Help - Search - Members - Calendar
Full Version: Monitoring With Fakesmc
Project OS X Forums > Snow Leopard Guides & Tutorials > Tools
Pages: 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20
Konsti
Guys, I reverted to my original, extracted DSDT and applied the fixes (HPET/TIMR/RTC/USBs/LPC/SATA/GFX0/GIGE etc.) and the system produces a kernel panic apparently due to FakeSMC.kext (way down in boot process). I reverted back to netkas' v2.5 and it rebooted fine (in /E/E/). When using rel218, in /S/L/E/ I get a panic. Cannot find this in Console.app > kernel.log. Sharing a photo of the screen might guide you at all for some help?

The reason for reverting back to the extracted DSDT with added fixes (and no code/device deletion) was to have proper sleep/thermalzone support, as I didn't have that when I started removing stuff by following MasterChief's DSDT steps...

Thanks in advance.
alphamerik
QUOTE (Freaky Chokra @ Jul 1 2010, 03:16 AM) *
With "kextstat | grep nv", this output:
CODE
Hack-iMac:PlugIns FreakyChokra$ kextstat | grep nV
   90    0 0xffffff7f812df000 0x10000    0x10000    com.PaztalomTechnologiez.nVmon (1) <76 68 30 9 7 4 3>

So I think it does get loaded. sad.gif [or I don't know actually, may be]


You can also do `sudo tail -f /var/log/system.log &` (instead of dmesg), and then kextunload and kextload, the log output will be easier to see (output isn't well labeled). Also check output of `kextload -nt FakeSMCnVclockPort.kext`.

Thanks for the info Mojodojo, guess it isn't a good idea to mix & match plugins from different versions.
Freaky Chokra
QUOTE (Mojodojo @ Jul 1 2010, 04:21 PM) *
Freaky Chokra, you couldn't simply get old plugin and get it worked, not in all cases. FakeSMC 2.7 is not compatable with older fakesmc versions. So if old plugins depending on old fakesmc but you have new one, plugin will not load.

Hi Mojodojo.
Thanks for that info. Actually, it was made clear a couple of post-pages back about not mixing old fakesmc + plugins. they don't work. Kernel Panic is sure.

Yet I tried as Alphamerik advised. Just to check if it did work this time. Just to get our ideas clear. wink.gif
&
QUOTE (alphamerik @ Jul 1 2010, 09:52 PM) *
You can also do `sudo tail -f /var/log/system.log &` (instead of dmesg), and then kextunload and kextload, the log output will be easier to see (output isn't well labeled). Also check output of `kextload -nt FakeSMCnVclockPort.kext`.

Thanks for the info Mojodojo, guess it isn't a good idea to mix & match plugins from different versions.

Alphamerik.
You told to use kextload with -nt for those plugins or even fakesmc (which I tried but didn't work).
But kextload or kextunload with -nt displays an error as under:
CODE
bash-3.2# kextload -nt FakeSMCnVclockPort.kext
kextload: unrecognized option `-nt'
Use kextutil(8) for development loading of kexts.
usage: kextload [options] [--] [kext] ...

use kextload -help for an explanation of each option
bash-3.2#


Now, What should I do??? sad.gif sad.gif sad.gif sad.gif
I really stumped!!!

Waiting for further answers. BUt I don't know if you guys have checked my DSDT.
I hope it is what you guys have added from posts #10 -> #23.

Or if any keys in FakeSMCnVClockPort.kext plugin needs to be added or modified. [I Don't Know... sad.gif ]

Respect & Regards,
Freaky Chorka :ph43r:
Konsti
This is the kernel panic screen: it's caused by nVmon (com.PaztalomTechnologiez.nVmon). I don't know what the plugin tries to read from memory etc. Can you be of assistance, guys? Thank you...

Logicpro8_user
Any more on implementing the SuperIO Chip=LPC47m182 is the next update?
alphamerik
QUOTE (Freaky Chokra @ Jul 1 2010, 06:31 PM) *
Alphamerik.
You told to use kextload with -nt for those plugins or even fakesmc (which I tried but didn't work).
But kextload or kextunload with -nt displays an error as under:

Waiting for further answers. BUt I don't know if you guys have checked my DSDT.


