Thanks for the report. I've also seen the "found dead pci device GFX0" Lilu message before a hang on my Mac Pro.
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 Latebloom - An experimental workaround for the 11.3+ race condition
- Thread starter Syncretic
- Start date
- Sort by reaction score
- Status
Thanks for the report. I've also seen the "found dead pci device GFX0" Lilu message before a hang on my Mac Pro.
Yeah, I also saw that the Broadcom 20702B0 card is Mac Pro 6,1 specific in Monterey.
I just updated my MacPro 4,1->5,1 from MacOS 11.3 to MacOS 11.5.2 via Software Update in the System Preferences app. Not a single boot failure during the update. Before I started the MacOS update, I updated to the latest version of h9826790's OpenCore all-in-one package, but in the config.plist moved the Latebloom kext addition to after Lilu and WhateverGreen and replaced it with what dataid listed in the Guide to using Latebloom. Thank you for the Latebloom workaround, it made my update experience much smoother than expected!
edit to add: I also downloaded the latest latebloom.kext from post #1 and replaced the version that already was in the OC all-in-one package (did not check if there was a version difference though)
edit to add: I also downloaded the latest latebloom.kext from post #1 and replaced the version that already was in the OC all-in-one package (did not check if there was a version difference though)
Code:
Warmboot
4D1FDA02-38C7-4A6A-9CC6-4BCCA8B30102:opencore-version REL-072-2021-08-02
51 0 0 0xd000 0xd000 AAA.LoadEarly.latebloom (0.19) D0667827-216F-33FF-9C91-1ED4972E4A28 <14 5 3 2 1>
2021-08-23 21:57:09.618478+0200 0x74 Default 0x0 0 0 kernel: (latebloom) _____[ !!! *** latebloom *** !!! ]: Hook placed successfully. Count = 0
Real model decoded from serial number: Mac Pro 2009 (MP4,1)
1 x Intel(R) Xeon(R) CPU W3680 @ 3.33GHz
CKxxxxxJ8PZ
Radeon RX 580
running at 11.5.2 20G95 Kernel 20.6.0
boot-args -lilubetaall -wegbeta agdpmod=pikera shikigva=80 unfairgva=1 mbasd=1 -wegtree -no_compat_check no32exec=0 latebloom=150 lb_range=20 lb_debug=0 -v
Hardware:
Hardware Overview:
Model Name: Mac Pro
Model Identifier: MacPro5,1
Processor Name: 6-Core Intel Xeon
Processor Speed: 3,33 GHz
Number of Processors: 1
Total Number of Cores: 6
L2 Cache (per Core): 256 KB
L3 Cache: 12 MB
Hyper-Threading Technology: Enabled
Memory: 48 GB
System Firmware Version: 9144.0.7.2.0
SMC Version (system): 1.39f5
SMC Version (processor tray): 1.39f5
System Software Overview:
System Version: macOS 11.5.2 (20G95)
Kernel Version: Darwin 20.6.0
Network:
Type: Ethernet
Hardware: Ethernet
Service Order: 0
Type: Ethernet
Hardware: Ethernet
Service Order: 1
Type: Ethernet
Hardware: Ethernet
Service Order: 2
Type: FireWire
Hardware: FireWire
Service Order: 3
Type: AirPort
Hardware: AirPort
Service Order: 4
Type: Ethernet
Hardware: Ethernet
Service Order: 5
Type: Ethernet
Hardware: Ethernet
Service Order: 6
Wi-Fi:
Card Type: AirPort Extreme (0x14E4, 0x111)
Firmware Version: Broadcom BCM43xx 1.0 (7.77.111.1 AirPortDriverBrcmNIC-1680.8)
Storage:
File System: NTFS
BSD Name: disk3s2
Medium Type: Rotational
Protocol: SATA
Internal: Yes
File System: APFS
BSD Name: disk4s1
Medium Type: SSD
Protocol: SATA
Internal: Yes
File System: Journaled HFS+
BSD Name: disk5
Medium Type: Rotational
Protocol: SATA
Internal: Yes
File System: APFS
BSD Name: disk4s3s1
Medium Type: SSD
Protocol: SATA
Internal: Yes
Physical Interconnect: SATA
Physical Interconnect: SATA
BSD Name: disk0
File System: MS-DOS FAT32
BSD Name: disk0s1
Content: EFI
BSD Name: disk0s2
Content: Apple_APFS
Physical Interconnect: SATA
BSD Name: disk3
File System: MS-DOS FAT32
BSD Name: disk3s1
Content: EFI
File System: NTFS
BSD Name: disk3s2
Content: Microsoft Basic Data
Physical Interconnect: SATA
BSD Name: disk1
File System: MS-DOS FAT32
BSD Name: disk1s1
Content: EFI
BSD Name: disk1s2
Content: Apple_RAID
File System: HFS+
BSD Name: disk1s3
Content: Apple_Boot
Physical Interconnect: SATA
BSD Name: disk2
File System: MS-DOS FAT32
BSD Name: disk2s1
Content: EFI
BSD Name: disk2s2
Content: Apple_RAID
File System: HFS+
BSD Name: disk2s3
Content: Apple_Boot
PCI:
Name: ATY,AMD,RadeonFramebuffer
Type: Display Controller
Bus: PCI
Slot: Slot-1
Type: Audio Device
Bus: PCI
Slot: Slot-1
Type: USB eXtensible Host Controller
Bus: PCI
Slot: Slot-4
USB:
USB 2.0 Bus:
Host Controller Driver: AppleUSBEHCIPCI
USB 2.0 Bus:
Host Controller Driver: AppleUSBEHCIPCI
USB 2.0 Hub:
Product ID: 0x0101
Speed: Up to 480 Mb/s
Location ID: 0xfd100000 / 1
Product ID: 0x9136
Speed: Up to 480 Mb/s
Location ID: 0xfd120000 / 3
Product ID: 0x1006
Speed: Up to 480 Mb/s
Location ID: 0xfd122000 / 7
Product ID: 0x0221
Speed: Up to 1.5 Mb/s
Location ID: 0xfd122200 / 11
Product ID: 0x0265
Speed: Up to 12 Mb/s
Location ID: 0xfd121000 / 5
Product ID: 0x9137
Speed: Up to 480 Mb/s
Location ID: 0xfd124000 / 6
Product ID: 0x9236
Speed: Up to 12 Mb/s
Location ID: 0xfd124300 / 9
Product ID: 0x8508
Speed: Up to 480 Mb/s
Location ID: 0xfd124200 / 8
Product ID: 0x2912
Speed: Up to 12 Mb/s
Location ID: 0xfd124100 / 10
Product ID: 0x8021
Speed: Up to 12 Mb/s
Location ID: 0xfd110000 / 2
Product ID: 0x2275
Speed: Up to 12 Mb/s
Location ID: 0xfd111000 / 4
USB Bus:
Host Controller Driver: AppleUSBUHCIPCI
USB Bus:
Host Controller Driver: AppleUSBUHCIPCI
USB Bus:
Host Controller Driver: AppleUSBUHCIPCI
USB Bus:
Host Controller Driver: AppleUSBUHCIPCI
USB Bus:
Host Controller Driver: AppleUSBUHCIPCI
USB Bus:
Host Controller Driver: AppleUSBUHCIPCI
USB 3.0 Bus:
Host Controller Driver: AppleUSBXHCIFL1100
Product ID: 0x9136
Speed: Up to 480 Mb/s
Location ID: 0x00400000 / 1
Product ID: 0x9137
Speed: Up to 480 Mb/s
Location ID: 0x00440000 / 2
Product ID: 0x9236
Speed: Up to 12 Mb/s
Location ID: 0x00443000 / 4
Product ID: 0x8508
Speed: Up to 480 Mb/s
Location ID: 0x00442000 / 5
Product ID: 0x2912
Speed: Up to 12 Mb/s
Location ID: 0x00441000 / 3
Memory:
Upgradeable Memory: Yes
Size: 16 GB
Type: DDR3
Speed: 1333 MHz
Size: 16 GB
Type: DDR3
Speed: 1333 MHz
Size: 16 GB
Type: DDR3
Speed: 1333 MHz
Size: Empty
Type: Empty
Speed: Empty
Last edited:
I am having no trouble with warm boots, but cold starts always require to boot into Catalina and log in, then restart. That works almost 100%.
