VT-X disabled by firmware update?

So, what is the official word from Parallels on this at this point? Surely they are looking into this?

I will let my mac go to sleep on it's own tomorrow and give that shot. There is still hope!!

 

mattham, my experience is much like yours. I have seen VT both on and off. One time was after a hard crash. Since then, it now only wants to do Software Mode 2....whatever that is.

 

so what are the different modes?

I have software mode 0

 

Hi All!

Thank you very much for helping us with this issue! We are filed a bug about Intel VT-x behavior on Mac Minis on bugreport.apple.com and we will keep you posted about its status.

 

Just tried a little experiment....

bought a new mac mini and upgraded it to 10.4.6 through Software update, installed Parallels Beta 2, and set up my XP session. Vt-x mode was off all through this. Updated firmware, tried the sleep trick, still running in software.

Then I fired up my XP session in Parallels, waited until XP had fully loaded and had no disk activity, then I yanked the power cable from the mini. After powering up and going back into my XP session, I now have VT-x running!

caveat emptor, YMMV, I am not responsible for damage to your Mac, etc.

 

Solo core mac Mini. 1 GB memory. I moved a MacBook pro image (winxp folder) and VT-x was working. Did an 10.4.6 update and sometime after the update, I start getting the message Firmware lockout of VT-x. XP was much slower.

Needed a bigger drive so I reinstalled on the MBP and copied the winxp folder to the mini mac. I now have VT-x working again.

 

Sleep tricked worked

Machine Name: Mac mini
Machine Model: Macmini1,1
CPU Type: Intel Core Solo
Number Of Cores: 1
CPU Speed: 1.5 GHz
Memory: 512 MB
Bus Speed: 667 MHz
Boot ROM Version: MM11.0055.B02
Serial Number: YM61204PU38
SMC Version: 1.3f2
L2 Cache: 2 MB

I started with PW 2.1 beta 1 before the firmware update but after 10.4.6..."VT-x is supported (reported by CPUID), but locked in MSR and can't be used" were results from VTX-check. I got the same results after the firmware update. I just installed beta 2 last night and paralles said that it was still using software 0 I think. Just installed beta3 just now. I wasn't thinking because I wastrying to install beta3 while xp installer was running in the VM. Needless to say I got the gray crash message saying that I need to restart my computer. So I did and then re-ran the beta3 installer. I was reading the forums when XP installer was going again so I went to the apple menu and put my machine to sleep. When it woke up I ran vtx-check and it reports enabled vt-x. Of course I didn't check before I put the machine to sleep. Anyway I hope this helps. I've got a feeling that it might have something to do with specific serials.
Nathan

 

This worked for me, too! I think my timeline is something like this:

• beta 1, worked with VT-X on [empirically; I didn't know then to look to make sure, but subsequently when I knew it was off, it was slower, so I'm going to assume that at this point it was on]

• tried to install beta 2. The installer went into an infinite loop, and was cramming my system.log with messages like the below at an alarming rate, so I had to force quite the installer:
  Apr 9 11:42:11 imini kernel[0]: Error:Unsupported il() call (0xc0105401) to hypervisor (hypervisor.c:122)
  Apr 9 11:42:11 imini kernel[0]: Error:Unsupported ioctl() call (0xc0105401) to hypervisor (hypervisor.c:122)
  Apr 9 11:42:11 imini kernel[0]: ctl() call (0xc0105401) to hypervisor (hypervisor.c:122)
  Apr 9 11:42:11 imini kernel[0]: or:Unsupported ioctl() call (0xc0105401) to hypervisor (hypervisor.c:122)
  Apr 9 11:42:11 imini kernel[0]: or:Unsupported ioctl() call (0xc0105401) to hypervisor (hypervisor.c:122)
  Apr 9 11:42:11 imini kernel[0]: Error:Unsupported ioctl() call (0xc0105) to hypervisor (hypervisor.c:122)
  Apr 9 11:42:11 imini kernel[0]: Error:Unsupported ioctl() call (0xc0101) to hypervisor (hypervisor.c:122)
  Apr 9 11:42:11 imini kernel[0]: Error:Unsupported ioctl() call (0xc0105401) to hypisor (hypervisor.c:122)
  Apr 9 11:42:11 imini kernel[0]: Error:Unsupport)ll (0xc0105401) to hypervisor (hypervisor.c:122)
  Apr 9 11:42:11 imini kernel[0]: Error:Unsupported ioctl() call (0xc0105401) to hyvisor (hypervisor.c:122)
  Apr 9 11:42:11 imini kernel[0]: Error:Unsupported ioctl() call (0xc0105401) to hypervisor (hypervisor.c:122)

• retrying a few times got similar problems, so I removed all references to Parallels, and installed again. This time it went without a hitch

• my initial attempts to run beta 2 were met with various crashes; panics; hung system; spontaneous reboot. Eventually I got it running smoothly. At this point I noticed it was obviously slower than beta 1 [albeit still perfectly fast enough for my requirements]. I had noticed the 'VT-X disabled by BIOS' message, but I assumed that that was just a new message for beta 2... after all, I figured why would something in the BIOS suddenly not work!

