OpenCore General Discussion

[ghost8282]

4 minutes ago, Pavo said:

but the MSR issue can be resolved by this adding this https://github.com/Pavo-IM/Proxintosh/blob/master/etc/modprobe.d/kvm.conf

Thank you Pavo, this could be useful in future! I didn't know about that settings..

[Pavo]

3 minutes ago, ghost8282 said:

Thank you Pavo, this could be useful in future! I didn't know about that settings..

You can look at all the settings I use with latest Qemu and KVM.

[olmirror]

1 hour ago, olmirror said:

Hi viorel, in the ssdt-EC that I have they are already named like that:

 

/*
 * Intel ACPI Component Architecture
 * AML/ASL+ Disassembler version 20180427 (64-bit version)(RM)
 * Copyright (c) 2000 - 2018 Intel Corporation
 *
 * Disassembling to non-symbolic legacy ASL operators
 *
 * Disassembly of iASLlRltqB.aml, Wed May 13 13:46:33 2020
 *
 * Original Table Header:
 *     Signature        "SSDT"
 *     Length           0x0000007D (125)
 *     Revision         0x02
 *     Checksum         0xC0
 *     OEM ID           "ACDT"
 *     OEM Table ID     "SsdtEC"
 *     OEM Revision     0x00001000 (4096)
 *     Compiler ID      "INTL"
 *     Compiler Version 0x20180427 (538444839)
 */
DefinitionBlock ("", "SSDT", 2, "ACDT", "SsdtEC", 0x00001000)
{
    External (_SB_.PCI0.LPCB, DeviceObj)    // (from opcode)

    Scope (\_SB.PCI0.LPCB)
    {
        Device (EC)
        {
            Name (_HID, "ACID0001")  // _HID: Hardware ID
            Method (_STA, 0, NotSerialized)  // _STA: Status
            {
                If (_OSI ("Darwin"))
                {
                    Return (0x0F)
                }
                Else
                {
                    Return (Zero)
                }
            }
        }
    }
}

 

 

I managed to load LPC by renaming it LPC0 but am still getting this in my output. Aaargh, so close... Anybody have any suggestions?

 

[viorel]

19 minutes ago, olmirror said:

 

I managed to load LPC by renaming it LPC0 but am still getting this in my output. Aaargh, so close... Anybody have any suggestions?

 

no renaming needed, delete all this pacth from config.plist

[olmirror]

18 minutes ago, viorel said:

no renaming needed, delete all this pacth from config.plist

Thanks, but I'm a bit confused. The ssdt-EC seems to be working so far, lpc and ec0 are loaded in IORegistryExplorer. Now I'm trying to get the ssdt-plug to run its course and enable p-states by setting plugin-type to one. I've tried several versions, this is the cleanest one I have generated with ssdtTime where I changed PR00 for CP00. Could you please tell me what exactly to get rid of in this ssdt. Thanks for your help!

 

/*
 * Intel ACPI Component Architecture
 * AML/ASL+ Disassembler version 20180427 (64-bit version)(RM)
 * Copyright (c) 2000 - 2018 Intel Corporation
 *
 * Disassembling to non-symbolic legacy ASL operators
 *
 * Disassembly of iASLOF9yF3.aml, Wed May 13 16:35:16 2020
 *
 * Original Table Header:
 *     Signature        "SSDT"
 *     Length           0x000000B1 (177)
 *     Revision         0x02
 *     Checksum         0xCC
 *     OEM ID           "CORP"
 *     OEM Table ID     "CpuPlug"
 *     OEM Revision     0x00003000 (12288)
 *     Compiler ID      "INTL"
 *     Compiler Version 0x20180427 (538444839)
 */
DefinitionBlock ("", "SSDT", 2, "CORP", "CpuPlug", 0x00003000)
{
    External (_SB_.SCK0.CP00, ProcessorObj)    // (from opcode)

    Scope (\_SB.SCK0.CP00)
    {
        Method (DTGP, 5, NotSerialized)
        {
            If (LEqual (Arg0, ToUUID ("a0b5b7c6-1318-441c-b0c9-fe695eaf949b")))
            {
                If (LEqual (Arg1, One))
                {
                    If (LEqual (Arg2, Zero))
                    {
                        Store (Buffer (One)
                            {
                                 0x03                                           
                            }, Arg4)
                        Return (One)
                    }

                    If (LEqual (Arg2, One))
                    {
                        Return (One)
                    }
                }
            }

            Store (Buffer (One)
                {
                     0x00                                           
                }, Arg4)
            Return (Zero)
        }

        Method (_DSM, 4, NotSerialized)  // _DSM: Device-Specific Method
        {
            Store (Package (0x02)
                {
                    "plugin-type",
                    One
                }, Local0)
            DTGP (Arg0, Arg1, Arg2, Arg3, RefOf (Local0))
            Return (Local0)
        }
    }
}

Edited May 13, 2020 by olmirror

[viorel]

57 minutes ago, olmirror said:

tell me what exactly

 

Sorry, I'm at work now and I don’t have access to the mac. Maybe someone will  help you.

