Messages in this thread | | | Date | Sun, 27 Dec 1998 06:46:29 +0000 (GMT) | From | Riley Williams <> | Subject | Re: /dev/one - why not /dev/repeat? |
| |
Hi Tim.
>>> What's wrong with the following C code being run as a usermode >>> program?
>> Uhm, you need a compiler to make it? Some of us like having vast >> amounts of power available from the shell, for minimal/thin >> client/embedded applications. This means small kernel additions >> that are vastly generic. (For vast flexabilty)
> Uhm, why can't the same compiler you'd use to build those kernel > additions be used to build Riley's program?
You've missed the important difference - HE didn't write any kernel addition as part of his script, so by getting them added to the kernel, he can pretend that his script is just that, entirely script, rather than being dependant on some kernel bloat that's written in C code...
Personally, I feel that the only non-kernel-specific scripts worth writing are those that work unchanged on Linux, FreeBSD, SunSolaris and other UNIX systems, and even if Linus was stupid enough to accept his ideas, it's unlikely he'd persuade enough of the other Unices to do so to make the ideas acceptable in the first place...
By "non-kernel-specific scripts", I mean scripts that are not directly part of the configuration and/or installation of a specific kernel, in the way that "Configure" and "Menuconfig" are. I specifically do NOT exclude "administrative scripts", such as the secure delete function that appears to be his raison d'etre for adding the proposed kernel bloat to satisfy his whims, since that's a facility that, if implemented, should work on ANY of the various flavours of UNIX...
Best wishes from Riley.
- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.rutgers.edu Please read the FAQ at http://www.tux.org/lkml/
| |