Jump to content

Clover General discussion

ErmaC

By ErmaC,
in Clover

 Share
30,171 posts in this topic

Recommended Posts

mrjayviper
mrjayviper
Posted
Posted

No matter, legacy BIOS can start from MBR and from GPT.

So I booted into my SSD drive via the USB installer. I am in the process of installing Clover but unsure which Bootloader option to choose.

 

I suppose:

 

1. I need to deselect "Install for UEFI booting" since I'm using legacy BIOS

2. probably deselect "Install Clover in the ESP"? I read in wikipedia that ESP is only present in UEFI systems?

3. as for the Bootloader options, the only suitable option seems to be "Do not update MBR and PBR sectors" since my SSD drive uses GPT

 

Thanks again for the help

Link to comment
Share on other sites
orenmac
orenmac
Posted
Posted

1. I need to deselect "Install for UEFI booting" since I'm using legacy BIOS

2. probably deselect "Install Clover in the ESP"? I read in wikipedia that ESP is only present in UEFI systems?

3. as for the Bootloader options, the only suitable option seems to be "Do not update MBR and PBR sectors" since my SSD drive uses GPT

1. Correct

2. ESP is hidden EFI partition on your hd. For start deselect this option (You can move your Clover files to EFI lather).

3. Correct

  • Like 1
Link to comment
Share on other sites
mrjayviper
mrjayviper
Posted
Posted

1. Correct

2. ESP is hidden EFI partition on your hd. For start deselect this option (You can move your Clover files to EFI lather).

3. Correct

are you saying I need to select the ESP option and deselect the other 2?

 

I'm asking because I did just that and I cannot boot without the USB stick.

 

It says on my screen:

------------

Verifying DMI Pool Data..........................

 

boot0: GPT

boot0: GPT

boot0: error

-----------

Link to comment
Share on other sites
Slice
Slice
Posted
Posted

No, the logic reversed.

Legacy BIOS means starting legacy Clover, not UEFI boot.

1. Correct.

2. Select  "Install Clover in the ESP" because you have GPT partitioning drive.

3. Deselect "Do not update MBR and PBR sectors" as you have to install boot sectors for legacy boot.

Yes, your system is legacy, not UEFI. But Clover creates own UEFI implementation after legacy boot and then your system will work from GPT drive and the Clover can boot OSX and Windows UEFI.

Sic! Windows 8-64bit UEFI can be started only from GPT drive. It requires UEFI environment. Legacy Clover do this!

  • Like 1
Link to comment
Share on other sites
RehabMan
RehabMan
Posted
Posted

another question please. So I got my setup working. I have a patched AppleHDA.kext (actual filename) which works fine if I placed it in SLE. But it wouldn't work if I placed it in kext/10.11.

 

Thanks again

Patched AppleHDA.kext and "dummyHDA" kexts must be installed to the system volume.

  • Like 2
Link to comment
Share on other sites
mrjayviper
mrjayviper
Posted
Posted

thanks for the helpful replies.

 

another question in relation to the responses given. Is it better to just place all KEXTs inside SLE then (assuming one is using a patched AppleHDA). instead of placing them in 2 places?

 

some KEXTs goes to kext/10.11 and HDA related KEXTs goes to SLE.

Link to comment
Share on other sites
mrjayviper
mrjayviper
Posted
Posted

the use of either clover folder or /s/l/e is a users pref. but not all kexts can be put in clover folder

I understand it's mostly user preference but if you're going to place a patched AppleHDA.kext in SLE anyway, seems logical to place everything in the same location.

 

but again, that comes down to user preference.

Link to comment
Share on other sites
liujianwei
liujianwei
Posted
Posted

coz it has no dependencies

I think Clover has some modify to make FakeSMC.kext work without install in SLE.Because I'm using yukon 88e8053 so I don't care about LAN kext and other kext that need install in Clover folder but I think Clover could load any kext just need some work so that it keep SLE vanilla.

