macOS Native CPU/IGPU Power Management

[RehabMan]

The system boots and looks like works properly with or without XCPM (plugin YES method), but the Intel Power Gadget is detecting the CPU Clock maximum peak value with AVX (4.2GHZ - AVX 3) and also looks like with XCPM there is a state parameter missing on C3. I tested with Plugin enabled (without XCPM) and with XCPM. I noticed some differences, which one would be recommended? Also, after doing the configuration I changed the BIOS settings to Manual and Sync All Cores to 4.2 GHZ, but the maximum still 3.9GHZ.



Thanks!

"Problem Reporting" files are incomplete.
Read FAQ, "Problem Reporting" again. Carefully. Attach all requested files/output.
https://www.tonymacx86.com/threads/faq-read-first-laptop-frequent-questions.164990/
Use the tool mentioned in the FAQ, that way it is less likely you'll omit something.

Note: Both of your ioreg files show "plugin-type" injected. So... no idea what you mean by pluginNo.ioreg vs. pluginYes.ioreg.

 

[fadis96]

hi

1. Description of problem:Installed the PM using the SSDT-XCPM.aml way, I have the same CPU as the example presented in the first thread/Post (I7-4770k haswell), however my P states aren't as many your example. could you please check that I have done this correctly
2. OS X version:10.13.3 /motherboard model17D47)/BIOS version:F9/processor: i7 4770k/graphics:Nvidia GTX 780
3. Procedure/Guide Used: SSDT-XCPM.aml + PluginType=Yes

other files required are attached and labeled.
If there are any abnormality in my reports, please let me know, I'm very keen to do this right.
thank you for your help!

 

[RehabMan]

hi

1. Description of problem:Installed the PM using the SSDT-XCPM.aml way, I have the same CPU as the example presented in the first thread/Post (I7-4770k haswell), however my P states aren't as many your example. could you please check that I have done this correctly
2. OS X version:10.13.3 /motherboard model17D47)/BIOS version:F9/processor: i7 4770k/graphics:Nvidia GTX 780
3. Procedure/Guide Used: SSDT-XCPM.aml + PluginType=Yes

other files required are attached and labeled.
If there are any abnormality in my reports, please let me know, I'm very keen to do this right.
thank you for your help!

No need to use config.plist/ACPI/SSDT/Generate/PluginType=true *and* SSDT-XCPM.aml.
The two solutions accomplish the same thing and are mutually exclusive.
Use one or the other, not both.

 

[fadis96]

No need to use config.plist/ACPI/SSDT/Generate/PluginType=true *and* SSDT-XCPM.aml.
The two solutions accomplish the same thing and are mutually exclusive.
Use one or the other, not both.

right. is there a preference to which I should use?
forgive me but does that explain why my p-states are less than the example provided in the guide?

 

[RehabMan]

right. is there a preference to which I should use?

I already wrote: "The two solutions accomplish the same thing".

forgive me but does that explain why my p-states are less than the example provided in the guide?

Pstates observed will vary depending on how long you monitor and what you do with the computer while monitoring.

 

[fadis96]

I already wrote: "The two solutions accomplish the same thing".



Pstates observed will vary depending on how long you monitor and what you do with the computer while monitoring.

I do apologies, yes I let it run longer and achieved the following (using the SSDT-XCPM method):

The plugin=Yes didn't work for me.

Thank you for your patience.

One last question, I removed the Dsdt and replaced it with SSDT-XCPM.
Is there anything on the Dsdt that I may need or is it redundant

 

[RehabMan]

I do apologies, yes I let it run longer and achieved the following (using the SSDT-XCPM method):
View attachment 311936
The plugin=Yes didn't work for me.

plugin=Yes is wrong.
It is spelled 'PluginType'.
Must be set at config.plist/ACPI/SSDT/Generate/PluginType
Must be boolean, and set to true (shown in Xcode as Yes).
Hence,
config.plist/ACPI/SSDT/Generate/PluginType=true

One last question, I removed the Dsdt and replaced it with SSDT-XCPM.

DSDT has nothing to do with SSDT-XCPM.aml
Whether you need patches to your DSDT is a separate question and is off-topic here.

 

[funkyfrank]

I have a question regarding this: You wrote in the first post "Drop OEM = no", but should I still drop the default tables Cpu0Ist and CpuPm on a Haswell system? Thanks.

EDIT: I left those table dropping, and everything is looking good, in IntelGAdget and also AppleIntelInfo.kext and "sysctl machdep.xcpm"... Now pikeralpha wrote in his original blog post that AppleSMBusPCI.kext is then not required anymore. But it still loads here. Is this an indicator that my config is still not 100% and can I block that kext with clover then?

 

[RehabMan]

I have a question regarding this: You wrote in the first post "Drop OEM = no", but should I still drop the default tables Cpu0Ist and CpuPm on a Haswell system? Thanks.

No need to drop those CPU related SSDTs. In fact, dropping them will likely cause problems, not solve any.

EDIT: I left those table dropping, and everything is looking good, in IntelGAdget and also AppleIntelInfo.kext and "sysctl machdep.xcpm"... Now pikeralpha wrote in his original blog post that AppleSMBusPCI.kext is then not required anymore. But it still loads here. Is this an indicator that my config is still not 100% and can I block that kext with clover then?

No "Problem Reporting" files attached.
Read FAQ, "Problem Reporting" again. Carefully. Attach all requested files/output.
https://www.tonymacx86.com/threads/faq-read-first-laptop-frequent-questions.164990/
Use the gen_debug.sh tool mentioned in the FAQ, that way it is less likely you'll omit something.

 

[funkyfrank]

No need to drop those CPU related SSDTs. In fact, dropping them will likely cause problems, not solve any.

Thanks, another question: Will this also work on my Vaio F11 having a very early Core i7 q720? Seems to be "Clarksfield" generation?