Re: lib/39215: getdevmajor()/major() types don't match

[David Holland <dholland%eecs.harvard.edu@localhost>]

The following reply was made to PR lib/39215; it has been noted by GNATS.

From: David Holland <dholland%eecs.harvard.edu@localhost>
To: Matthias Drochner <M.Drochner%fz-juelich.de@localhost>
Cc: Christos Zoulas <christos%zoulas.com@localhost>, 
gnats-bugs%NetBSD.org@localhost,
        lib-bug-people%NetBSD.org@localhost, gnats-admin%NetBSD.org@localhost,
        netbsd-bugs%NetBSD.org@localhost, dholland%eecs.harvard.edu@localhost
Subject: Re: lib/39215: getdevmajor()/major() types don't match
Date: Mon, 4 Aug 2008 17:14:45 -0400

 On Mon, Aug 04, 2008 at 11:05:20PM +0200, Matthias Drochner wrote:
  > > That's all not too serious, but it is more work than one would
  > > initially expect. I'll try to come up with patches. 
  > 
  > OK, here is it. I've just set devmajor/minor_t to "long"
  > to get compiler warnings. It should stay where it was,
  > at int32_t, for now. (Actually, I don't see a good reason to
  > use a fixed-size type here, it is all inside the API afaics.)
 
 Well, in general they should be the same size as dev_t, and dev_t
 needs to be a fixed size because it's a bunch of bits pasted together.
 
 They should probably always be printed as signed long long, though, so
 it can in the future be 64-bit on 32-bit machines, so most of your
 casts want adjusting. :-/
 
 -- 
    - David A. Holland / dholland%eecs.harvard.edu@localhost