Re: toolchain/38280: config(1) should emit devsw.h for devsw declarations in devsw.c

[Izumi Tsutsui <tsutsui%ceres.dti.ne.jp@localhost>]

The following reply was made to PR toolchain/38280; it has been noted by GNATS.

From: Izumi Tsutsui <tsutsui%ceres.dti.ne.jp@localhost>
To: cube%cubidou.net@localhost
Cc: gnats-bugs%NetBSD.org@localhost, tsutsui%ceres.dti.ne.jp@localhost
Subject: Re: toolchain/38280: config(1) should emit devsw.h for devsw 
declarations
         in devsw.c
Date: Sun, 23 Mar 2008 07:14:55 +0900

 > Wouldn't it be better to fix drivers to use devsw_name2chr() and
 > devsw_name2blk()?  Directly referencing something like that gives me
 > an ewww feeling.
 
 Main usage of it is
        makedev(cdevsw_lookup_major(&foo_cdevsw));
 in console cnattach() code.
 
 Should we also use devsw_name2chr() there?
 
 BTW, is mutex(9) (which is used in devsw_name2chr())
 valid at the point where consinit() is called first?
 ---
 Izumi Tsutsui