]> git.karo-electronics.de Git - mv-sheeva.git/blobdiff - drivers/crypto/omap-sham.c
Merge tag 'v2.6.38' of git://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux-2.6
[mv-sheeva.git] / drivers / crypto / omap-sham.c
index a081c7c7d03f883a0ed2df188269bb4d8a1e29b9..2e71123516e09f2d02667ddc11024196c8275da3 100644 (file)
 
 #define DEFAULT_TIMEOUT_INTERVAL       HZ
 
-#define FLAGS_FIRST            0x0001
 #define FLAGS_FINUP            0x0002
 #define FLAGS_FINAL            0x0004
-#define FLAGS_FAST             0x0008
+#define FLAGS_SG               0x0008
 #define FLAGS_SHA1             0x0010
 #define FLAGS_DMA_ACTIVE       0x0020
 #define FLAGS_OUTPUT_READY     0x0040
 #define FLAGS_INIT             0x0100
 #define FLAGS_CPU              0x0200
 #define FLAGS_HMAC             0x0400
-
-/* 3rd byte */
-#define FLAGS_BUSY             16
+#define FLAGS_ERROR            0x0800
+#define FLAGS_BUSY             0x1000
 
 #define OP_UPDATE      1
 #define OP_FINAL       2
 
+#define OMAP_ALIGN_MASK                (sizeof(u32)-1)
+#define OMAP_ALIGNED           __attribute__((aligned(sizeof(u32))))
+
+#define BUFLEN         PAGE_SIZE
+
 struct omap_sham_dev;
 
 struct omap_sham_reqctx {
@@ -97,8 +100,8 @@ struct omap_sham_reqctx {
        unsigned long           flags;
        unsigned long           op;
 
+       u8                      digest[SHA1_DIGEST_SIZE] OMAP_ALIGNED;
        size_t                  digcnt;
-       u8                      *buffer;
        size_t                  bufcnt;
        size_t                  buflen;
        dma_addr_t              dma_addr;
@@ -107,6 +110,8 @@ struct omap_sham_reqctx {
        struct scatterlist      *sg;
        unsigned int            offset; /* offset in current sg */
        unsigned int            total;  /* total request */
+
+       u8                      buffer[0] OMAP_ALIGNED;
 };
 
 struct omap_sham_hmac_ctx {
@@ -136,6 +141,7 @@ struct omap_sham_dev {
        int                     irq;
        struct clk              *iclk;
        spinlock_t              lock;
+       int                     err;
        int                     dma;
        int                     dma_lch;
        struct tasklet_struct   done_task;
@@ -194,53 +200,68 @@ static inline int omap_sham_wait(struct omap_sham_dev *dd, u32 offset, u32 bit)
 static void omap_sham_copy_hash(struct ahash_request *req, int out)
 {
        struct omap_sham_reqctx *ctx = ahash_request_ctx(req);
+       u32 *hash = (u32 *)ctx->digest;
+       int i;
+
+       /* MD5 is almost unused. So copy sha1 size to reduce code */
+       for (i = 0; i < SHA1_DIGEST_SIZE / sizeof(u32); i++) {
+               if (out)
+                       hash[i] = omap_sham_read(ctx->dd,
+                                               SHA_REG_DIGEST(i));
+               else
+                       omap_sham_write(ctx->dd,
+                                       SHA_REG_DIGEST(i), hash[i]);
+       }
+}
+
+static void omap_sham_copy_ready_hash(struct ahash_request *req)
+{
+       struct omap_sham_reqctx *ctx = ahash_request_ctx(req);
+       u32 *in = (u32 *)ctx->digest;
        u32 *hash = (u32 *)req->result;
        int i;
 
