DSDT fixes for Gigabyte boards

[Master Chief]

@Chief: this is because the PIC mode has been removed, right?

Spot on!

 

 

Hi again! Just want to check one thing:

 

I've got Legacy kexts in /Extra/Extensions, and IIRC I've been able to see them when I run kextstat | grep Legacy. Now I don't. Does this mean they aren't loaded at all?

 

These kexts include LegacyHDA, but also LegacyAGPM and LegacyACPI_SMC_PP.

 

kextstat | grep AppleIntel gives me only one CPU-type though (Penryn), and the SMbus devices are there, three of them. This shows that SBUS and EC are working correctly, right?

 

I'm just a bit confused, please unconfuse me.

What Beerkex'd said. And look at the index of kextstat and note the missing numbers... guess what

[William Parker]

@ Master Chief

 

Well I'm pretty close to throwing in the towel here unless some kind soul steps in to un-greek this thread.

 

@ keeza

 

Friend long time no see.

 

@mm67

 

Tried to follow your stripped 3 dsl but got lost. Attaching the last dsl I managed to patch properly & found working. Will be grateful if you could have a peek & if possible mod it.

 

Thank you all.

DSDT.dsl.zip

[xopher]

look at the index of kextstat and note the missing numbers... guess what

 

Aahaa!

[THX]

Can't see any major problems with it, boots just fine for me. Looks like you still have some more modifications to do.

Thanks!

I've stripped a USB device sections (EHCI, UHCI, UHC1-6) as you did in you dsdt - works for me now. But with other devices, such as EC, SBUS i'm totally stuck. IORegistryExplorer show me PX43 device under PCI0@0 section, seems it's a SBUS (i can rename PX43 to SBUS). I've tried to rename it according to you last stripped dsdt version, but lost IEEE1394

 

dsdt_v.1.12.zip

[iSoprano]

Could someone help me in letting me know what is my address for graphics card

 

lspci o/p

 

 

01:00.0 VGA compatible controller [0300]: nVidia Corporation G94 [GeForce 9600 GT] [10de:0622] (rev a1)

 

i.e. what should I put here

 

Device (POP2)

{

Name (_ADR, ?????????????)

}

 

Spot on!

 

 

 

What Beerkex'd said. And look at the index of kextstat and note the missing numbers... guess what

 

 

@chief, looks like more drivers are being loaded...it is good isn't it?

[mm67]

Could someone help me in letting me know what is my address for graphics card

 

lspci o/p

 

 

01:00.0 VGA compatible controller [0300]: nVidia Corporation G94 [GeForce 9600 GT] [10de:0622] (rev a1)

 

i.e. what should I put here

 

Device (POP2)

{

Name (_ADR, ?????????????)

}

 

Look for device PEGP in this.

dsdt.dsl.zip

[FKA]

Could someone help me in letting me know what is my address for graphics card

 

lspci o/p

 

 

01:00.0 VGA compatible controller [0300]: nVidia Corporation G94 [GeForce 9600 GT] [10de:0622] (rev a1)

 

i.e. what should I put here

 

Device (POP2)

{

Name (_ADR, ?????????????)

}

 

 

 

            Device (PEGP)
           {
               Name (_ADR, 0x00010000)
               Method (_STA, 0, NotSerialized)
               {
                   Return (0x0F)
               }

               Method (_PRW, 0, NotSerialized)
               {
                   Return (Package (0x02)
                   {
                       0x09, 
                       0x05
                   })
               }

               Device (GFX0)
               {
                   Name (_ADR, Zero)
                   Method (_DSM, 4, NotSerialized)
                   {
                       Store (Package (0x18)
                           {
                               "@0,compatible", 
                               Buffer (0x0B)
                               {
                                   "NVDA,NVMac"
                               }, 

                               "@0,device_type", 
                               Buffer (0x08)
                               {
                                   "display"
                               }, 

                               "@0,name", 
                               Buffer (0x0F)
                               {
                                   "NVDA,Display-A"
                               }, 

