MacOS – OSX 10.6.8 crash: suspected audio layer hard freeze on an iMac CWI39445

The following is a simple question that's difficult for me to diagnose because it causes a hard freeze that the system daemons appear to not be able to log. As a power user, I'm in full control of my system as root and expect, from circumstantial evidence, that this is related to an audio driver issue with the latest update to 10.6.8. I do not know how to fix the problem, however.

First, let's compare a standard case of improper shutdown, according to system.log, wherein I held down the power button to reboot the machine. Note that the hostname of my box in these logs has been replaced with hostname to prevent identifiability.

Oct 18 12:20:29 hostname com.apple.launchd.peruser.364919944[304] (com.apple.PreferenceSyncAgent): fork() failed, will try again in one second: Resource temporarily unavailable
Oct 18 12:20:29 hostname postfix/sendmail[98904]: warning: fork: Resource temporarily unavailable
^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@
^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@
^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@
^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@
^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@Oct 18 12:21:31 localhost com.apple.kextd[10]: Can't create kext cache under / - owner not root.
Oct 18 12:21:20 localhost com.apple.launchd[1]: *** launchd[1] has started up. ***

Now, let's look at system.log during one of these hard freezes:

Oct 18 14:27:18 localhost com.apple.kextd[10]: Can't create kext cache under / - owner not root.
Oct 18 14:27:09 localhost com.apple.launchd[1]: *** launchd[1] has started up. ***
Oct 18 14:27:19 localhost DirectoryService[11]: Improper shutdown detected
Oct 18 14:27:20 localhost blued[15]: Apple Bluetooth daemon started

kernel.log, standard improper shutdown:

Oct 18 12:18:14 hostname kernel[0]: Sound assertion "0 != result" failed in AppleMikeyDevice at line 765 goto handler
Oct 18 12:18:14 hostname kernel[0]: Sound assertion "0 != result" failed in AppleMikeyDevice at line 795 goto handler
Oct 18 12:18:14 hostname kernel[0]: Sound assertion "0 != dispatchToStateMachinePosition ( dequeuedEvent )" failed in AppleMikeyDevice at line 721 goto handler
Oct 18 12:18:14 hostname kernel[0]: Sound assertion "0 != result" failed in AppleMikeyDevice at line 765 goto handler
Oct 18 12:18:14 hostname kernel[0]: Sound assertion "0 != result" failed in AppleMikeyDevice at line 795 goto handler
Oct 18 12:18:14 hostname kernel[0]: Sound assertion "0 != dispatchToStateMachinePosition ( dequeuedEvent )" failed in AppleMikeyDevice at line 721 goto handler
Oct 18 12:18:28 hostname kernel[0]: Sound assertion "0 != result" failed in AppleMikeyDevice at line 765 goto handler
Oct 18 12:18:28 hostname kernel[0]: Sound assertion "0 != result" failed in AppleMikeyDevice at line 795 goto handler
Oct 18 12:18:28 hostname kernel[0]: Sound assertion "0 != dispatchToStateMachinePosition ( dequeuedEvent )" failed in AppleMikeyDevice at line 721 goto handler
Oct 18 12:18:28 hostname kernel[0]: Sound assertion "0 != result" failed in AppleMikeyDevice at line 765 goto handler
Oct 18 12:18:28 hostname kernel[0]: Sound assertion "0 != result" failed in AppleMikeyDevice at line 795 goto handler
Oct 18 12:18:28 hostname kernel[0]: Sound assertion "0 != dispatchToStateMachinePosition ( dequeuedEvent )" failed in AppleMikeyDevice at line 721 goto handler
Oct 18 12:21:31 localhost kernel[0]: npvhash=4095
Oct 18 12:21:31 localhost kernel[0]: PAE enabled
Oct 18 12:21:31 localhost kernel[0]: 64 bit mode enabled

kernel.log, hard freeze:

Oct 18 13:45:19 hostname kernel[0]: CoreServicesUIAg[1076] Unable to clear quarantine `Headers': 93
Oct 18 13:45:19 hostname kernel[0]: CoreServicesUIAg[1076] Unable to clear quarantine `Resources': 93
Oct 18 13:45:19 hostname kernel[0]: CoreServicesUIAg[1076] Unable to clear quarantine `Sparkle': 93
Oct 18 13:45:19 hostname kernel[0]: CoreServicesUIAg[1076] Unable to clear quarantine `Current': 93
Oct 18 14:27:18 localhost kernel[0]: npvhash=4095
Oct 18 14:27:18 localhost kernel[0]: PAE enabled
Oct 18 14:27:18 localhost kernel[0]: 64 bit mode enabled

Note that the complaints from CoreServicesUIAg are consistent across both hard freezes.

I have no additional information, insofar as no .hang or .crash appear to have been generated on my system and all other logs look fine. I did, however, leave this workstation locked overnight and did not suffer the same hard freeze, and I appear to be doing just fine typing out this message on the machine with audio disabled.

Addendum: core temperature and other hardware diagnostics appeared to be reporting within completely tolerable ranges. The last read I'd received before the second crash was somewhere in the 54-55°C range, which is pretty cool as far as CPUs are concerned.

Second Addendum: this article may be related, but I have no testing data. If this is plausible, I'll simply roll back to 10.6.7 and see what happens.

Is there any information available for this problem? If so, is there a simple hotfix or solution I can apply to my machine to regain stability without rolling back to a previous version?

Thank you in advance.