Clover General discussion

[Onixs]

I put the HFSPlus-64.efi from the driversUEFI64 folder into my drivers64 folder.

 

Right now I only have two folders drivers32 and drivers64. I deleted the UEFI folders since my system is not UEFI. Only FSinject-32.efi and HFSPlus.efi are in the drivers32 folder and FSInject-62.efi and HFSPlus-64.efi are in the drivers64 folder.

 

Boot time is still about 1 minute.

its taking time (too long) to read kernelcache

[pkdesign]

i see a lot of reference to biosblockio but no info other then this http://clover-wiki.zetam.org/Installation where and how do you activate it. or do you have to build clover a certain way for this function. maybe wiki needs an update?

 

As jamiethemorris said, you have to go through the installation process again and under CloverEFI choose the CloverEFI-64 BiosBlockIO option.

 

its taking time (too long) to read kernelcache

 

So is there something I can do to speed the process? Is there something in config.plist that I should change?

 

As I mentioned in my original post, Chameleon loads in under :30 seconds.

[Onixs]

i do have the same issue as yours with my old ich7 ide sata and it has been that way since the early stage of clover.

[bronxteck]

thanks tluck that did it but i had to use -s flag to change the build -t did not function for clover grower pro. pk design also thank you i saw the option in the svn built clover but not in my personal build. changing 

Default target(s) to use (ia32, x64, x64-mcp) [x64]: x64-mcp

worked.

[Slice]

Clover appeared in r3012. Do three of the front (clover version 3xxx) is the next revolution in guration bootloader, or is it just another version, like each previous (for example, no big changes)?

No functional change compared to 2999. 

30xx changes related to "How to build Clover", and changes in Ovmf that you are not using.

Three Clovers in official installer appeared by mistake. You may safely delete redundant. They are the same.

 

EDITED. But I am not sure about changes in EDK2. Looks like they didn't touch us.

[vusun123]

Did something happen or Clover in Legacy BIOS is really sucks these days ? I have been installing OS X onto my brother's laptop - a Sony Vaio SVE15117 using AMD Radeon 7650M, InsydeH2O BIOS and no UEFI option. Well I got everything to work ( GPU included ) but I faced many problems while installing and completing it. One is that FakeSMC doesn't want to be loaded at all, even injeting in EFI\Clover\kexts\10.10 or putting it in S/L/E without rebuilding cache. And Clover doesn't seem to automatically patch the AMD Graphics unless I manually patch it first, reboot got it working and then install the vanilla kext back. And one major problem: can't boot Legacy Windows ( be it 7 or 8.1 ). Can anyone address the issues here ?

[liujianwei]

Clover can write CPUFrequency into DMI but it can't make CPU run faster.

BusSpeed has more influence. I don't know all mechanics of this just observations. Resume is to set BusSpeed to be as close to real as possible. But again, Clover can not change real Speed. It informs kernel what is the speed.

BIOS can change real clocks, OEM dependent, I don't know how.

DMI and SMBIOS data stored in SystemTable created by UEFI BIOS. For legacy BIOS computers it was the task #1 to emulate this table to give an ability start OSX. Are you a programmer for further explanations?

So I can change CPUFrequency to 10GHz without crash?

And I think DMI stored in CMOS because overtime I plug in USB flash Drive and reboot my computer shows backup cmos.

I have installed Clover_v2k_r3021 in disk0s2 disk0s3 and disk1s1,I'm use clover on legacy MBR not GTP/MBR,I'm using Option A: BIOS-based PC (old motherboard) BIOS>MBR>PBR>boot>CLOVERX64.efi>OSLoader.Tried boot0af and boot0ss all works as described.The first problem is boot0ss said"This bootloader is a good choice when you have Windows installed on the same disk because Windows wants to have its partition active."But if I active windows Partition,then computer boot from windows,clover won't boot.

 

The second problem is when I active disk0s3 partition,and install boot0af to MBR,and BIOS>MBR>PBR>boot> works well,but the boot file in the disk0s3 boot CLOVERX64.efi in the disk0s2,use disk1s1 boot file the same thing happened.I'm sure because I use different theme on different partition clover,and I install boot0af in the MBR,because if I delete file "boot" in the disk0s3 my computer shows boot1:error(if I install boot0ss in the MBR it boot disk0s2 clover well).So the problem is file  disk0s2 disk0s3 and disk1s1's "boot" always boot disk0s2 CLOVERX64.efi,but I want they boot themselves partition clover.The only solution I can do is delete disk0s2,then disk0s3 and disk1s1 boot disk0s3's CLOVERX64.efi.

I have experienced this problem when I use niresh mavericks that install Clover in it's partition.So I tried many time and pretty sure this is the boot file problem.

The problem is legacy problem, since he's using the legacy firmware to create a UEFI environment then booting legacy from ANOTHER disk, the firmware doesn't know which disk to boot from because it can't figure out which bios id belongs to which drive from UEFI. Therefore DL register is probably filled with trash or with booted disk. Hence, BCD error because BCD can not be found on that disk. When you boot from the disk windows is on, then it works, right? There's not really a solution to that, but you can reinstall windows in UEFI instead and this problem won't exist.

What's ANOTHER disk?I have tried use HDD clover to boot SSD windows 7(plug in and sata0 and sata1),and boot is success.But use USB clover boot BCD error.

[Riley Freeman]

I just noticed that recent Clover versions have broken bdmesg here on my X79. Running it gives an "illegal instruction" message in Terminal. In Console it prints: 

 

ReportCrash[161]: Failed to create CSSymbolicatorRef for bash

 

followed by a crash report.