Edited May 13, 2020 by viorel

[MacNB]

17 hours ago, shhbz said:

here you go!!!

EFI.zip

 

 

Obviously it would be helpful if you find what you changed to break your system.

 

You have a couple of SSDT's related to USB and sleep.

Why do have SSDT-GPRW.aml and where did you get it from ?

 

Your SSDT-EC-USBX.aml seems modded from the OC sample, why do you need that and not the standard SSDT-EC.dsl ?

 

Also, your description of the problem is not clear when you say the wifi is trying to connect when you put the system to sleep.

How do you know that ?

 

Post your IOREG file.

 

[olmirror]

1 hour ago, viorel said:

 

Sorry, I'm at work now and I don’t have access to the mac. Maybe someone will  help you.

 

I'm no rush. Have a good one! - and thanks.

[shhbz]

1 hour ago, MacNB said:

 

 

Obviously it would be helpful if you find what you changed to break your system.

 

You have a couple of SSDT's related to USB and sleep.

Why do have SSDT-GPRW.aml and where did you get it from ?

 

Your SSDT-EC-USBX.aml seems modded from the OC sample, why do you need that and not the standard SSDT-EC.dsl ?

 

Also, your description of the problem is not clear when you say the wifi is trying to connect when you put the system to sleep.

How do you know that ?

 

Post your IOREG file.

 

Sorted all issues , thanks for the heads up 

 

Edited May 13, 2020 by shhbz

[arsradu]

@vit9696 remember that issues I was having with second stage boot showing only the loading bar and nothing else?

 

When you have a bit of time, you can take a look at the video here.  

 

So, looks like the issue was the iGPU (+ BIOS).

 

Anyway, this is what worked for me. 

In BIOS:

1. switched IGD to PEG (I initially set it to IGD for testing the iGPU itself). Which...after switching the ig-platform-id, showed similar results as before. So...not really an improvement.

2. Set Multi Monitor ON.

3. set video memory to 256MB (probably 64MB or 128MB would have been enough)

 

In config (and I haven't tested it with Clover, but I think it's gonna work, as well):

Set Device Properties, PCIRoot(0x0)Pci(0x2,0x0) - AAPL,ig-platform-id to default for Haswell IGPU: 0300220D with reversed bites (or 0x0D220003 without). Before, I was using a "headless" connectivity with ID 04001204 (or 0x04120004). Which helps with compute, but doesn't do anything for the FV issue.

 

I think I need to try a few other IDs because the default one, while it fixes the issue with second stage boot on FV, it also makes apps like iMovie and FCPX crash when opening.

 

So...either that ig-platform-id is not quite ok for all purposes, oor...there's something else I need to add to device properties to make it work for both.

 

Anyway, I just thought I'd share the result with you.  

 

 

Edited May 13, 2020 by arsradu

[ameenjuz]

@vit9696 

 thanks you fixed Deduplicate windows icon in latest commit changes
worked fine no issue in both macOS and windows 10 are working fine after resetnvram

 

[viorel]

51 minutes ago, olmirror said:

 

I'm no rush. Have a good one! - and thanks.

try

allmiror_OC_OC.zip

[MacNB]

1 hour ago, shhbz said:

Sorted all issues , thanks for the heads up 

 

 

Cool.

What was the solution so that others may learn in case they have similar issues.

After all, that's why we are here ....to share good knowledge 

[olmirror]

2 hours ago, viorel said:

try

allmiror_OC_OC.zip

Thanks!! I had to fiddle around with some settings in order to boot. It's getting late on this side of the pond so I'll let you know how it went tomorrow. Again, thanks for your time, effort and patience.

Edited May 13, 2020 by olmirror

[jsl2000]

On 5/11/2020 at 10:30 PM, hardcorehenry said:

Without rename this SSDT won’t load. Check log with Hackintool>logs>System if it applied correctly.

