Jump to content

10.6.8 officially released

shenor

By shenor,
in New Releases and Updates

 Share
371 posts in this topic

Recommended Posts

kerr
kerr
Posted
Posted

i noticed strange report by lspci.

 

 

pcilib: 0000:02:00.0 64-bit device address ignored.

 

 

it was not present with 10.6.7. could it be related to "pci configuration begin" thing ?

 

 

how to tell what device is at this address ?

Link to comment
Share on other sites
jpa
jpa
Posted
Posted
Updated to 10.6.8: Everything works fine on arch x86_64:

 

Asus P5K Vanilla

Firmware: 1201

AC'97 enabled

ACPI 2.0 tables enabled

graphic card: GTX 470

dsdt dumped on Linux while GTX470 installed, manually corrected all issues and patches.

 

smbios: MacPro4,1

Snow Leopard 10.6.8

 

Chameleon_2.0_RC5_r899 (prevents the 'Stop' at 'PCI configuration begin'. No need for rollback of AppleACPI.kext and IOPCI.kext) http://www.insanelymac.com/forum/index.php?showtopic=231075

 

Kext in /E/E:

- LegacyAppleAHCIPortJMicronJMB36xSATA.kext (eSATA only)

- PlatformUUID.kext

- UUID.kext

- fakesmc.kext 3.1.0 (prevents the 'Stop' at 'PCI configuration begin'. No need for rollback of AppleACPI.kext and IOPCI.kext) http://www.projectosx.com/forum/index.php?showtopic=1643

 

Kext in /S/L/E:

- AppleGraphicsPowerManagement.kext (modified for GTX 470)

- AppleHDA.kext (10.6.7 - 1.9.9f12 binary patched for ALC883 / microphone works)

- IOATAFamily.kext (modified for ICH9)

- AttansicL1Ethernet.kext (Mac Address changed to fit 00:1f)

 

Nvidia from latest MacBookPro update (If you need CUDA then keep it and don't update to the latest retail drivers. OpenCL works on both here):

- GeForceVADriver.bundle

- NVDAGF100Hal.kext

- GeForce.kext

- NVDANV40Hal.kext

- GeForce7xxxGLDriver.bundle

- NVDANV50Hal.kext

- GeForceGA.plugin

- NVDAResman.kext

- GeForceGLDriver.bundle

 

Kext removed from /S/L/E:

- AppleTyMCEDriver.kext (ECC Parity Check / You don't have it thus you don't need it)

 

Evoke:

while sleep 1; do rm -rf /System/Library/Extensions/AppleTyMCEDriver.kext; done

to prevent it from load while updating to 10.6.8

 

Cheers

 

please, upload the files used.

 

method?

 

works in my machine?

Link to comment
Share on other sites
Macslap
Macslap
Posted
Posted

I just updated to 10.6.8 (from 10.6.7) with no issues. As a matter of fact....my second monitor connected by DVI is now recognized (main monitor connected via HDMI).

 

Very happy hackintosh camper.

Link to comment
Share on other sites
Gringo Vermelho
Gringo Vermelho
Posted
Posted
Why are people requesting the 10.6.8 kexts separately?

Because they replaced them with 10.6.7 kexts without backing them up first and now they want to try putting them back.

Could you or anyone please upload AppleACPIPlatform and IOPCIFamily.kext from 10.6.7 for me

Already uploaded to this thread more than once. Read the thread.

i noticed strange report by lspci.

pcilib: 0000:02:00.0 64-bit device address ignored.

how to tell what device is at this address ?

uhhh..?! Gee I don't know, use LSPCI maybe?

Screen_shot_2011_06_26_at_1.11.35_PM.jpeg

Why does my wifi constantly disconnect and reconnect now? I'm using nawcoms 10.6.6 kernel.

Already discussed earlier in this thread.

Anyone with wireless issues should try adding the appropriate SMboardproduct to their /Extra/smbios.plist.

See the end of this post: http://www.insanelymac.com/forum/index.php...t&p=1703018

Link to comment
Share on other sites
Gringo Vermelho
Gringo Vermelho
Posted
Posted

I am using that version of FakeSMC and a newer version of Chameleon than was mentioned in noctrl's post and I still had the display freeze at "PCI Configuration Begin".

 

Don't get your hopes up.

Release_10.6.zip

EDIT: Do not blindly install everything from this archive!

Visit the HWSensors subforum over on the ProjectOSX forums to learn what each kext is for.

Link to comment
Share on other sites
oSxFr33k
oSxFr33k
Posted
Posted

@3.14r2 and others asking about the PCI messages.

 

My friend just did the combo update on his real mac and I had him boot into verbose and he has the same PCI messages we are seeing so its normal

 

@Gringo Vermelho,

 

Thanks so much for the fix.

 

I can confirm that adding this to my smbios.plist fixed my broadcom wireless:

 

<key>SMboardproduct</key>

<string>Mac-F22589C8</string>

 

 

Changing _ADR to _UID in the dsdt fixed the boot stopping from right after the PCI stuff.

 

 

This worked on my Dell XPS M1530 now to see if I need to do this for my Asus G51jx Laptop i7 quad first generation after I do the combo update first.

