Re: [idea] request_module(const char *fmt, ...);

[David Woodhouse]

[e-mail] said:
>   Q. What is the right design decision for Linux - to have a CONFIG_KMOD
>      dependency in the header that declares request_module() or to have
>      an extra function call in a non-performance critical code path that
>      invokes request_module() but avoid the dependency?

Personally, I think we should avoid the dependency, and put up with the tiny
amount of extra time it takes in non-critical code paths - if it's trying a
request_module but kmod isn't present, then it's almost invariably going to
fail to do what it was intending to do anyway - so who cares about a few extra 
cycles?

But The Master will speak, or more likely he'll just do what he thinks is best 
and we'll work it out from that :)

--
dwmw2