Clover General discussion

[The Real Deal]

Thanks for the new method to enable web drivers. My 980 Ti now works.

[PMheart]

Same. Just different developers place them in different places. Both work same way.

Thanks Slice. This means that I just set one of them to enable Cx is OK, right? 

[makk]

Hello and thank you for a great stupendous work!

 

3922 causes BIOS reset installed on SSD with my Laptop Asus K53E

 

Upon reboot the laptop bios goes into reconfiguration option (not sure what to call it) but isn't the normal EFI clover boot up selection screen, and isn't the BIOs configuration normal way.

 

with ESC usually get boot options to choose which drive, usb, network, to boot from

when choosing the SSD drive as normal goes to that strange BIOS configuration setting.  ( don't know what to call it)

probably BIOS pre-config setup.

 

But, this doesn't happen with >> USB stick. So can boot with USB stick to revert back to 3899.

 

I wanted to use the option Don't load "ASPM" which I read was implemented since 3904 or 5 whichever one.

I downloaded 3922 because that was what sourceforge had for automatic download.

 

DisableASPM = True 

 

and one last thing,

 

I'm not a programmer so I don't recognize programming language.  I'm a tech I blow xxx up and then fix.

 

there's a statement in a post referring to DisableASPM or APSM for AppleLPC on boot.

 

There's no Clover Button 

and when you type in a post

 

<key>ACPI</key>

    <dict>
        <key>DisableASPM</key>
        <true/>.   <<<<where does this go?  what do you select in XCODE ?  Binary, String, Number, Boolean?

 

    </dict>

 

 

This is what I ascertain from your post

 

hit the + on XCODE under ACPI

type:  DisableASPM then select Dictionary

 

Then what?    <true/>. how to get this?     String or Boolean?  in the Dictionary

 

 

 

Thanks a million!!  your config is awesome!  Great work!

 

Stu

 

[Slice]

 

Then what?    . how to get this?     String or Boolean?  in the Dictionary

 

 

This is Boolean.

Xcode shows it as YES/NO

[makk]

This is Boolean.

Xcode shows it as YES/NO

 

Thanks a million 

I suppose will not work in 3899.

 

What changed from 3899 to 3922 that caused my BIOS to reset?  

 

I installed with same options as previous to keep insync .  

Could it be that I installed wrong 64bit efi files?

 

Can I get 3904 somewhere?

 

Thank you again.

Stu

[Slice]

Yes, rev 3922 is broken. Take 3911 from sf.net.

I can understand what is wrong only next week.

[makk]

Yes, rev 3922 is broken. Take 3911 from sf.net.

I can understand what is wrong only next week.

 

Will do!  Thanks

Have a good week-end as well.  

 

How to say in Russian?  Have a good week 

 

Thank you

 

Stu

[Slice]

"Хороших выходных!"

[Mr MagOO]

Hi all,

@makiko : I think that "DisableASPM" fonction is broken in 3911 (see post             #11737            )

If it works for you, please let us know.

 

Thank you, have a good week end.

[ErmaC]

@Slice, or any other, who it may....

I'm going to split out the recent post about Clover/Polaris in a separate topic...

 

Cordially

 

ErmaC

 

 

EDIT: done I split it into a "new" topic Clover test and patches for Polaris GPU

PS: sorry fantomas1... the first post is your!

Edited November 12, 2016 by ErmaC
post update

[JorgeMax]

Can anyone explain to me what each option is for in the "Boot Graphics" section of the Clover?

 

Default Background Color;

UI Scale;

EFI Login Hi DPI;

Flag State.

 

thanks

[Slice]

Those values was asked during FileVault2 procedure.

More or less good values is written in config-sample.plist. I don't no more.

If set UISacle=2 then the Apple during boot will be twice larger. 

[Mork vom Ork]

Question: is it possible to get "Disable" feature for entries within "AddProperties" Array like we have for KextsToPatch like this: (following picture is just an example)

right now we are not able to "disable" one of your inserted properties. You could only disable the whole array by making it "#AddProperties".

Would be great to disable just some entries instead of whole function.

[Slice]

OK, will do this next week.

Question: is it possible to get "Disable" feature for entries within "AddProperties" Array like we have for KextsToPatch like this: (following picture is just an example)

 

right now we are not able to "disable" one of your inserted properties. You could only disable the whole array by making it "#AddProperties".

Would be great to disable just some entries instead of whole function.

You can disable one property by disable key "Device"

[Mork vom Ork]

OK, will do this next week.

You can disable one property by disable key "Device"

ah, good point, thanx. that will do it for me.

 

Just in case, anyone want's to do this: set "ATI" to "#ATI"!

Do NOT set "Device" to "#Device" - as i did in my first try. If you do this, CLOVER wont start   <--- so, now you have been warned.

[nmano]

future update Clover config without FakeSMC.kext

I dont know when?

 

 

 

[Slice]

future update Clover config without FakeSMC.kext

I dont know when?

As far as you solded SMC device into your motherboard.

It can't be virtual device, it must be real chip.