Looks like 10.6 moved a lot of functionality to kextutil, try:
`kextutil -nt FakeSMCnVclockPort.kext`

Of course I could be wrong but this looks like a bug somehow, probably not DSDT error.
Nvclock still works correctly, right?

Personally I haven't been keeping up with the newest revisions, have a build that works for me and haven't tried the new ones yet.
Mojodojo
I've added automatic software Fan Control support for Winbond chips. It could be buggy because I have no Winbond to test.
SMSC is the next.
Freaky Chokra
QUOTE (alphamerik @ Jul 2 2010, 05:06 AM) *
Looks like 10.6 moved a lot of functionality to kextutil, try:
`kextutil -nt FakeSMCnVclockPort.kext`

Of course I could be wrong but this looks like a bug somehow, probably not DSDT error.
Nvclock still works correctly, right?

Personally I haven't been keeping up with the newest revisions, have a build that works for me and haven't tried the new ones yet.

Hi alphamerik! Thanks a ton!
Here is the output:
CODE
bash-3.2# cd System/Library/Extensions/
bash-3.2# ls -al | grep Fak
drwxr-xr-x@   3 root  wheel   102 23 Jun 21:52 FakeSMC.kext
drwxr-xr-x    3 root  wheel   102 17 Jun 16:24 FakeSMCnVclockPort.kext
bash-3.2# kextutil -nt FakeSMCnVclockPort.kext
No kernel file specified; using running kernel for linking.
/System/Library/Extensions/FakeSMCnVclockPort.kext appears to be loadable (including linkage for on-disk libraries).
bash-3.2#


Respect & Regards,
Freaky Chokra :ph43r:
Logicpro8_user
QUOTE (Mojodojo @ Jul 2 2010, 08:18 AM) *
I've added automatic software Fan Control support for Winbond chips. It could be buggy because I have no Winbond to test.
SMSC is the next.


Great news! anything I can do?
kDawg
Will FakeSMCnVclockPort pick up two GFX cards? If so will it try and report the fans on each? I have two GFX cards but one is fanless. Not sure why I'm displaying two GPU fans. I'm using all the latest SVN builds with FakeSMCSuperIO.
Click to view attachment
Freaky Chokra
QUOTE (kDawg @ Jul 3 2010, 02:30 AM) *
Will FakeSMCnVclockPort pick up two GFX cards? If so will it try and report the fans on each? I have two GFX cards but one is fanless. Not sure why I'm displaying two GPU fans. I'm using all the latest SVN builds with FakeSMCSuperIO.
Click to view attachment

Hey! KDawg!

can you please share how you got the GPU Diode Temps? PLEASEEEEEEE!!!
Or, can you post your dsdt or any edited info.plist files in the fakesmcnvclockport.kext?
I'll try to match and do further things in my dsdt and my computer.

Please................. Please........... Please?????????

Respect & Regards,
Freaky Chokra ph34r.gif
kDawg
QUOTE (Freaky Chokra @ Jul 2 2010, 01:33 PM) *
Hey! KDawg!

can you please share how you got the GPU Diode Temps? PLEASEEEEEEE!!!
Or, can you post your dsdt or any edited info.plist files in the fakesmcnvclockport.kext?
I'll try to match and do further things in my dsdt and my computer.

Please................. Please........... Please?????????

Respect & Regards,
Freaky Chokra :ph43r:
Evidently there must be changes to FakeSMCnVclockPort that enabled them.
Freaky Chokra
QUOTE (kDawg @ Jul 3 2010, 04:44 AM) *
Evidently there must be changes to FakeSMCnVclockPort that enabled them.

Exactly what I asked you to provide. If you had edited any files, and by editing/correcting which values in that KEXT file?

