Jump to content

geeti

Members
  • Content Count

    29
  • Joined

  • Last visited

About geeti

  • Rank
    InsanelyMac Protégé
  1. Yeah, that's what I did.
  2. The picture was taken on the very early stage of clover booting output. FakeSMC, Lilu and any other kexts were not able to load, saying "not entitled to link kext".Checked the permission of kext binaries are all 777.Tried with different versions of clover - all same. RTVariables are also set: Hardware: Xeon E5 2648/Supermicro C612 Motherboard/RX580, almost copied everything from KGP's guide.Anyone has any idea?Thanks
  3. Very much appreciate. It's working now.. One issue left is, in my AppleIntelInfo.data output, the IA32_ENERGY_PERF_BIAS is 0. pikeralpha says it should be [1,15]. Did you have similar issue?
  4. Hi Fabiosun, Sorry for my late reply. It was a busy week at work! I have attached the clover folder as well as kext output and ioreg. Could you please take a look what's wrong? The problem is mach.xcpm.mode = 0 no matter what I've done. For the questions you asked me. 1. Yes. It's on 10.12.6 and I only see one P state(18). 2. The C states are enabled. ------update 1 ---------- I just used a patched bios to unlocked the 0xe2. From the appleintelinfo output, I saw this: MSR_PMG_CST_CONFIG_CONTROL.......(0xE2) : 0x403 ------------------------------------------ - I/O MWAIT Redirection Enable......... : 1 (enabled, IO read of MSR(0xE4) mapped to MWAIT) - CFG Lock............................. : 0 (MSR not locked) - C3 State Auto Demotion............... : 0 (disabled/unsupported) - C1 State Auto Demotion............... : 0 (disabled/unsupported) - C3 State Undemotion.................. : 0 (disabled/unsupported) - C1 State Undemotion.................. : 0 (disabled/unsupported) - Package C-State Auto Demotion........ : 0 (disabled/unsupported) - Package C-State Undemotion........... : 0 (disabled/unsupported) Does it mean the c states not enabled? ---------update 2----------- The kernel log shows this : 2017-08-12 10:41:13.750811-0400 0x77a Default 0x0 0 kernel: (IOPlatformPluginFamily) X86PlatformShim::sendStepper - Stepper dictionary from plist is NULL! 2017-08-12 10:41:13.750813-0400 0x77a Default 0x0 0 kernel: (IOPlatformPluginFamily) X86PlatformShim::sendStepper - Stepper dictionary from plist is NULL! 2017-08-12 10:41:13.881688-0400 0x77a Default 0x0 0 kernel: (IOPlatformPluginFamily) X86PlatformShim::start - Failed to send stepper 2017-08-12 10:41:13.881691-0400 0x77a Default 0x0 0 kernel: (IOPlatformPluginFamily) X86PlatformShim::start - Failed to send stepper I'm just so confused.. With the frequencyEdit script, the generated plist file does not contain any stepcontextdict. So how the x86platformshim is suppose to load stepper? Could you check with your working hackintosh, if the [macpro6,1].plist contains stepcontextdict? I'm actually using @nmano 's plist that he attached in last post. It does not contain stepcontextdict either. CLOVER.zip kextstat_output.txt Calamari.ioreg.zip
  5. Hi Fabiosum, Unfortunately the config you provided didn't work out either.. thanks anyway! I was thinking about SSDT and freqvectors. After running the freqVectorsEdit.sh, I noticed the Mac-[boardid].plist don't have the stepcontextdict any more. But according to some google search, piker said sending stepper requires the stepcontextdict. However, from the update log of freqVectorsEdit, I found Piker intentionally removed the stepcontextdict since one version... I'm just confused.. The second thing is, the ssdtPrGen doesn't support my CPU(e5-2648L v4). I added the parameter in "user defined.cfg". Will this generate a functional ssdt?
