[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: VFAT kernel considerations



Hi!

On Wed, May 20, 1998 at 04:03:54PM -0400, Martin-Eric Racine wrote:
> Exactly.  Write a file using VFAT, delete it under SingleTOS.
> You end up with a longname reference to a file that is gone.

And what's the problem? The long filenames in VFAT always have start
cluster 0, like empty files on FAT filesystems. The only problem with
that is you lose one directory entry, which can be a serious problem for
root directories, as they have fixed length. But VFAT causes directory
fragmentation even in regular usage, so I don't think that's too much of
a problem.

Also, I still don't see why enabling VFAT on the boot partition should
break old software. Remember, VFAT is case /preserving/, not case
sensetive, i.e. a file named "Test" can also be accessed as "test",
"TEST", and so on. Of course, lower case filenames get a VFAT entry,
too, but as long as the lower case name fits into standard FAT naming
rules (i.e. not more than 8+3, no additional dots and no illegal
characters like space), the corresponding short name will be exactly the
same, but of course in upper case, so it will be there without VFAT,
too.

But there's one reason to keep good old TOSFS of MiNT: Emulators like
STEmulator, STonX and so on need it to be able to access the host's
local filesystem, as this of course can't be done via Rwabs(), assuming
a standard FAT filesystem.


Ciao

Thomas


-- 
Thomas Binder (Gryf @ IRCNet)  gryf@hrzpub.tu-darmstadt.de
PGP-key available on request!  binder@rbg.informatik.tu-darmstadt.de

Attachment: pgpC16jRv4XUB.pgp
Description: PGP signature