My parameters are lbloom=200,20,1
Since boot testing and changing parameters are so time-consuming, what direction would be best? Should I go lower?
My parameters are lbloom=200,20,1
Since boot testing and changing parameters are so time-consuming, what direction would be best? Should I go lower?
What do these boot-args do?
-wegbeta agdpmod=pikera shikigva=80 unfairgva=1 -wegtree
-wegbeta agdpmod=pikera shikigva=80 unfairgva=1 -wegtree
This is off-topic, but I'll bite: -wegbeta is for using the WhateverGreen kext on unsupported versions of macOS, agdpmod=pikera is for enabling all video outputs when using iMac-type board-ID spoofing, shikigva=80 is for DRM on Catalina and earlier (80 is indeed the correct value for the Mac Pro 5,1, not 128 as suggested by the WEG documentation), unfairgva=1 is for DRM on Big Sur and later, and -wegtree is for using the modern device tree naming scheme to enable hardware acceleration. Most of these should really be included as device properties to keep the boot-args clean.
This isn't the best thread to ask we've already had the previous race condition thread derail. Recommend asking in the more general threads:
However the answer is within WhateverGreen's public documentation. But a simple TL;DR:
-wegbeta: Load WhateverGreen on unsupported OSes- Defined under kern_start, currently is set to load with Snow Leopard through Monterey
- Absolutely useless if you update your kexts, but many still seem to keep the arg
agdpmod=pikera:- Patches AppleGraphicsDisplayPolicy to allow display output building instead of relying on predefined display output mapping
- Useful for iMac Pro SMBIOS where only 3 displays are officially support, 2 being DisplayPort and the other LVDS and thus creating black screen output on Mac Pros
shikigva=80:- Fix DRM on Intel iGPU + AMD dGPUs using an iMac or Mac mini SMBIOS
- For iMac Pros and Mac Pro SMBIOS, 128 is a more suitable value as per WEG documentation
unfairgva=1:- Used to patch DRM in macOS 11 and newer
-wegtree:- Rebuilds device tree generated in IOService with ones more compliant to what macOS expects (ie. GPUs being named GFX0)
Thanks. I did not mean to derail the thread. Just looking for correlations between boot arguments and latebloom.kext successes and/or more reliable booting in general with the race conditions.
Note: there is a rare race condition in 11.2.3 and earlier. It just doesn't happen nearly as often as 11.3+ and later.
Trying to gain some insight on how the new way of booting unsupported post 11.2.3 may be improved or the race condition eliminated. I am still awestruck that VMs are not exhibiting any issues with booting. While they are not real machines, but do not always update their processes very quickly either.
Mainly looking because I have been having less success now with late bloom than when I first did. I don't know if it was just beginner's luck or a placebo effect. Trying to find what is the determiner on why my 3,1's and 5,1's are behaving badly with late bloom and seeing quite a bit of success based on reports. I know this is still in beta, just trying to find a way to make it more reliable. I'm not trying really hard on this. Mostly doing some minimal investigation and interested if there is a way out. I do want to spoof a Mac Book Pro and see how reliable Anka Developer VM is on it; curious that boots reliably and then try to dissect some of what it does when it preps an OS. Long shot. Just poking around.
Note: there is a rare race condition in 11.2.3 and earlier. It just doesn't happen nearly as often as 11.3+ and later.
Trying to gain some insight on how the new way of booting unsupported post 11.2.3 may be improved or the race condition eliminated. I am still awestruck that VMs are not exhibiting any issues with booting. While they are not real machines, but do not always update their processes very quickly either.
Mainly looking because I have been having less success now with late bloom than when I first did. I don't know if it was just beginner's luck or a placebo effect. Trying to find what is the determiner on why my 3,1's and 5,1's are behaving badly with late bloom and seeing quite a bit of success based on reports. I know this is still in beta, just trying to find a way to make it more reliable. I'm not trying really hard on this. Mostly doing some minimal investigation and interested if there is a way out. I do want to spoof a Mac Book Pro and see how reliable Anka Developer VM is on it; curious that boots reliably and then try to dissect some of what it does when it preps an OS. Long shot. Just poking around.
Last edited:
Their device trees are absolutely minuscule compared to a real machine's, and thus have less items to probe and even less chance of hitting the race condition. Additionally as everything is virtualized, there's less concerns with some 3rd party device's firmware taking too long to initialize and forcing IOService to stall
Test something on my 4.1 test rig with the latest APFS Driver by formatting the System disk with 11.5.2 createinstallmedia thumb drive.
The SSD was the only drive (but OCLP 0.24 and createinstallmedia USB Sticks)
Sadly the race condition is still there. Had 2 freezes during boot / installation, one in crypto, one in USB. The Thumb drives were installed in HFSplus.
transferred the SSD to my main machine (5,1), booted find with @Dayo's MyBootMgr (RefindPlus - OpenCore chain loading),
The SSD was the only drive (but OCLP 0.24 and createinstallmedia USB Sticks)
Sadly the race condition is still there. Had 2 freezes during boot / installation, one in crypto, one in USB. The Thumb drives were installed in HFSplus.
transferred the SSD to my main machine (5,1), booted find with @Dayo's MyBootMgr (RefindPlus - OpenCore chain loading),
Code:
Coldboot
4D1FDA02-38C7-4A6A-9CC6-4BCCA8B30102:opencore-version DBG-072-2021-08-02
50 0 0 0xd000 0xd000 AAA.LoadEarly.latebloom (0.19) D0667827-216F-33FF-9C91-1ED4972E4A28 <14 5 3 2 1>
2021-08-24 15:34:19.215391-0700 0x74 Default 0x0 0 0 kernel: (latebloom) _____[ !!! *** latebloom *** !!! ]: Hook placed successfully. Count = 0
Real model decoded from serial number: Mac Pro 2010 (MP5,1)
1 x Intel(R) Xeon(R) CPU W3670 @ 3.20GHz
CKxxxxxCEUG
NVIDIA GeForce GT 640
running at 11.5.2 20G95 Kernel 20.6.0
boot-args agdpmod=pikera wegtree=1 latebloom=150 lb_range=30 lb_debug=0 debug=0x144 keepsyms=1 -lilubetaall -no_compat_check
Hardware:
Hardware Overview:
Model Name: Mac Pro
Model Identifier: MacPro5,1
Processor Name: 6-Core Intel Xeon
Processor Speed: 3,2 GHz
Number of Processors: 1
Total Number of Cores: 6
L2 Cache (per Core): 256 KB
L3 Cache: 12 MB
Hyper-Threading Technology: Enabled
Memory: 48 GB
System Firmware Version: 9999.9.9.9.9
SMC Version (system): 9.99f99
SMC Version (processor tray): 1.39f11
System Software Overview:
System Version: macOS 11.5.2 (20G95)
Kernel Version: Darwin 20.6.0
Network:
Type: Ethernet
Hardware: Ethernet
Service Order: 0
Type: Ethernet
Hardware: Ethernet
