Please do NOT download the kext! This version may hang your system! Will update the kext when I located the problem.
A new updated version of AppleIntelCPUPowerManagementInfo.kext is now available for download. I have personally run it on a Sandy Bridge and Haswell configuration and there was no KP. Please try this version and let me know if it works on your hack. Thanks!
Update: The latest version of AppleIntelCPUPowerManagementInfo.kext lets you disable features from the Info.plist See Update-2 below for more info.
Many people already know how handy AppleIntelCPUPowerManagementInfo.kext can be, but the way I compiled the public version made it skip certain parts. Stuff that you may want to see in /var/log/system.log when you are trying to get full power management on your hack going. Or just to check if it is indeed working. This is why I made a new version. One with more output. Here’s the MSR log:
AICPUPMI: MSR_CORE_THREAD_COUNT......(0x35) : 0x40004
AICPUPMI: MSR_PLATFORM_INFO..........(0xCE) : 0x80838F3012200
AICPUPMI: MSR_PMG_CST_CONFIG_CONTROL.(0xE2) : 0x1E000005
AICPUPMI: MSR_PMG_IO_CAPTURE_BASE....(0xE4) : 0x1814
AICPUPMI: IA32_MPERF.................(0xE7) : 0xF7C78F89C
AICPUPMI: IA32_APERF.................(0xE8) : 0xEABEA97E5
AICPUPMI: MSR_FLEX_RATIO.............(0x194) : 0xE0000
AICPUPMI: MSR_IA32_PERF_STATUS.......(0x198) : 0x201300002300
AICPUPMI: MSR_IA32_PERF_CONTROL......(0x199) : 0x2600
AICPUPMI: IA32_CLOCK_MODULATION......(0x19A) : 0x0
AICPUPMI: IA32_THERM_STATUS..........(0x19C) : 0x883A0000
AICPUPMI: IA32_MISC_ENABLES..........(0x1A0) : 0x850089
AICPUPMI: MSR_MISC_PWR_MGMT..........(0x1AA) : 0x1
AICPUPMI: MSR_TURBO_RATIO_LIMIT......(0x1AD) : 0x23242526
AICPUPMI: IA32_ENERGY_PERF_BIAS......(0x1B0) : 0x5
AICPUPMI: MSR_POWER_CTL..............(0x1FC) : 0x4005F
AICPUPMI: MSR_RAPL_POWER_UNIT........(0x606) : 0xA0E03
AICPUPMI: MSR_PKG_POWER_LIMIT........(0x610) : 0xFFD00000EA82
AICPUPMI: MSR_PKG_ENERGY_STATUS......(0x611) : 0x8F048E
AICPUPMI: MSR_PKGC3_IRTL.............(0x60a) : 0x8842
AICPUPMI: MSR_PKGC6_IRTL.............(0x60b) : 0x886A
AICPUPMI: MSR_PKGC7_IRTL.............(0x60c) : 0x8891
AICPUPMI: MSR_PP0_CURRENT_CONFIG.....(0x601) : 0x1F40
AICPUPMI: MSR_PP0_POWER_LIMIT........(0x638) : 0xFFD0
AICPUPMI: MSR_PP0_ENERGY_STATUS......(0x639) : 0x4E19D2
AICPUPMI: MSR_PP0_POLICY.............(0x63a) : 0x0
AICPUPMI: MSR_CONFIG_TDP_NOMINAL.....(0x648) : 0x22
AICPUPMI: MSR_CONFIG_TDP_LEVEL1......(0x649) : 0x0
AICPUPMI: MSR_CONFIG_TDP_LEVEL2......(0x64a) : 0x0
AICPUPMI: MSR_CONFIG_TDP_CONTROL.....(0x64b) : 0x80000000
AICPUPMI: MSR_TURBO_ACTIVATION_RATIO.(0x64c) : 0x0
AICPUPMI: MSR_PKG_C2_RESIDENCY.......(0x60d) : 0x3F280E1392A
AICPUPMI: MSR_PKG_C3_RESIDENCY.......(0x3f8) : 0x0
AICPUPMI: MSR_PKG_C6_RESIDENCY.......(0x3f9) : 0x0
AICPUPMI: MSR_PKG_C7_RESIDENCY.......(0x3fa) : 0x0
Example of processor info:
AICPUPMI: Low Frequency Mode : 800 MHz
AICPUPMI: Clock Speed : 3400 MHz
AICPUPMI: Max Turbo Frequency: 3800 MHz
Example of IGPU info:
AICPUPMI: IGPU Current Frequency..: 1200 MHz
AICPUPMI: IGPU Min Frequency......: 200 MHz
AICPUPMI: IGPU Max Non-Turbo Freq.: 350 MHz
AICPUPMI: IGPU Max Turbo Frequency: 1200 MHz
AICPUPMI: IGPU Maximum limit......: No Limit
Example of C-State info:
AICPUPMI: CPU C3-Cores [ 0 2 3 ]
AICPUPMI: CPU C7-Cores [ 0 2 3 ]
AICPUPMI: CPU C3-Cores [ 0 1 2 3 ]
AICPUPMI: CPU C7-Cores [ 0 1 2 3 ]
AICPUPMI: CPU C6-Cores [ 0 1 2 3 ]
And yes indeed. This shows you that all processor cores have reached the available C-States. Which is good news. You may wonder why only MSR_PKG_C2_RESIDENCY is being used, but that is most likely something related to the IGPU drivers/settings and I was unable to solve this today (sorry, this was all I could do today).
Example of P-State info:
AICPUPMI: CPU P-States [ 8 17 34 35 36 (37) ] iGPU P-States [ (4) 24 ]
AICPUPMI: CPU P-States [ 8 17 34 (35) 36 37 ] iGPU P-States [ 4 (5) 24 ]
AICPUPMI: CPU P-States [ 8 17 (34) 35 36 37 ] iGPU P-States [ 4 5 (12) 24 ]
AICPUPMI: CPU P-States [ 8 17 34 35 (36) 37 ] iGPU P-States [ 4 5 (10) 12 24 ]
AICPUPMI: CPU P-States [ (8) 17 34 35 36 37 ] iGPU P-States [ 4 5 (8) 10 12 24 ]
AICPUPMI: CPU P-States [ 8 (17) 34 35 36 37 ] iGPU P-States [ 4 5 (7) 8 10 12 24 ]
AICPUPMI: CPU P-States [ (8) 17 34 35 36 37 ] iGPU P-States [ 4 5 (6) 7 8 10 12 24 ]
AICPUPMI: CPU P-States [ (8) 17 34 35 36 37 ] iGPU P-States [ 4 5 6 7 8 10 12 (19) 24 ]
AICPUPMI: CPU P-States [ (8) 17 34 35 36 37 ] iGPU P-States [ 4 5 6 7 8 10 12 (18) 19 24 ]
AICPUPMI: CPU P-States [ (8) 17 34 35 36 37 ] iGPU P-States [ 4 5 6 7 8 10 12 18 19 (22) 24 ]
AICPUPMI: CPU P-States [ (8) 17 34 35 36 37 ] iGPU P-States [ 4 5 6 7 8 10 12 18 19 22 (23) 24 ]
AICPUPMI: CPU P-States [ (8) 17 34 35 36 37 ] iGPU P-States [ 4 5 6 7 8 10 12 18 19 (21) 22 23 24 ]
AICPUPMI: CPU P-States [ (8) 17 34 35 36 37 ] iGPU P-States [ 4 5 6 7 8 10 12 (16) 18 19 21 22 23 24 ]
AICPUPMI: CPU P-States [ (8) 17 34 35 36 37 ] iGPU P-States [ 4 5 6 7 8 (9) 10 12 16 18 19 21 22 23 24 ]
AICPUPMI: CPU P-States [ 8 17 34 (35) 36 37 ] iGPU P-States [ 4 5 6 7 8 9 10 12 (13) 16 18 19 21 22 23 24 ]
Still not perfect, probably, but it is a lot better already. Anyway. Have fun with it, and let me know if it works for you 😉
Update:
Ok. This is a short run on my i5-4670K:
AICPUPMI: CPU P-States [ 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 (35) 36 37 38 ]
I’d say that is pretty impressive, but this is (currently) only possible with the next update of AppleIntelCPUPowerManagement.kext… if we manage to get it going on a wider range of processor models. This is why your support is so important to us, so please help us validate the kext and data for you processor model. Thank you!
Update-2: The currently available version of AppleIntelCPUPowerManagementInfo.kext lets you disable feature from the Info.plist. For this we have three (boolean) properties:
logMSRs (default value is true)
logIGPU (default value is false)
logCStates (default value is true)
Note: logIGPU is disabled by default because it appears to trigger a KP on certain motherboards. Hopefully these people will help me to fix this problem. Thanks!
Update-3:
MSR_PKG_C7_RESIDENCY is not supported by all processors, and even when it is supported by the processor, then its value can be zero all the time:
Package C6 state is the deepest C-state supported on discrete graphics systems with PCI Express Graphics (PEG).
Package C7 state is the deepest C-state supported on integrated graphics systems (or switchable graphics systems during integrated graphics mode). However, in most configurations, package C6 will be more energy efficient than package C7 state. As a result, package C7 state residency is expected to be very low or zero in most scenarios where the display is enabled. Logic internal to the processor will determine whether package C6 or package C7 state is the most efficient. There is no need to make changes in BIOS or system software to prioritize package C6 state over package C7 state.
Source: Intel datasheet 328897-001.pdf page 58.
I tried it here and it froze my entire system I had to hard reset.
I did not have AICPM loaded but NullCPM but that should not matter as this was not happening before where your tool would report the two states (lowest and highest) with NullCPM.
By the way I thought you would update this on your github repo, but I guess you only provided the binary…
That is too bad. Sorry. Need to have another look at the IGPU detection code, which I presume is the problem when you run the kext on SKU’s without an uncore part. And sure. I will update my Github repo, but did not had the time for it yesterday evening. Otherwise I would only have had time to update the repo and not provided a binary, and since having it checked by you guys is more important to me…
Edit: The Github repo is now updated (neglected for a while) so please check if you can compile it yourself and change one or more of these to see what triggers the hang (thanks):
#define REPORT_MSRS 1
#define REPORT_GPU_STATS 1
#define REPORT_C_STATES 1
Thanks for updating the repo at github. I compiled a new version and tried turning GPU_STATS off but it did not help, still froze my system and I had to hard reset.
Turning MSRS and C_STATES off makes no sense as then it beats the point of the tool. 🙂
The problem might be related to newly added MSR’s. Stuff that might not be supported by your processor, so turning it off one by one will give us a clue where we have to look for the problem.
You know I wonder if adding some if/defs would help to differentiate between different processor versions and the MSRs they support?
That ought to keep the things working…I’ll look into the source and see what I can hack in. 😉
Only if you recompile the source yourself. And I only added a few new MSR’s so the search for the error should be relatively easy.
So I set REPORT_GPU_STATS to 0 and updated reportMSRs() with this:
if (gCPUModel == CPU_MODEL_HASWELL)
{
IOLog(“AICPUPMI: MSR_CONFIG_TDP_NOMINAL…..(0x648) : 0x%llX\n”, (unsigned long long)rdmsr64(MSR_CONFIG_TDP_NOMINAL));
IOLog(“AICPUPMI: MSR_CONFIG_TDP_LEVEL1……(0x649) : 0x%llX\n”, (unsigned long long)rdmsr64(MSR_CONFIG_TDP_LEVEL1));
IOLog(“AICPUPMI: MSR_CONFIG_TDP_LEVEL2……(0x64a) : 0x%llX\n”, (unsigned long long)rdmsr64(MSR_CONFIG_TDP_LEVEL2));
IOLog(“AICPUPMI: MSR_CONFIG_TDP_CONTROL…..(0x64b) : 0x%llX\n”, (unsigned long long)rdmsr64(MSR_CONFIG_TDP_CONTROL));
IOLog(“AICPUPMI: MSR_TURBO_ACTIVATION_RATIO.(0x64c) : 0x%llX\n”, (unsigned long long)rdmsr64(MSR_TURBO_ACTIVATION_RATIO));
}
That makes it work for me on SB-E Xeon (0x2D gCPUModel)
If you need to use REPORT_GPU_STATE 0 then this check appears to be failing:
if ((READ_PCI8_NB(DEVEN) & DEVEN_D2EN_MASK))
Maybe we should change it to:
if ((READ_PCI8_NB(DEVEN) & DEVEN_D2EN_MASK) == DEVEN_D2EN_MASK)
Maybe your (UEFI) BIOS isn’t setting the bits as it should?
And can you check the changes in the repo? Thanks!
How do i download this kext. Still new to this stuff.
I don’t have a compiled kext that works for every setup so you will have to compile it yourself. The source code is part of my RevoBoot repository. Back then this was the quickest solution. Not the best, but not having to add a new repository was my goal. However. This will change sometime soon.
Kernel PANIC with this new ACPUPMI.
I correctly installed, permission and Kernel Cache …
Here are all the logs:
https://dl.dropboxusercontent.com/u/25204447/KP.rtf
Thank you very much Mr. P Alpha
But does it work for you?
KP with my E5-2695 V2 & P9X79 WS /w MP6,1 SMBios and a loading X86PlatformPlugin in 10.9.2. Let me know what you need from me.
On a side note, tried to generate P-states using your script, but hit an error related to unknown processor. Should it be working with this CPU?
Sleep works, but trying to get Turbo as I am stuck at x24 multiplier.
Thanks!