Will you provide that much help or not?
kDawg
QUOTE (Freaky Chokra @ Jul 2 2010, 11:45 PM) *
Exactly what I asked you to provide. If you had edited any files, and by editing/correcting which values in that KEXT file?

Will you provide that much help or not?

All I did was compile the latest build from the SVN repository. I didn't change a thing otherwise I would have told you. Have you been compiling the nightly builds?
Freaky Chokra
QUOTE (kDawg @ Jul 3 2010, 01:59 PM) *
All I did was compile the latest build from the SVN repository. I didn't change a thing otherwise I would have told you. Have you been compiling the nightly builds?

Oh Dear kDawg!
No. No. No.

I'm no programmer. sad.gif
And, Compiling from SVN... is not even Greek to me! Maybe Klingon or Na'avi! tongue.gif Lolz! wink.gif
No compiling anything, moi! However, thanks for that.
I still don't understand how can GPU Temp be disabled when GPU Fan RPM can now be seen! ?????

Respect & Regards,
Freaky Chokra ph34r.gif
kDawg
QUOTE (Freaky Chokra @ Jul 4 2010, 01:10 AM) *
Oh Dear kDawg!
No. No. No.

I'm no programmer. sad.gif
And, Compiling from SVN... is not even Greek to me! Maybe Klingon or Na'avi! tongue.gif Lolz! wink.gif
No compiling anything, moi! However, thanks for that.
I still don't understand how can GPU Temp be disabled when GPU Fan RPM can now be seen! ?????

Respect & Regards,
Freaky Chokra ph34r.gif
I'll send you the compiled kext when I get back to my computer.

EDIT:
Here is rev 259 and the latest as of 7.4.2010. This has FakeSMCSuperIO, FakeSMCSMBus, FakeSMCnVclockPort and FakeSMCIntelThermal in the plug-ins folder. This may get your GPU temps to display.
Click to view attachment
snackole
Does anyone have proper fan speeds with a 9600GT ?
mm67
Did some testing with my 2 Gigabyte boards and SuperIO plugin's fancontrol. One board is GA-EP45-UD3LR and on that one fancontrol works perfectly with both voltage and PWM controlled fans. My other board is GA-EP45-UD3 and on that one there is some problems with fancontrol. Both boards are using IT8718F as SuperIO chip but on this problem board Cpu fan header is connected to chip's third PWM output so there is no way to control it since plugin only handles outputs 1 & 2. Sys_fan2 header is connected to PWM output 2 as on other board but even that one has a problem, while idling or on light load fan works normally but on heavy load, like Prime95, control seems to go crazy. Fan speed starts going up and down, some times fan even stops for some seconds. On other board I couldn't replicate this behavior, fancontrol just works with light or heavy load. I did check that there is nothing wrong with header itself by booting in to Windows and using SpeedFan to control that header, no problems even with heavy loads. So the problem seems to be with SuperIO plugin's fancontrol, maybe it somehow messes fancontrol that PWM output 1 is not connected to anything ?
Freaky Chokra
QUOTE (kDawg @ Jul 4 2010, 08:00 PM) *
I'll send you the compiled kext when I get back to my computer.

EDIT:
Here is rev 259 and the latest as of 7.4.2010. This has FakeSMCSuperIO, FakeSMCSMBus, FakeSMCnVclockPort and FakeSMCIntelThermal in the plug-ins folder. This may get your GPU temps to display.
Click to view attachment

Hey kDawg! Thanks a million for the compiled version. wink.gif Really it worked without causing kernel panics. However......... Damn! Damn! Damn! sad.gif sad.gif sad.gif

Dear kDawg... sad.gif
Now I do get GPU temperature. But if you'd check my previous attachments [HERE] of screen shots... Heatsink Temperature is now missing!
And, VCore is now showing 3.00 Volts!!!!!!!!!!!!!!!!!!!!!!!!
And CPU & GPU Fan RPM is an Insanely Extremely HIGH Figure of -6128 RPM for Fan1 (cpu fan) and 5136 RPM for GPU Fan.