+       if (!hash)
+               return;
+
        if (likely(ctx->flags & FLAGS_SHA1)) {
                /* SHA1 results are in big endian */
                for (i = 0; i < SHA1_DIGEST_SIZE / sizeof(u32); i++)
-                       if (out)
-                               hash[i] = be32_to_cpu(omap_sham_read(ctx->dd,
-                                                       SHA_REG_DIGEST(i)));
-                       else
-                               omap_sham_write(ctx->dd, SHA_REG_DIGEST(i),
-                                                       cpu_to_be32(hash[i]));
+                       hash[i] = be32_to_cpu(in[i]);
        } else {
                /* MD5 results are in little endian */
                for (i = 0; i < MD5_DIGEST_SIZE / sizeof(u32); i++)
-                       if (out)
-                               hash[i] = le32_to_cpu(omap_sham_read(ctx->dd,
-                                                       SHA_REG_DIGEST(i)));
-                       else
-                               omap_sham_write(ctx->dd, SHA_REG_DIGEST(i),
-                                                       cpu_to_le32(hash[i]));
+                       hash[i] = le32_to_cpu(in[i]);
        }
 }
 
-static int omap_sham_write_ctrl(struct omap_sham_dev *dd, size_t length,
-                                int final, int dma)
+static int omap_sham_hw_init(struct omap_sham_dev *dd)
 {
-       struct omap_sham_reqctx *ctx = ahash_request_ctx(dd->req);
-       u32 val = length << 5, mask;
+       clk_enable(dd->iclk);
 
-       if (unlikely(!ctx->digcnt)) {
+       if (!(dd->flags & FLAGS_INIT)) {
+               omap_sham_write_mask(dd, SHA_REG_MASK,
+                       SHA_REG_MASK_SOFTRESET, SHA_REG_MASK_SOFTRESET);
 
-               clk_enable(dd->iclk);
+               if (omap_sham_wait(dd, SHA_REG_SYSSTATUS,
+                                       SHA_REG_SYSSTATUS_RESETDONE))
+                       return -ETIMEDOUT;
 
-               if (!(dd->flags & FLAGS_INIT)) {
-                       omap_sham_write_mask(dd, SHA_REG_MASK,
-                               SHA_REG_MASK_SOFTRESET, SHA_REG_MASK_SOFTRESET);
+               dd->flags |= FLAGS_INIT;
+               dd->err = 0;
+       }
 
-                       if (omap_sham_wait(dd, SHA_REG_SYSSTATUS,
-                                               SHA_REG_SYSSTATUS_RESETDONE))
-                               return -ETIMEDOUT;
+       return 0;
+}
 
-                       dd->flags |= FLAGS_INIT;
-               }
-       } else {
+static void omap_sham_write_ctrl(struct omap_sham_dev *dd, size_t length,
+                                int final, int dma)
+{
+       struct omap_sham_reqctx *ctx = ahash_request_ctx(dd->req);
+       u32 val = length << 5, mask;
+
+       if (likely(ctx->digcnt))
                omap_sham_write(dd, SHA_REG_DIGCNT, ctx->digcnt);
-       }
 
        omap_sham_write_mask(dd, SHA_REG_MASK,
                SHA_REG_MASK_IT_EN | (dma ? SHA_REG_MASK_DMA_EN : 0),
@@ -260,29 +281,26 @@ static int omap_sham_write_ctrl(struct omap_sham_dev *dd, size_t length,
                        SHA_REG_CTRL_ALGO | SHA_REG_CTRL_LENGTH;
 
        omap_sham_write_mask(dd, SHA_REG_CTRL, val, mask);
-
-       return 0;
 }
 
 static int omap_sham_xmit_cpu(struct omap_sham_dev *dd, const u8 *buf,
                              size_t length, int final)
 {
        struct omap_sham_reqctx *ctx = ahash_request_ctx(dd->req);
-       int err, count, len32;
+       int count, len32;
        const u32 *buffer = (const u32 *)buf;
 
        dev_dbg(dd->dev, "xmit_cpu: digcnt: %d, length: %d, final: %d\n",
                                                ctx->digcnt, length, final);
 
-       err = omap_sham_write_ctrl(dd, length, final, 0);
-       if (err)
-               return err;
+       omap_sham_write_ctrl(dd, length, final, 0);
+
+       /* should be non-zero before next lines to disable clocks later */
+       ctx->digcnt += length;
 
        if (omap_sham_wait(dd, SHA_REG_CTRL, SHA_REG_CTRL_INPUT_READY))
                return -ETIMEDOUT;
 
