Properly zero out high cluster data on non-FAT32 msdos filesystems

[siva]

Summary

This bugfix comes from the following commits from NetBSD:
http://cvsweb.netbsd.org/bsdweb.cgi/src/sys/msdosfs/Attic/denode.h.diff?r1=1.23&r2=1.24&only_with_tag=MAIN&f=h
http://cvsweb.netbsd.org/bsdweb.cgi/src/sys/fs/msdosfs/denode.h.diff?r1=1.9&r2=1.10&only_with_tag=MAIN&f=h

[emaste]

sys/fs/msdosfs/denode.h

214

I wonder why this wasn't putushort((dp)->deHighClust, FAT32((dep)->de_pmp) ? (dep)->de_StartCluster >> 16) : 0)

(Or for that matter, can de_StartCluster be > 0xffff in FAT12 / FAT16?)

[siva]

sys/fs/msdosfs/denode.h

214

Not sure why the more succinct version wasn't used, it has the same clarity as the explicitly named macros.
Also, based on http://www.beginningtoseethelight.org/fat16/index.htm, the "data location" section in the data entry diagram states that the maximum value on FAT16 can only be 0xFFFF, so there may not be a point to this. I guess the only way that this matters is if a value greater than 0xFFFF was accidentally written to that field, in which case there are more important bugs to consider.

[siva]

Abandoning because this change doesn't functionally change anything, it isn't actually fixing any bugs.