Jump to content

[Guide] Using DSDT with the Gigabyte GA-EP45-DS3L

blackosx

By blackosx,
in General Discussion

 Share
576 posts in this topic

Recommended Posts

mm67
mm67
Posted
Posted
Thanks for the clarification mm67.

 

But out of interest, comparing your original GA-EP45-UD3 acpidump DSDT.dsl to my original GA-EP45-DS3L acpidump DSDT.dsl, I find there isn't much difference between them. I have attached the DIFF result here.

ACPIdump_DSDT_Differences.txt.zip

 

Yes, should work if you take those PRT methods from your original dump. All other parts that have differences are already removed.

Link to comment
Share on other sites
blackosx
blackosx
Posted
  • Author
Posted
Yes, should work if you take those PRT methods from your original dump. All other parts that have differences are already removed.

I have taken your DSDT and made the appropriate changes for the GA-EP45-DS3L using the DIFF results from the linux acpidump. :angel:

 

Changes made are...

Amended Scope(_PR) for my processor - E7300 and changed style to match MasterChief's.

Commented out Device (WLAN)

Commented out Device (FRWR), and Method (_L1A....

Commented out PCI Routing Table data that's not in my original DSDT, and amended package number

Added Device (HDEF) for ALC888 (and Notify for HDEF in Method _L0D....

 

I will keep my eye on the Gigabyte DSDT fix thread for any differences/changes that develop. But for now it works great. Thanks mm67 :angel:

dsdt.dsl.zip

Link to comment
Share on other sites
thiagomorales
thiagomorales
Posted
Posted
I have taken your DSDT and made the appropriate changes for the GA-EP45-DS3L using the DIFF results from the linux acpidump. :ihw_wow:

 

Changes made are...

Amended Scope(_PR) for my processor - E7300 and changed style to match MasterChief's.

Commented out Device (WLAN)

Commented out Device (FRWR), and Method (_L1A....

Commented out PCI Routing Table data that's not in my original DSDT, and amended package number

Added Device (HDEF) for ALC888 (and Notify for HDEF in Method _L0D....

 

I will keep my eye on the Gigabyte DSDT fix thread for any differences/changes that develop. But for now it works great. Thanks mm67 :)

 

Hi Nick, recently tHe KiNg has released an new HDEF method, give a try (for ALC888, of course).

			Device (HDEF)
		 {
			 Name (_ADR, 0x001B0000)
			 Method (_PRW, 0, NotSerialized)
			 {
				 Return (Package (0x02)
				 {
					 0x0D, 
					 0x05
				 })
			 }

			 OperationRegion (HDCS, PCI_Config, 0x54, 0x04) // Power Management Control/Status (ICH9R-3169722.pdf / 17.1.19 / page 664).
			 Field (HDCS, DWordAcc, NoLock, Preserve)
			 {
					 ,   15, 
				 PMES,   1
			 }

			 Method (_DSM, 4, NotSerialized)
			 {
				 Store (Package (0x08)
					 {
						 "subsystem-id",
						 Buffer (0x04)
						 {
							 0xA0, 0x00, 0x00, 0x00
						 },

						 "subsystem-vendor-id",
						 Buffer (0x04)
						 {
							 0x6B, 0x10, 0x10, 0x10
						 },

						 "layout-id",
						 Buffer (0x04)
						 {
							 0x78, 0x03, 0x00, 0x00
						 },

						 "PinConfigurations",
						 Buffer (Zero) { }		
					 }, Local0)
				 MCDP (Arg2, RefOf (Local0))
				 Return (Local0)
			 }
		 }
	 }

 

It needs MCDP method:

Method (MCDP, 2, NotSerialized)											  // New Method V1.1 ? By Master Chief.
 {
	 If (LEqual (Arg0, Zero))												 // Function index: 0
	 {
		 Store (Buffer (One)
		 {
			 0x03
		 }, Arg1)
	 }
 }

Regards.

Link to comment
Share on other sites
blackosx
blackosx
Posted
  • Author
Posted
Hi Nick, recently tHe KiNg has released an new HDEF method, give a try (for ALC888, of course).

Hi thiagomorales

 

Yes, I know THe KiNG has released a revised HDEF method and I use it in my master DSDT that I have been working on with the aim of trying to match MasterChief's DSDT layout/structure. And the new HDEF code works great.

 

But as you have pointed out, it requires MasterChief's MCDP Method, which I didn't want to add to mm67's DSDT as at the moment nothing else in the DSDT uses it. Also as mm67's DSDT is so clean, I didn't want to add too much extra code to it.