                               "@1,compatible", 
                               Buffer (0x0B)
                               {
                                   "NVDA,NVMac"
                               }, 

                               "@1,device_type", 
                               Buffer (0x08)
                               {
                                   "display"
                               }, 

                               "@1,name", 
                               Buffer (0x0F)
                               {
                                   "NVDA,Display-B"
                               }, 

                               "NVCAP", 
                               Buffer (0x18)
                               {
                                   /* 0000 */    0x04, 0x00, 0x00, 0x00, 0x00, 0x00, 0x0B, 0x00, 
                                   /* 0008 */    0x04, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x07, 
                                   /* 0010 */    0x00, 0x00, 0x00, 0x00
                               }, 

                               "NVPM", 
                               Buffer (0x1C)
                               {
                                   /* 0000 */    0x01, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 
                                   /* 0008 */    0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 
                                   /* 0010 */    0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 
                                   /* 0018 */    0x00, 0x00, 0x00, 0x00
                               }, 

                               "VRAM,totalsize", 
                               Buffer (0x04)
                               {
                                   0x00, 0x00, 0x00, 0x40
                               }, 

                               "device_type", 
                               Buffer (0x0D)
                               {
                                   "NVDA,Parent"
                               }, 

                               "model", 
                               Buffer (0x17)
                               {
                                   "nVidia GeForce 9500 GT"
                               }, 

                               "rom-revision", 
                               Buffer (0x2B)
                               {
                                   "nVidia GeForce 9500 GT OpenGL Engine [EFI]"
                               }
                           }, Local0)
                       DTGP (Arg0, Arg1, Arg2, Arg3, RefOf (Local0))
                       Return (Local0)
                   }
               }
           }

[THX]

Whoops... i've continued to stripping my dsdt and get kernel panic on system boot. Can somebody be so kind to look into this

 

dsdt.dsl.zip

[Matthew L.]

Hmm, I could finally catch up on patching. Well, everything mentioned so far is done, but I'm having sound assertion errors at boot, which are related to mikey:

10/22/09 6:01:10 PM	kernel	Sound assertion "0 == mMikeyInterruptSource" failed in "/SourceCache/AppleHDA/AppleHDA-174.1.1/AppleMikeyDriver/AppleMikeyACPI.cpp" at line 260 goto Exit
10/22/09 6:01:10 PM	kernel	Sound assertion "0 != result" failed in "/SourceCache/AppleHDA/AppleHDA-174.1.1/AppleMikeyDriver/AppleMikeyI2C.cpp" at line 197 goto handler
10/22/09 6:01:10 PM	kernel	Sound assertion "mMikeyReceiverFound && 0 != result" failed in "/SourceCache/AppleHDA/AppleHDA-174.1.1/AppleMikeyDriver/AppleMikeyI2C.cpp" at line 247 goto Exit
10/22/09 6:01:10 PM	kernel	Sound assertion "0 != readRegisters ( (UInt8*)&mRegisterCache )" failed in "/SourceCache/AppleHDA/AppleHDA-174.1.1/AppleMikeyDriver/AppleMikeyI2C.cpp" at line 104 goto handler

 

Anyone has a clue?

I've attached my dsdt.dsl, which should be mostly identical to the current dsdt in post 1.

 

Thanks Master Chief, iSoprano, and mm67 for discovering great things!

 

Update: Auto-sleep is dead, too! I'm seeing constant HDD-led blinks, which indicates that OS X is sniffing the drive for something by a pre-defined repeat interval. (And no, I'm not using a TSSTCorp drive.)

Source_R.dsl.zip

[ccg]

@ Master Chief

 

My motherboard is a Giga-Byte EP45-DS3R (ICH10). I have a modified DSDT will all of the necessary mods for SL (HPET, CMOS reset, RTC, etc.) with EHCI, video, audio, LAN fixes.

 

After reading this thread, I am really confused.

 

Please, if you can find the time to review my DSDT and make the necessary changes.