Service Order: 1
Type: AirPort
Hardware: AirPort
Service Order: 2
Type: Ethernet
Hardware: Ethernet
Service Order: 3
Type: Ethernet
Hardware: Ethernet
Service Order: 4
Wi-Fi:
Card Type: AirPort Extreme (0x14E4, 0x8E)
Firmware Version: Broadcom BCM43xx 1.0 (5.106.98.102.31)
Storage:
File System: Journaled HFS+
BSD Name: disk1s2
Medium Type: Rotational
Protocol: SATA
Internal: Yes
File System: APFS
BSD Name: disk2s1
Medium Type: SSD
Protocol: SATA
Internal: Yes
File System: Journaled HFS+
BSD Name: disk3s1
Medium Type: Rotational
Protocol: SATA
Internal: Yes
File System: MS-DOS FAT32
BSD Name: disk3s2
Medium Type: Rotational
Protocol: SATA
Internal: Yes
File System: APFS
BSD Name: disk2s5s1
Medium Type: SSD
Protocol: SATA
Internal: Yes
File System: Journaled HFS+
BSD Name: disk4s2
Medium Type: Rotational
Protocol: SATA
Internal: Yes
File System: Journaled HFS+
BSD Name: disk7s2
Medium Type: Rotational
Protocol: SATA
Internal: Yes
File System: Journaled HFS+
BSD Name: disk8s2
Protocol: USB
Internal: No
File System: Journaled HFS+
BSD Name: disk9s2
Protocol: FireWire
Internal: No
File System: Journaled HFS+
BSD Name: disk10s2
Protocol: USB
Internal: No
Physical Interconnect: SATA
Physical Interconnect: SATA
BSD Name: disk0
File System: MS-DOS FAT32
BSD Name: disk0s1
Content: EFI
BSD Name: disk0s2
Content: Apple_APFS
Physical Interconnect: SATA
BSD Name: disk1
File System: MS-DOS FAT32
BSD Name: disk1s1
Content: EFI
File System: Journaled HFS+
BSD Name: disk1s2
Content: Apple_HFS
File System: Journaled HFS+
BSD Name: disk1s3
Content: Apple_Boot
Physical Interconnect: SATA
BSD Name: disk7
File System: MS-DOS FAT32
BSD Name: disk7s1
Content: EFI
File System: Journaled HFS+
BSD Name: disk7s2
Content: Apple_HFS
Physical Interconnect: SATA
BSD Name: disk3
File System: Journaled HFS+
BSD Name: disk3s1
Content: Apple_HFS
File System: MS-DOS FAT32
BSD Name: disk3s2
Content: DOS_FAT_32
Physical Interconnect: SATA
BSD Name: disk4
File System: MS-DOS FAT32
BSD Name: disk4s1
Content: EFI
File System: Journaled HFS+
BSD Name: disk4s2
Content: Apple_HFS
File System: Journaled HFS+
BSD Name: disk4s3
Content: Apple_Boot
PCI:
Type: NVM Express Controller
Bus: PCI
Slot: Slot-2
Name: NVDA,Display-D
Type: Display Controller
Bus: PCI
Slot: Slot-1
Name: pci10de,e1b
Type: NVDA,Parent
Bus: PCI
Slot: Slot-1
Type: USB eXtensible Host Controller
Bus: PCI
Slot: Slot-4
NVMExpress:
Generic SSD Controller:
Samsung SSD 970 EVO Plus 500GB:
Capacity: 500,11 GB (500.107.862.016 bytes)
TRIM Support: Yes
Model: Samsung SSD 970 EVO Plus 500GB
Revision: 2B2QEXM7
Link Width: x4
Link Speed: 5.0 GT/s
Detachable Drive: No
BSD Name: disk5
Partition Map Type: GPT (GUID Partition Table)
Removable Media: No
S.M.A.R.T. status: Verified
Volumes:
EFI:
Capacity: 209,7 MB (209.715.200 bytes)
File System: MS-DOS FAT32
BSD Name: disk5s1
Content: EFI
disk5s2:
Capacity: 499,9 GB (499.898.105.856 bytes)
BSD Name: disk5s2
Content: Apple_APFS
USB:
USB 2.0 Bus:
Host Controller Driver: AppleUSBEHCIPCI
Product ID: 0x1000
Speed: Up to 480 Mb/s
Location ID: 0xfd500000 / 1
USB Interface: 0
USB 2.0 Bus:
Host Controller Driver: AppleUSBEHCIPCI
Product ID: 0x1006
Speed: Up to 480 Mb/s
Location ID: 0xfa200000 / 1
Product ID: 0x0001
Speed: Up to 12 Mb/s
Location ID: 0xfa230000 / 3
Product ID: 0x0265
Speed: Up to 12 Mb/s
Location ID: 0xfa210000 / 4
Product ID: 0x0221
Speed: Up to 1.5 Mb/s
Location ID: 0xfa220000 / 2
USB Bus:
Host Controller Driver: AppleUSBUHCIPCI
USB Bus:
Host Controller Driver: AppleUSBUHCIPCI
USB Bus:
Host Controller Driver: AppleUSBUHCIPCI
USB Bus:
Host Controller Driver: AppleUSBUHCIPCI
USB Bus:
Host Controller Driver: AppleUSBUHCIPCI
USB Bus:
Host Controller Driver: AppleUSBUHCIPCI
USB 3.0 Bus:
Host Controller Driver: AppleUSBXHCIFL1100
Product ID: 0x8140
Speed: Up to 5 Gb/s
Location ID: 0x00600000 / 5
USB3.0 Hub:
Product ID: 0x0616
Speed: Up to 5 Gb/s
Location ID: 0x00500000 / 3
Product ID: 0x25a2
Speed: Up to 5 Gb/s
Location ID: 0x00520000 / 6
USB Interface: 0
Product ID: 0x8142
Speed: Up to 480 Mb/s
Location ID: 0x00200000 / 2
Product ID: 0x12ab
Speed: Up to 480 Mb/s
Location ID: 0x00210000 / 7
USB2.0 Hub:
Product ID: 0x0610
Speed: Up to 480 Mb/s
Location ID: 0x00100000 / 1
Bluetooth:
Bluetooth Low Energy Supported: Yes
Handoff Supported: No
Instant Hot Spot Supported: No
Manufacturer: Cambridge Silicon Radio
Composite Class Of Device: 0x380104
Connected: No
Class of Device: 0x05 0x25 0x2594
Memory:
Upgradeable Memory: Yes
Size: 16 GB
Type: DDR3
Speed: 1066 MHz
Size: 16 GB
Type: DDR3
Speed: 1066 MHz
Size: 8 GB
Type: DDR3
Speed: 1066 MHz
Size: 8 GB
Type: DDR3
Speed: 1066 MHzThat might well be entirely true for "Anka Developer VM", but I've experienced several problems with Big Sur 11.3+ on Parallels Desktop VM update procedures. If I remember correctly, I was utterly unable to update 11.2.3 to 11.3. After more than five hung update processes, I had to delete the 11.2.3 VM and create an 11.3 VM from scratch. All later updates (to 11.4, 11.5, 11.5.1 and 11.5.2) were eventually successful, but they all required several VM reboots (more than was to be expected) and, on two occasions, a manual VM reset, since the update was hung. I know the VM "hardware" is not really comparable to a Mac Pro 5,1, but I've sometimes wondered why the race condition should affect Parallels virtual machine update processes as well. Oddly enough, once they're installed/updated, the BS 11.3+ virtual machines themselves always boot without issues.
Another "blink and you'll miss it" visit from me. The moments I get to spend on this are precious and few. (And now, if you're of a certain age, you probably have that song stuck in your head - sorry! ;-)
While pondering the "out of order" PCI probe loops that @sfalatko brought to my attention, which I've subsequently seen on my own system, I discovered something mildly interesting.
PCIe bus 0 includes all onboard devices except the Ethernet controllers, the FireWire controller, and any cards in the PCIe slots (GPU, USB3, NVMe, etc.). I've modified latebloom to distinguish between these - I'm calling the bus 0 probes "Phase 1" (or "Onboard") and the remaining probes "Phase 2" (or "External"), and I've added a separate delay/range for "Phase 2," which allows for more flexibility when experimenting.
On my system (MP4,1->5,1 with no WiFi card), there are 55 devices on bus 0, so "Phase 2" starts on loop 56. (I don't know if a WiFi card would appear on bus 0 or not; hopefully, someone else can report on that, just for completeness if nothing else). I've found that I can lower my "Phase 1" timing quite a bit (to 15-30) if I keep my "Phase 2" timing relatively high (~200 +/- 50); this helps speed up the overall boot process. Interestingly, while I'd hoped to find that either "Phase 1" or "Phase 2" could effectively be set to 0 (meaning the entire problem was on one or the other), that's not the case - setting either one too low causes frequent failures, at least on my system.