 

Thanks, though. It's good to keep abreast of the newer techniques. :)

Link to comment
Share on other sites
mm67
mm67
Posted
Posted
Hi thiagomorales

 

Yes, I know THe KiNG has released a revised HDEF method and I use it in my master DSDT that I have been working on with the aim of trying to match MasterChief's DSDT layout/structure. And the new HDEF code works great.

 

But as you have pointed out, it requires MasterChief's MCDP Method, which I didn't want to add to mm67's DSDT as at the moment nothing else in the DSDT uses it. Also as mm67's DSDT is so clean, I didn't want to add too much extra code to it.

 

Thanks, though. It's good to keep abreast of the newer techniques. :)

 

If you are worried about the looks then you can define DSM like kdawg has done:

Method (_DSM, 4, NotSerialized)
{
Store (Package (0x08)
{
"subsystem-id" ,Buffer (0x01){0xA0, 0x00, 0x00, 0x00},
"subsystem-vendor-id" ,Buffer (0x04){0x6B, 0x10, 0x10, 0x10},
"layout-id" ,Buffer (0x01){0x0C, 0x00, 0x00, 0x00},
"PinConfigurations" ,Buffer (0x00){}
}, Local0)
DTGP (Arg0, Arg1, Arg2, Arg3, RefOf (Local0))
Return (Local0)
}

 

Looks pretty much like the style that I use :) And doesn't need method MCDP

Link to comment
Share on other sites
thiagomorales
thiagomorales
Posted
Posted
If you are worried about the looks then you can define it like kdawg has done:

Method (_DSM, 4, NotSerialized)
{
Store (Package (0x08)
{
"subsystem-id" ,Buffer (0x01){0xA0, 0x00, 0x00, 0x00},
"subsystem-vendor-id" ,Buffer (0x04){0x6B, 0x10, 0x10, 0x10},
"layout-id" ,Buffer (0x01){0x0C, 0x00, 0x00, 0x00},
"PinConfigurations" ,Buffer (0x00){}
}, Local0)
DTGP (Arg0, Arg1, Arg2, Arg3, RefOf (Local0))
Return (Local0)
}

 

Looks pretty much like the style that I use :) And doesn't need method MCDP

 

Great! I´ve update my DSDT with this, and get rid of MCDP method from LAN0 and HDEF.

Thanks!

 

				Method (_DSM, 4, NotSerialized)
			{
				Store (Package (0x08)
					{
						"subsystem-id" ,Buffer (0x04){0xA0, 0x00, 0x00, 0x00},
						"subsystem-vendor-id" ,Buffer (0x04){0x6B, 0x10, 0x10, 0x10},					
						"layout-id" ,Buffer (0x04){0x78, 0x03, 0x00, 0x00},					
						"PinConfigurations" ,Buffer (0x00){}								    
					}, Local0)
                   DTGP (Arg0, Arg1, Arg2, Arg3, RefOf (Local0))
                   Return (Local0)
               }

Link to comment
Share on other sites
jamonda
jamonda
Posted
Posted

I'm trying to follow you, but you are miles away from me...

Blackosx, isn't it time for a new EP45-DS3L DSDT.aml version? The last one is 11 days old! :)

As I said a few posts ago, a clean (mm67 like), fully commented file, with the new fixes/discoveries/styles (shutdown by mm67, HDEF by THe KiNG, DSM by kdawg) will be very welcome!!

Link to comment
Share on other sites
thiagomorales
thiagomorales
Posted
Posted

I´ve made some changes on DSDT posted by BlackOSX a few posts ago..:

- Using 'ThE KiNG' new HDEF injector for ALC888 using DTGP method (like mm67´s style)

- Change PR Scope to work with my Intel Core 2 Quad Q9550

- Added LAN0 Device

 

Model: GA-EP43-DS3L

 

I´ve saw errors UUID errors on console:

_CFGetHostUUIDString: unable to determine UUID for host. Error: 35

 

Regards

DSDT_mm67_thiagom.dsl.zip

Link to comment
Share on other sites
blackosx
blackosx
Posted
  • Author
Posted
If you are worried about the looks then you can define DSM like kdawg has done:

Thanks mm67

 

Blackosx, isn't it time for a new EP45-DS3L DSDT.aml version?

What was that I posted here then?

EDIT: I have added the latest version of this DSDT to the front page for use with the GA-EP45-DS3L.

