Jump to content

rongu

Donators
  • Content Count

    146
  • Joined

  • Last visited

  • Days Won

    1

Posts posted by rongu


  1. 1 hour ago, tonyx86 said:

    @rongu I don't work on the ThinkPad T61 very often (only updated to Mojave to see if it worked and it works well for a 2008 laptop).  If I remember to get the DSDT the next time I'm on that system, I'll post here.  In all honesty, the DSDT does not add any value to this thread.  Everything that I required to implement onejay09's brightness fix for the Thinkpad T61 is in my previous post.  I hope that helps.  I think the biggest revelation for me was that the GFX0._DSM doesn't require much if you use CLOVER's "Inject NVidia" option.  It significantly simplifies this brightness fix on the Thinkpad T61 and maybe other laptops as well (likely with different values for pwm-info and NVCAP).

    .

    Appreciate your reply, Thank You.


  2. 1 hour ago, hoangmytr said:
    Hi,

    I know those PMs drives have issues with MacOS/hackintosh as the booting drive. But anyone using them as storage devices only?

    I want to use them to store my data only, not booting the OS.

    Thanks! :thumbsup_anim:
     

    I am stuck with some PM981 too, bought a NVMe M.2 SSD Enclosure (10Gbps) from http://www.orico.cc/us/product/detail/951.html

     

    I had tried booting os x 10.14.5 with it and of course data storage works, doing about 500-600 MB/s with BlackMagic Speed test via USB3.1 Type-C.

     

    Never achieve speed what they claim.

     

     


  3. 5 hours ago, osy said:

    New release: https://github.com/osy86/HaC-Mini/releases/tag/v1.3

    • Enable iGPU. Fixes Preview freeze bug when opening JPEG files. If you previously disabled iGPU, please re-enable it in the BIOS!
    • Reworked the USB port injection. Should be cleaner.
    • Fixed a bug that caused AppleLPC to not be loaded.

    If you have a custom installation, make sure to merge the changes in the config.plist and replace the DSDT files.

    @osyYou should have at least reply or acknowledge LPC not loading bug than I dun have to waste my time hunting. I ask you at least 3 time or more.

    6 hours ago, osy said:

    @rongu Now I remember why I didn’t use iMac18,2/3. The USB-C DP doesn’t work with that config. Black screen on boot.

    That was when ?


  4. @kowbot Thanks, that means AppleLPC.kext was not loaded in your system (HVK) also. I have a HNK version and wonder why it not loading AppleLPC.kext despite device-id change in clover->devices->properties changes with "PciRoot(0)/Pci(31,0) device-id C19C0000" (according to osy it should load).

     

    At the end of the day, AppleLPC.kext is needed for proper CPU Power Management (https://pikeralpha.wordpress.com/2013/10/05/xnu-cpu-power-management/).

     

    To get AppleLPC loaded:

    - Disable device change in clover/devices/propeties by adding a "#" to "PciRoot(0)/Pci(31,0)",  see attach picture.

    - Install the attached AppleLPC_Injector.kext in clover/kexts/others.

    - save/reboot and verify with command "kextstat | grep AppleLPC" in your terminal again.

    - double check with IOregistryviewer

     

    HNK & HVK should have the same LPC controller (HM175 Chipset) pic8086,a152. Attached picture & terminal LPC output are mine.

     

    Can you or anybody test and reply, Thanks !

     

    Managed to test on a HVK system and confirmed AppleLPC.kext was not loaded without LPC injection.

    Last picture is a HVK (iMac18,2 profile) average idle frequency with AppleLPC.kext loaded. More test and time to tell.

     

    1829200062_Screenshot2019-07-25at7_06_26PM.png.d151405fc96f277367fe70db2e2a2393.png

     

     

     

    Screenshot 2019-07-25 at 7.05.59 PM.png

    AppleLPC_Injector.zip

     

    Screenshot 2019-07-25 at 11.17.54 PM.png

    Screenshot 2019-07-25 at 10.07.33 PM.png


  5. @kowbot Glad you found a way to manage the freezing issues and Thanks for sharing. You might want to include the BIOS version (0057 or 0056). I'm trying to borrow one HVK to test.