• eventually came to the conclusion that it wasn't just a new message, and installed the Mac mini firmware upgrade, expecting that to make VT-X work, and of course it didn't

• installed beta 3, thinking maybe that will fix it, but of course that didn't either. Then I found this forum!

• I shutdown the VM, then Parallels and all other apps, and pulled the powercord, and indeed VT-X is now functioning again, and performance is back to how it was when I first tried beta 1

I am speculating, but I wonder if the crashes I had with beta 2 caused VT-X to be disabled somehow, and that the Mac's firmware is not correctly resetting it, even after a soft restart. I would think that just shutting down and pulling power then would have also worked, rather than risking pulling the power while the Mac is still running.

 

Is there anyone here who got the Parallels Workstation to run natively (with u-processor support) on a miniMac w/ Core Solo? (even if you had to use the weird tricks)

I ask simply because I want the cheapest Windows-on-Mac system. As some people had mentioned before, restarting just to use one or two Windows apps sounds really irritating, this software sounds like it is really good when it's got u-proc. support! I was even considering looking for one of those old "Orange-PC" cards for my G4 tower to make Virtual PC run faster!

Thanks in advance for any responses.

 

I don't think the Core Solo supports VT-x. I believe only the dual core cpus do. Also there's some confusion based on what you read where that the Yonah based Core Duo is not supposed to support VT-x, but it clearly does, or at least some steppings of it do. Certainly Merom based Core Duo cpus will, and in my experience the current Core Duo does as well.

 

I have gotten VT-x working several times but after an upgrade on panic I lose it. Three time I have copied the image from my MBP and it start working on the Core Solo.

 

Hmm. So much for my theory. I dug around on Intel's web site and found a spec sheet for both Core Duo and Core Solo and it mentions Intel Virtualization without distinguishing between the two so I guess Core Solo supports it.

 

i presume you guys have seen this thread

http://forum.parallels.com/thread577.html

Matt

 

disabled firmware

I suspect the firmware in the Intel Mini has a disabled-on-purpose VT, just like the iBook, iMac, Mini, and eMac have only mirrored external monitor support. Apple disabled the extended desktop that is available on PowerMacs and PowerBooks, to differentiate between the consumer and professional lines. My 14" iBook does just fine with an extrnal monitor running at 1280x1024 while the 14" has a 1024x786 or nothing at all on it. The firmware was there, and could be turned on with unix. An enterprising programmer put together an Applescript hack that restored the settion, and I downloaded ScreenSpanningDoctor, ran the script, and everything works nicely.

I suspect if Apple has set the mini that way, it won't be long before someone figures out how to fix it!

 

what are people complaining about, i have 2 Mini's both updated and VT-x is working on Both, I even used the RECOVERY CD to downgrade one and reinstalled it again , same result ..

I even booted via Boot Camp into Native and use the Intel CPUID util...which also tells me that VT-X is available and supported...

Looks Like every one is trying to "Make" everyone else belive that this is an issue..

From My research and PHYSICAL testing I would say it ant an issue

or did I miss something ?

 

You missed the fact that you're only talking about 2 machines, which clearly do not accurately represent experience of everyone with Minis out there. On my Mini the VT-x was disabled by default. I used the sleep trick and now I have VT-x running. There's clearly a bug in the firmware that disables it at boot-up. I've also tried both firmwares on the mini.

That said, I rarely reboot my machine, so having it work after a sleep basically means it works 99.99% of the time.

 

No Sir I think you are missing the point... I tried to MAKE my maschines report that VT-X was disabled

Thas why I redid the whole firmware issue.. to try and force it

How come its disabled by defualt on yours but iI cant get 2 other maschines to do the same...

How come I cant get mines to do what yours does although I tried it on 2 maschines bought 3 weeks apart on 2 different contents..

I can boot both maschines and install the whole thing from scratch and it works...

I can boot it into native and the intel software tells me its enabled...

If my maschine and yours show different results, I boot mines with naked OS X (10.4.6) and it works, then as the OS & Parralles & the Firmware are the same... then somthing OTHER than the firmware is a miss here... pure logic..

So what else do you have loaded on your maschine ? (startup items, additional software) partiotions, os's) perhaps we should rule out everything else before calling wolf ..

 

Don't forget there's LOTS of people whose miniMacs aren't working with VT-x support until they do one of these tricks, and the firmware has fixed it for people. It doesn't seem like there's much question about it being the firmware thing, mate. That, and it's not All miniMacs, just a seemingly random bunch of them. If yours works, then just let it work! You just happen to be quite lucky in having TWO machines without the firmware problem!

 

Hi Djoh,

Thats why I am trying to get to the bottom of this as I have 2 Boxes and they dont have this issue

In Such situations its a good idea to try and find out what is different on the Boxes, including looking at

What OS's are installed
What OS patches are loaded
Do they Have Bottcamp installed
Did they have XOM Installed


Who did they find out if VT-X is enabled (PVM, or someother thing)..

Just writing here to the Par - Team its broken.. withouth giving them something to be getting on with ant real helpfull same as people just Jumping on the Band wagon "The Firmware..."

 

Well I have a mini Core Solo if that helps. However I don't believe that people on here have been reporting 100% sucess with the Core Duo and 100% failure with the mini Core Solo. It comes back to the point that 2 machines does not a accurate sample make, regardless of what you tried to force them to do.