Clover General discussion

[Zulakis]

But sorry, documentation is very large. What should I read about the problem?

Clover uses the same boot0, boot1h, boot as Chameleon. If it is implemented in Chameleon why not in Clover?

The real difference is in boot1h 

Chameleon's

kBoot2Sectors		EQU		(448 * 1024 - 512) / kSectorBytes	; max size of 'boot' file in sectors

For Clover I change 448 -> 472

Is it multiboot violation?

You should probably read http://www.gnu.org/software/grub/manual/multiboot/multiboot.html#Specification.

 

'boot' should be multiboot-compliant. It is not important for boot1 to be multiboot compliant.

 

Do you know why chainloading boot1f32alt does work and chainloading boot1h does not work?

[Slice]

You should probably read http://www.gnu.org/software/grub/manual/multiboot/multiboot.html#Specification.

 

'boot' should be multiboot-compliant. It is not important for boot1 to be multiboot compliant.

 

Do you know why chainloading boot1f32alt does work and chainloading boot1h does not work?

No matter what 'boot' is it. When it runs no other jump is possible.

There is other possibility: boot GRUB from Chameleon menu or from Clover one. It has no relation to 'multiboot'.

Why you send me to read something? Can you say in two words what is wrong in Clover's sources?

 

As I said Clover's boot1h differs from Chameleon's one by one constant. So.... chainloading, multiboot etc.

[WaldMeister]

Consider me the exception, i'm not the only 1 with this issue. See             #1898            

Any suggestion is welcome. It boots fine with Chameleon, but i would not want to revert to it. 

I attached a screenshot of the drivers, without those, i can not boot OS X, well, if i remove some, the boot time will increase by 2 min or so.

 

I even started playing with my bios options, it's unlocked, but it did not make a difference. 

 

Thanks in advance.

Replaced the boot.efi file using Pandora, it is still a no go, i can not get past this screen. Removing the OSXaptiosomething, well it makes my system boot slow, tried this also, with no result. 

Still using UEFI boot, Clover rev 1914.

[Zulakis]

No matter what 'boot' is it. When it runs no other jump is possible.

There is other possibility: boot GRUB from Chameleon menu or from Clover one. It has no relation to 'multiboot'.

Why you send me to read something? Can you say in two words what is wrong in Clover's sources?

 

As I said Clover's boot1h differs from Chameleon's one by one constant. So.... chainloading, multiboot etc.

 

Multiboot is a specification of 'how' a OS-Image (in this case 'boot') should be loaded from a bootloader. It specifies:

1. The format of an OS image as seen by a boot loader.
2. The state of a machine when a boot loader starts an operating system.
3. The format of information passed by a boot loader to an operating system.

Do you understand?

 

I don't know if chainloading boot1h from chameleon does work... If you say it only differs from that one constant, then it probably won't aswell.

In chameleon it's not important as atleast 'boot' is multiboot compliant.. In this case, chainloading boot1 is necessary because boot can't be chainloaded directly.

 

Still, chainloading boot1h should probably get fixed. Making boot multiboot compliant would be even better, but boot1h should be enough I think. At all the necessity to use boot0 should definetely be removed IMO so you can choose the partition you boot from yourself. Also, it would make loading clover from logical partitions possible.

 

Edit:

Sorry if it isn't totally clear, I am not a native english speaker. Just ask and I will clarify for you. Thanks!

[ameris_cyning]

I am having an issue with Haswell and Mavericks.

 

I prefer Clover EFI because I can use my bluetooth keyboard in the EFI setup screen (what is the proper name of EFI BIOS?).

 

However, Mavericks and Mountain Lion cannot boot with Clover. It will just show a black screen like it had an unsupported kernel as soon as it is done finishing with loading the kernel cache. This was a vanilla install. This does not happen in Chameleon, but it is the only solution to this issue.

 

My specs are in my sig.

 

Expanded upon.

[kidalive]

No matter what 'boot' is it. When it runs no other jump is possible.

There is other possibility: boot GRUB from Chameleon menu or from Clover one. It has no relation to 'multiboot'.

Why you send me to read something? Can you say in two words what is wrong in Clover's sources?

 

As I said Clover's boot1h differs from Chameleon's one by one constant. So.... chainloading, multiboot etc.

