Jump to content

KernelAndKextPatches 10.13x,10.14.x,10.15.x X99/X299


nmano
244 posts in this topic

Recommended Posts

28 minutes ago, jlcdgd said:

They work together perfectly with xcpm and really smooth in my case...

Thanks vector

@jlcdgd, @vector sigma:

There is a real increase in performance, however with PmDrvr, the readings of HWMonitor seem strange...<_< especially in the instant use of the CPU load. I don't know why, and I don't know if the use of this kext is actually an advantage, apart from the performances ... Here I trust vector sigma that certainly knows more than me! At the moment I preferred to eliminate it, also due to an increase in core temperature ...

Schermata 2020-05-01 alle 18.28.23.png

Link to comment
Share on other sites

6 minutes ago, vector sigma said:

yes, Clover enables xcpm (for unsupported CPUs) only if you set KernelAndKextPatches->KernelXCPM = true. For supported CPUs it just will work.

If your is an unsupported one, KernelAndKextPatches->KernelXCPM = false (or just remove the key) will work as well, I guess. Otherwise you will have two programs that will interefere each other. 

 

P.S. you can disable it also in Clover GUI at boot (binaries patching menu). 

Ok...

With KernelXCPM off (Not ticked) I get better scores in Mojave But some "periodic peaks" in performance that create some playback errors in protools ("related to buffer sizes" which has no sense).

Also higher temperatures.

 

You can see peaks here:

Peaks.thumb.png.c842896e8c3746875b83538cc3318682.png

 

With KernelXCPM off (Not ticked) I get slightly better scores in Catalina but some "periodic peaks" in performance too. Protools uses more CPU resources but is more stable than Mojave...

Also higher temperatures.

 

Geekbench  in Catalina:

922645446_KernelXCPMOff.thumb.png.b3a4dde7573b52978576801ec8d78c5b.png

 

So in my conclusion the combination is better in both systems. No other patches applied...

 

 

Link to comment
Share on other sites

12 minutes ago, lucke said:

@jlcdgd, @vector sigma:

There is a real increase in performance, however with PmDrvr, the readings of HWMonitor seem strange...<_< especially in the instant use of the CPU load. I don't know why, and I don't know if the use of this kext is actually an advantage, apart from the performances ... Here I trust vector sigma that certainly knows more than me! At the moment I preferred to eliminate it, also due to an increase in core temperature ...

Schermata 2020-05-01 alle 18.28.23.png

 

Of Course vector sigma knows a lot off more than me... I was a "coder" a hundred years ago so...

In my case 5960X Asus X99 EWS temperature is lower and my system is way more stable with protools (in fact much better than a MacPro with 8 cores) if I use a combination of two.

Just trying to help if I can do something.

Brumbauer (the gentelman who did the kext) is a master too.

 

Best wishes and thanks for reporting...

 

 

 

 

 

 

  • Like 1
Link to comment
Share on other sites

Tomorrow I can share my EFI if you need it. Higher geekbench means higher temperatures too.

Anyway you can check with Intel that you won’t temperatures can go way higher and you won’t have any problem...

 

My CPU has at least 7 years, my motherboard like 5...

Link to comment
Share on other sites

42 minutes ago, jlcdgd said:

Tomorrow I can share my EFI if you need it. Higher geekbench means higher temperatures too.

Anyway you can check with Intel that you won’t temperatures can go way higher and you won’t have any problem...

 

My CPU has at least 7 years, my motherboard like 5...

Ok Thanks!

I use TSCAdjust, in idle my temperatures are crazy, they go from 35 to 70 degrees celsius with pmdrvr and the problem of using the cpu at 0% remains ... If it were not for these problems, the use of this kext can actually increase the performance, as it was for High Sierra & Mojave... at least I think ...!

Edited by lucke
Link to comment
Share on other sites

17 hours ago, vector sigma said:

 

 

18 hours ago, lucke said:

 

Please find enclosed my EFI folder...

Is the best combination I can have. And everything works really fine.

