Become a MacRumors Supporter for $50/year with no ads, ability to filter front page stories, and private forums.
Status
The first post of this thread is a WikiPost and can be edited by anyone with the appropiate permissions. Your edits will be public.
Status
Not open for further replies.

eVasilis

macrumors 6502
Jan 13, 2010
425
182
Unfortunately even after replacing both kexts I still have ionvme panic if my expansion board with the OWC PCIE is attached. Also it hangs if ethernet is connected. This tells me that Ionvme is just a red Herring. It needs to be tested by more people though.
Did you try replacing all three kexts, IOAHCIFamily, IONVMeFamily and IOPCIFamily?
 

startergo

macrumors 603
Sep 20, 2018
5,022
2,283
For those in this thread who have "trust issues" with what other members write about PCIe booting (on a cMP 3,1):
I noticed something odd. The startup disk is not in about this Mac. Is it because the PCIE is installed in slot1? Not sure.
 

startergo

macrumors 603
Sep 20, 2018
5,022
2,283
Replaced IOAHCIFamily.kext, but it does not fix the problem.
Attempted to replace both IO80211Family.kext and IO80211FamilyV2.kext, but symbols don't match:
Code:
sudo kmutil install --volume-root /System/Volumes/Update/mnt1/ --update-all
checking collections...
updated extension 'com.Areca.ArcMSR' found on disk at '/System/Volumes/Update/mnt1/System/Library/Extensions/ArcMSR.kext' (mach-o UUID changed from <unknown> to 01FC5EBD-0A34-3E77-9216-53B89DDD8CE4)
updated extension 'com.apple.driver.usb.AppleUSBVHCICommonRSM' found on disk at '/System/Volumes/Update/mnt1/System/Library/Extensions/AppleUSBCommon.kext/Contents/PlugIns/AppleUSBVHCICommonRSM.kext' (mach-o UUID changed from <unknown> to D2DACBE0-CDE2-3D7A-8AA4-632130700764)
updated extension 'com.apple.driver.AirPort.BrcmNIC' found on disk at '/System/Volumes/Update/mnt1/System/Library/Extensions/IO80211Family.kext/Contents/PlugIns/AirPortBrcmNIC.kext' (mach-o UUID changed from D75BF2E4-37CF-3131-AEA1-92D089E0092E to 729DC69F-0B01-3807-B946-AE96C4584DA1)
updated extension 'com.promise.driver.stex' found on disk at '/System/Volumes/Update/mnt1/System/Library/Extensions/PromiseSTEX.kext' (mach-o UUID changed from <unknown> to AE253556-66D8-38D2-B9D1-179B78F29153)
updated extension 'com.apple.driver.IO80211NetBooter' found on disk at '/System/Volumes/Update/mnt1/System/Library/Extensions/IO80211Family.kext/Contents/PlugIns/IO80211NetBooter.kext' (mach-o UUID changed from F99B53F8-BA4B-37FE-BBFE-4D86A7BF5B92 to 56B5847C-8C9E-33D0-AA4C-027AB3618940)
updated extension 'com.apple.iokit.IO80211Family' found on disk at '/System/Volumes/Update/mnt1/System/Library/Extensions/IO80211Family.kext' (mach-o UUID changed from 0FEA34AB-B3C6-36E4-A63B-3FB8611D6F60 to 27EDE1C0-32A1-3BC9-A93A-49B4E7AD2167)
updated extension 'com.apple.iokit.IO80211FamilyV2' found on disk at '/System/Volumes/Update/mnt1/System/Library/Extensions/IO80211FamilyV2.kext' (mach-o UUID changed from 13EF976B-6D04-33F3-B9F4-46178D344179 to FA0689F9-1CFE-351D-8B83-117E0BEEB442)
updated extension 'com.apple.driver.IO80211NetBooterV2' found on disk at '/System/Volumes/Update/mnt1/System/Library/Extensions/IO80211FamilyV2.kext/Contents/PlugIns/IO80211NetBooterV2.kext' (mach-o UUID changed from 3B7F5205-F857-310A-94FF-556BF7D51C33 to BF28380F-8530-3A49-BB66-1E78C2FAFDBB)
rebuilding release collections: boot, system
rebuilding release collections:
    boot kernel collection
    system kext collection
failed to build release collection:
    One or more binaries has an error which prevented linking.  See other errors.
    Could not use 'com.apple.driver.AppleBCMWLANCoreMac' because: Failed to bind '__ZN16IO80211Interface23setInterfaceOFDMDesenseE18apple80211_channelx' in 'com.apple.driver.AppleBCMWLANCoreMac' (at offset 0x7312b in __TEXT, __text) as could not find a kext which exports this symbol
    com.apple.driver.AppleBCMWLANCoreMac:
        Failed to bind '__ZN16IO80211Interface23setInterfaceOFDMDesenseE18apple80211_channelx' in 'com.apple.driver.AppleBCMWLANCoreMac' (at offset 0x7312b in __TEXT, __text) as could not find a kext which exports this symbol
Error Domain=KMErrorDomain Code=31 "Error occurred while building a collection:
    1: One or more binaries has an error which prevented linking.  See other errors.
    2: Could not use 'com.apple.driver.AppleBCMWLANCoreMac' because: Failed to bind '__ZN16IO80211Interface23setInterfaceOFDMDesenseE18apple80211_channelx' in 'com.apple.driver.AppleBCMWLANCoreMac' (at offset 0x7312b in __TEXT, __text) as could not find a kext which exports this symbol
com.apple.driver.AppleBCMWLANCoreMac specific:
    1: Failed to bind '__ZN16IO80211Interface23setInterfaceOFDMDesenseE18apple80211_channelx' in 'com.apple.driver.AppleBCMWLANCoreMac' (at offset 0x7312b in __TEXT, __text) as could not find a kext which exports this symbol
" UserInfo={NSLocalizedDescription=Error occurred while building a collection:
    1: One or more binaries has an error which prevented linking.  See other errors.
    2: Could not use 'com.apple.driver.AppleBCMWLANCoreMac' because: Failed to bind '__ZN16IO80211Interface23setInterfaceOFDMDesenseE18apple80211_channelx' in 'com.apple.driver.AppleBCMWLANCoreMac' (at offset 0x7312b in __TEXT, __text) as could not find a kext which exports this symbol
com.apple.driver.AppleBCMWLANCoreMac specific:
    1: Failed to bind '__ZN16IO80211Interface23setInterfaceOFDMDesenseE18apple80211_channelx' in 'com.apple.driver.AppleBCMWLANCoreMac' (at offset 0x7312b in __TEXT, __text) as could not find a kext which exports this symbol
}
 