NTFS driver 3.10 [Flags: R/W].
NTFS volume name Windows 8 Pro, version 3.1.
NTFS-fs error (device /dev/disk0s6, pid 93): ntfs_restart_page_header_is_valid(): $LogFile version 2.0 is not supported.  (This driver supports version 1.1 only.)
NTFS-fs error (device /dev/disk0s6, pid 93): ntfs_restart_page_header_is_valid(): $LogFile version 2.0 is not supported.  (This driver supports version 1.1 only.)
NTFS-fs error (device /dev/disk0s6, pid 93): ntfs_logfile_check(): Did not find any restart pages in $LogFile and it was not empty.
NTFS-fs warning (device /dev/disk0s6, pid 93): ntfs_system_inodes_get(): $LogFile is not clean.  Will not be able to remount read-write.  Mount in Windows.
NTFS-fs warning (device /dev/disk0s6, pid 93): ntfs_system_inodes_get(): Windows is hibernated.  Will not be able to remount read-write.  Run chkdsk.
NTFS volume name Source, version 3.1.
NTFS-fs error (device /dev/disk0s7, pid 96): ntfs_restart_page_header_is_valid(): $LogFile version 2.0 is not supported.  (This driver supports version 1.1 only.)
NTFS-fs error (device /dev/disk0s7, pid 96): ntfs_restart_page_header_is_valid(): $LogFile version 2.0 is not supported.  (This driver supports version 1.1 only.)
NTFS-fs error (device /dev/disk0s7, pid 96): ntfs_logfile_check(): Did not find any restart pages in $LogFile and it was not empty.
NTFS-fs warning (device /dev/disk0s7, pid 96): ntfs_system_inodes_get(): $LogFile is not clean.  Will not be able to remount read-write.  Mount in Windows.

Whar`s wrong with this? I`m using four efi drivers below. What`s OsxFatBinaryDrv-64.efi used for? Booting FAT partition?

I`m using OsxAptioFixDrv.efi, OsxFatBinaryDrv.efi, FSInject.efi, VBoxHFS.efi. No HFSPlus.efi and NTFS.efi included in the latest Clover release. I checked the clover wiki link it says VBoxHFS.efi might be  slower than  HFSPlus.efi. Where to find this efi driver and the NTFIdriver?

 

 

[Andy Vandijck]

 

NTFS driver 3.10 [Flags: R/W].
NTFS volume name Windows 8 Pro, version 3.1.
NTFS-fs error (device /dev/disk0s6, pid 93): ntfs_restart_page_header_is_valid(): $LogFile version 2.0 is not supported.  (This driver supports version 1.1 only.)
NTFS-fs error (device /dev/disk0s6, pid 93): ntfs_restart_page_header_is_valid(): $LogFile version 2.0 is not supported.  (This driver supports version 1.1 only.)
NTFS-fs error (device /dev/disk0s6, pid 93): ntfs_logfile_check(): Did not find any restart pages in $LogFile and it was not empty.
NTFS-fs warning (device /dev/disk0s6, pid 93): ntfs_system_inodes_get(): $LogFile is not clean.  Will not be able to remount read-write.  Mount in Windows.
NTFS-fs warning (device /dev/disk0s6, pid 93): ntfs_system_inodes_get(): Windows is hibernated.  Will not be able to remount read-write.  Run chkdsk.
NTFS volume name Source, version 3.1.
NTFS-fs error (device /dev/disk0s7, pid 96): ntfs_restart_page_header_is_valid(): $LogFile version 2.0 is not supported.  (This driver supports version 1.1 only.)
NTFS-fs error (device /dev/disk0s7, pid 96): ntfs_restart_page_header_is_valid(): $LogFile version 2.0 is not supported.  (This driver supports version 1.1 only.)
NTFS-fs error (device /dev/disk0s7, pid 96): ntfs_logfile_check(): Did not find any restart pages in $LogFile and it was not empty.
NTFS-fs warning (device /dev/disk0s7, pid 96): ntfs_system_inodes_get(): $LogFile is not clean.  Will not be able to remount read-write.  Mount in Windows.

Whar`s wrong with this? I`m using four efi drivers below. What`s OsxFatBinaryDrv-64.efi used for? Booting FAT partition?

OS X FAT binary driver is for loading the Mach-O executables.

The problem above is an unclean NTFS partition.

Start Windows, do a disk check and reboot...

