Jump to content

Clover General discussion


ErmaC
29,818 posts in this topic

Recommended Posts

Hi, guys.

 

So, during the past few days I think I formatted and installed Yosemite in both Legacy and UEFI mode at least 10 times, just to see where does my iMessage issue come from. And I made a very interesting discovery (and very annoying at the same time): Mac OS doesn't actually completely erase the freaking partitions!
And here's the proof: after completing making the bootable USB stick with Clover as boot loader (legacy or UEFI doesn't seem to matter anymore), I left a file in the Trash. Not on purpose, but I just forgot to empty the trash, before rebooting into the USB and starting reinstalling the system. A file that could have never been from the system, because it was a tutorial I got from the internet and I didn't need anymore. So there is no way that would have been from the installation. But bear with me. 

So I booted up into the USB, as I normally do, I reached the installer, I went to Disk Utility, unmounted the Yosemite partition (though I don't think that's mandatory since it seems to unmount it previous to formatting anyway) then I went to the Erase tab and clicked Erase. I accepted the pop-up and waited patiently until it did its job.
Then I closed the window and resumed the installation, as usual.

Well, after everything was done and I was back into the Desktop, what do you think I found in the Trash? The file I deleted previous to the formatting and the "new" installation. So who knows what else might it have kept, even after formatting the freaking partition. I'm so pissed off right now! How is it possible, for such an operating system to have such issues? And no, I don't think that's something specific to Yosemite. I think Mavericks does it too. I just don't have enough patience to reinstall that as well right now.

The first time this happened I thought it was just me. I thought the fact that I got a "packages" shortcut (from when you make the bootable USB drive) is just my imagination. But now I'm sure.

So, since none of my "clean" installations was actually clean, can you, please, guide me on how the hell can I erase a partition, for real?

Link to comment
Share on other sites

Use the security options available under the erase tab in Disk Utility and select the most secure one: i think it writes zeros on the entire disk and do it several times, complying with some US security standards. If even this fails to actually delete the disk, download a live Ubuntu disk and use it to do the job.

 

Good luck!


P.S.: and yeah, this is freaking scary and explains something that happened some time ago when I was doing some tests with my AMD rig. That time, I thought it was some confusion of mine, but perhaps it wasn't me that was confused after all.

  • Like 1
Link to comment
Share on other sites

Use the security options available under the erase tab in Disk Utility and select the most secure one: i think it writes zeros on the entire disk and do it several times, complying with some US security standards. If even this fail to actually delete the disk, download a live Ubuntu disk and use it to do the job.

 

Good luck!

 

P.S.: and yeah, this is freaking scary and explain something that happened some time ago when I was doing some tests with my AMD rig. That time, I thought it was some confusion of mine, but perhaps it wasn't me that was confused after all.

Thanks, man. I'll try that tonight, when I get back home. I'm on my way to work right now. I've got Ubuntu on another hard drive. I use it mostly to compile Android. So I don't actually need a live CD/usb. But does gparted know about Mac partitions? HSF+ i think, right? can it interact with them and actually get the job done? I guess there is only one way to find out. :)

 

Thanks again. I'll report back. :)

Link to comment
Share on other sites

Clover 2721

In debug.log as I resolve "Cpu0Ist" I see true

3:323  0:007  Dropping 3 tables
3:330  0:007  Drop table 0 signature="DMAR" (52414D44)
3:390  0:060  set table: 52414D44,                0 to drop:
3:406  0:015  Drop table 1 signature="SSDT" (54445353) table-id="Cpu0Ist" (0074734930757043)
3:534  0:127  set table: 54445353,   74734930757043 to drop:
3:549  0:015  Drop table 2 signature="SSDT" (54445353) table-id="CpuPm" (0000006D50757043)
3:661  0:112  set table: 54445353,       6D50757043 to drop:  true

Thanks

 

Fabio

OK,

If this is a question about correct dropping SSDT tables then see your boot.log/preboot.log/debug.log

1:571  0:001  Get Acpi Tables List from RSDT:
1:575  0:003   Found table: FACP  A M I len=132
1:577  0:001   Found table: APIC  A M I len=256
1:579  0:001   Found table: FPDT  A M I len=68
1:580  0:001   Found table: MCFG  OEMMCFG. len=60
1:582  0:001   Found table: HPET  A M I len=56
1:584  0:001   Found table: SSDT  CpuPm len=839976
1:586  0:001   Found table: EINJ  AMI EINJ len=304
1:588  0:001   Found table: ERST  AMI ERST len=560
1:590  0:001   Found table: HEST  AMI HEST len=168
1:593  0:003   Found table: BERT  AMI BERT len=48
1:595  0:001  Calibrated TSC frequency =3199974350 =3199MHz

There is no DMAR table. No need to drop it.