Yeah I know. We still need to figure out what the problem is, and thanks – data for your CPU has been added in ssdtPRGen.sh v7.4
There is new remark I had not with old version such as :
Package (0x04)
Remark 5063 – Initializer list shorter than declared package length ^
BTW Pike what do you mean by
Warning: ‘system-type’ may be set improperly (3 instead of 1)
in SSDTPRGEN Last version ?
This is not the place to address issues in ssdtPRGen.sh so please e-mail me the generated SSDT and the output of ssdtPRGen.sh Thanks.
About the warning. That is used to initialise FADT->Preferred_PM_Profile (see ACPI specification) and the Mac Pro is a workstation (3) – hence the warning, because you are using 1.
sorry Pike. My bad.
you are right. Email sent with all informations.
have a happy new year
Still have Kernel Panic with this last updated version…
Weird. I checked it with my Intel i5-4670K (same CPU?) and have no issues whatsoever. No KP nothing.
should i test this with 10.8 or 10.9 ?
Should work with all kind of versions OS X. Confirmed to work on OS X 10.8.3 and OS X 10.9.2
I also have KP with Xeon E5 2690 V2 on 10.9.2
Can you compile the source code yourself?
I didn’t compile Kexts by myself but I could give i a try, if you tell where to get the source
https://github.com/Piker-Alpha/RevoBoot/archive/clang.zip
You can find it in the: RevoBoot/i386/libsaio/cpu/tools/PMinfo directory
Hi Pike, same error as with your version:
http://s14.directupload.net/file/d/3490/lwyhupex_jpg.htm
I tried to compile on 10.8.4 and on 10.9.2, but same result
Does it work when you set this to false in AppleCPUPowerManagementInfo.h
bool igpuEnabled = true
Oops. I forgot to change it to false in case there’s no IGPU enabled/visible. New version should be better.
MY IGPU is not visble/enabled this may be the problem for me 🙂
Ok. That might explain it because I forgot to change the flag when we detect that the IGPU is disabled/hidden. My mistake. Sorry. Let’s see if did it right this time 😉
WORKING !
Yah! Finally!
The new version is working now for me. Do you want the log output? CPU PM is not yet working correctly for my E5-2690 V2. I generated SSDT with your script. (maybe you could add the CPU-Data in your script. For now i add it manually for every new version…) In normal mode I’m locked to multiplicator 30. Deleting AICPUPM.kext enables 12 and 30. I Have MacPro6,1 SMBIOS and your recently discovered cpu-type for IvyBridge-E.
Any hints for getting PM to work or something i could help you with?
Great! Thanks for testing it!
And sure. I’ll add the missing CPU data in the next update. And yes please. The output in the log is essential in order to fix power management.
Thank YOU Pike!
Here’s the output:
MWAIT C-States…………………: 4384
MSR_CORE_THREAD_COUNT……(0x35) : 0xA0014
MSR_PLATFORM_INFO……….(0xCE) : 0xC10EC811E00
MSR_PMG_CST_CONFIG_CONTROL.(0xE2) : 0x8407
MSR_PMG_IO_CAPTURE_BASE….(0xE4) : 0x10414
IA32_MPERF……………..(0xE7) : 0x2C73CADF8C
IA32_APERF……………..(0xE8) : 0x2797FFFAFF
MSR_FLEX_RATIO………….(0x194) : 0x0
MSR_IA32_PERF_STATUS…….(0x198) : 0x1DC800001E00
MSR_IA32_PERF_CONTROL……(0x199) : 0x1E00
IA32_CLOCK_MODULATION……(0x19A) : 0x0
IA32_THERM_STATUS……….(0x19C) : 0x883A0000
IA32_MISC_ENABLES……….(0x1A0) : 0x850089
MSR_MISC_PWR_MGMT……….(0x1AA) : 0x400000
MSR_TURBO_RATIO_LIMIT……(0x1AD) : 0x2121212121222324
IA32_ENERGY_PERF_BIAS……(0x1B0) : 0x7
MSR_POWER_CTL…………..(0x1FC) : 0x2104005B
MSR_RAPL_POWER_UNIT……..(0x606) : 0xA1003
MSR_PKG_POWER_LIMIT……..(0x610) : 0x684E0005E8410
MSR_PKG_ENERGY_STATUS……(0x611) : 0xA1D3C22
MSR_PKGC3_IRTL………….(0x60a) : 0x0
MSR_PKGC6_IRTL………….(0x60b) : 0x0
MSR_PP0_CURRENT_CONFIG…..(0x601) : 0x14149480000520
MSR_PP0_POWER_LIMIT……..(0x638) : 0x80000000
MSR_PP0_ENERGY_STATUS……(0x639) : 0x84AC893
MSR_PP0_POLICY………….(0x63a) : 0x0
MSR_PKG_C2_RESIDENCY…….(0x60d) : 0x0
MSR_PKG_C3_RESIDENCY…….(0x3f8) : 0x0
MSR_PKG_C6_RESIDENCY…….(0x3f9) : 0x0
Low Frequency Mode……………..: 1200 MHz
Clock Speed (Max. Non-Turbo Freq.).: 3000 MHz
Maximum Turbo Frequency…………: 3600 MHz
CPU P-States [ (30) ]
CPU C6-Cores [ 2 5 6 7 8 10 11 15 ]
CPU C6-Cores [ 2 3 4 5 6 7 8 9 10 11 12 13 15 16 ]
CPU P-States [ (12) 30 ]
CPU C6-Cores [ 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 ]
And that’s the sysctl machdep.xcpm output:
machdep.xcpm.mode: 0
machdep.xcpm.asserts_enabled: 0
machdep.xcpm.asserts_traced: 0
machdep.xcpm.cpu_wakeup_energy_cost_selector: 0
machdep.xcpm.disable_quiesce: 0
machdep.xcpm.mp_match: 0
machdep.xcpm.disable_idle_self_select: 0
machdep.xcpm.mp_load_txfr_coeff: 128
machdep.xcpm.scalability_cpi_threshold: 4
machdep.xcpm.scalability_cpi_demotion_threshold: 5
machdep.xcpm.scalability_reeval_interval: 1000
machdep.xcpm.scalability_eval_ratio_min: 17
machdep.xcpm.scalability_detection_enabled: 0
machdep.xcpm.hard_plimit_max_100mhz_ratio: 0
machdep.xcpm.hard_plimit_min_100mhz_ratio: 0
machdep.xcpm.soft_plimit_max_100mhz_ratio: 0
machdep.xcpm.soft_plimit_min_100mhz_ratio: 0
machdep.xcpm.ratio_changes_hf: 0
machdep.xcpm.ratio_changes_lf: 0
machdep.xcpm.ratio_change_limited: 0
machdep.xcpm.ratio_change_hf_limit: 2
machdep.xcpm.ratio_change_lf_limit: 1
machdep.xcpm.ratio_change_ratelimit_ns: 500000
machdep.xcpm.ratio_changes_total: 0
machdep.xcpm.maxbusdelay: 0
machdep.xcpm.maxintdelay: 0
machdep.xcpm.mbd_mode: 1
machdep.xcpm.mbd_applications: 0
machdep.xcpm.mbd_relaxations: 0
machdep.xcpm.forced_idle_ratio: 100
machdep.xcpm.forced_idle_period: 30000000
machdep.xcpm.deep_idle_log: 0
machdep.xcpm.qos_txfr: 1
machdep.xcpm.q_migration_ttd_min: 1000000
machdep.xcpm.q_migration_ttd_max: 100000000
machdep.xcpm.preidle_spin_tsc: 0
machdep.xcpm.qos_ratio_change_limited: 0
machdep.xcpm.qos_ratio_change_hf_limit: 8
machdep.xcpm.qos_ratio_change_lf_limit: 1
machdep.xcpm.qos_ratelimiting_enabled: 1
machdep.xcpm.quiesce_trace: 0
machdep.xcpm.deep_idle_count: 0
machdep.xcpm.deep_idle_last_stats: n/a
machdep.xcpm.deep_idle_total_stats: n/a
I know XCPM is only working for Haswell CPUs, but deleting the AICPUPM.kext is working better for me…(Nullcpupm.kext deleted of course!)
This is not good:
MSR_PKGC3_IRTL………….(0x60a) : 0×0
MSR_PKGC6_IRTL………….(0x60a) : 0×0
See if you can somehow change this in the BIOS.
0xE2 and 0x1AD are also wrong. If these cannot be changed via BIOS settings, then you need to do this from the bootloader.
Hi Pike, just tried to change some BIOS settings, but nothing really changed.
Any further possibilities or do we have to wait?
I’m not familiar with the Clover source code, but if you can setup an USB stick with RevoBoot, then testing can continue (we need to figure out the MSR’s that need to be changed, and that can be done with Revoboot so much easier).
Thanks for the hint! I managed compiling and installing it.
After booting the grey apple logo screen apperars and after about 5sec the system will hang.
How do I activate verbose mode with revoboot?
This is not the right place to discuss this kind of stuff. Please use Github issues for it. Thanks!
Are you doing this in 10.9.2 beta though?
Sorry, I am currently using my iPad and I don’t know if this question was for me or someone else… Running into a limitation of WordPress on the iPad.
Yes, I am on 10.9.2
Sorry it was for Mark…as he said he was not getting PM working. I have a v1 Xeon E5 and not having PM either, even with 10.9.2.
Although everything is getting set up with X86PlatformPlugin in the IOReg, the AICPM does not detect my CPU…
Ok, but are all MSR’s initialised correctly now? If not, you should definitely start experimenting with RevoBoot on a USB stick as that will be the only way for me to help you guys – the goal here is of course is to figure out what is wrong so that all other boot loader developers can fix their code.
Don’t forget that the Clovers developer(s) only recently added the missing ACPI methods to their code, which is one of the many things that we figured out, a long ago already, so… I don’t really trust their code (with all due respect for Slice and his gang).
Edit:
It came to my attention that some Clover folks think that this was an attempt te have people switch to RevoBoot, but that is not true.
Thing is. I do not want you, nor anyone else for that matter, to switch to RevoBoot… but since RevoBoot is the only boot loader that I support, and thus if you want my help, then I kindly ask you to make it as easy as possible for me, by using RevoBoot. Be it temporarily… just so that we can try to fix problems.
The MSRs aren’t getting all initialized, but I believe that to be due to the AICPM failing to detect the CPU and thus do its thing properly, either because it ends up using a Virtual CPU or Null CPU at the end instead of SB or IB.
I’m on 10.9.2 beta.
Pike, how can I install Revoboot? it won’t let me compile as i get error:
[MKDIR] i386/config/SETTINGS
[CP] i386/config/settings-template.h i386/config/SETTINGS/default.h
======================================================
Running: make MODEL= mountain-lion
================= RevoBoot/i386/util =================
make[2]: *** No rule to make target `%.s’, needed by `/Users/test/Desktop/RevoBoot-clang’. Stop.
an older version will also not compile:
./libsa/../config/SETTINGS/default.h:331:99: note: expanded from macro
‘STATIC_NVRAM_ROM’
…0x03, 0x04, 0x05, 0x06, 0x07, 0x08, 0x09, 0x0A, 0x0B, 0xNN, 0xNN, 0xNN, …
Rename your ‘RevoBoot-clang’ directory to ‘RevoBoot’ (our Makefiles suck) and read the note below #define STATIC_NVRAM_ROM in settings-template, which reads: “Note: Use your MAC address for the last six bytes!“
How do we actually use RevoBoot?
There was some obscure page on Wiki about RevoDisk and RevoStick but just putting files onto a hdd or USB does not do anything…
Sorry I know this is off topic here but maybe others would benefit from your writing a few words of instructions here.
The idea was simple; replace the Chameleon /boot file, which at that time (back in 2009) was the only boot loader we could use for OS X. And I understand that there is a need for additional information, but let’s add the missing bits to the Wiki so that everyone can find it.
Installed and no KP
Tested on I7 4770K :
kernel[0] AICPUPMI: CPU P-States [ (8) 35 37 38 39 ] iGPU P-States [ 4 5 7 8 9 10
If you need more information ask away
KP on an AsRock X79 Extreme 9 with a 2690 V2 IB Ep cpu
Please download the source code and make this change: #define REPORT_GPU_STATS 0
If that makes it works, then we can see why/what fails on your motherboard, because Mark has the same CPU and no KP.
Thank you pikeralpha but i Have no skills to do what you are advicing me to do.
If you can guide throw steps I can do
Thank you again
Hello Pike,
#define REPORT_GPU_STATS 0 not solved my freeze on sudo touch xxxx
Setting logIPGStyle = 0 helped me and now I can use your latest AICPUPMI
I have Cx Residency to 0x0 in output but maybe because I don’t know how to set plugin-type = 1 in SSDT or in chameleon bootloader if needed for my IB EP Cpu.
In X79 Extreme 9 bios enabling or disablig C states is not useful
Yeah. Some setups require you to make changes to one or more settings in the Info.plist, to disable stuff, or you may keep running into crashes. I would like to fix this so can you please comment out this line and see if that helps:
wrmsr64(199, (pState << 8));
Thanks!
p.s. About the plugin-type. See the SSDT created by sstPRGen.sh
Hi Pike – I enabled iGPU in info.plist and NO KP this time but the readout appears incorrect, as I am using only the discrete Nvidia GT660 is this the reason? Any thoughts are appreciated.
Shared with Droplr: ‘Pike.txt’
http://d.pr/f/Ar9e
It will not log data when the IGPU is disabled/hidden by the BIOS. We have to do this or it will crash. Does that answer your question?