  6. Thanks for your help! I meant I did the frequencyvectors script, replacing macpro6,1 with imac17,1 or imac14,1. But same result... no xcpm... I started to thinking what did I do wrong with my installation. I've been using hackintosh for 5 years and installed over 5 machines. This is the first server system I've been using. The motherboard is C612(Supermicro x10sra), with a Intel Xeon E5-2648L v4. Is this similar to your system? If you have time, could you please take a look at my clover configs? There are multiple config, but the one I'm using is config_test.plist. Please note I disabled all the kerneltopatch except "xpcm performance fix 1".(I've tried enabled all those patches). Also attached the appleintelinfo output. Even I updated the freqVector(tried different models), I still got this from kernel log: 2017-08-05 23:20:44.675184-0400 0x822 Default 0x0 0 kernel: (IOPlatformPluginFamily) X86PlatformShim::sendPStates - Success! 2017-08-05 23:20:44.675466-0400 0x822 Default 0x0 0 kernel: (IOPlatformPluginFamily) X86PlatformShim::sendPStates - Success! 2017-08-05 23:20:44.675468-0400 0x822 Default 0x0 0 kernel: (IOPlatformPluginFamily) X86PlatformShim::sendPStates - Success! 2017-08-05 23:20:44.683935-0400 0x822 Default 0x0 0 kernel: (IOPlatformPluginFamily) X86PlatformShim::sendStepper - Stepper dictionary from plist is NULL! 2017-08-05 23:20:44.683938-0400 0x822 Default 0x0 0 kernel: (IOPlatformPluginFamily) X86PlatformShim::sendStepper - Stepper dictionary from plist is NULL! 2017-08-05 23:20:44.684205-0400 0x822 Default 0x0 0 kernel: (IOPlatformPluginFamily) X86PlatformShim::start - Failed to send stepper 2017-08-05 23:20:44.684207-0400 0x822 Default 0x0 0 kernel: (IOPlatformPluginFamily) X86PlatformShim::start - Failed to send stepper Please forgive my dumb questions.. how to tell c1e-disabled? CLOVER.zip appleintelinfo.rtf
  7. Still not... Could you enlighten me how you would do? Thanks!
  8. I was using the configs you provided for 12.6, but still got machdep.xcpm.mode: 0. Does the SSDT matter? Because ssdtprgen.sh does not support my CPU, I added those value manually to generate this ssdt. Also, I saw the long thread you had with pikeralpha on github. You also had the xcpm.mode=0 with your 2699. Did you figure it out in the end?
  9. Updated the original post. It's on Sierra. Will look at this thread. Thanks
  10. Hi nmano, Is there any different with configuring Xeon E5 v4, on Sierra? I'm using a E5-2648L v4, with the configs from your guide, but it seems the CPU power management is not engaged. Although in ioreg, the pm are loaded, but the CPU freq/power stay at max. The AppleIntelInfo shows only one p-state(max). Thanks!
  11. I just finally understood it. When I use smbios of a iMac, the CPU register IA32_ENERGY_PERF_BIAS is set to 0x1, which is highest performance. At this setting, the CPU freq is maintained at a high level. WHen I use a macbook pro smbios, the same register is set to 0x5, which is performance and power balanced. At this setting, the CPU is shifting more agressively, and average frequency is much lower. I think these are all presets in x86platformplugs, so I'll just take what Apple want it to be. Thank you so much!
  12. Same result. As long as I put the SSDT in clover, no matter whether I drop the oem table or generate p/c from clover, the CPU frequency will stay high. But the CPU pm will be enabled in ioreg. Odd...
  13. Here's the ioreg. Please note this is not the PC I asked you about today. It's from a few days ago, with Xeon E3-1240v3.(Post #156 in this thread). Calamari 2.zip
  14. You inspired me to try out the clover PM. It seems out with clover PM, the CPU frequency is being adjusted much more better. My previous confusion in this thread is about this. With SSDT and dropped pm/Ist0 tables, the CPU frequency keeps around max. Do you happen to know why? I see almost everyone is using ssdtPRGen created ssdt, but it just seems not that good.
×