[Stanwell Park]

May be you should read again here             #32            ?

I studied the post again but strange is my Darwindump showing CurrentCloverBootedConfig.plist is totally different from my other two config files (which are correct). Something is overriding my own config.

 

Bootloader Config .plist files

/CurrentCloverBootedConfig.plist :

/disk0s1-EFI/EFI/CLOVER/config.plist : Clover

/disk0s1-EFI/EFI/CLOVER/OEM/P67 Professional/UEFI/config.plist : Clover

[Maniac10]

Are you sure "P67 Professional" is your board name? Check its name in the boot-log:

0:100  0:000  Clover revision: 1914  running on To be filled by O.E.M.
0:100  0:000  ... with board Z68AP-D3

[Slice]

 

I studied the post again but strange is my Darwindump showing CurrentCloverBootedConfig.plist is totally different from my other two config files (which are correct). Something is overriding my own config.

 

Bootloader Config .plist files

/CurrentCloverBootedConfig.plist :

/disk0s1-EFI/EFI/CLOVER/config.plist : Clover

/disk0s1-EFI/EFI/CLOVER/OEM/P67 Professional/UEFI/config.plist : Clover

 

Yes, I think the problem that you use old DarwinDumper with genconfig not compatible with current Clover version.

For good result you should use clover-genconfig same revision as Clover. Or just use latest DarwinDumper that can choose reliable genconfig.

[Stanwell Park]

Maniac10 wrote:

Are you sure "P67 Professional" is your board name? Check its name in the boot-log:

0:100 0:000 Clover revision: 1914 running on To be filled by O.E.M.0:100 0:000 ... with board Z68AP-D3

Yes. I'm sure my board name with board "P67 Professional" shown in the log. My board is asrock P67 fatal1ty professional

Yes, I think the problem that you use old DarwinDumper with genconfig not compatible with current Clover version.

For good result you should use clover-genconfig same revision as Clover. Or just use latest DarwinDumper that can choose reliable genconfig.

I assume the genconfig is getting current config and this isn't wrong because all my CPU status and another patches gone.

 

If I want to completely remove clover and reinstall again, what are the correct steps?

[Stanwell Park]

Yes, I think the problem that you use old DarwinDumper with genconfig not compatible with current Clover version.

For good result you should use clover-genconfig same revision as Clover. Or just use latest DarwinDumper that can choose reliable genconfig.

I will try to download latest version try again.

[Andy Vandijck]

I switched from legacy boot to full UEFI boot on an Asus P8H77-M LE.

I use Clover rev 1920.

Everything works except one thing: when I sleep the computer it goes to sleep just fine but wake just reboots the computer and it boots stating an error has occurred.

I've attached my bdmesg output.

Anybody have any idea's on how to fix sleep?

bdmesg.txt.zip

[artur_pt]

hello Andy

 

that happen only in uefi boot ?

 

in legacy boot the sleep works ?

 

must be efi drivers..

 

good hack

[Slice]

Andy,

This is well known problem. Sleep is good but wake as in table

            Lion           ML
CloverEFI   Good           Good
UEFI        Good           BAD

I have no solution for this yet.

Some tricks.

1. Use videocard with UEFI VBIOS (AMD Radeon HD7xxx)

2. Exclude CsmVideoDxe.efi

3. Use FakeSMC v3.3 

[Andy Vandijck]

Andy,

This is well known problem. Sleep is good but wake as in table

            Lion           ML
CloverEFI   Good           Good
UEFI        Good           BAD

I have no solution for this yet.

Some tricks.

1. Use videocard with UEFI VBIOS (AMD Radeon HD7xxx)

2. Exclude CsmVideoDxe.efi

3. Use FakeSMC v3.3

I have HD6450 Silent which is not UEFI GOP compliant so I can't disable csm.

Also tried the IGPU which is Intel HD2500 but it messes up the picture on the HDMI so I unfortunetely can't test that card (gonna get me a DVI to VGA adapter and test that port).

Then I could however exclude csm and disable csm in BIOS.

Maybe I indeed gotta get me a newer gpu.

[alexanderq]

Multiboot is a specification of 'how' a OS-Image (in this case 'boot') should be loaded from a bootloader. It specifies:

Do you understand?

 

I don't know if chainloading boot1h from chameleon does work... If you say it only differs from that one constant, then it probably won't aswell.

