Jump to content
ErmaC

Chameleon 2.4svn Official PKG Installer

4,337 posts in this topic

Recommended Posts

Hmm, well, as I said in post #3254, I did move line #788 to line #797 in order to stop the device id replacement; that should have been just the same as your suggested move too line #793.

 

What I don't understand is why my generated boot file worked Ok booting Mav and Yos but caused systematic laptop reset when attempting to boot 10.11. That baffles me...

 

In terms of 10.11 AzulFB + HD5000 kexts patching, it's pretty much the same as for Yosemite:

1) add Intel mobile HD4600 id to Info.plist file of each kext, i.e change from this

<key>IOPCIPrimaryMatch</key><string>0x0d268086 0x0a268086 0x0a2e8086 0x0d228086 0x04128086</string>

to this

<key>IOPCIPrimaryMatch</key><string>0x0d268086 0x0a268086 0x0a2e8086 0x0d228086 0x04128086 0x04168086</string>

in both kexts.

 

2) using hex editor 0xED, replace Intel desktop HD4600 id by that of Intel mobile HD4600 in the binary file found in the Contents/MacOS subfolder of the kext, i.e. change this

86 80 12 04

to this

86 80 16 04

There's only 1 instance in the binary file of each kext in 10.11 as far as I've seen so far.

 

 

Could you send me your modified gma.c file so that I compare with my earlier mod? Thanks.

Share this post


Link to post
Share on other sites
Advertisement

 

 

2) using hex editor 0xED, replace Intel desktop HD4600 id by that of Intel mobile HD4600 in the binary file found in the Contents/MacOS subfolder of the kext, i.e. change this

86 80 12 04

to this

86 80 16 04

There's only 1 instance in the binary file of each kext in 10.11 as far as I've seen so far.

is what the bootloader do by default, ie as before, as at svn, as the one in the download section! :|  :cry:  :wacko:  :blink:  :huh:  :o  :shock:  :mad:

Why don't work?

mmmh, have you try to remove all your DSDT  patch like this:

 

      Device (IGPU)

        {

            Name (_ADR, 0x00020000)

            Method (_DSM, 4, NotSerialized)

            {

                Store (Package (0x08)

                    {

                        "AAPL,slot-name", 

                        "Built-in", 

                        "device-id", 

                        Buffer (0x04)

                        {

                             0x16, 0x04, 0x00, 0x00

                        }, 

                        "AAPL,ig-platform-id", 

                        Buffer (0x04)

                        {

                             0x06, 0x00, 0x26, 0x0A

                        }, 

                        "hda-gfx", 

                        Buffer (0x0A)

                        {

                            "onboard-1"

                        }

                    }, Local0)

                DTGP (Arg0, Arg1, Arg2, Arg3, RefOf (Local0))

                Return (Local0)

            }

???

Share this post


Link to post
Share on other sites

I understand what you mean with id replacement Micky, but in my case it's never worked with vanilla kexts. Neither in Yos, nor in ElCapitan. I've tried without the HD4600 id 0416 in the DSDT IGPU patch of course, keeping just the ig-platform-id injection; I've tried without any DSDT IGPU patch at all, just using InjectIntel-ig=0600260a or IntelAzulFB=12 options; I've tried without any DSDT at all; I've even tried all 15 AzulFB/ig-platform-id values; all to no good! Never managed to suss out why... All I can think of is some conflict or incompatibility between Intel desktop HD4600 id 0412 and layout-id 0a260006.

 

I can only get full QE/CI by patching the kexts for mobile id 0416 and use AzulFB #12, hence my requirement to avoid IGPU device id replacement in the bootloader.

 

Re: Instant Menu screen corruption, it happens only with the test file you've sent me, so it's clearly linked to it. It is systematic and happens whether I use my patched DSDT or not or whether I remove the IGPU patch from the DSDT. I don't think it's related to DSDT as I understand that the table has not even loaded at Instant Menu time, has it?

 

So really, all I need is the same as ErmaC's r2737 but without the id replacement and without Instant Menu delay + screen corruption. Maybe there are certain options to provide at "make" time?