There is one SSDT table and it has signature CpuPm. You successfully dropped it.

There is no table with signature Cpu0Ist. No need to drop it.

  • Like 1
Link to comment
Share on other sites

Hi there Slice, i think that this can be interesting, about booting with apple's logo...

 

Original Post link from MiniHack

 

Pasted:

 

I think Pike has the answer….http://pikeralpha.wo...gessound-files/ (see towards end of post). If I understand correctly - and possible I don't! - then the boot loader will need to recognise and load the efires images/logos mentioned. So guess it is probably just waiting on Clover/Chameleon implementation in due course.

Link to comment
Share on other sites

I'm trying to see if I can replace the DSDT.aml override on my X79 with Clover patches. There's only a few edits that I make and most are covered, although there's some bits I'd like to change. I think the key might be through binary DSDT patching but I didn't have much success with a test last night.

 

First off, there are two values in the nvidia gfx injection that I want to change. The card is listed as Asus GeForce on PCI Slot 0 but I want it as NVIDIA GeForce on Slot-1 as this is how I always inject it. It's purely cosmetic but that's how I like it.

 

More importantly I need to inject fake device IDs for one of my USB3 ports. The USB injection only seems to pick up my rear hub-attached ports (it renames ASMX to XHC1 in the process). However, I don't use these as they cause KPs here (I'm on Snow). Instead I inject fake IDs for the front ports which allows the USB3 driver to load and enable only those.

 

I tried adding the replacement ven/dev id to the XHCI section but it didn't seem to get applied. In any case, it might have gotten applied to both ports and I only want the front ones patched.

 

Any tips on how I could go about applying these patches?

Link to comment
Share on other sites

Thanks. The first link took care of the slot name.

 

The second link is what I tried last night. The XHCI injection didn't work. I'm thinking it's probably because I'm trying to fake the vendor as well as the device. Still, I need it to only inject it for one instance. In other words, I need to fake the IDs for the port on PEX2, but not the one on PEX3.

 

This is what I need but the samples link to applelife.ru doesn't work.

Link to comment
Share on other sites

Hi there Slice, i think that this can be interesting, about booting with apple's logo...

 

Original Post link from MiniHack

 

Pasted:

 

I think Pike has the answer….http://pikeralpha.wo...gessound-files/ (see towards end of post). If I understand correctly - and possible I don't! - then the boot loader will need to recognise and load the efires images/logos mentioned. So guess it is probably just waiting on Clover/Chameleon implementation in due course.

 

This is Chameleon way to load manually system files. Clover will not do this. It must be boot.efi affair. Some more research needed.

 

I'm trying to see if I can replace the DSDT.aml override on my X79 with Clover patches. There's only a few edits that I make and most are covered, although there's some bits I'd like to change. I think the key might be through binary DSDT patching but I didn't have much success with a test last night.

 

First off, there are two values in the nvidia gfx injection that I want to change. The card is listed as Asus GeForce on PCI Slot 0 but I want it as NVIDIA GeForce on Slot-1 as this is how I always inject it. It's purely cosmetic but that's how I like it.

 

More importantly I need to inject fake device IDs for one of my USB3 ports. The USB injection only seems to pick up my rear hub-attached ports (it renames ASMX to XHC1 in the process). However, I don't use these as they cause KPs here (I'm on Snow). Instead I inject fake IDs for the front ports which allows the USB3 driver to load and enable only those.

 

I tried adding the replacement ven/dev id to the XHCI section but it didn't seem to get applied. In any case, it might have gotten applied to both ports and I only want the front ones patched.

 

Any tips on how I could go about applying these patches?

Sorry, but automatic DSDT patches by Clover are far from perfect because artificial intellect is still not created.

My proposition is to prepare DSDT patched by Clover, save it, and then patch manually for all you want to achieve the best DSDT.

