cozmonate / hwsensors Goto Github PK

Tachometer key names in the Info.plist for ACPISensors.kext are wrong for what the code expects because of the special way tachometers are added as sensors internally. The end result is that if you have any ACPI tachometers (fans), you will get a KP as soon as the menu extra accesses them.

The problem is they are in the Info.plist as FANx, where 'x' is hex digit. But the code expects them to be FxAc, where 'x' is hex digit. This means the look up in the dictionary KPs when ACPISensors::getSensorValue is called for kFakeSMCTachometerSensor.

The fix I implemented in my fork is to change the Info.plist key values to match what the code is expecting, for example for first fan, F0Ac. Location of fork: https://github.com/RehabMan/HWSensors

I noticed it's OS X target is set to 10.7. Is 10.6 possible? Or is it using features specific to 10.7+?

Right now, I have a hacked version of slice's HwMonitor that works with your plugins, but it would be better to use yours...

I tried setting the target to 10.6 and it didn't work. Haven't investigated any further... not sure if the problem is small or big. Comment? Thanks!

I had never used the static analyser in xcode, but was playing around with it... There are a couple of potential issues uncovered:

geforcesensors.cpp

    // map device memory
    if ((device->pcidev = (IOPCIDevice*)provider)) {

        device->pcidev->setMemoryEnable(true);

        if ((device->mmio = device->pcidev->mapDeviceMemoryWithIndex(0))) {
            nv_debug(device, "memory mapped successfully\n");
        }
        else {
            nv_error(device, "failed to map memory\n");
            return false;
        }
    }
    else {
        nv_error(device, "failed to assign PCI device\n");  // STATAN
        return false;
    }

The nv_error macro definitely dereferences 'device' here which is NULL. Result would be a KP.

nouveau_gpio.cpp

static u16 dcb_gpio_parse(struct nouveau_device *device, int idx, int ent, u8 *ver, u8 *len, struct dcb_gpio_func *gpio)
{
    u16 data = dcb_gpio_entry(device, idx, ent, ver, len);
    if (data) {
        if (*ver < 0x40) {
            u16 info = nv_ro16(device, data);

*ver is random stack garbage here. Passed from dcb_gpio_match, from nouveau_gpio_find, where it is an uninitialized auto.

There's a few others (dead stores) and one warning about 'self' in LaunchAtLoginController.m that might be worth looking at.

This is all using Xcode 4.6.

Hello Kozlek,

I'm using ATI Radeon HD 5450 from XFX. Upto GPUSensors V 2.8 everything worked fine, but after you refactored GPU sensors I always got this error in console (from version 2.9)

"19/01/13 7:49:15.000 AM kernel[0]: RadeonSensors (pci1@0): found ATI Radeon 0x68f9
19/01/13 7:49:15.000 AM kernel[0]: RadeonSensors (pci1@0): [Error] unable to locate ATY,bin_image
19/01/13 7:49:15.000 AM kernel[0]: RadeonSensors (pci1@0): [Warning] using device-id list to resolve temperature sensor type
19/01/13 7:49:15.000 AM kernel[0]: RadeonSensors (pci1@0): adding EverGreen thermal sensor"

Could you please resolve it?

There's seems to be a problem with the keys imported from Clover EFI.
I have for example 2 REV keys:

Today I just updated to the latest git repo version and now I have no cpu frequencies or individual core temperatures showing in HWMonitor. I have a 1st gen core i5 system. Logs aren't showing anything weird either.

Why is the PKG removed? Do I need to compile this with Xcode?

Thanks.

While merging your latest changes (haven't done it in a while), it seems the build is broken. I'm getting an error from radeon_atombios.h, which #include "atombios.h" and atombios.h which is nowhere to be found in the repo?

Perhaps you forgot to 'git add' it.

When I put my system to sleep and resume from sleep it will not return from sleep and system crashes

Just to confirm I reinstalled an older build 5.0.20 and it resumes from sleep just fine without KP!