 

 

It boots fast and runs great thanks again!!

Link to comment
Share on other sites
genmodorg
genmodorg
Posted
Posted
App Store does not work anymore: no GUID available. I delete NetworkInterface.plist a couple of times, no success after each reboot. iTunes works. Any ideas?

I'm also having this problem; tried the suggested fix to no avail.

 

P6T w/ vanilla SL 10.6.8

Link to comment
Share on other sites
MacKonsti
MacKonsti
Posted
Posted
I am using that version of FakeSMC and a newer version of Chameleon than was mentioned in noctrl's post and I still had the display freeze at "PCI Configuration Begin".

Don't get your hopes up.

 

I agree with you. Nothing changed with the newer FakeSMC v3.1.x and latest Chameleon R10xx. On my real MacBook Pro, after the update and in verbose mode, I read:

 

IOAPIC: Version 0x20 Vectors 64:87
ACPI: System State [S0 S3 S4 S5] (S3)
PFM64 0xf10000000, 0xf0000000
[ PCI configuration begin ]
console relocated to 0xf10030000
PCI configuration changed (bridge=5 device=1 cardbus=0)
[ PCI configuration end, bridges 7 devices 17 ]
AppleIntelCPUPowerManagement: (built 16:44:45 Jun  7 2011) initialization complete
mbinit: done (64 MB memory set for mbuf pool)

 

See the "console relocated to 0xf10030000" message. Where does this come from? Could this be the key to solving the pause?

 

Can anyone see or guess?!? Could this be related to FakeSMC and the related model of our hackintosh? On my mobo, I declare it as MacPro3,1...

Link to comment
Share on other sites
mitch_de
mitch_de
Posted
Posted
@3.14r2 and others asking about the PCI messages.

Changing _ADR to _UID in the dsdt fixed the boot stopping from right after the PCI stuff.

 

This worked on my Dell XPS M1530 now to see if I need to do this for my Asus G51jx Laptop i7 quad first generation after I do the combo update first.

It boots fast and runs great thanks again!!

 

Can you post an bigger code part of that area (as code box) to look into ?

So that may happen only to some bios (= mainboards), which before 10.6.8 / 10.7 didnt habe any problems ?

Or does that happen to all bios and depends onyl of gpu (i read mostly fermi gpu users have that problem).

Link to comment
Share on other sites
oSxFr33k
oSxFr33k
Posted
Posted

@mitch_de,

 

Post # 127

 

http://www.insanelymac.com/forum/index.php...t&p=1703409

 

Third link from above link.

 

http://www.insanelymac.com/forum/index.php?showtopic=207635

 

Around the middle of the page it shows before and after all I did was change _ADR to _UID in my case.

 

I am also using the latest version of FakeSMC

Link to comment
Share on other sites
Gringo Vermelho
Gringo Vermelho
Posted
Posted

No, it happens on some hacks with ATI cards too. I think the 5750? I don't keep up with ATI stuff (have my hands full as it is).

 

It does continue to boot but the display stops updating after the PCI configuration Begin message. When I'm not booting in verbose mode, I see the Apple logo but no spinner, no animation. It's just a graphics freeze. Once it arrives at the desktop everything is back to normal.

 

/EDIT - patched 10.6.8 IOPCIFamily.kext now available, see this post:

http://www.insanelymac.com/forum/index.php...t&p=1706087

 

I believe the video freeze is a symptom of some underlying factor that we don't understand yet - I don't think it can possibly be a video driver issue because at that point (PCI Configuration Begin) the video drivers haven't loaded yet. It's too early in the boot process. You can verify this by looking at kernel.log via console.app - if you look at the timestamps you'll see that the video drivers load much later.

 

Konsti: Does your MacBook Pro have nvidia graphics?

 

One minor issue: App Store does not work anymore: no GUID available. I delete NetworkInterface.plist a couple of times, no success after each reboot. iTunes works. Any ideas?

Deleting NetworkInterfaces.plist is pointless if you haven't made any configuration changes - it will just generate a new one identical to the old one on the next reboot. You need to do some of this:

 

Set EthernetBuiltIn=y in /Extra/com.apple.Boot.plist.

 

If EthernetBuiltIn=y doesn't do the trick by itself, you probably need to set your PCI Root UID manually as well - at the boot prompt, type PciRootUID=0 - if App Store still doesn't work then reboot and try PciRootUID=1.

 

You should also delete NetworkInterfaces.plist between reboots so that it can be rebuilt to reflect your changes.

 

Once you find which one works for you, apply this DSDT fix to allow Chameleon to detect and report your PCI Root UID correctly:

http://www.insanelymac.com/forum/index.php?showtopic=207635

Note that if your PCI Root UID is 1 rather than 0 (common on Gigabyte boards) you should probably use one (0x01) rather than zero....I don't know if changing it from 1 to 0 actually hurts anything though.

Link to comment
Share on other sites
oSxFr33k
oSxFr33k
Posted
Posted

@kerr,

 

I see that also under LSPCI but twice. Could that be something to do with Thunderbolt which now shows up in the System Profiler?

