>>>>> " " == Robert Love <rml@tech9.net> writes:
> CLONE_CRED. I suspect 90% of the cases retain the same
> credentials anyhow. Copy-on-write? :)
Ben,
Making the credentials a monolithic block like you appear to be
doing just doesn't make sense. If you look at the way things like
fsuid/fsgid/groups[] are used, you will see that almost all those that
filesystems that care are making their own private copies.
It would be a lot more useful to split out fsuid/fsgid/groups[] as
per the *BSD ucred, and then allow filesystems to reference the
resulting struct instead (using COW semantics).
That way too, 'struct file' could finally contain a reference to a
full copy of the filesystem credentials, and we could get rid of
the 'struct file' crud in address_space_operations like readpage().
See
http://www.fys.uio.no/~trondmy/src/bsdcred/linux-2.5.1-pre11_cred.dif
for my earlier attempt at doing this sort of thing...
Cheers,
Trond
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
This archive was generated by hypermail 2b29 : Sun Jun 23 2002 - 22:00:24 EST