Here is KP Console Log:

2/20/13 4:45:33.000 PM kernel[0]: Backtrace 0xffffff800065011f 0xffffff800064fc99 0xffffff800064fc00 0xffffff7f810bbb2b 0xffffff7f810bb760 0xffffff800063008c 0xffffff800062c2d0
2/20/13 4:45:33.000 PM kernel[0]: Kernel Extensions in backtrace:
2/20/13 4:45:33.000 PM kernel[0]: org.netkas.driver.FakeSMC(418.0)[FB3ED069-8CE9-30D9-A3C0-B4A65632A36E]@0xffffff7f810b7000->0xffffff7f810c9fff
2/20/13 4:45:33.000 PM kernel[0]: dependency: com.apple.iokit.IOACPIFamily(1.4)[A35915E8-C1B0-3C0F-81DF-5515BC9002FC]@0xffffff7f80d40000
2/20/13 4:45:33.000 PM kernel[0]: Sleep failure code 0x00000000 0x21006700

almost perfect except mobo's temperature(syst) and voltage readings. extracted sensors info from aida if it helps.

mb-product = P7H55-M PRO
mb-manufacturer = ASUS

Let me know if I can be of any assistance.

Aida dump LINK

Is it possible to add GPU ram usage to the HWmonitor Menubar. I do retouching for a living and it would help alot to be able to monitor how much vram i have left.

Thanks

Mine is Nvidia GTX275.
If NVClockx.kext is loaded, everything is OK. GPU temp and GPU fan are well shown.
But, if NouveauSensors.kext, the latest version,is loaded, GPU temp is OK, no GPU Fan is shown.
OS is 10.8
kernel log as below:
kernel[0] [AGPM Controller] build GPUDict by Vendor10deDevice05e2
kernel[0] NouveauSensors: VBIOS successfully read from PRAMIN
kernel[0] NouveauSensors: BIT VBIOS found
kernel[0] NouveauSensors: detected an NV50 generation card (0x0a0300b1) with 896 Mib of GDDR3 memory
kernel[0] NouveauSensors:: probing I2C busses...
kernel[0] NouveauSensors:: bus: 2 device: b2
kernel[0] NouveauSensors:: bus: 2 device: a0
kernel[0] NouveauSensors:: bus: 2 device: 92
kernel[0] NouveauSensors:: bus: 2 device: 74
kernel[0] NouveauSensors:: bus: 2 device: 6e
kernel[0] NouveauSensors:: bus: 3 device: 5c
kernel[0] NouveauSensors:: found Analog Devices ADT7473 monitoring chip

I just downloaded the latest build (Thanks for the continued development!) and seem to have a problem when trying to compile with Xcode that prevents the process completing?

File '/Users/Me/Downloads/HWSensors-master/GPUSensors/GPUSensors-Prefix.pch' has been modified since the precompiled header was built

It could be me doing something wrong but I used the same method as I did yesterday that worked.

I notice that when I click on the HWMonitor icon on the menu bar, that the sensors are displayed with old values for a second or so. That means, for example, if my laptop fan has stopped where previously it was running, for one second the popup shows the fan running at its previous RPM instead of 0.

Is it possible to force an update to HWMonitor's cache of sensors before showing the popup?

Hi kozlek, I am currently trying to learn Clover and setup a USB stick for OS X installation.

Using the latest build (Not tried any older) the install process hangs at a grey screen with a dreaded spinning beach ball.

I have manually copied the kext to EFI/Kexts/10.8 (Tried other locations too) but no where I have tried it works.

Apologies if this isn't something the file should be able to do but if I use the latest build from Slice the boot and installation process works fine.

The value of Power Fan should be in Chassis fan.

I have double checked on the MB and the BIOS-screen.

I can provide a screenshot or other info if needed. :)

Thanks!

not it can defect fan speed, but it would be better to set fan speed.