 

Oh, here are bdmesg files for you: one when I boot without AzulFB injection (-> no QE/CI) and one with AzulFB injection (-> QE/CI).

bdmesg_No-AzulFB-injection.zip bdmesg_with_AzulFB-injection.zip

Share this post


Link to post
Share on other sites

I understand what you mean with id replacement Micky,...

attachicon.gifbdmesg_No-AzulFB-injection.zip attachicon.gifbdmesg_with_AzulFB-injection.zip

Yes, istead of patch binary on filesystem...just deceive OSx before, so if a fake 0412 is detected the driver should be ok with this...

Thanks for the files, but now I will see them tomorrow... here .... 3:39 AM, -_-  good night

Share this post


Link to post
Share on other sites

Hervè, to be vanilla, your test should be done with vanilla kexts too, otherwise if you had replaced the id in the kext binary then is normal that cannot work ;)

Share this post


Link to post
Share on other sites

:whistle: As I said in my previous posts, 'done it, 'been there, 'does not work. :thumbsdown_anim:

 

I have not consulted the HD4600 Yosemite thread for a while but I'm not sure anyone has actually succeeded with mobile id replacement to desktop HD4600 0412 + vanilla kexts + Azul FB injection.

 

Again, in the case of my E6440 with i5-4300M and mobile Intel HD graphics 4600 (id 0x0416), booting Yos 10.10.x or El Capitan DBx with:

  • Enoch r2737 (with 0416 id replacement by id 0412) + boot option InjectIntel-ig=0600260a (or IntelAzulFB=12) + vanilla AzulFB & HD5000Graphics kexts = framebuffer support only, no QE/CI
  • Clover 3224 + InjectIntel=Yes + FakeID=04120000 + FBConfig=0600260a + vanilla AzulFB & HD5000Graphics kexts = framebuffer support only, no QE/CI

In both cases, same results with or without DSDT injection of device id 0412 and/or ig-platform-id 0x0a260006.

 

That's why I believe Desktop HD 4600 (id 0412) + Azul FB #12 (id 0x0a260006) = incompatibility.

 

Maybe this id replacement business works for Haswell mobile HD4400 and/or Iris, but it certainly does not for mobile HD4600, at least on the Dell E6440!

Share this post


Link to post
Share on other sites

:whistle: As I said in my previous posts, 'done it, 'been there, 'does not work. :thumbsdown_anim:

 

I have not consulted the HD4600 Yosemite thread for a while but I'm not sure anyone has actually succeeded with mobile id replacement to desktop HD4600 0412 + vanilla kexts + Azul FB injection.

 

Again, in the case of my E6440 with i5-4300M and mobile Intel HD graphics 4600 (id 0x0416), booting Yos 10.10.x or El Capitan DBx with:

  • Enoch r2737 (with 0416 id replacement by id 0412) + boot option InjectIntel-ig=0600260a (or IntelAzulFB=12) + vanilla AzulFB & HD5000Graphics kexts = framebuffer support only, no QE/CI
  • Clover 3224 + InjectIntel=Yes + FakeID=04120000 + FBConfig=0600260a + vanilla AzulFB & HD5000Graphics kexts = framebuffer support only, no QE/CI
In both cases, same results with or without DSDT injection of device id 0412 and/or ig-platform-id 0x0a260006.

 

That's why I believe Desktop HD 4600 (id 0412) + Azul FB #12 (id 0x0a260006) = incompatibility.

 

Maybe this id replacement business works for Haswell mobile HD4400 and/or Iris, but it certainly does not for mobile HD4600, at least on the Dell E6440!

 

Enoch 2748 is avaiable in the download section.

 

I "turned off" the device id substitution for your gfx IDs.

 

Let me know

 

ErmaC

Share this post


Link to post
Share on other sites

Great, I'll go try it out. Thanks.
 
