mirror of
https://github.com/torvalds/linux.git
synced 2024-11-14 16:12:02 +00:00
Merge git://git.kernel.org/pub/scm/linux/kernel/git/herbert/crypto-2.6
* git://git.kernel.org/pub/scm/linux/kernel/git/herbert/crypto-2.6: crypto: talitos - sparse fix crypto: talitos - Stop leaking memory in error path crypto: talitos - Fix GFP flag usage crypto: talitos - Preempt overflow interrupts crypto: talitos - Correct dst != src case handling crypto: talitos - Remove calls to of_node_put
This commit is contained in:
commit
7a76d89232
@ -99,6 +99,9 @@ struct talitos_private {
|
||||
/* next channel to be assigned next incoming descriptor */
|
||||
atomic_t last_chan;
|
||||
|
||||
/* per-channel number of requests pending in channel h/w fifo */
|
||||
atomic_t *submit_count;
|
||||
|
||||
/* per-channel request fifo */
|
||||
struct talitos_request **fifo;
|
||||
|
||||
@ -263,15 +266,15 @@ static int talitos_submit(struct device *dev, struct talitos_desc *desc,
|
||||
|
||||
spin_lock_irqsave(&priv->head_lock[ch], flags);
|
||||
|
||||
head = priv->head[ch];
|
||||
request = &priv->fifo[ch][head];
|
||||
|
||||
if (request->desc) {
|
||||
/* request queue is full */
|
||||
if (!atomic_inc_not_zero(&priv->submit_count[ch])) {
|
||||
/* h/w fifo is full */
|
||||
spin_unlock_irqrestore(&priv->head_lock[ch], flags);
|
||||
return -EAGAIN;
|
||||
}
|
||||
|
||||
head = priv->head[ch];
|
||||
request = &priv->fifo[ch][head];
|
||||
|
||||
/* map descriptor and save caller data */
|
||||
request->dma_desc = dma_map_single(dev, desc, sizeof(*desc),
|
||||
DMA_BIDIRECTIONAL);
|
||||
@ -335,6 +338,9 @@ static void flush_channel(struct device *dev, int ch, int error, int reset_ch)
|
||||
priv->tail[ch] = (tail + 1) & (priv->fifo_len - 1);
|
||||
|
||||
spin_unlock_irqrestore(&priv->tail_lock[ch], flags);
|
||||
|
||||
atomic_dec(&priv->submit_count[ch]);
|
||||
|
||||
saved_req.callback(dev, saved_req.desc, saved_req.context,
|
||||
status);
|
||||
/* channel may resume processing in single desc error case */
|
||||
@ -842,7 +848,7 @@ static int sg_to_link_tbl(struct scatterlist *sg, int sg_count,
|
||||
|
||||
/* adjust (decrease) last one (or two) entry's len to cryptlen */
|
||||
link_tbl_ptr--;
|
||||
while (link_tbl_ptr->len <= (-cryptlen)) {
|
||||
while (be16_to_cpu(link_tbl_ptr->len) <= (-cryptlen)) {
|
||||
/* Empty this entry, and move to previous one */
|
||||
cryptlen += be16_to_cpu(link_tbl_ptr->len);
|
||||
link_tbl_ptr->len = 0;
|
||||
@ -874,7 +880,7 @@ static int ipsec_esp(struct ipsec_esp_edesc *edesc, struct aead_request *areq,
|
||||
unsigned int cryptlen = areq->cryptlen;
|
||||
unsigned int authsize = ctx->authsize;
|
||||
unsigned int ivsize;
|
||||
int sg_count;
|
||||
int sg_count, ret;
|
||||
|
||||
/* hmac key */
|
||||
map_single_talitos_ptr(dev, &desc->ptr[0], ctx->authkeylen, &ctx->key,
|
||||
@ -978,7 +984,12 @@ static int ipsec_esp(struct ipsec_esp_edesc *edesc, struct aead_request *areq,
|
||||
map_single_talitos_ptr(dev, &desc->ptr[6], ivsize, ctx->iv, 0,
|
||||
DMA_FROM_DEVICE);
