Clover General discussion

[kushwavez]

@Slice Mine is 8086,a103. If I don't use CtlnaAHCIPort.kext the device is not recognisable and the system cannot boot. So what do you suggest instead of that kext?

[Matgen84]

9 minutes ago, Slice said:

As well for my 10 series chipset. But I am sure it is cosmetic. The same kext works the same manner. It is AboutMac.app shows wrong.

Vendor name = Generic

 

why?

 

 

I'm agree with you. It is just cosmetic. Thanks for your explanations 

It's a little bit strange : why vendor name is generic !

[Slice]

30 minutes ago, kushwavez said:

@Slice Mine is 8086,a103. If I don't use CtlnaAHCIPort.kext the device is not recognisable and the system cannot boot. So what do you suggest instead of that kext?

FakeID = 0xA1028086

or Properties

PciRoot(0x0)/Pci(0x17,0x0)

device-id  = <02 a1 00 00>

 

But I really don't understand why I have vendor=Generic. It seems Apple's AHCI controller has some mark it is Apple. And the kext CtlnaAHCIPort  ignores this mark.

[kushwavez]

Unfortunately FakeID doesn't work. "Waiting on boot-uuid-media" then after a while "Still waiting for root device"

 

EDIT:

1 hour ago, iCanaro said:

 

OK I added your IDs to the injector, if you tell me I'd be grateful archivio.zip

Unfortunately same with yours, even though I see my dev-id in the Info.plist.

 

Well, I'm good with CtlnaAHCIPort, but yes just a FakeID or smth like that would be nice option, less kexts are always better.  

 

 

Edited November 29, 2020 by kushwavez

[Alpha22]

for series 9, I use this kext no problem

 

 

 

[Matgen84]

17 minutes ago, kushwavez said:

Unfortunately FakeID doesn't work. "Waiting on boot-uuid-media" then after a while "Still waiting for root device"

 

EDIT:

Unfortunately same with yours, even though I see my dev-id in the Info.plist.

 

 

Well, I'm good with CtlnaAHCIPort, but yes just a FakeID or smth like that would be nice option, less kexts are always better.  

 

 

 

Do you try Item 3 instead of Item 0 for @iCanaro Kexts/Info.plist 

Edited November 29, 2020 by Matgen84

[kushwavez]

I'm pretty sure that the order doesn't matter, but sure I can try.

EDIT: yes, doesn't matter. Not working unfortunately.

Edited November 29, 2020 by kushwavez

[Jief_Machak]

Hi everyone.

I'd like help to identify a potential Clover bug.

If you have a bootable OC configuration (no need that everything is configured, just bootable), check that you can boot with quirks DevirtualiseMmio enabled, and nothing in MmioWhitelist. It's supposed to be ok from IvyBridge.

Then check the same quirks in Clover config.plist. For me, when this is enabled in Clover, there is a very early crash. And you ?

 

[kushwavez]

@Jief_Machak 

Notebook 1:

Clover v5127: DevirtualiseMmio enabled : halting at "HID: Legacy Shim 2" (very early, as you mentioned)

2020-11-29_20-14-32_CLOVERX64.efi.log

-> Disabling DevirtualiseMmio: booting OK

 

OpenCore 0.6.3: DevirtualiseMmio enabled: halting at exactly the same spot "HID: Legacy Shim 2"

opencore-2020-11-29-201458.txt

-> Disabling DevirtualiseMmio: booting OK

Edited November 29, 2020 by kushwavez

[Jief_Machak]

In your case, it seems to do the same. So that's ok. In my case, it doesn't...

Because Clover now use OC "kernel engine", a quirk is supposed to act the same.

Anyone else can test ?

@kushwavez : thanks for the test.

[PG7]

26 minutes ago, Jief_Machak said:

Hi everyone.

I'd like help to identify a potential Clover bug.

If you have a bootable OC configuration (no need that everything is configured, just bootable), check that you can boot with quirks DevirtualiseMmio enabled, and nothing in MmioWhitelist. It's supposed to be ok from IvyBridge.

Then check the same quirks in Clover config.plist. For me, when this is enabled in Clover, there is a very early crash. And you ?

 

I'm here with Ivybridge and I have both bootloaders ready.I'm here with Ivybridge and I have both bootloaders ready.

[PG7]

39 minutes ago, Jief_Machak said:

Hi everyone.

I'd like help to identify a potential Clover bug.

If you have a bootable OC configuration (no need that everything is configured, just bootable), check that you can boot with quirks DevirtualiseMmio enabled, and nothing in MmioWhitelist. It's supposed to be ok from IvyBridge.

Then check the same quirks in Clover config.plist. For me, when this is enabled in Clover, there is a very early crash. And you ?

 

 

 

Dell Latitude E6330
Ivybridge Intel i7-3520M
intel HD4000

tested here with OpenCore 0.6.3 and DevirtualiseMmio active = Crach

Clover r5127 with DevirtualiseMmio active = Crash

 

[iCanaro]

46 minutes ago, Jief_Machak said:

Then check the same quirks in Clover config.plist. For me, when this is enabled in Clover, there is a very early crash. And you ?

 

however, when you have Devirtualisemmio skills then depending on the platform, it is necessary to see if with all the MMIO active you work so the hack. On AMD platform it is better that you get them all and then add them in the list and check if 1 or more give problems. On AMD X570 (in mine I have 13 MMIO voices) and TRX40, this is very important.