 

Thanks very much,

ccg

DSDT_EP45_DS3R.8800GT.zip

[blackosx]

After a couple of days reading (trying to learn) an testing I have managed to complete the USB part of this guide. So I am now looking for the next part of this, which I guess will be Speedstepping, cleaning up my current DSDT, and anything else I can gather from here.

 

But out of interest, I have just thrown iSoprano's DSDT - Update 3, from the front page, at my system and it works great (apart from the orange icons which I can fix), with speedstepping too. But that is built for his ICH9 system? I am surprised it works as I thought DSDT's should be tailored for each individual system and I shouldn't really use anybody else's.. Maybe I am wrong?

[mitch_de]

Hi, i used the latest (22/10/2009) dsdt from post one on my GA-EP-35 DS3, Modell iMac5,1

No problems compared to my also working own dsdt.

 

But main thing i want to fix wasnt fixed : SLEEP does not work !!!!

PC try to shut down but fans + mainboards stays on (10.5.8, vanilla).

I dont think its an usb/ehci problem, because same with my own dsdt and that worked until 10.5.7.

Also no usb/ehci probs to see in the system.log (not with yours / not with mine).

 

HINT: I would excude the C-States+Pstates CPU part of all shared dsdt.aml !!!

If someone has an complete different CPU (not Q6600) at least non voodoopower based (better)speedstep(=Pstate) tools like Mark-i, which uses AHCI or vodoopower based with UseAHCI flag may produce KPs by wrong Volts/Pstate settings . Same can happen if using AppleIntelCPUPM (=Cstate)!!

 

I used my own Ctstate+Pstate part from my C2D and all working - excpet SLEEP.

[blackosx]

Hi mitch_de

 

With my own DSDT, sleep, auto sleep, USB, EHCI, UHCI worked fine. And with the tweaks from the 1st part of this thread (Master Chief's USB fixes) I could wake my system with a mouse & keyboard.

 

Surprisingly (because it's for a EP35-DS3) iSoprano's DSDT also gives me the same features, although I added to the Device (SATA) section with this fix. (It has already been mentioned, except for Device (PRIM), in this thread by xopher).

Device (SATA)
           {
               Name (_ADR, 0x001F0002)
               Method (_DSM, 4, NotSerialized)
               {
                   Store (Package (0x02)
                       {
                           "device-id", 
                           Buffer (0x04)
                           {
                               0x81, 0x26, 0x00, 0x00
                           }
                       }, Local0)
                   DTGP (Arg0, Arg1, Arg2, Arg3, RefOf (Local0))
                   Return (Local0)
               }
           }

 

My SMBIOS currently has SMproductname set to MacPro1,1. But I did read in this thread that Master Chief aims to build his DSDT's to be as close as possible to a MacPro3,1. So Maybe I can look at changing that.

 

But I take your point that iSoprano uses a Q6600 and I use E7300 which are completely different. I don't want to cause any damage to my system (is that a possibility?). So I will use his DSDT for comparison purposes to continue building mine with.

 

Thanks

[milanca]

Ripped out a little too much? Here's mine:

   30	2 0x555fd000 0x3000	 0x2000	 com.apple.iokit.IOSMBusFamily (1.1) <5 4 3>
  67	0 0x5ca45000 0x2000	 0x1000	 com.apple.driver.AppleSMBusPCI (1.0.2d0) <14 5 4 3>
  79	0 0x5536a000 0x9000	 0x8000	 com.apple.driver.AppleSMBusController (1.0.2d0) <30 14 13 5 4 3>

Concentrate on the Devices SBUS/EC for a starter.

 

Thanks Master Chief Not too much, more likely didn't add everything. Fixed all of it so far.

