magnifico Posted August 7, 2014 Share Posted August 7, 2014 I have OsxAptioFix. Look why have you two osxaptio? not normal..delete Ops Link to comment Share on other sites More sharing options...
Riley Freeman Posted August 7, 2014 Share Posted August 7, 2014 EDIT: does loading an existing and probably older IONetworkingFamily kext actually works? 5:512 0:009 Extra kext: EFI\CLOVER\kexts\10.10\IONetworkingFamily.kext I had issues with my ethernet disappearing when I tried loading the ethernet kext from the EFI partition. I think it mostly would happen after installing software that rebuilt the kext cache. I thought it was related to not having ionetworkingfamily with it but it didn't seem to matter. In the end I just put the ethernet kext back into /S/L/E for safety. Link to comment Share on other sites More sharing options...
crusher Posted August 7, 2014 Share Posted August 7, 2014 You should read the installer's description of OsxAptioFixDrv and OsxLowMemoryFixDvr, it is suggested to use only one of those drivers. EDIT: does loading an existing and probably older IONetworkingFamily kext actually works? 5:512 0:009 Extra kext: EFI\CLOVER\kexts\10.10\IONetworkingFamily.kext Look only work and I manually go to sleep mode and after wake don't work USB 3.0 and after 2-5min System restart. Link to comment Share on other sites More sharing options...
TimeWalker75a Posted August 7, 2014 Share Posted August 7, 2014 I'm not sure... It needs some stuff to be in nvram though...? No, why would it? If you are installing on a brand new machine that you have put together there's no data stored in NVRAM. And as I learned today having some garbage data in there may result in installation failing for no apparent reason. Link to comment Share on other sites More sharing options...
JC. Posted August 7, 2014 Share Posted August 7, 2014 Just completely remove it and then reinstall and it will give you the default. How do I completely remove Clover? Is "rm -rf" in the EFI volume sufficient? Or is something more required? Because I deleted just the config.plist file and reinstalled, and no new config.plist file was created. Link to comment Share on other sites More sharing options...
jamiethemorris Posted August 7, 2014 Share Posted August 7, 2014 Hmmmmm, I'm not sure about this at all. I'll PM you a link since I don't want to post it here... I don't believe thats the identifier of the kext. You might also be using the wrong notation or something. Thanks. I did some googling but it doesn't look like anyone has really attempted what I'm trying to do which I find odd. I thought Clover used the name of the binary rather than the identifier? @Funky frank you might want to try AGPM instead of AppleGraphicsPowerManagement. I believe that is the name of the identifier. It should also be possible to create an info.plist only kext to inject the right information, unfortunately I have tried that haven't been able to get it to work. The information I was using to do this is here: http://www.projectosx.com/forum/index.php?showtopic=798 Also, remember that you have to specifically tell Clover that this is an info.plist patch in case you haven't done that. <key>InfoPlistPatch</key> <true/> It also requires 2 reboots to be injected into the kernel cache: Side note: The patch is supposed to be done in the cache. If you patch an Info.plist to allow the loading of a kext, then it is not yet present in the cache. You will need to reboot twice. First with the key NoCaches to allow FSInject to load the kext and a second time for the cache to be patched successfully. If that still doesn't work it may be helpful to also enable debugging for KernelAndKextPatches to see where it's going wrong: <key>Debug</key> <true/> Link to comment Share on other sites More sharing options...
>Federico< Posted August 7, 2014 Share Posted August 7, 2014 Then use the stuff that the iso is using..... Like whatever drivers and the config.plist. I've tried without success, I think, it's my personal think, that the problem is the installation in the ESP, but if I deselect it I get boot1: error (tried all possible method) Link to comment Share on other sites More sharing options...
Funky frank Posted August 8, 2014 Share Posted August 8, 2014 I don't believe thats the identifier of the kext. You might also be using the wrong notation or something. I think there is some bug in Clover's plist find/replace algorithm...? I tried identifiers: "AppleGraphicsPowerManagement" and "AGPM" This is the config.plist entry: <key>KextsToPatch</key> <array> <dict> <key>Comment</key> <string>Replace IGPU to GFX0</string> <key>Find</key> <string>IGPU</string> <key>InfoPlistPatch</key> <true/> <key>Name</key> <string>AGPM</string> <key>Replace</key> <string>GFX0</string> </dict> </array> It does not replace all IGPU to GFX0. So I changed it manually. BTW. if I use a "&" symbol (resulted from writing "IGPU -> GFX0" as name in clover configurator) in key "Name", there will be an error in the log: Aug 8 03:08:58 localhost kernel[0]: "name" not a kext Aug 8 03:08:58 localhost kernel[0]: Error unserializing info dictionary for device tree entry Driver-12414000: OSUnserializeXML: duplicate dictionary key near line 428 Aug 8 03:08:58 localhost kernel[0]: . Link to comment Share on other sites More sharing options...