-       ctx->digcnt += length;
-
        if (final)
                ctx->flags |= FLAGS_FINAL; /* catch last interrupt */
 
@@ -298,16 +316,11 @@ static int omap_sham_xmit_dma(struct omap_sham_dev *dd, dma_addr_t dma_addr,
                              size_t length, int final)
 {
        struct omap_sham_reqctx *ctx = ahash_request_ctx(dd->req);
-       int err, len32;
+       int len32;
 
        dev_dbg(dd->dev, "xmit_dma: digcnt: %d, length: %d, final: %d\n",
                                                ctx->digcnt, length, final);
 
-       /* flush cache entries related to our page */
-       if (dma_addr == ctx->dma_addr)
-               dma_sync_single_for_device(dd->dev, dma_addr, length,
-                                          DMA_TO_DEVICE);
-
        len32 = DIV_ROUND_UP(length, sizeof(u32));
 
        omap_set_dma_transfer_params(dd->dma_lch, OMAP_DMA_DATA_TYPE_S32, len32,
@@ -317,9 +330,7 @@ static int omap_sham_xmit_dma(struct omap_sham_dev *dd, dma_addr_t dma_addr,
        omap_set_dma_src_params(dd->dma_lch, 0, OMAP_DMA_AMODE_POST_INC,
                                dma_addr, 0, 0);
 
-       err = omap_sham_write_ctrl(dd, length, final, 1);
-       if (err)
-               return err;
+       omap_sham_write_ctrl(dd, length, final, 1);
 
        ctx->digcnt += length;
 
@@ -371,15 +382,29 @@ static size_t omap_sham_append_sg(struct omap_sham_reqctx *ctx)
        return 0;
 }
 
+static int omap_sham_xmit_dma_map(struct omap_sham_dev *dd,
+                                       struct omap_sham_reqctx *ctx,
+                                       size_t length, int final)
+{
+       ctx->dma_addr = dma_map_single(dd->dev, ctx->buffer, ctx->buflen,
+                                      DMA_TO_DEVICE);
+       if (dma_mapping_error(dd->dev, ctx->dma_addr)) {
+               dev_err(dd->dev, "dma %u bytes error\n", ctx->buflen);
+               return -EINVAL;
+       }
+
+       ctx->flags &= ~FLAGS_SG;
+
+       /* next call does not fail... so no unmap in the case of error */
+       return omap_sham_xmit_dma(dd, ctx->dma_addr, length, final);
+}
+
 static int omap_sham_update_dma_slow(struct omap_sham_dev *dd)
 {
        struct omap_sham_reqctx *ctx = ahash_request_ctx(dd->req);
        unsigned int final;
        size_t count;
 
-       if (!ctx->total)
-               return 0;
-
        omap_sham_append_sg(ctx);
 
        final = (ctx->flags & FLAGS_FINUP) && !ctx->total;
@@ -390,30 +415,68 @@ static int omap_sham_update_dma_slow(struct omap_sham_dev *dd)
        if (final || (ctx->bufcnt == ctx->buflen && ctx->total)) {
                count = ctx->bufcnt;
                ctx->bufcnt = 0;
-               return omap_sham_xmit_dma(dd, ctx->dma_addr, count, final);
+               return omap_sham_xmit_dma_map(dd, ctx, count, final);
        }
 
        return 0;
 }
 
-static int omap_sham_update_dma_fast(struct omap_sham_dev *dd)
+/* Start address alignment */
+#define SG_AA(sg)      (IS_ALIGNED(sg->offset, sizeof(u32)))
+/* SHA1 block size alignment */
+#define SG_SA(sg)      (IS_ALIGNED(sg->length, SHA1_MD5_BLOCK_SIZE))
+
+static int omap_sham_update_dma_start(struct omap_sham_dev *dd)
 {
        struct omap_sham_reqctx *ctx = ahash_request_ctx(dd->req);
-       unsigned int length;
+       unsigned int length, final, tail;
+       struct scatterlist *sg;
 