In chameleon it's not important as atleast 'boot' is multiboot compliant.. In this case, chainloading boot1 is necessary because boot can't be chainloaded directly.

 

Still, chainloading boot1h should probably get fixed. Making boot multiboot compliant would be even better, but boot1h should be enough I think. At all the necessity to use boot0 should definetely be removed IMO so you can choose the partition you boot from yourself. Also, it would make loading clover from logical partitions possible.

 

Edit:

Sorry if it isn't totally clear, I am not a native english speaker. Just ask and I will clarify for you. Thanks!

And maybe it will fix legacy booting problem because i noticed that to boot from win 7 installed on C: works only if i have another installation of windows on D: and have bootmgr on both partitons.And this is not allways working.

[Stanwell Park]

I will try to download latest version try again.

 

I came home and took a look at the config.plist. There was an extra "/" under one key. I think Clover could not recognise the config and loaded the default one. I fixed the config.plist and clover loaded both /EFI/CLOVER/config.plist and /EFI/CLOVER/OEM/P67 Professional/UEFI/config.plist without any issue.

 

My desktop now is 99% working and only one minor issue left which is sleep. It takes over two minutes to wake up screen after sleep.

[ameris_cyning]

Hey guys

 

I love Clover. I adore it. If I had my way it would displace Chameleon.

 

But I can't use it. Why? I don't know, and my posts inquiring into this issue in this very thread were ignored.

 

So lets try this all over again, shall we?

 

Clover EFI does not work for me.

 

It works up to the point when the kernel takes over. It turns into a black screen acting like it is an incompatible kernel. It doesn't matter what OS X I try to boot in, it is the same damn thing.

 

The install mechanism I chose was for UEFI installation and I am using the vanilla kernel. Rest of specs are in my signature

[Slice]

Your spec is good. But I can say nothing without some digits.

Can you provide DarwinDumper report, for example?

Also it is very desirable debug.log when booted by Clover.

	<key>GUI</key>
	<dict>
		<key>DebugLog</key>
		<true/>

[ameris_cyning]

DarwinDumper_2.7.8_Chameleon_2.2_Mav_notsure.zip

[kidalive]

OS X FAT binary driver is for loading the Mach-O executables.

The problem above is an unclean NTFS partition.

Start Windows, do a disk check and reboot...

IS IT OK TO DELTE THIS DRIVER? I`M BOOTING mac 10.8 AND wINDOWS 8.

[Slice]

Hey guys

 

I love Clover. I adore it. If I had my way it would displace Chameleon.

 

But I can't use it. Why? I don't know, and my posts inquiring into this issue in this very thread were ignored.

 

So lets try this all over again, shall we?

 

Clover EFI does not work for me.

 

It works up to the point when the kernel takes over. It turns into a black screen acting like it is an incompatible kernel. It doesn't matter what OS X I try to boot in, it is the same damn thing.

 

The install mechanism I chose was for UEFI installation and I am using the vanilla kernel. Rest of specs are in my signature

I looked into your dumps and have the follow question.

1. How did you try to boot? Legacy or UEFI? You said about CloveEFI but you chose "UEFI installation". They are different.

2. In your folder /driver64uefi there are redundant driver EmuVariableUefi-64.efi (delete it!!!) but absent OsxAptioFix. So why you have reboot before kernel start.

3. For Chameleon you use nvda_drv=1. Why not for Clover?

[ameris_cyning]

I looked into your dumps and have the follow question.

1. How did you try to boot? Legacy or UEFI? You said about CloveEFI but you chose "UEFI installation". They are different.

2. In your folder /driver64uefi there are redundant driver EmuVariableUefi-64.efi (delete it!!!) but absent OsxAptioFix. So why you have reboot before kernel start.

3. For Chameleon you use nvda_drv=1. Why not for Clover?

 

uefi and not legacy 

 

I installed it on my /EFI drive and set it as the main boot device.

 

I shall delete the redundant drivers and try again

 

As for nvda_drv = 1, that I should have taken out a while ago. It shouldn't be there.

 

 

Edit: I love ya, dude. Works now.

[Andy Vandijck]

IS IT OK TO DELTE THIS DRIVER? I`M BOOTING mac 10.8 AND wINDOWS 8.

If you boot UEFI I wouldn't...

It's needed to load the kernel(cache) and drivers, etc..