Re: PATCH for FAT in 2.3.28
Martin Dalecki (dalecki@cs.net.pl)
Sat, 13 Nov 1999 20:27:26 +0100
Alexander Viro wrote:
>
> On Sat, 13 Nov 1999, Martin Dalecki wrote:
>
> > The following is fixing the compilation of the fat FS as a runtime
> > loadable
> > module. I have tested the patch on my ZIP drive. Apparently there is
> > only
> > one one other filesystem out there which is using the same partial page
> > write interface, which appears as a leftover from older VFS versions.
>
> Wrong.
>
> > (At least for fat I can't see what the special function was doing
> > different
> > then the generic one. At least it didn't touch any special FAT fields in
> > the inode.)
>
> Watch carefully. FAT (_and_ HPFS) doesn't allow files with holes. First of
> all it _does_ touch FAT-specific fields (albeit i_size) would probably be
> OK with some tweaking. Yes, it might make sense to move it into buffer.c.
> But it's _not_ an equivalent of the generic one. Heck, have you ever tried
> the result of your changes?
Yes I have. Apparently it's working for the simple cases ;-).
Thank's for the advice on the way this function is supposed to work.
Anyway: alloc_pages() is missing from the list of exported symbols and
causes fat.o to not work as a module.
--Marcin
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/