Jump to content
30960 posts in this topic

Recommended Posts

10 hours ago, Pene said:

Hi Slice,

Thanks.

Meanwhile. he tested with changed Rank (Attributes=2, before your latest changes) but unfortunately there was no change, install still failed. So changing this in smbios didn't really help

He said he will send me the memory chips so that I can make more investigations. But it will take a while till he sends them.

What is your memory rank in dmidecode -t 17?

Also, your error is the exact one that my friend is reporing? 

Hi Pene,

I see real iMac18,3 also has Rank=1

Handle 0x0001, DMI type 17, 40 bytes
0000: 11 28 01 00 00 00 fe ff 40 00 40 00 00 10 0d 00 
0010: 01 02 1a 80 00 60 09 03 04 05 06 01 00 00 00 00 
0020: 60 09 00 00 00 00 b0 04 

Memory Device
	Array Handle: 0x0000
	Error Information Handle: Not Provided
	Total Width: 64 bits
	Data Width: 64 bits
	Size: 4096 MB
	Form Factor: SODIMM
	Set: None
	Locator: DIMM0
	Bank Locator: BANK 0
	Type: <OUT OF SPEC>
	Type Detail: Synchronous
	Speed: 2400 MHz (0.4 ns)
	Manufacturer: 0x802C
	Serial Number: 0x21211463
	Asset Tag:  
	Part Number: 0x344154463531323634485A2D3247334232202020
	Rank: 1
	Configured Clock Speed: 2400 MHz (0.4 ns)
	Minimum voltage:  Unknown
	Maximum voltage:  Unknown
	Configured voltage:  1.200 V

What is the size of your DIMM? Before my latest commit there was a problem with 32Gb modules. Ticket 595.

13 hours ago, Slice said:

But SMBIOS v3 has other entry point and other GUID so I can't recommend to use it until all standards will be followed.

Up to iMac15,1 Apple used SmbiosVersion 2.4.

Since iMac17,1 I see SmbiosVersion 3.0.

  • Like 1
15 hours ago, Slice said:

Hi Pene,

I see real iMac18,3 also has Rank=1

What is the size of your DIMM? Before my latest commit there was a problem with 32Gb modules. Ticket 595.

The dimm where install fails is a rank1 8gb x2 chips each (Corsair, total 16gb)

Install was ok with rank 2 8gb x2 chips (Gskill, total 16gb)

And also was ok on with rank 2 16gb x2 chips  (Corsair, total 32gb)

That's why I thought it is probably related to rank, as there was no other major difference between the chips where it worked and failed. Same size also as you see.

He was using iMac19,2 model.

 

Edited by Pene
7 hours ago, Taza4776 said:

hi... how do i fix my ssd..? it show as external....

 

Try this patch in KextsToPatch section (config.plist file of course) :)

<dict>
				<key>Comment</key>
				<string>External icons patch</string>
				<key>Disabled</key>
				<false/>
				<key>Find</key>
				<data>
				RXh0ZXJuYWw=
				</data>
				<key>InfoPlistPatch</key>
				<false/>
				<key>MatchOS</key>
				<string>10.13.x,10.14.x,10.15.x</string>
				<key>Name</key>
				<string>com.apple.driver.AppleAHCIPort</string>
				<key>Replace</key>
				<data>
				SW50ZXJuYWw=
				</data>
			</dict>

 

  • Like 1
8 hours ago, Matgen84 said:

 

Try this patch in KextsToPatch section (config.plist file of course) :)


<dict>
				<key>Comment</key>
				<string>External icons patch</string>
				<key>Disabled</key>
				<false/>
				<key>Find</key>
				<data>
				RXh0ZXJuYWw=
				</data>
				<key>InfoPlistPatch</key>
				<false/>
				<key>MatchOS</key>
				<string>10.13.x,10.14.x,10.15.x</string>
				<key>Name</key>
				<string>com.apple.driver.AppleAHCIPort</string>
				<key>Replace</key>
				<data>
				SW50ZXJuYWw=
				</data>
			</dict>

 

 

it still the same result ssd on external..., i patch it on config.plist KextsToPatch section.  

Edited by Taza4776
  • Sad 1
7 hours ago, Taza4776 said:

 

it still the same result ssd on external..., i patch it on config.plist KextsToPatch section.  

 

Do you a NVMe SSD? Pike R. Alpha had created a kext-to-patch that can be added to your config.plist to deal with this issue:

 

<dict>
<key>Comment</key>
<string>IONVMeFamily Pike R. Alpha Patch#External</string>
<key>Disabled</key>
<false/>
<key>Name</key>
<string>IONVMeFamily</string>
<key>Find</key>
<data>SIXAdAeAiwgBAAAQ</data>
<key>Replace</key>
<data>kJCQkJCAiwgBAAAQ</data>
</dict>

 

  • Like 2
4 hours ago, Matgen84 said:

 

Do you a NVMe SSD? Pike R. Alpha had created a kext-to-patch that can be added to your config.plist to deal with this issue:

 


<dict>
<key>Comment</key>
<string>IONVMeFamily Pike R. Alpha Patch#External</string>
<key>Disabled</key>
<false/>
<key>Name</key>
<string>IONVMeFamily</string>
<key>Find</key>
<data>SIXAdAeAiwgBAAAQ</data>
<key>Replace</key>
<data>kJCQkJCAiwgBAAAQ</data>
</dict>

 

my sdd = ADATA XPG SX8200 PRO 1TB M.2 2280 PCIE GEN3X4 SSD. would it work?

2 hours ago, Taza4776 said:

my sdd = ADATA XPG SX8200 PRO 1TB M.2 2280 PCIE GEN3X4 SSD. would it work?