 

This happens with both r2999 and r3021. Going back to r2988 fixed it.

[pkdesign]

Oh man. Was giving it another look this evening and I thought maybe I should rebuild my kext. So I fired up Kext Utility and voila, my machine boots in under :30 seconds just as it does with Chameleon.

 

Hopefully this will help someone else. Always be sure to rebuild your kext.

[Maniac10]

I just noticed that recent Clover versions have broken bdmesg here on my X79. Running it gives an "illegal instruction" message in Terminal. In Console it prints: 

 

ReportCrash[161]: Failed to create CSSymbolicatorRef for bash

 

followed by a crash report.

 

This happens with both r2999 and r3021. Going back to r2988 fixed it.

 

Working fine with r3027.

[wern apfel]

had an strange issue with clover, it seems that the windowserver is not starting

no problem with chameleon

it's an sandy bridge laptop, with optimus enabled the windowserver is working

but it's also working with nullcpupm loaded aicpupm is enabled

can someone please help me?

[Dr. Hurt]

I'm having a strange issue.

 

Clover GUI shows boot "Microsoft EFI" from "Yosemite" HFS+ partition.

And boot "Clover EFI" from EFI partition!!

 

Thats after clean legacy install on the Yosemite partition.

 

My partition scheme is as follows:

 

1- EFI (ms-win boot loader is here)

2- Yosemite (clover is here)

3- Recovery

4- Data

5- Windows installation files

6- Windows 8.1 UEFI

 

EDIT 1:

Fixed the "boot Clover EFI from EFI" entry by reformatting the EFI partition.

I still get the boot "Microsoft EFI boot menu from Yosemite" option though!!!

[fusion71au]

@Dr Hurt,

 

I found that with VBoxHFS-64.efi (built into recent builds of Clover EFI boot file), there was a spurious Windows bootmgr.efi entry appearing on my OSX partition (and also longer boot up times).  Others have reported the same issue over at ProjectOSX....

Replacing the "boot" file with one from Clover versions earlier than v2988 (which had HFSPlus.efi built in) solved the issue for me.  I think @Slice is aware of the issue and we might in future be able to specify CloverGrowerPro to compile with our choice of HFSPlus.efi.

[fc bayern]

Is it just me , I dont know, but the latest clover version 2988 slown down the boot time by 4-5 seconds - before the progres bar appears it waits couple of seconds. WIth previos versions that was not the case, it showed the progress bar almost immediatelly.This I can confirm on my 2 hackintosh systems, hack Mac pro and hack macbookpro.

This is still issue with the newer versions of clover. Any idea how to speed up the boot time -I'm using Hfsplus.efi  jus to to mention.

[Maniac10]

The delay disappeared for me, it's even faster than ever. Just try BiosBlockIO instead.

[fc bayern]

OK, i will try this and post the result.

[crusher]

Working fine with r3027.

 

Do you have drivers32UEFi folder and CLOVERx32.efi file.

[Maniac10]

No, I build my own 64bit only version.

[Goddessavril]

There's the darkness of 30s~1min before i get to the clover GUI. And it's usually 5s or so before the upgradation to yosemite 10.10.1.  Anyone knows what causes the slowness?

clover version: r2999 and r3021

[crusher]

No, I build my own 64bit only version.

You install Legacy support?

[fc bayern]

The delay disappeared for me, it's even faster than ever. Just try BiosBlockIO instead.

That worked.Good hack

[Maniac10]

You install Legacy support?

Yes, I boot with BiosBlockIO.

[teeodor]

Hi,i install clover on a usb stick.

In my optiplex 755 the usb stick boot good,i se the screen of clover but i don't see the harddisk and the mouse doesn't work.

I see only the usb stick and the hd recovery.

In another pc work good

[pkdesign]

Hmm, something odd has happened. The :20 second delay after BIOS is back. I definitely installed using BiosBloack and rebuilt kexts but still the delay is there again. I've build two separate USB sticks and they boht do it.

 

Any more ideas, thoughts?

[Philip Petev]

Hi guys.

There is something wrong with the latest OS X version 10.10.1. Tried a fresh install with flash drive, created with this method and something strange happens every time: Clover boots the install system from the flash drive, the system begins to boot up and when the moment comes to fire up the graphics (HD4000), there is just black blank screen for a moment and a reboot of the computer follows. That also happened with the newly installed 10.10.1 system, installed by the legacy way (Chameleon), but booted with Clover on the first run (booting the same system from the legacy USB drive with Chameleon works). There is no such problem with OS X system, installed first as 10.10 and then updated to 10.10.1 via the App Store. Tried several things:

- several Clover versions

- several SMBIOS personalities

- rebuilding the kext cache several times (the freshly installed 10.10.1 was booted up with Chameleon in that case)

but no go.

Testes on both my board (GA-H77N-WIFI) and another one (GA-Z77N-WIFI), maybe other H77/Z77 boards (Gigabyte or not) are affected too.

The problem doesn't occur on my laptop (ProBook 4340s), tested the same flash drive (with the EFI data for my laptop of course) and there is no problem with both the installation and the newly installed system.

 

I have one presumptions: something has been changed in the latest OS X installer app, affecting the UEFI boot on the boards with AMI UEFI.

 

Edit: I've uploaded here my data: Clover debug logs from both the working (10.10. updated to 10.10.1) and non-working (10.10.1 fresh) installs, a copy of ioreg (taken from the working install), my EFI folder and system.log from the non-working install.