I have tried TSCAdjustReset.kext, but my system does not work according to the developers is created for X299 systems (not my case):

https://github.com/interferenc/TSCAdjustReset

 

If you think my EFI can be improved in any way I will try...

in my BIOS IEST is Disabled.

Best deserves

 

EFI.zip

Link to comment
Share on other sites

2 hours ago, jlcdgd said:

 

Please find enclosed my EFI folder...

Is the best combination I can have. And everything works really fine.

I have tried TSCAdjustReset.kext, but my system does not work according to the developers is created for X299 systems (not my case):

https://github.com/interferenc/TSCAdjustReset

 

If you think my EFI can be improved in any way I will try...

in my BIOS IEST is Disabled.

Best deserves

 

EFI.zip

I had a look at your EFI. I would say that if the machine works well, you should leave it.

Anyway, I share my EFI, where I spent some time to make it performant, including the compilation of all SSDT.

If you want to take a look...

lucke EFI.zip

Link to comment
Share on other sites

1 hour ago, lucke said:

I had a look at your EFI. I would say that if the machine works well, you should leave it.

Anyway, I share my EFI, where I spent some time to make it performant, including the compilation of all SSDT.

If you want to take a look...

lucke EFI.zip

Thank you.

I will take a look. My experience says that less is more. Maybe not true in some cases, but I try to avoid as much SSDTs as possible, as you can see. And everything works in a very clean way... 

Who knows this is the hackintosh world, but even real macs have some problems

1 hour ago, lucke said:

I had a look at your EFI. I would say that if the machine works well, you should leave it.

Anyway, I share my EFI, where I spent some time to make it performant, including the compilation of all SSDT.

If you want to take a look...

lucke EFI.zip

Thank you.

I will take a look. My experience says that less is more. Maybe not true in some cases, but I try to avoid as much SSDTs as possible, as you can see. And everything works in a very clean way... 

Who knows this is the hackintosh world, but even real macs have some problems :)

 

1 hour ago, lucke said:

I had a look at your EFI. I would say that if the machine works well, you should leave it.

Anyway, I share my EFI, where I spent some time to make it performant, including the compilation of all SSDT.

If you want to take a look...

lucke EFI.zip

Thank you.

I will take a look. My experience says that less is more. Maybe not true in some cases, but I try to avoid as much SSDTs as possible, as you can see. And everything works in a very clean way... 

Who knows this is the hackintosh world, but even real macs have some problems :)

 

1 hour ago, lucke said:

I had a look at your EFI. I would say that if the machine works well, you should leave it.

Anyway, I share my EFI, where I spent some time to make it performant, including the compilation of all SSDT.

If you want to take a look...

lucke EFI.zip

Thank you.

I will take a look. My experience says that less is more. Maybe not true in some cases, but I try to avoid as much SSDTs as possible, as you can see. And everything works in a very clean way... 

Who knows this is the hackintosh world, but even real macs have some problems :)

 

1 hour ago, lucke said:

I had a look at your EFI. I would say that if the machine works well, you should leave it.

Anyway, I share my EFI, where I spent some time to make it performant, including the compilation of all SSDT.

If you want to take a look...

lucke EFI.zip

Thank you.

I will take a look. My experience says that less is more. Maybe not true in some cases, but I try to avoid as much SSDTs as possible, as you can see. And everything works in a very clean way... 

Who knows this is the hackintosh world, but even real macs have some problems :)

 

  • Like 1
Link to comment
Share on other sites

4 hours ago, jlcdgd said:

 

Please find enclosed my EFI folder...

Is the best combination I can have. And everything works really fine.

I have tried TSCAdjustReset.kext, but my system does not work according to the developers is created for X299 systems (not my case):

https://github.com/interferenc/TSCAdjustReset

 

If you think my EFI can be improved in any way I will try...

in my BIOS IEST is Disabled.

Best deserves

 

EFI.zip

I would add  CPUFriend.kext

Link to comment
Share on other sites

19 minutes ago, pitrysha said:

I don’t know how to explain correctly, it controls the CPU multipliers

https://github.com/acidanthera/CPUFriend/releases

Look at this, it could help:

https://www.olarila.com/topic/5693-guide-ssdt-with-pikes-pm-script-and-use-with-cpufriend/

and:

I think it is not necessary for haswell or broadwell...

