Re: sysctl(3) [Was: CVS commit: syssrc/sys/arch]

[woods%weird.com@localhost (Greg A. Woods)]

[ On Sunday, August 26, 2001 at 10:06:52 (-0700), Jason R Thorpe wrote: ]
> Subject: Re: sysctl(3) [Was: CVS commit: syssrc/sys/arch]
>
> On Mon, Aug 27, 2001 at 02:41:19AM +1000, Simon Burge wrote:
> 
>  > > Log Message:
>  > > 
>  > > sysctl(2) -> sysctl(3) (in comments)
>  > 
>  > Offhandedly, I've always been curious as to why there's a sysctl(3)
>  > when sysctl() is a system call...
> 
> sysctl(3) is a wrapper around __sysctl(2).

which is a pretty lame excuse if you ask me!  ;-)

The leading double-underscore pretty much means it's "hands-off" from
any application API, and:

        $ man __sysctl
        man: no entry for __sysctl in the manual.
 
What's even more lame is that the source for __sysctl() is generated on
the fly:

        printf '#include "SYS.h"\nRSYSCALL(__sysctl)\n' | cpp | as

YUCK!  :-)

Of course that's far from the worst of weirdness in syscall land....

-- 
                                                        Greg A. Woods

+1 416 218-0098      VE3TCP      <gwoods%acm.org@localhost>     
<woods%robohack.ca@localhost>
Planix, Inc. <woods%planix.com@localhost>;   Secrets of the Weird 
<woods%weird.com@localhost>