|
||||
|
||||
return talitos_submit(dev, desc, callback, areq);
|
||||
ret = talitos_submit(dev, desc, callback, areq);
|
||||
if (ret != -EINPROGRESS) {
|
||||
ipsec_esp_unmap(dev, edesc, areq);
|
||||
kfree(edesc);
|
||||
}
|
||||
return ret;
|
||||
}
|
||||
|
||||
|
||||
@ -1009,6 +1020,8 @@ static struct ipsec_esp_edesc *ipsec_esp_edesc_alloc(struct aead_request *areq,
|
||||
struct talitos_ctx *ctx = crypto_aead_ctx(authenc);
|
||||
struct ipsec_esp_edesc *edesc;
|
||||
int src_nents, dst_nents, alloc_len, dma_len;
|
||||
gfp_t flags = areq->base.flags & CRYPTO_TFM_REQ_MAY_SLEEP ? GFP_KERNEL :
|
||||
GFP_ATOMIC;
|
||||
|
||||
if (areq->cryptlen + ctx->authsize > TALITOS_MAX_DATA_LEN) {
|
||||
dev_err(ctx->dev, "cryptlen exceeds h/w max limit\n");
|
||||
@ -1022,7 +1035,7 @@ static struct ipsec_esp_edesc *ipsec_esp_edesc_alloc(struct aead_request *areq,
|
||||
dst_nents = src_nents;
|
||||
} else {
|
||||
dst_nents = sg_count(areq->dst, areq->cryptlen + ctx->authsize);
|
||||
dst_nents = (dst_nents == 1) ? 0 : src_nents;
|
||||
dst_nents = (dst_nents == 1) ? 0 : dst_nents;
|
||||
}
|
||||
|
||||
/*
|
||||
@ -1040,7 +1053,7 @@ static struct ipsec_esp_edesc *ipsec_esp_edesc_alloc(struct aead_request *areq,
|
||||
alloc_len += icv_stashing ? ctx->authsize : 0;
|
||||
}
|
||||
|
||||
edesc = kmalloc(alloc_len, GFP_DMA);
|
||||
edesc = kmalloc(alloc_len, GFP_DMA | flags);
|
||||
if (!edesc) {
|
||||
dev_err(ctx->dev, "could not allocate edescriptor\n");
|
||||
return ERR_PTR(-ENOMEM);
|
||||
@ -1337,6 +1350,7 @@ static int __devexit talitos_remove(struct of_device *ofdev)
|
||||
if (hw_supports(dev, DESC_HDR_SEL0_RNG))
|
||||
talitos_unregister_rng(dev);
|
||||
|
||||
kfree(priv->submit_count);
|
||||
kfree(priv->tail);
|
||||
kfree(priv->head);
|
||||
|
||||
@ -1466,9 +1480,6 @@ static int talitos_probe(struct of_device *ofdev,
|
||||
goto err_out;
|
||||
}
|
||||
|
||||
of_node_put(np);
|
||||
np = NULL;
|
||||
|
||||
priv->head_lock = kmalloc(sizeof(spinlock_t) * priv->num_channels,
|
||||
GFP_KERNEL);
|
||||
priv->tail_lock = kmalloc(sizeof(spinlock_t) * priv->num_channels,
|
||||
@ -1504,6 +1515,16 @@ static int talitos_probe(struct of_device *ofdev,
|
||||
}
|
||||
}
|
||||
|
||||
priv->submit_count = kmalloc(sizeof(atomic_t) * priv->num_channels,
|
||||
GFP_KERNEL);
|
||||
if (!priv->submit_count) {
|
||||
dev_err(dev, "failed to allocate fifo submit count space\n");
|
||||
err = -ENOMEM;
|
||||
goto err_out;
|
||||
}
|
||||
for (i = 0; i < priv->num_channels; i++)
|
||||
atomic_set(&priv->submit_count[i], -priv->chfifo_len);
|
||||
|
||||
priv->head = kzalloc(sizeof(int) * priv->num_channels, GFP_KERNEL);
|
||||
priv->tail = kzalloc(sizeof(int) * priv->num_channels, GFP_KERNEL);
|
||||
if (!priv->head || !priv->tail) {
|
||||
@ -1559,8 +1580,6 @@ static int talitos_probe(struct of_device *ofdev,
|
||||
|
||||
err_out:
|
||||
talitos_remove(ofdev);
|
||||
if (np)
|
||||
of_node_put(np);
|
||||
|
||||
return err;
|
||||
}
|
||||
|
Loading…
Reference in New Issue
Block a user