Jump to content

AnVAL (ACPI Loader)

valv

By valv,
in Chameleon

 Share
1538 posts in this topic

Recommended Posts

JBraddock
JBraddock
Posted
Posted
@Valv,

 

How does Restart and Shutdown is handling with your bootloader? I am still using OpenHaltRestart.kext. I don't know the reason but most of the time when I choose to shutdown or restart the computer, I am being welcomed with a black screen loading some texts and at the end of it saying CPU Halted etc.

 

Would you mind telling me how I should be using built-in, if any, shutdown and restart fixes?

How is your solution on HDX18?

WMware Fusion was causing the problem so I removed that and everything seems fine now. I'll try removing OpenHaltRestart.kext and see what happens.

valv
valv
Posted
  • Author
Posted

after a long time of meditation, we got u some good news: Beta is here...once again

thanks to the great work of cparm & the team.

stuff added:

  • blackosx's new themes support
  • fermi support
  • p-states & c-states (c4 included)
  • memory detection

hope u guyz could test it further :wacko:

available here

scrax
scrax
Posted
Posted
after a long time of meditation, we got u some good news: Beta is here...once again

thanks to the great work of cparm & the team.

stuff added:

  • blackosx's new themes support
  • fermi support
  • p-states & c-states (c4 included)
  • memory detection

hope u guyz could test it further :wacko:

available here

 

Are the source on the repo updated?

 

what's the differences with the trunk, "just" blackosx's theme support and fermi support?

 

I like how chameleon is growin' after the opening of the public repo.

JBraddock
JBraddock
Posted
Posted
after a long time of meditation, we got u some good news: Beta is here...once again

thanks to the great work of cparm & the team.

stuff added:

  • blackosx's new themes support
  • fermi support
  • p-states & c-states (c4 included)
  • memory detection

hope u guyz could test it further :D

available here

Valv, do I need to remove legacy ACPI_SMC_PlatformPlugin that you created for me before adding c-states and p-states key to the boot.plist?

valv
valv
Posted
  • Author
Posted
Valv, do I need to remove legacy ACPI_SMC_PlatformPlugin that you created for me before adding c-states and p-states key to the boot.plist?
I don't think it's needed. go ahead.
JBraddock
JBraddock
Posted
Posted
I don't think it's needed. go ahead.

I added the following keys to the boot.plist.

<key>GeneratePStates</key>
<string>Yes</string>
<key>GenerateCStates</key>
<string>Yes</string>

The rest of my boot.plist is identical to yours including DropSSDT=No and also excluding oemSSDT etc. Memory detection seems to work.

Memory Slots:

 ECC:	Disabled

BANK 0/DIMM0:

 Size:	2 GB
 Type:	DDR2 SDRAM
 Speed:	800 MHz
 Status:	OK
 Manufacturer:	Apple Computer Inc.
 Part Number:	M470T5663QZ3-CF73Q
 Serial Number:	*/ deleted by me

BANK 2/DIMM2:

 Size:	2 GB
 Type:	DDR2 SDRAM
 Speed:	800 MHz
 Status:	OK
 Manufacturer:	Apple Computer Inc.
 Part Number:	M470T5663QZ3-CF73Q
 Serial Number:	*/deleted by me

However, C-States and P-States couldn't be generated.

I"ll make another reboot to catch the full error messages.

 

Ok. Here is the error messages.

 

- Invalid characters found in ProcessorsOP! (repeats a few times)

- DSDT CPUs not found: C-States not generated

- DSDT CPUs not found: P-States not generated

 

Here is my DSDT file. @Vavl, I have only your DSDT hacks applied.

dsdt.aml.zip

scrax
scrax
Posted
Posted
For what i can see it was released 13th so the last mozodojo mod are not yet merged, or am I wrong?

 

Valv, please can you confirm my thoughts?

Your booter was my default booter until two days ago when i've tried the mozodojo's one.

 

On the repo I see that your last change is dated 13 july 2010, so I can't understand how could this build have the same code posted in the last day by mozodojo, rekursor and zef.

 

Is the memory detection different from the one on the trunk? If not i've already reported that id dosn't work on my zotac mobo...

JBraddock
JBraddock
Posted
Posted
issued from mozodojo's topic:

 

You need to DropSSDT or not to use GenerateCStates feature.

I tried the followings.

1) GeneratePStates, GenerateCStates, DropSSDT = Yes:

Same above-mentioned error messages from Chameleon and the following errors on Console. I also lost P-States, which you activated via legacykext.