 

 

Fix_Shutdown.zip

Thanks for your advice.

In order to check this kernel logs how can I get it as shown here ?

[hardcorehenry]

7 hours ago, jsl2000 said:

Thanks for your advice.

In order to check this kernel logs how can I get it as shown here ?

[shhbz]

15 hours ago, MacNB said:

 
Cool.
What was the solution so that others may learn in case they have similar issues.
After all, that's why we are here ....to share good knowledge 

Well for my system Catalina turned out to be disaster even after patches applied lot of bugs were there, went back with to Mojave and all the patches i executed worked perfect.
Need to test further will report 

 

Update: all my ACPI patch was fine nothing fancy there, it just Catalina issue and getting rid of Disabling real EC in ACPI patch as it was not needed.

Edited May 14, 2020 by shhbz

[jsl2000]

4 hours ago, hardcorehenry said:

Thanks for your prompt reply and help which worked for me !

[Edit]

In my Legacy booting OpenCore 0.5.7 & 0.5.8 hackintoshs (Asus P6TSE & P5Q PRO) I have to enabled their patched DSDT.aml plus these _PTS->ZPTS patch & SSDT-ZPTS.aml, otherwise I can not get true power off of CPU, monitor, & HDDs.

Edited May 14, 2020 by jsl2000

[olmirror]

 

15 hours ago, viorel said:

try

allmiror_OC_OC.zip

Allright, I did manage to reduce the acpi errors with your files, even though I had to make some adjustments, renaming lpcb to lpc0 and I disabled ssdt-awac as it produced acpi erors. I also had to make some adjustments to the config file that you provided as my system won't boot with DevirtualizeMmio enabled, I enabled and disabled AppleCpuPmCfgLock and AppleXcpmCfgLock but it doesn't make a difference and I enabled AppleXcpmExtraMsrs since my system won't boot without it.

 

These are the errors left in the log:

 

I found some patches in ACPI that rename _dsm to XDSM or ZDSM since the errors keep referring to _DSM. But they don't seem to make any difference either.

 

I'm also uploading my IORegistry output just in case someone wants to look.

 

All I want is to get rid of PMDrvr.kext from my system, enable package c-states and have my cpu run at its full potential. At this point without PMDrvr I have a 30% performance drop. In clover I was able to manage getting fully enabled power management by enabling PluginType and entering the corresponding xcpm patches.

 

I looked over the CPUfriend instructions but can't seem to get a complete grasp of how and what. Pretty confusing.

 

But then again, maybe I should be happy with how far I got. Thanks again viorel.

 

OC.zip

 

Edited May 14, 2020 by olmirror

[shhbz]

On 4/21/2020 at 4:05 PM, gorans said:

 

- (Lilu) SMCProcessor      scpu: @ failed to get system model

- (AppleACPIPlatform) <AppleACPIPlatform`newObjectFromACPIObject(acpi_object const*)> ACPI: cannot translate ACPI object 14

- (IONVMeFamily) <IONVMeFamily`IONVMeDebugAssert> AppleNVMe Assert failed: ( 0 != data )

- (pmtelemetry) <pmtelemetry`com_apple_driver_pmtelemetry::_pch_init()> Unsupported CPU

- (pmtelemetry) <pmtelemetry`com_apple_driver_pm_pch_reporter::setup(com_apple_driver_pmtelemetry*, IOReportLegend*)> Unsupported PCH

- (AppleMobileFileIntegrity) <AppleMobileFileIntegrity`AMFILoadTrustedKeysFromNVRam> AMFILoadTrustedKeysFromNVRam: no nvram variable

and buttload of (IOAudioFamily) and (Sandbox) messages.

 

any progress here to get rid of these issue. as for me too all end sorted but logs mention these would be more quick to load if some can be sorted

[arsradu]

Update:

 

So I thought the issue with iMovie and FCPX could be the SMBIOS. So I tried different ones (iMac14,2, MacPro6,1)...with no success.

 

Basically I got it narrowed down to two scenarios, both related to the iGPU:

1. with default AAPL,ig-platform-id (0300220D) - FV2 second stage boot loading perfectly on a transparent background, but iMovie and FCPX crashing on startup.

2. with "headless" AAPL,ig-platform-id (04001204) - FV2 second stage boot loading only the loading bar on black background, but iMovie and FCPX working fine.