Screenshot:
Click to view attachment

This all is so weird and strange that this is happening to my setup only! sad.gif Even after I have edited the DSDT with the codes in posts #10 -> #23.
And, no other changes! sad.gif sad.gif sad.gif

Respect & Regards,
Freaky Chokra ph34r.gif
Mojodojo
mm67, you can control up to 5 FANs. Just duplicate Fan1 or Fan0 node and rename to Fan2, Fan3, Fan4. You can connect Processor or System temperature input to any FAN, or can connect nothing: in this case "Start Throttle" value will be used. I think i must add some debug info about fan index because the FAN numbers on your MB can be different from used by chip.

I will see what can I do with Fan Control problems on some boards.
Mojodojo
QUOTE (Freaky Chokra @ Jul 5 2010, 08:12 AM) *
Hey kDawg! Thanks a million for the compiled version. wink.gif Really it worked without causing kernel panics. However......... Damn! Damn! Damn! sad.gif sad.gif sad.gif

Dear kDawg... sad.gif
Now I do get GPU temperature. But if you'd check my previous attachments [HERE] of screen shots... Heatsink Temperature is now missing!
And, VCore is now showing 3.00 Volts!!!!!!!!!!!!!!!!!!!!!!!!
And CPU & GPU Fan RPM is an Insanely Extremely HIGH Figure of -6128 RPM for Fan1 (cpu fan) and 5136 RPM for GPU Fan.

Screenshot:
Click to view attachment

This all is so weird and strange that this is happening to my setup only! sad.gif Even after I have edited the DSDT with the codes in posts #10 -> #23.
And, no other changes! sad.gif sad.gif sad.gif

Respect & Regards,
Freaky Chokra ph34r.gif


There is your FakeSMC and plugins installed? In E/E or S/L/E? Do not install into E/E, plugins will not work.
mm67
QUOTE (Mojodojo @ Jul 5 2010, 08:19 AM) *
mm67, you can control up to 5 FANs. Just duplicate Fan1 or Fan0 node and rename to Fan2, Fan3, Fan4. You can connect Processor or System temperature input to any FAN, or can connect nothing: in this case "Start Throttle" value will be used. I think i must add some debug info about fan index because the FAN numbers on your MB can be different from used by chip.

I will see what can I do with Fan Control problems on some boards.


Adding Fan2 did enable controlling of Cpu fan header but there is still some problems:

1. Sys_Fan1 header outputs so low signal that a 120 mm Akasa Apache fan doesn't spin up by itself when I turn on the system and since the fan isn't running when SuperIO plugin starts it disables control of that output. If I push the fan a little bit to help it start while booting then plugin sees the fan and takes control.

2. Enabling control of Cpu fan makes also that one to to start spinning up and down on heavy load so at some point all PWM fans might be turned off under heavy load.
гык-sse2
QUOTE (Freaky Chokra @ Jul 5 2010, 09:12 AM) *
Hey kDawg! Thanks a million for the compiled version. wink.gif Really it worked without causing kernel panics. However......... Damn! Damn! Damn! sad.gif sad.gif sad.gif

Dear kDawg... sad.gif
Now I do get GPU temperature. But if you'd check my previous attachments [HERE] of screen shots... Heatsink Temperature is now missing!
And, VCore is now showing 3.00 Volts!!!!!!!!!!!!!!!!!!!!!!!!
And CPU & GPU Fan RPM is an Insanely Extremely HIGH Figure of -6128 RPM for Fan1 (cpu fan) and 5136 RPM for GPU Fan.

Screenshot:
Click to view attachment

This all is so weird and strange that this is happening to my setup only! sad.gif Even after I have edited the DSDT with the codes in posts #10 -> #23.
And, no other changes! sad.gif sad.gif sad.gif

Respect & Regards,
Freaky Chokra ph34r.gif


It's not with your system only, it's now with ALL Winbonds (testing came from beta back to alpha and it's on applelife.ru). Mozo is fighting with it, because we (developers) have IT8718F chips, and only users have WB83627 ones.
Freaky Chokra
QUOTE (гык-sse2 @ Jul 5 2010, 04:54 PM) *
It's not with your system only, it's now with ALL Winbonds (testing came from beta back to alpha and it's on applelife.ru). Mozo is fighting with it, because we (developers) have IT8718F chips, and only users have WB83627 ones.