-       ctx->flags |= FLAGS_FAST;
+       if (!ctx->total)
+               return 0;
+
+       if (ctx->bufcnt || ctx->offset)
+               return omap_sham_update_dma_slow(dd);
+
+       dev_dbg(dd->dev, "fast: digcnt: %d, bufcnt: %u, total: %u\n",
+                       ctx->digcnt, ctx->bufcnt, ctx->total);
+
+       sg = ctx->sg;
 
-       length = min(ctx->total, sg_dma_len(ctx->sg));
-       ctx->total = length;
+       if (!SG_AA(sg))
+               return omap_sham_update_dma_slow(dd);
+
+       if (!sg_is_last(sg) && !SG_SA(sg))
+               /* size is not SHA1_BLOCK_SIZE aligned */
+               return omap_sham_update_dma_slow(dd);
+
+       length = min(ctx->total, sg->length);
+
+       if (sg_is_last(sg)) {
+               if (!(ctx->flags & FLAGS_FINUP)) {
+                       /* not last sg must be SHA1_MD5_BLOCK_SIZE aligned */
+                       tail = length & (SHA1_MD5_BLOCK_SIZE - 1);
+                       /* without finup() we need one block to close hash */
+                       if (!tail)
+                               tail = SHA1_MD5_BLOCK_SIZE;
+                       length -= tail;
+               }
+       }
 
        if (!dma_map_sg(dd->dev, ctx->sg, 1, DMA_TO_DEVICE)) {
                dev_err(dd->dev, "dma_map_sg  error\n");
                return -EINVAL;
        }
 
+       ctx->flags |= FLAGS_SG;
+
        ctx->total -= length;
+       ctx->offset = length; /* offset where to start slow */
 
-       return omap_sham_xmit_dma(dd, sg_dma_address(ctx->sg), length, 1);
+       final = (ctx->flags & FLAGS_FINUP) && !ctx->total;
+
+       /* next call does not fail... so no unmap in the case of error */
+       return omap_sham_xmit_dma(dd, sg_dma_address(ctx->sg), length, final);
 }
 
 static int omap_sham_update_cpu(struct omap_sham_dev *dd)
@@ -433,8 +496,17 @@ static int omap_sham_update_dma_stop(struct omap_sham_dev *dd)
        struct omap_sham_reqctx *ctx = ahash_request_ctx(dd->req);
 
        omap_stop_dma(dd->dma_lch);
-       if (ctx->flags & FLAGS_FAST)
+       if (ctx->flags & FLAGS_SG) {
                dma_unmap_sg(dd->dev, ctx->sg, 1, DMA_TO_DEVICE);
+               if (ctx->sg->length == ctx->offset) {
+                       ctx->sg = sg_next(ctx->sg);
+                       if (ctx->sg)
+                               ctx->offset = 0;
+               }
+       } else {
+               dma_unmap_single(dd->dev, ctx->dma_addr, ctx->buflen,
+                                DMA_TO_DEVICE);
+       }
 
        return 0;
 }
@@ -454,14 +526,7 @@ static void omap_sham_cleanup(struct ahash_request *req)
        spin_unlock_irqrestore(&dd->lock, flags);
 
        if (ctx->digcnt)
-               clk_disable(dd->iclk);
-
-       if (ctx->dma_addr)
-               dma_unmap_single(dd->dev, ctx->dma_addr, ctx->buflen,
-                                DMA_TO_DEVICE);
-
-       if (ctx->buffer)
-               free_page((unsigned long)ctx->buffer);
+               omap_sham_copy_ready_hash(req);
 
        dev_dbg(dd->dev, "digcnt: %d, bufcnt: %d\n", ctx->digcnt, ctx->bufcnt);
 }
@@ -489,8 +554,6 @@ static int omap_sham_init(struct ahash_request *req)
 
        ctx->flags = 0;
 
-       ctx->flags |= FLAGS_FIRST;
-
        dev_dbg(dd->dev, "init: digest size: %d\n",
                crypto_ahash_digestsize(tfm));
 
