Jump to content

Clover General discussion

ErmaC

By ErmaC,
in Clover

 Share
29,866 posts in this topic

Recommended Posts

Sherlocks
Sherlocks
Posted
Posted

Search in DSDT

Clover read DVMT in DSDT? I dont see dvmt. I have a skylake laptop. Apple changed to need DVMT higher than HD4600 or down. Broadwell and skylake DVMT need 96mb in OSX. But many broadwell and skylake have DVMT 32mb. If owner of BW an SKL, have to change DVMT in BIOS. But many laptop cant change DVMT. Manufacture prevent this feature. Sometime can change DVMT. But recent Manufacture perfectly prevent this trick. Example insydeh20. So i just wonder that we can control DVMT in Clover? Or already define DVMT in DSDT?

 

나의 LG-F410S 의 Tapatalk에서 보냄

Link to comment
Share on other sites
Slice
Slice
Posted
Posted

May be graphics chip, some of its registers. And they can be shown in DSDT like these

            OperationRegion (IGDP, PCI_Config, 0x40, 0xC0)
            Field (IGDP, AnyAcc, NoLock, Preserve)
...
            OperationRegion (IGDM, SystemMemory, ASLB, 0x2000)
            Field (IGDM, AnyAcc, NoLock, Preserve)

I don't know.

Clover read DVMT in DSDT? I dont see dvmt. I have a skylake laptop. Apple changed to need DVMT higher than HD4600 or down. Broadwell and skylake DVMT need 96mb in OSX. But many broadwell and skylake have DVMT 32mb. If owner of BW an SKL, have to change DVMT in BIOS. But many laptop cant change DVMT. Manufacture prevent this feature. Sometime can change DVMT. But recent Manufacture perfectly prevent this trick. Example insydeh20. So i just wonder that we can control DVMT in Clover? Or already define DVMT in DSDT?

나의 LG-F410S 의 Tapatalk에서 보냄

 

Clover do nothing with DVMT.

Link to comment
Share on other sites
RehabMan
RehabMan
Posted
Posted

May be graphics chip, some of its registers. And they can be shown in DSDT like these

            OperationRegion (IGDP, PCI_Config, 0x40, 0xC0)
            Field (IGDP, AnyAcc, NoLock, Preserve)
...
            OperationRegion (IGDM, SystemMemory, ASLB, 0x2000)
            Field (IGDM, AnyAcc, NoLock, Preserve)
I don't know.

 

This is PCI config and other graphics system memory. But reading the Linux PRMs for Intel graphics I know this:

- DVMT-prealloc is referenced as "DSM" or "stolen ram" in the PRMs/Intel docs

- this "stolen ram" is a memory region in physical memory dedicated to the graphics chip

- via various chipset registers, its base address and size is set (the docs are not super clear)

- there are also other registers which point inside the "stolen ram" region (eg. trying to move one will require updating other registers that point inside it)

- probably any sort of UEFI memory map would have to be updated to reserve this "stolen ram" region if it were changed by Clover...

 

Clover do nothing with DVMT.

The thought is that it might be able to do something with this very nasty problem due to BIOS vendors hiding and disabling various settings like this one...

 

Looks pretty complicated though.

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

@Slice

 

I have issue when trying to use Settings=Special feature documented here. According to the post, I made a special Sierra.plist so I can use a different AMD framebuffer patch for 10.12 (as you may know, they've changed since El Capitan).

 

But it seems like doing so, the patch in Sierra.plist is ignored and that it's the one from main config.plist that's used in every case. So I tried the same patch with a normal config.plist on a USB Pen and it worked fine.

 

Patch lenght is superior in Sierra.

 

When I boot using the USB pen, I have this in log :

20:825  0:000   ATIController: 7000

20:825  0:000   ATIDataLength: 96

When I use config.plist with Settings=Sierra in GUI from HDD, I have this when I boot :

15:920  0:000   ATIController: 7000

15:920  0:000   ATIDataLength: 64

Obviously, that's the patch for El Capitan :(.

 

Also, I've noticed that's the same for Kernel and Kext Patches. None of the ones in Sierra.plist will be applied. You can't see it in attached logs but I've tried when I used this feature instead of ATIConnectorsData/ATIConnectorsPatch.

 

Is there a mistake somewhere ?

 

Thanks.

Configs.zip

Logs.zip

Link to comment
Share on other sites
Slice
Slice
Posted
Posted

It has been a while since I looked at it, but from what I remember Clover cannot find the PCI audio device on (some) Skylake because it is searching for PCI class incorrectly...