Thanks гык-sse2!

Well, looking at the way things are, I'm going back to 218. sad.gif sad.gif sad.gif

But, will keep checking for updates.

However, this time I tried mixing up rev 218 and kDawg's rev 259 versions' plugins with each other 2 dozen times, and got interesting results.

Some sensors would get enabled, Fans would report wrong figures.
Fans are correct then GPU/CPU temperatures would absolutely go missing. iStat menus can not read them.
So such are the errors. Ok? sad.gif
If you wish I can post the screen shots.

And Lastly, I also tried playing with dependency figures in the .plist file of nVclockport.kext and IntelThermal.kext

Those made it possible to load older+newer plugins with older rev 218 FakeSMC.kext
and newer rev 259 FakeSMC load similarly older+newer plugin set.
I'm reporting this, coz previously it was not possible.
We have had reports of Kernel Panics etc. right?

Well, till next time. Hope you guys can solve this soon.

Respect & Regards,
Freaky Chokra ph34r.gif
FormerlyKnownAs
rev259 is giving me some messed up voltages and fan speeds

Click to view attachment
CODE
NVDANV50HAL loaded and registered.
FakeSMCIntelThermal: CPU family 0x6, model 0x17, stepping 0x7
FakeSMCIntelThermal: Found 4 cores 0 threads
FakeSMCIntelThermal: CPU Tjmax 100
SuperIO: Probing Fintek
SuperIO: Found Fintek F71882 Super I/O chip on 0xa00
DSMOS has arrived
PTKawainVi: started
Vendor ID: 10de, Device ID: 5e2
Architecture: 800000
nvidia GeForce GTX 260
Binding key TG0D/nBinding key TG0H/nBinding key F0Ac/nSuperIO: Binding key Th0H
SuperIO: Binding key TN0P
SuperIO: Binding key VC0C
SuperIO: Binding key F1Ac
AppleRTL8169Ethernet: Ethernet address
ACPI_SMC_GPU_CtrlLoop: NOTE: ignoring mismatched primary control with ID:18




Rgds

D
Mojodojo
Check new version on SVN.
Freaky Chokra
QUOTE (Mojodojo @ Jul 6 2010, 11:30 AM) *
Check new version on SVN.

Hi Mojodojo!
I just checked on the svn site. It's now revised to Rev 283. But I think those are source files. wink.gif
Well, that probably means we have to compile them ourselves. But I am not into that at all. sad.gif
How do folks like me try get the latest SVN files to try, install, and report back with any feedback? sad.gif

How can I compile them, if I do get Xcode for Snow Leo?

Respect & Regards,
Freaky Chokra ph34r.gif
Logicpro8_user
Nice rev 277! is smsc supported in the version?
Cyberdog!
test with
GA-P35-DS3P + Palit 8800GT (Chameleon 2.0RC4 + DSDT)



rev277 give me 3 new problem with istatPro

no CPU fan
no Motherboard temp
no Northbridge temp


A big thank's for your work
Mojodojo
277 is obsolete. Try 283. But nVlockPort, ACPI, x3100 will not work with new fakesmc, you need to wait me or Slice or usr-ss2 will update those plugins.
mm67
Problem with my GA-EP45-UD3 and using fan control with heavy load seems to be related to using Processor as speed source. If I use System as speed source fan speed stays stable. Since bios controlled fan control works just fine on Cpu fan header I could easily live with this. Now the main problem for me is that fan connected to Sys_Fan1 header doesn't start at boot because bios outputs too low signal and when plugin starts it disables control on that channel because it sees no rpm signal there. Would it be possible to force plugin to ignore rpm signal and start control anyway ?
FormerlyKnownAs
QUOTE (Mojodojo @ Jul 6 2010, 03:30 AM) *
Check new version on SVN.