tsialex

Contributor
Jun 13, 2016
13,455
13,601
For those in this thread who have "trust issues" with what other members write about PCIe booting (on a cMP 3,1):
Seems you don't get my point, until someone reproduces your find, it's dubious/not proved from a scientific stand point. Without a dump to understand and reproduce how your BootROM was created and prove it that is booting reliably while several others can't, it's useless for our diagnostic.
 

nekton1

macrumors 65816
Apr 15, 2010
1,094
777
Asia
Seems you don't get my point, until someone reproduces your find, it's dubious/not proved from a scientific stand point. Without a dump to understand and reproduce how your BootROM was created and prove it that is booting reliably while several others can't, it's useless for our diagnostic.
Have it your own way tsialex. I think others here (Pri-est maybe) have a reliably booting cMP3,1. When I have time this week, I'll build a High Sierra boot disk and dump the bootrom with ROMtool for your science (in which I have spent 40+ years working) experiment.
 

startergo

macrumors 603
Sep 20, 2018
5,022
2,283
iMac 2020 with thunderbolt on BS beta:
 
  • Like
Reactions: JohnD and cdf

startergo

macrumors 603
Sep 20, 2018
5,022
2,283
A glimmer of hope that Apple will fix this issue perhaps? Too bad they didn't get a PROHIBITED symbol. :p I hope they report their issue, and I hope it relates to our issue, though Apple's fix could further break our machines.
Yes it looks like they will not fix it for the next release and there will be a lot of screaming.when the systems get updated.
 

JohnD

macrumors regular
Jun 2, 2005
150
97
Los Angeles, California
So, with all the PCIe issues 11.3/11.4 created for us - from NVMe, to SATA SSD on PCIe, to Thunderbolt, etc. - why is nobody seemingly having any issues with the GPU? I've run GPU benchmarks and stress tests for hours with no issue. I've also copied several TB's of data between my SATA SSD on PCIe and internal HDD's, while booted off that PCIe SSD, and not had any kernel panics. The ONLY problem I've had with any 11.3 beta or 11.4b1 is booting. Maybe I should install a USB3 card and test further, but my CalDigit card doesn't work in Big Sur, and that's my only spare.
 
  • Like
Reactions: LucMac and Stex

eVasilis

macrumors 6502
Jan 13, 2010
425
182
... - why is nobody seemingly having any issues with the GPU? I've run GPU benchmarks and stress tests for hours with no issue...
I can't even pretend I have the knowledge most of the people in this forum have; I just like to tinker. Still, this is, to my understanding, a valid point: if there is an incompatibility with PCIe in cMPs, it will definitely have an impact on GPUs. So, what if the problem is in the way 11.3 and 11.4 handle storage and kexts such as IOStorageFamily?