@@ -499,21 +562,7 @@ static int omap_sham_init(struct ahash_request *req)
 
        ctx->bufcnt = 0;
        ctx->digcnt = 0;
-
-       ctx->buflen = PAGE_SIZE;
-       ctx->buffer = (void *)__get_free_page(
-                               (req->base.flags & CRYPTO_TFM_REQ_MAY_SLEEP) ?
-                               GFP_KERNEL : GFP_ATOMIC);
-       if (!ctx->buffer)
-               return -ENOMEM;
-
-       ctx->dma_addr = dma_map_single(dd->dev, ctx->buffer, ctx->buflen,
-                                       DMA_TO_DEVICE);
-       if (dma_mapping_error(dd->dev, ctx->dma_addr)) {
-               dev_err(dd->dev, "dma %u bytes error\n", ctx->buflen);
-               free_page((unsigned long)ctx->buffer);
-               return -EINVAL;
-       }
+       ctx->buflen = BUFLEN;
 
        if (tctx->flags & FLAGS_HMAC) {
                struct omap_sham_hmac_ctx *bctx = tctx->base;
@@ -538,10 +587,8 @@ static int omap_sham_update_req(struct omap_sham_dev *dd)
 
        if (ctx->flags & FLAGS_CPU)
                err = omap_sham_update_cpu(dd);
-       else if (ctx->flags & FLAGS_FAST)
-               err = omap_sham_update_dma_fast(dd);
        else
-               err = omap_sham_update_dma_slow(dd);
+               err = omap_sham_update_dma_start(dd);
 
        /* wait for dma completion before can take more data */
        dev_dbg(dd->dev, "update: err: %d, digcnt: %d\n", err, ctx->digcnt);
@@ -560,15 +607,12 @@ static int omap_sham_final_req(struct omap_sham_dev *dd)
                use_dma = 0;
 
        if (use_dma)
-               err = omap_sham_xmit_dma(dd, ctx->dma_addr, ctx->bufcnt, 1);
+               err = omap_sham_xmit_dma_map(dd, ctx, ctx->bufcnt, 1);
        else
                err = omap_sham_xmit_cpu(dd, ctx->buffer, ctx->bufcnt, 1);
 
        ctx->bufcnt = 0;
 
-       if (err != -EINPROGRESS)
-               omap_sham_cleanup(req);
-
        dev_dbg(dd->dev, "final_req: err: %d\n", err);
 
        return err;
@@ -576,6 +620,7 @@ static int omap_sham_final_req(struct omap_sham_dev *dd)
 
 static int omap_sham_finish_req_hmac(struct ahash_request *req)
 {
+       struct omap_sham_reqctx *ctx = ahash_request_ctx(req);
        struct omap_sham_ctx *tctx = crypto_tfm_ctx(req->base.tfm);
        struct omap_sham_hmac_ctx *bctx = tctx->base;
        int bs = crypto_shash_blocksize(bctx->shash);
@@ -590,48 +635,56 @@ static int omap_sham_finish_req_hmac(struct ahash_request *req)
 
        return crypto_shash_init(&desc.shash) ?:
               crypto_shash_update(&desc.shash, bctx->opad, bs) ?:
-              crypto_shash_finup(&desc.shash, req->result, ds, req->result);
+              crypto_shash_finup(&desc.shash, ctx->digest, ds, ctx->digest);
 }
 
 static void omap_sham_finish_req(struct ahash_request *req, int err)
 {
        struct omap_sham_reqctx *ctx = ahash_request_ctx(req);
+       struct omap_sham_dev *dd = ctx->dd;
 
        if (!err) {
                omap_sham_copy_hash(ctx->dd->req, 1);
                if (ctx->flags & FLAGS_HMAC)
                        err = omap_sham_finish_req_hmac(req);
+       } else {
+               ctx->flags |= FLAGS_ERROR;
        }
 
-       if (ctx->flags & FLAGS_FINAL)
+       if ((ctx->flags & FLAGS_FINAL) || err)
                omap_sham_cleanup(req);
 
