Jump to content

Chameleon 2.4svn Official PKG Installer


ErmaC
4,261 posts in this topic

Recommended Posts

Hi, tested with Enoch+FileNVRAM.dylib_rc1 , El Capitan.

Worked, NVIDIA web drivers got loaded - one first boot, didnt tried second time.

Q: I have nvram...plist in /Extra, which was detected and used ( i think) on first boot.

I can see that new .nvram.plist was made after the first boot in / .

The content or structure gives some ? - for example the new key NVRAMFile with some (human unreadable data)   :)

Screenshoot shows generated .nvram.plist

post-110586-0-78549000-1490053382_thumb.jpg

 

The old (used with older NVRAMdylib) and found by the new version looks a bit different in structure and content.

post-110586-0-13875600-1490053487_thumb.jpg

 

 

Thanks

  • Like 1
Link to comment
Share on other sites

Hi, tested with Enoch+FileNVRAM.dylib_rc1 , El Capitan.

Worked, NVIDIA web drivers got loaded - one first boot, didnt tried second time.

Q: I have nvram...plist in /Extra, which was detected and used ( i think) on first boot.

I can see that new .nvram.plist was made after the first boot in / .

The content or structure gives some ? - for example the new key NVRAMFile with some (human unreadable data)   :)

Screenshoot shows generated .nvram.plist

attachicon.gifBildschirmfoto 2017-03-21 um 00.45.05.jpg

 

The old (used with older NVRAMdylib) and found by the new version looks a bit different in structure and content.

attachicon.gifBildschirmfoto 2017-03-21 um 00.52.03.jpg

 

 

Thanks

Thanks to you! Is too much asking to remove the FakeSMC.kext and rebuild the cache?

New FileNVRAM.dylib is mostly ok, I have to fix only one bug. You can restore your previous module for now, thanks again!

So really, the issue remains as before in the Yosemite partition which was Ok again once I reverted to boot file of r2848.

If still regard a memory allocation in xml.c maybe is due to the fact that some plist are not present. So I added a check, please try that attached.

Edited by Micky1979
boot file removed
Link to comment
Share on other sites

No, i will try that (check working without Fakesmc on HD or on perlinked kernel).

The second boot was also OK, now new .nvram.plist is used ( even i dont understand the new key NVRAMFile :) ) NvidiaWeb Drivers got loaded again.

FileNVRAM develop version rc1 (Micky1979): scanning for nvram file:
	scanning hd(2,3)/
	scanning hd(2,2)/
	scanning hd(1,7)/
	scanning hd(1,6)/
	scanning hd(1,5)/
	scanning hd(1,4)/
	scanning hd(1,3)/
	scanning hd(1,2)/
	scanning hd(0,2)/
	found nvram.plist at /Volumes/Mac_OSX [hd(0,2)]

Now i remove Fakesmc, cache  and reboot .. 

Link to comment
Share on other sites

No, i will try that (check working without Fakesmc on HD or on perlinked kernel).

The second boot was also OK, now new .nvram.plist is used ( even i dont understand the new key NVRAMFile :) ) NvidiaWeb Drivers got loaded again.

FileNVRAM develop version rc1 (Micky1979): scanning for nvram file:
	scanning hd(2,3)/
	scanning hd(2,2)/
	scanning hd(1,7)/
	scanning hd(1,6)/
	scanning hd(1,5)/
	scanning hd(1,4)/
	scanning hd(1,3)/
	scanning hd(1,2)/
	scanning hd(0,2)/
	found nvram.plist at /Volumes/Mac_OSX [hd(0,2)]

Now i remove Fakesmc, cache  and reboot .. 

NVRAMFile is the path to the loaded file. If you paste the hex data into Hex Fiend.app you will see the human readable path!

Link to comment
Share on other sites

... back ( after: sudo rm -f /System/Library/Caches/com.apple.kext.caches/Startup/kernelcache && sudo kextcache -system-prelinked-kernel )

booted also, without FakeSMC :)

 

Q:

I looked in the kextstat, can it be that those two entries below came by the new  internal boot injection for FameSMC / NVRAM? I am sure there are no FakeSMC or Nvram kexts anymore.

