Re: Kernel -> user boundary
Re: Kernel -> user boundary
- Subject: Re: Kernel -> user boundary
- From: Marek Kozubal <email@hidden>
- Date: Wed, 29 Jan 2003 03:58:14 -0500 (EST)
On Mon, 27 Jan 2003, Godfrey van der Linden wrote:
>
>IS there anything I can do or use to initiate a
>
>kernel -> user crossing without user parts help?
>
>
Not really. Do you have a user land app that you are talking to? If
>
so it is pretty easy to emulate a kernel initiated event.
>
>
What do you want to happen in user land when the kernel hits something.
I also need to be able to do this sort of communication. Though I'm fine
with having some setup at the beginning in the user -> kernel direction.
What my kext needs to do is send the user land process some information,
and then wait for a reply back from the user land process with an answer
to the query (the user daemon is doing some IOKit stuff during the wait
time). The reply time is on the order of 15 seconds so I need to be able
to release the kernel funnels while I'm waiting for a reply. This is all
in a file system kext.
In a previous conversation I had w/someone, it was suggested that socket's
might be the way to go for this communication but perhaps there is
something more appropriate. ioctl and sysctl dont feel appropriate which
is whats "suggested" for file system communcaition. I'm not familiar
w/the mach ports and the like, would these be better?
The amount of data being communicated is variable length but on the order
of 3 to 100 bytes or so.
So I guess, which communications API should I start rummaging though to
implement this? And what kind of setup from the user daemon will be
required to the kext if the kext itself cannot initiate the connection?
---
Marek Kozubal
email@hidden
_______________________________________________
darwin-kernel mailing list | email@hidden
Help/Unsubscribe/Archives:
http://www.lists.apple.com/mailman/listinfo/darwin-kernel
Do not post admin requests to the list. They will be ignored.