Re: to CONFIG_KMOD or not to CONFIG_KMOD
Keith Owens (kaos@ocs.com.au)
Thu, 30 Dec 1999 22:55:00 +1100
On Thu, 30 Dec 1999 11:27:14 +0000 (GMT),
Tigran Aivazian <tigran@sco.COM> wrote:
>What is the official status on the situations like:
>
>check some resource;
>if (resource == NULL) {
> char buf[32];
> sprintf(buffer, "foo-bar-XYZ");
> request_module(buffer);
> if (resource is still NULL)
> return -ENODEV; /* or whatever appropriate */
>}
>
>The #ifdef CONFIG_KMODE around this type of code can be omitted (because
>request_module() will be just noop) e.g. as is done in phone_open() (the
>new Linux Telephony thingy) or it can be left for performance when any CPU
>cycle is important e.g. as in fs/exec.c:search_binary_handler()
>
>Does everyone agree with such state of things or do you think one should
>go through the whole source and clean it up removing all #ifdef
>CONFIG_KMOD as they are strictly speaking superfluous?
I would like to see all the #ifdef removed. With the current code,
switching CONFIG_KMOD on or off recompiles far too much of the kernel.
-
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/