22    0 0xffffff7f82902000 0x19000    0x19000    org.netkas.driver.FakeSMC (1421) 721BC019-EC98-335A-9955-90C0E3954B71 <11 7 5 4 3 1>

38    0 0xffffff7f8291b000 0x7000     0x7000     com.xZenue.kext.FileNVRAM (6.9.5) 97780849-340F-3CCC-9EE4-4A9C7BD63C42 <7 5 4 3 1>

 

in the bdmesg is no info about like " no FakeSMC found - use internal ..." 

 

 
 

Link to comment
Share on other sites

FileNVRAM.kext is embedded in FileNVRAM.dylib,

FakeSMC.kext is embedded in Enoch and is v1421 as your log show :D. My bdmesg show "Loading embedded FakeSMC: success!"

EDIT
Use DarwinDumper and post the kernel log, if you see:

Mar 20 22:19:25 localhost kernel[0] <Notice>: FakeSMC (Chameleon embedded, commit 9b07a8f) v6.24.1421 Copyright 2017 netkas, slice, usr-sse2, kozlek, navi, THe KiNG, RehabMan. All rights reserved. 

you will know that is loaded by Enoch

Link to comment
Share on other sites

NVRAMFile is the path to the loaded file. If you paste the hex data into Hex Fiend.app you will see the human readable path!

 

Thanks, yep now i can see the key content - in my case    /Volumes/Mac_OSX/.nvram.plist

 

 

Yep, now i found also the Loading embedded FakeSMC: success!

information near the end :geek:  of the bdmesg  . :thumbsup_anim:
 
 
Perhaps sometimes we will see: Patchings for AMD Ryzen: done & success! 
:) 
  • Like 1
Link to comment
Share on other sites

 

Thanks, yep now i can see the key content - in my case    /Volumes/Mac_OSX/.nvram.plist

 

 

Yep, now i found also the Loading embedded FakeSMC: success!

information near the end of the bdmesg  .

 

 

good!


 

 
Perhaps sometimes we will see: Patchings for AMD Ryzen: done & success! 
:) 

 

that is hard w/o having an AMD system..

  • Like 1
Link to comment
Share on other sites

Hi Hervè, I've corrected a miscalculation, please try and report, should work now. Also if you can, test the news below  :) .

PS anyway to disable the kexts patcher you can type "SkipKextPatcher=Yes" (something similar also for the kernel patcher: just "SkipKP=No")

 

for all

Enoch rc1 and FileNVRAM.dylib rc1

 

Enoch rc1 has the ability to patch kernel and kexts (prelinked or not) like in the previous test versions made available some days ago in this Topic.

I've removed the internal nvram support because I worked with meklort to fix some little bugs in the old code (continue reading below).

News:

now Enoch know if you load the FakeSMC.kext or not. So what?

Well, if not found on the filesystem or not prelinked now Enoch load its embedded version automatically.

If you glad or you want make tests for me (please), you can remove it and see it loaded by Enoch.  Please report back, Ermac and I would release that code.

 

FileNVRAM.dylib module attached is a develop version of my fork (unpushed). As I said I worked with meklort to fix it, but this is radically changed and I need feedback to make a new pull request to the mantainer (meklort). So as always the module goes in Extra/module/ of the boot partition. 

PS this module dump the nvram in /.nvram.plist (hidden), but if you have the old one in /Extra/ will read this for the first time (..and w/o apporting changes to it). Later the new file location will be always on the root.

(you don't need NVRAM Dumper Pro.app to dump the nvram, and if active you should keep of it)

 

Please report back.

Thanks for this new boot file.

But for Z87 hackintosh it has the same issue as previous boot r.2851 both of which can not load NvidiaWebDriver at 10.12.x even booted with nvda_drv=1

Previous Enoch r.2848 can load it properly with the same prelinkedkernel.

Link to comment
Share on other sites

Thanks for this new boot file.

But for Z87 hackintosh it has the same issue as previous boot r.2851 both of which can not load NvidiaWebDriver at 10.12.x even booted with nvda_drv=1

Previous Enoch r.2848 can load it properly with the same prelinkedkernel.

