Jump to content

[pre-release] macOS Catalina


1,679 posts in this topic

Recommended Posts

8 hours ago, PMheart said:

--- orig-kernel_patcher.c	2019-06-04 05:26:39.000000000 +0800
+++ patched.c	2019-06-04 05:24:43.000000000 +0800
@@ -483,6 +483,18 @@
 STATIC UINT8 HSieMojSearchModel[]   = {0x89, 0xc1, 0xc0, 0xe9, 0x04};
 // Need to use YosECSieSearchExt, LionReplaceModel
 
+// Catalina
+/*
+ This patch searches
+  mov eax, r12   ||   mov eax, r12
+  shr al, 0x4    ||   shr eax, 0x10
+ and replaces to
+  mov eax, FakeModel || mov eax, FakeExt
+  nop                || nop
+*/
+STATIC UINT8 CataSearchModel[]      = {0x44, 0x89, 0xE0, 0xC0, 0xE8, 0x04};
+STATIC UINT8 CataSearchExt[]        = {0x44, 0x89, 0xE0, 0xC1, 0xE8, 0x10};
+STATIC UINT8 CataReplaceMovEax[]    = {0xB8, 0x00, 0x00, 0x00, 0x00, 0x90}; // mov eax, val || nop
 
 BOOLEAN PatchCPUID(UINT8* bytes, UINT8* Location, INT32 LenLoc,
                    UINT8* Search4, UINT8* Search10, UINT8* ReplaceModel,
@@ -552,14 +564,22 @@
     return;
   }
 // High Sierra/Mojave patterns
-// Sherlocks: 10.13/10.14.DP1
+// Sherlocks: 10.13/10.14
   DBG_RT(Entry, "CPUID: try High Sierra/Mojave patch...\n");
-  if (PatchCPUID(kernelData, &StrMsr8b[0], sizeof(StrMsr8b), &HSieMojSearchModel[0],
+  if (PatchCPUID(kernelData, &StrMsr8b[0], sizeof(StrMsr8b), &CataSearchModel[0],
                  &YosECSieSearchExt[0], &LionReplaceModel[0], &LionReplaceModel[0],
                  sizeof(HSieMojSearchModel), Entry)) {
     DBG_RT(Entry, "...done!\n");
     return;
   }
+// Catalina patterns
+// PMheart: 10.15.DP1
+  DBG_RT(Entry, "CPUID: try Catalina patch...\n");
+  if (PatchCPUID(kernelData, &StrMsr8b[0], sizeof(StrMsr8b), &HSieMojSearchModel[0],
+                 &CataSearchExt[0], &CataReplaceMovEax[0], &CataReplaceMovEax[0],
+                 sizeof(CataSearchModel), Entry)) {
+    return;
+  }
 }
 
 // new way by RehabMan 2017-08-13

 

 

Hi @PMHeart

This CPUID patch, when used, seems to hang the kernel.

Edited by Pene
Link to comment
Share on other sites

4 minutes ago, Pene said:

 

Hi @PMHeart

This CPUID patch, when used, seems to hang the kernel.

Hi, indeed, I made a very stupid typo... Sorry!

 

Please see: 

 

 

Or... Do you mean that it is even broken with my corrections?!

Link to comment
Share on other sites

7 hours ago, PMheart said:

As for Kernel Lapic patch, the code looks very different, I haven't come up with an idea how to patch it, sorry, it has been too late here, I feel like taking a nap now...

 

Sorry,

PMheart

News on Kernel Lapic:

 

Looks like the method we used to patch Lapic panic is kind of complex nowadays, and the data in 10.15 kernel really looks brand-new. For now I only want to post the patch (NOT guaranteed to work), and wait for the next Beta to check further.

 

Please try this kernel patch for 10.15 Beta 1:

9E 00 00 74 0E 8B -> 9E 00 00 EB 22 8B

  • Like 1
Link to comment
Share on other sites

I have Catalina up and running but cannot get video kexts loaded.  I'm hoping it's just a matter of waiting for Lilu and WEG to be updated for 10.15.  Also, my Broadcom Wifi card caused a KP when using AirportBrcmFixup.kext.  When I reverted to FakePCIID.kext and FakePCIID_Broadcom_WiFi.kext, wifi worked fine.  If anyone has any tips for loading intel IGPU kexts, let me know.

Link to comment
Share on other sites

33 minutes ago, PMheart said:

Or... Do you mean that it is even broken with my corrections?!

No, didn't notice that post.

Committed as r4940, now need to test and see.

Edited by Pene
Link to comment
Share on other sites

6 minutes ago, Pene said:

No, didn't notice that post.

 Committed as r4940, now need to test and see.