Edit: it's... absolutely perfect!  :thumbsup_anim: Boots my Mav/Yos/ElCap partitions, no change of HD4600 mobile id 0416 and no more Instant Menu screen corruption. Only the Apple logo boot screen is a bit different than with earlier versions (we're back to legacy logo by default but no progress bar except at the very end and, then, without Apple logo) but not to worry. Many thanks ErmaC.
 
r2748:
post-851564-0-36834700-1438391284_thumb.jpg
 
FYI, the boot screen behaviour is as follows (same computer and same Themes folder) :
r2737:
post-851564-0-20789700-1438393531.jpg post-851564-0-13800500-1438393548.jpg
 
r2748:
post-851564-0-02516400-1438393555.jpg post-851564-0-37598700-1438393562.jpg

Share this post


Link to post
Share on other sites

Looks like for El Capitan DP4/DP5 and PB3 Enoch not work anymore....

Until now the CSR (Code Signing Restrictions) was set to 0x01 (00000001) (Thx Pike)
so only untrusted kext was allowed...

Pls try this UNTESTED versions:

  • CsrActiveConfig (Hex 0x00) (Binary 00000000) (Decimal 0)
     
  • CsrActiveConfig (Hex 0x67) (Binary 01100111) (Decimal 103)
     
  • CsrActiveConfig (Hex 0x65) (Binary 01100101) (Decimal 101)
     

Let me know...

ErmaC

Share this post


Link to post
Share on other sites

boot in clover, touch /s/l/e then boot in chameleon

Thanks and tried as advised, got the same kernel Panic as before.

Just touch /S/L/E seems not enough for me.

I guess Clover and Chameleon applied different ways to handle kernelcache.

Share this post


Link to post
Share on other sites

I test it as yours as below with DB6.

2732 and below, you can boot with /System/Library/PrelinkedKernels/prelinkedkernel.

2737, 2748 can boot only with /System/Library/Caches/com.apple.kext.caches/Startup/kernelcache.

 

If prelinkedkernel and kernelcache didn't exist, it will boot as -f as loading all kexts, and get KP as yours.

So, you need prelinkedkernel or kernelcache, get it with Clover first.

 

[edit]

Another findings, test with DB6, 2732 and below boot fine with prelinkedkernel, 2737, 2748 can't boot with kernelcache as your reply.

2737, 2748 didn't work here with DB6.

I think, that's could be the problem of Chameleon.

 

update:

Test with DB6.

1.Rebuild prelinkedkernel and kernelcache by Kext Utility 2.6.4 with Clover 2353.

Chameleon 2732 can boot with prelinkedkernel, 2737, 2748 can't, (delete kernelcache first)

reboot as pics below.

attachicon.gifreboot.jpg

 

2.Rebuild prelinkedkernel and kernelcache by Kext Utility 2.6.4 with Chameleon 2732, and 2748.

Chameleon 2732, 2737, 2748, Clover 2353 all can boot with prelinkedkernel. (delete kernelcache first)

boot fine.

attachicon.gifboot.jpg

 

Something wrong with 2737, 2748, i think. FYI.

can you pleeeeease :drool:  test this boot file made for @Hervé five days ego?

Share this post


Link to post
Share on other sites

My new kext injection is huge. A lot of code is required to strip off the code signature/symbols of the kext(s) and to fix up the addresses and sizes so I was thinking of having a (command line) tool/app to prepare the kext for injection, with just the data and binary, so that the source code can be smaller and faster. What do you think?

Share this post


Link to post
Share on other sites

yep find a kext and installing it somewhere  is not far to do a transformation in "data" ready for the bootloader. You mean this as already done by some firmware?

 

Pike you need only the info.plist and the /MacOS/binary, right? 

or you need only some segment of the mach-o binary?...like creating then a .bin file?

 

...maybe I can help with an app

Share this post


Link to post
Share on other sites

update #3285 and 2737-Hervé, all work fine. :thumbsup_anim: :thumbsup_anim: :blush:

THX the CSR config is ok, then the problem is elsewhere.

Guys I'v just made an app to save the NVRAM to the file system... If I've to do not pay for a certificate for the helper tool the game is over!

Share this post


Link to post
Share on other sites

My new kext injection is huge. A lot of code is required to strip off the code signature/symbols of the kext(s) and to fix up the addresses and sizes so I was thinking of having a (command line) tool/app to prepare the kext for injection, with just the data and binary, so that the source code can be smaller and faster. What do you think?

I think there is a way to create kernelcache ab ovo at bootloader time including custom kernel and all kexts from le, sle and ee.