-       clear_bit(FLAGS_BUSY, &ctx->dd->flags);
+       clk_disable(dd->iclk);
+       dd->flags &= ~FLAGS_BUSY;
 
        if (req->base.complete)
                req->base.complete(&req->base, err);
 }
 
-static int omap_sham_handle_queue(struct omap_sham_dev *dd)
+static int omap_sham_handle_queue(struct omap_sham_dev *dd,
+                                 struct ahash_request *req)
 {
        struct crypto_async_request *async_req, *backlog;
        struct omap_sham_reqctx *ctx;
-       struct ahash_request *req, *prev_req;
+       struct ahash_request *prev_req;
        unsigned long flags;
-       int err = 0;
-
-       if (test_and_set_bit(FLAGS_BUSY, &dd->flags))
-               return 0;
+       int err = 0, ret = 0;
 
        spin_lock_irqsave(&dd->lock, flags);
+       if (req)
+               ret = ahash_enqueue_request(&dd->queue, req);
+       if (dd->flags & FLAGS_BUSY) {
+               spin_unlock_irqrestore(&dd->lock, flags);
+               return ret;
+       }
        backlog = crypto_get_backlog(&dd->queue);
        async_req = crypto_dequeue_request(&dd->queue);
-       if (!async_req)
-               clear_bit(FLAGS_BUSY, &dd->flags);
+       if (async_req)
+               dd->flags |= FLAGS_BUSY;
        spin_unlock_irqrestore(&dd->lock, flags);
 
        if (!async_req)
-               return 0;
+               return ret;
 
        if (backlog)
                backlog->complete(backlog, -EINPROGRESS);
@@ -646,7 +699,22 @@ static int omap_sham_handle_queue(struct omap_sham_dev *dd)
        dev_dbg(dd->dev, "handling new req, op: %lu, nbytes: %d\n",
                                                ctx->op, req->nbytes);
 
-       if (req != prev_req && ctx->digcnt)
+
+       err = omap_sham_hw_init(dd);
+       if (err)
+               goto err1;
+
+       omap_set_dma_dest_params(dd->dma_lch, 0,
+                       OMAP_DMA_AMODE_CONSTANT,
+                       dd->phys_base + SHA_REG_DIN(0), 0, 16);
+
+       omap_set_dma_dest_burst_mode(dd->dma_lch,
+                       OMAP_DMA_DATA_BURST_16);
+
+       omap_set_dma_src_burst_mode(dd->dma_lch,
+                       OMAP_DMA_DATA_BURST_4);
+
+       if (ctx->digcnt)
                /* request has changed - restore hash */
                omap_sham_copy_hash(req, 0);
 
@@ -658,7 +726,7 @@ static int omap_sham_handle_queue(struct omap_sham_dev *dd)
        } else if (ctx->op == OP_FINAL) {
                err = omap_sham_final_req(dd);
        }
-
+err1:
        if (err != -EINPROGRESS) {
                /* done_task will not finish it, so do it here */
                omap_sham_finish_req(req, err);
@@ -667,7 +735,7 @@ static int omap_sham_handle_queue(struct omap_sham_dev *dd)
 
        dev_dbg(dd->dev, "exit, err: %d\n", err);
 
-       return err;
+       return ret;
 }
 
 static int omap_sham_enqueue(struct ahash_request *req, unsigned int op)
@@ -675,18 +743,10 @@ static int omap_sham_enqueue(struct ahash_request *req, unsigned int op)
        struct omap_sham_reqctx *ctx = ahash_request_ctx(req);
        struct omap_sham_ctx *tctx = crypto_tfm_ctx(req->base.tfm);
        struct omap_sham_dev *dd = tctx->dd;
-       unsigned long flags;
-       int err;
 
        ctx->op = op;
 
-       spin_lock_irqsave(&dd->lock, flags);
-       err = ahash_enqueue_request(&dd->queue, req);
-       spin_unlock_irqrestore(&dd->lock, flags);
-
-       omap_sham_handle_queue(dd);
-
-       return err;
+       return omap_sham_handle_queue(dd, req);
 }
 