The attached version (0.21) is intended for experimenters. If you're using latebloom with satisfactory results, this version does not provide any new features or advantages, so "upgrading" is neither necessary nor particularly desirable.
If you're interested in helping to gather more data, please consider trying version 0.21, playing around with "Phase 1/Phase 2," and posting your results. The distinction between single- and multi-threaded probes may end up being a red herring, but I think it looks interesting enough to investigate.
The "Phase 2" arguments work just like their "Phase 1" counterparts; delay in milliseconds, +/- range in milliseconds, unspecified arguments use defaults (the "Phase 1" values, in this case), empty specified arguments are interpreted as 0. If no "Phase 2" arguments are specified, latebloom works as before, using the "latebloom=" and "lb_range=" values for every loop. If a "Phase 2" delay is specified with no associated range, the "Phase 1" range is used. If a "Phase 2" range is specified with no associated delay, the "Phase 1" delay is used with the specified "Phase 2" range (i.e. you can use the same delay for both phases, changing only the range - sadly, this isn't as interesting as it sounds). Note that while
New boot-args:
Modified boot-args:
Example:
is equivalent to
All boot-args from prior versions still work as before. Since "Phase 2" assumes its defaults from "Phase 1," version 0.21 can be used as a drop-in replacement for prior versions; other than slightly different log messages, its behavior should be effectively identical.
It may be a while before I can revisit this, but feel free to play around with v0.21 and see if it makes a difference on your system. I'm curious to see if there's any consistency between systems regarding the use of high/low values on the separate phases.
Just to repeat - If a previous version of latebloom is doing what you need it to do, and you're not interested in experimenting, and you're not keen to "fix what ain't broke," please don't bother with this version.
(I have labeled the attached version "DEBUG" only because it's really just another experiment, not a new feature release. If we see interesting "Phase 1/Phase 2" results, I may re-label it, or produce a new version that further exploits whatever we learn from this one. Also, I have named it as @TECK requested, and it's just a straight ZIP file. I previously used tar/tgz files to preserve ownership/permissions, but that's not really necessary for use with OpenCore.)
While pondering the "out of order" PCI probe loops that @sfalatko brought to my attention, which I've subsequently seen on my own system, I discovered something mildly interesting.
IOPCIFamily.kext enumerates PCIe bus 0 while single-threaded, then launches multi-threaded probes for all remaining buses. This explains the occasional "out of order" display, but may also provide some clues toward a more solid solution.PCIe bus 0 includes all onboard devices except the Ethernet controllers, the FireWire controller, and any cards in the PCIe slots (GPU, USB3, NVMe, etc.). I've modified latebloom to distinguish between these - I'm calling the bus 0 probes "Phase 1" (or "Onboard") and the remaining probes "Phase 2" (or "External"), and I've added a separate delay/range for "Phase 2," which allows for more flexibility when experimenting.
On my system (MP4,1->5,1 with no WiFi card), there are 55 devices on bus 0, so "Phase 2" starts on loop 56. (I don't know if a WiFi card would appear on bus 0 or not; hopefully, someone else can report on that, just for completeness if nothing else). I've found that I can lower my "Phase 1" timing quite a bit (to 15-30) if I keep my "Phase 2" timing relatively high (~200 +/- 50); this helps speed up the overall boot process. Interestingly, while I'd hoped to find that either "Phase 1" or "Phase 2" could effectively be set to 0 (meaning the entire problem was on one or the other), that's not the case - setting either one too low causes frequent failures, at least on my system.
The attached version (0.21) is intended for experimenters. If you're using latebloom with satisfactory results, this version does not provide any new features or advantages, so "upgrading" is neither necessary nor particularly desirable.
If you're interested in helping to gather more data, please consider trying version 0.21, playing around with "Phase 1/Phase 2," and posting your results. The distinction between single- and multi-threaded probes may end up being a red herring, but I think it looks interesting enough to investigate.
The "Phase 2" arguments work just like their "Phase 1" counterparts; delay in milliseconds, +/- range in milliseconds, unspecified arguments use defaults (the "Phase 1" values, in this case), empty specified arguments are interpreted as 0. If no "Phase 2" arguments are specified, latebloom works as before, using the "latebloom=" and "lb_range=" values for every loop. If a "Phase 2" delay is specified with no associated range, the "Phase 1" range is used. If a "Phase 2" range is specified with no associated delay, the "Phase 1" delay is used with the specified "Phase 2" range (i.e. you can use the same delay for both phases, changing only the range - sadly, this isn't as interesting as it sounds). Note that while
latebloom=0 still causes the kext to do nothing at all, lb_delay2=0 is perfectly valid, and will cause "Phase 2" probes to run with no delays. (latebloom=1 is currently the fastest that "Phase 1" can be run.)New boot-args:
lb_delay2=NNNN - The "Phase 2" delay (milliseconds) (defaults to the latebloom= value)lb_range2=NNNN - The "Phase 2" range (defaults to the lb_range= value)lbloom=delay1,range1,debug,delay2,range2(As before, omitted arguments take their defaults, while present or implied arguments become 0. For example,
lbloom=100,50,1 assumes delay2=delay1 and range2=range1; lbloom=100,50,1, (note trailing comma) sets delay2=0 and range2=range1. (The trailing comma implies the next argument, whether it's present or not.))Example:
latebloom=30 lb_range=10 lb_debug=1 lb_delay2=200 lb_range2=50lbloom=30,10,1,200,50All boot-args from prior versions still work as before. Since "Phase 2" assumes its defaults from "Phase 1," version 0.21 can be used as a drop-in replacement for prior versions; other than slightly different log messages, its behavior should be effectively identical.
It may be a while before I can revisit this, but feel free to play around with v0.21 and see if it makes a difference on your system. I'm curious to see if there's any consistency between systems regarding the use of high/low values on the separate phases.
Just to repeat - If a previous version of latebloom is doing what you need it to do, and you're not interested in experimenting, and you're not keen to "fix what ain't broke," please don't bother with this version.
(I have labeled the attached version "DEBUG" only because it's really just another experiment, not a new feature release. If we see interesting "Phase 1/Phase 2" results, I may re-label it, or produce a new version that further exploits whatever we learn from this one. Also, I have named it as @TECK requested, and it's just a straight ZIP file. I previously used tar/tgz files to preserve ownership/permissions, but that's not really necessary for use with OpenCore.)