Hello Pike
Can you show me how to do so the output place
thanks
“Update 2” AICPM causes me KP
This version has three switches (see Info.plist) that can be used to disable one or more features: logIGPU, logCStates and logMSRs Set one at a time to false/NO and see where the KP happens. Thanks for testing!
Hello Pike
I’m sorry but I did not say where he created the KP
Tried with default settings nothing KP
PlogMSRs (true)
logIGPU (false)
logCStates (true)
By changing logIGPU in (true) ko KP
PlogMSRs (true)
logIGPU (true)
logCStates (true)
Ok. So when logIGPU is set to true, then you get a KP. I suspect a (UEFI) BIOS error. One that isn’t clearing the bit stating that your processor is one without uncore (IGPU) part. Do you have a setting in the BIOS to manually disable it?
Hello Pike,
I am starting a Hackintosh build with the following components:
(2) E5-2687W (ES)
Asus Z9PE-D8 WS (bios 5304)
EVGA GTX 760
32GB G-Skill DDR3 (2133)
256GB Samsung Pro SSD
I have never attempted this before so any help/suggestions are greatly appreciated!
Thank you!
Hi Pike,
current version is again KPing 😦
IGPU logging disabled of course.
I have no idea what I did. Let’s see. Can you please set logMSRs (or logCStates) to false and see if that stops it from crashing? This will give me an idea where to look for the error. Thanks.
It even KPs with all 3 options set to “false”
Leaving logIGPU defaults
I tried logCStates = No I have KP
I tried logMRs = No I have KP
Hello Pike, current version (update 2) gives KP, tried all flags PlogMSRs,logIGPU and LogCStates. The version before (update 2) was working OK on Asus R4E i7-3930K Mavericks 10.9.2, thanks!!
Same with me!
Thank’s for your work Pike, compiled version of v3.3 AICPUPMI and working fine, tested on SB-E i7-3930K Mavericks 13C44.
AICPUPMI: logMSRs……………………….: 1
AICPUPMI: logIGPU………………………..: 0
AICPUPMI: logCStates…………………….: 1
AICPUPMI: logIPGStyle……………………: 1
Nothing new to report back, no C-tsates, just a bunch of P-States (only without AICPUPM).
AICPUPMI: CPU P-States [ (12) 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 ]
Pike, is your AICPMI supposed to work in Safe Mode as well?
If you recall, I have to boot 10.9.2 into Safe Mode since my graphics does not work otherwise. When using ACPI_SMC_Platform upon loading your AICPMI it KPs and that’s with default settings (iGPU false), and a SB Xeon-E5 CPU…
Pike, this is interesting:
After disabling EIST in the BIOS, the CPU reached Turbomode 3,3Ghz. But only after deleting AICPUPM.kext
So now I have 1,2Ghz, 3,0Ghz and newly: 3,3Ghz. I never ever had 3,3GHz before!
Time to get up your info.kext again! Maybe this resolves the bad MSRs… 🙂
Have you tried to compile the source code from the Github repo?
No, that was the old code. I will give the new commit later a try.
Thanks.
Ok, the new commit works good! I just copied over my settings.h and everything worked.
Could you please upload the last working AICPUPMI.kext again? I can’t find it anymore on my HDD.
The compiled kext can be found in:
~/Library/Developer/Xcode/DerivedData/AppleIntelCPUPowerManagementInfo-*/Build/Products/Debug
You have used that for your testing? If yes. Good. Then we can add back stuff. One piece of code after the other, to see what I did wrong (currently a bit busy writing a new AppleHDA script).
Edit: First set of changes is now committed.
Hi Pike, with the latest commit, the kext is working for me again. Thanks!
But the critical MSRs didn’t change in the meantime.
Mark, Are you using a 2690 V2 Cpu?
If so, can you share compiled AppleIntelCPUPowerManagement.kext and AICPUPMI.kext
In my setup turboosts is an utopia at this moment. I have only two steps (12/30) till a 3120 Mhz due to my 104 blck over clock state
Tip: Do not use anything but 100 MHz because Apple does all kind of calculations based on 100 MHz and is not expecting a greater value.
Thank you for tip.
I am using same Windows bios settings and I have no problem, Istat Pro programgive a value of 3.120 Ghz and 1.248 Ghz ( I achieve only two steps with standard myhack installation without any additional kext)
Your tip is for stability of my system?
Come on. You know better. OS X is not Windows. It doesn’t matter what Windows supports. Apple simply uses 100 MHz in their routines to calculate P-States and what not. And you do want power management and a stable hack, correct?
yes Pike, i would like to say tha i have same behavior with default bcl
and yes, i would like to have corrects speedsteps and correct PM
I am learning something, now I can see these values:
Jan 10 03:37:27 localhost kernel[0]: AICPUPMI: CPU P-States [ 33 36 ]
Jan 10 03:37:27 localhost kernel[0]: AICPUPMI: CPU P-States [ 12 33 36 ]
Jan 10 04:11:06 localhost kernel[0]: AICPUPMI: CPU P-States [ 33 36 ]
Jan 10 04:11:06 localhost kernel[0]: AICPUPMI: CPU P-States [ 12 33 36 ]
Jan 10 10:28:21 localhost kernel[0]: AICPUPMI: CPU P-States [ 33 36 ]
Jan 10 10:28:21 localhost kernel[0]: AICPUPMI: CPU P-States [ 12 33 36 ]
With Kozlek hwsensors I can see also 31 and 32
I would like to tell that I tried with 100 Mhz and all ten core are “stressed” at 3,3 Ghz.
at 104 Mhz they reach 3.432 Ghz, same behavior of my windows 8.1 setup I did some test with cinebench and I see all 20 thread
Thank you again
Sorry pikeralpha, I have to learn more
I achieve same result with or without AppleIntelCPUPowerManagementInfo.kext launching :
cat /var/log/system.log | grep “AICPUPMI:”
as you say in other forum. so something for me is not working as it should be.
With the actual version I have immediately a cpu KP, with version found on a your thread in other forum, with your instructions, system starts, but I have same results with or without kext installation
Pike,
I used your hint and updated the cpu.c of revoboot to write the missing MSR values:
MSR_PKGC3_IRTL………….(0x60a) : 0×0
MSR_PKGC6_IRTL………….(0x60b) : 0×0
I used your values and the wrmsr64 function to update the registers.
So now I have some kind of working Turbo Mode. AICPUPMI.kext told me the CPU reached P-Sate 33 and 36! Additionally to 12 and 30.
What does this tell us? I don’t really understand the relation between the C-State registers and “working” turbo-mode.
Again without aicpupm.kext. With that i’m locked to P-State 30.
I Also checked the other MSR 0x1AD. Comparing it the the Intel reference, the value seems correct to me.
What are the values that you used (0x88nn)? I would also add a value to MSR_PKGC7_IRLT as all three are checked by XCPM.
Here’s another thing to consider. The value of MSR 0x1AD (MSR_TURBO_RATIO_LIMIT) is 0×2121212121222324 but that gets shifted (16 bits) to the right in _xcpm_dvfs_configure and thus becomes: 0x212121212122 afterwards. Meaning that Apple only wants a 6 core value. Tops. So what happens when you change that into say: 0x1F2021222324 ?
Also. Don’t forget to check the (un)demotion bits for the C-States.
I thought Mark had a Xeon-E5 v2, and there’s no XCPM for IB only Haswell in the kernel, no?
Both are supported. FrequencyVectors are Haswell only. See plist files.
Those are the values:
wrmsr64(MSR_PKGC3_IRTL, 0x8842);
wrmsr64(MSR_PKGC6_IRTL, 0x886A);
wrmsr64(MSR_TURBO_RATIO_LIMIT, 0x1F2021222324);
wrmsr64(MSR_PKGC7_IRTL, 0x8891);
But the latest values, you told me to try, don’t change anything. I’m not quite sure if C7 is even available on IB-EP.
How do you know, XCPM checks that values? Do you have the sourcecode of it? Where can I get it?
But is XCPM really the solution for Macpro61? Considering the original MP61 ioreg file, AICPUPM gets loaded and thus the should put our efforts on that?
Correct. There is no C7 support (see AppleIntelCPUPowerManagementInfo.kext for a link to the documentation) but then it expects a value of 0xFFFF.
And I don’t think that anyone but Apple has the source code, but we decompile stuff to see what it does (see also: https://pikeralpha.wordpress.com/2013/08/20/hopper-disassembler/)
The rest will have to wait for now (need to go back to work first).
Mark, based on your 0xE2 MSR you have no C limit set, so C7 should be available. After all that’s something definable in your BIOS settings. At least in mine I can control whether to enable specific C-states in BIOS, and whether there’s a limit or not.
AppleIntelCPUPowerManagement.kext will change the status of the first three bits. Look here:
_SandybridgeCStateRegInit:
../..
movl $226, %ecx
rdmsr
movl %edx, %edx
andl $4194271224, %eax // Mask (0xF9FF7FF8)
orq $7, %rax
movl $226, %ecx
wrmsr
_IvybridgeCStateRegInit:
../..
movl $226, %ecx
rdmsr
movl %edx, %Ed
andl $3791618040, %eax // Mask (0xE1FF7FF8
orq $7, %rax
movl $226, %ecx
wrmsr
So go ahead and wipe them in your (UEFI) BIOS / boot loader and don’t be surprised when they are set (again). The same thing happens in the kernel with XCPM, be it with different routines.
ok, but writing wrmsr64(MSR_PKGC7_IRTL, 0×FFFF);
immediatly resets my system.
/me ducks (Oops. 0xFFFF is a check for invalid data).
What did you do to cpu.c to write the MSRs and what ever the values that you wrote?
Hi johnd,
i added that to the function void initTurboRatios():
void initTurboRatios()
{
wrmsr64(MSR_PKGC3_IRTL, 0x8842);
wrmsr64(MSR_PKGC6_IRTL, 0x886A);
wrmsr64(MSR_TURBO_RATIO_LIMIT, 0x1F2021222324);
wrmsr64(MSR_PKGC7_IRTL, 0xFFFF);
with the 0xE2 MSR you’re right, but C7 is simply not supported by that CPU.
Would SB Xeon also work with XCPM if IB Xeon is supported?
Hey man thanks for all your work. I extracted a couple of acpi from the mac pro 6,1 efi firmware update. I dont know if is useful for you but here is.
https://mega.co.nz/#!eg4UBQhB!RK66NBDyTaCgk7DoyQUrGEnseYQOdptE1A6jfIwlpac
You can get an entire Mac Pro 6,1 Darwin Dump, compliments of mrengles, from http://organicdata.net/mrengles/MacPro61_DarwinDumper_SysDiagnose.zip
Thanks, but I already have all the data that we need.
Please note that Samantha, my late sister, wrote a Perl script for this and anyone interested in using acpi-extract.pl can now download it.
Hi there,
I was wondering if there is a way to manually disable CPU throttling OR increase the CPU’s ‘Low Frequency Mode’ from 800MHZ to something higher?
My 17′ macbook 2011 i7 CPU speed is stuck at using 0.8GHZ because of a thermal sensor gone wrong on the logic board ($600 repair) – I will manage heat/fans manually to protect the CPU.
thanks in advance. Great thread btw.
Why don’t you try using NullCPUPowerManagement.kext ?
It generally sets a lowest and highest ratio and then the system throttles between the two.
does it work fine with 10.9.1 Mavericks?
Because i remember reading somewhere on a site that as of Mavericks, the NullCPU kext would no longer work correctly.
does it work fine with 10.9.1 Mavericks?
Because i remember reading somewhere on a site that as of Mavericks, the NullCPU kext would no longer work correctly.
It works fine for me.
Is 4930K supported with this script? Tried (with 10.9.2b) and while it did create a SSDT, I still get no PM (and also no sleep in clover), multiplier stuck at x34. Any idea?
Try this: http://www.insanelymac.com/forum/topic/295200-testers-needed-cpu-power-management-for-sb-and-ib-xeons-or-i7-39xx-on-x79-or-c60x-chipset/
Sure. The 4930K is supported.
About broken sleep. You are not alone, but this has been discussed time and time again. The problem is that dropping all, or the wrong set of, SSDT table(s) will break stuff. The best thing for you to do is to use a patched DSDT. One that doesn’t rely on methods and variables in any SSDT table because honestly… automatic patching of ACPI tables, in any boot loader, has so its limits. This is obviously one of it.
Well, actually I get best results when I do not patch anything in clover, and do not drop any tables. But still neither sleep nor CPU PM. The big problem for me is that there seems to be no definitive explanation as to what actually causes these problems, it is more like “try and error”. Any idea where to look at? Ah and as it seems – if the post johnd mentioned is correct – there is more to enable CPU PM than just creating a custom SSDT. PS: also tried with a custom DSDT, that also fixes my USB problems but still no dice.
In my book that is not a “best result” because without proper power management and sleep being broken, it totally sucks.
Fixing ACPI issues can be though at times, yes, but the problem can be solved. Start by looking at the Methods that are being called from your DSDT. Also check variables that might be used. There simply has to be something that people like you are missing.
Sure. ssdtPRGen.sh is just a tool, like AppleIntelCPUPowerManagementInfo.kext is, but these are the tools that I use to solve power management issues, and help me to understand what is going on when it fails.
Also. In my view we have way too many people who solely rely on auto-patching by boot loaders like Clover and other tools. Good for newbies, but the result is that almost nobody seem to understand stuff anymore.
