Got a tip for us?
Let us know
Become a MacRumors Supporter for $50/year with no ads, ability to filter front page stories, and private forums.
MP 1,1-5,1 testing TB3 AIC with MP 5,1
- Thread starter Flint Ironstag
- Start date
-
- Tags
- cmp egpu gigabyte mojave 10.14 tb3
- Sort by reaction score
Do you usually have both ports working? Also, can you verify that the USB controller is present under PCI in System Information?
Just as an FYI, that may be taken with a grain of salt here as well ...
I've been using the GC-TR 1.0 with the DSM v23 firmware flashed to it, using SSDT entry and using a Dell TB19 dock with it for a while (a month or so), just needed the reboot and it was working great.
Last week I had an Akitio Node arrive to play with the idea of doing some eGPU usage, I did a few reboots with the node and dock connected and started to get some progress on having it work with the eGPU (the card showed up in PCI and if i powered off the node i would get the message about "GPU not disconnected properly", but the eGPU menulet would not come up and manually adding it showed no GPUs)
Anyways I've been running this in a dual X5680 hack (a non-EFI board even), I know this isn't a hack site, however I'm posting here since it is also a Westmere/Nehalem CPU chipset (X58) like the 4,1 / 5,1 is, so if you get some panics it could be coming from this card. Interesting to note that during reboot i would actually get display out via the eGPU (RX 580) instead of the internal PCIe GTX650) but then attempting to boot macOS Big Sur 11.6.1 it would panic / hang, I was going to start down the path of swapping in an internal RX 580 and having an external RX 580 however the constant machine panics that started one I began attempting to use it as eGPU prevented me from doing that.
Eventually I was able to get some info and it was NMI timeout panics. so with that in mind if anyone was interested in it i'll drop it into the spoiler below, only really The panics continued regardless of the akitio node being connected once the issue began.
With the card removed I've had no further panics, I also tried this card in another hack, a T5810 (a haswell-e xeon X99 chipset) and the panics also occurred in that machine. I don't understand why the card was fine for a month then once eGPU was tried it has seemingly failed.
Caught one of the NMI timeout panics, only seen when I started to use the TB3 card to attempt to drive an eGPU.
But now that I've tried it would seem that just having the card connected now causes panics, the node works fine with a windows laptop so my mind says there must be some kind of fault in the TB3 card.
I did have the resistor removed and the jumper installed to make the card work, I also had the USB2 header ran into a USB header on the board so it was doing proper USB 2 and 3 for the TB dock I had connected.
in the PCI section:
It showed up in Graphics/Displays too but still wasn't usable
The NMI timeout as reported under the X99 chipset was a bit different, however they both seemed to have the same root cause.
I've been using the GC-TR 1.0 with the DSM v23 firmware flashed to it, using SSDT entry and using a Dell TB19 dock with it for a while (a month or so), just needed the reboot and it was working great.
Last week I had an Akitio Node arrive to play with the idea of doing some eGPU usage, I did a few reboots with the node and dock connected and started to get some progress on having it work with the eGPU (the card showed up in PCI and if i powered off the node i would get the message about "GPU not disconnected properly", but the eGPU menulet would not come up and manually adding it showed no GPUs)
Anyways I've been running this in a dual X5680 hack (a non-EFI board even), I know this isn't a hack site, however I'm posting here since it is also a Westmere/Nehalem CPU chipset (X58) like the 4,1 / 5,1 is, so if you get some panics it could be coming from this card. Interesting to note that during reboot i would actually get display out via the eGPU (RX 580) instead of the internal PCIe GTX650) but then attempting to boot macOS Big Sur 11.6.1 it would panic / hang, I was going to start down the path of swapping in an internal RX 580 and having an external RX 580 however the constant machine panics that started one I began attempting to use it as eGPU prevented me from doing that.
Eventually I was able to get some info and it was NMI timeout panics. so with that in mind if anyone was interested in it i'll drop it into the spoiler below, only really The panics continued regardless of the akitio node being connected once the issue began.
With the card removed I've had no further panics, I also tried this card in another hack, a T5810 (a haswell-e xeon X99 chipset) and the panics also occurred in that machine. I don't understand why the card was fine for a month then once eGPU was tried it has seemingly failed.