it's working with your default settings, latebloom_check v10 also
Code:
Warmboot
4D1FDA02-38C7-4A6A-9CC6-4BCCA8B30102:opencore-version REL-071-2021-07-02
56 0 0 0xe000 0xe000 AAA.LoadEarly.latebloom (0.21) 6C8398F6-162F-36B9-A430-D6D98797BF4A <14 5 3 2 1>
2021-08-26 16:06:41.076959-0700 0x74 Default 0x0 0 0 kernel: (latebloom) _____[ !!! *** latebloom *** !!! ]: Hook placed successfully. Count = 0 :: 30,10,1,200,50
Real model decoded from serial number: Mac Pro 2010 (MP5,1)
1 x Intel(R) Xeon(R) CPU W3670 @ 3.20GHz
CKxxxxxCEUG
NVIDIA GeForce GT 640
running at 11.5.2 20G95 Kernel 20.6.0
boot-args keepsyms=1 debug=0x100 -lilubetaall latebloom=30 lb_range=10 lb_debug=1 lb_delay2=200 lb_range2=50 -v -no_compat_check
Hardware:
Hardware Overview:
Model Name: Mac Pro
Model Identifier: MacPro5,1
Processor Name: 6-Core Intel Xeon
Processor Speed: 3,2 GHz
Number of Processors: 1
Total Number of Cores: 6
L2 Cache (per Core): 256 KB
L3 Cache: 12 MB
Hyper-Threading Technology: Enabled
Memory: 48 GB
System Firmware Version: 9999.999.999.999.999
SMC Version (system): 9.9999
SMC Version (processor tray): 1.39f11
System Software Overview:
System Version: macOS 11.5.2 (20G95)
Kernel Version: Darwin 20.6.0
Network:
Type: Ethernet
Hardware: Ethernet
Service Order: 0
Type: Ethernet
Hardware: Ethernet
Service Order: 1
Type: AirPort
Hardware: AirPort
Service Order: 2
Type: Ethernet
Hardware: Ethernet
Service Order: 3
Type: Ethernet
Hardware: Ethernet
Service Order: 4
Wi-Fi:
Card Type: AirPort Extreme (0x14E4, 0x8E)
Firmware Version: Broadcom BCM43xx 1.0 (5.106.98.102.31)
Storage:
File System: Journaled HFS+
BSD Name: disk1s2
Medium Type: Rotational
Protocol: SATA
Internal: Yes
File System: APFS
BSD Name: disk2s1
Medium Type: SSD
Protocol: SATA
Internal: Yes
File System: Journaled HFS+
BSD Name: disk3s1
Medium Type: Rotational
Protocol: SATA
Internal: Yes
File System: MS-DOS FAT32
BSD Name: disk3s2
Medium Type: Rotational
Protocol: SATA
Internal: Yes
File System: Journaled HFS+
BSD Name: disk4s2
Medium Type: Rotational
Protocol: SATA
Internal: Yes
File System: APFS
BSD Name: disk2s5s1
Medium Type: SSD
Protocol: SATA
Internal: Yes
File System: Journaled HFS+
BSD Name: disk5s2
Medium Type: Rotational
Protocol: SATA
Internal: Yes
File System: Journaled HFS+
BSD Name: disk6s2
Protocol: USB
Internal: No
File System: Journaled HFS+
BSD Name: disk9s2
Protocol: USB
Internal: No
File System: Journaled HFS+
BSD Name: disk10s2
Protocol: FireWire
Internal: No
Physical Interconnect: SATA
Physical Interconnect: SATA
BSD Name: disk0
File System: MS-DOS FAT32
BSD Name: disk0s1
Content: EFI
BSD Name: disk0s2
Content: Apple_APFS
Physical Interconnect: SATA
BSD Name: disk1
File System: MS-DOS FAT32
BSD Name: disk1s1
Content: EFI
File System: Journaled HFS+
BSD Name: disk1s2
Content: Apple_HFS
File System: Journaled HFS+
BSD Name: disk1s3
Content: Apple_Boot
Physical Interconnect: SATA
BSD Name: disk5
File System: MS-DOS FAT32
BSD Name: disk5s1
Content: EFI
File System: Journaled HFS+
BSD Name: disk5s2
Content: Apple_HFS
Physical Interconnect: SATA
BSD Name: disk3
File System: Journaled HFS+
BSD Name: disk3s1
Content: Apple_HFS
File System: MS-DOS FAT32
BSD Name: disk3s2
Content: DOS_FAT_32
Physical Interconnect: SATA
BSD Name: disk4
File System: MS-DOS FAT32
BSD Name: disk4s1
Content: EFI
File System: Journaled HFS+
BSD Name: disk4s2
Content: Apple_HFS
File System: Journaled HFS+
BSD Name: disk4s3
Content: Apple_Boot
PCI:
Type: NVM Express Controller
Bus: PCI
Slot: Slot-2
Name: NVDA,Display-D
Type: Display Controller
Bus: PCI
Slot: Slot-1
Name: pci10de,e1b
Type: NVDA,Parent
Bus: PCI
Slot: Slot-1
Type: USB eXtensible Host Controller
Bus: PCI
Slot: Slot-4
NVMExpress:
Generic SSD Controller:
Samsung SSD 970 EVO Plus 500GB:
Capacity: 500,11 GB (500.107.862.016 bytes)
TRIM Support: Yes
Model: Samsung SSD 970 EVO Plus 500GB
Revision: 2B2QEXM7
Link Width: x4
Link Speed: 5.0 GT/s
Detachable Drive: No
BSD Name: disk7
Partition Map Type: GPT (GUID Partition Table)
Removable Media: No
S.M.A.R.T. status: Verified
Volumes:
EFI:
Capacity: 209,7 MB (209.715.200 bytes)
File System: MS-DOS FAT32
BSD Name: disk7s1
Content: EFI
disk7s2:
Capacity: 499,9 GB (499.898.105.856 bytes)
BSD Name: disk7s2
Content: Apple_APFS
USB:
USB 2.0 Bus:
Host Controller Driver: AppleUSBEHCIPCI
Product ID: 0x1006
Speed: Up to 480 Mb/s
Location ID: 0xfa200000 / 1
Product ID: 0x0001
Speed: Up to 12 Mb/s
Location ID: 0xfa230000 / 4
Product ID: 0x0265
Speed: Up to 12 Mb/s
Location ID: 0xfa210000 / 2
Product ID: 0x0221
Speed: Up to 1.5 Mb/s
Location ID: 0xfa220000 / 3
USB 2.0 Bus:
Host Controller Driver: AppleUSBEHCIPCI
Product ID: 0x1000
Speed: Up to 480 Mb/s
Location ID: 0xfd500000 / 1
USB Interface: 0
USB Bus:
Host Controller Driver: AppleUSBUHCIPCI
USB Bus:
Host Controller Driver: AppleUSBUHCIPCI
USB Bus:
Host Controller Driver: AppleUSBUHCIPCI
USB Bus:
Host Controller Driver: AppleUSBUHCIPCI
USB Bus:
Host Controller Driver: AppleUSBUHCIPCI
USB Bus:
Host Controller Driver: AppleUSBUHCIPCI
USB 3.0 Bus:
Host Controller Driver: AppleUSBXHCIFL1100
Product ID: 0x8140
Speed: Up to 5 Gb/s
Location ID: 0x00600000 / 2
USB3.0 Hub:
Product ID: 0x0616
Speed: Up to 5 Gb/s
Location ID: 0x00500000 / 4
Product ID: 0x25a2
Speed: Up to 5 Gb/s
Location ID: 0x00520000 / 7
USB Interface: 0
Product ID: 0x8142
Speed: Up to 480 Mb/s
Location ID: 0x00200000 / 1
Product ID: 0x12ab
Speed: Up to 480 Mb/s
Location ID: 0x00210000 / 8
USB2.0 Hub:
Product ID: 0x0610
Speed: Up to 480 Mb/s
Location ID: 0x00100000 / 5
Bluetooth:
Bluetooth Low Energy Supported: Yes
Handoff Supported: No
Instant Hot Spot Supported: No
Manufacturer: Cambridge Silicon Radio
Composite Class Of Device: 0x380104
Connected: No
Class of Device: 0x05 0x25 0x2594
Memory:
Upgradeable Memory: Yes
Size: 16 GB
Type: DDR3
Speed: 1066 MHz
Size: 16 GB
Type: DDR3
Speed: 1066 MHz
Size: 8 GB
Type: DDR3
Speed: 1066 MHz
Size: 8 GB
Type: DDR3
Speed: 1066 MHz
Last edited:
Used the default settings for .21. Works so far!
Edit: So I just did 20 warm boots alternating between Big Sur/Monterey. All successful! @Syncretic you are awesome! Thanks for everything! Genius.
Edit: So I just did 20 warm boots alternating between Big Sur/Monterey. All successful! @Syncretic you are awesome! Thanks for everything! Genius.