Well, the problem is simply that I am not an ACPI expert …. so I simply do not know! And yes auto patching is a problem, that seems to make things only worse, so I already avoid that. What could also be a problem is that I seem to have a newer X79 rev. But as I said this is just over my head. Also still haven’t found a good “beginners guide” or something similar, to ease the steep learning curve with ACPI. Also most people just seem to do “try and error”, there are almost no explanations to what is actually happening with all these edits. So without expert help, maybe I should simply give up now, already wasted so much time trying all kinds of stuff. Yes, that sucks BIG time, but what can I do?
Feel free to e-mail me a zipped up Darwindump output and I will ask one of the interns here to take a look at it.
I’ll gladly do that (already have prepared the files), but to which email adress? I don’t see any on this website, or am I blind?
See ssdtPRGen.sh
Not 4930K is NOT supported, at least the values are wrong. So, do NOT use the script with 4930K – it won’t take you anywhere, sorry.
May I ask what script that it is that you are referring to?
In case you are talking about ssdtPRGen.sh then have a look for yourself:
http://ark.intel.com/nl/products/77780/Intel-Core-i7-4930K-Processor-12M-Cache-up-to-3_90-GHz
In short. All values are fine.
Errm, I just tested that and with the values ssdtPRgen provided I am stuck at x12. Using the correct values gives me 12x – 39x. Sorry to say that but it seems you never really tested that – it just does NOT work! If you don’t believe me, try for yourself … For starters: your APSS values start from 800MHz whereas the 4930K can only go as low as x12 if I recall it correctly. So still sure your script is fine?
Correct. I never owned an Intel i7-4930K and thus I have never been able to verify it. Not personally. However. There are many pages about ssdtPRGen.sh on forums where I ask people for help. To improve this script, but when that doesn’t happen… err.
Here is one example: I was unaware of the error that certain processors were misplaced, and used 800 MHz instead of 1200 MHz, but luckily someone e-mailed me about it… and as a result I fixed it earlier today.
So after so many long hours. So many updates. You still can only blame me for missing something, that I do in the little free time I have… instead of trying to help me and solve the problem.
Errrm. Sorry, but I DID help you by telling that something might be wrong with your script! Just. Scroll. up. But all you did is insisting that it works just fine – but as it seems without ever testing it … So I do not blame you for making mistakes – I am developing myself, and I know this just happens – but rather for blaming ME, when in fact it was an error on your side. That’s the whole point. And I do not mean it personal, but I have seen this too often, esp. in the so called “Hackintosh” community. So yeah, instead of finger pointing we should go on with actual problem solving 😉 Just my 2c.
You should read the thread (all pages) to see that I kept on asking for help from people, constantly, and that when something was reported as broken… that I fixed it. However. After Samantha died and Jeroen took control of his own life I have to do everything myself. Next to being a husband and building a home. That is when delays happen. I’m only human and sometimes forget about stuff being broken, that’s when friendly reminders come in handy. But the way you did it doesn’t help anyone.
Hey, I did not want to step on anyones toes! But excuse me, I do not see what did I wrong? I even provided you with a hint _why_ these values might be wrong just above (ie 800 vs 1200). Sorry that I am no expert, I just do as best as I can.
Thanks. I already moved/changed the processor data, but I only do that after I got a confirmation from someone who showed me the output of AppleIntelCPUPowerManagementInfo.kext The problem was that it crashed for you (and others). Something that apparently was solved by someone else, but who didn’t share his change(s) with me. Which is too bad for people like you, since you guys are ultimately paying the price for it.
Listen. It’s not about stepping on toes, but about being fair. Let’s not forget. I do this as a volunteer in my free time. Please keep that in mind. And when things seem to slow down, then I usually are trying to locate and solve problems. You know. For processor and stuff that I don’t even own myself.
Yeah for “noobs” like me it is double and triple hard believe me, esp if you also have unexpected (and unexplainable) problems with your setup. But after all I might be one of the first persons who actually enjoys an almost perfect working LGA2011 PM on a hack. (at least it looks like so). AICPUPMI still shows some unexpected values it seems, but I have already posted that log in the IM thread. So, let us just be happy we’re actually one step further with that issue, OK 😉
Still no X79 SpeedStep
Pike,
What is the final word on the EIST, should that be enable or disabled?
Looking at my bits in the MSR it looks like it’s enabled and there’s nowhere in my BIOS settings to allow me to disable it (so I’ll have to hack the firmware again and do it in code if necessary)
I have everything set to disabled in the (UEFI) BIOS for my Haswell CPU, but XCPM enables it when I boot up with OS X so I don’t think that it really matters. At least for my setup. This way we know that XCPM really works with, for my setup, and not that some EFI module is at work.
Sandy Bridge and Ivy Bridge CPU’s on a X79 motherboard are a completely different story. So it seems. We have a lot more testing to do to get a grip on power management for these boards. The problem is that I am working on a new updated version of AppleHDA8Series.sh and that takes time. In fact. It is taking me too much time, now that I only have a limited amount of free time available for hacking stuff, and thus it may look like we are slowing down, but most of you have been waiting for ages already so… what is a little extra wait anyway.
I wouldn’t say things are slowing down, as progress was made with AICPM on that link I posted above at insanelymac forum.
Turbo modes appear to be working, but lower performance states are still missing.
I was talking about my personal involvement, not that of anyone else. And hey. I may be wrong, but I don’t know why you would need to patch AppleIntelCPUPowerManagement.kext since Mark reported that he got two more Turbo P-States when he initialised MSR_PKGC3_IRTL/MSR_PKGC6_IRTL. Which in my book means that it worked without patching the kext. And the fact that he only got two extra P-States might because the value of his MSR 0x1AD is not what Apple expects (shifted).
Hey Pikeralpha,
The updated kext freezes or KPs on my system even if I load it after boot.
3930k/Rampage IV Extreme
I compiled it from source and it now works. The linked version above gave me KPs.
Yes. The version linked to (under Update-2) includes new code. Stuff that may not work on every setup. Something I had to figure out the hard way.
kernel panic on original 2006 Mac Pro ( apple hardware )
what I see :
AICPUPMI:MWAIT C-states : 32
panic(cpu 0 caller 0xffffff80002dbe2e):kernel trap…. type 13 general protection
fault cr2:0xffffff8091565000 erorr code 0 fault cpu 0
Thank you for your report Dirk, but what switches are enabled/disabled (see Info.plist)?
Also get panics as soon as I load this …
i have to disable logIPGStyle, if its enabled, i got kp
here is my dmesg,
AICPUPMI: logMSRs……………………….: 1
AICPUPMI: logIGPU……………………….: 0
AICPUPMI: logCStates…………………….: 1
AICPUPMI: logIPGStyle……………………: 0
AICPUPMI: MWAIT C-States…………………: 4384
AICPUPMI: MSR_CORE_THREAD_COUNT……(0x35) : 0x40008
AICPUPMI: MSR_PLATFORM_INFO……….(0xCE) : 0x100070012200
AICPUPMI: MSR_PMG_CST_CONFIG_CONTROL.(0xE2) : 0x1E000400
AICPUPMI: MSR_PMG_IO_CAPTURE_BASE….(0xE4) : 0x10414
AICPUPMI: IA32_MPERF……………..(0xE7) : 0x1117610C58
AICPUPMI: IA32_APERF……………..(0xE8) : 0x10F5A25DB2
AICPUPMI: MSR_FLEX_RATIO………….(0x194) : 0xE0000
AICPUPMI: MSR_IA32_PERF_STATUS…….(0x198) : 0x24FD00002200
AICPUPMI: MSR_IA32_PERF_CONTROL……(0x199) : 0x2200
AICPUPMI: IA32_CLOCK_MODULATION……(0x19A) : 0x0
AICPUPMI: IA32_THERM_STATUS……….(0x19C) : 0x883E0000
AICPUPMI: IA32_MISC_ENABLES……….(0x1A0) : 0x850089
AICPUPMI: MSR_MISC_PWR_MGMT……….(0x1AA) : 0x400000
AICPUPMI: MSR_TURBO_RATIO_LIMIT……(0x1AD) : 0x23242526
AICPUPMI: IA32_ENERGY_PERF_BIAS……(0x1B0) : 0x0
AICPUPMI: MSR_POWER_CTL…………..(0x1FC) : 0x4005F
AICPUPMI: MSR_RAPL_POWER_UNIT……..(0x606) : 0xA1003
AICPUPMI: MSR_PKG_POWER_LIMIT……..(0x610) : 0x83B6001482F8
AICPUPMI: MSR_PKG_ENERGY_STATUS……(0x611) : 0x1CA7143
AICPUPMI: MSR_PKGC3_IRTL………….(0x60a) : 0x8850
AICPUPMI: MSR_PKGC6_IRTL………….(0x60b) : 0x8868
AICPUPMI: MSR_PKGC7_IRTL………….(0x60c) : 0x886D
AICPUPMI: MSR_PP0_CURRENT_CONFIG…..(0x601) : 0x18141494000004B0
AICPUPMI: MSR_PP0_POWER_LIMIT……..(0x638) : 0x0
AICPUPMI: MSR_PP0_ENERGY_STATUS……(0x639) : 0x163FA84
AICPUPMI: MSR_PP0_POLICY………….(0x63a) : 0x0
AICPUPMI: MSR_PKG_C2_RESIDENCY…….(0x60d) : 0x0
AICPUPMI: MSR_PKG_C3_RESIDENCY…….(0x3f8) : 0x0
AICPUPMI: MSR_PKG_C6_RESIDENCY…….(0x3f9) : 0x0
AICPUPMI: MSR_PKG_C7_RESIDENCY…….(0x3fa) : 0x0
AICPUPMI: CPU Low Frequency Mode………….: 1600 MHz
AICPUPMI: CPU Maximum non-Turbo Frequency….: 3400 MHz
AICPUPMI: CPU Maximum Turbo Frequency……..: 3800 MHz
AICPUPMI: CPU P-States [ (16) 34 ]
AICPUPMI: CPU C6-Cores [ 2 3 4 ]
AICPUPMI: CPU C6-Cores [ 2 3 4 5 6 ]
AICPUPMI: CPU C6-Cores [ 2 3 4 5 6 7 ]
AICPUPMI: CPU C3-Cores [ 0 1 ]
AICPUPMI: CPU C6-Cores [ 0 1 2 3 4 5 6 7 ]
AICPUPMI: CPU C3-Cores [ 0 1 3 5 ]
AICPUPMI: CPU C3-Cores [ 0 1 2 3 4 5 ]
AICPUPMI: CPU P-States [ 16 34 35 (36) ]
AICPUPMI: CPU P-States [ 16 (25) 34 35 36 ]
AICPUPMI: CPU P-States [ 16 25 34 35 36 (37) ]
AICPUPMI: CPU C3-Cores [ 0 1 2 3 4 5 7 ]
AICPUPMI: CPU C3-Cores [ 0 1 2 3 4 5 6 7 ]
AICPUPMI: CPU P-States [ 16 25 (30) 34 35 36 37 ]
AICPUPMI: CPU P-States [ 16 (20) 25 30 34 35 36 37 ]
Hi pike, AICPMI kext found on tonymack place in your post is it legit to use?
with this one i have no KP and I would like to know if it useful for testing or I have to use version you posted here
Thank you
Yes. Quite old already but better than nothing. I’ll try to resolve the KP issues some time this weekend.