will do .. was just being lazy smile.gif
Mojodojo
QUOTE (mm67 @ Jul 6 2010, 02:01 PM) *
Problem with my GA-EP45-UD3 and using fan control with heavy load seems to be related to using Processor as speed source. If I use System as speed source fan speed stays stable. Since bios controlled fan control works just fine on Cpu fan header I could easily live with this. Now the main problem for me is that fan connected to Sys_Fan1 header doesn't start at boot because bios outputs too low signal and when plugin starts it disables control on that channel because it sees no rpm signal there. Would it be possible to force plugin to ignore rpm signal and start control anyway ?


yes, it's simple. I'll add "force" parameter into fan control.

P/S/ maybe it's not needed. You can set fan name in "FAN Names" node to force its detection and controller activation
Konsti
Dear friends, since no member had a look at my KP problem with v218 of the plugin (posted a few days ago) due to my smaller/compact DSDT, I was forced this weekend to repeat the process from scratch and use NullCPUPowerManagement to reboot to a 'clean' desktop and extract the DSDT via DSDT-SE.

I managed to get my mobo back in shape, with the original (plus USB modifications) DSDT but using the plugins (v218) in /S/L/E/ I discovered that:

1. With the current Scope (\_TZ) from untouched DSDT, the plugin SMCITEController.kext doesn't produce any results. On the other hand, FakeSMCSuperIO.kext detects my ITE chipset normally. Any ideas?

CODE
Scope (\_TZ)
{
Device (FAN)
{
Name (_HID, EisaId ("PNP0C0B"))
Method (_INI, 0, NotSerialized)
{
Store (TP1H, CTOS)
Store (TP1L, CTHY)
}
}

ThermalZone (THRM)
{
Name (_AL0, Package (0x01)
{
FAN
})
Method (_AC0, 0, NotSerialized)
{
If (Or (PLCY, PLCY, Local7))
{
Return (TP2H)
}
Else
{
Return (TP1H)
}
}

Name (_PSL, Package (0x01)
{
\_PR.CPU0
})
Name (_TSP, 0x3C)
Name (_TC1, 0x04)
Name (_TC2, 0x03)
Method (_PSV, 0, NotSerialized)
{
If (Or (PLCY, PLCY, Local7))
{
Return (TP1H)
}
Else
{
Return (TP2H)
}
}

Method (_CRT, 0, NotSerialized)
{
Return (TRPC)
}

Method (_TMP, 0, NotSerialized)
{
And (SENF, 0x01, Local6)
If (LEqual (Local6, 0x01))
{
Return (RTMP ())
}
Else
{
Return (0x0B86)
}
}

Method (_SCP, 1, NotSerialized)
{
If (Arg0)
{
Store (One, PLCY)
}
Else
{
Store (Zero, PLCY)
}

Notify (\_TZ.THRM, 0x81)
}

Method (STMP, 2, NotSerialized)
{
Store (Arg1, DW00)
If (Arg0)
{
STHY (DB00, DB01, DW00)
}
Else
{
STOS (DB00, DB01, DW00)
}
}
}
}

2. With a modified/simplified Scope (\_TZ) as per slice's post here SMCITEController.kext finds my chipset.

CODE

Scope (\_TZ) // Provide basic ThermalZone support
{
Device (FAN)
{
Name (_HID, EisaId ("PNP0C0B"))
}

ThermalZone (THRM)
{
Method (_TMP, 0, NotSerialized)
{
Store (FAN, Local0)
Return (Local0)
}
}
} // End of Scope (\_TZ)

Is it safe to replace the big chunk of code with such a smaller piece of code?

3. Also, another question: is it possible to make the pugin in such a way so we speed-up the detection of chipsets when using FakeSMCSuperIO.kext by setting the detection order preference in the .plist file? I mean, detecting ITE chips is third attempt, after Winbond and the other one; if I could tell the kext to first look for ITE and then the others, I could save some boot-up time!