Code:
Warmboot
4D1FDA02-38C7-4A6A-9CC6-4BCCA8B30102:opencore-version REL-072-2021-08-02
51 0 0 0xe000 0xe000 AAA.LoadEarly.latebloom (0.21) 6C8398F6-162F-36B9-A430-D6D98797BF4A <14 5 3 2 1>
Timestamp Thread Type Activity PID TTL
Real model decoded from serial number: Mac Pro 2010 (MP5,1)
1 x Intel(R) Xeon(R) CPU X5680 @ 3.33GHz
[Removed by me]
Radeon RX 580
running at 11.5.2 20G95 Kernel 20.6.0
boot-args -lilubetaall -wegbeta agdpmod=pikera shikigva=80 unfairgva=1 mbasd=1 -wegtree -no_compat_check no32exec=0 -v latebloom=30 lb_range=10 lb_debug=1 lb_delay2=200 lb_range2=50
Hardware:
Hardware Overview:
Model Name: Mac Pro
Model Identifier: MacPro5,1
Processor Name: 6-Core Intel Xeon
Processor Speed: 3.33 GHz
Number of Processors: 1
Total Number of Cores: 6
L2 Cache (per Core): 256 KB
L3 Cache: 12 MB
Hyper-Threading Technology: Enabled
Memory: 48 GB
System Firmware Version: 9144.0.7.2.0
SMC Version (system): 1.39f11
SMC Version (processor tray): 1.39f11
System Software Overview:
System Version: macOS 11.5.2 (20G95)
Kernel Version: Darwin 20.6.0
Network:
Type: Ethernet
Hardware: Ethernet
Service Order: 0
Type: Ethernet
Hardware: Ethernet
Service Order: 1
Type: FireWire
Hardware: FireWire
Service Order: 2
Type: AirPort
Hardware: AirPort
Service Order: 3
Type: Ethernet
Hardware: Ethernet
Service Order: 4
Type: Ethernet
Hardware: Ethernet
Service Order: 5
Wi-Fi:
Card Type: AirPort Extreme (0x14E4, 0x111)
Firmware Version: Broadcom BCM43xx 1.0 (7.77.111.1 AirPortDriverBrcmNIC-1680.8)
Storage:
File System: Journaled HFS+
BSD Name: disk0s2
Medium Type: SSD
Protocol: SATA
Internal: Yes
File System: NTFS
BSD Name: disk2s2
Medium Type: SSD
Protocol: SATA
Internal: Yes
File System: APFS
BSD Name: disk4s1
Medium Type: SSD
Protocol: SATA
Internal: Yes
File System: APFS
BSD Name: disk5s1
Medium Type: SSD
Protocol: SATA
Internal: Yes
File System: APFS
BSD Name: disk5s3
Medium Type: SSD
Protocol: SATA
Internal: Yes
File System: APFS
BSD Name: disk8s1
Medium Type: SSD
Protocol: PCI-Express
Internal: Yes
File System: Journaled HFS+
BSD Name: disk6s2
Medium Type: Rotational
Protocol: SATA
Internal: Yes
File System: Journaled HFS+
BSD Name: disk9s1
Protocol: USB
Internal: No
File System: APFS
BSD Name: disk8s7s1
Medium Type: SSD
Protocol: PCI-Express
Internal: Yes
File System: Journaled HFS+
BSD Name: disk10s3
Protocol: FireWire
Internal: No
Physical Interconnect: SATA
Physical Interconnect: SATA
BSD Name: disk0
File System: MS-DOS FAT32
BSD Name: disk0s1
Content: EFI
File System: Journaled HFS+
BSD Name: disk0s2
Content: Apple_HFS
Physical Interconnect: SATA
BSD Name: disk1
File System: MS-DOS FAT32
BSD Name: disk1s1
Content: EFI
BSD Name: disk1s2
Content: Apple_APFS
Physical Interconnect: SATA
BSD Name: disk2
File System: MS-DOS FAT32
BSD Name: disk2s1
Content: EFI
File System: NTFS
BSD Name: disk2s2
Content: Microsoft Basic Data
Physical Interconnect: SATA
BSD Name: disk6
File System: MS-DOS FAT32
BSD Name: disk6s1
Content: EFI
File System: Journaled HFS+
BSD Name: disk6s2
Content: Apple_HFS
File System: Journaled HFS+
BSD Name: disk6s3
Content: Apple_Boot
Physical Interconnect: SATA
BSD Name: disk3
File System: MS-DOS FAT32
BSD Name: disk3s1
Content: EFI
BSD Name: disk3s2
Content: Apple_APFS
PCI:
Type: NVM Express Controller
Bus: PCI
Slot: Slot-2
Name: ATY,AMD,RadeonFramebuffer
Type: Display Controller
Bus: PCI
Slot: Slot-1
Type: Audio Device
Bus: PCI
Slot: Slot-1
Type: USB eXtensible Host Controller
Bus: PCI
Slot: Slot-4
NVMExpress:
Generic SSD Controller:
HP SSD EX900 500GB:
Capacity: 500.11 GB (500,107,862,016 bytes)
TRIM Support: Yes
Model: HP SSD EX900 500GB
Revision: S0614B0
Link Width: x4
Link Speed: 5.0 GT/s
Detachable Drive: No
BSD Name: disk7
Partition Map Type: GPT (GUID Partition Table)
Removable Media: No
S.M.A.R.T. status: Verified
Volumes:
EFI:
Capacity: 209.7 MB (209,715,200 bytes)
File System: MS-DOS FAT32
BSD Name: disk7s1
Content: EFI
disk7s2:
Capacity: 499.9 GB (499,898,105,856 bytes)
BSD Name: disk7s2
Content: Apple_APFS
USB:
USB 2.0 Bus:
Host Controller Driver: AppleUSBEHCIPCI
USB 2.0 Bus:
Host Controller Driver: AppleUSBEHCIPCI
Product ID: 0x9126
Speed: Up to 480 Mb/s
Location ID: 0xfd300000 / 1
USB Receiver:
Product ID: 0xc52b
Speed: Up to 12 Mb/s
Location ID: 0xfd330000 / 4
Product ID: 0x4102
Speed: Up to 12 Mb/s
Location ID: 0xfd320000 / 6
Product ID: 0x0012
Speed: Up to 12 Mb/s
Location ID: 0xfd310000 / 7
Product ID: 0x9226
Speed: Up to 12 Mb/s
Location ID: 0xfd360000 / 5
Product ID: 0x8508
Speed: Up to 480 Mb/s
Location ID: 0xfd350000 / 2
Product ID: 0x1105
Speed: Up to 12 Mb/s
Location ID: 0xfd340000 / 3
USB Bus:
Host Controller Driver: AppleUSBUHCIPCI
USB Bus:
Host Controller Driver: AppleUSBUHCIPCI
USB Receiver:
Product ID: 0xc52b
Speed: Up to 12 Mb/s
Location ID: 0x3a200000 / 1
USB Bus:
Host Controller Driver: AppleUSBUHCIPCI
USB Bus:
Host Controller Driver: AppleUSBUHCIPCI
USB Bus:
Host Controller Driver: AppleUSBUHCIPCI
USB Bus:
Host Controller Driver: AppleUSBUHCIPCI
USB 3.0 Bus:
Host Controller Driver: AppleUSBXHCIFL1100
USB3.1 Hub:
Product ID: 0x0620
Speed: Up to 5 Gb/s
Location ID: 0x00500000 / 3
Rugged USB-C:
Product ID: 0x108c
Speed: Up to 5 Gb/s
Location ID: 0x00540000 / 8
Rugged USB-C:
USB Interface: 0
USB3.0 Hub:
Product ID: 0x0616
Speed: Up to 5 Gb/s
Location ID: 0x00530000 / 7
Product ID: 0x4500
Location ID: 0x00300000
Bluetooth USB Host Controller:
Product ID: 0x828d
Location ID: 0x00330000
USB2.1 Hub:
Product ID: 0x0610
Speed: Up to 480 Mb/s
Location ID: 0x00100000 / 1
USB2.0 Hub:
Product ID: 0x0610
Speed: Up to 480 Mb/s
Location ID: 0x00130000 / 4
Bluetooth:
Bluetooth Low Energy Supported: Yes
Handoff Supported: Yes
Instant Hot Spot Supported: Yes
Manufacturer: Broadcom
Composite Class Of Device: 0x380104
Connected: Yes
Manufacturer: Apple (0x3, 0x31C)
Class of Device: 0x05 0x25 0x2594
EDR Supported: No
eSCO Supported: No
SSP Supported: No
Connected: No
Class of Device: 0x05 0x20 0x2580
Connected: No
Class of Device: 0x00 0x00 0x0000
Memory:
Upgradeable Memory: Yes
Size: 16 GB
Type: DDR3
Speed: 1333 MHz
Size: 16 GB
Type: DDR3
Speed: 1333 MHz
Size: 16 GB
Type: DDR3
Speed: 1333 MHz
Size: Empty
Type: Empty
Speed: Empty
Last edited:
I appreciate the effort and all the testing that's going into latebloom but at the end of the day it's a pointless kludge which avoids but doesn't address the actual problem (whatever it is!).