thank you
Pike, only commenting this line /* wrmsr64(199, (pState << 8)); */ : (as you suggested) with AICPUPMI 3.2 I have this result:
Jan 28 11:13:28 fabios-Mac-Pro.local com.apple.kextd[16]: WARNING – Invalid signature -67062 0xFFFFFFFFFFFEFA0A for kext "/Users/fabio/Desktop/AppleIntelCPUPowerManagementInfo.kext"
Jan 28 11:13:28 fabios-Mac-Pro kernel[0]: AICPUPMI: v3.2 Copyright © 2012-2014 Pike R. Alpha. All rights reserved
Jan 28 11:13:28 fabios-Mac-Pro kernel[0]: AICPUPMI: logMSRs……………………….: 1
Jan 28 11:13:28 fabios-Mac-Pro kernel[0]: AICPUPMI: logIGPU……………………….: 0
Jan 28 11:13:28 fabios-Mac-Pro kernel[0]: AICPUPMI: logCStates…………………….: 1
Jan 28 11:13:28 fabios-Mac-Pro kernel[0]: AICPUPMI: logIPGStyle……………………: 1
Jan 28 11:13:28 fabios-Mac-Pro kernel[0]: AICPUPMI: MWAIT C-States…………………: 4384
Jan 28 11:13:28 fabios-Mac-Pro kernel[0]: AICPUPMI: MSR_CORE_THREAD_COUNT……(0x35) : 0xA0014
Jan 28 11:13:28 fabios-Mac-Pro kernel[0]: AICPUPMI: MSR_PLATFORM_INFO……….(0xCE) : 0xC10E4811E00
Jan 28 11:13:28 fabios-Mac-Pro kernel[0]: AICPUPMI: MSR_PMG_CST_CONFIG_CONTROL.(0xE2) : 0x401
Jan 28 11:13:28 fabios-Mac-Pro kernel[0]: AICPUPMI: MSR_PMG_IO_CAPTURE_BASE….(0xE4) : 0x10414
Jan 28 11:13:28 fabios-Mac-Pro kernel[0]: AICPUPMI: IA32_MPERF……………..(0xE7) : 0x3E6311EF8
Jan 28 11:13:28 fabios-Mac-Pro kernel[0]: AICPUPMI: IA32_APERF……………..(0xE8) : 0x3C3CF6F37
Jan 28 11:13:28 fabios-Mac-Pro kernel[0]: AICPUPMI: MSR_FLEX_RATIO………….(0x194) : 0x0
Jan 28 11:13:28 fabios-Mac-Pro kernel[0]: AICPUPMI: MSR_IA32_PERF_STATUS…….(0x198) : 0x221B00001E00
Jan 28 11:13:28 fabios-Mac-Pro kernel[0]: AICPUPMI: MSR_IA32_PERF_CONTROL……(0x199) : 0x1E00
Jan 28 11:13:28 fabios-Mac-Pro kernel[0]: AICPUPMI: IA32_CLOCK_MODULATION……(0x19A) : 0x0
Jan 28 11:13:28 fabios-Mac-Pro kernel[0]: AICPUPMI: IA32_THERM_STATUS……….(0x19C) : 0x884C0000
Jan 28 11:13:28 fabios-Mac-Pro kernel[0]: AICPUPMI: IA32_MISC_ENABLES……….(0x1A0) : 0x850089
Jan 28 11:13:28 fabios-Mac-Pro kernel[0]: AICPUPMI: MSR_MISC_PWR_MGMT……….(0x1AA) : 0x400001
Jan 28 11:13:28 fabios-Mac-Pro kernel[0]: AICPUPMI: MSR_TURBO_RATIO_LIMIT……(0x1AD) : 0x2121212121222324
Jan 28 11:13:28 fabios-Mac-Pro kernel[0]: AICPUPMI: IA32_ENERGY_PERF_BIAS……(0x1B0) : 0x0
Jan 28 11:13:28 fabios-Mac-Pro kernel[0]: AICPUPMI: MSR_POWER_CTL…………..(0x1FC) : 0x2104005B
Jan 28 11:13:28 fabios-Mac-Pro kernel[0]: AICPUPMI: MSR_RAPL_POWER_UNIT……..(0x606) : 0xA1003
Jan 28 11:13:28 fabios-Mac-Pro kernel[0]: AICPUPMI: MSR_PKG_POWER_LIMIT……..(0x610) : 0x68960005AFFFF
Jan 28 11:13:28 fabios-Mac-Pro kernel[0]: AICPUPMI: MSR_PKG_ENERGY_STATUS……(0x611) : 0x2C5EC673
Jan 28 11:13:28 fabios-Mac-Pro kernel[0]: AICPUPMI: MSR_PKGC3_IRTL………….(0x60a) : 0x0
Jan 28 11:13:28 fabios-Mac-Pro kernel[0]: AICPUPMI: MSR_PKGC6_IRTL………….(0x60b) : 0x0
Jan 28 11:13:28 fabios-Mac-Pro kernel[0]: AICPUPMI: MSR_PP0_CURRENT_CONFIG…..(0x601) : 0x14149480000640
Jan 28 11:13:28 fabios-Mac-Pro kernel[0]: AICPUPMI: MSR_PP0_POWER_LIMIT……..(0x638) : 0x80000000
Jan 28 11:13:28 fabios-Mac-Pro kernel[0]: AICPUPMI: MSR_PP0_ENERGY_STATUS……(0x639) : 0x195D9191
Jan 28 11:13:28 fabios-Mac-Pro kernel[0]: AICPUPMI: MSR_PP0_POLICY………….(0x63a) : 0x0
Jan 28 11:13:28 fabios-Mac-Pro kernel[0]: AICPUPMI: MSR_PKG_C2_RESIDENCY…….(0x60d) : 0x0
Jan 28 11:13:28 fabios-Mac-Pro kernel[0]: AICPUPMI: MSR_PKG_C3_RESIDENCY…….(0x3f8) : 0x0
Jan 28 11:13:28 fabios-Mac-Pro kernel[0]: AICPUPMI: MSR_PKG_C6_RESIDENCY…….(0x3f9) : 0x0
Jan 28 11:13:28 fabios-Mac-Pro kernel[0]: AICPUPMI: CPU Low Frequency Mode………….: 1200 MHz
Jan 28 11:13:28 fabios-Mac-Pro kernel[0]: AICPUPMI: CPU Maximum non-Turbo Frequency….: 3000 MHz
Jan 28 11:13:28 fabios-Mac-Pro kernel[0]: AICPUPMI: CPU Maximum Turbo Frequency……..: 3600 MHz
Jan 28 11:13:29 fabios-Mac-Pro kernel[0]: AICPUPMI: CPU P-States [ (12) 26 30 ]
Jan 28 11:13:29 fabios-Mac-Pro kernel[0]: AICPUPMI: CPU C6-Cores [ 2 3 5 9 10 13 14 15 16 ]
Jan 28 11:13:29 fabios-Mac-Pro kernel[0]: AICPUPMI: CPU P-States [ (12) 26 29 30 ]
Jan 28 11:13:29 fabios-Mac-Pro kernel[0]: AICPUPMI: CPU C6-Cores [ 2 3 4 5 6 7 8 9 10 11 13 14 15 16 ]
Jan 28 11:13:30 fabios-Mac-Pro kernel[0]: AICPUPMI: CPU P-States [ (12) 26 28 29 30 ]
Jan 28 11:13:30 fabios-Mac-Pro kernel[0]: AICPUPMI: CPU C6-Cores [ 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 ]
Jan 28 11:13:30 fabios-Mac-Pro kernel[0]: AICPUPMI: CPU P-States [ (12) 21 26 28 29 30 ]
Jan 28 11:13:32 fabios-Mac-Pro kernel[0]: AICPUPMI: CPU P-States [ (12) 21 25 26 28 29 30 ]
Jan 28 11:13:32 fabios-Mac-Pro kernel[0]: AICPUPMI: CPU P-States [ (12) 21 25 26 27 28 29 30 ]
Jan 28 11:13:33 fabios-Mac-Pro kernel[0]: AICPUPMI: CPU P-States [ (12) 21 22 25 26 27 28 29 30 ]
Jan 28 11:13:33 fabios-Mac-Pro kernel[0]: AICPUPMI: CPU P-States [ (12) 21 22 23 25 26 27 28 29 30 ]
Jan 28 11:13:34 fabios-Mac-Pro kernel[0]: AICPUPMI: CPU P-States [ (12) 18 21 22 23 25 26 27 28 29 30 ]
Jan 28 11:13:35 fabios-Mac-Pro kernel[0]: AICPUPMI: CPU P-States [ (12) 18 19 21 22 23 25 26 27 28 29 30 ]
Jan 28 11:13:36 fabios-Mac-Pro kernel[0]: AICPUPMI: CPU P-States [ (12) 16 18 19 21 22 23 25 26 27 28 29 30 ]
Jan 28 11:13:37 fabios-Mac-Pro kernel[0]: AICPUPMI: CPU P-States [ (12) 14 16 18 19 21 22 23 25 26 27 28 29 30 ]
Jan 28 11:13:39 fabios-Mac-Pro kernel[0]: AICPUPMI: CPU P-States [ (12) 14 16 18 19 21 22 23 24 25 26 27 28 29 30 ]
Jan 28 11:13:40 fabios-Mac-Pro kernel[0]: AICPUPMI: CPU P-States [ (12) 14 16 17 18 19 21 22 23 24 25 26 27 28 29 30 ]
Jan 28 11:13:41 fabios-Mac-Pro kernel[0]: AICPUPMI: CPU P-States [ (12) 13 14 16 17 18 19 21 22 23 24 25 26 27 28 29 30 ]
Jan 28 11:13:44 fabios-Mac-Pro kernel[0]: AICPUPMI: CPU P-States [ (12) 13 14 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 ]
Jan 28 11:13:47 fabios-Mac-Pro kernel[0]: AICPUPMI: CPU P-States [ (12) 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 ]
Cool! That’s more like it. Thank you for testing this!
p.s. I see that your UEFI BIOS uses 3GHz at boot time, rather than using 3.6GHz and thus your boot time can be reduced (by using the 600 MHz headroom).
following data are achieved changing from auto to 36 in max ratio default bios setting
AICPUPMI: MSR_TURBO_RATIO_LIMIT changes conseguently.
I would also say that I achieve same results without dropping OEM SSDT and without using custom SSDT.
Is it possible that all this is driven only by ASRock firmware?
PS. Same results with Speedstep disable or enabled in bios
Jan 29 09:04:47 fabios-Mac-Pro kernel[0]: AICPUPMI: v3.2 Copyright © 2012-2014 Pike R. Alpha. All rights reserved
Jan 29 09:04:47 fabios-Mac-Pro kernel[0]: AICPUPMI: logMSRs……………………….: 1
Jan 29 09:04:47 fabios-Mac-Pro kernel[0]: AICPUPMI: logIGPU……………………….: 0
Jan 29 09:04:47 fabios-Mac-Pro kernel[0]: AICPUPMI: logCStates…………………….: 1
Jan 29 09:04:47 fabios-Mac-Pro kernel[0]: AICPUPMI: logIPGStyle……………………: 1
Jan 29 09:04:47 fabios-Mac-Pro kernel[0]: AICPUPMI: MWAIT C-States…………………: 4384
Jan 29 09:04:47 fabios-Mac-Pro kernel[0]: AICPUPMI: MSR_CORE_THREAD_COUNT……(0x35) : 0xA0014
Jan 29 09:04:47 fabios-Mac-Pro kernel[0]: AICPUPMI: MSR_PLATFORM_INFO……….(0xCE) : 0xC10E4811E00
Jan 29 09:04:47 fabios-Mac-Pro kernel[0]: AICPUPMI: MSR_PMG_CST_CONFIG_CONTROL.(0xE2) : 0x401
Jan 29 09:04:47 fabios-Mac-Pro kernel[0]: AICPUPMI: MSR_PMG_IO_CAPTURE_BASE….(0xE4) : 0x414
Jan 29 09:04:47 fabios-Mac-Pro kernel[0]: AICPUPMI: IA32_MPERF……………..(0xE7) : 0x43878DE
Jan 29 09:04:47 fabios-Mac-Pro kernel[0]: AICPUPMI: IA32_APERF……………..(0xE8) : 0x4A282F5
Jan 29 09:04:47 fabios-Mac-Pro kernel[0]: AICPUPMI: MSR_FLEX_RATIO………….(0x194) : 0x0
Jan 29 09:04:47 fabios-Mac-Pro kernel[0]: AICPUPMI: MSR_IA32_PERF_STATUS…….(0x198) : 0x25A100002100
Jan 29 09:04:47 fabios-Mac-Pro kernel[0]: AICPUPMI: MSR_IA32_PERF_CONTROL……(0x199) : 0x2400
Jan 29 09:04:47 fabios-Mac-Pro kernel[0]: AICPUPMI: IA32_CLOCK_MODULATION……(0x19A) : 0x0
Jan 29 09:04:47 fabios-Mac-Pro kernel[0]: AICPUPMI: IA32_THERM_STATUS……….(0x19C) : 0x88470000
Jan 29 09:04:47 fabios-Mac-Pro kernel[0]: AICPUPMI: IA32_MISC_ENABLES……….(0x1A0) : 0x850089
Jan 29 09:04:47 fabios-Mac-Pro kernel[0]: AICPUPMI: MSR_MISC_PWR_MGMT……….(0x1AA) : 0x400001
Jan 29 09:04:47 fabios-Mac-Pro kernel[0]: AICPUPMI: MSR_TURBO_RATIO_LIMIT……(0x1AD) : 0x2424242424242424
Jan 29 09:04:47 fabios-Mac-Pro kernel[0]: AICPUPMI: IA32_ENERGY_PERF_BIAS……(0x1B0) : 0x0
Jan 29 09:04:47 fabios-Mac-Pro kernel[0]: AICPUPMI: MSR_POWER_CTL…………..(0x1FC) : 0x2104005B
Jan 29 09:04:47 fabios-Mac-Pro kernel[0]: AICPUPMI: MSR_RAPL_POWER_UNIT……..(0x606) : 0xA1003
Jan 29 09:04:47 fabios-Mac-Pro kernel[0]: AICPUPMI: MSR_PKG_POWER_LIMIT……..(0x610) : 0x68960005AFFFF
Jan 29 09:04:47 fabios-Mac-Pro kernel[0]: AICPUPMI: MSR_PKG_ENERGY_STATUS……(0x611) : 0xC536FFD
Jan 29 09:04:47 fabios-Mac-Pro kernel[0]: AICPUPMI: MSR_PKGC3_IRTL………….(0x60a) : 0x0
Jan 29 09:04:47 fabios-Mac-Pro kernel[0]: AICPUPMI: MSR_PKGC6_IRTL………….(0x60b) : 0x0
Jan 29 09:04:47 fabios-Mac-Pro kernel[0]: AICPUPMI: MSR_PP0_CURRENT_CONFIG…..(0x601) : 0x14149480000640
Jan 29 09:04:47 fabios-Mac-Pro kernel[0]: AICPUPMI: MSR_PP0_POWER_LIMIT……..(0x638) : 0x80000000
Jan 29 09:04:47 fabios-Mac-Pro kernel[0]: AICPUPMI: MSR_PP0_ENERGY_STATUS……(0x639) : 0x89546FD
Jan 29 09:04:47 fabios-Mac-Pro kernel[0]: AICPUPMI: MSR_PP0_POLICY………….(0x63a) : 0x0
Jan 29 09:04:47 fabios-Mac-Pro kernel[0]: AICPUPMI: MSR_PKG_C2_RESIDENCY…….(0x60d) : 0x0
Jan 29 09:04:47 fabios-Mac-Pro kernel[0]: AICPUPMI: MSR_PKG_C3_RESIDENCY…….(0x3f8) : 0x0
Jan 29 09:04:47 fabios-Mac-Pro kernel[0]: AICPUPMI: MSR_PKG_C6_RESIDENCY…….(0x3f9) : 0x0
Jan 29 09:04:47 fabios-Mac-Pro kernel[0]: AICPUPMI: CPU Low Frequency Mode………….: 1200 MHz
Jan 29 09:04:47 fabios-Mac-Pro kernel[0]: AICPUPMI: CPU Maximum non-Turbo Frequency….: 3000 MHz