 

Here is the smbus stat

28	2 0xffffff7f80713000 0x2000	 0x2000	 com.apple.iokit.IOSMBusFamily (1.1) <5 4 3>
55	0 0xffffff7f80885000 0x2000	 0x2000	 com.apple.driver.AppleSMBusPCI (1.0.2d0) <9 5 4 3>
64	0 0xffffff7f80ff6000 0xb000	 0xb000	 com.apple.driver.AppleSMBusController (1.0.2d0) <28 9 8 5 4 3>

 

Thanks for your help bud, i really appreciate it. Latest hint was really great, removing all the PIC bytes, PICM,PIC1-5 and _PRT returns only APIC code. Got rid of IFs there, BUF and entire section LNKA-F, LNK0-1 required for PIC; and with previous tweaks OSFL, _INI methods my compiled .aml is ~13kb. I also removed bunch of unneeded devices like floppy, parallel, serial, mem, etc. System works just fine so far including sleep and wake.

 

Higher CPU temp. problem remained. I think speedstepping code is wrong, or something related to it (it used to work fine before these new tweaks, its the same code from iSoprano). Now bottom multiplier sits at x7 instead of x6, Q6600 native are x6 and x9.

 

@mitch_de,

 

i also tried iSoprano dsdt and it worked fine for me except the sleep. His board is an L variant so it works for him but doesn't work on our board, EP35-DS3. I managed to get it sleep and wake with Master Chief's tweaks, actually even some more than just sleep/wake, getting closer to MacPro3,1. Here is my latest DSDT.

dsdt.aml.zip

[Master Chief]

@ Master Chief

 

Well I'm pretty close to throwing in the towel here unless some kind soul steps in to un-greek this thread.

Repeat after me: "I will NOT quit". And again: "I will NOT quit".

 

p.s. I am busy with follow ups, so I'm afraid that I don't have time for this at the moment. Sorry, but that's the way it is.

[Master Chief]

Hmm, I could finally catch up on patching. Well, everything mentioned so far is done, but I'm having sound assertion errors at boot, which are related to mikey:

 

<snip />

 

Update: Auto-sleep is dead, too! I'm seeing constant HDD-led blinks, which indicates that OS X is sniffing the drive for something by a pre-defined repeat interval. (And no, I'm not using a TSSTCorp drive.)

Oh my ... I helped you to get rid of the junk, to get Speedstep going, and you had the most perfect PR scope on this entire forum, but you copied the junk back in You fool Fix that please!

 

BTW: You can simply comment out the mikey device... to prevent the errors in kernel.log until we are ready for it.

[tapper00`]

First, I would like to thank Master Chief for taking the time to help those of us with Gigabyte boards. After seeing what you have done with the Asus P5K boards, I look forward to your knowledge and assistance. I know you don't personally have one (Gigabyte Board). If I could help with a fund.... count me in. If enough Gigabyte users are willing to chip in we could get you an EP35-DS3P on eBay for about $70 shipped.

 

Secondly, @mm67 I remember reading in the vanilla speedstep thread that you had speedstep working with only the CST info added to your DSDT, but I notice that your latest DSDT has PCT and P-states added. How is speedstep working for you? Mainly interested in c states (temps and voltage while at idle).

 

Thanks all in advance!!

[William Parker]

Repeat after me: "I will NOT quit". And again: "I will NOT quit".

 

p.s. I am busy with follow ups, so I'm afraid that I don't have time for this at the moment. Sorry, but that's the way it is.

 

 

Yes sir, I wish it were that easy.

The computer runs DTP apps that are part of my bread & butter. Besides good temps & wake from mouse should keep me happy. This is still work in progress, & I trust you will sometime in future, whenever you find time, help my dsdt. I will wait.

Meantime I am exploring avenues of progress in other sections, namely boot screen. White screen followed by apple logo followed by the whirly & desktop. That's how Apple does it. PC MBs fill in all sorts of gibberish in between which is not at all elegant. A diagnostic screen may be verbose, but why should it become an eyesore every time I use my computer. I am planning to employ professional services of a Bios Engineer to achieve this.

My best wishes to you - Happy cracking.

[xjasx]

White screen followed by apple logo followed by the whirly & desktop. That's how Apple does it. PC MBs fill in all sorts of gibberish in between which is not at all elegant. A diagnostic screen may be verbose, but why should it become an eyesore every time I use my computer. I am planning to employ professional services of a Bios Engineer to achieve this.

