[pnfs] [PATCH 1/5] pnfs: Several non-critical fixups.

[Benny Halevy]

On May. 06, 2008, 16:45 -0700, Dean Hildebrand <seattleplus at gmail.com> wrote:
<snip>

> @@ -499,10 +499,11 @@ decode_and_add_ds(uint32_t **pp, struct filelayout_mount_type *mt)
>  	 */
>  	if (!ds->ds_clp) {
>  		err = nfs4_pnfs_ds_create(mds_srv, ds);
> -		printk(KERN_ERR
> -		       "%s nfs4_pnfs_ds_create returned %d\n", __func__, err);
> -		if (err)
> +		if (err) {
> +			printk(KERN_ERR "%s nfs4_pnfs_ds_create error %d\n",
> +			       __func__, err);
>  			goto out_err;
> +		}
>  	}

Why is this non-critical?
This looks like a real bug fix to me.... 


> @@ -1184,7 +1184,7 @@ pnfs_update_layout_commit(struct inode *inode,
>  	struct nfs_page *nfs_page = nfs_list_entry(head->next);
>  	int status;
>  
> -	dprintk("--> %s inode %p layout range: %Zd@%Lu\n", __func__, inode,
> +	dprintk("--> %s inode %p layout range: %Zd@%Zu\n", __func__, inode,
>  				(size_t)(npages * PAGE_SIZE),
>  				(loff_t)idx_start * PAGE_SIZE);
>  


Why is that?
Do you get a warning in your environment?
How about this instead?

 	dprintk("--> %s inode %p layout range: %Zd@%Lu\n", __func__, inode,
 				(size_t)(npages * PAGE_SIZE),
-				(loff_t)idx_start * PAGE_SIZE);
+				(u64)idx_start * PAGE_SIZE);