Subject: Re: howto debug kernel
To: None <jonathan@DSG.Stanford.EDU>
From: Mike Long <mikel@shore.net>
List: tech-kern
Date: 05/12/1997 18:58:18
>Date: Mon, 12 May 1997 14:34:31 -0700
>From: Jonathan Stone <jonathan@DSG.Stanford.EDU>

>Michael Graff <explorer@flame.org> writes:
>
>>>Jason Thorpe <thorpej@nas.nasa.gov> writes:
>>
>>> ...not really, but it works like:
>
>>How about I commit this?
>
>Minor nits:
> * Should also document addlog() :)
>
> * Could perhaps document kernel printf()  (see above!)
>
>Less minor nits:
>
> * All of log(), addlog(),  printf(), tprintf() etc. are really just
>   wrappers to kprintf(), but specifying different output  (to /dev/klog,
>   to console, to both; to ctty of (session of) current process; etc.

The base manpage should be kprintf(9), and MLINKS should be used to
generate links for all of its variants.

>   There should really be a manpage documenting the printf() control
>   strings used by all the kernel printf() functions, including  %b and %:.
>   (I couldn't find them anywhere in the  current manpages.)

Also formats that aren't supported, like %X.  I've been bitten by that
one a number of times.

Isn't %b supposed to be deprecated?  If so, you should note that.

>A name change for the kernel printf() function might make the
>documentation cleaner., too.  We already have printf(3) and printf(1),
>do we really want printf(9) too?

I think that changing the name of a function that is used by almost
every file in the kernel tree, for the sake of more easily
distinguishable manpage names, is an extremely bad idea.

-- 
Mike Long <mikel@shore.net>                http://www.shore.net/~mikel
"Every normal man must be tempted at times to spit on his hands,
hoist the black flag, and begin slitting throats." -- H.L. Mencken