 

that would be verbose mode, remove the -v flag from com.apple.Boot.plist

[mitch_de]

"...speedstep working with only the CST info added to your DSDT, but I notice that your latest DSDT has PCT and P-states added...:"

For my knowdledge CState und PState Powermanagement are different. speedstep (voodoopower, mark-i) use Pstates NOT CStates for throttling.

So you dont need to have Cstate things in the dsdt - i also havent them and vodoopower (useAHCI) P-64 or mark-i worked useing that dsdt Pstates.

Cstates PM are used by Apples AppleCPUPowermangement, which i disabled / not use.

I am not 100% sure what i talked about (above) but 98%

 

So normally working PStates is enough to get an good powermanagement. I would not make to much work for Ctstates (with AppleIntelCPUPowermanagement).

 

Thanks to all here for so much detail informations.

 

Also thanks for Info from other user that in the post #1 the Mainboard is not EP-35-DS3 its an EP-35DS3L.which is a bit different. I was a bit ??? / thinked that because of using 888 insted of 889A at HDEF.

[William Parker]

that would be verbose mode, remove the -v flag from com.apple.Boot.plist

 

Oh dear! Surely I must know that. That is not what I am talking about.

Gigabyte boards have no way of hiding the AHCI device detection screen, PCI device list screen & "Verifying DMI pool data" screens. This is also followed by a small flashing of few lines from Netkas's 10.5 PC-EFI booter.

Except for the latter part all else is BIOS related & have nothing to do with any OS. It is easy to replace the Boot screen which I have. But the other things, well another story.

My apologies for getting off topic here. Anyone wishing to discuss this issue please open a new thread & desist from further discussions here or please wait until I have some good news.

So good friend, any ideas? Please PM me.

[mm67]

First, I would like to thank Master Chief for taking the time to help those of us with Gigabyte boards. After seeing what you have done with the Asus P5K boards, I look forward to your knowledge and assistance. I know you don't personally have one (Gigabyte Board). If I could help with a fund.... count me in. If enough Gigabyte users are willing to chip in we could get you an EP35-DS3P on eBay for about $70 shipped.

 

Secondly, @mm67 I remember reading in the vanilla speedstep thread that you had speedstep working with only the CST info added to your DSDT, but I notice that your latest DSDT has PCT and P-states added. How is speedstep working for you? Mainly interested in c states (temps and voltage while at idle).

 

Thanks all in advance!!

 

Nope, I still have only CST added to my dsdt, P-states and and PCT are vanilla from the board itself. I just fixed the value of CFGD so that correct PSS tables are loaded from bios.

[mitch_de]

Thanks Master Chief

Here is the smbus stat

28	2 0xffffff7f80713000 0x2000	 0x2000	 com.apple.iokit.IOSMBusFamily (1.1) <5 4 3>
55	0 0xffffff7f80885000 0x2000	 0x2000	 com.apple.driver.AppleSMBusPCI (1.0.2d0) <9 5 4 3>
64	0 0xffffff7f80ff6000 0xb000	 0xb000	 com.apple.driver.AppleSMBusController (1.0.2d0) <28 9 8 5 4 3>

@mitch_de,

i also tried iSoprano dsdt and it worked fine for me except the sleep. His board is an L variant so it works for him but doesn't work on our board, EP35-DS3. I managed to get it sleep and wake with Master Chief's tweaks, actually even some more than just sleep/wake, getting closer to MacPro3,1. Here is my latest DSDT.

Thanks.

One more thing

Which .kext you are using like OpenHaltrestart and sleepenabler ?

Maybe also on my system is (even dsdt 100% perfect) an sleep problem with some HD(3, 2 SATA, 1 IDE>SATA adapter or my DVD(SATA)). I heared that sometimes beside USB/EHCI also hardware devices may be the reson for sleep problems.