Caught one of the NMI timeout panics, only seen when I started to use the TB3 card to attempt to drive an eGPU.
But now that I've tried it would seem that just having the card connected now causes panics, the node works fine with a windows laptop so my mind says there must be some kind of fault in the TB3 card.
I did have the resistor removed and the jumper installed to make the card work, I also had the USB2 header ran into a USB header on the board so it was doing proper USB 2 and 3 for the TB dock I had connected.
Code:
AKiTiO Node:
Vendor Name: inXtron
Device Name: AKiTiO Node
Mode: Thunderbolt 3
Device ID: 0x308
Vendor ID: 0x41
Device Revision: 0x1
UID: 0x0041537E14E1D500
Route String: 3
Firmware Version: 23.1
Port (Upstream):
Status: Device connected
Link Status: 0x2
Speed: Up to 40 Gb/s x1
Current Link Width: 0x2
Link Controller Firmware Version: 0.34.0in the PCI section:
Code:
display:
Type: VGA-Compatible Controller
Driver Installed: Yes
Tunnel Compatible: Yes
Pause Compatible: No
MSI: No
Bus: PCI
Slot: Thunderbolt@26,0,0
Vendor ID: 0x1002
Device ID: 0x67df
Subsystem Vendor ID: 0x148c
Subsystem ID: 0x2377
Revision ID: 0x00e7
Link Width: x4
Link Speed: 8.0 GT/s
pci1002,aaf0:
Type: Audio Device
Driver Installed: Yes
Tunnel Compatible: Yes
Pause Compatible: No
MSI: Yes
Bus: PCI
Slot: Thunderbolt@26,0,1
Vendor ID: 0x1002
Device ID: 0xaaf0
Subsystem Vendor ID: 0x148c
Subsystem ID: 0xaaf0
Revision ID: 0x0000
Link Width: x4
Link Speed: 8.0 GT/sIt showed up in Graphics/Displays too but still wasn't usable
Code:
Display:
Type: External GPU
Bus: PCIe
PCIe Lane Width: x4
Vendor: AMD (0x1002)
Device ID: 0x67df
Revision ID: 0x00e7The NMI timeout as reported under the X99 chipset was a bit different, however they both seemed to have the same root cause.
Code:
Machine-check capabilities: 0x0000000007000c16
family: 6 model: 60 stepping: 3 microcode: 61
signature: 0x306c3
Intel(R) Xeon(R) CPU E5-1650 v3 @ 3.50GHz
22 error-reporting banks
Processor 0: IA32_MCG_STATUS: 0x0000000000000005
IA32_MC4_STATUS(0x411): 0xbe00000000800400
IA32_MC4_ADDR(0x412): 0x<ptr>
IA32_MC4_MISC(0x413): 0x<ptr>
IA32_MC17_STATUS(0x445): 0xfe200000000c110a
IA32_MC17_ADDR(0x446): 0x00000000d1000000
IA32_MC17_MISC(0x447): 0x04fc381602402086
Processor 1: IA32_MCG_STATUS: 0x0000000000000005
IA32_MC4_STATUS(0x411): 0xbe00000000800400
IA32_MC4_ADDR(0x412): 0x<ptr>
IA32_MC4_MISC(0x413): 0x<ptr>
IA32_MC17_STATUS(0x445): 0xfe200000000c110a
IA32_MC17_ADDR(0x446): 0x00000000d1000000
IA32_MC17_MISC(0x447): 0x04fc381602402086
Processor 2: IA32_MCG_STATUS: 0x0000000000000005
IA32_MC4_STATUS(0x411): 0xbe00000000800400
IA32_MC4_ADDR(0x412): 0x<ptr>
IA32_MC4_MISC(0x413): 0x<ptr>
IA32_MC17_STATUS(0x445): 0xfe200000000c110a
IA32_MC17_ADDR(0x446): 0x00000000d1000000
IA32_MC17_MISC(0x447): 0x04fc381602402086
Processor 3: IA32_MCG_STATUS: 0x0000000000000005
IA32_MC4_STATUS(0x411): 0xbe00000000800400
IA32_MC4_ADDR(0x412): 0x<ptr>
IA32_MC4_MISC(0x413): 0x<ptr>
IA32_MC17_STATUS(0x445): 0xfe200000000c110a
IA32_MC17_ADDR(0x446): 0x00000000d1000000