Thanks! Could you please try it? (i.e pass a value that would not trigger a panic, if you should be using your Z390 as shown in your signature, you could try 0x0306C3 (or 0x3C as how Clover/XNU parses it), which is Haswell)

 

And then run something like

 

sysctl machdep.cpu.model

 

See if the result is 60. (0x3C in hex)

Link to comment
Share on other sites

Here's a weird thing. Not that I've got it going (ACPI stall, then get to the apple logo then blank screen) but what it has done is split my volume up into the OS and data. But the data has some other stuff in it - screenshot:

 

498474286_ScreenShot2019-06-04at3_25_33PM.png.428133567fcfd13031168a588cf29b14.png

1709449383_ScreenShot2019-06-04at3_26_34PM.png.abec143e71ab88bf1726f0c2474a819b.png

No sign of the apps that were installed previous. So just to be clear, I had a few volumes - OSX, OSXbac and OSXHS now I have four - OSX and OSX data, OSXbak and OSXHS.

I'll try a clean install now :)

This is on my Zenbook

Link to comment
Share on other sites

52 minutes ago, PMheart said:

Thanks! Could you please try it? (i.e pass a value that would not trigger a panic, if you should be using your Z390 as shown in your signature, you could try 0x0306C3 (or 0x3C as how Clover/XNU parses it), which is Haswell)

 

And then run something like

 

sysctl machdep.cpu.model

 

See if the result is 60. (0x3C in hex)

Sure.

Yes, it seems to be working fine,

  • Like 1
Link to comment
Share on other sites

2 minutes ago, Pene said:

Sure.

Yes, it seems to be working fine,

Great! Thanks a lot!

21 minutes ago, Cyberdevs said:

Hi,

 

Just want to get everyone's forgiveness. :P

FakeCPUID is totally broken under Clover r4939, and r4940 is fine as confirmed by @Pene

If possible, please remove it and re-upload r4940!

 

Thanks,

PMheart

  • Like 4
Link to comment
Share on other sites

28 minutes ago, PMheart said:

Great! Thanks a lot!

Hi,

 

Just want to get everyone's forgiveness. :P

FakeCPUID is totally broken under Clover r4939, and r4940 is fine as confirmed by @Pene

If possible, please remove it and re-upload r4940!

 

Thanks,

PMheart

you are a hero! So what's the dvmt and Usb port limit patch in 10.15 db1 please :blush:

Link to comment
Share on other sites

2 hours ago, Ricky1994 said:

you are a hero! So what's the dvmt and Usb port limit patch in 10.15 db1 please :blush:

As for DVMT - I doubt it works with a raw patch, please wait for the update of WhateverGreen.

 

If you still need it, OK, please tell me which exact FB kext you'd like to patch? (Or rather all?! :D)

 

Regarding USB patches, I am not sure either, just made these patches based off what we have done under 10.14.5:

 

EDIT: Please see 

 

And do NOT use the ones below!

 

IOUSBHostFamily #1
Find: 83FB0F0F87
Repl: 83FB3F0F87

IOUSBHostFamily #2
Find: 83E30FD3E34109DF
Repl: 83E33FD3E34109DF

AppleUSBXHCI #1
Find: 0F0F83
Repl: 3F0F83

 

Edited by PMheart
  • Like 6
Link to comment
Share on other sites

27 minutes ago, daliansky said:

try it:


#1:
com.apple.driver.usb.AppleUSBXHCI
find: 83E40F0F
replace: 83E43F0F

#2:
com.apple.driver.usb.AppleUSBXHCI
find: 83F90F0F
replace: 83F93F0F

 

Xnip2019-06-04_16-40-42.png

Hi,

 

I doubt #1 is really fine, are you even patching that bitmask?!

Correct my mistake if any, though.

Edited by PMheart
  • Like 1
Link to comment
Share on other sites

19 minutes ago, PMheart said:

Hi,

 

I doubt #1 is really fine, are you even patching that bitmask?!

Correct my mistake if any, though.

hi, @PMheart

I have confirmed, remove the #1 patch, just add the #2 patch. It works just like your patch.

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

4 minutes ago, daliansky said:

hi, @PMheart

I have confirmed, remove the #1 patch, just add the #2 patch. It works just like your patch.

Yes, your #2 is identical to mine.

 

By the way, do we need to patch IOUSBHostFamily as used to?

Edited by PMheart
Link to comment
Share on other sites

9 hours ago, Max.1974 said:

How are you fix the audio? tks!!! 

using the latest versions of lilu and it's plugins and using those boot args: -lilubetaall -lilubeta -alcbeta

Link to comment
Share on other sites

Guest
This topic is now closed to further replies.
×
×
  • Create New...