I guess the latter is a request to the developers. Many thanks for your time.
mm67
QUOTE (Mojodojo @ Jul 6 2010, 02:40 PM) *
yes, it's simple. I'll add "force" parameter into fan control.

P/S/ maybe it's not needed. You can set fan name in "FAN Names" node to force its detection and controller activation


Adding a name to Fan1 did the trick, now fan starts while desktop is loading, thanks.
Slice
QUOTE (Mojodojo @ Jul 6 2010, 12:14 PM) *
277 is obsolete. Try 283. But nVlockPort, ACPI, x3100 will not work with new fakesmc, you need to wait me or Slice or usr-ss2 will update those plugins.

Ya torchu!

OK, will create a recent revision if it is not obsolete yet. 277 is tested by me both in Snow and Leo. Now I convert it to Tiger but withiout NVidia, it is more complex.
FormerlyKnownAs
QUOTE (Konsti @ Jul 6 2010, 01:48 PM) *
Dear friends, since no member had a look at my KP problem with v218 of the plugin (posted a few days ago) due to my smaller/compact DSDT, I was forced this weekend to repeat the process from scratch and use NullCPUPowerManagement to reboot to a 'clean' desktop and extract the DSDT via DSDT-SE.


I'm sure you know this - when testing you should either work on a back-up/test install or keep a current backup that you can revert to in times of trouble.

I think the ThermalZone patch is as in red below (someone please correct if wrong!)
CODE
Scope (\_TZ)
{
Device (FAN)
{
Name (_HID, EisaId ("PNP0C0B"))
Method (_INI, 0, NotSerialized)
{
Store (TP1H, CTOS)
Store (TP1L, CTHY)
}
}

ThermalZone (THRM)
{
Name (_AL0, Package (0x01)
{
FAN
})
Method (_AC0, 0, NotSerialized)
{
If (Or (PLCY, PLCY, Local7))
{
Return (TP2H)
}
Else
{
Return (TP1H)
}
}

Name (_PSL, Package (0x01)
{
\_PR.CPU0
})
Name (_TSP, 0x3C)
Name (_TC1, 0x04)
Name (_TC2, 0x03)
Method (_PSV, 0, NotSerialized)
{
If (Or (PLCY, PLCY, Local7))
{
Return (TP1H)
}
Else
{
Return (TP2H)
}
}

Method (_CRT, 0, NotSerialized)
{
Return (TRPC)
}

Method (_TMP, 0, NotSerialized) // get current temp
{
Store (FAN1, Local0)
Return (Local0)
}

Else
{
Return (0x0B86)
}
}

Method (_SCP, 1, NotSerialized)
{
If (Arg0)
{
Store (One, PLCY)
}
Else
{
Store (Zero, PLCY)
}

Notify (\_TZ.THRM, 0x81)
}

Method (STMP, 2, NotSerialized)
{
Store (Arg1, DW00)
If (Arg0)
{
STHY (DB00, DB01, DW00)
}
Else
{
STOS (DB00, DB01, DW00)
}
}
}
}
BenReilly
First off, many thanks to those who painstakingly made these kexts. My hats off to you.

Version 277 works fine on my ASUS P7P55D-LE build, though it says there that my GPU is 'Unknown nVidia card' and that I should submit my PCI-ID to the people who made this? I have an nVidia Geforce GT 240 512mb GDDR3 128bit and its Id is 0ac3, does that help anyone?
vincent99
In the last rev 298 i don't found the winbond plugin. Why?

Slice
QUOTE (vincent99 @ Jul 7 2010, 05:42 PM) *
In the last rev 298 i don't found the winbond plugin. Why?

Use SuperIO instead.
vincent99
QUOTE (Slice @ Jul 7 2010, 04:28 PM) *
Use SuperIO instead.


Ok, thanks. The winbond worked very very well, with superio the speed of fan processor is wrong (but i think that it can be resolve by me), and tha fan exhaust is vanish. Winbond worked amazing, by the way ....
FormerlyKnownAs
Hi,

PlugIns rev301 and fakeSMC rev296 - have lost CPU fan, CPU heatsink and north bridge temp.

