At 00:32 13/11/01, Alexander Viro wrote:
On Mon, 12 Nov 2001, Andreas Gruenbacher wrote:
> There is one difference between the interfaces you are complaining about
> above and the proposed EA interface for EA's: In those interfaces you have
> wildcard parameters that are used for who-knows-what, depending on a
> command-like parameter, including use as a value, use as a pointer to a
> value/struct, etc.
Yes, and? You've got more than enough material for the same kind of
abuse. What's more, you _already_ have it - in some of the subfunctions
*data is read from, in some - written to, in some - ignored. Worse
yet, in some subfunctions we put structured data in there, in some -
just a chunk of something.
With all that, who had said that a year down the road we won't get a
dozen of new syscalls hiding behind that one?
Sorry, folks, but idea of private extendable syscall table (per-filesystem,
no less) doesn't look like a good thing. That's _the_ reason why ioctl()
is bad.
Al,
Out of interest, which access interface(s) would you like to see used?
Giving a few suggestions you would be happy with would be a lot easier on
anyone trying to develop a filesystem API than for them having to come up
with one after the other until one is found which you approve of... (-;
Best regards,
Anton
--
"I've not lost my mind. It's backed up on tape somewhere." - Unknown
--
Anton Altaparmakov <aia21 at cam.ac.uk> (replace at with @)
Linux NTFS Maintainer / WWW: http://linux-ntfs.sf.net/
ICQ: 8561279 / WWW: http://www-stu.christs.cam.ac.uk/~aia21/
|