     

    @kowbot "I disable VT-d in bios how working Parallels os OS X? "

    At this point of time, the biggest bottleneck with hypervisors is Direct I/O and with VT-d enable, it should help with I/O intensive virtualization. To me VT-d is good to have not must have.

     

    If you are a PRO providing a virtual service, hosting guests, running something where you need the performance ........ then VT-d is very important.

    Normal day to day, casual run additional OS ...then not that important, I can't tell if they were any difference.

     

    By the way @kowbot can you do me a favour and post your terminal output with the command "kextstat |grep AppleLPC", Thanks !


  6. 4 hours ago, kowbot said:

    hmm... i give up my nuc is still hard freeze randomly when < 10mins...

    @kowbotHey, have you tried resetting your BIOS and select Low Power Enable or Balance Enable under PrimaryPower Settings and see how it goes....

     

    Other things your can try to isolate your issues by

    -Removing your windows' m.2 storage

    -Running your system with single stick of ram

    -Disable wifi/bt in BIOS

     


  7. 1 hour ago, KKKW said:

    Thank you very much, @rongu! That works to me.

     

    I DID use the same version of files in your package but I installed them in /L/E/ (or /S/L/E, I tried both, none works).

     

    Putting them into EFI/Clover/kext/Other works to me.

     

    Also, I tried your steps to enable iGPU, it works! 

     

    Thank you again!

     

    @KKKW, glad it work out for you.

     

    To match our PEGP to injected AGPM profile add the following patch below or after "change GFX0 to IGPU". See Pic1

    - Add a patch "change PEGP to GFX0" in Clover->DSDT->Patches, find hex "50454750", replace hex "47465830"

     

    Save and reboot, verify with IOregistryviewer see Pic2

    That should be all for AGPM..... i hope :)

     

    Pic1.jpg

    Pic2.png


  8. 5 minutes ago, kowbot said:

    thx. 

    i use brcmFirmwareRepo.kext and brcmPatchRam2.kext and u posted config.plist bluetooth not detected anymore.

     

    but with HaC mini config.plist work bluetooth whats difference between?

    You should only used BrcmFirmwareRepo.kext & BrcmPatchRAM2 in /Library/Extensions OR BrcmFirmwareData.kext & BrcmPatchRAM2 in clover/kexts/others.

     

    HaC mini config.plist work bluetooth ???


  9. On 7/23/2019 at 6:50 PM, kowbot said:

    thx for sharing.

     

    u find enable VT-d and brcmpatch both work?

    No, I cannot get them to work together (VT-d conflict with BrcmFirmwareData.kext). Work around, disable Bluetooth and enable VT-d in the BIOS.

    On 7/23/2019 at 9:39 PM, Dagvelo said:

    do the sleep/wake is working for you ?

     

    I have reboot from wake issues on 10.14.6 but none on 10.14.5, both with NTFY patch disabled.

    @Dagvelo I have no trouble with sleep/wake 10.14.6 with NTFY patch disabled.


  10. 9 hours ago, osy said:

    Any reason why you changed the model to iMac? The thermal profiles are different and I think the Mac mini profile is closer to the NUC. 

    I am not sure about the thermal profiles but I think is also important to match the hardware profile.

    If you choose a macmini81 the system expect IGPU to be around and always on, I checked dsdt & etc and wonder what happen if call were being made in routine related IGPU which in this case is always off (disable).

     

    In term of hardware profile, I think iMac18,3 is closer as it has a GFX0, and a IGPU connected to nothing, like our NUC (from some Intel NUC spec). That why I guess my preview.app's problem goes away as the log tells me the problem is due to some HW Jpeg decoding. With iMac18,3 we have IGPU doing the decoding. Also by selecting iMac18,3 we enable quicksync, this one I had check.

     

    I think it is easier to deal with thermal profiles of the CPU by tweaking (ssdtgen, HWPValue or whatever), please correct me if I  am wrong.

     

    At the end of the day, we will probably find out what best as we go along.