IA32_MC17_MISC(0x447): 0x04fc381602402086
Processor 6: IA32_MCG_STATUS: 0x0000000000000005
IA32_MC4_STATUS(0x411): 0xbe00000000800400
IA32_MC4_ADDR(0x412): 0x<ptr>
IA32_MC4_MISC(0x413): 0x<ptr>
IA32_MC17_STATUS(0x445): 0xfe200000000c110a
IA32_MC17_ADDR(0x446): 0x00000000d1000000
IA32_MC17_MISC(0x447): 0x04fc381602402086
Processor 7: IA32_MCG_STATUS: 0x0000000000000005
IA32_MC4_STATUS(0x411): 0xbe00000000800400
IA32_MC4_ADDR(0x412): 0x<ptr>
IA32_MC4_MISC(0x413): 0x<ptr>
IA32_MC17_STATUS(0x445): 0xfe200000000c110a
IA32_MC17_ADDR(0x446): 0x00000000d1000000
IA32_MC17_MISC(0x447): 0x04fc381602402086
Processor 8: IA32_MCG_STATUS: 0x0000000000000005
IA32_MC4_STATUS(0x411): 0xbe00000000800400
IA32_MC4_ADDR(0x412): 0x<ptr>
IA32_MC4_MISC(0x413): 0x<ptr>
IA32_MC17_STATUS(0x445): 0xfe200000000c110a
IA32_MC17_ADDR(0x446): 0x00000000d1000000
IA32_MC17_MISC(0x447): 0x04fc381602402086
Processor 9: IA32_MCG_STATUS: 0x0000000000000005
IA32_MC4_STATUS(0x411): 0xbe00000000800400
IA32_MC4_ADDR(0x412): 0x<ptr>
IA32_MC4_MISC(0x413): 0x<ptr>
IA32_MC17_STATUS(0x445): 0xfe200000000c110a
IA32_MC17_ADDR(0x446): 0x00000000d1000000
IA32_MC17_MISC(0x447): 0x04fc381602402086
Processor 10: IA32_MCG_STATUS: 0x0000000000000005
IA32_MC4_STATUS(0x411): 0xbe00000000800400
IA32_MC4_ADDR(0x412): 0x<ptr>
IA32_MC4_MISC(0x413): 0x<ptr>
IA32_MC17_STATUS(0x445): 0xfe200000000c110a
IA32_MC17_ADDR(0x446): 0x00000000d1000000
IA32_MC17_MISC(0x447): 0x04fc381602402086
Processor 11: IA32_MCG_STATUS: 0x0000000000000005
IA32_MC4_STATUS(0x411): 0xbe00000000800400
IA32_MC4_ADDR(0x412): 0x<ptr>
IA32_MC4_MISC(0x413): 0x<ptr>
IA32_MC17_STATUS(0x445): 0xfe200000000c110a
IA32_MC17_ADDR(0x446): 0x00000000d1000000
IA32_MC17_MISC(0x447): 0x04fc381602402086
mp_kdp_enter() timed-out on cpu 7, NMI-ing
mp_kdp_enter() NMI pending on cpus: 0 1 2 3 4 5 6 8 9 10 11
mp_kdp_enter() timed-out during locked wait after NMI;expected 12 acks but received 1 after 35529925 loops in 1745959985 ticks
panic(cpu 7 caller 0xffffff80103c5fc9): "Machine Check at 0xffffff80102a3860, registers:\n" "CR0: 0x000000008001003b, CR2: 0x00007f8d364cc000, CR3: 0x0000000016113000, CR4: 0x00000000001626e0\n" "RAX: 0x000000000000018b, RBX: 0x0000000000000045, RCX: 0x0000000000000044, RDX: 0xffffff80110c7500\n" "RSP: 0xffffffa0accb8dd0, RBP: 0xffffffa0accb8e10, RSI: 0x0000000000000000, RDI: 0xffffff80110c7500\n" "R8: 0x0000000000000001, R9: 0x0000000272891542, R10: 0x0000000000000000, R11: 0x000006e6a14e0d30\n" "R12: 0xffffff86af2e1140, R13: 0x0000000000000000, R14: 0x0000000000000000, R15: 0xffffff80110c7500\n" "RFL: 0x0000000000000097, RIP: 0xffffff80102a3860, CS: 0x0000000000000008, SS: 0x0000000000000010\n" "Error code: 0x0000000000000000\n"@/System/Volumes/Data/SWE/macOS/BuildRoots/220e8a1b79/Library/Caches/com.apple.xbs/Sources/xnu/xnu-7195.141.8/osfmk/i386/trap_native.c:168
Backtrace (CPU 7), Frame : Return Address
0xffffff8010159ad0 : 0xffffff801028d25d
0xffffff8010159b20 : 0xffffff80103d49d3
0xffffff8010159b60 : 0xffffff80103c4fca
0xffffff8010159bb0 : 0xffffff8010231a2f