 

Note that Apple changed the class match for AppleHDAController because they made the same mistake:

 

10.7.5 and 10.11.5 happen to be the two versions I have installed on my desktop:

10.7.5: IOPCIClassMatch=0x04020000&0xFFFE0000

10.11.5: IOPCIClassMatch=0x04010000&0xFFFD0000

 

It is possible Clover code was based on the old/wrong IOPCIClassMatch.

 

I don't use Clover audio inject (prefer ACPI or /Devices/Arbitrary), so I didn't bother trying to fix it...

Thanks for the note. I accepted class code 0401 in Clover rev 3569. It is AC97 Audio while 0403 is HD Audio.

But industry standard is for 0401. Looks like some vendors will return to 0401 (Intel, Apple, Creative).

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

I'm trying to build rev 3569 with CloverGrowerPro but the process is failing. I'm on 10.9.5 with Xcode 6.2. Can somebody help me?

Building ... /Users/administrator/CloverGrowerPro/edk2/MdePkg/Library/UefiUsbLib/UefiUsbLib.inf [X64]
[CC] VideoBiosPatchLib
/bin/sh: /Users/administrator/src/opt/local/bin/nasm: No such file or directory
make: *** [/Users/administrator/CloverGrowerPro/edk2/Build/Clover/RELEASE_GCC49/X64/MdePkg/Library/BaseLib/BaseLib/OUTPUT/X64/Thunk16.obj] Error 127


build.py...
 : error 7000: Failed to execute command
make tbuild [/Users/administrator/CloverGrowerPro/edk2/Build/Clover/RELEASE_GCC49/X64/MdePkg/Library/BaseLib/BaseLib]


build.py...
 : error F002: Failed to build module
/Users/administrator/CloverGrowerPro/edk2/MdePkg/Library/BaseLib/BaseLib.inf [X64, GCC49, RELEASE]

- Failed -
Build end time: 04:43:10, Jun.22 2016
Build total time: 00:00:13

Cloverx64 release  ERROR!!

EDIT: Copying /Users/administrator/CloverGrowerPro/toolchain/* to /Users/administrator/src/opt/local should work?

 

EDIT 2: Yes, after copying the content of ~/CloverGrowerPro/toolchain to ~/src/opt/local I was able to build!

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

Hi Slice, UseUARTFrequency behavior is changed? Made a Skylake hack that was working well with this since last Clover update. Now seems not working well.

Thanks

Yes, changed. Now default is False. If you want to use it you should write explicitly in config.plist.

For most people it doesn't needed and leads to cracked sound.

 

I'm trying to build rev 3569 with CloverGrowerPro but the process is failing. I'm on 10.9.5 with Xcode 6.2. Can somebody help me?

Building ... /Users/administrator/CloverGrowerPro/edk2/MdePkg/Library/UefiUsbLib/UefiUsbLib.inf [X64]
[CC] VideoBiosPatchLib
/bin/sh: /Users/administrator/src/opt/local/bin/nasm: No such file or directory
make: *** [/Users/administrator/CloverGrowerPro/edk2/Build/Clover/RELEASE_GCC49/X64/MdePkg/Library/BaseLib/BaseLib/OUTPUT/X64/Thunk16.obj] Error 127


build.py...
 : error 7000: Failed to execute command
make tbuild [/Users/administrator/CloverGrowerPro/edk2/Build/Clover/RELEASE_GCC49/X64/MdePkg/Library/BaseLib/BaseLib]


build.py...
 : error F002: Failed to build module
/Users/administrator/CloverGrowerPro/edk2/MdePkg/Library/BaseLib/BaseLib.inf [X64, GCC49, RELEASE]

- Failed -
Build end time: 04:43:10, Jun.22 2016
Build total time: 00:00:13

Cloverx64 release  ERROR!!
EDIT: Copying /Users/administrator/CloverGrowerPro/toolchain/* to /Users/administrator/src/opt/local should work?

 

EDIT 2: Yes, after copying the content of ~/CloverGrowerPro/toolchain to ~/src/opt/local I was able to build!

 

Not a problem, look this line

/Users/administrator/src/opt/local/bin/nasm: No such file or directory

Take here

http://www.nasm.us/pub/nasm/releasebuilds/2.12.01/macosx/nasm-2.12.01-macosx.zip

unpack and place to the folder

Link to comment
Share on other sites
Riley Freeman
Riley Freeman
Posted
Posted

Thanks for the updated builds Sherlocks. Nice to have them while waiting for the next official release :)

 

I've moved to using AptioFix2Drv on my hacks and am back to having the occasional failed boot (error allocated pages yadda yadda). It's the reason I stayed on the original AptioFix, but now I'd rather move with the times.