 static int omap_sham_update(struct ahash_request *req)
@@ -709,21 +769,13 @@ static int omap_sham_update(struct ahash_request *req)
                        */
                        omap_sham_append_sg(ctx);
                        return 0;
-               } else if (ctx->bufcnt + ctx->total <= 64) {
+               } else if (ctx->bufcnt + ctx->total <= SHA1_MD5_BLOCK_SIZE) {
+                       /*
+                       * faster to use CPU for short transfers
+                       */
                        ctx->flags |= FLAGS_CPU;
-               } else if (!ctx->bufcnt && sg_is_last(ctx->sg)) {
-                       /* may be can use faster functions */
-                       int aligned = IS_ALIGNED((u32)ctx->sg->offset,
-                                                               sizeof(u32));
-
-                       if (aligned && (ctx->flags & FLAGS_FIRST))
-                               /* digest: first and final */
-                               ctx->flags |= FLAGS_FAST;
-
-                       ctx->flags &= ~FLAGS_FIRST;
                }
-       } else if (ctx->bufcnt + ctx->total <= ctx->buflen) {
-               /* if not finaup -> not fast */
+       } else if (ctx->bufcnt + ctx->total < ctx->buflen) {
                omap_sham_append_sg(ctx);
                return 0;
        }
@@ -761,12 +813,14 @@ static int omap_sham_final(struct ahash_request *req)
 
        ctx->flags |= FLAGS_FINUP;
 
-       /* OMAP HW accel works only with buffers >= 9 */
-       /* HMAC is always >= 9 because of ipad */
-       if ((ctx->digcnt + ctx->bufcnt) < 9)
-               err = omap_sham_final_shash(req);
-       else if (ctx->bufcnt)
-               return omap_sham_enqueue(req, OP_FINAL);
+       if (!(ctx->flags & FLAGS_ERROR)) {
+               /* OMAP HW accel works only with buffers >= 9 */
+               /* HMAC is always >= 9 because of ipad */
+               if ((ctx->digcnt + ctx->bufcnt) < 9)
+                       err = omap_sham_final_shash(req);
+               else if (ctx->bufcnt)
+                       return omap_sham_enqueue(req, OP_FINAL);
+       }
 
        omap_sham_cleanup(req);
 
@@ -836,6 +890,8 @@ static int omap_sham_cra_init_alg(struct crypto_tfm *tfm, const char *alg_base)
        struct omap_sham_ctx *tctx = crypto_tfm_ctx(tfm);
        const char *alg_name = crypto_tfm_alg_name(tfm);
 
+       pr_info("enter\n");
+
        /* Allocate a fallback and abort if it failed. */
        tctx->fallback = crypto_alloc_shash(alg_name, 0,
                                            CRYPTO_ALG_NEED_FALLBACK);
@@ -846,7 +902,7 @@ static int omap_sham_cra_init_alg(struct crypto_tfm *tfm, const char *alg_base)
        }
 
        crypto_ahash_set_reqsize(__crypto_ahash_cast(tfm),
-                                sizeof(struct omap_sham_reqctx));
+                                sizeof(struct omap_sham_reqctx) + BUFLEN);
 
        if (alg_base) {
                struct omap_sham_hmac_ctx *bctx = tctx->base;
@@ -932,7 +988,7 @@ static struct ahash_alg algs[] = {
                                                CRYPTO_ALG_NEED_FALLBACK,
                .cra_blocksize          = SHA1_BLOCK_SIZE,
                .cra_ctxsize            = sizeof(struct omap_sham_ctx),
-               .cra_alignmask          = 0,
+               .cra_alignmask          = OMAP_ALIGN_MASK,
                .cra_module             = THIS_MODULE,
                .cra_init               = omap_sham_cra_init,
                .cra_exit               = omap_sham_cra_exit,
@@ -956,7 +1012,7 @@ static struct ahash_alg algs[] = {
                .cra_blocksize          = SHA1_BLOCK_SIZE,
                .cra_ctxsize            = sizeof(struct omap_sham_ctx) +
                                        sizeof(struct omap_sham_hmac_ctx),
-               .cra_alignmask          = 0,
+               .cra_alignmask          = OMAP_ALIGN_MASK,
                .cra_module             = THIS_MODULE,
                .cra_init               = omap_sham_cra_sha1_init,
                .cra_exit               = omap_sham_cra_exit,
@@ -980,7 +1036,7 @@ static struct ahash_alg algs[] = {
                .cra_blocksize          = SHA1_BLOCK_SIZE,
                .cra_ctxsize            = sizeof(struct omap_sham_ctx) +
                                        sizeof(struct omap_sham_hmac_ctx),
-               .cra_alignmask          = 0,
+               .cra_alignmask          = OMAP_ALIGN_MASK,
                .cra_module             = THIS_MODULE,
                .cra_init               = omap_sham_cra_md5_init,
                .cra_exit               = omap_sham_cra_exit,
@@ -993,7 +1049,7 @@ static void omap_sham_done_task(unsigned long data)
        struct omap_sham_dev *dd = (struct omap_sham_dev *)data;
        struct ahash_request *req = dd->req;
        struct omap_sham_reqctx *ctx = ahash_request_ctx(req);
-       int ready = 1;
+       int ready = 0, err = 0;
 
