Jump to content

DSDT for Asus P8P67-M PRO

Time2Retire

By Time2Retire,
in Patches

 Share
834 posts in this topic

Recommended Posts

Time2Retire
Time2Retire
Posted
  • Author
Posted
I did a test-run of Prime95, does it mean anything if the temp of all core goes from 41 to 51°C?

Turbo ratio active?

 

Found out that 0xFC in PSD is causing this, 0xFE is working fine.

Ouch! I said to be using 0xFE remember?

 

Wow. The cooler arrived. What a big thing. Difficult to install. Heck even reading the installation instruction LOL

Link to comment
Share on other sites
flAked
flAked
Posted
Posted
Ouch! I said to be using 0xFE remember?

Apparently, 0xFC works as well if I use PSD with just the one Package definition (had the old one with two).

 

But 0xFC is used in the iMac ACPI tables?

 

 

Aha, so 0xFC does work with the iMac12,2 definition, but not MBP8,3 (maybe because SMCPDRC is not matched on iMac12,2 and Power Status is not there, so something is not loading on iMac12,2).

Did I say it's driving me nuts? It worked for one boot *g

 

 

Have you seen this? http://www.tonymacx86.com/viewtopic.php?f=79&t=20181

The FSB is still wrong, it's showing 5000Mhz with this new release.

Link to comment
Share on other sites
Time2Retire
Time2Retire
Posted
  • Author
Posted
Apparently, 0xFC works as well if I use PSD with just the one Package definition (had the old one with two).

 

But 0xFC is used in the iMac ACPI tables?

I see both 0xFC and 0xFE are there, but it is hard to tell what is used without knowing the capability bits. We need an MSR report to see what it is using.

 

Have you seen this? http://www.tonymacx86.com/viewtopic.php?f=79&t=20181

The FSB is still wrong, it's showing 5000Mhz with this new release.

Thanks. I didn't. Went there to make a post about it.

 

Still. It is good to see that folks with Asrock, Intel and MSI motherboards can now finally boot with a Chameleon branch, thanks to bcc9 for running into that issue.

 

I however don't trust their AICPUPM deal. Looks very wrong to me. Why would anyone want to use a MacPro3,1 with a Sandy Bridge setup? And OC'ing and AICPUPM are a no go anyway. Just look at the turbo ratio... with may show DEF0 instead of valid values. Just one of the things that people forget about :D

 

And how exactly was speed stepping (p-states) verified to work correctly?

Link to comment
Share on other sites
flAked
flAked
Posted
Posted
And how exactly was speed stepping (p-states) verified to work correctly?

Exactly! Thats the kicker.

 

I compiled the new HWsensors package (with IntelCPUMonitor) but no changes, still no frequency showing up via SMC.

 

We need a reliable tool that is up to date for SB.

Link to comment
Share on other sites
flAked
flAked
Posted
Posted
Seems like Turbo ratio's are in fact working. Sweet. However. Without AICPUPM loaded my Geekbench / Cinebench scores are much higher.

How do you test it?

 

I made a kernel extension that writes the current CPUfreq into the IORegistry, but I always get 331xxxx something, not the real frequency.

 

I used the code from dynamic_data.h so I have tscFrequency and the others available.

Link to comment
Share on other sites
Time2Retire
Time2Retire
Posted
  • Author
Posted
How do you test it?

I disabled "Turbo Ratio" in the BIOS to get a base score in Geekbench and Cinebench. After this I enabled turbo again and set all four multipliers to the same value. Starting with 1 and so on. Showing me that turbo does work.

 

I made a kernel extension that writes the current CPUfreq into the IORegistry, but I always get 331xxxx something, not the real frequency.

 

I used the code from dynamic_data.h so I have tscFrequency and the others available.

What MSR(s) do you read? You probably need kernel data.

 

I searched and found this in xnu/osfmk/i386/rtlclock.c:

		/*
	 * Set min/max to actual.
	 * ACPI may update these later if speed-stepping is detected.
	 */
	gPEClockFrequencyInfo.cpu_frequency_min_hz = cycles;
	gPEClockFrequencyInfo.cpu_frequency_max_hz = cycles;

So if sysctl isn't updated... the p-states don't work?

Link to comment
Share on other sites
flAked
flAked
Posted
Posted
What MSR(s) do you read? You probably need kernel data.

MSR_PLATFORM_INFO and MSR_FLEX_RATIO

 

BTW; is Lion DP3 working with RevoBoot?

 

This could be work?

We would need the source for it, it's hard to google because msr-tools is a linux project as well.

Link to comment
Share on other sites
winnimac
winnimac
Posted
Posted
SpeedStepper v1.0 (BETA)

 

SpeedStepper is a command-line tool that patches AppleIntelCPUPowerManagement.kext to load on ASUS P67 boards.