Maniac10 Posted August 8, 2014 Share Posted August 8, 2014 @Funky Frank: shouldn't the patch's name be "AppleGraphicsPowerManagement" instead of just "AGPM"? Hence the: Aug 8 03:08:58 localhost kernel[0]: "name" not a kext Link to comment Share on other sites More sharing options...
toleda Posted August 8, 2014 Share Posted August 8, 2014 It does not replace all IGPU to GFX0. So I changed it manually. The correct dsdt patch. AGPM natively looks for IGPU, no edit required. SysDef model must be the same processor generation as your installed processor. <dict> <key>Comment</key> <string>Rename GFX0 to IGPU</string> <key>Find</key> <data> R0ZYMA== </data> <key>Replace</key> <data> SUdQVQ== </data> </dict> 1 Link to comment Share on other sites More sharing options...
jamiethemorris Posted August 8, 2014 Share Posted August 8, 2014 The correct dsdt patch. AGPM natively looks for IGPU, no edit required. SysDef model must be the same processor generation as your installed processor. <dict> <key>Comment</key> <string>Rename GFX0 to IGPU</string> <key>Find</key> <data> R0ZYMA== </data> <key>Replace</key> <data> SUdQVQ== </data> </dict> That's probably the best way to do it. Make sure you revert your changes to the kext otherwise that won't work. Link to comment Share on other sites More sharing options...
naujoks Posted August 8, 2014 Share Posted August 8, 2014 Hi guys, as you might remember, my problem is a black screen just when the GUI is supposed to start (GTX 780, Sabertooth P67, 2600k). I've attached the debug.log. As you can see, the reason for the halt is "Custom boot screen not used because entry has unset use graphics" and "Failed to initialize boot screen: Aborted!" Would anyone know how I can fix this? debug.rtf Link to comment Share on other sites More sharing options...
naujoks Posted August 8, 2014 Share Posted August 8, 2014 Thats not any error, that's just that you don't have Boot/CustomLogo set so there is no custom boot screen.... That's not your problem, it's something else. Puzzling though that things should stop right there... Any idea what actually might be wrong? Link to comment Share on other sites More sharing options...
TimeWalker75a Posted August 8, 2014 Share Posted August 8, 2014 Because the second stage of the installer needs this stuff, without it you can't complete install. How do you propose differentiating these two? I highly doubt that it needs any kind of info at second stage. Image a situation: you get a compatible CPU and a board from a friend/dumpster and decide you wan't to test drive OSX on it. The board is BIOS-based and you are using CloverEFI to install OSX. At what point in time would it be possible for Clover to store or read ay kind of nvram variables on this machine? Yet the install would continue flawlessly in both stages, provided the configuration in config.plist was set properly in order to allow this setup to boot. Link to comment Share on other sites More sharing options...
naujoks Posted August 8, 2014 Share Posted August 8, 2014 That's just the last thing that happens before boot.efi takes over, it's not strange at all. The custom boot screen gets drawn right before hand off. You probably need to drop BGRT table, use OsxAptioFix (or OsxLowMemFix) if UEFI booting, and try without CsmVideo and vbios patch if you have them I'm already using OsxAptioFix and don't have CsmVideo or vbios patched. I'm afraid I haven't heard about BGRT tables before, so don't know what to do about that... Link to comment Share on other sites More sharing options...
TimeWalker75a Posted August 8, 2014 Share Posted August 8, 2014 The first stage saves nvram variables about the installation, without working nvram then the installation cant continue and you get this error about another installation didn't finish, probably the error that you received in the first place. Perhaps if you run the .app from within OS, yes.. But how are you supposed to perform a clean install using CloverEFI or UEFI with locked NVRAM access if nothing gets written to NVRAM (there are no scripts to run and save it)? Yet you somehow do, because when running the installer on a machine that has never ran OSX before you are greeted with the Recover-like UI first and only then the second stage invokes after a reboot, because Clover is able to locate the boot.efi on the temp store where first stage has extracted the setup contents. Link to comment Share on other sites More sharing options...
TimeWalker75a Posted August 8, 2014 Share Posted August 8, 2014 I don't understand how the contents of nvram matter for the second stage if CloverEFI wouldn't be able to save them at any point during or after the first stage... hence the logical conclusion that none of the variables are needed to proceed to the second stage. If however you have an nvram.plist on any of the disks/partition Clover picks it up and messes up the first stage of the installation. At least this is my experience. 1 Link to comment Share on other sites More sharing options...