@jsl, off course that does not work, I posted a FileNVRAM.dylib to work with nvram (the only requirement to load the NVDIA Web drivers in 10.12+). Even tried to install it and load the drivers through the NVIDIA web manager (to add the required argument in nvram) and reboot if see if that work???

 

FileNVRAM.dylib = nvram support = You can load the NVDIA Web drivers in Sierra

 

Anyway, I made a new build with a new key that you can put in kernel.plist or just type at boot time:

NVIDIAWebDrv=Yes/No

only for Sierra, and for who did not install FileNVRAM.dylib.

Edited by Micky1979
boot file removed
  • Like 3
Link to comment
Share on other sites

@jsl, off course that does not work, I posted a FileNVRAM.dylib to work with nvram (the only requirement to load the NVDIA Web drivers in 10.12+). Even tried to install it and load the drivers through the NVIDIA web manager (to add the required argument in nvram) and reboot if see if that work???

 

FileNVRAM.dylib = nvram support = You can load the NVDIA Web drivers in Sierra

 

Anyway, I made a new build with a new key that you can put in kernel.plist or just type at boot time:

NVIDIAWebDrv=Yes/No

only for Sierra, and for who did not install FileNVRAM.dylib.

 

Thanks for your prompt response and help.

In previous test I have included FileNVRAM.dylib in /Extra/modules, but it did not work.

For this new test Boot r.2852 it still not working yet at both 10.12.3 and 10.12.4.

Only r.2848 working perfectly for my Z87 hackintosh now.

Another choice is Clover r.4035 which is also working too !

Link to comment
Share on other sites

@jsl, off course that does not work, I posted a FileNVRAM.dylib to work with nvram (the only requirement to load the NVDIA Web drivers in 10.12+). Even tried to install it and load the drivers through the NVIDIA web manager (to add the required argument in nvram) and reboot if see if that work???

 

FileNVRAM.dylib = nvram support = You can load the NVDIA Web drivers in Sierra

 

Anyway, I made a new build with a new key that you can put in kernel.plist or just type at boot time:

NVIDIAWebDrv=Yes/No

only for Sierra, and for who did not install FileNVRAM.dylib.

 

Every new (3) test boot files, NOT work for me.

Only (old) boot works in my computer.

 

 

I reinstall a Mac OS, but I don't find Enoch r2852, where is possibile to download it?

 

IMG_1954.jpg

Link to comment
Share on other sites

@Gigamaxx, you know that you are talking about an AMD cpu? Sorry but here you need a developer that already own such that processor and that know what to do and which provide code..

And boot0:error  just  means that Enoch is bad installed, maybe you can have better results formatting your pen drive in Fat32.

Thanks, yes it's an AMD, but it's Ryzen! Lol! I'll try fornatting in Fat32 and give it a go. Do you know where to turn next, as it seems these CPUs don't like Clover UEFI booting at all, and at least in legacy mode I get some error codes?

Link to comment
Share on other sites

In previous test I have included FileNVRAM.dylib in /Extra/modules, but it did not work.

For this new test Boot r.2852 it still not working yet at both 10.12.3 and 10.12.4.

Only r.2848 working perfectly for my Z87 hackintosh now.

Web Drivers need nvd_drv 31 (hex byte that means 1) as nvram key/object in Sierra, unlike in El Capitan that is among the bootArgs and is a string (nvda_drv=1)

 

For this new test Boot r.2852 it still not working yet at both 10.12.3 and 10.12.4.

Only r.2848 working perfectly for my Z87 hackintosh now.

 

Then I suspect you did something wrong because here work (is the same code of 2848 that I personally wrote). The requirement as I said, unlike 2848, is that key in kernel.plist:

<key>NVIDIAWebDrv</key>
	<true/>

and offcourse if you have an Extra folder inside the Sierra's partition the kernel.plist with this key must be there. Otherwise inside the Extra folder of the boot partition.

The bdmesg should looks like:

[ KEXTS PATCHER START ]
	Patching NVDAStartupWeb:
	  1 substitutions made!

Another choice is Clover r.4035 which is also working too !

 

Clover is my primary bootloader, and as soon I finished here I hope some code will finish in Clover as well. Also FileNVRAM.kext is runnable in Clover using a special boot arg made on purpose (meklort, the mantainer, accepted and commited this).