Jan 29 09:04:47 fabios-Mac-Pro kernel[0]: AICPUPMI: CPU Maximum Turbo Frequency……..: 3600 MHz
Jan 29 09:04:48 fabios-Mac-Pro kernel[0]: AICPUPMI: CPU P-States [ (12) 29 33 ]
Jan 29 09:04:48 fabios-Mac-Pro kernel[0]: AICPUPMI: CPU C6-Cores [ 2 3 4 8 9 10 11 12 14 15 16 ]
Jan 29 09:04:49 fabios-Mac-Pro kernel[0]: AICPUPMI: CPU P-States [ (12) 29 30 33 ]
Jan 29 09:04:49 fabios-Mac-Pro kernel[0]: AICPUPMI: CPU C6-Cores [ 2 3 4 5 7 8 9 10 11 12 13 14 15 16 ]
Jan 29 09:04:49 fabios-Mac-Pro kernel[0]: AICPUPMI: CPU C6-Cores [ 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 ]
Jan 29 09:04:50 fabios-Mac-Pro kernel[0]: AICPUPMI: CPU P-States [ (12) 14 29 30 33 ]
Jan 29 09:04:50 fabios-Mac-Pro kernel[0]: AICPUPMI: CPU C6-Cores [ 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 ]
Jan 29 09:04:51 fabios-Mac-Pro kernel[0]: AICPUPMI: CPU P-States [ (12) 14 28 29 30 33 ]
Jan 29 09:04:51 fabios-Mac-Pro kernel[0]: AICPUPMI: CPU P-States [ (12) 14 25 28 29 30 33 ]
Jan 29 09:04:53 fabios-Mac-Pro kernel[0]: AICPUPMI: CPU P-States [ 12 14 25 28 29 30 32 (33) ]
Jan 29 09:04:55 fabios-Mac-Pro kernel[0]: AICPUPMI: CPU P-States [ (12) 14 25 26 28 29 30 32 33 ]
Jan 29 09:04:57 fabios-Mac-Pro kernel[0]: AICPUPMI: CPU P-States [ (12) 14 25 26 27 28 29 30 32 33 ]
Jan 29 09:04:58 fabios-Mac-Pro kernel[0]: AICPUPMI: CPU P-States [ (12) 14 24 25 26 27 28 29 30 32 33 ]
Jan 29 09:05:00 fabios-Mac-Pro kernel[0]: AICPUPMI: CPU P-States [ (12) 14 24 25 26 27 28 29 30 31 32 33 ]
Jan 29 09:05:01 fabios-Mac-Pro kernel[0]: AICPUPMI: CPU P-States [ (12) 14 19 24 25 26 27 28 29 30 31 32 33 ]
Jan 29 09:05:02 fabios-Mac-Pro kernel[0]: AICPUPMI: CPU P-States [ 12 14 19 23 24 25 26 27 28 29 30 31 32 (33) ]
Jan 29 09:05:03 fabios-Mac-Pro kernel[0]: AICPUPMI: CPU P-States [ (12) 14 19 22 23 24 25 26 27 28 29 30 31 32 33 ]
Jan 29 09:05:46 fabios-Mac-Pro kernel[0]: AICPUPMI: CPU P-States [ (12) 14 17 19 22 23 24 25 26 27 28 29 30 31 32 33 ]
Jan 29 09:05:48 fabios-Mac-Pro kernel[0]: AICPUPMI: CPU P-States [ 12 14 17 19 21 22 23 24 25 26 27 28 29 30 31 32 (33) ]
Jan 29 09:05:54 fabios-Mac-Pro kernel[0]: AICPUPMI: CPU P-States [ (12) 14 17 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 ]
Jan 29 09:06:09 fabios-Mac-Pro kernel[0]: AICPUPMI: CPU P-States [ (12) 14 16 17 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 ]
Jan 29 09:06:56 fabios-Mac-Pro kernel[0]: AICPUPMI: CPU P-States [ 12 14 16 17 19 20 21 22 23 24 25 26 27 28 29 30 31 32 (33) 34 ]
Jan 29 09:07:47 fabios-Mac-Pro kernel[0]: AICPUPMI: CPU P-States [ (12) 14 15 16 17 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 ]
Jan 29 09:07:58 fabios-Mac-Pro kernel[0]: AICPUPMI: CPU P-States [ (12) 13 14 15 16 17 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 ]
Jan 29 09:08:06 fabios-Mac-Pro kernel[0]: AICPUPMI: CPU P-States [ (12) 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 ]
Sorry, but it doesn’t work like that. Your boot loader should do something like this:
https://github.com/Piker-Alpha/RevoBoot/blob/clang/i386/libsaio/cpu.c#L184
I tried on my system P9X79 Pro, 4960X, but it causes the system to lock up on kextload. Ive tried disabling the options in info.plist, but even all 3 at false causes a lock up or sometimes kernel panic.
You need to compile the kext yourself.
Hey Pike, is there a reason why AICPUPMI outputs all P-States up to 23 (my turbo frequency is 2300MHz) while HWMonitor and Intel Power Gadget report at maximum only 2000MHz? Is that expected? I’m using an SSDT created by the latest version of your script. Thank you!
Can’t tell. Need more info like CPU model and the used version of IPG.
i5-2467M
Intel Power Gadget 3.0.1
Thanks!
Just noticed these apps actually show more than 2GHz sometimes but it stays at turbo frequency for a really small period of time. It happens like when dragging something with lots of constraints in Xcode. However if I run “yes” to the CPUs, the frequency stays fixed at 2GHz. Don’t know if that’s normal either. I believe it should get more often to real turbo 2.3GHz frequency, at least on Windows it stays between 2 and 2.3 almost the whole time.
You may want to read this: http://www.apple.com/osx/advanced-technologies/
In short; Apple improved their power management routines to reduce power consumption and extend battery life. This should also help people with desktop computers since it runs more power efficiently on Mavericks. Maybe that is why Microsoft Windows spikes up to 2.3GHz when it shouldn’t, or is IPG just so much different for Windows?
Hello Pike
3930K but I’m trying on KP also changing
: PlogMSRs (true) a (false)
logIGPU (false)
logCStates (true) a (false)
advice
Thanks
Try to compile the kext yourself, because that worked for others.
Hi Pike,
also a little late, but I compiled AICPM 3.3 from the GitHub Repo here as an upadate, because I was still using v1.0… and your build didn’t work. So, yes, I followed your advice (+ left the IGPU logging = false for now).
Guess what: Success. It works flawlessly on my (old) Acer V3-771G.
I’m sending you the output from bdmesg & the kext (just for the records):
CPU: Brand String: Intel(R) Core(TM) i7-3610QM CPU @ 2.30GHz
CPU: Vendor/Family/ExtFamily: 0x756e6547/0x6/0x0
CPU: Model/ExtModel/Stepping: 0x3a/0x3/0x9
CPU: MaxCoef/CurrCoef: 0x0/0x17
CPU: MaxDiv/CurrDiv: 0x0/0x0
CPU: TSCFreq: 2294MHz
CPU: FSBFreq: 99MHz
CPU: CPUFreq: 2294MHz
CPU: NoCores/NoThreads: 4/8
CPU: Features: 0x000002ff
29.06.14 15:21:27,000 kernel[0]: AICPUPMI: v3.3 Copyright © 2012-2014 Pike R. Alpha. All rights reserved
29.06.14 15:21:27,000 kernel[0]: AICPUPMI: logMSRs……………………….: 1
29.06.14 15:21:27,000 kernel[0]: AICPUPMI: logIGPU……………………….: 0
29.06.14 15:21:27,000 kernel[0]: AICPUPMI: logCStates…………………….: 1
29.06.14 15:21:27,000 kernel[0]: AICPUPMI: logIPGStyle……………………: 1
29.06.14 15:21:27,000 kernel[0]: AICPUPMI: MWAIT C-States…………………: 135456
29.06.14 15:21:27,000 kernel[0]: AICPUPMI: MSR_CORE_THREAD_COUNT……(0x35) : 0x40008
29.06.14 15:21:27,000 kernel[0]: AICPUPMI: MSR_PLATFORM_INFO……….(0xCE) : 0x80C10E0011700
29.06.14 15:21:27,000 kernel[0]: AICPUPMI: MSR_PMG_CST_CONFIG_CONTROL.(0xE2) : 0x1E008404
29.06.14 15:21:27,000 kernel[0]: AICPUPMI: MSR_PMG_IO_CAPTURE_BASE….(0xE4) : 0x20414
29.06.14 15:21:27,000 kernel[0]: AICPUPMI: IA32_MPERF……………..(0xE7) : 0x711AC17EA
29.06.14 15:21:27,000 kernel[0]: AICPUPMI: IA32_APERF……………..(0xE8) : 0x70B130965
29.06.14 15:21:27,000 kernel[0]: AICPUPMI: MSR_FLEX_RATIO………….(0x194) : 0x100000
29.06.14 15:21:27,000 kernel[0]: AICPUPMI: MSR_IA32_PERF_STATUS…….(0x198) : 0x1E1A00001700
29.06.14 15:21:27,000 kernel[0]: AICPUPMI: MSR_IA32_PERF_CONTROL……(0x199) : 0x1700
29.06.14 15:21:27,000 kernel[0]: AICPUPMI: IA32_CLOCK_MODULATION……(0x19A) : 0x8
29.06.14 15:21:27,000 kernel[0]: AICPUPMI: IA32_THERM_STATUS……….(0x19C) : 0x88330000
29.06.14 15:21:27,000 kernel[0]: AICPUPMI: IA32_MISC_ENABLES……….(0x1A0) : 0x850089
29.06.14 15:21:27,000 kernel[0]: AICPUPMI: MSR_MISC_PWR_MGMT……….(0x1AA) : 0x400001
29.06.14 15:21:27,000 kernel[0]: AICPUPMI: MSR_TURBO_RATIO_LIMIT……(0x1AD) : 0x1F1F2021
29.06.14 15:21:27,000 kernel[0]: AICPUPMI: IA32_ENERGY_PERF_BIAS……(0x1B0) : 0x4
29.06.14 15:21:27,000 kernel[0]: AICPUPMI: MSR_POWER_CTL…………..(0x1FC) : 0x14005F
29.06.14 15:21:27,000 kernel[0]: AICPUPMI: MSR_RAPL_POWER_UNIT……..(0x606) : 0xA1003
29.06.14 15:21:27,000 kernel[0]: AICPUPMI: MSR_PKG_POWER_LIMIT……..(0x610) : 0x800081C200DC8168
29.06.14 15:21:27,000 kernel[0]: AICPUPMI: MSR_PKG_ENERGY_STATUS……(0x611) : 0x976D7E
29.06.14 15:21:27,000 kernel[0]: AICPUPMI: MSR_PKGC3_IRTL………….(0x60a) : 0x883B
29.06.14 15:21:27,000 kernel[0]: AICPUPMI: MSR_PKGC6_IRTL………….(0x60b) : 0x8850
29.06.14 15:21:27,000 kernel[0]: AICPUPMI: MSR_PKGC7_IRTL………….(0x60c) : 0x8857
29.06.14 15:21:27,000 kernel[0]: AICPUPMI: MSR_PP0_CURRENT_CONFIG…..(0x601) : 0x1814149480000380
29.06.14 15:21:27,000 kernel[0]: AICPUPMI: MSR_PP0_POWER_LIMIT……..(0x638) : 0x0
29.06.14 15:21:27,000 kernel[0]: AICPUPMI: MSR_PP0_ENERGY_STATUS……(0x639) : 0x5F4EF0
29.06.14 15:21:27,000 kernel[0]: AICPUPMI: MSR_PP0_POLICY………….(0x63a) : 0x0
29.06.14 15:21:27,000 kernel[0]: AICPUPMI: MSR_CONFIG_TDP_NOMINAL…..(0x648) : 0x17
29.06.14 15:21:27,000 kernel[0]: AICPUPMI: MSR_CONFIG_TDP_LEVEL1……(0x649) : 0x120000000000000
29.06.14 15:21:27,000 kernel[0]: AICPUPMI: MSR_CONFIG_TDP_LEVEL2……(0x64a) : 0x120000000000000
29.06.14 15:21:27,000 kernel[0]: AICPUPMI: MSR_CONFIG_TDP_CONTROL…..(0x64b) : 0x80000000
29.06.14 15:21:27,000 kernel[0]: AICPUPMI: MSR_TURBO_ACTIVATION_RATIO.(0x64c) : 0x0
29.06.14 15:21:27,000 kernel[0]: AICPUPMI: MSR_PKG_C2_RESIDENCY…….(0x60d) : 0x0
29.06.14 15:21:27,000 kernel[0]: AICPUPMI: MSR_PKG_C3_RESIDENCY…….(0x3f8) : 0x0
29.06.14 15:21:27,000 kernel[0]: AICPUPMI: MSR_PKG_C6_RESIDENCY…….(0x3f9) : 0x0
29.06.14 15:21:27,000 kernel[0]: AICPUPMI: MSR_PKG_C7_RESIDENCY…….(0x3fa) : 0x0
29.06.14 15:21:27,000 kernel[0]: AICPUPMI: CPU Low Frequency Mode………….: 1200 MHz
29.06.14 15:21:27,000 kernel[0]: AICPUPMI: CPU Maximum non-Turbo Frequency….: 2300 MHz
29.06.14 15:21:27,000 kernel[0]: AICPUPMI: CPU Maximum Turbo Frequency……..: 3300 MHz
29.06.14 15:21:33,000 kernel[0]: AICPUPMI: CPU C7-Cores [ 0 1 2 3 4 5 6 7 ]
29.06.14 15:21:37,000 kernel[0]: AICPUPMI: CPU C3-Cores [ 0 1 2 3 4 5 6 7 ]
29.06.14 15:21:38,000 kernel[0]: AICPUPMI: CPU C6-Cores [ 0 1 2 3 4 5 6 7 ]
29.06.14 15:29:20,000 kernel[0]: AICPUPMI: CPU P-States [ 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 (33) ]
Well, like all the others say: Please keep up the good work, because most of our machines won’t be running this smooth without you (or at all).
