Mac Pro 3,1, PCIe SSD Boot Problems

[pacmanz]

Hello,

Wondering if someone could advise how to troubleshoot the following.

I updated from El Capitan (not having the following problem), to Sierra (I'm on an unsupported machine for this OS, but used dosdudes patch, aside from the following issue everything works great). I seem to only be able to option key boot into my SSD, and get the circle with slash icon on straight boot up. I have reset NVRAM, I have replaced the NVRAM battery, Control selected the drive on option key boot, selected the drive in Startup Manager, and disconnected and reconnected the PCIe card (slot 2, directly above GPU in slot 1). I have removed all other drives and optical drives. The SSD (Crucial) is mounted on the Apricorn Velocity Solo x1 PCIe card. I also have a flashed GTX 680 2GB GPU installed. On option key boot, everything runs fine and the machine performs great.

I know the 3,1 PCIe was always a bit flaky but not clear what might have changed going from El Cap to Sierra and/or if there is a way via terminal or verbose mode to troubleshoot this better or provide you folk the information to help me troubleshoot it better. Crucial's site is currently down for maintenance but I'm going to doublecheck to confirm I have the most current firmware on my SSD as soon as I can.

Is it possible that this just won't work?

 

[harryb2448]

When I had a Mac Pro and looked at this, the suggestion was one could do one and one only kext update. Suggest inquiring over at Mac Rumours where they are more into Hackintoshes than here.

 

[pacmanz]

Hmmm, yea. I wonder if this has to do with the AppleAHCIPort.kext. This is all way above my head at this point, I'd imagine it's not safe to just yank the kext file from El Capitan and swap it out with the one from Sierra? I'll keep digging though.

 

[pacmanz]

Checked the .fsck_hfs.log after boot up and saw the following:

fsck_hfs.log

/dev/rdisk4s2: fsck_hfs started at Tue Jan 15 16:10:28 2019
/dev/rdisk4s2: /dev/rdisk4s2: ** /dev/rdisk4s2 (NO WRITE)
/dev/rdisk4s2: Executing fsck_hfs (version hfs-366.70.3).
/dev/rdisk4s2: fsck_hfs completed at Tue Jan 15 16:10:28 2019

** /dev/rdisk3s2 (NO WRITE)
/dev/rdisk3s2: fsck_hfs started at Tue Jan 15 16:10:28 2019
/dev/rdisk3s2: /dev/rdisk3s2: ERROR: volume / is mounted with write access. Re-run with (-l) to freeze volume.
/dev/rdisk3s2: fsck_hfs completed at Tue Jan 15 16:10:28 2019

Where disk4s2 is the one the machine should be booting into, it's booting to disk3s2 instead. Anyone know if that NO WRITE error is important?