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: Sun, 01 Jun 2014 16:50:16 +0200

 Hi,
 
 after counselling by Martin Husemann: new patch proposals.
 
 The patch for distrib/sets/lists/comp/mi.orig is now retracted.
 
 The patch for sys/fs/cd9660 is slightley modified to reflect
 my grown deference towards the semi-public API/ABI of cd9660.
 cd9660_extern.h will not have (uint32_t) as replacement of
 inappropriate (int), but rather (unsigned int).
 I expect 16 bit int to fail miserably anyway, so it is only
 about not breaking the ABI for machines with 64 bit (int).
 The change of signedness can hardly affect existing ISO 9660
 filesystems, because it shows up only above 4 TiB size.
 
 The patch for sbin/mount_cd9660 was shrunk by the assumption
 that new /sbin/mount_cd9660 will always be compiled in a *.h
 environment and run on a kernel which implement the new session
 offset feature from the sys/fs/cd9660 patch.
 I considered to remove only the after-mount test for success of
 the new option. But actually it makes few sense to allow an
 old kernel header environment at compile time, but to assume a
 new kernel at run time.
 So there are two alternative patches for choice:
 - The older one, ready for old and new kernels.
     Date: Sun, 18 May 2014 16:39:13 +0200
 - The new one, ready only for new kernels.
 
 The new patches are tested by the ISO filesystem which also
 serves for the upcomming large file patch. All seems well,
 except of course 2 failed large file tests.
 
 
 Have a nice day :)
 
 Thomas