I have a big big problem..every time hwsensors is running..the system locks up .. . This usually happens after the login screen. I log in and then everything loades..but when the hwsensors app start gray screen appers....any ideas ?

Hi!

I have a question about the installer that is available from the bitbucket repo. Does it install FakeSMC to /E/E or /S/L/E? I'm used to install FakeSMC + Plugins to /E/E in order to keep my OS X clean. If the installer does not install to /E/E, would there be any change to provide an option to choose between /E/E and /S/L/E?

Thank you very much in advance!

SystemUIServer crashes after waking from sleep when the HWMonitor menu bar extension is installed.

Here is part of my log file after waking from sleep:

2012-12-18 9:07:34.069 PM SystemUIServer[1077]: An uncaught exception was raised
2012-12-18 9:07:34.069 PM SystemUIServer[1077]: *** Collection <NSArrayM: 0x7fde3a07f660> was mutated while being enumerated.
2012-12-18 9:07:34.070 PM SystemUIServer[1077]: (
0 CoreFoundation 0x00007fff862f00a6 __exceptionPreprocess + 198
1 libobjc.A.dylib 0x00007fff8b0223f0 objc_exception_throw + 43
2 CoreFoundation 0x00007fff86383f98 __NSFastEnumerationMutationHandler + 232
3 HWMonitorExtra 0x0000000107f9333e -[HWMonitorEngine updateSmartSensors] + 226
4 HWMonitorExtra 0x0000000107f8c6e4 -[HWMonitorExtra updateSmartSensors] + 28
5 Foundation 0x00007fff83dd4612 __NSThread__main + 1345
6 libsystem_c.dylib 0x00007fff86068742 _pthread_start + 327
7 libsystem_c.dylib 0x00007fff86055181 thread_start + 13
)
2012-12-18 9:07:34.070 PM SystemUIServer[1077]: * Terminating app due to uncaught exception 'NSGenericException', reason: '* Collection <__NSArrayM: 0x7fde3a07f660> was mutated while being enumerated.'
*** First throw call stack:
(
0 CoreFoundation 0x00007fff862f00a6 __exceptionPreprocess + 198
1 libobjc.A.dylib 0x00007fff8b0223f0 objc_exception_throw + 43
2 CoreFoundation 0x00007fff86383f98 __NSFastEnumerationMutationHandler + 232
3 HWMonitorExtra 0x0000000107f9333e -[HWMonitorEngine updateSmartSensors] + 226
4 HWMonitorExtra 0x0000000107f8c6e4 -[HWMonitorExtra updateSmartSensors] + 28
5 Foundation 0x00007fff83dd4612 NSThread__main + 1345
6 libsystem_c.dylib 0x00007fff86068742 _pthread_start + 327
7 libsystem_c.dylib 0x00007fff86055181 thread_start + 13
)
2012-12-18 9:07:40.634 PM com.apple.launchd.peruser.501[1062]: (com.apple.SystemUIServer.agent[1077]) Job appears to have crashed: Abort trap: 6
2012-12-18 9:07:41.349 PM SystemUIServer[2120]: MenuCracker 2.2 (/Users/bobscratchit/src/HWSensors/Binaries/MenuCracker/MenuCracker.menu)
See http://sourceforge.net/projects/menucracker
MenuCracker is now loaded. Ready to accept new menu extras.
2012-12-18 9:07:41.350 PM SystemUIServer[2120]: failed to instantiate and get the principal class of bundle: NSBundle </Users/bobscratchit/src/HWSensors/Binaries/MenuCracker/MenuCracker.menu> (loaded)
2012-12-18 9:07:41.411 PM SystemUIServer[2120]: MenuCracker: Allowing "HWMonitorExtra".
2012-12-18 9:07:41.411 PM SystemUIServer[2120]: Could not load menu extra NSBundle </Users/bobscratchit/src/HWSensors/Binaries/MenuCracker/MenuCracker.menu> (loaded) for Class (null)
2012-12-18 9:07:43.146 PM ReportCrash[2114]: Saved crash report for SystemUIServer[1077] version 1.7 (369.2) to /Users/bobscratchit/Library/Logs/DiagnosticReports/SystemUIServer_2012-12-18-210743_bobscratchit-pc.crash

This has been an issue for some time. Reporting it so you're aware. Latest git sources. Mountain Lion 10.8.2 yields this:

Hi,
Is it possible to use the application "smcFanControl"?