Newer versions are just for testing and posted in this thread by Micky in the last few days.

Right, and since I wrote hundreds of lines of new code with the help of no one.. also I want to finish and debug it before something will be published. New functionalities at the cost that someone have the patience to help in debugging. Lately I made my time on it, someone else should for kindness at least understand the meaning of testing/debugging. To the end as jsl said Clover works well, but I just love programming.. with what I can off course :wink_anim:.

EDIT

this bootloader is alive with the help of ErmaC, bungo, Pike R. Alpha/Anv and MinusZwei (yosemite/el capitan time), Bronya for amd, Zenith432 that have made commits lately, me. Big thanks to @crazybirdy (main tester), mich_de lately and Hervè that has notify me a bug (what I'm looking for). Off course the old chameleon team and all who write codes. Also I have to say that Chameleon has an intensive use of "made by meklort" if you look at the source code that in truth also  appear in other bootloaders ;), and I assure you that code made in past is incredibly actual and well done.

Thanks, yes it's an AMD, but it's Ryzen! Lol! I'll try fornatting in Fat32 and give it a go. Do you know where to turn next, as it seems these CPUs don't like Clover UEFI booting at all, and at least in legacy mode I get some error codes?

Enoch is different from Clover, and as I said I have no hardaware to test and still I have no datas about this CPU.. so nothing can be done.

In addition the bootloader cannot do miracles. In fact that support should already be done inside the modded AMD kernel... so that this really should be the first step (but no one has provided data). What to do then?

  • Like 5
Link to comment
Share on other sites

Web Drivers need nvd_drv 31 (hex byte that means 1) as nvram key/object in Sierra, unlike in El Capitan that is among the bootArgs and is a string (nvda_drv=1)

 

 

 

 

Then I suspect you did something wrong because here work (is the same code of 2848 that I personally wrote). The requirement as I said, unlike 2848, is that key in kernel.plist:

NVIDIAWebDrv
and offcourse if you have an Extra folder inside the Sierra's partition the kernel.plist with this key must be there. Otherwise inside the Extra folder of the boot partition.

The bdmesg should looks like:

[ KEXTS PATCHER START ]	Patching NVDAStartupWeb:	  1 substitutions made!
Clover is my primary bootloader, and as soon I finished here I hope some code will finish in Clover as well. Also FileNVRAM.kext is runnable in Clover using a special boot arg made on purpose (meklort, the mantainer, accepted and commited this).

 

Right, and since I wrote hundreds of lines of new code with the help of no one.. also I want to finish and debug it before something will be published. New functionalities at the cost that someone have the patience to help in debugging. Lately I made my time on it, someone else should for kindness at least understand the meaning of testing/debugging. To the end as jsl said Clover works well, but I just love programming.. with what I can off course :wink_anim:.

EDIT

this bootloader is alive with the help of ErmaC, bungo, Pike R. Alpha/Anv and MinusZwei (yosemite/el capitan time), Bronya for amd, Zenith432 that have made commits lately, me. Big thanks to @crazybirdy (main tester), mich_de lately and Hervè that has notify me a bug (what I'm looking for). Off course the old chameleon team and all who write codes. Also I have to say that Chameleon has an intensive use of "made by meklort" if you look at the source code that in truth also  appear in other bootloaders ;), and I assure you that code made in past is incredibly actual and well done.

 

Enoch is different from Clover, and as I said I have no hardaware to test and still I have no datas about this CPU.. so nothing can be done.

In addition the bootloader cannot do miracles. In fact that support should already be done inside the modded AMD kernel... so that this really should be the first step (but no one has provided data). What to do then?

What kind of data do you need? Later I will try to extract what I can grasp from CPU z or Aida in Windows. Any hints on how and what to provide you with, any help would be greatly appreciated?

This is all I have on my iPad at the moment.

 

Link to comment
Share on other sites

Web Drivers need nvd_drv 31 (hex byte that means 1) as nvram key/object in Sierra, unlike in El Capitan that is among the bootArgs and is a string (nvda_drv=1)

 

 

Then I suspect you did something wrong because here work (is the same code of 2848 that I personally wrote). The requirement as I said, unlike 2848, is that key in kernel.plist:

<key>NVIDIAWebDrv</key>
	<true/>

and offcourse if you have an Extra folder inside the Sierra's partition the kernel.plist with this key must be there. Otherwise inside the Extra folder of the boot partition.