 

It patches 10 distinct write operations to MSR 0x2E, that is locked by the ASUS BIOS.

 

Supported is only AICPUPM v142.6.0 (32/64bit).

 

 

WARNING: this is beta software and meant for further analysis to get SpeedStep working on P67. It is not a complete solution. Test it at your own risk.

 

Usage: Make a backup of the kext first, SpeedStepper will directly write to the binary given. You can run it multiple times and see if everything was patched.

 

 

speedstepper.zip

 

 

 

How to use?

thx

Link to comment
Share on other sites
flAked
flAked
Posted
Posted
How to use?

thx

Please don't quote the whole post.

 

The instructions are there, what exactly is unclear?

 

It is not a public release yet (still beta), I will make a installer package later on.

Link to comment
Share on other sites
flAked
flAked
Posted
Posted
This or this could be help?

Also find this on GitHub.

All Linux I'm afraid. I tested MSR Tools in 32bit and it shows a current frequency of 1,6Ghz, but it doesn't change when starting Prime95. But the tool is from 2008, so...

 

To be honest, I'm a bit lost on the next steps. I could try the iMac12,1 (me having the i5) and see if that gives the matching I get with MBP8,3. I guess I'm still puzzled about the occasional boot hang, that I can't explain. Maybe by booting with Chameleon something gets reset?

 

Or because the SMBIOS-fix gives me such slow boot times, fix SSDT_USB - a while back you said something about fixing it, whats the latest?

Link to comment
Share on other sites
Time2Retire
Time2Retire
Posted
  • Author
Posted
To be honest, I'm a bit lost on the next steps. I could try the iMac12,1 (me having the i5) and see if that gives the matching I get with MBP8,3. I guess I'm still puzzled about the occasional boot hang, that I can't explain. Maybe by booting with Chameleon something gets reset?

The Extensions.mkext might get changed yes. This has caused me trouble in the past. Where does it stop? What is the last line you see?

 

Or because the SMBIOS-fix gives me such slow boot times, fix SSDT_USB - a while back you said something about fixing it, whats the latest?

 

This is the static data I use instead of loading /Extra/ACPI/SSDT_USB.aml:

#if STATIC_SSDT_USB_TABLE_INJECTION
#define STATIC_SSDT_USB_TABLE_DATA \
/* 1046 Bytes to convert to little-endian: */ \
0x54445353, 0x00000416, 0x5041E501, 0x20454C50, 0x45627355, 0x00646378, 0x00001000, 0x4C544E49, \
0x20110316, 0x5C3F4110, 0x42535F2E, 0x4943505F, 0x48825B30, 0x43484522, 0x415F0831, 0x000C5244, \
0x06001D00, 0x535F2E5C, 0x57505F42, 0x505F3344, 0x5F085752, 0x0A443353, 0x535F0803, 0x030A4434, \
0x1F49825B, 0x4E425548, 0x44415F08, 0x825B0152, 0x52501E4B, 0x5F084E54, 0x01524441, 0x50555F08, \
0x04071243, 0x0000FF0A, 0x505F0800, 0x1512444C, 0x00810A10, 0x00000000, 0x300A0000, 0x00001C0A, \
0x00000000, 0x5034825B, 0x08315452, 0x5244415F, 0x555F0801, 0x08124350, 0x0AFF0A04, 0x080000FF, \
0x444C505F, 0x0A101512, 0x00000081, 0x00000000, 0x1C0AE10A, 0x00000000, 0x825B0000, 0x54525035, \
0x415F0832, 0x020A5244, 0x50555F08, 0x04081243, 0xFF0AFF0A, 0x5F080000, 0x12444C50, 0x810A1015, \
0x00000000, 0x0A000000, 0x001D0AE1, 0x00000000, 0x35825B00, 0x33545250, 0x44415F08, 0x08030A52, \
0x4350555F, 0x0A040812, 0x00FF0AFF, 0x505F0800, 0x1512444C, 0x00810A10, 0x00000000, 0xE10A0000, \
0x00001D0A, 0x00000000, 0x5035825B, 0x08345452, 0x5244415F, 0x5F08040A, 0x12435055, 0xFF0A0408, \
0x0000FF0A, 0x4C505F08, 0x10151244, 0x0000810A, 0x00000000, 0x0AE10A00, 0x0000001E, 0x5B000000, \
0x52503582, 0x5F083554, 0x0A524441, 0x555F0805, 0x08124350, 0x0AFF0A04, 0x080000FF, 0x444C505F, \
0x0A101512, 0x00000081, 0x00000000, 0x1E0AB10A, 0x00000000, 0x825B0000, 0x54525035, 0x415F0836, \
0x060A5244, 0x50555F08, 0x04081243, 0xFF0AFF0A, 0x5F080000, 0x12444C50, 0x810A1015, 0x00000000, \
0x0A000000, 0x001E0AB1, 0x00000000, 0x35825B00, 0x37545250, 0x44415F08, 0x08070A52, 0x4350555F, \
0x0A040812, 0x00FF0AFF, 0x505F0800, 0x1512444C, 0x00810A10, 0x00000000, 0xB00A0000, 0x00001E0A, \
0x00000000, 0x5035825B, 0x08385452, 0x5244415F, 0x5F08080A, 0x12435055, 0xFF0A0408, 0x0000FF0A, \
0x4C505F08, 0x10151244, 0x0000810A, 0x00000000, 0x0AB10A00, 0x0000001E, 0x5B000000, 0x451B4982, \
0x08324348, 0x5244415F, 0x1A00000C, 0x2E5C0600, 0x5F42535F, 0x33445750, 0x5752505F, 0x33535F08, \
0x08030A44, 0x4434535F, 0x825B030A, 0x5548184A, 0x5F084E42, 0x01524441, 0x174C825B, 0x4E545250, \
0x44415F08, 0x5F080152, 0x12435055, 0xFF0A0407, 0x08000000, 0x444C505F, 0x0A101512, 0x00000081, \
0x00000000, 0x1C0A300A, 0x00000000, 0x825B0000, 0x54525034, 0x415F0831, 0x08015244, 0x4350555F, \
0x0A040812, 0x00FF0AFF, 0x505F0800, 0x1512444C, 0x00810A10, 0x00000000, 0xE10A0000, 0x00001C0A, \
0x00000000, 0x5035825B, 0x08325452, 0x5244415F, 0x5F08020A, 0x12435055, 0xFF0A0408, 0x0000FF0A, \
0x4C505F08, 0x10151244, 0x0000810A, 0x00000000, 0x0AE10A00, 0x0000001D, 0x5B000000, 0x52503582, \
0x5F083354, 0x0A524441, 0x555F0803, 0x08124350, 0x0AFF0A04, 0x080000FF, 0x444C505F, 0x0A101512, \
0x00000081, 0x00000700, 0x1D0AE10A, 0x00000000, 0x825B0000, 0x54525035, 0x415F0834, 0x040A5244, \
0x50555F08, 0x04081243, 0xFF0AFF0A, 0x5F080000, 0x12444C50, 0x810A1015, 0x00000000, 0x0A000000, \
0x001E0AE1, 0x00000000, 0x34825B00, 0x35545250, 0x44415F08, 0x08050A52, 0x4350555F, 0x0A040712, \
0x000000FF, 0x4C505F08, 0x10151244, 0x0000810A, 0x00000000, 0x0AB10A00, 0x0000001E, 0x5B000000, \
0x52503582, 0x5F083654, 0x0A524441, 0x555F0806, 0x08124350, 0x0AFF0A04, 0x080000FF, 0x444C505F, \
0x0A101512, 0x00000081, 0x00000000, 0x1E0AE10A, 0x00000000, 0x00000000

#endif

Copy this into config/ACPI/data.h and you are good to go. I'll also attach the DSL variant shortly...

 

p.s. Don't forget to set #define STATIC_SSDT_USB_TABLE_INJECTION to 1

 

Hehe. I finally got my activation e-mail. Tagged for moderation. How lovely. What a sweet start.

 

Update: I have attached a mini-perl script that I used to convert the data by entering the following command in a terminal window:

sudo perl ./am2struct.pl SSDT_USB.aml

ssdt_usb.dsl.zip

aml2struct.pl.zip

Link to comment
Share on other sites
flAked
flAked
Posted
Posted
The Extensions.mkext might get changed yes. This has caused me trouble in the past. Where does it stop? What is the last line you see?

ACPI_SMC_PlatformPlugin::registerLPCDriver - failed to locate SMC driver

 

Sometimes, I get a couple of printfs from vmware net afterwards, and the console is still responsive, so I can write jibberish and press enter.

Link to comment
Share on other sites
Time2Retire
Time2Retire
Posted
  • Author
Posted
ACPI_SMC_PlatformPlugin::registerLPCDriver - failed to locate SMC driver

 

Sometimes, I get a couple of printfs from vmware net afterwards, and the console is still responsive, so I can write jibberish and press enter.

Strange. Well. At least you don't get that stupid error message. That I tell you is driving me nuts. I don't get it. Shake hands?

 

I hope that Apple fixes the panics in their next upgrade because I have seen a couple of nasty KP's today (in iCal).

 

Did you install the Mac OS X 10.6.7 Update for iMac (early 2011) 1.0 from http://support.apple.com/kb/DL1383

Link to comment
Share on other sites
flAked
flAked
Posted
Posted
Strange. Well. At least you don't get that stupid error message. That I tell you is driving me nuts. I don't get it. Shake hands?

We'll crack it in the next few days, hopefully.

 

I'll try the USB fix tomorrow morning in conjunction with the fixed SMBIOS, thanks!