Link to comment
Share on other sites
LockDown
LockDown
Posted
Posted (edited)

Common would be, fakesmc & lan kexts in clover folder, patched applehda or dummyhda in /s/l/e

I understand it's mostly user preference but if you're going to place a patched AppleHDA.kext in SLE anyway, seems logical to place everything in the same location.

 

but again, that comes down to user preference.

Try AppleALC that works in clover folder Edited by ellaosx
  • Like 2
Link to comment
Share on other sites
tluck
tluck
Posted
Posted

@tetonne - did you remove the logical volume? if you have the logical volume that the installer likes to put in place, it usually leads to confusing dual icons.

 

run:

 

diskutil cs revert / 

Link to comment
Share on other sites
Micky1979
Micky1979
Posted
Posted

Why Fakesmc.kext don't need to install the system volume?

mostly because AppleHDA binary use this function:

setProperty(kIOAudioDeviceLocalizedBundleKey, "AppleHDA.kext").

to find xml.zlib files inside AppleHDA.kext/Contents/Resources, but this path contains only vanilla files. I've tried to patch the binary and pass to it another kext name using Clover patches with no luck. Anyway we can't pass aDummyHDA.kext as argument to that funcion if the kext is inside an umounted EFI partition, also because Clover does not seem to use a unix path separator ('\' instead of '/') for the kexts allocated in the memory-map, so anything made would be useless.

FakeSMC instead has no external bundle to deal with, so simply works because everything needs is already in memory!

  • Like 3
Link to comment
Share on other sites
RehabMan
RehabMan
Posted
Posted

Why Fakesmc.kext don't need to install the system volume?

AppleHDA is a complex kext. It loads various configuration files dynamically from Contents/Resources via OSKextRequestResource. OSKextRequestResource does not work when the kext bundle is injected by Clover. It is for the same reason that BrcmFirmwareRepo.kext does not work from CLOVER/kexts.

 

There are other kexts that don't work with injection. For example, any kext that can be unloaded by the system will not be able to reload (the system has the IOKitPersonality in the IOCatalog, but it can't find the kext to load it). The way the original BrcmPatchRAM2.kext was written it would load, do its thing (push firmware), then unload. It worked on fresh boot and it even worked after sleep if sleep was initiated "soon enough" after fresh boot. But it didn't work if enough time passed to allow the kernel to unload the kext (the kernel employs "lazy unloading"). It is for this reason, that I added some code to BrcmPatchRAM2 to keep it resident even if it was "done", which then allowed it to work from Clover/kexts.

  • Like 8
Link to comment
Share on other sites
liujianwei
liujianwei
Posted
Posted

Common would be, fakesmc & lan kexts in clover folder, patched applehda or dummyhda in /s/l/e

Try AppleALC that works in clover folder

 

mostly because AppleHDA binary use this function:

setProperty(kIOAudioDeviceLocalizedBundleKey, "AppleHDA.kext").

to find xml.zlib files inside AppleHDA.kext/Contents/Resources, but this path contains only vanilla files. I've tried to patch the binary and pass to it another kext name using Clover patches with no luck. Anyway we can't pass aDummyHDA.kext as argument to that funcion if the kext is inside an umounted EFI partition, also because Clover does not seem to use a unix path separator ('\' instead of '/') for the kexts allocated in the memory-map, so anything made would be useless.

FakeSMC instead has no external bundle to deal with, so simply works because everything needs is already in memory!

 

AppleHDA is a complex kext. It loads various configuration files dynamically from Contents/Resources via OSKextRequestResource. OSKextRequestResource does not work when the kext bundle is injected by Clover. It is for the same reason that BrcmFirmwareRepo.kext does not work from CLOVER/kexts.

 

There are other kexts that don't work with injection. For example, any kext that can be unloaded by the system will not be able to reload (the system has the IOKitPersonality in the IOCatalog, but it can't find the kext to load it). The way the original BrcmPatchRAM2.kext was written it would load, do its thing (push firmware), then unload. It worked on fresh boot and it even worked after sleep if sleep was initiated "soon enough" after fresh boot. But it didn't work if enough time passed to allow the kernel to unload the kext (the kernel employs "lazy unloading"). It is for this reason, that I added some code to BrcmPatchRAM2 to keep it resident even if it was "done", which then allowed it to work from Clover/kexts.