Edited by lucke
Link to comment
Share on other sites

8 minutes ago, lucke said:

Look at this, it could help:

https://www.olarila.com/topic/5693-guide-ssdt-with-pikes-pm-script-and-use-with-cpufriend/

and:

I think it is not necessary for haswell or broadwell...

This confirms that it controls the processor, whether it is needed or not, each decides in his own way, I just expressed my opinion

Link to comment
Share on other sites

Just now, pitrysha said:

This confirms that it controls the processor, whether it is needed or not, each decides in his own way, I just expressed my opinion

Of course, surely it was just my opinion. I was just saying that I don't think it's necessary because in the topic on olarila, I wrote a post months ago, to understand if I could use cpufriend, MaLd0n replied that it was useless for me to use it for my configuration, which as I noticed you also have a x99... That's it... :yes:

Link to comment
Share on other sites

On 5/1/2020 at 3:12 PM, vector sigma said:

@lucke I was referring to the above statement, which clearly is not the case :)

I was referring to the fact the when running hackintosh mac os is not requesting the cpu to go higher when it could go higher so it is not using the CPU at maximum capacity under load but on an real mac pro it works as it should that can be seen through the fact that the core req is at maximum frequency under load, on hackintosh it doesn't push req to the maximum. That's why we need the performance patch for the moment to always req maximum from the CPU. It is not optimal but we don't have other option for the moment.

Link to comment
Share on other sites

On 4/28/2020 at 1:12 PM, vector sigma said:

 

 

On 5/1/2020 at 6:05 PM, lucke said:

Vector Sigma thanks for your efforts, My system is working better than ever before...

I can make the system work properly with the very last version of OpenCore...

Clover and Kernel work perfect in a x99 EWS but not in a x99 EWS 3.1... interesting. Both have the very last version of BIOS same BIOS set up, even I have tried with the same EFI... In the x99 EWS 3.1 works better without the kext but I get 40% less of geekbench (same CPU)

TSCAdjustReset.kext vs VoodooTSCSync.kext... Both work. But I thing for X99 VoodooTSCSync.kext is the right one

CPUfriend.kext does not make any change in any system except for having like 2 to 4 degrees Celsius more ¿? Have tried a lot of times.

Thanks to all of you

@nmano

 

I wil love to see patches for opencore with the last beta version of Catalina

 

:rolleyes:

 

Best

  • Like 1
Link to comment
Share on other sites

I have waited over a month to update to 10.15.4. I failed a month ago and am continuing to fail now. I have been web searching this last month, but this problem is persistent. I know this isn't a kernal and kext patch, but I've depended on your guys X99 patches for a couple of years.

The error I get is :    Stuck at [EB]#LOG:EXITBS:START]. 

I'm on Clover 5116, but I've tried 5112 etc and no go.

I tried a fresh 10.15.4 usb install too. I appreciate any possible help. Sincerely johnm

IMG_0009.jpg

Edited by jmacie
Link to comment
Share on other sites

20 hours ago, jmacie said:

I have waited over a month to update to 10.15.4. I failed a month ago and am continuing to fail now. I have been web searching this last month, but this problem is persistent. I know this isn't a kernal and kext patch, but I've depended on your guys X99 patches for a couple of years.

The error I get is :    Stuck at [EB]#LOG:EXITBS:START]. 

I'm on Clover 5116, but I've tried 5112 etc and no go.

I tried a fresh 10.15.4 usb install too. I appreciate any possible help. Sincerely johnm

IMG_0009.jpg

 

You have the kernel debugging turned on. Check nvram arguments (or ones in the config.plist?) and clean anything related to debugging like "debug=0x44 kcsuffix=development” if any, in short:

sudo nvram -d boot-args

and readd any needed bootArgs (but I guess you already have them in the config.plist?) then try with:

sudo rm -f /System/Library/PrelinkedKernels/prelinkedkernel.de*

then 

sudo kextcache -invalidate /

then 

sudo touch /System/Library/Extensions && sudo kextcache -u /

reboot and see if  something changed. 

