Jump to content


  • Content count

  • Joined

  • Last visited

About mchaggis

  • Rank
    InsanelyMac Protégé
  1. Sorry, I had to send my Mac Pro away for repair with a PSU problem so I won't be able to test your script for now. I'll hopefully get to try it on a non-critical disk when I get the machine back.
  2. It seems to be working on my main disk now too. The process was similar, but this time I tried not installing the MS AHCI drivers before laying the patch. This resulted in a BSOD (no driver, I presume). I then had to re-lay the original MBR, boot back into Windows and install the MS AHCI drivers (I did it only 1 of the 2 SATA devices shown, the one with my HD attached to it). Then I re-relaid the patched MBR, booted into Windows and it detected the AHCI device. From then on it was plain-sailing, I just repeated the process for the second SATA controller. For some reason my 09 Mac Pro presents a 2-port SATA controller and a 4-port SATA controller separately, which is why I had to do it twice. For reference this time I laid the MBR patch using a linux LiveCD and dd. The patch was the 440 byte version I modified (by removing the last 6 bytes), rather than the 446 byte original patch from johnsock. The main stumbling block seems to be Windows drivers, and getting everything set up in the right order so Windows knows to use the right one after you transition to AHCI mode. Edit: I've since realised that the '2-port SATA controller' has not been picked up as SATA/AHCI, it still seems to be operating in legacy mode. I presume the original code was written to enable one device which controls all 4 internal HD ports AND the 2 optical ports on the older Macs; wheras Apple seems to have put 2 controllers on the 09 Mac Pro - one for the internal bays and one for the optical ports. That sound plausible? This probably contributed to my BSOD problems when I tried to install it the second time - I may have installed the MS AHCI driver on the wrong one of the 2 available ports on my first attempt.
  3. I got it working too, but using the 'original' MBR (good_ahci.mbr) from johnsock. I didn't do it exactly like in the post above though. Comparing the MBRs from 3 different Vista installs on different disks (johnsock's plus 2 of mine: one Bootcamp with OSX on the same disk; the other no Bootcamp, straight Vista) I found they were identical with the exception of the final 6 bytes (0x0 to 0x1B7 were identical). So I dd'd the first 0x1B8 (440 bytes) of johnsock's code onto /dev/diskX, not rdisk as suggested - when I umounted the disk in question /dev/rdiskX was no longer available. Is there a way to dd onto a mounted disk in MacOS? It told me the resource was unavailable until I umounted it. The first attempt gave a BSOD (but too fast to see the code). Before trying this I'd installed the AHCI driver as suggested. I then tried again with 'Last Good Conf' so the AHCI driver was no longer installed. This worked! It detected the AHCI devices at startup, then I loaded the Intel Storage Matrix driver and no complaints. ISM when opened shows a tree for the AHCI device with all my attached SATA drives underneath it, including NCQ enabled. I'm going to try it on the other disk (MacOS + Vista BootCamp) soon, but I'm going to get a problem dd'ing to the active disk. Is this why you use fdisk by any chance? If you're wondering, I used dd because I know exactly what it does - I wasn't sure if fdisk might try to overwrite the whole MBR with only 440 bytes of data and lose the bytes which seem to be unique per disk. More to come...
  4. johsock: yes the 09 Mac Pro does use ICH10 (http://www.intel.com/Assets/PDF/prodbrief/xeon-5500.pdf). Is there a reason for having to do so much installing of drivers and restarting? Is it just because the Intel SMB driver comes as an app and will only install if it can see the correct device? Looking forward to trying this
  5. Firstly, thanks a million to ludacrisvp for the OP, it's a beacon of light I haven't tried it yet though for 2 reasons: firstly it would be for my work computer and I can't afford the time either to do the procedure or for it not to work right now. Secondly, I have one of the new 09 Mac Pros so I wasn't sure if the GRUB module would work with the new chipset/southbridge. Any ideas? And to johnsock - your method certainly sounds a lot simpler. If you have the time I'm sure I wouldn't be the only one offering to have your proverbial babies if you could come up with something relatively usable If either of you (or anyone else) has something working and want to test it on an 09 Mac Pro, feel free to get in touch and I can test it for you. As an aside: I installed Vista x64 without using Bootcamp (partitioned as GPT) and it booted fine, but still no AHCI. Can someone explain at what point the SATA controller is put into legacy mode in this config? It would seem to me (as a techy but without much knowledge of EFI and bootloaders) that Mac EFI must default to legacy mode but MacOS X must enable it during boot. Isn't this weird? Would there be any way of getting rEFIt to enable AHCI mode prior to a boot loader running? Like I say, I don't know a huge amount about this but if it helps, great...