Re: Dual load -- was Re: sub-KEXT question
Re: Dual load -- was Re: sub-KEXT question
- Subject: Re: Dual load -- was Re: sub-KEXT question
- From: David Elliott <email@hidden>
- Date: Mon, 9 Feb 2009 22:34:31 -0500
On Feb 9, 2009, at 6:22 PM, Kevin Brock wrote:
The other oddity with this sample is that on 10.5 if I do touch/
killall/kextstat I see:
117 0 0x5b529000 0x5000 0x4000
com.apple.dts.driver.SimpleDriver (2.0) <6 5 4>
118 0 0x5b4a1000 0x4000 0x3000
com.apple.dts.driver.SimpleDriver_10_4 (2.0) <6 5 4>
And I thought that the IOProbeScore was supposed to prevent that...
The IOProbeScore prevents the driver class instance from starting. It
does this because the driver class instance in one driver will wind up
with a higher probe score than the other one when all is said and done.
By that I mean after the probe methods of both instances have
completed because that is when the kernel finally makes the decision.
For the probe methods to be called in the first place it is of course
necessary to load both kernel extensions and instantiate the driver
class in each.
After a few minutes it seems kextd will unload kexts without any
active object instances.
-Dave
_______________________________________________
Do not post admin requests to the list. They will be ignored.
Darwin-kernel mailing list (email@hidden)
Help/Unsubscribe/Update your Subscription:
This email sent to email@hidden