I've reverted to Catalina which will be supported for a few more years. I'll probably sell my beloved cMP over the next 12 months or retire it as a NAS.
I've reverted to Catalina which will be supported for a few more years. I'll probably sell my beloved cMP over the next 12 months or retire it as a NAS.
@plunger
How is your experience conducive to clarifying the circumstances of the race condition or to improving the latebloom kext? If it doesn’t add anything useful to that aim, why are you sharing it?
How is your experience conducive to clarifying the circumstances of the race condition or to improving the latebloom kext? If it doesn’t add anything useful to that aim, why are you sharing it?
You actually don't know what the problem is; you think it's a race condition. It's a bit like string theory in fundamental physics, a huge amount of (wasted) effort by lots of talented people over 20+ years down a likely dead end.
I think that improving the latebloom kext is a waste of time and effort. I think this thread should be shut down and the focus should shift to actually identifying and then solving the problem.
I think that improving the latebloom kext is a waste of time and effort. I think this thread should be shut down and the focus should shift to actually identifying and then solving the problem.
Last edited:
You are correct that we don't know the true cause of the problem, that latebloom is a kludge, and you're not alone in your assessment that the path of least resistance may well be to sell your old Mac and get a more modern (supported) one. Depending on how Apple prices the next generation of Mac Pro, and what it has "under the hood," that option may become attractive to a lot more people.
What you seem to have missed is that while latebloom provides an admittedly kludgy way for some users to continue using the latest MacOS on their aging systems, it's also an ongoing experiment that's trying to figure out what the underlying cause is. I'm not trying to improve latebloom per se; as a tool to get recalcitrant old Macs to boot Big Sur/Monterey, it's probably about as good as it's going to get. By adding new tweaks like the most recent "Phase 1/Phase 2" separation, I'm effectively feeling around the edges of the problem, trying to find something that will provide the insight we need to create a proper solution. We'll never know the cause of the problem if we don't look for it. I've spent countless hours poring over the public source code, my disassembled kernels and kexts, and debugger traces, looking for clues to what's causing this frustrating mess; the latebloom results posted in this thread provide additional data points.
A race condition/deadlock is the most likely culprit, given the evidence we have. It's easy to think that something about the pre-2013 hardware is simply too old, but the fact is that some modern Hackintoshes exhibit the same problems, so it's not confined to some Mac-specific oddity. And in a system as complex and diverse as MacOS, finding a race condition is devilishly difficult. It's true that this problem may be insurmountable, but again - we'll never know unless we try.
I respect your position that this may well be a dead end, and that you may be better served by abandoning your old Mac in favor of a newer system. You certainly could be right on both counts. However, as @PeterHolbrook has pointed out, simply naysaying and suggesting that "this thread should be shut down" does nothing toward gathering data points or solving the problem. If you have some insight to offer on what the problem is, or how better to find it, or how to fix it, I encourage you to share your thoughts - here in this thread, or in one of your own. However, if all you're offering are arguments why this is a pointless exercise, I'd prefer that you direct your energies elsewhere. I'd like to keep this thread focused on trying to locate and solve the problem - or at least finding ways to ameliorate the symptoms.
I've invested a lot of money and time into my 2010 Mac Pro and, by many comparisons, it still offers me amazing performance and customisation. I don't need a gaming computer with 160fps or one that can encode 4K video in 'x' faster than realtime. I would have happily stuck at 11.2.3 for the next 10 years or until my capacitors started to blow apart, but am incredibly thankful and appreciative to Syncretic and the other developers who are providing Latebloom as a functional work-around to maximise the value of our computers with 11.3+ and Monterey... It doesn't really matter how we get to a loaded and functioning OS. If never knowing the true nature of the holdups during boot-up is a 'thing', then it really doesn't matter if we can skirt around them like an adept athlete.
@Syncretic
Many thanks for your work, highly appreciated.
What surprising me, while I have 4 cMP (5,1) they all have the same issues discussed on this thread when uppgrading to 11.3. I am now starting experimenting with latebloom.
But, I also have 3 macMini (mid 2012) and 2 MacBook Pro's (mid 2012) and none of these 5 machines have had any issues with 11.3, 11.4 and 11.5. They work flawless with OC 0.7.2, no issues at all.
Any ideas why this is the case?
Many thanks for your work, highly appreciated.
What surprising me, while I have 4 cMP (5,1) they all have the same issues discussed on this thread when uppgrading to 11.3. I am now starting experimenting with latebloom.
But, I also have 3 macMini (mid 2012) and 2 MacBook Pro's (mid 2012) and none of these 5 machines have had any issues with 11.3, 11.4 and 11.5. They work flawless with OC 0.7.2, no issues at all.
Any ideas why this is the case?
latebloom v2.1
Still working good enough for me as I continue using BS 11.5.2 and Monterey Beta5 on all my cMP 5,1 Machines.
Still working good enough for me as I continue using BS 11.5.2 and Monterey Beta5 on all my cMP 5,1 Machines.