Neso Posted August 8, 2014 Share Posted August 8, 2014 Clover rev. 2806. Custom kext patching fixed, debug=true started working, i can see kext patching procedure in output. Everything back to normal. Thanks apianti. Link to comment Share on other sites More sharing options...
Maniac10 Posted August 9, 2014 Share Posted August 9, 2014 (edited) I just tested moving the "KernelAndKextPatches" section into my mavericks custom entry and the patches don't work. 4:131 0:000 KextsToPatch: 11 requested 4:131 0:000 KextToPatch 0: AppleUSBXHCI (Enable PCI power management) Kext bin patch, data len: 9 4:131 0:000 KextToPatch 1: AppleUSBXHCI (Disable XHCI 1.0 10.9) Kext bin patch, data len: 4 4:131 0:000 KextToPatch 2: AppleAHCIPort (Fix Orange Disk) Kext bin patch, data len: 8 4:131 0:000 KextToPatch 3: AppleHDA (ALC889 Codec) Kext bin patch, data len: 4 4:131 0:000 KextToPatch 4: AppleUSBXHCI (Disable sleep code) Kext bin patch, data len: 7 4:131 0:000 KextToPatch 5: AirPortAtheros40 (Wi-Fi Region to 0x64) Kext bin patch, data len: 7 4:131 0:000 KextToPatch 6: AppleUSBXHCI (Patch MSI or pin interrupts 2) Kext bin patch, data len: 8 4:131 0:000 KextToPatch 7: AppleUSBXHCI (Disable Intel & Fresco USB3.0 10.9) Kext bin patch, data len: 9 4:131 0:000 KextToPatch 8: IOAHCIBlockStorage (TRIM Enabler) Kext bin patch, data len: 9 4:131 0:000 KextToPatch 9: AppleUSBXHCI (Patch MSI or pin interrupts 1) Kext bin patch, data len: 7 4:131 0:000 KextToPatch 10: AppleHDA (AppleHDA xml>zml) Kext bin patch, data len: 6 4:132 0:000 Custom entry boot CUSTOM_BOOT_DISABLED (0x0) 4:132 0:000 Custom entry boot CUSTOM_BOOT_DISABLED (0x0) 28:822 0:000 Kernel and Kext Patches: 28:822 0:000 Allowed: n 28:822 0:000 Debug: n 28:822 0:000 KernelCpu: n 28:822 0:000 Lapic: n 28:822 0:000 AICPUPM: n 28:822 0:000 AppleRTC: y 28:822 0:000 KernelPm: n 28:822 0:000 FakeCPUID: 0x0 28:822 0:000 ATIController: <null string> 28:822 0:000 ATIDataLength: 0 28:822 0:000 KextPatch[0]: 9 bytes, ApeSXCpa,8 28:822 0:000 KextPatch[1]: 4 bytes, ApeSXCpa(Ó8 28:822 0:000 KextPatch[2]: 8 bytes, ApeHIotHpa 28:822 0:000 KextPatch[3]: 4 bytes, ApeD 28:822 0:000 KextPatch[4]: 7 bytes, ApeSXC 28:822 0:000 KextPatch[5]: 7 bytes, Arotteo4 28:822 0:000 KextPatch[6]: 8 bytes, ApeSXC 28:822 0:000 KextPatch[7]: 9 bytes, ApeSXC 28:822 0:000 KextPatch[8]: 9 bytes, IACBoktrg 28:822 0:000 KextPatch[9]: 7 bytes, ApeSXC 28:822 0:000 KextPatch[10]: 6 bytes, ApeD EDIT: restored the KernelAndKextPatches section to its original place and it's still not working for me in rev 2807. 4:077 0:000 KextsToPatch: 14 requested 4:077 0:000 KextToPatch 0: AppleUSBXHCI (Enable PCI power management) Kext bin patch, data len: 9 4:077 0:000 KextToPatch 1: AppleUSBXHCI (Disable XHCI 1.0 10.9) Kext bin patch, data len: 4 4:077 0:000 KextToPatch 2: IOATAFamily (IOATAFamily Faster Boot) Kext bin patch, data len: 4 4:077 0:000 KextToPatch 3: AppleAHCIPort (Fix Orange Disk) Kext bin patch, data len: 8 4:077 0:000 KextToPatch 4: AppleUSBXHCI (Disable XHCI 1.0 10.10) Kext bin patch, data len: 11 4:077 0:000 KextToPatch 5: AppleHDA (ALC889 Codec) Kext bin patch, data len: 4 4:077 0:000 KextToPatch 6: AppleUSBXHCI (Disable sleep code) Kext bin patch, data len: 7 4:077 0:000 KextToPatch 7: AppleUSBXHCI (Disable Intel & Fresco USB3.0 10.10) Kext bin patch, data len: 9 4:077 0:000 KextToPatch 8: AirPortAtheros40 (Wi-Fi Region to 0x64) Kext bin patch, data len: 7 4:077 0:000 KextToPatch 9: AppleUSBXHCI (Patch MSI or pin interrupts 2) Kext bin patch, data len: 8 4:077 0:000 KextToPatch 10: AppleUSBXHCI (Disable Intel & Fresco USB3.0 10.9) Kext bin patch, data len: 9 4:077 0:000 