Just a thought.
 
Last edited:

PeterHolbrook

macrumors 68000
Sep 23, 2009
1,625
441
Anyone tested yet 11.3?
Currently installing it on a Parallels 16.1.2 virtual machine. Not interesting for you, I'm afraid. I'm also looking forward to hearing some news about Titan-Ridge Thunderbolt compatibility on a Mac Pro 5,1 that boots Big Sur off an HDD.
 

JohnD

macrumors regular
Jun 2, 2005
150
97
Los Angeles, California
Anyone tested yet 11.3?
I wiped my 11.4b1 PCIe SSD in the test machine, and am currently doing a fresh install of 11.2.3, from which I'll upgrade to 11.3GM. Not expecting anything different, but it'll be a fresh install that I can play with swapping KEXT's, etc. I must say, it's nice having a spare machine with drives/OS's you don't care about corrupting, on a 2nd monitor, while working from home. :) EDIT: That is, on cool days that I'm not paying for air conditioning... we have a short heat wave coming in the next couple days... 96F/35.5C - won't have any excess equipment running!
 

eVasilis

macrumors 6502
Jan 13, 2010
425
182
I wiped my 11.4b1 PCIe SSD in the test machine, and am currently doing a fresh install of 11.2.3, from which I'll upgrade to 11.3GM. Not expecting anything different, but it'll be a fresh install that I can play with swapping KEXT's, etc. I must say, it's nice having a spare machine with drives/OS's you don't care about corrupting, on a 2nd monitor, while working from home. :)
Would a clean install be any different?
 

cdf

macrumors 68020
Jul 27, 2012
2,256
2,583
Downloading it now to do a straight upgrade from 11.2.3, NVMe installed on Syba PCIe slot 2. See signature for model.
Use a test installation of 11.2.3 or make sure to have a back up before proceeding.
 

JohnD

macrumors regular
Jun 2, 2005
150
97
Los Angeles, California
I performed a clean install of 11.2.3 to a SATA SSD on PCIe card, which went well as expected. I then upgraded it to 11.3, which didn't go well as expected. I lost count of how many times the progress bar got stuck (I have a post-it with a marker stuck on the screen so I know when it gets to that point), and got several prohibited symbols. I finally got to the desktop! What next? Replace some KEXT's with 11.2.3's KEXT's per the GitHub link? @startergo had symbol issues with IO80211Family.kext and IO80211FamilyV2.kext

 

TECK

macrumors 65816
Nov 18, 2011
1,129
478
Use a test installation of 11.2.3 or make sure to have a back up before proceeding.

I'm running OC 0.6.8 with the Plistlib Generator setup.py example configuration (see #348 for my config.plist).

From 11.2.3, I updated to 11.3 with Software Update. It got stuck once at the progress bar (see below why) forcing me to manually shutdown from the power button. After power-up, it went through installation process rebooting twice and I was presented with the 11.3 setup screen.

After installation, I determined the following pattern for my machine: If there are no computer problems previously detected, the reboot will get stuck at Progress Bar, no matter what. (that explains why upon initial upgrade, Progress Bar got stuck). In other words, everything works if I'm welcomed by the "you shut down your computer because of a problem." error. Once I'm logged in 11.3 and I try to shutdown the Mac (which in 11.2.3 would fix this error message), I'm stuck at Progress Bar boot screen. As long as I don't shutdown, which will reset the shut down error, I can reboot fine many times (I tested the reboot 10 times). I also have a BootROM reconstructed by @tsialex, which I refreshed one month ago.

IMG_4779.JPG IMG_4780.JPG

I'm able to constantly repeat this exact pattern every time. Anything else is functional in 11.3, I'm able to login with my Apple Watch, perform AirDrop, Wifi, Bluetooth, etc. See my signature for all hardware upgrades.

In my case, I might have better success than others because the chip used into Syba SI-PEX40129 (see #341 for chip details), which is the same as HighPoint 7101A and reported to have proper NVMe functionality with Big Sur 11.3. See also #337 for Sonnet Allegro USB3C-2PM-E PCIe card test.

Posting now from 11.3, these screenshots:

1619477581162.png


Screen Shot 2021-04-26 at 6.56.40 PM.png


Screen Shot 2021-04-26 at 6.53.50 PM.png


Screen Shot 2021-04-26 at 6.54.33 PM.png


Screen Shot 2021-04-26 at 7.55.31 PM.png
 
Last edited:
Status
Not open for further replies.
Register on MacRumors! This sidebar will go away, and you'll see fewer ads.