crypto: inside-secure - move request dequeueing into a workqueue

This patch moves the request dequeueing into a workqueue to improve the
coalescing of interrupts when sending requests to the engine; as the
engine is capable of having one single interrupt for n requests sent.
Using a workqueue allows to send more request at once.

Suggested-by: Ofer Heifetz <oferh@marvell.com>
Signed-off-by: Antoine Tenart <antoine.tenart@free-electrons.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
This commit is contained in:
Antoine Ténart
2017-12-14 15:26:51 +01:00
committed by Herbert Xu
parent 5290ad6e9a
commit 8472e77810
4 changed files with 29 additions and 26 deletions
+16 -13
View File
@@ -429,8 +429,6 @@ void safexcel_dequeue(struct safexcel_crypto_priv *priv, int ring)
struct safexcel_request *request;
int ret, nreq = 0, cdesc = 0, rdesc = 0, commands, results;
priv->ring[ring].need_dequeue = false;
do {
spin_lock_bh(&priv->ring[ring].queue_lock);
backlog = crypto_get_backlog(&priv->ring[ring].queue);
@@ -445,8 +443,6 @@ void safexcel_dequeue(struct safexcel_crypto_priv *priv, int ring)
spin_lock_bh(&priv->ring[ring].queue_lock);
crypto_enqueue_request(&priv->ring[ring].queue, req);
spin_unlock_bh(&priv->ring[ring].queue_lock);
priv->ring[ring].need_dequeue = true;
goto finalize;
}
@@ -455,7 +451,6 @@ void safexcel_dequeue(struct safexcel_crypto_priv *priv, int ring)
if (ret) {
kfree(request);
req->complete(req, ret);
priv->ring[ring].need_dequeue = true;
goto finalize;
}
@@ -471,9 +466,7 @@ void safexcel_dequeue(struct safexcel_crypto_priv *priv, int ring)
} while (nreq++ < EIP197_MAX_BATCH_SZ);
finalize:
if (nreq == EIP197_MAX_BATCH_SZ)
priv->ring[ring].need_dequeue = true;
else if (!nreq)
if (!nreq)
return;
spin_lock_bh(&priv->ring[ring].lock);
@@ -628,13 +621,18 @@ static inline void safexcel_handle_result_descriptor(struct safexcel_crypto_priv
static void safexcel_handle_result_work(struct work_struct *work)
{
struct safexcel_work_data *data =
container_of(work, struct safexcel_work_data, work);
container_of(work, struct safexcel_work_data, result_work);
struct safexcel_crypto_priv *priv = data->priv;
safexcel_handle_result_descriptor(priv, data->ring);
}
if (priv->ring[data->ring].need_dequeue)
safexcel_dequeue(data->priv, data->ring);
static void safexcel_dequeue_work(struct work_struct *work)
{
struct safexcel_work_data *data =
container_of(work, struct safexcel_work_data, work);
safexcel_dequeue(data->priv, data->ring);
}
struct safexcel_ring_irq_data {
@@ -665,7 +663,10 @@ static irqreturn_t safexcel_irq_ring(int irq, void *data)
*/
dev_err(priv->dev, "RDR: fatal error.");
} else if (likely(stat & EIP197_xDR_THRESH)) {
queue_work(priv->ring[ring].workqueue, &priv->ring[ring].work_data.work);
queue_work(priv->ring[ring].workqueue,
&priv->ring[ring].work_data.result_work);
queue_work(priv->ring[ring].workqueue,
&priv->ring[ring].work_data.work);
}
/* ACK the interrupts */
@@ -846,7 +847,9 @@ static int safexcel_probe(struct platform_device *pdev)
priv->ring[i].work_data.priv = priv;
priv->ring[i].work_data.ring = i;
INIT_WORK(&priv->ring[i].work_data.work, safexcel_handle_result_work);
INIT_WORK(&priv->ring[i].work_data.result_work,
safexcel_handle_result_work);
INIT_WORK(&priv->ring[i].work_data.work, safexcel_dequeue_work);
snprintf(wq_name, 9, "wq_ring%d", i);
priv->ring[i].workqueue = create_singlethread_workqueue(wq_name);
+1 -1
View File
@@ -459,6 +459,7 @@ struct safexcel_config {
struct safexcel_work_data {
struct work_struct work;
struct work_struct result_work;
struct safexcel_crypto_priv *priv;
int ring;
};
@@ -489,7 +490,6 @@ struct safexcel_crypto_priv {
/* queue */
struct crypto_queue queue;
spinlock_t queue_lock;
bool need_dequeue;
} ring[EIP197_MAX_RINGS];
};
@@ -358,8 +358,8 @@ static int safexcel_handle_inv_result(struct safexcel_crypto_priv *priv,
if (enq_ret != -EINPROGRESS)
*ret = enq_ret;
if (!priv->ring[ring].need_dequeue)
safexcel_dequeue(priv, ring);
queue_work(priv->ring[ring].workqueue,
&priv->ring[ring].work_data.work);
*should_complete = false;
@@ -448,8 +448,8 @@ static int safexcel_cipher_exit_inv(struct crypto_tfm *tfm)
crypto_enqueue_request(&priv->ring[ring].queue, &req->base);
spin_unlock_bh(&priv->ring[ring].queue_lock);
if (!priv->ring[ring].need_dequeue)
safexcel_dequeue(priv, ring);
queue_work(priv->ring[ring].workqueue,
&priv->ring[ring].work_data.work);
wait_for_completion_interruptible(&result.completion);
@@ -495,8 +495,8 @@ static int safexcel_aes(struct skcipher_request *req,
ret = crypto_enqueue_request(&priv->ring[ring].queue, &req->base);
spin_unlock_bh(&priv->ring[ring].queue_lock);
if (!priv->ring[ring].need_dequeue)
safexcel_dequeue(priv, ring);
queue_work(priv->ring[ring].workqueue,
&priv->ring[ring].work_data.work);
return ret;
}
+6 -6
View File
@@ -381,8 +381,8 @@ static int safexcel_handle_inv_result(struct safexcel_crypto_priv *priv,
if (enq_ret != -EINPROGRESS)
*ret = enq_ret;
if (!priv->ring[ring].need_dequeue)
safexcel_dequeue(priv, ring);
queue_work(priv->ring[ring].workqueue,
&priv->ring[ring].work_data.work);
*should_complete = false;
@@ -470,8 +470,8 @@ static int safexcel_ahash_exit_inv(struct crypto_tfm *tfm)
crypto_enqueue_request(&priv->ring[ring].queue, &req->base);
spin_unlock_bh(&priv->ring[ring].queue_lock);
if (!priv->ring[ring].need_dequeue)
safexcel_dequeue(priv, ring);
queue_work(priv->ring[ring].workqueue,
&priv->ring[ring].work_data.work);
wait_for_completion_interruptible(&result.completion);
@@ -556,8 +556,8 @@ static int safexcel_ahash_enqueue(struct ahash_request *areq)
ret = crypto_enqueue_request(&priv->ring[ring].queue, &areq->base);
spin_unlock_bh(&priv->ring[ring].queue_lock);
if (!priv->ring[ring].need_dequeue)
safexcel_dequeue(priv, ring);
queue_work(priv->ring[ring].workqueue,
&priv->ring[ring].work_data.work);
return ret;
}