For example, on my X570 if I have DevirtualiseMmio, then the hack no longer shuts down unless you turn off the power outlet 

 

PS: then another important thing, the MMIO voices could also change from one version of the BIOS to another, I always take my X570 example, at first I had 4 MMIO voices, then with new BIOS 13 and now that I have already flashed BIOS for ZEN3 I did not verify.... how many do I have? more, less, the same?

Edited November 29, 2020 by iCanaro

[Jief_Machak]

My point is : is that quirks does the same when enabled in Clover and when enabled in OC. Same can mean crash, it’s ok.

Of course with the same configuration in OC and Clover.

[PG7]

11 minutes ago, Jief_Machak said:

My point is : is that quirks does the same when enabled in Clover and when enabled in OC. Same can mean crash, it’s ok.

Of course with the same configuration in OC and Clover.

 

in my case yes to a great similarity

Edited November 29, 2020 by PG7

[iCanaro]

28 minutes ago, Jief_Machak said:

My point is : is that quirks does the same when enabled in Clover and when enabled in OC. Same can mean crash, it’s ok.

Of course with the same configuration in OC and Clover.

 

OK I understood the purpose, regardless of the result, to see if the behavior between OC and Clover is identical or different.
On Z370 hack with DevirtualiseMmio enabled or disabled the behavior between OC and Clover is identical; no problem, the hack apparently works the same way in all situations

[iCanaro]

I wanted to see what was happening on the X570 with new BIOS and DevirtualiseMmio enabled and with Clover nothing changes, apparently identical operation, no anomaly detected.
On OC 062 instead a nice mess, you have blackscreen, which never happened, then if you add boot-arg agdpmod=pikera you get to the desktop, but you don't have any USB working.. in short, enabling DevirtualiseMmio creates nice problems even worse than in the past.

[Jief_Machak]

7 hours ago, iCanaro said:

I wanted to see what was happening on the X570 with new BIOS and DevirtualiseMmio enabled and with Clover nothing changes, apparently identical operation, no anomaly detected.
On OC 062 instead a nice mess, you have blackscreen, which never happened, then if you add boot-arg agdpmod=pikera you get to the desktop, but you don't have any USB working.. in short, enabling DevirtualiseMmio creates nice problems even worse than in the past.

You are saying there is a behaviour difference ? Works with Clover and not with OC ?

[Matgen84]

10 hours ago, Jief_Machak said:

Hi everyone.

I'd like help to identify a potential Clover bug.

If you have a bootable OC configuration (no need that everything is configured, just bootable), check that you can boot with quirks DevirtualiseMmio enabled, and nothing in MmioWhitelist. It's supposed to be ok from IvyBridge.

Then check the same quirks in Clover config.plist. For me, when this is enabled in Clover, there is a very early crash. And you ?

 

Hi @Jief_Machak I just read several post about DevirtualiseMmio and MmioWhitelist with IvyBridge.

I follow convert guide to build Clover config.plist: the settings are the same for Clover and functional OC for my Asus Z77 (i7 3770K)

 

	<key>MmioWhitelist</key>
        <array/>
        <key>Quirks</key>
        <dict>
            <key>AvoidRuntimeDefrag</key>
            <true/>
            <key>DevirtualiseMmio</key>
            <false/>

There is no crash if disable. Only crash when  DevirtualiseMmio is enable like @PG7

[Slice]

10 hours ago, Jief_Machak said:

Hi everyone.

I'd like help to identify a potential Clover bug.

If you have a bootable OC configuration (no need that everything is configured, just bootable), check that you can boot with quirks DevirtualiseMmio enabled, and nothing in MmioWhitelist. It's supposed to be ok from IvyBridge.

Then check the same quirks in Clover config.plist. For me, when this is enabled in Clover, there is a very early crash. And you ?

 

The quirk is sometimes needed for Z390. Should be Off for IvyBridge.

[Jief_Machak]

@Matgen84 @Slice : The point is NOT to know if it should be enabled or not. The point is : is the behavior the same in Clover and OC ?

For me it's not. So I'd say there is a bug in Clover.

@Matgen84 @iCanaro Do you have the "same" behavior if you enable DevirtualiseMmio ?

[iCanaro]

on the X570 I don't have the same behavior

 

on the Z370 I already knew that if I enabled or did not enable DevirtualiseMmio it made no difference, so it is not possible to detect differences in behavior between Clover and OC

[iCanaro]

35 minutes ago, Jief_Machak said:

Do you have the "same" behavior if you enable DevirtualiseMmio ?

 

specifying that the tests are performed with OC062 and Clover 5127, in the Z68 I have the same behavior, enabling DevirtualiseMmio Kernel Panic to boot with both OC and Clover

[Matgen84]

48 minutes ago, Jief_Machak said:

Do you have the "same" behavior if you enable DevirtualiseMmio ?

 

The tests are performed with OC 0.6.4 Beta and Clover r5127, in my Z77 Ivybridge. Like @iCanaro I have the same behavior, DevirtualiseMmio enable --->> Kernel panic with both bootloader.

[naiclub]

สปอยเลอร์

This version, booting with clover5127, does not work, I just tried to access it and it appears that it cannot boot any more.

I switched back to 5126 recently compiled last time.