Thorsten Kukuk wrote:
> >
> > This statement is wrong... unfortunately, this is not a realistic
> > statement for kernel-using daemons.
>
> No, it is right, even for kernel-using daemons. You have to copy
> the relevant data you need from the kernel header file to a local
> one. I think this is the only point where Linus and Uli Drepper
> agree about.
> Only if you have a kernel module you must use kernel header files
> and no glibc header files.
>
> Thorsten
>
Linus and Ulrich are the main benefactors for jumping through hoops on
this one. Everyone who have to deal with these things on a regular basis
-- mount, autofs, etc -- know it's utterly unrealistic.
(Oh, and it's particularly pitiful since glibc still #includes kernel
headers. I had compiling gcc break for me because of such a hidden
include recently.)
-hpa
--
<[EMAIL PROTECTED]> at work, <[EMAIL PROTECTED]> in private!
"Unix gives you enough rope to shoot yourself in the foot."
http://www.zytor.com/~hpa/puzzle.txt