Or could you add some support to make us be able to use fan speed control on hackintosh?

I know it's easier said than done.... It's just an idea. :P

Thanks for your work.
I have installed your branch on x220 lenovo under ML.

CPU frequencies and temperatures detected, SSD detected as well.
But I get this message: SuperIODevice: [Fatal] found unsupported ship ID=0xffff

I think this is related to IT87 chip. Can I do something to integrate this chip to your FakeSMC branch ?

I on a l502x non uefi device with OS X 10.8.2 (iatkos ML2 with custom DSDT and other kext from this package http://www.insanelymac.com/forum/topic/257950-guide-dell-xps-15-l502x-early-2011-snow-leopard-install-possibly-l702x-too/page__st__1540#entry1822872)
Whenever I install your full package with HWSensor and and all the Fakesmc plugins I start getting weird bugs. My text cursor starts trailing text when typing, flash video has a lot of ghosting, a lot of UI freezing and gfx glitches are to be seen.

However, when I trash the plugins and the app, I get only frequent UI freezes but not the other issues (gfx glitches are there but they occur much less and cover much less of the screen real estate).

Edit: I installed the latest package from bitbucket (5.1.9 -> https://bitbucket.org/kozlek/hwsensors/downloads/Installer%205.1.9.zip)

I have a dual Xeon build (12 cores); I only have six cores showing up in HW Monitor.

I alluded to this problem in issue #25, but I think my comment was ignored. Quote:
"Either that, or there is a race condition (are calls to ::start guaranteed to be serialized?)"

Well, the answer to that question "are calls to ::start guaranteed to be serialized?" is NO. The system will start kernel extensions in parallel using multiple cores to make startup faster. This means that kernel extensions must protect shared data against simultaneous access during the startup sequence using a mutex or other mechanism.

I have created a new fork that has this fix and a few others at: https://github.com/RehabMan/OS-X-FakeSMC-kozlek

You will see the changes in commit: 0ce9001366

In particular, access to 'keys' must be protected, and certain operations which are two part must be handled as one atomic operation (you will see this with getVacantGPUIndex/takeVacantGPUIndex). You cannot allow two threads to acquire an index with getVacant at a time before one has a chance to claim it with takeVacant, as they will both end up with the same index.

There is quite a few changes because it is much easier to implement locks when using a single exit point.

I've just upgraded to the latest git version of your hwsensors from an older version (from around september). I immediately get a kernel panic upon boot up and it mentions gpusensors. Once I delete gpusensors I am able to boot and restore an older fakesmc and plugins.

My gpu is a nvidia geforce 560 ti 1GB

del

Hey, i was just wondering how i do install, i just made a hackintosh and i need your fakeSMC :S

Just something I noticed while merging in some of your recent changes. You are using snprintf inside the loop to space-pad the passed in 'name' to getKey:

   while ((key = OSDynamicCast(FakeSMCKey, iterator->getNextObject()))) {

            // Made the key name valid (4 char long): add trailing spaces if needed
            char validKeyNameBuffer[5];
            snprintf(validKeyNameBuffer, 5, "%-4s", name);

            ...

Since 'name' is invariant, the creation of validKeyNameBuffer can be moved outside the loop for efficiency:

   // Made the key name valid (4 char long): add trailing spaces if needed
   char validKeyNameBuffer[5];
   snprintf(validKeyNameBuffer, 5, "%-4s", name);

   while ((key = OSDynamicCast(FakeSMCKey, iterator->getNextObject()))) {

             ...