 

Have a good night, for now :thumbsup_anim:

 

 

Found the CPU-i code (by mozodojo), maybe we could update it for SB?

CPUi103src.zip

  • Like 1
Link to comment
Share on other sites
flAked
flAked
Posted
Posted
Did you install the Mac OS X 10.6.7 Update for iMac (early 2011) 1.0 from http://support.apple.com/kb/DL1383

I extracted the important bits, but am now on 10.6.8 10k521.

 

So first thing I was greeted with this sunday was the LPC error. Swapped RevoBoot with a version with MBP8,3 and it booted. Nothing changed since yesterday, and the other version was booting fine yesterday as well.

Booting into Windows did not change anything. So maybe I should power off, wait 15min and see if I get the LPC error with this version as well.

 

Ehm, this is a shocker!

 

post-210592-1306061457_thumb.jpgpost-210592-1306061467_thumb.jpg

The multipliers are of course wrong, but it shows a changing frequency if I start prime95. What?!

 

iLeopod just sent me these links (thanks!), sources to PStatechanger and VoodooPstate:

http://osx86drivers.svn.sourceforge.net/vi.../PStateChanger/

http://osx86drivers.svn.sourceforge.net/vi...s/VoodooPState/

Link to comment
Share on other sites
Time2Retire
Time2Retire
Posted
  • Author
Posted
Ehm, this is a shocker!

 

The multipliers are of course wrong, but it shows a changing frequency if I start prime95. What?!

I have MSRTools but I only see Initlializing (twice) and then is goes blank. System.log shows a bunch of these errors"

Could not connect to MSR Tools driverCould not connect to MSR Tools driver

You don't get these? What is that driver anyway?

 

I have seen it working once while I was stripping the DSDT weeks ago but after that I always get this blank screen. Are you using the factory _PSS object or did you make one yourself?

 

I don't know what PStatechanger and VoodooPstate is. Do you?

Link to comment
Share on other sites
flAked
flAked
Posted
Posted
You don't get these? What is that driver anyway?

 

I have seen it working once while I was stripping the DSDT weeks ago but after that I always get this blank screen. Are you using the factory _PSS object or did you make one yourself?

Reading the MSR's is only possible in kernel space, so all of these tools need a kext. MSR Tools only works in 32bit.

 

I'm using MBP8,3 definition and my crafted SSDT_PR.

 

 

This is the one, it uses 0xFE in PSD and has no TSD. (Remember folks, this is for i5-2500K only)

SSDT_PR.aml.zip

Link to comment
Share on other sites
Time2Retire
Time2Retire
Posted
  • Author
Posted
Reading the MSR's is only possible in kernel space, so all of these tools need a kext. MSR Tools only works in 32bit.

Ok but what kext might that be? This way I can check if I have it, or that some update or action of me removed it.

 

I'm using MBP8,3 definition and my crafted SSDT_PR.

 

This is the one, it uses 0xFE in PSD and has no TSD. (Remember folks, this is for i5-2500K only)

SSDT_PR.aml.zip

I'll have a look at it. Thanks.

 

BTW. The 10K524 update changed the error a little:

ACPI_SMC_CtrlLoop::loadStepperProgram - pmCPUControl (PMIOCSETPSTATESTEPPER) failed, result = 0x4

The version number is still the same (4.7.0) but now it is using uppercase (PMIOCSETPSTATESTEPPER) and the 0x4 was 0x16 before.

Link to comment
Share on other sites
flAked
flAked
Posted
Posted
Ok but what kext might that be? This way I can check if I have it, or that some update or action of me removed it.

info.olofsson.driver.CMSRToolsDriver (1.0.1d1)

Kext_MSRTools.kext is inside the bundle.

 

BTW. The 10K524 update changed the error a little:

AICPUPM still the same?

Link to comment
Share on other sites
Time2Retire
Time2Retire
Posted
  • Author
Posted 
info.olofsson.driver.CMSRToolsDriver (1.0.1d1)

Kext_MSRTools.kext is inside the bundle.

Ok. The kext loads fine but the screen goes blank.

 

AICPUPM still the same?

Same version number yes, but it might have changed anyway. Let me check something... Yes. Still the same!

 

BTW. The error is not there when I use MacBookPro8,3 but I want to use iMac12,2 :)

 

Going to do a move trick once again...

Link to comment
Share on other sites
flAked
flAked
Posted
Posted

For the SSDT_USB, the P8P67 has

Name (_S4D, 0x02)
Name (_S3D, 0x02)
Name (_S2D, 0x02)
Name (_S1D, 0x02)

 

And in yours

Name (_S3D, 0x03)
Name (_S4D, 0x03)

 

Is there a reason to not set the states for S1 and S2? I find the section 7.2.15 in the ACPI specs rather cryptic, can you explain why it is 0x03?

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