     

    PS, I missed one very important step regarding enabling IGPU, we need a patch "change GFX0 to IGPU" for everything else to work properly especially the AGPM.


  11. For those interested in enabling Intel Quicksync (IGPU) of NUC8i7HNK/HVK.

     

    Changelog

       2019-0725-00

          - Add iMac18,2 as an option

          - Add iMac18,2 GPUs profile in dAGPM.kext

          - Tested on NUC8i7HVK

     

    Requirements:

    - HaC_mini 1.2 (osy, NUC8 Hero !)

    - Add "-disablegfxfirmware" into Clover->boot arguments

    - Add a patch "change GFX0 to IGPU" in Clover->DSDT->Patches, find hex "47465830", replace hex "49475055"

    - Add a patch "change PEGP to GFX0" in Clover->DSDT->Patches, find hex "50454750", replace hex "47465830"

    - Check Intel Inject box [/] and set 0x59120003 for ig-platform-id under Clover->Graphics

    - Select iMac18,2 or iMac18,3 under Clover->SMBIOS, save and reboot to BIOS menu ( I prefer iMac18,2)

    - Enable IGPU under Graphics, save and exit, boot to Desktop

    - Optional, copy dAGPM.kext into Clover/kexts/others (GFX0 & IGPU AGPM profile are from iMac18,2 & iMac18,3 respectively)

     

    To verify Quicksync feature, download and run Intel® Power Gadget. Other than normal CPU (blue bar), you should also see additional green bar (Intel iGPU status) on the right.

     

    Go to Gigabyte website and do some browsing or play iTune trailer movies, you should see the green frequency varies according to different load. CPU frequency will plunge as iGPU load increase in some case.

     

    Another test is to download Trial version of MacX Video Converter Pro, make sure INTEL/AMD/NVIDIA box [l] is checked when launch and click the blue info thing, you should see "Hardware Encoding Supported YES" message.

     

    For those using Final Cut will benefit for sure.

     

    Have fun :)

    1098918298_Screenshot2019-07-22at10_57_16PM.thumb.png.3992617b9bf71604d82d52d2424d287b.png773750512_Screenshot2019-07-22at10_57_37PM.thumb.png.59adb8e18bdf6900ecca907c861bbd26.png

    Screenshot 2019-07-22 at 11.04.20 PM.png

    Screenshot 2019-07-22 at 11.01.34 PM.png

    Screenshot 2019-07-22 at 10.58.29 PM.png

     

     

     

    dAGPM.zip


  12. 12 hours ago, KKKW said:

    Thank you, @osy

     

    I had a NUC (NUC8i7HVK) and it was running the Hackintosh by everything on tonymacx86.com forum. But after I found your guide then I erased my disk to try your steps, because the glitch issue is the one I have suffered enough.

     

    There is a critical issue I am having now - I used a BCM94352Z wifi/bt module to replace the one in the NUC before, and it was working well after using the AirportBrcmFixup.kext, BrcmFirmwareRepo.kext and BrcmPatchRAM2.kext mentioned in https://www.tonymacx86.com/threads/broadcom-wifi-bluetooth-guide.242423/. But it is not working now. 

     

    The system cannot recognize the wifi/bt module, instead, it shows a BCM20702A0 USB device. Do you know how do I fix it?

     

    I saw @rongu mentioned he is using BCM94352Z as well? @rongu could you share how do you make it work? 

     

    Thanks in advanced!

    Both Wifi/BT works fine

    On my set, I have to disable VT-d in the bios under Secuirty tab. dart=0 in Clover boot argument actually doing the same thing but does not always work.

     

    After you complete the above, just install the following to Clover/kexts/Other, see attached

    AirportBrcmFixup.kext

    BrcmFirmwareData.kext

    BrcmPatchRAM2.kext

     

    BCM94352Z.zip

     

    Note, the attached files are not my work, they were created by a group of wonderful people and they are here for convenient sake.

    File source, https://bitbucket.org/RehabMan/

     


  13. 16 minutes ago, kwjenkins said:

    thanks for the feedback all, I'll look in these directions and see if I can get to that install.