 

Is there any data needed to try and make AptioFix2 more reliable? It would be great if it could be relied on to boot 100% of the time like the old version.

Link to comment
Share on other sites
insanelyDeepak
insanelyDeepak
Posted
Posted

hey @slice me too in queue CGP not working on myHack but on macMini perfect 

TOOLCHAIN_DIR: /Users/Deepak/CloverGrowerPro/toolchain
Initializing workspace
Loading previous configuration from /Users/Deepak/CloverGrowerPro/edk2/Conf/BuildEnv.sh
WORKSPACE: /Users/Deepak/CloverGrowerPro/edk2
EDK_TOOLS_PATH: /Users/Deepak/CloverGrowerPro/edk2/BaseTools
CONF_PATH: /Users/Deepak/CloverGrowerPro/edk2/Conf
Building tools as they are not found
/Applications/Xcode.app/Contents/Developer/usr/bin/make -C Source/C
Attempting to detect ARCH from 'uname -m': x86_64
Detected ARCH of X64 using uname.
mkdir -p .
/Applications/Xcode.app/Contents/Developer/usr/bin/make -C Common
gcc -Wno-deprecated-declarations  -c -MD -fshort-wchar -fno-strict-aliasing -Wall -Werror -Wno-deprecated-declarations -Wno-self-assign -nostdlib -c -g  -I .. -I ../Include/Common -I ../Include/ -I ../Include/IndustryStandard -I ../Common/ -I .. -I . -I ../Include/X64/  BasePeCoff.c -o BasePeCoff.o
dyld: Library not loaded: /Users/Mac2/CloverGrowerPro/toolchain/lib/libcloog-isl.4.dylib
  Referenced from: /Users/Deepak/CloverGrowerPro/toolchain/bin/../libexec/gcc/x86_64-apple-darwin15.5.0/4.9.3/cc1
  Reason: image not found
gcc: internal compiler error: Trace/BPT trap: 5 (program cc1)
libbacktrace could not find executable to open
Please submit a full bug report,
with preprocessed source if appropriate.
See <http://gcc.gnu.org/bugs.html> for instructions.
make[2]: *** [BasePeCoff.o] Error 4
make[1]: *** [Common] Error 2
make: *** [Source/C] Error 2
Cloverx64 release  ERROR!!


Ejecting RAM disk
"disk1" unmounted.
"disk1" ejected.
Link to comment
Share on other sites
Slice
Slice
Posted
Posted

 

hey @slice me too in queue CGP not working on myHack but on macMini perfect 

TOOLCHAIN_DIR: /Users/Deepak/CloverGrowerPro/toolchain
Initializing workspace
Loading previous configuration from /Users/Deepak/CloverGrowerPro/edk2/Conf/BuildEnv.sh
WORKSPACE: /Users/Deepak/CloverGrowerPro/edk2
EDK_TOOLS_PATH: /Users/Deepak/CloverGrowerPro/edk2/BaseTools
CONF_PATH: /Users/Deepak/CloverGrowerPro/edk2/Conf
Building tools as they are not found
/Applications/Xcode.app/Contents/Developer/usr/bin/make -C Source/C
Attempting to detect ARCH from 'uname -m': x86_64
Detected ARCH of X64 using uname.
mkdir -p .
/Applications/Xcode.app/Contents/Developer/usr/bin/make -C Common
gcc -Wno-deprecated-declarations  -c -MD -fshort-wchar -fno-strict-aliasing -Wall -Werror -Wno-deprecated-declarations -Wno-self-assign -nostdlib -c -g  -I .. -I ../Include/Common -I ../Include/ -I ../Include/IndustryStandard -I ../Common/ -I .. -I . -I ../Include/X64/  BasePeCoff.c -o BasePeCoff.o
dyld: Library not loaded: /Users/Mac2/CloverGrowerPro/toolchain/lib/libcloog-isl.4.dylib
  Referenced from: /Users/Deepak/CloverGrowerPro/toolchain/bin/../libexec/gcc/x86_64-apple-darwin15.5.0/4.9.3/cc1
  Reason: image not found
gcc: internal compiler error: Trace/BPT trap: 5 (program cc1)
libbacktrace could not find executable to open
Please submit a full bug report,
with preprocessed source if appropriate.
See <http://gcc.gnu.org/bugs.html> for instructions.
make[2]: *** [BasePeCoff.o] Error 4
make[1]: *** [Common] Error 2
make: *** [Source/C] Error 2
Cloverx64 release  ERROR!!


Ejecting RAM disk
"disk1" unmounted.
"disk1" ejected.

cd .... Clover

./ebuild.sh

Link to comment
Share on other sites
Slice
Slice
Posted
Posted

Clover doesn't load theme is set in config.plist , it always used randomly mostly embedded 

It set by NVRAM which has a priority.