Extract from install log attached below. Can reproduce from command line:

kextcache -system-prelinked-kernel
Kernel file /mach_kernel does not contain requested arch: i386
FakeSMC.kext is not authentic; omitting.
/System/Library/Extensions/FakeSMC.kext is not authentic; omitting from prelinked kernel.
Can't use FakeSMC.kext - not linked.
Prelink failed for org.hwsensors.driver.MotherboardSensors; aborting prelink.
Failed to generate prelinked kernel.

Installer log extract:

Sep 30 14:46:26 Akita-X.local installd[842]: PackageKit: Failed to move active dyld cache file /var/db/dyld/dyld_shared_cache_x86_64 to /var/run: No such file or directory (2)
Sep 30 14:46:26 Akita-X.local installd[842]: PackageKit: Moved active dyld cache file dyld_shared_cache_x86_64.map to /var/run
Sep 30 14:46:26 Akita-X.local install_monitor[850]: Temporarily excluding: /Applications, /Library, /System, /bin, /private, /sbin, /usr
Sep 30 14:46:26 Akita-X.local installd[842]: PackageKit: kextcache -system-caches
Sep 30 14:46:27 Akita-X.local installd[842]: PackageKit: kextcache -update-volume / -Installer
Sep 30 14:46:28 Akita-X.local installd[842]: kextcache: FakeSMC.kext is not authentic; omitting.
Sep 30 14:46:28 Akita-X.local installd[842]: kextcache: //System/Library/Extensions/FakeSMC.kext is not authentic; omitting from prelinked kernel.
Sep 30 14:46:31 Akita-X.local installd[842]: kextcache: Can't use FakeSMC.kext - not linked.
Sep 30 14:46:31 Akita-X.local installd[842]: kextcache: Prelink failed for org.hwsensors.driver.MotherboardSensors; aborting prelink.
Sep 30 14:46:31 Akita-X.local installd[842]: kextcache: Failed to generate prelinked kernel.
Sep 30 14:46:31 Akita-X.local installd[842]: kextcache: Child process /usr/sbin/kextcache[856] exited with status 71.
Sep 30 14:46:31 Akita-X.local installd[842]: kextcache: Error 107 rebuilding /System/Library/Caches/com.apple.kext.caches/Startup/kernelcache.
Sep 30 14:46:31 Akita-X.local installd[842]: PackageKit: releasing Spotlight indexing
Sep 30 14:46:31 Akita-X.local install_monitor[850]: Re-included: /Applications, /Library, /System, /bin, /private, /sbin, /usr
Sep 30 14:46:31 Akita-X.local installd[842]: PackageKit: releasing backupd
Sep 30 14:46:31 Akita-X.local installd[842]: PackageKit: allow user idle system sleep
Sep 30 14:46:31 Akita-X.local installd[842]: PackageKit: Install Failed: Error Domain=PKInstallErrorDomain Code=121 "An error occurred while updating system extension information." UserInfo=0x7f9d1491c020 {NSUnderlyingError=0x7f9d14946990 "The operation couldn’t be completed. (kextcache error 71.)", NSLocalizedDescription=An error occurred while updating system extension information., arguments=(

This is something I've done with my projects. I add a link to the source code repo in the Info.plist. When I first started dong my hack project, I was frustrated in trying to find the source to various kext packages being passed around. It is contrary to the license for most open source projects to distribute without a link or copy of the source code. So, what I decided to do is add a link right in the Info.plist, so there is no question where the software comes from:

Example from my fork of this project:

       <key>Source Code</key>
       <string>https://github.com/RehabMan/OS-X-FakeSMC-kozlek</string>

The idea being the string is something you can plug into your browser and arrive at a project page.

What do you think?

Not Use GPUSensors.kext System Run Run。

Use GPUSensors.kext System Collapse，Get into System After,Whole Screen Is White Then Collapse。

I am sorry!

My English is very poor!

