This is the mail archive of the
gcc@gcc.gnu.org
mailing list for the GCC project.
Re: kernel-2.2.1-undefined references.
- To: hpa at transmeta dot com (H. Peter Anvin)
- Subject: Re: kernel-2.2.1-undefined references.
- From: Alan Cox <alan at cymru dot net>
- Date: Wed, 17 Feb 1999 16:37:18 +0000 (GMT)
- Cc: egcs at egcs dot cygnus dot com, torvalds at transmeta dot com, hpa at transmeta dot com, alan dot cox at linux dot org
> > Why is that a weakness? If the compiler's optimizer says "this code runs
> > faster when not inlined" (I can think of a couple of reasons why it would
> > even be right doing that...), I'm not qualified to contradict it.
>
> No, you have it the wrong way around. The human should be assumed
> smarter than the compiler: if you say that this should be inlined, the
> compiler better not contradict you.
Ive been away so this is a bit of a late final reply
I think both sides are right.
o The compiler should by default do what it feels is right
That includes inlining and uninlining code
o The compiler has to have a way you can tell it "or else".
extern __inline_or_i_beat_you_to_death__
or similar