 

For both scenarios, BIOS was set to PEG and multi-monitor enabled (the amount of shared memory doesn't seem to matter in this case, but you should probably have at least 64MB).

For the second scenario, same result would be accomplished if you disable multi-monitor in BIOS.

 

Now, to me, none of these was ideal. So I kept testing and testing...until I found a scenario in which both FV2 boot screen and iMovie/FCPX work. And I'm just putting it out here, in case someone else wants to try it:

1. in BIOS, set Graphics to IGD  (while still connected to the dedicated GPU).

2. in config, set Device Properties for the iGPU to AAPL,ig-platform-id = 0300220D (default).

 

That's it.  Now both the FV2 boot screen and iMovie work properly.

 

Also, Lilu + WEG are required for best results.

 

So, ideally, this is what you wanna see in Graphics.

Edited May 14, 2020 by arsradu

[MacNB]

9 minutes ago, arsradu said:

Update:

 

So I thought the issue with iMovie and FCPX could be the SMBIOS. So I tried different ones (iMac14,2, MacPro6,1)...with no success.

 

Basically I got it narrowed down to two scenarios, both related to the iGPU:

1. with default AAPL,ig-platform-id (0300220D) - FV2 second stage boot loading perfectly on a transparent background, but iMovie and FCPX crashing on startup.

2. with "headless" AAPL,ig-platform-id (04001204) - FV2 second stage boot loading only the loading bar on black background, but iMovie and FCPX working fine.

 

For both scenarios, BIOS was set to PEG and multi-monitor enabled (the amount of shared memory doesn't seem to matter in this case, but you should probably have at least 64MB).

For the second scenario, same result would be accomplished if you disable multi-monitor in BIOS.

 

Now, to me, none of these was ideal. So I kept testing and testing...until I found a scenario in which both FV2 boot screen and iMovie/FCPX work. And I'm just putting it out here, in case someone else wants to try it: set Graphics to IGD in BIOS (while still connected to the dedicated GPU).

 

That's it.  Now both the FV2 boot screen and iMovie work properly.

 

Also, Lilu + WEG are required for best results.

 

I admire your perseverance 

 

On my ageing i7-3770K, the only way to have FCPX working (and by "working" I mean that it uses the RX580 for Rendering, Decoding and Encoding of both H.264 & HEVC) is to DISABLE the IGPU and use iMacPro1,1 SMBIOS id.

 

With the IGPU enabled, headless ig-platform-id (assigned by WeG), H.264 encode/decode is handled by the IGPU, HEVC is handled by the CPU and   the rendering is done by the RX580 - i.e. not really a useable system.

 

Using iMac13,2 with -shikigva=32 shiki-id=Mac-7BA5B2D9E42DDD94 boot-args (to use the board-id of iMacPro1,1) did NOT enable HEVC nor H.264 encode/decode on the RX580 - it just does not work. So I cannot use the natural iMac13,2 id but have to resort to iMacPro1,1.

On real Mac's I think they have H/W video multiplexor/switcher for IGPU & dGPU and I suspect what happens is that IGPU frame is copied to the dGPU after dGPU initialisation and the video out switched to the dGPU. Resulting in apparent seamless transition during second stage boot.

Without the IGPU, there's clearly another S/W step taking place on real Macs to get the seamless transition. 

 

I have not yet enable FV2 so cannot confirm what you see during boot.

With my IGPU disabled, I always get the black screen between the two stages when I don't even have FV2.

Just thinking out loud.

 

[viorel]

3 hours ago, olmirror said:

maybe I should be happy with how far I got. Thanks again viorel.

 Sorry, but I do not have this motherboard, I can only say a few thoughts - if AWAC does not work
 then try SSDT-RTC0 from open core samples.
look for help from those who have similar
 hardware https://www.insanelymac.com/forum/topic/334342-imac-pro-x99-live-the-future-now-with-macos-1014-mojave-successful-buildsuccessful-guide/

[arsradu]

45 minutes ago, MacNB said:

 

I admire your perseverance 

 

On my ageing i7-3770K, the only way to have FCPX working (and by "working" I mean that it uses the RX580 for Rendering, Decoding and Encoding of both H.264 & HEVC) is to DISABLE the IGPU and use iMacPro1,1 SMBIOS id.

 

With the IGPU enabled, headless ig-platform-id (assigned by WeG), H.264 encode/decode is handled by the IGPU, HEVC is handled by the CPU and   the rendering is done by the RX580 - i.e. not really a useable system.

