Re: Kernel Language support
Alan Cox (alan@lxorguk.ukuu.org.uk)
Sun, 19 Jan 1997 14:39:34 +0000 (GMT)
> That's almost 12000 printk's in the kernel. Not that I'm saying it
> wouldn't be worth doing, just saying that it'd be more than one
> afternoon's project.
>
> FWIW, I liked the idea about the
> #define ERROR_CONDITION "Error Condition";
> printk(KERN_LEVEL, ERROR_CONDITION);
The problem is almost that what you need to be able to do is tell gcc
that printk's argument lives in a different segment via an indirect pointer.
I'm not sure if there is enough gcc magic to do that