 

I don't have A SSD NVMe, so I can test this patch. Try it and you will have got answer. Rember it's just for orange icons. You can do that from booting with a usb pendrive

4 hours ago, Taza4776 said:

my sdd = ADATA XPG SX8200 PRO 1TB M.2 2280 PCIE GEN3X4 SSD. would it work?

That patch has no effect on my Samsung 970 Pro.

This, however, solves the problem: 1.2.0 (RELEASE).zip

If you prefer, you can also solve it with a SSDT patch. I chose the kext solution i attached, but you can google for it if you wish.

 

[EDIT] Forgot to mention that the kext is a Lilu plugin. You need Lilu in order to use it.

 

Biciolino

Edited by biciolino
  • Like 1
1 hour ago, telepati said:

right-click and remove Prefpan is ok? or should uninstall differently?

It is ok.

 

1 hour ago, telepati said:

the latest clover will install new clover.app or we must install manually from .zip?

Just download the app and run it. More info here.

  • Thanks 1
On 11/20/2019 at 1:35 PM, Taza4776 said:

hi... how do i fix my ssd..? it show as external....


It's easy, just add a device with address of your NVME. You can see your NVME in IOREG is pcixxx,xx, you must declare it with DSDT or config.plist :)

  • Like 2
On 8/21/2019 at 6:54 PM, Pene said:

It is for more complex cases, mostly for cases where you have two GPUs in the system (both are enabled from BIOS), but monitor is connected only to one, and you wish to be able to control the settings for OSX according to which GPU the monitor is connected when system boots up.

It is not needed for simpler cases where only one GPU is being used.

 

It can help in some other cases too, but here is a use case, where it is really needed:

- You have an Intel GPU and also a discrete GPU, but the discrete GPU is not supported by OSX (for example, all latest Nvidia cards).

- You want to use Nvidia GPU when you boot Windows, but usually Intel GPU when you boot OSX. And maybe you want to boot OSX sometimes also with the NVidia card, when monitor is connected to it.

- You have a DP or HDMI switch connected to the monitor, and can switch the monitor between the Intel and the Nvidia card. Or alternatively, you have two inputs on your monitor, and connect each monitor input to a different GPU, and select it from the monitor before powering up the system.

(Note: such configuration is usually supported with the following UEFI BIOS settings: CSM=Disabled, Primary Graphics Adapter=Onboard, IGPU-Multi-Monitor=Enabled, which results, on most UEFI BIOSes, in POST being displayed to where the monitor is connected).

 

In the above case, it would be useful for Clover to detect which card is active when you boot (meaning, to which card the monitor is connected), so that you can disable the inactive GPU when OSX boots, in order to avoid problems regarding Multi-GPU setup within OSX (and use OSX, instead, as having a single GPU, regardless of which GPU you are booting with).

- To do this, in the above case, you can now set properties for secondary (inactive) gpu in config.plist (at Devices/Properties/SecondaryGPU). An example property for to inject for the secondary device could be a device-id=0xffff, which will result in disabling the inactive GPU (and it will always disable the card that is not used when booting up, whether it is Intel or Discrete).

- Note, that it this case, it can be also useful to set in config.plist GUI/ShowOptimus=true (this will result in showing at the bottom of Clover's GUI whether monitor is currently connected to Intel or to Discrete GPU).

- Properties can be set also for PrimaryGPU, if needed (at Devices/Properties/PrimaryGPU).

With the correct settings, you can achieve OSX booting properly also with Discrete GPU when monitor is connected to it (without acceleration, of course, for the example above of an unsupported card), and also with Intel GPU when monitor is connected to that one. 

 


Hi, I can’t work out from your post here a) if this feature can help me get to a better place, or b) just actually how to use it if that is the case. 
 

my situation: two GPUs, one GTX 1080 and one Radeon WX Pro 7100. 
 

I have disabled the GTX 1080 using Devices / AddProperties / NVidia + name/IOName/class-code

 

i have both cards attached to separate inputs on my display and so dual boot. So the desire is to have as painless use of the Radeon output for MacOS. 
 

Currently, the machine boots to MacOS, and stops updating the screen once the “IOConsoleUser .. ScreenLockState...” line is displayed. Notably this is via the DisplayPort from the GTX 1080. If I at this point change inputs on the display, I have my nice login screen via the Radeon. 
 

So - I could live with this I guess. But does this PrimaryGPU/SecondaryGPU offer me anything better? If so, how?

9 hours ago, Drinahn said:


Hi, I can’t work out from your post here a) if this feature can help me get to a better place, or b) just actually how to use it if that is the case. 
 

my situation: two GPUs, one GTX 1080 and one Radeon WX Pro 7100. 
 

I have disabled the GTX 1080 using Devices / AddProperties / NVidia + name/IOName/class-code

 

i have both cards attached to separate inputs on my display and so dual boot. So the desire is to have as painless use of the Radeon output for MacOS. 
 

Currently, the machine boots to MacOS, and stops updating the screen once the “IOConsoleUser .. ScreenLockState...” line is displayed. Notably this is via the DisplayPort from the GTX 1080. If I at this point change inputs on the display, I have my nice login screen via the Radeon. 
 

So - I could live with this I guess. But does this PrimaryGPU/SecondaryGPU offer me anything better? If so, how?

The feature is meant to be used for injecting properties depending on which card is the monitor connected to when system boots up. It checks which card has the GOP active (that will be the card where you see the POST output), and allows you to inject properties judging by that. For example, if you want to always disable the card which was not active when booting, you can inject to SecondaryGPU.

As in your case you always want to disable a specific card, it is better to inject directly to it, as you are already doing.

×
×
  • Create New...