    I rebuilt that boot drive a couple times last night and it was always giving the same result.

    Just remember to disable IGPU in the bios and follow osy's instruction you should be good.

     

    Also disable VT-d if you install BrcmFirmwareData.kext (BCM94352Z wifi/BT replacement), mine refuse to boot successfully. dart=0 don't always work !

     


  14. On 7/15/2019 at 12:20 AM, osy said:

    The line I highlighted, which is a device-id override for PciRoot(0)/Pci(31,0). This should let AppleLPC match on the device without any other patches. 

     

    I’m using an Apple WiFi card so I don’t need BrcmFirmwareData.kext. 

     

    @osy

    For some reason, AppleLPC.kext was not loaded with the patch you directed me to "The setting is here: https://github.com/osy86/HaC-Mini/blob/master/Clover/config.plist#L167"

     

    Would you be able to show me the output of your nuc8 with the command in terminal "kextstat |grep AppleLPC" ?

     

    This my nuc8 output with LPC injection and I get nothing with device-id override for PciRoot(0)/Pci(31,0).

     

    Screenshot 2019-07-19 at 4.23.54 PM.png

     

    Thanks !


  15. 8 minutes ago, tonyx86 said:

    As I previously reported, after upgrading my Thinkpad T61 to Mojave 10.14.5 (using DosDude's Mojave patcher), @onejay09's brightness fix (patching DSDT with PNLF and GFX0._DSM and running onejay09's script) still works (Display slider still controls brightness).  I recently noticed that when running Mojave, IOHWControl was not loading under AGPM as viewed in IORegistryExplorer (see attached screenshot showing IOHWControl in IORegistryExplorer).  IOHWControl was loading in High Sierra 10.13.6, but not in Mojave 10.14.5.  I'm not sure what I was missing in my DSDT patches, but the "fix" for this is to use CLOVER's "Inject NVidia" option (I wasn't using the CLOVER option before, since I injected all NVidia attributes via DSDT GFX0._DSM).  Now, IOHWControl loads in both High Sierra and Mojave.

     

    With the use of CLOVER's "Inject NVidia" option, my DSDT patches now include only the PNLF patch and the following elements in my GFX0._DSM:

    • define "@0,pwm-info": this is required for proper brightness control.  Brightness control "worked" without this in my DSDT, but brightness adjustment wasn't "smooth" across the entire adjustment range.  With this defined, brightness adjustment is smooth across the entire adjustment range.
    • define "NVCAP": this is required for the internal and external display to work properly.  Without "NVCAP" defined, the internal LCD display works when an external monitor is NOT plugged into the Thinkpad VGA port; however, if "NVCAP" is not defined, neither the internal LCD nor the external display work when an external display is connected to the VGA port.
    With these adjustments (enabling CLOVER's "Inject NVidia" option and including only "@0,pwm-info" and "NVCAP" in my DSDT's GFX0._DSM), my Thinkpad T61's display brightness slider works perfectly in High Sierra 10.13.6 and Mojave 10.14.5 (with DosDude's Mojave patcher).
     
    My new GFX0._DSM includes only following:
     
    
                If (LEqual (Arg2, Zero)){Return (Buffer (){0x03})}
                Return (Package ()
                {
                    "@0,pwm-info", 
                    Buffer ()
                    {
                        /* 0000 */  0x01, 0x14, 0x00, 0x64, 0xA8, 0x61, 0x00, 0x00,
                        /* 0008 */  0x1C, 0x02, 0x00, 0x00, 0x64, 0x00, 0x00, 0x00,
                        /* 0010 */  0x00, 0x04, 0x00, 0x00
                    }, 
                    "NVCAP", 
                    Buffer ()
                    {
                        /* 0000 */  0x04, 0x00, 0x00, 0x00, 0x00, 0x00, 0x01, 0x00,
                        /* 0008 */  0x1E, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x0A,
                        /* 0010 */  0x00, 0x00, 0x00, 0x00
                    }
                }
    

    Screen Shot 2019-07-14 at 9.32.01 AM.png

    Can you share your dsdt with the patch mentioned, Thanks !

×