0xffffff8010159bd0 : 0xffffff801028ca7d
0xffffff8010159cf0 : 0xffffff801028cd73
0xffffff8010159d60 : 0xffffff8010a9d8fa
0xffffff8010159dd0 : 0xffffff80103c5fc9
0xffffff8010159ec0 : 0xffffff8010aa169e
0xffffff8010159ed0 : 0xffffff801023228f
0xffffffa0accb8e10 : 0xffffff80102b5039
0xffffffa0accb8e50 : 0xffffff80102a90e4
0xffffffa0accb8e90 : 0xffffff80102de226
0xffffffa0accb8f20 : 0xffffff80103b0846
0xffffffa0accb8f60 : 0xffffff80103ce04a
0xffffffa0accb8f80 : 0xffffff80103c476b
0xffffffa0accb8fd0 : 0xffffff8010231bdd
0xffffffa0adc536a0 : 0xffffff801035a5a1
0xffffffa0adc53810 : 0xffffff8010359106
0xffffffa0adc53900 : 0xffffff8010a0493d
0xffffffa0adc53a20 : 0xffffff8010a01b51
0xffffffa0adc53a50 : 0xffffff8012e7b410
0xffffffa0adc53b10 : 0xffffff8010560ea0
0xffffffa0adc53b70 : 0xffffff80104febc6
0xffffffa0adc53be0 : 0xffffff80104fd831
0xffffffa0adc53c20 : 0xffffff80133af0e8
0xffffffa0adc53c70 : 0xffffff80133aed32
0xffffffa0adc53cb0 : 0xffffff80133ae50c
0xffffffa0adc53d60 : 0xffffff80134abe47
0xffffffa0adc53da0 : 0xffffff80134abae2
0xffffffa0adc53e10 : 0xffffff80134c64fd
0xffffffa0adc53f20 : 0xffffff80134c5ec7
0xffffffa0adc53fa0 : 0xffffff801023113e
Kernel Extensions in backtrace:
com.apple.iokit.IOStorageFamily(2.1)[AEB16297-5950-3E23-84C4-2E9928878BA1]@0xffffff8012e6e000->0xffffff8012e7ffff
com.apple.filesystems.apfs(1677.141.2)[A2E36135-081F-3EE7-9F8A-53103C506BE8]@0xffffff80133ab000->0xffffff801351afff
dependency: com.apple.driver.AppleEFINVRAM(2.1)[511FC3DC-D63F-3993-9C5C-613AF51D837C]@0xffffff80116fe000->0xffffff8011707fff
dependency: com.apple.driver.AppleEffaceableStorage(1.0)[7AC5E118-B326-3E46-980F-4706B9C5ABF0]@0xffffff8011711000->0xffffff8011716fff
dependency: com.apple.iokit.CoreAnalyticsFamily(1)[3040B1E0-7488-35C7-9DB9-ED6C9EF965E7]@0xffffff8011b6d000->0xffffff8011b73fff
dependency: com.apple.iokit.IOStorageFamily(2.1)[AEB16297-5950-3E23-84C4-2E9928878BA1]@0xffffff8012e6e000->0xffffff8012e7ffff
dependency: com.apple.kec.corecrypto(11.1)[379F0025-A70B-3A5D-A92F-3C557D6E4E68]@0xffffff8013548000->0xffffff80135d9fff
dependency: com.apple.security.AppleImage4(3.0.0)[7B75FFFD-E85A-3686-8D37-27B84E4DFE6D]@0xffffff801177d000->0xffffff801178dfff
Process name corresponding to current thread: kernel_task
Boot args: serverperfmode=1 -v npci=0x2000 dart=0 -no_compat_check debug=0x144 -lilubetaall shikigva=128 unfairgva=1 mbasd=1 -cdfon no32exec=0
Mac OS version:
20G224
Kernel version:
Darwin Kernel Version 20.6.0: Tue Oct 12 18:33:42 PDT 2021; root:xnu-7195.141.8~1/RELEASE_X86_64
Kernel UUID: ABC69550-60C2-34FE-B307-C24A8C39309C
KernelCache slide: 0x0000000010000000
KernelCache base: 0xffffff8010200000
Kernel slide: 0x0000000010010000
Kernel text base: 0xffffff8010210000
__HIB text base: 0xffffff8010100000
System model name: iMacPro1,1 (Mac-7BA5B2D9E42DDD94)
System shutdown begun: NO
Panic diags file available: NO (0xe00002bc)
Hibernation exit count: 0
System uptime in nanoseconds: 7610710580047
Last Sleep: absolute base_tsc base_nano
Uptime : 0x000006ec01b3c71d
