crypto: caam - Add cache coherency support

Freescale i.MX6 ARM platforms do not support hardware cache coherency.
This patch adds cache coherency support to the CAAM driver.

Signed-off-by: Victoria Milhoan <vicki.milhoan@freescale.com>
Tested-by: Horia Geantă <horia.geanta@freescale.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>

3 files changed, 23 insertions, 1 deletions

diff --git a/drivers/crypto/caam/caamhash.c b/drivers/crypto/caam/caamhash.c
index dae1e8099969..9622a81f4336 100644
--- a/drivers/crypto/caam/caamhash.c
+++ b/drivers/crypto/caam/caamhash.c
@@ -127,7 +127,7 @@ struct caam_hash_state {
 	int buflen_0;
 	u8 buf_1[CAAM_MAX_HASH_BLOCK_SIZE] ____cacheline_aligned;
 	int buflen_1;
-	u8 caam_ctx[MAX_CTX_LEN];
+	u8 caam_ctx[MAX_CTX_LEN] ____cacheline_aligned;
 	int (*update)(struct ahash_request *req);
 	int (*final)(struct ahash_request *req);
 	int (*finup)(struct ahash_request *req);
diff --git a/drivers/crypto/caam/caamrng.c b/drivers/crypto/caam/caamrng.c
index 5095337205b8..a1d21d5fb2ff 100644
--- a/drivers/crypto/caam/caamrng.c
+++ b/drivers/crypto/caam/caamrng.c
@@ -108,6 +108,10 @@ static void rng_done(struct device *jrdev, u32 *desc, u32 err, void *context)
 
 	atomic_set(&bd->empty, BUF_NOT_EMPTY);
 	complete(&bd->filled);
+
+	/* Buffer refilled, invalidate cache */
+	dma_sync_single_for_cpu(jrdev, bd->addr, RN_BUF_SIZE, DMA_FROM_DEVICE);
+
 #ifdef DEBUG
 	print_hex_dump(KERN_ERR, "rng refreshed buf@: ",
 		       DUMP_PREFIX_ADDRESS, 16, 4, bd->buf, RN_BUF_SIZE, 1);
diff --git a/drivers/crypto/caam/jr.c b/drivers/crypto/caam/jr.c
index b8b5d47acd7a..b7ec1ad38841 100644
--- a/drivers/crypto/caam/jr.c
+++ b/drivers/crypto/caam/jr.c
@@ -202,6 +202,13 @@ static void caam_jr_dequeue(unsigned long devarg)
 		userdesc = jrp->entinfo[sw_idx].desc_addr_virt;
 		userstatus = jrp->outring[hw_idx].jrstatus;
 
+		/*
+		 * Make sure all information from the job has been obtained
+		 * before telling CAAM that the job has been removed from the
+		 * output ring.
+		 */
+		mb();
+
 		/* set done */
 		wr_reg32(&jrp->rregs->outring_rmvd, 1);
 
@@ -351,12 +358,23 @@ int caam_jr_enqueue(struct device *dev, u32 *desc,
 
 	jrp->inpring[jrp->inp_ring_write_index] = desc_dma;
 
+	/*
+	 * Guarantee that the descriptor's DMA address has been written to
+	 * the next slot in the ring before the write index is updated, since
+	 * other cores may update this index independently.
+	 */
 	smp_wmb();
 
 	jrp->inp_ring_write_index = (jrp->inp_ring_write_index + 1) &
 				    (JOBR_DEPTH - 1);
 	jrp->head = (head + 1) & (JOBR_DEPTH - 1);
 
+	/*
+	 * Ensure that all job information has been written before
+	 * notifying CAAM that a new job was added to the input ring.
+	 */
+	wmb();
+
 	wr_reg32(&jrp->rregs->inpring_jobadd, 1);
 
 	spin_unlock_bh(&jrp->inplock);