Jump to content


  • Content Count

  • Joined

  • Last visited

About omni

  • Rank
    InsanelyMac Geek

Profile Information

  • Gender

Recent Profile Visitors

The recent visitors block is disabled and is not being shown to other users.

  1. It's nice that you guys are beta testing Apple's unfinished code, but perhaps we should wait till they come out officially with a final build before trying to get it working?
  2. @RemoteWorks, Why are you using MSRDumper when I clearly wrote in the OP to use AppleIntelCPUPowerManagementInfo from Pike for this test? Also, I've given instructions in my OP on how to do PM debugging - please read that again and follow through.
  3. @theconnactic As I don't have an IB Xeon I can't say myself, but I believe others did report that it worked for them. @stinga11 Would that patch apply to Mavericks though? If so I'll add it to the OP.
  4. If you meant to use MSRpatcher to unlock 0xE2 MSR that will not do. It patches other registers related to CPUPM, but it cannot unlock 0xE2. Nothing can once it's locked, so you either have to edit your BIOS, or use one of those apps that someone made, but latter might not work on all boards. You can also try to patch my AICPM.kext from the front page for MSR 0xE2 writes - that's what's causing your KPs. There's a script floating around that will find and disable those writes to that register.
  5. I was considering selling it actually, but would not sell just the board as I have no use for CPUs and RAM without it...Besides this whole hack was built water-cooled and I probably wouldn't keep any of that either then.
  6. I had a same problem in 10.9.2, and as I posted on Netkas's forum, I ultimately resolved it by replacing IOPCIFamily and AppleACPIFamily from 10.9.1. If you do not want to replace kexts, alternative was suggested by mysticalos to raise certain PCI voltages in BIOS. My BIOS does not offer the needed voltage options, so I had to resort to kext replacement. Regarding the CPUs, I used Pike's script as a base but then I manually add and correct things. You won't be generating SSDT all the time, so might as well invest some time into customizing it to your own needs.
  7. It is inevitable...I'm just tired of various small annoying glitches that happen almost on every boot. For me OS X took a dive with 10.9.2...wish I could go back to 10.6.8. I spent almost 4 months getting this hackintosh up to par and right when I thought CPUPM will be it, Apple made 10.9.2 unusable for me. I don't want to be replacing kexts to get it to work especially if things used to work OOB. They made "improvements" that only made things worse and I'm not going to waste any more of my time on it. Sigh.
  8. @RobLans I have a SB Xeon which is 0x2D just like your CPU, and have no issues with locked multipliers or anything of sorts. Give my MSRpatcher v002 a try, or alternatively enable C states in your BIOS because your MSR 0xE2 does not look right with 0x1E000400 - that last 0 should 3, 5 or 7. Your Cx_RESIDENCY are all 0, too and shouldn't be.
  9. Speaking of Chameleon, I'm having a real hard time getting to boot Ubuntu with it since by default Ubuntu uses MBR to boot, and Chameleon does not support that (ORLY? it's 2014! HELLO?) I guess because I have a drive for each OS and because I picked Chameleon to be the primary booting drive it craps out. GREAT! That's why I've installed Ubuntu and will be switching to Linux soon...
  10. MSRpatcher v007 posted, this one should actually work... Please test.
  11. If you use v006, it will output before/after MSR 0xE2 value when you do bdmesg, so I suggest trying that one... @fabiosun, That's just crazy, that your Cx_RESIDENCY is changing like that...I'm not sure how that works with sleep/awake then. Is it possible that something in your BIOS is causing this to happen (some automated setting)? @yangshun1029, Your output looks great, you got C3/C6/C7 states and performance states from minimum all the way to Turbo Boost. Have you run any benches to see whether the performance is keeping up?
  12. Ah FFS...it's this stupid wysiwyg editor that craps out all the time on long posts. Fixed now, please give it a try.
  13. I'm posting a v006 of MSRpatcher.dylib - please test it out, without any other chameleon patches you might have applied for MSR changes, so they don't clash and skew results. Thanks!
  14. VoodooTSCSync causes my clock to go crazy, either backwards or forwards... I'm putting up a new MSRpatcher.dylib up, not sure why it crashes because I tested it here.