Jul 24 21:40:54 SnowKittys-MacBook-Pro kernel[0]: ACPI_SMC_PlatformPlugin::pushCPU_CSTData - _CST evaluation failed
Jul 24 21:40:54 SnowKittys-MacBook-Pro kernel[0]: ACPI_SMC_PlatformPlugin::registerLPCDriver - WARNING - LPC device initialization failed: C-state power management not initialized
Jul 24 21:40:54 SnowKittys-MacBook-Pro kernel[0]: ACPI_SMC_PlatformPlugin::pushCPU_CSTData - _CST evaluation failed

2) GeneratePStates, DropSSDT = Yes: Same with the above result.

3) GeneratePStates, GenerateCStates = Yes. I removed DropSSDT=No key from my boot.plist. Same result. Not ACPI_SMC_PlatformPlugin related errors.

 

My rams are manufactured by Samsung, which also is being reported prior to Chamaleon menu but under System Profiler it is reported as manufactured by Apple.

JBraddock
JBraddock
Posted
Posted
I tried the followings.

1) GeneratePStates, GenerateCStates, DropSSDT = Yes:

Same above-mentioned error messages from Chameleon and the following errors on Console. I also lost P-States, which you activated via legacykext.

Jul 24 21:40:54 SnowKittys-MacBook-Pro kernel[0]: ACPI_SMC_PlatformPlugin::pushCPU_CSTData - _CST evaluation failed
Jul 24 21:40:54 SnowKittys-MacBook-Pro kernel[0]: ACPI_SMC_PlatformPlugin::registerLPCDriver - WARNING - LPC device initialization failed: C-state power management not initialized
Jul 24 21:40:54 SnowKittys-MacBook-Pro kernel[0]: ACPI_SMC_PlatformPlugin::pushCPU_CSTData - _CST evaluation failed

2) GeneratePStates, DropSSDT = Yes: Same with the above result.

3) GeneratePStates, GenerateCStates = Yes. I removed DropSSDT=No key from my boot.plist. Same result. Not ACPI_SMC_PlatformPlugin related errors.

 

My rams are manufactured by Samsung, which also is being reported prior to Chamaleon menu but under System Profiler it is reported as manufactured by Apple.

I had to move Scope (_PR) to the beginning of DSDT file. I got the tip from this post. Apparently, this is a bug and has been fixed by mozodojo. @Valv, may be you didn't include this fix to your branch.

 

I also included the following keys to the boot.plist. And now it's generated P-States and C-States. 

<key>GeneratePStates</key>
<string>Yes</string>
<key>GenerateCStates</key>
<string>Yes</string>
<key>DropSSDT</key>
<string>Yes</string>

 

Previously, I had 4 PerformanceStateArray and now I have 9.

post-562417-1280000097_thumb.jpg

But I don't see any difference in terms of temperature. It is almost the same values while using P-States only.

 

I wonder if I should remove ACPI_SMC_PlatformPlugin and set my Mac model to MacBookPro5,5. I am using 5,4 although it is not the closest match to my CPU.

Any ideas?

 

Edit:

I removed ACPI_SMC_PlatformPlugin and set my Mac model to MacBookPro5,5. No change. ThermalProfile part was empty.

Then I created ACPI_SMC_PlatformPlugin according to MacBookPro5,5 but no luck. Coolbook shows 9 P-States but there is no temperature change.

 

'sudo setpci -s 0:1f.0 0xa6.b' returns 00. To my knowledge it should return 80 when C-States are working.

cparm
cparm
Posted
Posted
'sudo setpci -s 0:1f.0 0xa6.b' returns 00. To my knowledge it should return 80 when C-States are working.

 

not exactly, afaik 80 mean that c-state can go deeper than C2 state

 

Unlocked C-State Transition — RO. This bit is set by hardware when a processor power state (C-State)

transition deeper than C2 occurs and the C- STATE_CONFIG_LOCK bit is not set. This bit is cleared by PLTRST#

and is not writable by software.

danielkza
danielkza
Posted
Posted
not exactly, afaik 80 mean that c-state can go deeper than C2 state

The wording on the docs. is a bit unclear: it may mean the bit is set right after a transition to C3/C4 occurs, and cleared when the CPU wakes up, which would make it unreliable for determining the if deeper sleep states are enable, but not currently active.

mm67
mm67
Posted
Posted
The wording on the docs. is a bit unclear: it may mean the bit is set right after a transition to C3/C4 occurs, and cleared when the CPU wakes up, which would make it unreliable for determining the if deeper sleep states are enable, but not currently active.

 

There is nothing unclear about PLTRST# signal, it is only asserted at boot up so the bit that gets set by C3/C4 stays on until next boot.