KextToPatch 11: IOAHCIBlockStorage (TRIM Enabler) Kext bin patch, data len: 9 4:077 0:000 KextToPatch 12: AppleUSBXHCI (Patch MSI or pin interrupts 1) Kext bin patch, data len: 7 4:077 0:000 KextToPatch 13: AppleHDA (AppleHDA xml>zml) Kext bin patch, data len: 6 4:077 0:000 Custom entry boot CUSTOM_BOOT_DISABLED (0x0) 4:077 0:000 Custom entry boot CUSTOM_BOOT_DISABLED (0x0) 11:757 0:000 Kernel and Kext Patches: 11:757 0:000 Allowed: n 11:757 0:000 Debug: n 11:757 0:000 KernelCpu: n 11:757 0:000 Lapic: n 11:757 0:000 AICPUPM: n 11:757 0:000 AppleRTC: y 11:757 0:000 KernelPm: n 11:757 0:000 FakeCPUID: 0x0 11:757 0:000 ATIController: <null string> 11:757 0:000 ATIDataLength: 0 11:757 0:000 KextPatch[0]: 9 bytes, ApeSXC 11:757 0:000 KextPatch[1]: 4 bytes, ApeSXC 11:757 0:000 KextPatch[2]: 4 bytes, IAAaiy 11:757 0:000 KextPatch[3]: 8 bytes, ApeHIot 11:757 0:000 KextPatch[4]: 11 bytes, ApeSXC 11:757 0:000 KextPatch[5]: 4 bytes, ApeD¼ 11:757 0:000 KextPatch[6]: 7 bytes, ApeSXC 11:757 0:000 KextPatch[7]: 9 bytes, ApeSXCDpa$ 11:757 0:000 KextPatch[8]: 7 bytes, Arotteo4âpa$ 11:757 0:000 KextPatch[9]: 8 bytes, ApeSXC 11:757 0:000 KextPatch[10]: 9 bytes, ApeSXC 11:757 0:000 KextPatch[11]: 9 bytes, IACBoktrg 11:757 0:000 KextPatch[12]: 7 bytes, ApeSXC 11:757 0:000 KextPatch[13]: 6 bytes, ApeD Edited August 9, 2014 by Maniac10 Link to comment Share on other sites More sharing options...
TheBoatman Posted August 9, 2014 Share Posted August 9, 2014 No patching does not work as expected in Clover >=2797, I would stick with Slice's SourceForge for a stable Clover until he returns from his holiday......or Apianti fixes the bugs that have been introduced. 1 Link to comment Share on other sites More sharing options...
chris1111 Posted August 9, 2014 Share Posted August 9, 2014 yes i'm still working on it, i know it doesn't work, not sure why Neso reported it worked... it may have for him though. Dont put to mutch pressure on you steps back for A time and coming back on force Link to comment Share on other sites More sharing options...
HackBook Pro Posted August 9, 2014 Share Posted August 9, 2014 Not all mice work. Sorry! So, there's no way for me to use my Razer mouse in Clover? Is there a way to fix it so that I don't need to unplug/re-plug it to get it to work in the OSs? Link to comment Share on other sites More sharing options...
Neso Posted August 9, 2014 Share Posted August 9, 2014 After the 2796...orange icons have come back and trim was not working. Debug = true in my config.plist didn't work, so couldn't help apianti with output. Waited for >2804, and when 2806 come i've installed. Forget that i had Debug = true left in my config.plist and on reboot had kernel and kext patching debug output on screen for the first time. When in desktop hard drive icons are back to metallic, with: <dict><key>Name</key><string>AppleAHCIPort</string><key>Find</key><data>RXh0ZXJuYWw=</data><key>Replace</key><data>SW50ZXJuYWw=</data></dict> so for me at least everything back to normal with 2806... Legacy boot, on asus rampage III gene. ESP installed clover. Link to comment Share on other sites More sharing options...
magnifico Posted August 9, 2014 Share Posted August 9, 2014 Oi Api..listen me please...after i use this command on yose sudo kextcache -prelinked-kernel dont can boot with him mhmm Link to comment Share on other sites More sharing options...
pico joe Posted August 9, 2014 Share Posted August 9, 2014 boot with clover or other bootloader from flashdrive or you can fix from recovery hd --> repair permission Link to comment Share on other sites More sharing options...
Recommended Posts