crypto: caam - consolidate memory barriers from job ring en/dequeue

Memory barriers are implied by the i/o register write implementation
(at least on Power). So we can remove the redundant wmb() in
caam_jr_enqueue, and, in dequeue(), hoist the h/w done notification
write up to before we need to increment the head of the ring, and
save an smp_mb.

Signed-off-by: Kim Phillips <[e-mail]>
Signed-off-by: Herbert Xu <[e-mail]>

drivers/crypto/caam/jr.c

···
 		userdesc = jrp->entinfo[sw_idx].desc_addr_virt;
 		userstatus = jrp->outring[hw_idx].jrstatus;
 
-		smp_mb();
+		/* set done */
+		wr_reg32(&jrp->rregs->outring_rmvd, 1);
 
 		jrp->out_ring_read_index = (jrp->out_ring_read_index + 1) &
 					   (JOBR_DEPTH - 1);
···
 
 			jrp->tail = tail;
 		}
-
-		/* set done */
-		wr_reg32(&jrp->rregs->outring_rmvd, 1);
 
 		spin_unlock_bh(&jrp->outlock);
 
···
 	jrp->inp_ring_write_index = (jrp->inp_ring_write_index + 1) &
 				    (JOBR_DEPTH - 1);
 	jrp->head = (head + 1) & (JOBR_DEPTH - 1);
-
-	wmb();
 
 	wr_reg32(&jrp->rregs->inpring_jobadd, 1);