This was my (quite) bone-headed mistake. Applied.
Charles
On Wed, 2005-11-16 at 15:29 +0100, Martin Walter wrote:
> Hi,
>
> the strcmp in inode.c results in 0 if "nfs". So IS_NFS fails.
>
> So would you please patch:
>
> #########################################################################
> --- unionfs-20051115-1700/inode.c.orig 2005-11-15 23:00:18.000000000 +0100
> +++ unionfs-20051115-1700/inode.c 2005-11-16 15:06:44.674402087 +0100
> @@ -855,7 +855,7 @@
> if (inode->i_op && inode->i_op->permission) {
> retval = inode->i_op->permission(inode, submask, nd);
> #ifdef NFS_SECURITY_HOLE
> -#define IS_NFS(inode) (strcmp("nfs", (inode)->i_sb->s_type->name))
> +#define IS_NFS(inode) (strcmp("nfs", (inode)->i_sb->s_type->name) == 0)
> if ((retval == -EACCES) && (submask & MAY_WRITE) &&
> IS_NFS(inode)) {
> retval = generic_permission(inode, submask, NULL);
> #########################################################################
>
> Thanx,
> Martin
>
>
_______________________________________________
unionfs mailing list
unionfs@mail.fsl.cs.sunysb.edu
http://www.fsl.cs.sunysb.edu/mailman/listinfo/unionfs