Sleep : 0x0000000000000000 0x0000000000000000 0x0000000000000000
Wake : 0x0000000000000000 0x000000576b6d1263 0x0000000000000000
From studying firmware dumps, I've noticed that data gets written to the flash memory when different devices are connected, apparently for communication between software and firmware. This interaction also depends on GPIOs, but because these are lacking on unsupported hardware, it's possible that the card ends up in some undetermined state (persistently because of the data written to the flash), causing the panics. You may be able to fix the issue by reflashing a clean firmware.
Interesting. By a “clean” firmware do you mean another copy of the modded firmware or the original firmware from before the flash? I’ve heard of other modded firmwares being available but I’m not an expert on all the nuances of those. does the rev 2 card use the same firmware? Ideally I’d like to have the TB3 card working again without the panics.
The modified firmware that you originally flashed should be relatively clean. I say "relatively" because the files in circulation are made from dumps, which contain data from prior use themselves. If you want to try this experiment, make sure to dump your current firmware first, so that we can take a look.
While it's an Apple firmware, it's unfortunately a regression. In my testing, however, both ports worked (maybe a V1 vs V2 thing?).
I'm wondering about this as well.. should I go with the v2 titan ridge (known to work) or the new maple ridge version?
Last edited:
Neither, of course. I use the original approach:
OpenCore on the Mac Pro
This guide explains how to use the excellent OpenCore boot loader to install, run, and update the latest version of macOS on the MacPro5,1, resulting in a clean, unaltered operating system just like on a supported Mac.
forums.macrumors.com
As long as it loads the SSDT, the particular OC setup doesn't matter. The state of the art for the SSDT is what HackinDROM can produce. I use a modified version (in particular, without the "Intel JHL7540..." name, because, for instance, the 2019 Mac Pro doesn't display such information in System Information > Hardware > PCI).
No, because there's no macOS support for Maple Ridge.
Titan Ridge is currently the best option.
If anyone is inclined to do more testing with Apple firmware, these are all potential candidates:
Mac-1E7E29AD0135F9BC
Mac-7BA5B2DFE22DDD8C
Mac-226CB3C6A851A671
Mac-827FAC58A8FDFA22
Mac-827FB448E656EC26
Mac-937A206F2EE63C01
Mac-63001698E7A34814
Mac-AA95B1DDAB278B95
Mac-1E7E29AD0135F9BC
Mac-7BA5B2DFE22DDD8C
Mac-226CB3C6A851A671
Mac-827FAC58A8FDFA22
Mac-827FB448E656EC26
Mac-937A206F2EE63C01
Mac-63001698E7A34814
Mac-AA95B1DDAB278B95
GUIDE - How to Enable Thunderbolt 4 Hotplug on macOS
How to Enable Thunderbolt 4 Hotplug on macOS [Clover/OpenCore] An EliteMacx86 Exclusive Guide - This guide covers implementation of Thunderbolt 4 Hotplug on macOS for built-in and Add In Card (AIC) Thunderbolt hardware. Overview Recently, Intel and Apple announced their new Thunderbolt...
elitemacx86.com