I used to get 3 fan readings on my desktop from LPCSensors (nuvoton). Works on commit ee211. Does not work on 76425. Also tried latest commit 16460 and same result: not working. Let me know if you need additional debug info.

Board: GA-Z77-DS3H
Gfx: Gigabyte GV-N66TOC-2GD NVIDIA GTX 660 Ti
OSX: 10.8.2

I just installed the latest HWSensors.pkg. As soon as I add the GPU Plugin and reboot, my system freezes if HWMonitor is enabled. iStat Menues 4 freezes too, but only occasionally. (Unfortunately I don't get any kernel panic message or anything, it just freezes.)

I was about to send a darwindumper dump and downloaded the tool, but I'm not sure what information it gathers, that I don't want to publicily share. Do you need all of it or only specific parts of it?

Just tried the latest build, no panics but my GPU temp and fan speed (GTX285) are no longer displayed.

Would this be anything to do with the previous problem someone had?

With NVClockX loaded it is impossible to use verbose mode or single user mode. It causes kernel panic. Normal boot works fine, but every time if I use verbose mode or single user mode, I get kernel panic.

Last message from nvclockx is: overclocking

I have 10.7.4 and nvidia gtx 570 in sli ( same problem occurs with only one card too).

Please tell me what kind of information you need more.

Thanks in advance
Smolderas

Hi !

Please can you add support for this GPU ?

for now i'm getting this :
12/9/12 4:59:23.000 PM kernel[0]: NTFS driver 3.10 [Flags: R/W].RadeonMonitor: [Error] unknown device id
12/9/12 4:59:23.000 PM kernel[0]: RadeonMonitor: [Fatal] can't initialize driver
12/9/12 5:13:05.000 PM kernel[0]: RadeonMonitor: [Error] unknown device id

And here is the data of the card:

Chipset Model: AMD Radeon HD 7xxx
Type: GPU
Bus: PCIe
PCIe Lane Width: x16
VRAM (Total): 3072 MB
Vendor: ATI (0x1002)
Device ID: 0x6798
Revision ID: 0x0000

Also this :

Device Id:1002 6798
Subsystem Id:1462 2770

Thanks in advance !

The code to evaluate ACPI methods in ACPISensors::getSensorValue is using the UInt64 version of IOACPIPlatform::evaluateInteger. This appears to cause a KP, where using the UInt32 version (as does slice's branch) does not.

There is also one more case in PTIDSensors where this method is used. My DSDT doesn't have the PTID, so I didn't test that, although changing it might be a good idea as well...

My branch: https://github.com/RehabMan/HWSensors

While merging your latest changes I noticed an issue in updateSmcSensorsList (HwMonitorEngine.m):

- (void)updateSmcSensorsList:(NSArray *)sensors
{
    [_sensorsLock lock];

    if (!sensors) return; // [self updateSmcSensors];
    ...

I'm not sure under what circumstances it may be called with nil sensors, but in that case the result would be deadlock later, as the mutex is acquired and not released.

Fix:

- (void)updateSmcSensorsList:(NSArray *)sensors
{
    if (!sensors) return; // [self updateSmcSensors];

    [_sensorsLock lock];

   ...

Hi kozlek,
Could you add LPCSensors config for P67A-UD3R-B3 in next update?

Here it goes:
http://www.mediafire.com/?fy93ena7j1113wk

Thanks! :)

There is no problem of booting up the hack. But console shows this stuff which is worth taking a look into it.
12/16/12 4:29:25.000 PM kernel[0]: HWSensors Project Copyright 2012 netkas, slice, usr-sse2, kozlek, navi, THe KiNG. All rights reserved.
12/16/12 4:29:25.000 PM kernel[0]: waitForSystemMapper
12/16/12 4:29:25.000 PM kernel[0]: Backtrace 0xffffff800065011f 0xffffff800064fc99 0xffffff800064fc00 0xffffff7f81312cdf 0xffffff7f81312917 0xffffff800063008c 0xffffff800062c2d0
12/16/12 4:29:25.000 PM kernel[0]: Kernel Extensions in backtrace:
12/16/12 4:29:25.000 PM kernel[0]: org.netkas.driver.FakeSMC(4.3.21)[A9B2B0D8-6D78-3581-A99A-610E278A645E]@0xffffff7f8130e000->0xffffff7f81320fff
12/16/12 4:29:25.000 PM kernel[0]: dependency: com.apple.iokit.IOACPIFamily(1.4)[A35915E8-C1B0-3C0F-81DF-5515BC9002FC]@0xffffff7f807fb000