Then you set

DSDT->Name=DSDT.aml which is your custom DSDT.aml

Fixes: NewWay_80000000 + FixRegions_10000000 that needed because of Floating regions.

 

 

 

 

This is what I need but the samples link to applelife.ru doesn't work.

At this moment applelife.ru is down.

  • Like 1
Link to comment
Share on other sites

OK,

If this is a question about correct dropping SSDT tables then see your boot.log/preboot.log/debug.log

 

1:571  0:001  Get Acpi Tables List from RSDT:
1:575  0:003   Found table: FACP  A M I len=132
1:577  0:001   Found table: APIC  A M I len=256
1:579  0:001   Found table: FPDT  A M I len=68
1:580  0:001   Found table: MCFG  OEMMCFG. len=60
1:582  0:001   Found table: HPET  A M I len=56
1:584  0:001   Found table: SSDT  CpuPm len=839976
1:586  0:001   Found table: EINJ  AMI EINJ len=304
1:588  0:001   Found table: ERST  AMI ERST len=560
1:590  0:001   Found table: HEST  AMI HEST len=168
1:593  0:003   Found table: BERT  AMI BERT len=48
1:595  0:001  Calibrated TSC frequency =3199974350 =3199MHz

There is no DMAR table. No need to drop it.

There is one SSDT table and it has signature CpuPm. You successfully dropped it.

There is no table with signature Cpu0Ist. No need to drop it.

Thanks for your reply.

 

Fabio

Link to comment
Share on other sites

Hello, i updated clover to 2721 from 2703 and everything went fine but when i boot and i go to clover pref panel, it says that it booted with 2703 and the installed version is 2721, any ideea why? It's like having both versions of clover installed :s

Link to comment
Share on other sites

I'm facing an odd experience with FACP fix implemented in Clover on my Eee PC1201N.

Thie laptop needs the restart fix because by default it doesn't support Reset Register (values are 00 00). So i have to enter 0x64 0xFE in config in order to gain proper reboot. Without the fix the laptop just stays at black screen and nothing happens. With reset address and value specified in config restart starts working .. but a subsequent issue arises. Every restart causes CMOS reset. If i don't enable the fix or use something like EvoReboot the CMOS remains intact after reboot, which led me to thinking it's something related to how Clover adjusts the FACP table. 

 

Enclosed are stock BIOS FACP/FACS tables and tables fixed by Clover.

1201N_facp_issue.zip

 

Edit:

Disregard this for the time being! I guess the problem lies somewhere else .. there's a driver in my system that accesses RTC and corrupts CMOS checksum. I will try to deal with it hoping the problem goes away.

 

UPD:

Yes, completely disregard the above issue - unrelated to Clover by all means. The problem comes from accessing CMOS without SMI interrupt and thus CRC corruption occurs.

Link to comment
Share on other sites

@nsxcp2005 - ok. can you boot it in safe mode or single user mode? should take a different path on loading drivers right? might help you narrow the issue. what about GMA 950 gfx kext for mavericks - does it exist? the installer usually uses a more generic driver.

Link to comment
Share on other sites

I have a question about the new Clover 2728 Refers to Yosemite DP2

 

How can I set the default volume? in the conig.plist? 

 

With the "Univ unique identification." From the Disk Utility information - that does not work

 

and with your entry eg hd (1,2) also does not work.. 

 

AND how can you change the font color of the label partition

 

Thanks for info

Link to comment
Share on other sites

Has anyone had success installing clover on a PCIe SSD? I have a Revodrive that works fine with Chameleon installed on the EFI partition of the drive but clover bootloader can't see it even from another boot disk.

I am using Clover in conjunction with generic two port/slot bootable pci-e add in card using I think the Marvel 9128 chipset for the controller. Clover sees both SSDs I have in the enclosure at the moment and boots from either my main or backup drive I have in it for testing new updates or for changes I make to the OS.
Link to comment
Share on other sites

I have a question about the new Clover 2728 Refers to Yosemite DP2

 

How can I set the default volume? in the conig.plist

 

With the "Univ unique identification." From the Disk Utility information - that does not work

 

and with your entry eg hd (1,2) also does not work.. 

 

AND how can you change the font color of the label partition

 

Thanks for info

Link to comment
Share on other sites

×
×
  • Create New...