Click to view attachment
CODE
FakeSMC: Opensource SMC device emulator by netkas © 2009
FakeSMC: Monitoring plugins support by mozodojo © 2010
FakeSMC: Original idea of plugins and code sample by usr-sse2 © 2010
com.apple.AppleFSCompressionTypeZlib kmod start
com.apple.AppleFSCompressionTypeZlib load succeeded
AppleIntelCPUPowerManagementClient: ready
///
BSD root: disk6, major 14, minor 22
Kernel is LP64
systemShutdown false
IPv6 packet filtering initialized, default to accept, logging disabled
NVDANV50HAL loaded and registered.
Waiting for DSMOS...
kxld[org.mozodojo.FakeSMCSuperIO]: The following symbols are unresolved for this kext:
kxld[org.mozodojo.FakeSMCSuperIO]: __ZTV14FakeSMCBinding
Can't load kext org.mozodojo.FakeSMCSuperIO - link failed.
Failed to load executable for kext org.mozodojo.FakeSMCSuperIO.
Kext org.mozodojo.FakeSMCSuperIO failed to load (0xdc008016).
Failed to load kext org.mozodojo.FakeSMCSuperIO (error 0xdc008016).
IntelThermal: CPU family 0x6, model 0x17, stepping 0x7
IntelThermal: Found 4 cores 0 threads
IntelThermal: CPU Tjmax 100
DSMOS has arrived
PTKawainVi: started
Vendor ID: 10de, Device ID: 5e2
Architecture: 800000
nvidia GeForce GTX 260
AppleRTL8169Ethernet: Ethernet address
NTFS driver 3.2 [Flags: R/W].
NTFS volume name Windows7 Boot, version 3.1.
NTFS volume name Windows7 x64, version 3.1.
ACPI_SMC_GPU_CtrlLoop: NOTE: ignoring mismatched primary control with ID:18



Cheers
D
vincent99
I think there is a problem with last rev 304. fakesmc doesn't load never. I compile it correctly as in previous rev, but now this rev doesn't load.
Slice
There is rewriting of all kexts so new revision is not finished yet.
Mojodojo
FakeSMC not loading on startup fixed.
vincent99
QUOTE (Mojodojo @ Jul 10 2010, 07:15 AM) *
FakeSMC not loading on startup fixed.


Ok.

I have lost yet speed of fans (processor and exaust, my two fans) Speed of my 8800 gt is wrong yet (too fast)

winbond on p5q

With previous winbond plugin that values were corrected
Mojodojo
In rev307, compiled binaries anyone can now download from assembla. If you lost your fans you can force its detection. See info.plist.
JaS
Just wanted to say thanks for the latest version of FakeSMCSuperIO and the new FakeSMC 2.7.2 , detection is working great. still testing to see how fan control is working. Specs in sig

JaS

With Software Fan control enabled



EDIT: there were fan speed reporting issues with fan control enabled. 0 fan speed being reported or negative numbers, disabling the fan control in the info plist of FakeSMCSuperIO and allowing the bios to control them fixed the issue. Im sure you will get it worked out smile.gif

Without Software Fan control enabled

Mojodojo
QUOTE (JaS @ Jul 10 2010, 10:51 AM) *
EDIT: there were fan speed reporting issues with fan control enabled. 0 fan speed being reported or negative numbers, disabling the fan control in the info plist of FakeSMCSuperIO and allowing the bios to control them fixed the issue. Im sure you will get it worked out smile.gif


I think I've fixed it. Try the new binary, have uploaded it to assembla.
JaS
QUOTE (Mojodojo @ Jul 10 2010, 05:24 AM) *
I think I've fixed it. Try the new binary, have uploaded it to assembla.


Hey Mojodojo,

Sorry to report that the issue still arises with Software Fan Control Enabled see pics below
Click to view attachment
Click to view attachment
This is a "lo-fi" version of our main content. To view the full version with more information, formatting and images, please click here.
Invision Power Board © 2001-2012 Invision Power Services, Inc.