The bdmesg should looks like:

[ KEXTS PATCHER START ]
	Patching NVDAStartupWeb:
	  1 substitutions made!
I have
<key>NVIDIAWebDrv</key>

    <true/>

in my kernel.plist in both EFI/Extra and inside the Extra folder of the boot 10.12.3/10.12.4 partitions.
Nvidia Web Driver was not loaded if booted by this new Boot r.2852 version.
However Boot r.2848 always worked without this issue.
Maybe it's my hardware incompatibility.
Waiting for someone else posts the same issue as mine.
Link to comment
Share on other sites

ehi Micky1979 I try a new thing :)

for the first time I would try to patch the famous FileNVRAM.kext (I download a file for El Capitan... I think, for security I attached it if you want to see)

in this way the new boot file provided by Micky1979 work perfectly and also better

now I have also the nvidia driver selected:

 

Senza_titolo.jpg

 

if you have an another and better FileNVRAM.kext to make me test in Sierra, I will do ^_^

i'm waiting for you, best regards

:thumbsup_anim:

 

PS: so r2852 is called by the boot I use

PS2: salutami peppe!

FileNVRAM.kext.zip

Link to comment
Share on other sites

 

I have
<key>NVIDIAWebDrv</key>

    <true/>

in my kernel.plist in both EFI/Extra and inside the Extra folder of the boot 10.12.3/10.12.4 partitions.
Nvidia Web Driver was not loaded if booted by this new Boot r.2852 version.
However Boot r.2848 always worked without this issue.
Maybe it's my hardware incompatibility.
Waiting for someone else posts the same issue as mine.

 

The goal is to use the NVDIA Web manager to use the nvram like in a real mac.  r2848 just use a workaround by patching the kext (if nvda_drv=1 is present in the Kernel Flags) to look in a fake node created by the bootloader in that event. But, if we are testing to make the nvram in a working state ... using the same method used in r2848... the NVDIA Web manager will always report the default drivers in use also if in reality the web drivers are loaded. For this reason I don't want this do be automatic and I added "NVIDIAWebDrv" key.

 

Try this one athached, yesterday (in the rush) may I commited a typo that prevent "NVIDIAWebDrv" to work correctly. Also try to rebuild the kernel cahe if again has no effect.

 

 

ehi Micky1979 I try a new thing :)

for the first time I would try to patch the famous FileNVRAM.kext (I download a file for El Capitan... I think, for security I attached it if you want to see)

in this way the new boot file provided by Micky1979 work perfectly and also better

now I have also the nvidia driver selected:

 

Senza_titolo.jpg

 

if you have an another and better FileNVRAM.kext to make me test in Sierra, I will do ^_^

i'm waiting for you, best regards

:thumbsup_anim:

 

PS: so r2852 is called by the boot I use

PS2: salutami peppe!

Ciao Gabriele, the FileNVRAM.kext posted by you is really old w/o latest changes. Try FileNVRAM.dylib posted by me on previous pages (it goes in Extra/modules inside the boot partition, i.e where the boot file reside).

Anyway as I said a bug  is present (I think the last to completely fix the module) that cause a double xml footer to be present on the dumped file. Is also present in the old kext you posted here otherwise the NVDIA Web Driver's should worked as expected.

 

PS the FileNVRAM.kext posted by you has the "com.apple.iokit.IOStorageFamily" key set, that means that will not work on older version of OSX and also that probably will not work on newer incoming one. 

The original does not have this limitation.

Edited by Micky1979
boot file removed
  • Like 1
Link to comment
Share on other sites

The goal is to use the NVDIA Web manager to use the nvram like in a real mac.  r2848 just use a workaround by patching the kext (if nvda_drv=1 is present in the Kernel Flags) to look in a fake node created by the bootloader in that event. But, if we are testing to make the nvram in a working state ... using the same method used in r2848... the NVDIA Web manager will always report the default drivers in use also if in reality the web drivers are loaded. For this reason I don't want this do be automatic and I added "NVIDIAWebDrv" key.