        if (ctx->flags & FLAGS_OUTPUT_READY) {
                ctx->flags &= ~FLAGS_OUTPUT_READY;
@@ -1003,15 +1059,18 @@ static void omap_sham_done_task(unsigned long data)
        if (dd->flags & FLAGS_DMA_ACTIVE) {
                dd->flags &= ~FLAGS_DMA_ACTIVE;
                omap_sham_update_dma_stop(dd);
-               omap_sham_update_dma_slow(dd);
+               if (!dd->err)
+                       err = omap_sham_update_dma_start(dd);
        }
 
-       if (ready && !(dd->flags & FLAGS_DMA_ACTIVE)) {
-               dev_dbg(dd->dev, "update done\n");
+       err = dd->err ? : err;
+
+       if (err != -EINPROGRESS && (ready || err)) {
+               dev_dbg(dd->dev, "update done: err: %d\n", err);
                /* finish curent request */
-               omap_sham_finish_req(req, 0);
+               omap_sham_finish_req(req, err);
                /* start new request */
-               omap_sham_handle_queue(dd);
+               omap_sham_handle_queue(dd, NULL);
        }
 }
 
@@ -1019,7 +1078,7 @@ static void omap_sham_queue_task(unsigned long data)
 {
        struct omap_sham_dev *dd = (struct omap_sham_dev *)data;
 
-       omap_sham_handle_queue(dd);
+       omap_sham_handle_queue(dd, NULL);
 }
 
 static irqreturn_t omap_sham_irq(int irq, void *dev_id)
@@ -1041,6 +1100,7 @@ static irqreturn_t omap_sham_irq(int irq, void *dev_id)
        omap_sham_read(dd, SHA_REG_CTRL);
 
        ctx->flags |= FLAGS_OUTPUT_READY;
+       dd->err = 0;
        tasklet_schedule(&dd->done_task);
 
        return IRQ_HANDLED;
@@ -1050,8 +1110,13 @@ static void omap_sham_dma_callback(int lch, u16 ch_status, void *data)
 {
        struct omap_sham_dev *dd = data;
 
-       if (likely(lch == dd->dma_lch))
-               tasklet_schedule(&dd->done_task);
+       if (ch_status != OMAP_DMA_BLOCK_IRQ) {
+               pr_err("omap-sham DMA error status: 0x%hx\n", ch_status);
+               dd->err = -EIO;
+               dd->flags &= ~FLAGS_INIT; /* request to re-initialize */
+       }
+
+       tasklet_schedule(&dd->done_task);
 }
 
 static int omap_sham_dma_init(struct omap_sham_dev *dd)
@@ -1066,15 +1131,6 @@ static int omap_sham_dma_init(struct omap_sham_dev *dd)
                dev_err(dd->dev, "Unable to request DMA channel\n");
                return err;
        }
-       omap_set_dma_dest_params(dd->dma_lch, 0,
-                       OMAP_DMA_AMODE_CONSTANT,
-                       dd->phys_base + SHA_REG_DIN(0), 0, 16);
-
-       omap_set_dma_dest_burst_mode(dd->dma_lch,
-                       OMAP_DMA_DATA_BURST_16);
-
-       omap_set_dma_src_burst_mode(dd->dma_lch,
-                       OMAP_DMA_DATA_BURST_4);
 
        return 0;
 }