Code:
Warmboot
4D1FDA02-38C7-4A6A-9CC6-4BCCA8B30102:opencore-version REL-071-2021-07-02
56 0 0 0xe000 0xe000 AAA.LoadEarly.latebloom (0.21) 6C8398F6-162F-36B9-A430-D6D98797BF4A <14 5 3 2 1>
2021-08-27 18:08:19.150613-0400 0x74 Default 0x0 0 0 kernel: (latebloom) _____[ !!! *** latebloom *** !!! ]: Hook placed successfully. Count = 0 :: 30,10,1,200,50
Real model decoded from serial number: Mac Pro 2012 (MP5,1)
2 x Intel(R) Xeon(R) CPU X5690 @ 3.47GHz
CMxxxxxF4MG
NVIDIA GeForce GTX 680
running at 11.5.2 20G95 Kernel 20.6.0
boot-args keepsyms=1 debug=0x100 -lilubetaall latebloom=30 lb_range=10 lb_debug=1 lb_delay2=200 lb_range2=50 -no_compat_check
Hardware:
Hardware Overview:
Model Name: Mac Pro
Model Identifier: MacPro5,1
Processor Name: 6-Core Intel Xeon
Processor Speed: 3.46 GHz
Number of Processors: 2
Total Number of Cores: 12
L2 Cache (per Core): 256 KB
L3 Cache (per Processor): 12 MB
Hyper-Threading Technology: Enabled
Memory: 48 GB
System Firmware Version: 9999.999.999.999.999
SMC Version (system): 9.9999
SMC Version (processor tray): 1.39f11
System Software Overview:
System Version: macOS 11.5.2 (20G95)
Kernel Version: Darwin 20.6.0
Network:
Type: Ethernet
Hardware: Ethernet
Service Order: 0
Type: Ethernet
Hardware: Ethernet
Service Order: 1
Type: AirPort
Hardware: AirPort
Service Order: 2
Type: Ethernet
Hardware: Ethernet
Service Order: 3
Wi-Fi:
Card Type: AirPort Extreme (0x14E4, 0x8E)
Firmware Version: Broadcom BCM43xx 1.0 (5.106.98.102.31)
Storage:
File System: APFS
BSD Name: disk1s1
Medium Type: SSD
Protocol: SATA
Internal: Yes
File System: APFS
BSD Name: disk1s2
Medium Type: SSD
Protocol: SATA
Internal: Yes
File System: APFS
BSD Name: disk6s1
Medium Type: Rotational
Protocol: SATA
Internal: Yes
File System: APFS
BSD Name: disk9s1
Medium Type: SSD
Protocol: PCI-Express
Internal: Yes
File System: Journaled HFS+
BSD Name: disk2s2
Medium Type: Rotational
Protocol: SATA
Internal: Yes
File System: Journaled HFS+
BSD Name: disk4s2
Medium Type: Rotational
Protocol: SATA
Internal: Yes
File System: Journaled HFS+
BSD Name: disk5s1
Medium Type: Rotational
Protocol: SATA
Internal: Yes
File System: NTFS
BSD Name: disk7s3
Medium Type: SSD
Protocol: PCI-Express
Internal: Yes
File System: APFS
BSD Name: disk9s2s1
Medium Type: SSD
Protocol: PCI-Express
Internal: Yes
Physical Interconnect: SATA
Physical Interconnect: SATA
BSD Name: disk0
File System: MS-DOS FAT32
BSD Name: disk0s1
Content: EFI
BSD Name: disk0s2
Content: Apple_APFS
Physical Interconnect: SATA
BSD Name: disk2
File System: MS-DOS FAT32
BSD Name: disk2s1
Content: EFI
File System: Journaled HFS+
BSD Name: disk2s2
Content: Apple_HFS
Physical Interconnect: SATA
BSD Name: disk5
File System: Journaled HFS+
BSD Name: disk5s1
Content: Apple_HFS
BSD Name: disk5s2
Content: Microsoft LDM Metadata
BSD Name: disk5s3
Content: Microsoft Reserved
Physical Interconnect: SATA
BSD Name: disk4
File System: MS-DOS FAT32
BSD Name: disk4s1
Content: EFI
File System: Journaled HFS+
BSD Name: disk4s2
Content: Apple_HFS
Physical Interconnect: SATA
BSD Name: disk3
BSD Name: disk3s1
Content: Apple_APFS
PCI:
Name: NVDA,Display-D
Type: Display Controller
Bus: PCI
Slot: Slot-1
Name: pci10de,e0a
Type: NVDA,Parent
Bus: PCI
Slot: Slot-1
Type: NVM Express Controller
Bus: PCI
Slot: Slot-3
Type: NVM Express Controller
Bus: PCI
Slot: Slot-4
Type: USB eXtensible Host Controller
Bus: PCI
Slot: Slot-2@9,0,0
Type: USB eXtensible Host Controller
Bus: PCI
Slot: Slot-2@11,0,0
Type: USB eXtensible Host Controller
Bus: PCI
Slot: Slot-2@8,0,0
Type: USB eXtensible Host Controller
Bus: PCI
Slot: Slot-2@10,0,0
NVMExpress:
Generic SSD Controller:
Samsung SSD 970 PRO 1TB:
Capacity: 1.02 TB (1,024,209,543,168 bytes)
TRIM Support: Yes
Model: Samsung SSD 970 PRO 1TB
Revision: 1B2QEXP7
Link Width: x1
Link Speed: 5.0 GT/s
Detachable Drive: No
BSD Name: disk7
Partition Map Type: GPT (GUID Partition Table)
Removable Media: No
S.M.A.R.T. status: Verified
Volumes:
NO NAME:
Capacity: 104.9 MB (104,857,600 bytes)
File System: MS-DOS FAT32
BSD Name: disk7s1
Content: EFI
disk7s2:
Capacity: 16.8 MB (16,777,216 bytes)
BSD Name: disk7s2
Content: Microsoft Reserved
WIN11:
Capacity: 999.56 GB (999,557,169,152 bytes)
Free: 739.02 GB (739,019,014,144 bytes)
Writable: No
File System: NTFS
BSD Name: disk7s3
Mount Point: /Volumes/WIN11
Content: Microsoft Basic Data
Generic SSD Controller:
CT2000P2SSD8:
Capacity: 2 TB (2,000,398,934,016 bytes)
TRIM Support: Yes
Model: CT2000P2SSD8
Revision: P2CR033
Link Width: x4
Link Speed: 5.0 GT/s
Detachable Drive: No
BSD Name: disk8
Partition Map Type: GPT (GUID Partition Table)
Removable Media: No
S.M.A.R.T. status: Verified
Volumes:
EFI:
Capacity: 209.7 MB (209,715,200 bytes)
File System: MS-DOS FAT32
BSD Name: disk8s1
Content: EFI
disk8s2:
Capacity: 2 TB (2,000,189,177,856 bytes)
BSD Name: disk8s2
Content: Apple_APFS
USB:
USB 2.0 Bus:
Host Controller Driver: AppleUSBEHCIPCI
USB 2.0 Bus:
Host Controller Driver: AppleUSBEHCIPCI
Product ID: 0x1006
Speed: Up to 480 Mb/s
Location ID: 0xfd300000 / 2
Apple Optical USB Mouse:
Product ID: 0x0304
Speed: Up to 1.5 Mb/s
Location ID: 0xfd330000 / 4
Product ID: 0x0220
Speed: Up to 1.5 Mb/s
Location ID: 0xfd320000 / 3
Product ID: 0xa100
Speed: Up to 480 Mb/s
Location ID: 0xfd100000 / 5
USB Bus:
Host Controller Driver: AppleUSBUHCIPCI
USB Bus:
Host Controller Driver: AppleUSBUHCIPCI
Product ID: 0x2018
Speed: Up to 12 Mb/s
Location ID: 0x1a200000 / 1
USB Bus:
Host Controller Driver: AppleUSBUHCIPCI
Product ID: 0x4500
Location ID: 0x5a100000
Bluetooth USB Host Controller:
Product ID: 0x8215
Location ID: 0x5a110000
USB Bus:
Host Controller Driver: AppleUSBUHCIPCI
USB Bus:
Host Controller Driver: AppleUSBUHCIPCI
USB Bus:
Host Controller Driver: AppleUSBUHCIPCI
USB 3.0 Bus:
Host Controller Driver: AppleUSBXHCIFL1100
USB 3.0 Bus:
Host Controller Driver: AppleUSBXHCIFL1100
USB3.0 Hub :
Product ID: 0x8110
Speed: Up to 5 Gb/s
Location ID: 0x01800000 / 2
USB2.0 Hub :
Product ID: 0x2811
Speed: Up to 480 Mb/s
Location ID: 0x01400000 / 3
Product ID: 0x6204
Speed: Up to 12 Mb/s
Location ID: 0x01440000 / 4
USB2.0 Hub :
Product ID: 0x2811
Speed: Up to 480 Mb/s
Location ID: 0x01430000 / 6
Product ID: 0x190f
Speed: Up to 480 Mb/s
Location ID: 0x01433000 / 7
USB 3.0 Bus:
Host Controller Driver: AppleUSBXHCIFL1100
USB 3.0 Bus:
Host Controller Driver: AppleUSBXHCIFL1100
Product ID: 0x55aa
Speed: Up to 5 Gb/s
Location ID: 0x03800000 / 1
Bluetooth:
Bluetooth Low Energy Supported: No
Handoff Supported: No
Instant Hot Spot Supported: No
Manufacturer: Broadcom
Composite Class Of Device: 0x380104
Connected: No
Manufacturer: Broadcom (0x5, 0x240C)
Class of Device: 0x05 0x20 0x0580
EDR Supported: Yes
eSCO Supported: No
SSP Supported: Yes
Memory:
Upgradeable Memory: Yes
Size: 16 GB
Type: DDR3
Speed: 1333 MHz
Size: 16 GB
Type: DDR3
Speed: 1333 MHz
Size: 16 GB
Type: DDR3
Speed: 1333 MHz
Size: Empty
Type: Empty
Speed: Empty
Size: Empty
Type: Empty
Speed: Empty
Size: Empty
Type: Empty
Speed: Empty
Size: Empty
Type: Empty
Speed: Empty
Size: Empty
Type: Empty
Speed: Empty- Status