Edited by vector sigma
  • Like 2
Link to comment
Share on other sites

Bless vector sigma, i appreciate everything, will get on this asap, johnm

@vector sigma , I tried these commands and rebooted not working, but I'll try again later. I do appreciate you helping, johnm

Edited by jmacie
Link to comment
Share on other sites

Hi guys, thanks to all the info and help in this thread. I am running a X99 system with 10.14.6. Please could you confirm that I am using the correct performance patch for 10.14.6:

 

Comment    String  <-> XCPM Performance fix by @PMheart 
Disabled   Boolean <-> No
Find       Data<-> C1E30848 63D389D0 48C1EA20 
MatchOS    String <-> 10.14.x,10.15.x
Replace    Data<->C1E308B8 00FF0000 31D29090 

 

 

On 4/27/2020 at 7:29 AM, latze said:

 

It is at 2550 because the performance patch ( 41C1E608 B800FF00 0031D290 90 ) sets it like that. See the FF in the code, that sets the frequency. To set it to your max frequency ( 4300 ) change it to 2B so the replace code will be 41C1E608 B8002B00 0031D290 90

 

 

@ latze I am having the same issue Intel Power Gadget stuck at 2550 with the performance patch for 10.14.6. How could I modify the patch so that it reports my Max Freq?

 

Thank you

 

1956488004_Screenshot2020-05-12at10_27_24.thumb.png.efc78dc254fdfa692446bfdc6e10ea36.png

Link to comment
Share on other sites

On 2/9/2020 at 12:08 PM, yapan4 said:

Thank you @nmano

No success this time. I do try many times - no boot.

Note: for successful boot two kernel patches must be enabled in my case and i am not sure in second one (please see attachment)

Screenshot 2020-02-09 at 18.56.36.png

Still boot successful with old bootstrap patch.

Hi

I am wondering how you can add the kernel patches to opencore config.plist. Can you help me, please? Thanks

Link to comment
Share on other sites

In the quest for achieving good CPU performance with XCPM while maintaining the power saving to a good level I played a little with the frequency vectors in the past days. I managed to achieve similar results with the one I was getting when I was using the performance patch but still hitting low frequency. So right now I am currently running with the new clover that @vector sigma provided with just the kernelXCPM option enabled and the CPUFriend kext with the generated CPUFriendDataProvider kext attached for whom wants to try it. 

 

Tested with iMacPro1,1 SMBIOS on i7 5820K . 

CPUFriend.kext.zip

CPUFriendDataProvider.kext.zip

Link to comment
Share on other sites

4 hours ago, jrv10 said:

Hi guys, thanks to all the info and help in this thread. I am running a X99 system with 10.14.6. Please could you confirm that I am using the correct performance patch for 10.14.6:

 


Comment    String  <-> XCPM Performance fix by @PMheart 
Disabled   Boolean <-> No
Find       Data<-> C1E30848 63D389D0 48C1EA20 
MatchOS    String <-> 10.14.x,10.15.x
Replace    Data<->C1E308B8 00FF0000 31D29090 

 

 

 

@ latze I am having the same issue Intel Power Gadget stuck at 2550 with the performance patch for 10.14.6. How could I modify the patch so that it reports my Max Freq?

 

Thank you

 

1956488004_Screenshot2020-05-12at10_27_24.thumb.png.efc78dc254fdfa692446bfdc6e10ea36.png

 

 

If you are seeing 25.50 at Core Req you are using the right one :D it wouldn't reach that level otherwise.

 

To see your actual Max Freq there which from the pic seem to be 3.8GHz you need to change the replace string to C1E308B8 00260000 31D29090 ( 26 instead of FF => 26 in HEX is 38 )

 

But I recommend you to try my other method that I just posted instead of performance patch. 

Link to comment
Share on other sites

×
×
  • Create New...