Thanks from Germany.
Hi There, thanks for the tone of info and your work… Just to report that on a GA-X79-UP4 and a 4930K I got KPs whatever the options I set in the Info.plist…
Found an older version that does’t KP, only reported P-states is 39 which explained why my 6 cores are stuck at 3.9 ghz… I’m clueless… Generated the sdst with ssdtPRgen -w 3 and got “Number of Turbo States: 5 (3500-3900 MHz)
Number of P-States: 28 (1200-3900 MHz)”
Anyone, Any idea of where I could start searching ??
just in case someone would have some time here’s the bdmesg output : http://pastebin.com/N7geHUpK
Hello Pike!
With logIPGStyle=true, I get:
AICPUPMI: CPU P-States [ 8 9 10 11 12 13 14 15 16 (17) 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 ]
logIPGStyle=false:
AICPUPMI: CPU P-States [ 8 17 34 35 (36) 37 38 ]
Why is that?
logIPGStyle=true shows all power state changes. Not just the P-States triggered by AppleIntelCPUPowermanagement.kext or XCPM so what you see is pretty normal so to speak.
Hello Pikeralpha,
i am testing my powermanagement and see if i have the right values with my configuration :
Probook 4540s,i3-3110m, HD 4000. Using Clover UEFI, SSDT created with your script. OS X Yosemite.
with DPCIManager i get only 3 pstats ( 12,22,24 )
i have installed your kext and and i get this
Oct 26 11:13:39 Samirs-MacBook-Pro kernel[0]: AICPUPMI: v3.3 Copyright © 2012-2014 Pike R. Alpha. All rights reserved
Oct 26 11:13:39 Samirs-MacBook-Pro kernel[0]: AICPUPMI: logMSRs……………………….: 1
Oct 26 11:13:39 Samirs-MacBook-Pro kernel[0]: AICPUPMI: logIGPU……………………….: 1
Oct 26 11:13:39 Samirs-MacBook-Pro kernel[0]: AICPUPMI: logCStates…………………….: 1
Oct 26 11:13:39 Samirs-MacBook-Pro kernel[0]: AICPUPMI: logIPGStyle……………………: 1
Oct 26 11:13:39 Samirs-MacBook-Pro kernel[0]: AICPUPMI: MWAIT C-States…………………: 135456
Oct 26 11:13:39 Samirs-MacBook-Pro kernel[0]: AICPUPMI: MSR_CORE_THREAD_COUNT……(0x35) : 0x20004
Oct 26 11:13:39 Samirs-MacBook-Pro kernel[0]: AICPUPMI: MSR_PLATFORM_INFO……….(0xCE) : 0x80C10E0011800
Oct 26 11:13:39 Samirs-MacBook-Pro kernel[0]: AICPUPMI: MSR_PMG_CST_CONFIG_CONTROL.(0xE2) : 0x1E008404
Oct 26 11:13:39 Samirs-MacBook-Pro kernel[0]: AICPUPMI: MSR_PMG_IO_CAPTURE_BASE….(0xE4) : 0x20414
Oct 26 11:13:39 Samirs-MacBook-Pro kernel[0]: AICPUPMI: IA32_MPERF……………..(0xE7) : 0x527F66A1AA
Oct 26 11:13:39 Samirs-MacBook-Pro kernel[0]: AICPUPMI: IA32_APERF……………..(0xE8) : 0x499AF454C0
Oct 26 11:13:39 Samirs-MacBook-Pro kernel[0]: AICPUPMI: MSR_FLEX_RATIO………….(0x194) : 0x110000
Oct 26 11:13:39 Samirs-MacBook-Pro kernel[0]: AICPUPMI: MSR_IA32_PERF_STATUS…….(0x198) : 0x20D300001800
Oct 26 11:13:39 Samirs-MacBook-Pro kernel[0]: AICPUPMI: MSR_IA32_PERF_CONTROL……(0x199) : 0x1800
Oct 26 11:13:39 Samirs-MacBook-Pro kernel[0]: AICPUPMI: IA32_CLOCK_MODULATION……(0x19A) : 0x0
Oct 26 11:13:39 Samirs-MacBook-Pro kernel[0]: AICPUPMI: IA32_THERM_STATUS……….(0x19C) : 0x88350000
Oct 26 11:13:39 Samirs-MacBook-Pro kernel[0]: AICPUPMI: IA32_MISC_ENABLES……….(0x1A0) : 0x850089
Oct 26 11:13:39 Samirs-MacBook-Pro kernel[0]: AICPUPMI: MSR_MISC_PWR_MGMT……….(0x1AA) : 0x400001
Oct 26 11:13:39 Samirs-MacBook-Pro kernel[0]: AICPUPMI: MSR_TURBO_RATIO_LIMIT……(0x1AD) : 0x18181818
Oct 26 11:13:39 Samirs-MacBook-Pro kernel[0]: AICPUPMI: IA32_ENERGY_PERF_BIAS……(0x1B0) : 0xF
Oct 26 11:13:39 Samirs-MacBook-Pro kernel[0]: AICPUPMI: MSR_POWER_CTL…………..(0x1FC) : 0x4005F
Oct 26 11:13:39 Samirs-MacBook-Pro kernel[0]: AICPUPMI: MSR_RAPL_POWER_UNIT……..(0x606) : 0xA1003
Oct 26 11:13:39 Samirs-MacBook-Pro kernel[0]: AICPUPMI: MSR_PKG_POWER_LIMIT……..(0x610) : 0x815E00DC8118
Oct 26 11:13:39 Samirs-MacBook-Pro kernel[0]: AICPUPMI: MSR_PKG_ENERGY_STATUS……(0x611) : 0xEECE644
Oct 26 11:13:39 Samirs-MacBook-Pro kernel[0]: AICPUPMI: MSR_PKGC3_IRTL………….(0x60a) : 0x883B
Oct 26 11:13:39 Samirs-MacBook-Pro kernel[0]: AICPUPMI: MSR_PKGC6_IRTL………….(0x60b) : 0x8850
Oct 26 11:13:39 Samirs-MacBook-Pro kernel[0]: AICPUPMI: MSR_PKGC7_IRTL………….(0x60c) : 0x8857
Oct 26 11:13:39 Samirs-MacBook-Pro kernel[0]: AICPUPMI: MSR_PP0_CURRENT_CONFIG…..(0x601) : 0x1814149480000380
Oct 26 11:13:39 Samirs-MacBook-Pro kernel[0]: AICPUPMI: MSR_PP0_POWER_LIMIT……..(0x638) : 0x0
Oct 26 11:13:39 Samirs-MacBook-Pro kernel[0]: AICPUPMI: MSR_PP0_ENERGY_STATUS……(0x639) : 0x735E83B
Oct 26 11:13:39 Samirs-MacBook-Pro kernel[0]: AICPUPMI: MSR_PP0_POLICY………….(0x63a) : 0x0
Oct 26 11:13:39 Samirs-MacBook-Pro kernel[0]: AICPUPMI: MSR_PP1_CURRENT_CONFIG…..(0x602) : 0x1814149480000170
Oct 26 11:13:39 Samirs-MacBook-Pro kernel[0]: AICPUPMI: MSR_PP1_POWER_LIMIT……..(0x640) : 0x0
Oct 26 11:13:39 Samirs-MacBook-Pro kernel[0]: AICPUPMI: MSR_PP1_ENERGY_STATUS……(0x641) : 0xF621A3
Oct 26 11:13:39 Samirs-MacBook-Pro kernel[0]: AICPUPMI: MSR_PP1_POLICY………….(0x642) : 0x10
Oct 26 11:13:39 Samirs-MacBook-Pro kernel[0]: AICPUPMI: MSR_CONFIG_TDP_NOMINAL…..(0x648) : 0x18
Oct 26 11:13:39 Samirs-MacBook-Pro kernel[0]: AICPUPMI: MSR_CONFIG_TDP_LEVEL1……(0x649) : 0xC0000000000000
Oct 26 11:13:39 Samirs-MacBook-Pro kernel[0]: AICPUPMI: MSR_CONFIG_TDP_LEVEL2……(0x64a) : 0xC0000000000000
Oct 26 11:13:39 Samirs-MacBook-Pro kernel[0]: AICPUPMI: MSR_CONFIG_TDP_CONTROL…..(0x64b) : 0x80000000
Oct 26 11:13:39 Samirs-MacBook-Pro kernel[0]: AICPUPMI: MSR_TURBO_ACTIVATION_RATIO.(0x64c) : 0x0
Oct 26 11:13:39 Samirs-MacBook-Pro kernel[0]: AICPUPMI: MSR_PKG_C2_RESIDENCY…….(0x60d) : 0x165D9BEAD8
Oct 26 11:13:39 Samirs-MacBook-Pro kernel[0]: AICPUPMI: MSR_PKG_C3_RESIDENCY…….(0x3f8) : 0x23DED330
Oct 26 11:13:39 Samirs-MacBook-Pro kernel[0]: AICPUPMI: MSR_PKG_C6_RESIDENCY…….(0x3f9) : 0x1C8A38A638
Oct 26 11:13:39 Samirs-MacBook-Pro kernel[0]: AICPUPMI: MSR_PKG_C7_RESIDENCY…….(0x3fa) : 0x10868955018
Oct 26 11:13:39 Samirs-MacBook-Pro kernel[0]: AICPUPMI: CPU Low Frequency Mode………….: 1200 MHz
Oct 26 11:13:39 Samirs-MacBook-Pro kernel[0]: AICPUPMI: CPU Maximum non-Turbo Frequency….: 2400 MHz
Oct 26 11:13:39 Samirs-MacBook-Pro kernel[0]: AICPUPMI: CPU Maximum Turbo Frequency……..: 2400 MHz
Oct 26 11:13:39 Samirs-MacBook-Pro kernel[0]: AICPUPMI: IGPU Current Frequency………….: 350 MHz
Oct 26 11:13:39 Samirs-MacBook-Pro kernel[0]: AICPUPMI: IGPU Minimum Frequency………….: 350 MHz
Oct 26 11:13:39 Samirs-MacBook-Pro kernel[0]: AICPUPMI: IGPU Maximum Non-Turbo Frequency…: 650 MHz
Oct 26 11:13:39 Samirs-MacBook-Pro kernel[0]: AICPUPMI: IGPU Maximum Turbo Frequency…….: 1000 MHz
Oct 26 11:13:39 Samirs-MacBook-Pro kernel[0]: AICPUPMI: IGPU Maximum limit……………..: No Limit
Oct 26 11:13:40 Samirs-MacBook-Pro kernel[0]: AICPUPMI: CPU P-States [ (18) 20 24 ] iGPU P-States [ (7) ]
Oct 26 11:13:40 Samirs-MacBook-Pro kernel[0]: AICPUPMI: CPU C3-Cores [ 0 1 2 ]
Oct 26 11:13:40 Samirs-MacBook-Pro kernel[0]: AICPUPMI: CPU C6-Cores [ 0 2 3 ]
Oct 26 11:13:40 Samirs-MacBook-Pro kernel[0]: AICPUPMI: CPU C7-Cores [ 0 2 3 ]
Oct 26 11:13:41 Samirs-MacBook-Pro kernel[0]: AICPUPMI: CPU P-States [ (12) 17 18 20 24 ] iGPU P-States [ (7) ]
Oct 26 11:13:41 Samirs-MacBook-Pro kernel[0]: AICPUPMI: CPU C3-Cores [ 0 1 2 3 ]
Oct 26 11:13:41 Samirs-MacBook-Pro kernel[0]: AICPUPMI: CPU C6-Cores [ 0 1 2 3 ]
Oct 26 11:13:41 Samirs-MacBook-Pro kernel[0]: AICPUPMI: CPU C7-Cores [ 0 1 2 3 ]
Oct 26 11:13:42 Samirs-MacBook-Pro kernel[0]: AICPUPMI: CPU P-States [ 12 17 18 19 20 (24) ] iGPU P-States [ (7) ]
Oct 26 11:13:43 Samirs-MacBook-Pro kernel[0]: AICPUPMI: CPU P-States [ 12 17 18 19 20 21 (24) ] iGPU P-States [ (7) ]
Oct 26 11:14:06 Samirs-MacBook-Pro kernel[0]: AICPUPMI: CPU P-States [ (12) 16 17 18 19 20 21 24 ] iGPU P-States [ (7) ]
Oct 26 11:14:21 Samirs-MacBook-Pro kernel[0]: AICPUPMI: CPU P-States [ 12 16 17 18 19 20 21 23 (24) ] iGPU P-States [ (7) ]
Oct 26 11:14:21 Samirs-MacBook-Pro kernel[0]: AICPUPMI: CPU P-States [ (12) 16 17 18 19 20 21 22 23 24 ] iGPU P-States [ (7) ]
Oct 26 11:14:57 Samirs-MacBook-Pro kernel[0]: AICPUPMI: CPU P-States [ (12) 15 16 17 18 19 20 21 22 23 24 ] iGPU P-States [ (7) ]
Oct 26 11:18:29 Samirs-MacBook-Pro kernel[0]: AICPUPMI: CPU P-States [ (12) 13 15 16 17 18 19 20 21 22 23 24 ] iGPU P-States [ (7) ]
Oct 26 11:19:00 Samirs-MacBook-Pro kernel[0]: AICPUPMI: CPU P-States [ (12) 13 14 15 16 17 18 19 20 21 22 23 24 ] iGPU P-States [ (7) ]
