Subject: Re: CVS commit: src/sys/sys
To: None <perry@piermont.com>
From: YAMAMOTO Takashi <yamt@mwd.biglobe.ne.jp>
List: source-changes
Date: 12/28/2005 12:24:49
> >> Modified Files:
> >> 	src/sys/sys: endian.h
> >> 
> >> Log Message:
> >> inline -> __inline for selected things (specifically, the header
> >> file is POSIX visible and the usage isn't #ifdef _KERNEL)
> >> 
> >> Discussed with core. I may need to go farther than this but for now
> >> compilations done with gcc -ansi (i.e. firefox) will be okay.
> >
> > can you explain how posix is related here?
> 
> Things like various dev/ headers aren't a problem per se since they
> only use "inline" in #ifdef _KERNEL instances. If you're writing
> highly NetBSD specific userland code that grovels stuff deep enough
> that you need to #define _KERNEL in a userland program, you can be
> expected to make tiny sacrifices like not compiling with -ansi. :)
> 
> I'm unaware of any instances of "inline" that now exist that aren't
> protected by #ifdef _KERNEL or otherwise "kernel only" header files --
> if there are any, they are mistakes (either in that the code should be
> #ifdef _KERNEL or because I missed fixing them back.)
> 
> Perry

ok, then, posix isn't related at all, right?

YAMAMOTO Takashi