 

I´ve made some changes on DSDT posted by BlackOSX a few posts ago..:

- Using 'ThE KiNG' new HDEF injector for ALC888 using DTGP method (like mm67´s style)

- Change PR Scope to work with my Intel Core 2 Quad Q9550

- Added LAN0 Device

 

Model: GA-EP43-DS3L

 

I´ve saw errors UUID errors on console:

_CFGetHostUUIDString: unable to determine UUID for host. Error: 35

Good work Thiago :unsure:

And thanks for adding comments too, as this keeps the file up to date.

Have you the PlatformUUID.kext in /E/E ? as without it, I also get the UUID error.

 

EDIT: I have added your changes to my DSDT.dsl and posted it on the front page. I did make changes which were...

• Revised the CST latencies back from the figures I was testing with.

• A general clean up for alignment/readability.

• Removed extra PCI Routing Table code from mm67's original, which I had previously commented out.

• Removed Device (FRWR), which I had previously commented out.

• Removed Device (WLAN), which I had previously commented out.

 

Note: When compiling with the new HDEF DTGP method, I receive 1 Remark (Effective AML buffer length is zero ^ ) when compiling? It works fine though.

Link to comment
Share on other sites
mm67
mm67
Posted
Posted
Thanks mm67

 

 

What was that I posted here then?

 

 

Good work Thiago :unsure:

And thanks for adding comments too, as this keeps the file up to date.

Have you the PlatformUUID.kext in /E/E ? as without it, I also get the UUID error.

 

No need for PlatformUUID.kext if you use modified PC EFI 10.5 by Asere.

 

boot.zip

Link to comment
Share on other sites
mm67
mm67
Posted
Posted
So at last, I can get rid of PlatformUUID.kext?

I try to keep on top of all the latest developments but I never knew about this one.

Thanks again mm67. :unsure:

 

I don't think that very many people have heard of this version, I have been using it for couple of weeks. My kexts are now down to fakesmc, OSXRestart and IOAHCIBlockStorageInjector.

Link to comment
Share on other sites
blackosx
blackosx
Posted
  • Author
Posted
I don't think that very many people have heard of this version, I have been using it for couple of weeks. My kexts are now down to fakesmc, OSXRestart and IOAHCIBlockStorageInjector.

Yes, I have just read the forum where this was posted.

So it must be a fork of Chameleon RC3 as I don't think Netkas has ever released the source for PC-EFI. The next version of Chameleon will have more changes/additions etc, I wonder if this UUID idea will make it in?

Link to comment
Share on other sites
jamonda
jamonda
Posted
Posted
Thanks mm67

 

 

What was that I posted here then?

EDIT: I have added the latest version of this DSDT to the front page for use with the GA-EP45-DS3L.

 

 

Good work Thiago :)

And thanks for adding comments too, as this keeps the file up to date.

Have you the PlatformUUID.kext in /E/E ? as without it, I also get the UUID error.

 

EDIT: I have added your changes to my DSDT.dsl and posted it on the front page. I did make changes which were...

• Revised the CST latencies back from the figures I was testing with.

• A general clean up for alignment/readability.

• Removed extra PCI Routing Table code from mm67's original, which I had previously commented out.

• Removed Device (FRWR), which I had previously commented out.

• Removed Device (WLAN), which I had previously commented out.

 

Note: When compiling with the new HDEF DTGP method, I receive 1 Remark (Effective AML buffer length is zero ^ ) when compiling? It works fine though.

 

woohoo! does this new version include the shutdown fix by mm67?

Link to comment
Share on other sites
thiagomorales
thiagomorales
Posted
Posted
woohoo! does this new version include the shutdown fix by mm67?

 

Yes. It´s include new PTS method that fix shutdown on GB.

 

No need for PlatformUUID.kext if you use modified PC EFI 10.5 by Asere.

 

Well, I did not use any patch or kext to fix the problem of UUID. In DSDT.aml old (was using the DSDT stickping modified from: http://www.insanelymac.com/forum/index.php...p;#entry1355552) I saw no error 'UUID' in console ..

I'll see how we can fix the 'UUID' for DSDT without using boot kext or modified.

 

Regards.

Link to comment
Share on other sites
jamonda
jamonda
Posted
Posted
I don't think that very many people have heard of this version, I have been using it for couple of weeks. My kexts are now down to fakesmc, OSXRestart and IOAHCIBlockStorageInjector.

 

 

Is it necessary to add any key to boot.plist in order use this UUID detection functionality?

Link to comment
Share on other sites
mm67
mm67
Posted
Posted
Is it necessary to add any key to boot.plist in order use this UUID detection functionality?