Dynamic value are always have priority on static value.

Link to comment
Share on other sites
insanelyDeepak
insanelyDeepak
Posted
Posted

It set by NVRAM which has a priority.

Dynamic value are always have priority on static value.

Thanks i found it in NVRAM  

 

 

dyld: Library not loaded: /Users/Mac2/CloverGrowerPro/toolchain/lib/libcloog-isl.4.dylib

cd .... Clover

./ebuild.sh

i was using Toolchain folder from macMini  :hysterical: , CGP confused between Users and don't find directory 

Link to comment
Share on other sites
Micky1979
Micky1979
Posted
Posted

Yes, changed. Now default is False. If you want to use it you should write <true/> explicitly in config.plist.

For most people it doesn't needed and leads to cracked sound.

In fact that way works (streamed audio, normal was working anyway). Thanks

Link to comment
Share on other sites
Sherlocks
Sherlocks
Posted
Posted

It will great if we can patch DVMT with Clover :drool::drool:

 

I have HP Pavilio Pro 15-AC169TX and can't change DVMT with Shell Command.

 

I hope "super coders" can solve this issue :thumbsup_anim:

Yes. DVMT is important laptop in furture. Manufacture prevent bios mod(shell, linuxshell, mod). It needs for perfect work.

 

Maybe its hard work.

 

나의 LG-F410S 의 Tapatalk에서 보냄

Link to comment
Share on other sites
@ROBASEFR
@ROBASEFR
Posted
Posted

Rev 3561 - first Sierra compatible revision

 

Rev 3577

Legacy boot improvements.

 

Legacy boot is a process to boot OSes through 16bit instructions in real mode like it did old PC BIOS-based computer before Clover, before UEFI.

It was needed to boot Windows XP or Windows 7-32bit, old Linux or modern systems if the drive formatted to MBR Partition Table.

But Clover is already 32/64bit program working in EFI environment in protected mode. It can't get back to real mode. To do this we made a trick with loading PBR boot sector and start it in real mode.

It's a pity it is not always work because devices like SATA controller occur in undefined state. :(

 

There was three methods for Legacy boot before "LegacyBiosDefault", "PBRtest", "PBR" set by config.plist

	<key>Boot</key>
	<dict>
		<key>Legacy</key>
		<string>PBRtest</string>

"LegacyBiosDefault" assumed UEFI BIOS with CSM module capable to do LegacyBoot.

"PBR" is the most commonly used method for legacy Clover. I have one computer with PATA drive working with this method.

"PBRtest" is somehow different. May be someone will be happy with it.

 

Now I propose one more method "PBRsata". It is the same as "PBR" but making a SATA reset before boot. AFAIK the SATA controller is a main reason why legacy boot is not working.

Hope someone will be happy.

 

PS. The version is released.

Hello Slice,

 

I have EX58-UD5 and use "PBR" and Clover EFi 64bits BiosBlockIO, because i don't have UEFI...

My experiences with the latest Clover boot updates :

 

Clover r 3551 is ok for me for OSX 10.11.5,  but Clover versions > 3551 don'work !

 

Clover r3661 is ok for me for OSX 10.12, but Clover versions > 3561 don'work !

 

Why is this ?

 

Do you believe I could have better results with "PBRsata" option and do I have to change to Clover EFi 64bits SATA in the same time?

 

Thank you for your advice

Link to comment
Share on other sites
Slice
Slice
Posted
Posted

Hello Slice,

 

I have EX58-UD5 and use "PBR" and Clover EFi 64bits BiosBlockIO, because i don't have UEFI...

My experiences with the latest Clover boot updates :

 

Clover r 3551 is ok for me for OSX 10.11.5,  but Clover versions > 3551 don'work !

 

Clover r3661 is ok for me for OSX 10.12, but Clover versions > 3561 don'work !

 

Why is this ?

 

Do you believe I could have better results with "PBRsata" option and do I have to change to Clover EFi 64bits SATA in the same time?

 

Thank you for your advice

CloverEFi (boot6, boot7) lives own life other then CloverGUI (CloverX64.efi). "PBRsata" is a part of CloverX64.efi.

  • Like 1
Link to comment
Share on other sites
Micky1979
Micky1979
Posted
Posted

..

HID = Human Interface Device (?)

Shim = old API 

Are you using some old kexts like ApplePS2Controller (or some trackpad kexts)? Not sure but try to boot with no those kexts if any, and unplug everything (also the mouse/keyboard).... just to make a test..

Otherwise if means "_HID" ... ACPI problem with your DSDT.. and need a GURU

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