Some quick thoughts:
> Permute the location of files. E.g. 'permute(A, B, C)' is equivalent to A->B,
> B->C and C->A. This is essentially a series of renames done as a single atomic
> operation.
Hmm. Can we choose a more specific name than 'permute'? To me,
->permute() tells me just as much about the operation as
->do_something(). {multi,bulk,mass}_rename()? renamev()?
Maybe it's just me.
> to be done as an atomic operation. We could add whiteout support to filesystem
> ops to perform the creation or removal of whiteouts atomically, but it would
> complicate many filesystem ops needlessly.
>
> Alternatively we can add a generic permute operation and add whiteout support to
> the VFS which utilizes this to perform the operations atomically.
I certainly like the sound of this.
> +static void sort_parents3(struct dentry **p)
> +void sort_parents(struct dentry **p, unsigned *nump)
Yikes, that's a bunch of fiddly code. Is it *really* worth all that to
avoid calling the generic sort helpers?
> + if (WARN_ON(num > PERMUTE_MAX) ||
> + WARN_ON(num < 2))
> + return -EINVAL;
And in other places this is a BUG? Why not, like the syscall, limit the
arguments to three if we're serious about that limitation?
- z
--
To unsubscribe from this list: send the line "unsubscribe linux-fsdevel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