Just make the result to be correct.

Share this post


Link to post
Share on other sites

Hey guys,

 

Can someone help me with this kernel panic? Running 10.11 DB5. Tried the last 3 Enoch's revisions but no luck. I've upgraded from 10.10.4 with success. Got the DB2 update, worked fine, then the DB3, same thing, then after DB4 kernel panic. Since I had the boot loader on usb stick, I've plugged the hd to my iMac (27-late 2013). Removed the kexts for my Hac (from S/L/E), which basically where just 3, FakeSMC, VoodooHDA and AppleIntelE1000e (all the most recent versions), and booted from this hd on iMac. Updated to DB5 (I thought maybe will fix the error this way), installed the kexts again, used Kext Utility 2.6.4 to rebuild cache and hooked the hd back to my Hac (PC). Same kernel panic like in DB4. My specs - signature.

 

Thanks to anyone who could help, point me in the right direction, to fix it.

post-159649-0-26629300-1438894838_thumb.jpg

Share this post


Link to post
Share on other sites

Hey guys,

 

Can someone help me with this kernel panic? Running 10.11 DB5. Tried the last 3 Enoch's revisions but no luck. I've upgraded from 10.10.4 with success. Got the DB2 update, worked fine, then the DB3, same thing, then after DB4 kernel panic. Since I had the boot loader on usb stick, I've plugged the hd to my iMac (27-late 2013). Removed the kexts for my Hac (from S/L/E), which basically where just 3, FakeSMC, VoodooHDA and AppleIntelE1000e (all the most recent versions), and booted from this hd on iMac. Updated to DB5 (I thought maybe will fix the error this way), installed the kexts again, used Kext Utility 2.6.4 to rebuild cache and hooked the hd back to my Hac (PC). Same kernel panic like in DB4. My specs - signature.

 

Thanks to anyone who could help, point me in the right direction, to fix it.

No idea. This also happened to me with dp1 on the first/second attempt, then gone by itself...

anyway appear to be USB 2 related

Share this post


Link to post
Share on other sites

It's not related to the boot loader but the USB kext (possibly modified? DSDT EHCx patch required?). Best to move the posts in the relevant 10.11 section...

Share this post


Link to post
Share on other sites

That's the"annoying" thing! I've never had any errors (kernel panics) like this before. And I've run successfully OS X from 10.6 to 10.10 (and I might add 10.11 DB1-DB3) on this Hac (PC) (just one hardware change, new graphics card; old was 9800 GT). (It's as close to a Mac Pro 4,1 as it can be.)

 

Anyway, thanks for taking the time to read my post. :)


It's not related to the boot loader but the USB kext (possibly modified? DSDT EHCx patch required?). Best to move the posts in the relevant 10.11 section...

Didn't modify any kext. All are stock. Unless Apple modified/changed something from DB3 to DB4 (which probably happened). Tried with/without dsdt in Extra. Same thing. dsdt was created in Yosemite with MaciASL (no errors/warnings).

 

Anyway, if you say is not a boot loader thing, I'll look in other more appropriate threads. Thanks.

Share this post


Link to post
Share on other sites

Hey guys,

 

Can someone help me with this kernel panic? Running 10.11 DB5. Tried the last 3 Enoch's revisions but no luck. I've upgraded from 10.10.4 with success. Got the DB2 update, worked fine, then the DB3, same thing, then after DB4 kernel panic. Since I had the boot loader on usb stick, I've plugged the hd to my iMac (27-late 2013). Removed the kexts for my Hac (from S/L/E), which basically where just 3, FakeSMC, VoodooHDA and AppleIntelE1000e (all the most recent versions), and booted from this hd on iMac. Updated to DB5 (I thought maybe will fix the error this way), installed the kexts again, used Kext Utility 2.6.4 to rebuild cache and hooked the hd back to my Hac (PC). Same kernel panic like in DB4. My specs - signature.

 

Thanks to anyone who could help, point me in the right direction, to fix it.

 

You can remove /System/Library/Extensions/IOUSBHostFamily.kext/Contents/PlugIns/AppleUSBUHCIPCI.kext, but usb may not work.

