card working in a board that does not support thunderbolt also, but it does not work even half as well as the system that supports thunderbolt it also does some strange things to the video sometimes.

Why do you think the problem is the thunderbolt when your error says FocusritePCIe this would lead me to believe that it is a problem with a Focusrite PCI card not the Thunderbolt card. What Focusrite device do you have?

 

[PTuser]

@scottkendall
"Why do you think the problem is the thunderbolt when your error says FocusritePCIe this would lead me to believe that it is a problem with a Focusrite PCI card not the Thunderbolt card. What Focusrite device do you have?"

although the driver name is misdescribed, it is actually the thunderbolt driver. Its a redline16. I get what you say about compromises given no header. Im doing ok with it given this, as I'm getting full power on both ports. Im certain my system is to blame in regard to this driver. Im just not knowledgeable enough to troubleshoot it.

I can get past the panic with hot plugging, however its interaction with core audio is still not happening, so Im only able to control the units functions via software.