[nmano]

#Maintained by: N.Mano
# 1. Under Device PCI0_SMC
# 2. Fixed some X99_Warnings
into device label SMC parent_label PCI0 remove_entry;
into device label PCI0 insert begin
Device (SMC)\n
{\n
Name (_HID, EisaId ("APP0001"))  // _HID: Hardware ID\n
                    Name (_CID, "smc-napa")  // _CID: Compatible ID\n
                    Name (_STA, 0x0B)  // _STA: Status\n
                    Name (_CRS, ResourceTemplate ()  // _CRS: Current Resource Settings\n
                    {\n
                       
})\n
                Return (Package ()\n
                {\n
                    "plugin-type",\n 
                    One\n
                })\n
}\n
end;

This patches is made changes ioregistry

manoranjan’s Mac Pro.zip

[nmano]

Its sample config

Its worked with FakeSMC.kext

But I dont see any Error.

config.plist.zip

F22J All.zip

[Slice]

Its sample config

Its worked with FakeSMC.kext

But I dont see any Error.

DSDT is the description only. FakeSMC works without this.

You need real device to work without FakeSMC.kext.

FakeSMC created virtual device but it works only inside kernel.

SMCHelper is another story. It just keep own SMC keys database to answer on boot.efi questions. It can be improved in future but it also has no deal with DSDT.

As far as you solded SMC device into your motherboard.

It can't be virtual device, it must be real chip.

For example.

You can write into DSDT that you have AMD Radeon HD770. Will you get a full Radeon graphics if you have no such device? 

[mhaeuser]

How does a DSDT edit make it a virtual device? As you said right before, it's a description, not a vDevice.

[Slice]

How does a DSDT edit make it a virtual device? As you said right before, it's a description, not a vDevice.

I meant the Device written into DSDT in nmano's question.

[mhaeuser]

I meant the Device written into DSDT in nmano's question.

 

Yes, I get that, but you cited your own quote stressing it must be a real device.

That adding a device to DSDT is not enough should be obvious.

[smolderas]

Hello Slice,

 

you proposed to write the "nvram.plist" into the EFI partition of the boot volume, since It is not encrypted with filevault2.

At least for the legacy clover with the change of the function "findFirstAppleBootDevice" from the file "rc.shutdown.d/80.save_nvram_plist.local" the content of the nvram will be saved to the EFI partition of the boot device (core storage):

function findFirstAppleBootDevice {
    local rootDevice=$(LC_ALL=C diskutil info /|awk '{ if ($1 == "Device" && $2 == "Node:") {print $3}}')
    recoveryDisk=$(LC_ALL=C diskutil info /|awk '{ if ($1 == "Recovery" && $2 == "Disk:") {print $3}}')
    device=$(LC_ALL=C diskutil info $recoveryDisk|awk '{ if ($1 == "Part" && $2 == "of" && $3 == "Whole:") {print $4}}')
        # Find EFI partition signature
        index=$(LC_ALL=C /usr/sbin/gpt -r show "$device" 2>/dev/null | \
         awk 'toupper($7) == "C12A7328-F81F-11D2-BA4B-00A0C93EC93B" {print $3; exit}')
        if [[ ! $index =~ ^[0-9]+$ ]];then
            index=
        fi
    # If index found return the device and partition (like diskXsY)
    [[ -n "$index" ]] && echo "${device##*/}s$index"
}

Cheers,

Smolderas

 

 

Edit:

Here is the diff:

41,44c41,44
<     local devices=$(LC_ALL=C ls /dev/disk* | sed -n '/\/dev\/disk[0-9]*$/p')
<     # Iterate over all devices
<     for device in $devices; do
<         # Find first partition with Apple Boot partition signature
---
>     local rootDevice=$(LC_ALL=C diskutil info /|awk '{ if ($1 == "Device" && $2 == "Node:") {print $3}}') 
>     recoveryDisk=$(LC_ALL=C diskutil info /|awk '{ if ($1 == "Recovery" && $2 == "Disk:") {print $3}}')
>     device=$(LC_ALL=C diskutil info $recoveryDisk|awk '{ if ($1 == "Part" && $2 == "of" && $3 == "Whole:") {print $4}}')
>         # Find EFI partition signature
46,49c46,47
<          awk 'toupper($7) == "426F6F74-0000-11AA-AA11-00306543ECAC" {print $3; exit}')
<         if [[ $index =~ ^[0-9]+$ ]];then
<             break
<         else
---
>          awk 'toupper($7) == "C12A7328-F81F-11D2-BA4B-00A0C93EC93B" {print $3; exit}')
>         if [[ ! $index =~ ^[0-9]+$ ]];then
52d49
<     done

[J Lamp]

I have a question regarding the "other" kexts folder. I'm running Sierra and had a some customized plist injection in FakeSMC in the 10.12 folder that was not loading. Apparently the unedited, older FakeSMC I had in the "other" folder was overriding it. Deleting the kexts from the other kexts folder fixed the issue.

 

My understanding was that the other kexts folder was for unrecognized OSs (i.e. a new beta etc.) and not as a general override folder. Do I have that wrong or is this behaving oddly?

 

Thanks.