 

No, no keys are needed there or in smbios.plist, it uses Gigabyte's serialnumber to generate UUID key if you don't define something else in smbios.plist.

Link to comment
Share on other sites
jamonda
jamonda
Posted
Posted
No, no keys are needed there or in smbios.plist, it uses Gigabyte's serialnumber to generate UUID key if you don't define something else in smbios.plist.

 

 

You mean I can exclude the HD UID number I put in UUID kext plist and smbios.plist?

 

Netkas describes something about UUID for PC EFI 10.5. Only the modified version of his boot file (by Asere) will make UUID kext unnecessary?

 

Thanks.

Link to comment
Share on other sites
mm67
mm67
Posted
Posted
You mean I can exclude the HD UID number I put in UUID kext plist and smbios.plist?

 

Netkas describes something about UUID for PC EFI 10.5. Only the modified version of his boot file (by Asere) will make UUID kext unnecessary?

 

Thanks.

 

Yes, you don't need to add UUID codes anywhere when you use this modified PC EFI

Link to comment
Share on other sites
jamonda
jamonda
Posted
Posted

The last DSDT.dsl is working good with my EP45-UD3L. I changed only the CPU parameters. I`m also using the boot file of post #386. Now I have only MC OSXRestart, Netkas fakesmc, LegacyHDA and OrangeIconFix kexts in /E/E. No need for PlatformUUID.kext, nor UUID string in SMBios.plist.

Thank you very much fellows.

Link to comment
Share on other sites
thiagomorales
thiagomorales
Posted
Posted (edited)

I did some checks and updated DSDT Device (LAN0) that was in the wrong place. I also added "device type" in it.

 

Make a test with this new DSDT and see if the UUID errors will disappear. (I'm at work atm)

 

Regards.

blackosx_101209_dsdt_revised_thiagom.zip

Edited by thiagomorales
Link to comment
Share on other sites
blackosx
blackosx
Posted
  • Author
Posted
The last DSDT.dsl is working good with my EP45-UD3L. I changed only the CPU parameters. I`m also using the boot file of post #386. Now I have only MC OSXRestart, Netkas fakesmc, LegacyHDA and OrangeIconFix kexts in /E/E. No need for PlatformUUID.kext, nor UUID string in SMBios.plist.

Thank you very much fellows.

Okay.

 

So you are running the EP45-UD3L? I thought you were running this motherboard? It would help if you add a signature so we can all see what hardware you're running.

 

Have you started using the DSDT I posted? as mm67's original DSDT is for the EP45-UD3 and I amended it for the EP45-DS3L by removing some PCI Routing Table data. If you are using the one I posted then you are missing that code for your board. You must have been reading the last couple of pages as you were posting in them? Go back to this post and read onwards to see what we did. Especially this bit...

Yes, should work if you take those PRT methods from your original dump. All other parts that have differences are already removed.
Your system might run as it is, but it makes sense to have the DSDT contain the original code for your motherboard.

 

 

I did some checks and updated DSDT Device (LAN0) that was in the wrong place. I also added "device type" in it.

 

Make a test with this new DSDT and see if the UUID errors will disappear. (I'm at work atm)

I didn't check your placement of the LAN0 code, I just took it that you would have it right. But well spotted ;)

It had a compile error which I fixed; the method _DSM Store package value needed to read 0x06, though I still get UUID errors with it.

 

But I have an interesting puzzle at the moment, in that my ethernet is working without any LAN0 code in DSDT. I did once start using the <key>EthernetBuiltin</key> in my com.apple.Boot.plist but I have removed that. I have just rebooted using a very old DSDT without networking and again my ethernet is working?

Link to comment
Share on other sites
thiagomorales
thiagomorales
Posted
Posted
Okay.

I didn't check your placement of the LAN0 code, I just took it that you would have it right. But well spotted ;)

It had a compile error which I fixed; the method _DSM Store package value needed to read 0x06, though I still get UUID errors with it.

 

But I have an interesting puzzle at the moment, in that my ethernet is working without any LAN0 code in DSDT. I did once start using the <key>EthernetBuiltin</key> in my com.apple.Boot.plist but I have removed that. I have just rebooted using a very old DSDT without networking and again my ethernet is working?

 

I updated the 'DSM Package Store' in attachment. Thanks.

I´m trying to fix UUID errors through DSDT..I believe that UUID errors is caused because system can not find the built-in ethernet..The Device LAN0 works with old DSDT in same place, and i did get any UUID errors.

 

When i´m at home i´ll try to look on my old DSDT..

Regards.

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