Thank you very much for your help,we can circumvent AppleHDA by use AppleALC which ellaosx mentioned.But I don't find a good way to list it.IORegistryExplorer IOResources can show it,but kextstat in Terminal doesn't. 

It's Clover and DSDT problem.Clover auto load USB drive EFI folder ignore my SSD EFI folder so Clover doesn't inject audio id 1.And there is no working DSDT audio id inject patch on internet.Now my Realtek ALC885 works as build in instead of my USB C-Media Audio Card with AppleALC.I don't set first boot device to usb.I think because my computer is legacy bios based so boot file like to boot usb clover EFI folder.

What's bdmesg,where does it come from?Where is the log that bdmesg read from?

Link to comment
Share on other sites
frankiee
frankiee
Posted
Posted

Hi guys,

 

I really could need some help with a very strange problem that is a total mystery for me. My machine ran OK for about 2 years without any problems, but - around the time I upgraded to 10.11 four months ago - I get a lot of KPs. These are almost always "type 14=page fault" errors, and the offending process is "launchd" most of the time and they only happen when:

  • I am running 10.11 (also with a fresh installation, never had this with 10.10)
  • I do sleep before (this is absolutely needed to reproduce the problem)
  • Mostly on a subsequent restart / shutdown, but sometimes also "out of nowhere"
  • When I test my RAM with rember after waking
  • I do have all  8 DIMMs installed, i.e. running with 64GB of RAM

I have / had no problems with:

  • Memtest, running for about 20 hours
  • Running Rember in OS X when I do not sleep before
  • Windows
  • OS X 10.10
  • Edit: the problem also goes away when removing 4 DIMMs, does not matter which ones (swapped back and forth)

Since I also did a lot of Clover updates in the meantime, and some things do not work anymore as they did before (safe boot, booting with AptioFixDrv - the original, not the "2" version) I cannot even rule out that this might be related to Clover. And I did try a LOT of things to rule out other factors, for example disabling USB2, disabling USB3, removing PCI cards, and I even swapped my mobo!

 

So, what could it be? Any help greatly appreciated!

  • Like 1
Link to comment
Share on other sites
liujianwei
liujianwei
Posted
Posted

Hi guys,

 

I really could need some help with a very strange problem that is a total mystery for me. My machine ran OK for about 2 years without any problems, but - around the time I upgraded to 10.11 four months ago - I get a lot of KPs. These are almost always "type 14=page fault" errors, and the offending process is "launchd" most of the time and they only happen when:

  • I am running 10.11 (also with a fresh installation, never had this with 10.10)
  • I do sleep before (this is absolutely needed to reproduce the problem)
  • Mostly on a subsequent restart / shutdown, but sometimes also "out of nowhere"
  • When I test my RAM with rember after waking
  • I do have all  8 DIMMs installed, i.e. running with 64GB of RAM

I have / had no problems with:

  • Memtest, running for about 20 hours
  • Running Rember in OS X when I do not sleep before
  • Windows
  • OS X 10.10

Since I also did a lot of Clover updates in the meantime, and some things do not work anymore as they did before (safe boot, booting with AptioFixDrv - the original, not the "2" version) I cannot even rule out that this might be related to Clover. And I did try a LOT of things to rule out other factors, for example disabling USB2, disabling USB3, removing PCI cards, and I even swapped my mobo!

 

So, what could it be? Any help greatly appreciated!

I also get unstable when I'm using 10.11,doesn't have problem when I'm using 10.9.I have turn off EIST to see if the problem solved.I will keep report.

Link to comment
Share on other sites
frankiee
frankiee
Posted
Posted

I also get unstable when I'm using 10.11,doesn't have problem when I'm using 10.9.I have turn off EIST to see if the problem solved.I will keep report.

 

Do you have similar symptoms like me? I.e. "type 14=page fault" KPs after sleep?

Link to comment
Share on other sites
 Share
Go to topic listing
×
×
  • Create New...