Re: Kernel Extensions
Re: Kernel Extensions
- Subject: Re: Kernel Extensions
- From: Joshua Graessley <email@hidden>
- Date: Mon, 17 Nov 2003 10:15:36 -0800
On Nov 16, 2003, at 1:14 PM, Justin Walker wrote:
>
On Friday, November 14, 2003, at 05:21 PM, George Andre wrote:
>
>
A word of warning about the word of warning :-}
>
None of the symbols in the kernel can be relied on, unless Apple has
>
committed to supporting their presence and meaning. While most of
>
them are unlikely to change (there's no percentage in making arbitrary
>
changes when you are intending to import updates from the outside), it
>
is possible that they will be made unavailable, in favor of more
>
supportable "accessor functions" or some similar scheme.
>
>
The odds-makers in Vegas and London have yet to publish their take on
>
this event.
If I was a betting man, I wouldn't bet on all those nifty symbols
hanging around for long. Then again, I have insider information, so
perhaps betting would get me locked up ;). As mentioned at WWDC earlier
this year, we are working on all those accessor functions and a cleaner
but more limited API for kernel extensions.
If you can find any way to do your work outside of the kernel, you will
have a much more pleasant experience in almost every aspect. Debugging
kernel panics isn't fun. Dealing with your kext breaking whenever the
kernel changes isn't fun. Finding good documentation on writing kernel
extensions isn't easy. Writing to the kernel "APIs" isn't fun either.
They're only APIs in the loses sense. The reality is that nearly every
symbol is exported, giving you an opportunity to exploit various design
decisions and flaws.
-josh
[demime 0.98b removed an attachment of type application/pkcs7-signature which had a name of smime.p7s]
_______________________________________________
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.