 

Using iMac13,2 with -shikigva=32 shiki-id=Mac-7BA5B2D9E42DDD94 boot-args (to use the board-id of iMacPro1,1) did NOT enable HEVC nor H.264 encode/decode on the RX580 - it just does not work. So I cannot use the natural iMac13,2 id but have to resort to iMacPro1,1.

On real Mac's I think they have H/W video multiplexor/switcher for IGPU & dGPU and I suspect what happens is that IGPU frame is copied to the dGPU after dGPU initialisation and the video out switched to the dGPU. Resulting in apparent seamless transition during second stage boot.

Without the IGPU, there's clearly another S/W step taking place on real Macs to get the seamless transition. 

 

I have not yet enable FV2 so cannot confirm what you see during boot.

With my IGPU disabled, I always get the black screen between the two stages when I don't even have FV2.

Just thinking out loud.

 

 

Hehe! :)) Thank you.

 

I also got both H.264 and HEVC (H.265) working with this setup. It was working before, as well. So it's not due to the changes I've made here. I think that's because WEG integrated it a while ago. So...if you haven't updated your WEG kext, now would be the time to do so.

 

We've got the same dGPU (well, similar, not identical, but the same chip), so...in theory, it should work for you, as well. With both default and "headless" device properties. I'm also using SMBIOS iMacPro1,1 so...that's one more similar thing between our builds. But, if I'm not mistaken, I think everything was working smoothly and everything enabled with other SMBIOSes, as well. So....I'm not sure that's still relevant. At least with newer WEG. I know there was a time when this was indeed the only SMBIOS with which you could get HEVC. But I don't think it's the case anymore.

 

 

As for the gap you mentioned in between first and second stage, I do get that, too, regardless of anything. I'm not sure there's a fix for that. If there is, I'd love to get that fixed.  And, if I'm correct, I think some of the older real Macs had it, too.

 

The thing I was trying to fix though was the second stage boot, and only when FV2 was enabled. With normal boots everything was fine. Apple logo + loading bar on a black background. But with FV2 enabled, I was only getting the loading bar. Now, if I'm not mistaken, a "real Mac" boot with FV would have both stages on a transparent background, with no Apple logo, but instead, with your user logo up top. Which...in my case, it's exactly what I see. :))

 

So, aside from the small gap between those stages, I think I pretty much got a prefect boot.  I don't really use FV2 either. As I mentioned in one of my previous messages, I just wanted to see if it works. Which it does. Then I got into the second stage issue...and now that one's fixed, as well.  

Edited May 14, 2020 by arsradu

[shhbz]

21 minutes ago, arsradu said:

 

Hehe! :)) Thank you.

 

I also got both H.264 and HEVC (H.265) working with this setup. It was working before, as well. So it's not due to the changes I've made here. I think that's because WEG integrated it a while ago. So...if you haven't updated your WEG kext, now would be the time to do so.

 

We've got the same dGPU (well, similar, not identical, but the same chip), so...in theory, it should work for you, as well. With both default and "headless" device properties. I'm also using SMBIOS iMacPro1,1 so...that's one more similar thing between our builds. But, if I'm not mistaken, I think everything was working smoothly and everything enabled with other SMBIOSes, as well. So....I'm not sure that's still relevant. At least with newer WEG. I know there was a time when this was indeed the only SMBIOS with which you could get HEVC. But I don't think it's the case anymore.

 

 

As for the gap you mentioned in between first and second stage, I do get that, too, regardless of anything. I'm not sure there's a fix for that. And, if I'm correct, I think some of the older real Macs had it, too.

 

The thing I was trying to fix though was the second stage boot, and only when FV2 was enabled. With normal boots everything was fine. Apple logo + loading bar on a black background. But with FV2 enabled, I was only getting the loading bar. Now, if I'm not mistaken, a "real Mac" boot with FV would have both stages on a transparent background, with no Apple logo, but instead, with your user logo up top. Which...in my case, it's exactly what I see. :))

 

So, aside from the small gap between those stages, I think I pretty much got a prefect boot.  I don't really use FV2 either. As I mentioned in one of my previous messages, I just wanted to see if it works. Which it did. Then I got into the second stage issue...and now that one's fixed, as well.  

this seems interesting but the 2 questions which arise in my mind is

1.does your sleep/wake work with graphics set to IGD

2.have you tried encoding, decoding a file and rendering a small timeline adding some effects does the DGPU works better that way.