 

Platform Reset: The ICH10 asserts PLTRST# to reset devices on the platform (e.g., SIO, FWH, LAN, (G)MCH, TPM, etc.). The ICH10 asserts PLTRST# during power-up and when S/W initiates a hard reset sequence through the Reset Control register (I/O Register CF9h). The ICH10 drives PLTRST# inactive a minimum of 1 ms after both PWROK and VRMPWRGD are driven high. The ICH10 drives PLTRST# active a minimum of 1 ms when initiated through the Reset Control register (I/O Register CF9h).

danielkza
danielkza
Posted
Posted
There is nothing unclear about PLTRST# signal, it is only asserted at boot up so the bit that gets set by C3/C4 stays on until next boot.

 

Platform Reset: The ICH10 asserts PLTRST# to reset devices on the platform (e.g., SIO, FWH, LAN, (G)MCH, TPM, etc.). The ICH10 asserts PLTRST# during power-up and when S/W initiates a hard reset sequence through the Reset Control register (I/O Register CF9h). The ICH10 drives PLTRST# inactive a minimum of 1 ms after both PWROK and VRMPWRGD are driven high. The ICH10 drives PLTRST# active a minimum of 1 ms when initiated through the Reset Control register (I/O Register CF9h).

I was to lazy read the docs on PLTRST#, my bad.

rednous
rednous
Posted
Posted

hi, valv :)

 

Just tested AnVAL5 beta2 and here's the last two lines of my kernel log:

kernel[0]: ACPI_SMC_PlatformPlugin::pushCPU_CSTData - _CST evaluation failed
kernel[0]: ACPI_SMC_PlatformPlugin::registerLPCDriver - WARNING - LPC device initialization failed: C-state power management not initialized

 

I completely removed all the code related to CPU power management and only left the processor scope in my DSDT. Im using the following keys in the boot.plist:

	<key>DropSSDT</key>
<string>Yes</string>
<key>UpdateACPIVersion</key>
<string>Yes</string>	
<key>GeneratePStates</key>
<string>Yes</string>
<key>GenerateCStates</key>
<string>Yes</string>
<key>EnableC4State</key>
<string>Yes</string>
<key>oemDSDT</key>
<string>No</string>
<key>oemFADT</key>
<string>No</string>
<key>oemSSDT</key>
<string>Yes</string>
<key>FixFSB</key>
<string>No</string>
<key>ForceFSB</key>
<string>266</string>
<key>busratio</key>
<string>110</string>

 

It seems C and P states aren't working and the CPU temps are 6-7 degrees higher than with working C and P states.

rednous
rednous
Posted
Posted
Hi All,

 

As I understand it, ATI HD 3650 is not supported in SL, but it is listed as one of the AnVal bootloaders supported graphics cards. Does this mean that the card will be supported but without the appropriate kext I won't get Open GL QE/CI etc?

 

Thanks

 

If you own ATI HD 3650 go for the current AnVAL boot loader, install, test and report here :)

k20kev
k20kev
Posted
Posted

Installed iAtkos SL with AnVal Bootlader and ATI HD 3650 (dev id 9598). I deleted all ATI kexts and I can boot but no QE/CI OpenCL or resolution change - so pretty much useless.

 

My first try I used ATI Frambuffer and ATI enabler that came with iAtkos but just got a black screen at the GUI stage.

 

Any help will be much appreicated, but I'm guessing that nothing can be done. It seems strange that the ATIX2000 kext contains my dev id, along with many others, but doesn't support it.

kizwan
kizwan
Posted
Posted
Installed iAtkos SL with AnVal Bootlader and ATI HD 3650 (dev id 9598). I deleted all ATI kexts and I can boot but no QE/CI OpenCL or resolution change - so pretty much useless.

 

My first try I used ATI Frambuffer and ATI enabler that came with iAtkos but just got a black screen at the GUI stage.

 

Any help will be much appreicated, but I'm guessing that nothing can be done. It seems strange that the ATIX2000 kext contains my dev id, along with many others, but doesn't support it.

May I know what is your computer specification? Based on the symptom you explained, make me think that this is notebook computer. If it is, you can forget about it, it will not work. Even though this is desktop computer, I don't think you can get it to work if it doesn't work with combination of enabler & stock ATI kexts. Your only option is RadeonHD.kext but don't hope it will enabled QE/CI. Ahhh...this make me miss my old computer with ATI Radeon HD 2600XT. It works wonderfully on Mac OS X since Apple Mac Pro at that time use the same graphic card. New Apple Mac Pro come with ATI Radeon HD 5770 & ATI Radeon HD 5870.

 Share
Go to topic listing
×
×
  • Create New...