 

pcilib: 0000:0b:00.0 64-bit device address ignored.

pcilib: 0000:01:00.0 64-bit device address ignored.

Link to comment
Share on other sites
Gringo Vermelho
Gringo Vermelho
Posted
Posted

It's possible that all this PCI Configuration Begin stuff is related to Thunderbolt support being added to OS X.

http://en.wikipedia.org/wiki/Thunderbolt_%28interface%29

I guess this could also explain why it's affecting graphics.

 

It looks to me like the address referred to in LSPCI is that of the PCI Bridge. This is the place where the PCI Root UID goes..

We're definitely onto something but we need some of the ACPI code geniuses to help us. I have a strong feeling that this issue can be fixed in DSDT by adding something to the PCI Bridge code.

Link to comment
Share on other sites
kerr
kerr
Posted
Posted

oSxFr33k, Thubnderbolt is a non essential device for OS X to work. On the other hand you don't have it so no device, no problem. Right ? LSPCI shows what devices you have and now the 64 bit address of one of them is igonred. The question is whether an update messes up with either the device configuration or the LSPCI software itself and making it unable to read specific device.

 

Generally speaking, i think that we should first figure out what OS X reconfigures and why. Then it will be easy to fix it.

Link to comment
Share on other sites
Gringo Vermelho
Gringo Vermelho
Posted
Posted

Yes...but no, you are missing the bigger picture I think.

 

Be careful with your wording - LSPCI is not ignoring a device, it is ignoring the 64-bit address of a device.

LSPCI ignoring, not recognizing or complaining about a device has no bearing on the functionality of the device. If anything, LSPCI just needs to be updated.

AppleACPIPlatform and IOPCIFamily.kext were updated in 10.6.8, it's extremely likely that they were updated because of Thunderbolt. If this is true, then it doesn't matter if we have Thunderbolt hardware or not, 10.6.8 adding support for it is now directly affecting us.

my chip audio is ALC 889..

I have installed the old ones kext and it does not work, as I can resolve?

If you want to keep using your old fixes you need to replace AppleHDA with the one from 10.6.7 as well.

Otherwise you will have to start over and patch your audio in the same way that it's done for Lion - AppleHDA was updated with 10.6.8 and it works like the one in Lion now.

 

If you're using VoodooHDA you might need to delete AppleHDA. Sometimes AppleHDA loads even if it does not work with your audio device and this can cause a conflict with VoodooHDA.

Link to comment
Share on other sites
oSxFr33k
oSxFr33k
Posted
Posted
oSxFr33k, Thubnderbolt is a non essential device for OS X to work. On the other hand you don't have it so no device, no problem. Right ? LSPCI shows what devices you have and now one of them is igonred. The question is whether an update messes up with either the device configuration or the LSPCI software itself and making it unable to read specific device.

 

Generally speaking, i think that we should first figure out what OS X reconfigures and why. Then it will be easy to fix it.

 

 

I wasn't thinking your right. I will look at my other XPS M1530 LSPCI before I update it to 10.6.8 and take a screenshot then compare it to the other XPS M1530 10.6.8.

 

@MacKonsti,

 

SInce you have the real MacBookPro could you run LSCPI for us to see if its just an LSPCI application issue or not? I am using EVO's LSPCI.

 

 

EDITED about an Hour later:

 

I checked my Wifes D830 didn't have time yet to check the other XPS M1530 but it could be the Cardbus readout or the Card reader?

 

My LSCPI on the XPS M1530 does not show any cardbus details so I am suspecting that could be it and it does have two slots thus the two ignored PCI addresses.

 

Can anyone else confirm a LSCPI before and after a 10.6.8 update?

Link to comment
Share on other sites
kerr
kerr
Posted
Posted
Be careful with your wording - LSPCI is not ignoring a device, it is ignoring the 64-bit address of a device.

 

oh, yes sorry

 

AppleACPIPlatform and IOPCIFamily.kext were updated in 10.6.8

 

IOPCIFamily.kext is updated but the the first one - no. In fact i don't have this kext. Where is it ? In /S/L/E i have AppleACPIPlatform.kext and AppleAPIC.kext and both of them are from 01.08.2009.

Link to comment
Share on other sites
magnifico
magnifico
Posted
Posted
1) Google/forum search for it

 

2) Use Pacifist to extract it from the 10.6.7 combo update

 

3) Wait for someone to attach it here for you

 

I would attach mine if it wasn't patched in several places to support my hardware.

I replace old kext ...but not work :rolleyes:

Link to comment
Share on other sites
koliberK
koliberK
Posted
Posted

I do some tests on my Dell E6400 on spare HDD.

On SL OSX 10.6.7 installed combo update 10.6.8 from apple website. After replaced kexts: AppleACPIPlatform.kext v.1.3.5 and IOPCIFamily.kext v.2.6 (from 10.6.7) and system boot up, but is very slow. No WiFi and PM.

 

Just like kerr pointed, we should wait and figure out what update reconfigures.

 

AppleACPIPlatform.kext v.1.3.5 and IOPCIFamily.kext v.2.6 from 10.6.7 [491.84 K]

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