 

Try this one athached, yesterday (in the rush) may I commited a typo that prevent "NVIDIAWebDrv" to work correctly. Also try to rebuild the kernel cahe if again has no effect.

 

attachicon.gifboot_rc2.zip

 

Thanks for this new boot rc2 file, but unfortunately it got KP at first reboot after rebuilt prelinkedkernel.

After fixation and reboot it can login desktop, but both NvidiaWebDriver and all audio from ALC-1150, DP audio from Intel HD 4600, and HDMI audio from GTX-980 were disabled.

The KP picture was attached:

post-70188-0-58011400-1490193721_thumb.jpeg

Link to comment
Share on other sites

The KP picture was attached:

thanks

 

After fixation and reboot it can login desktop, but both NvidiaWebDriver and all audio from ALC-1150, DP audio from Intel HD 4600, and HDMI audio from GTX-980 were disabled.

That only means that the prelinkedkernel has been reconstructed incorrectly, and it is usual thing and symptom of existing problems (some tool also still create the /Extra/Extensions.mkext ... that is  prehistoric thing that will not be loaded by the kernel but that ended up in the memory map as garbage). Mind that the bootloader read things in memory but does nothing on the filesystem.

So I'll advice you to perform the same steps described in this post:             #3981            

Link to comment
Share on other sites

Finally I solved all issue with the FileNVRAM.dylib (e.g. iMessage working by set my MLB/ROM, and my Sierra load the Web drivers by just enabling them through the official "NVIDIA Driver Manager"  :thumbsup_anim:, just like a real Mac).

In Enoch I fixed some potential bugs reading kernel.plist and kexts.plist.

 

@mitch_de since there was a bug with the old module I published, I advice you to:

  1. boot without FileNVRAM.dylib (or kext if any)
  2. delete all hidden .nvram.plist created on each partition that may have it.
  3. install the new Enoch and the new FileNVRAM.dylib and reboot once with new stuff.

off course before rebooting again ensure that you selected the Web Driver in the NVIDIA Driver Manager. Let me know.

 

@Hervé, a test by you is really appreciated to see if I fixed the problem encountered by you. @jsl, the same, plz

@gabryele, looks your Web drivers working in Sierra as well..

 

 

Files:

[EDIT use the boot file at             #4135            ]

FileNVRAM.dylib_rc3.zip

 

examples of how use new plist:

kernel.plist.zip

kexts.plist.zip

Edited by Micky1979
boot_rc3 removed
  • Like 4
Link to comment
Share on other sites

Finally I solved all issue with the FileNVRAM.dylib (e.g. iMessage working by set my MLB/ROM, and my Sierra load the Web drivers by just enabling them through the official "NVIDIA Driver Manager"  :thumbsup_anim:, just like a real Mac).

In Enoch I fixed some potential bugs reading kernel.plist and kexts.plist.

 

@mitch_de since there was a bug with the old module I published, I advice you to:

  1. boot without FileNVRAM.dylib (or kext if any)
  2. delete all hidden .nvram.plist created on each partition that may have it.
  3. install the new Enoch and the new FileNVRAM.dylib and reboot once with new stuff.

off course before rebooting again ensure that you selected the Web Driver in the NVIDIA Driver Manager. Let me know.

 

@Hervé, a test by you is really appreciated to see if I fixed the problem encountered by you. @jsl, the same, plz

@gabryele, looks your Web drivers working in Sierra as well..

 

 

Files:

attachicon.gifboot_rc3.zip

attachicon.gifFileNVRAM.dylib_rc3.zip

 

examples of how use new plist:

attachicon.gifkernel.plist.zip

attachicon.gifkexts.plist.zip

Thanks for this new Boot rc3 file and your help.

However no matter how I have tried these Boot rc1, rc2 & rc3 files not working in my Z87 hackintosh.

At present only Enoch r.2848 and Clover r.4035 work.

I'll keep on trying to find out the reason why !

Link to comment
Share on other sites

Hi,

 

i did: 

 

  1. boot without FileNVRAM.dylib (or kext if any)
  2. delete all hidden .nvram.plist created on each partition that may have it.
  3. install the new Enoch and the new FileNVRAM.dylib and reboot once with new stuff.