Oct 26 12:19:40 localhost kernel[0]: AICPUPMI: v3.3 Copyright © 2012-2014 Pike R. Alpha. All rights reserved
Oct 26 12:19:40 localhost kernel[0]: AICPUPMI: logMSRs……………………….: 1
Oct 26 12:19:40 localhost kernel[0]: AICPUPMI: logIGPU……………………….: 1
Oct 26 12:19:40 localhost kernel[0]: AICPUPMI: logCStates…………………….: 1
Oct 26 12:19:40 localhost kernel[0]: AICPUPMI: logIPGStyle……………………: 1
Oct 26 12:19:40 localhost kernel[0]: AICPUPMI: MWAIT C-States…………………: 135456
Oct 26 12:19:40 localhost kernel[0]: AICPUPMI: MSR_CORE_THREAD_COUNT……(0x35) : 0x20004
Oct 26 12:19:40 localhost kernel[0]: AICPUPMI: MSR_PLATFORM_INFO……….(0xCE) : 0x80C10E0011800
Oct 26 12:19:40 localhost kernel[0]: AICPUPMI: MSR_PMG_CST_CONFIG_CONTROL.(0xE2) : 0x1E008404
Oct 26 12:19:40 localhost kernel[0]: AICPUPMI: MSR_PMG_IO_CAPTURE_BASE….(0xE4) : 0x20414
Oct 26 12:19:40 localhost kernel[0]: AICPUPMI: IA32_MPERF……………..(0xE7) : 0x56CADEE8C
Oct 26 12:19:40 localhost kernel[0]: AICPUPMI: IA32_APERF……………..(0xE8) : 0x565282672
Oct 26 12:19:40 localhost kernel[0]: AICPUPMI: MSR_FLEX_RATIO………….(0x194) : 0x100000
Oct 26 12:19:40 localhost kernel[0]: AICPUPMI: MSR_IA32_PERF_STATUS…….(0x198) : 0x214E00001800
Oct 26 12:19:40 localhost kernel[0]: AICPUPMI: MSR_IA32_PERF_CONTROL……(0x199) : 0x1818
Oct 26 12:19:40 localhost kernel[0]: AICPUPMI: IA32_CLOCK_MODULATION……(0x19A) : 0x0
Oct 26 12:19:40 localhost kernel[0]: AICPUPMI: IA32_THERM_STATUS……….(0x19C) : 0x883A0000
Oct 26 12:19:40 localhost kernel[0]: AICPUPMI: IA32_MISC_ENABLES……….(0x1A0) : 0x850089
Oct 26 12:19:40 localhost kernel[0]: AICPUPMI: MSR_MISC_PWR_MGMT……….(0x1AA) : 0x400000
Oct 26 12:19:40 localhost kernel[0]: AICPUPMI: MSR_TURBO_RATIO_LIMIT……(0x1AD) : 0x18181818
Oct 26 12:19:40 localhost kernel[0]: AICPUPMI: IA32_ENERGY_PERF_BIAS……(0x1B0) : 0x0
Oct 26 12:19:40 localhost kernel[0]: AICPUPMI: MSR_POWER_CTL…………..(0x1FC) : 0x14005F
Oct 26 12:19:40 localhost kernel[0]: AICPUPMI: MSR_RAPL_POWER_UNIT……..(0x606) : 0xA1003
Oct 26 12:19:40 localhost kernel[0]: AICPUPMI: MSR_PKG_POWER_LIMIT……..(0x610) : 0x815E00DC8118
Oct 26 12:19:40 localhost kernel[0]: AICPUPMI: MSR_PKG_ENERGY_STATUS……(0x611) : 0x5154EB
Oct 26 12:19:40 localhost kernel[0]: AICPUPMI: MSR_PKGC3_IRTL………….(0x60a) : 0x883B
Oct 26 12:19:40 localhost kernel[0]: AICPUPMI: MSR_PKGC6_IRTL………….(0x60b) : 0x8850
Oct 26 12:19:40 localhost kernel[0]: AICPUPMI: MSR_PKGC7_IRTL………….(0x60c) : 0x8857
Oct 26 12:19:40 localhost kernel[0]: AICPUPMI: MSR_PP0_CURRENT_CONFIG…..(0x601) : 0x1814149480000380
Oct 26 12:19:40 localhost kernel[0]: AICPUPMI: MSR_PP0_POWER_LIMIT……..(0x638) : 0x0
Oct 26 12:19:40 localhost kernel[0]: AICPUPMI: MSR_PP0_ENERGY_STATUS……(0x639) : 0x3ADF53
Oct 26 12:19:40 localhost kernel[0]: AICPUPMI: MSR_PP0_POLICY………….(0x63a) : 0x0
Oct 26 12:19:40 localhost kernel[0]: AICPUPMI: MSR_PP1_CURRENT_CONFIG…..(0x602) : 0x1814149480000170
Oct 26 12:19:40 localhost kernel[0]: AICPUPMI: MSR_PP1_POWER_LIMIT……..(0x640) : 0x0
Oct 26 12:19:40 localhost kernel[0]: AICPUPMI: MSR_PP1_ENERGY_STATUS……(0x641) : 0x2B5F3
Oct 26 12:19:40 localhost kernel[0]: AICPUPMI: MSR_PP1_POLICY………….(0x642) : 0x10
Oct 26 12:19:40 localhost kernel[0]: AICPUPMI: MSR_CONFIG_TDP_NOMINAL…..(0x648) : 0x18
Oct 26 12:19:40 localhost kernel[0]: AICPUPMI: MSR_CONFIG_TDP_LEVEL1……(0x649) : 0xC0000000000000
Oct 26 12:19:40 localhost kernel[0]: AICPUPMI: MSR_CONFIG_TDP_LEVEL2……(0x64a) : 0xC0000000000000
Oct 26 12:19:40 localhost kernel[0]: AICPUPMI: MSR_CONFIG_TDP_CONTROL…..(0x64b) : 0x80000000
Oct 26 12:19:40 localhost kernel[0]: AICPUPMI: MSR_TURBO_ACTIVATION_RATIO.(0x64c) : 0x0
Oct 26 12:19:40 localhost kernel[0]: AICPUPMI: MSR_PKG_C2_RESIDENCY…….(0x60d) : 0x0
Oct 26 12:19:40 localhost kernel[0]: AICPUPMI: MSR_PKG_C3_RESIDENCY…….(0x3f8) : 0x0
Oct 26 12:19:40 localhost kernel[0]: AICPUPMI: MSR_PKG_C6_RESIDENCY…….(0x3f9) : 0x0
Oct 26 12:19:40 localhost kernel[0]: AICPUPMI: MSR_PKG_C7_RESIDENCY…….(0x3fa) : 0x0
Oct 26 12:19:40 localhost kernel[0]: AICPUPMI: CPU Low Frequency Mode………….: 1200 MHz
Oct 26 12:19:40 localhost kernel[0]: AICPUPMI: CPU Maximum non-Turbo Frequency….: 2400 MHz
Oct 26 12:19:40 localhost kernel[0]: AICPUPMI: CPU Maximum Turbo Frequency……..: 2400 MHz
Oct 26 12:19:40 localhost kernel[0]: AICPUPMI: IGPU Current Frequency………….: 350 MHz
Oct 26 12:19:40 localhost kernel[0]: AICPUPMI: IGPU Minimum Frequency………….: 350 MHz
Oct 26 12:19:40 localhost kernel[0]: AICPUPMI: IGPU Maximum Non-Turbo Frequency…: 650 MHz
Oct 26 12:19:40 localhost kernel[0]: AICPUPMI: IGPU Maximum Turbo Frequency…….: 1000 MHz
Oct 26 12:19:40 localhost kernel[0]: AICPUPMI: IGPU Maximum limit……………..: No Limit
Oct 26 12:19:40 localhost kernel[0]: AICPUPMI: CPU P-States [ (12) 24 ] iGPU P-States [ (7) ]
Oct 26 12:19:40 localhost kernel[0]: AICPUPMI: CPU C7-Cores [ 2 3 ]
Oct 26 12:19:40 localhost kernel[0]: AICPUPMI: CPU P-States [ (12) 23 24 ] iGPU P-States [ (7) ]
Oct 26 12:19:40 localhost kernel[0]: AICPUPMI: CPU P-States [ (12) 16 23 24 ] iGPU P-States [ (7) ]
Oct 26 12:19:40 localhost kernel[0]: AICPUPMI: CPU P-States [ (12) 16 17 23 24 ] iGPU P-States [ (7) ]
Oct 26 12:19:40 localhost kernel[0]: AICPUPMI: CPU P-States [ (12) 16 17 20 23 24 ] iGPU P-States [ (7) ]
Oct 26 12:19:40 localhost kernel[0]: AICPUPMI: CPU P-States [ (12) 13 16 17 20 23 24 ] iGPU P-States [ (7) ]
Oct 26 12:19:40 localhost kernel[0]: AICPUPMI: CPU P-States [ 12 13 16 17 20 22 23 (24) ] iGPU P-States [ (7) ]
Oct 26 12:19:40 localhost kernel[0]: AICPUPMI: CPU P-States [ (12) 13 15 16 17 20 22 23 24 ] iGPU P-States [ (7) ]
Oct 26 12:19:40 localhost kernel[0]: AICPUPMI: CPU P-States [ (12) 13 15 16 17 19 20 22 23 24 ] iGPU P-States [ (7) ]
Oct 26 12:19:40 localhost kernel[0]: AICPUPMI: CPU P-States [ (12) 13 14 15 16 17 19 20 22 23 24 ] iGPU P-States [ (7) ]
Oct 26 12:19:40 localhost kernel[0]: AICPUPMI: CPU P-States [ (12) 13 14 15 16 17 19 20 21 22 23 24 ] iGPU P-States [ (7) ]
Oct 26 11:19:17 Samirs-MacBook-Pro kernel[0]: AICPUPMI: CPU C3-Cores [ 1 3 ]
Oct 26 11:19:17 Samirs-MacBook-Pro kernel[0]: AICPUPMI: CPU C7-Cores [ 0 1 2 3 ]
Oct 26 11:19:18 Samirs-MacBook-Pro kernel[0]: AICPUPMI: CPU C3-Cores [ 0 1 2 3 ]
Oct 26 11:19:22 Samirs-MacBook-Pro kernel[0]: AICPUPMI: CPU P-States [ (12) 13 14 15 16 17 18 19 20 21 22 23 24 ] iGPU P-States [ (7) ]
Oct 26 11:20:38 Samirs-MacBook-Pro kernel[0]: AICPUPMI: CPU C6-Cores [ 0 1 ]
Oct 26 11:20:43 Samirs-MacBook-Pro kernel[0]: AICPUPMI: CPU C6-Cores [ 0 1 2 ]
Oct 26 11:20:44 Samirs-MacBook-Pro kernel[0]: AICPUPMI: CPU C6-Cores [ 0 1 2 3 ]
Oct 26 11:22:56 Samirs-MacBook-Pro kernel[0]: AICPUPMI: CPU P-States [ 12 13 14 15 16 17 18 19 20 21 22 23 (24) ] iGPU P-States [ 7 (8) ]
Oct 26 11:25:06 Samirs-MacBook-Pro kernel[0]: AICPUPMI: CPU P-States [ 12 13 14 15 16 17 18 19 20 21 (22) 23 24 ] iGPU P-States [ 7 8 (9) ]
But DPCIManager still show that i have only 3 pstats working (12,22,24).
When i enable -xcpm and kernelpm on clover, i get 4 pstats (12,18,21,24).
Can you tell me if its normale that i get only 3 or 4 pstats ? Is that a way to get more pstats ?
Thank you.
Yes. Your Intel i3-3110m lacks the Intel Turbo Boost feature so this is normal.
Thank you piker for the quick answer,
what about the -xcpm and kernelPM ? do you advise me to use it to get the 4th pstats (12,18,21,24) or not use it and stay with the 3 pstats (12,22,24) ?
Whats best in my case ?
If you can XCPM working without errors, then yes. Of course!
Hi Piker Alpha!
The AICPMinfo don’t work in Yosemite.
[img]http://sharebin.net/Lxc.png[/img]
Is possible fix that?
Thanks!
Don’t use discontinued software. Use my new AppleIntelInfo kext (self compiled).
Please note that I will change the name some time soon. I’ll do that to get it approved by Apple.
I can’t find AppleIntelInfo for download!
The source code – you need to compile it yourself – can be found in my Github repository
Hi, Pike!! I have one notebook G75VW with I7 3610qm, Yosemite and Clover. I would to use all my States but my ssdt generated by pike script didn’t work. Can you Help me? I had only 2 states x12 and x23. I attached my ssdt and config.plist
Thanks!! =D