Re: kern/48808

["Thomas Schmitt" <scdbackup%gmx.net@localhost>]

The following reply was made to PR kern/48808; it has been noted by GNATS.

From: "Thomas Schmitt" <scdbackup%gmx.net@localhost>
To: gnats-bugs%NetBSD.org@localhost
Cc: 
Subject: Re: kern/48808
Date: Wed, 14 May 2014 11:26:40 +0200

 Martin Husemann:
 > Forward binary compat like this is not needed - a new mount_cd9660 binary
 > can expect a new kernel always.
 
 mount_cd9660(8) is not the exclusive caller of mount(2) with
 struct iso_args. So i felt obliged to provide a detection method
 in mount_cd9660.h (struct old_iso_args_v_1_6).
 
 I'm a userlander and not far from calling mount(2) in own software.
 It would be decisive for me to know whether my args.ssector was
 taken into respect.
 
 Next i (in my new role as kernel hacker) had to test whether this
 distinction works. This was most easy to do in mount_cd9660(8).
 
 We can tear down all ifdef ISOFSMNT_SSECTOR in mount_cd9660(8)
 if they are deemed too ugly.
 But they are of educational use for anybody who wants to learn
 about this revolution after 20 years of stability.
 
 Errata:
    * The old struct is exposed here to serve as size benchmark for the reply
  - * of mount(2) with ISOFSMNT_SSECTOR. If after successful mount(2), the 
inquiry
  + * of mount(2) with MNT_GETARGS. If after successful mount(2), the inquiry
    * by MNT_GETARGS yields sizeof(struct old_iso_args_v_1_6) then the