Last reboot (after using new boot / new FileNVRAM.dylib, new kernel / kext.plist structure,

i got Chameleon error HALTED (screenshoot). 

post-110586-0-20788000-1490263339_thumb.jpg

 

Attached kexts.plist and kernel.plist.

 

Now i am on backup partition (older Chameleon ) and copy boot RC1 (which worked) back on / .

I let new FileNVRAM.dylib (RC3) and try to boot with boot RC1 +  FileNVRAM.dylib RC3....

 

 

 

 

 

 

@mitch_de since there was a bug with the old module I published, I advice you to:

  1. boot without FileNVRAM.dylib (or kext if any)
  2. delete all hidden .nvram.plist created on each partition that may have it.
  3. install the new Enoch and the new FileNVRAM.dylib and reboot once with new stuff.

off course before rebooting again ensure that you selected the Web Driver in the NVIDIA Driver Manager. Let me know.

 

Files:

attachicon.gifboot_rc3.zip

attachicon.gifFileNVRAM.dylib_rc3.zip

 

examples of how use new plist:

attachicon.gifkernel.plist.zip

attachicon.gifkexts.plist.zip

kexts.plist.zip

kernel.plist.zip

  • Like 1
Link to comment
Share on other sites

????

Went back to RC1 + FileNVRAM RC1 (which worked before i tried RC 3) = same chameleon Error , but now in line 365 (RC3 line 371) ?

How can that be?

Did RC3 something with the prelinked kernel so that RC1 fail also?

 

EDIT: I now set ALL patch to NO in kexts.plist - RC1 + FileNVRAM RC1 worked again :)

<?xml version="1.0" encoding="UTF-8"?>
<!DOCTYPE plist PUBLIC "-//Apple//DTD PLIST 1.0//EN" "http://www.apple.com/DTDs/PropertyList-1.0.dtd">
<plist version="1.0">
<dict>
	<key>AICPMPatch</key>
	<false/>
	<key>AppleRTCPatch</key>
	<false/>
	<key>OrangeIconFixSata</key>
	<false/>
	<key>TrimEnablerSata</key>
	<false/>
</dict>
</plist>

Are the key values always boolean (also newer RC3) like in RC1?

 

Something wrong with kext patcher, because i set all to NO,

bdmesg shows i patched something!!!

GA-EP35:~ andreasm$ bdmesg
Enoch (2852 Micky1979 rc1)
Logging started: 2017/03/23, 11:10:38
....
Boot UUID of 'Mac_OSX'  (GPT HFS+): A887B8AC-171B-3F08-B2E1-2C85B032D013
Kernel Cache file path (OS X 10.11 and newer): /System/Library/PrelinkedKernels/prelinkedkernel
Loading Kernel Cache from: 'Mac_OSX' (GPT HFS+)
Booting on OS X 10.11.6 (15G1217)
...
[ KERNEL PATCHER START ]

	Will patch for 15.6.0 kernel version.

	KernelBooter_kexts state:  enabled!
	KernelPm           state: disabled!
	KernelLapicError   state: disabled!
	KernelLapicVersion state: disabled!
	KernelHaswell      state: disabled!
	KernelcpuFamily    state: disabled!
	KernelSSE3         state: disabled!
	- Searching for booter extensions pattern:
	Found El Capitan SIP pattern: patched!
	Found EL Capitan EXT pattern: patched!
	2 substitution(s) made.
Kernel patcher: done!

[ KEXTS PATCHER START ]
	Patching IOAHCIBlockStorage: 1 substitutions made!  
	Patching AppleAHCIPort: 0 substitutions made!
Kexts patcher: done!


Why   Patching IOAHCIBlockStorage: 1 substitutions made!  , if i set all to NO / this key isnt set to anything?!

The patched  IOAHCIBlockStorage isnt in the kext.plist (not set), and in such a case - patch part isnt set to YES/NO, the patcher should always (default) NOT!!! patch this part.

 Perhaps thats the reason for the error / halted - chameleon tries to patch something should not patch OR this key isnt in the kext.plist at all > memory error?

  • Like 1
Link to comment
Share on other sites

×
×
  • Create New...