 

My kextstat (onyl those which are "of ACPI or dsdt Interest" listed, HPET,RTC, HDA loaded but lot shown here.

ami$ kextstat
I[size=1]ndex Refs Address    Size       Wired      Name (Version) <Linked Against>
   1    1 0x0        0x0        0x0        com.apple.kernel (9.8.0)
....
   19    3 0x54825000 0x3e000    0x3d000    com.apple.driver.AppleACPIPlatform (1.2.5) <18 17 12 7 5 4>
  21    0 0x54879000 0x2000     0x1000     com.psystar.driver.[b]OpenHaltRestart[/b] (1.0.3) <12>
   24    0 0x549c2000 0x2000     0x1000     sk.triaxis.kext.[b]SleepEnabler[/b] (1.0.0) <7 5 4>
  30    0 0x5a761000 0x4000     0x3000     com.apple.driver.[b]AppleACPIButtons[/b] (1.2.5) <23 19 18 7 6 5 4 2>
  31    0 0x5a80f000 0x3000     0x2000     com.apple.driver.[b]AppleACPIPCI[/b] (1.2.5) <19 18 17 12>
  32    2 0x5a875000 0x3000     0x2000     com.apple.iokit.IOSMBusFamily (1.1) <6 5 4>
  33    0 0x5a8d3000 0x5000     0x4000     com.apple.driver.[b]AppleACPIEC[/b] (1.2.5) <32 19 18 12>
   61    2 0x54723000 0xa000     0x9000     com.apple.driver.[b]AppleSMC[/b] (2.3.1d1) <18 7 6 5 4>
  62    1 0x5472d000 0xe000     0xd000     com.apple.driver.[b]IOPlatformPluginFamily[/b] (3.4.0a17) <12 4>
  63    0 0x5473b000 0xf000     0xe000     com.apple.driver.[b]ACPI_SMC_PlatformPlugin[/b] (3.4.0a17) <62 61 18 17 12 7 6 5 4>
  65    0 0x54760000 0x5000     0x4000     com.psystar.driver.RealtekR1000 (1.8.1) <64 17 7 6 5 4 2>
  66    0 0x54765000 0x3000     0x2000     com.apple.driver.[b]AppleLPC[/b] (1.3.1) <17 6 5 4>
  67    0 0x5a5f9000 0x2000     0x1000     com.apple.driver.[b]AppleSMBusPCI[/b] (1.0.1d1) <6 5 4>
 78    0 0x5a964000 0x9000     0x8000     com.apple.driver.[b]AppleSMBusController[/b] (1.0.1d1) <32 18 17 6 5 4>[/size]

No AppleIntelCPU (disabled), iMac5,1 Modell

[Matthew L.]

Oh my ... I helped you to get rid of the junk, to get Speedstep going, and you had the most perfect PR scope on this entire forum, but you copied the junk back in You fool Fix that please!

 

BTW: You can simply comment out the mikey device... to prevent the errors in kernel.log until we are ready for it.

 

Hehe, thanks.

 

The reason for using the alternate way is the call for PNOT in _WAK and EC. I haven't had enough PDCs for PNOT, only a PDC0 defined in CPU0, nothing more. Should I just send the PNOT method and all of its callers to hell, so I can restore my old _PR scope?

 

What 'bout the dead auto-sleep?

 

BTW, it's just the message that bothers me about mikey, no downsides in using the OS, so it stays as it is.

 

Thanks in advance!

 

Update: Hmm... Today OS X said that he won't bother the DVD drive, so will do auto sleep. (I'm sure that was the case earlier too, but I've started checking every new DSDT.aml build with sleep and wake with keyboard/mouse/power button.) Yea, it did, but after wake is when it's started thinking otherwisely. All in all, first sleep works, then wake makes me enjoy the infamous no-sleep-coz-of-optical-drive.

 

Update 2: Just checked reboot and shutdown w/o OHR kext, and only the latter (shutdown) works.

[spacr]

Anyone know if it's safe to use the Dsdt from update 2 on the ep45 board? I'm using the version 11 beta bios. I've tried to compare it with my current one but I'm not sure what to add. If someone out there wants to upload a compatible dsdt I would be extremely grateful. If not then any advise would be appreciated as well.