Or, fix it follow this.. http://www.insanelymac.com/forum/topic/306777-guide-usb-fix-el-capitan-1011/

Share this post


Link to post
Share on other sites

Looks like for El Capitan DP4/DP5 and PB3 Enoch not work anymore....

 

Until now the CSR (Code Signing Restrictions) was set to 0x01 (00000001) (Thx Pike)

so only untrusted kext was allowed...

 

Pls try this UNTESTED versions:

Let me know...

 

ErmaC

 

 

Hi ErmaC,

 

Do they set BooterConfig=0x28 too, like Clover?

 

crazybirdy

Share this post


Link to post
Share on other sites

  • Recently Browsing   0 members

    No registered users viewing this page.

  • Similar Content

    • By Rohan20
      Opencore is running really really slow on Catalina 10.15.4. So It was running perfectly before I put the ssdt-pnlf for enabling Brightness control. it is running really slow and I verified my config with the sanity checker and its all correct. can someone please help. I am attaching my oc folder.

       
      https://www.dropbox.com/s/drkek0eaz19ina5/OC.zip?dl=0
       
    • By StarFighter_77
      I built this hack years ago now. Started with Lion, now on El Cap. I want to update now to High Sierra.

      I think I messed up my Clover when I downloaded the update .pkg and just installed it without taking my time or customizing the install. Was on Clover r3423, now I think I'm on r5104

      I've been searching and reading this site and others for a few days before making this thread, but I need help compiling what I've learned to get my system running correctly again and ready to update to High Sierra. Then, eventually, I want to update my graphics card, switch to UEFI (if necessary) and update to Mojave.

      I couldn't get system to boot. Could get to Clover and all the options, except the Shell... I hit enter on the Shell and I get black screen and nothing.

      The only way I got it to boot was to add -v to boot args, and change smbios from 14,2 to 13,2. Not sure why that worked still. Also I'm running two monitors, but I can only now run one of them off the Intel 3000 Graphics and the second one off my GFX Card.

      So now I'm afraid to shut down or reboot.... I could really use some help with Clover.

      I'm pretty sure my config file in Clover is not right. Also, some of the files in my Clover backups are not in the new Clover install. (2.5k 5104)

      If someone could please take a look at the files I've attached and help me sort this out, I would be grateful. Thank you, Thank you, Thank you.

      I've attached
      - The last boot log
      - The current Clover folder from the EFI partition after performing Clover update. (.zip)
      - The Clover backup that was working before Clover update (.zip)
      - Library/Extensions Folder (screen shot)
      - System/Library/Extensions Folder (.zip)

      Here is my current hardware:

      Mobo: GIGABYTE GA-Z68MX-UD2H-B3

      CPU: Intel Core i7-2600K

      GPU: GIGABYTE ATI Radeon HD6870 1GB DDR5 2DVI/HDMI/2x

      RAM: Corsair Vengeance Blue 16 GB DDR3 SDRAM Dual Channel Memory Kit CMZ16GX3M4A1600?C9B

      HD0 (for OS and Apps): OCZ Technology 120 GB Vertex Series SATA II Solid State Drive

      HD1 (file storage, Adobe scratch disk): Samsung 1 TB Spinpoint 7200 RPM 32MB Cache SATA

      BURNER: LG CD/DVD Burner Model GH22NS50

      CPU COOLER: Corsair Cooling Hydro Series H50 All in One High-performanc?e CPU Cooler CWCH50-1

      P/S: Antec CP-850 850 Watt CPX Power Supply Unit

      CASE: Antec Performance One P183 V3 Case
      CLOVER.zip
      CloverBackup.zip
      bootlog.txt

      S_L_E.zip
    • By dgsga
      Can I propose a new subforum be created for the new OpenCorePkg OpenCore front end being created by vit9696 and others, it is a fantastic piece of work:
      https://github.com/acidanthera/OpenCorePkg
      Even at version 0.1 it runs my Mojave 10.14.4 setup very nearly flawlessly. It consists of a 10KB bootstrap BootX64.efi and a 200KB OpenCore.efi OS loader. All configuration is done using a very well documented config.plist 
       
       
×