Looks good.
Reviewed-by: Christoph Hellwig <hch@xxxxxx>
Two little nitpicks:
On Wed, May 27, 2009 at 08:30:29AM -0400, Jeff Layton wrote:
> access_flags_to_mode(ppace[i]->access_req,
> ppace[i]->type,
> - &(inode->i_mode),
> + &(fattr->cf_mode),
> &user_mask);
> if (compare_sids(&(ppace[i]->sid), pgrpsid))
> access_flags_to_mode(ppace[i]->access_req,
> ppace[i]->type,
> - &(inode->i_mode),
> + &(fattr->cf_mode),
> &group_mask);
> if (compare_sids(&(ppace[i]->sid), &sid_everyone))
> access_flags_to_mode(ppace[i]->access_req,
> ppace[i]->type,
> - &(inode->i_mode),
> + &(fattr->cf_mode),
> &other_mask);
If you touch these lines please also remove the superflous braces.
> + cifs_i->delete_pending = fattr->cf_flags & CIFS_FATTR_DELETE_PENDING;
> +
> + /*
> + * Can't safely change the file size here if the client is writing to
> + * it due to potential races.
> + */
> spin_lock(&inode->i_lock);
> if (is_size_safe_to_change(cifs_i, fattr->cf_eof)) {
> - /*
> - * We can not safely change the file size here if the client
> - * is writing to it due to potential races.
> - */
Why isn't this comment introduced in the correct location in the
patch adding it?
--
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