The issue is, i can only get any data when the menu bar extension is installed, but while it is installed, i get Kernel panics when i try to open disk utility

Almost 200MB now. I think the problem is checking in the binaries. There is currently 306 copies of HWSensors.pkg, and then even more older binaries (before the package installer).

I wonder if these can be pruned somehow? There is this link: http://stackoverflow.com/questions/11255802/after-deleting-a-binary-file-from-git-history-why-is-my-repository-still-large. And this one: http://stevelorek.com/how-to-shrink-a-git-repository.html

I noticed that Info.plist IONameMatch for ACPISensors uses MONITOR. With the current Intel iasl compiler, you cannot compile a DSDT which contains an _HID 'MONITOR', as the compiler is a bit more picky. In my branch I have added 'FAN00000' and 'MON00000'

FAN00000 because that is what we currently use on the 4530s ProBook DSDT in wide use.
-and-
MON00000 because that is what I probably should have named it when I did this a few months ago...

My branch: https://github.com/RehabMan/HWSensors

hi,
first of all I want you to thank you for all your great work. Everything works great, just cosmetics errors I hope.

Issue:
Chameleon detects my cpu i5 3750K family 0x6 model 0x3a.. But after when boot log comes it says:

kernel[0]: ACPI_SMC_PlatformPlugin::start - waitForService(resourceMatching(AppleIntelCPUPowerManagement) timed out
kernel[0]: WARNING: IOPlatformPluginUtil : getCPUIDInfo: this is an unknown CPU model 0x3a
kernel[0]: -- power management may be incomplete or unsupported
kernel[0]: [AGPM Controller] unknownPlatform

I got full log here cause here cant upload files I guess:
http://www.insanelymac.com/forum/index.php?showtopic=282123

1. ACPISensors.cpp:15:1 <Spelling=/Applications/Xcode.app/Contents/Developer/Platforms/MacOSX.platform/Developer/SDKs/MacOSX10.8.sdk/System/Library/Frameworks/Kernel.framework/Headers/libkern/c++/OSMetaClass.h:1702:5>: current parser token 'const'
2. ACPISensors.cpp:15:1 <Spelling=/Applications/Xcode.app/Contents/Developer/Platforms/MacOSX.platform/Developer/SDKs/MacOSX10.8.sdk/System/Library/Frameworks/Kernel.framework/Headers/libkern/c++/OSMetaClass.h:1701:39>: LLVM IR generation of declaration 'ACPIMonitor::gMetaClass'
3. ACPISensors.cpp:15:1 <Spelling=/Applications/Xcode.app/Contents/Developer/Platforms/MacOSX.platform/Developer/SDKs/MacOSX10.8.sdk/System/Library/Frameworks/Kernel.framework/Headers/libkern/c++/OSMetaClass.h:1701:39>: Generating code for declaration 'ACPIMonitor::gMetaClass'

clang: error: unable to execute command: Segmentation fault: 11

Installed a EVGA GTX 660 2gb SC last week. HWmonitor wont show any temps or items for it. Istats menu wont show video ram either. Any way to fix gpusensors so my card works?
im using v1.0.119

Thanks

Not really an issue:
I suggest to add new profiles for the Asus P8Z77-V LK and P8Z77-M. They are equal.
The chip being used is the Nuvoton,NCT6779D.

Link to the Info.plist: http://www.sendspace.com/file/qggi2x