From: Kim Phillips <[email protected]>
since we ack early, the re-read interrupt status in talitos_error
may be already updated with a new value. Pass the error ISR value
directly in order to report and handle the error based on the correct
error status.
Also remove unused error tasklet.
Signed-off-by: Kim Phillips <[email protected]>
Signed-off-by: Lee Nipper <[email protected]>
---
In this series, patches 1 and 3 authored by Kim, patch 2 by Lee.
drivers/crypto/talitos.c | 12 +++---------
1 files changed, 3 insertions(+), 9 deletions(-)
diff --git a/drivers/crypto/talitos.c b/drivers/crypto/talitos.c
index 4c47fe5..b6705a3 100644
--- a/drivers/crypto/talitos.c
+++ b/drivers/crypto/talitos.c
@@ -127,7 +127,6 @@ struct talitos_private {
/* request callback tasklet */
struct tasklet_struct done_task;
- struct tasklet_struct error_task;
/* list of registered algorithms */
struct list_head alg_list;
@@ -469,16 +468,13 @@ static void report_eu_error(struct device *dev, int ch, struct talitos_desc *des
/*
* recover from error interrupts
*/
-static void talitos_error(unsigned long data)
+static void talitos_error(unsigned long data, u32 isr, u32 isr_lo)
{
struct device *dev = (struct device *)data;
struct talitos_private *priv = dev_get_drvdata(dev);
unsigned int timeout = TALITOS_TIMEOUT;
int ch, error, reset_dev = 0, reset_ch = 0;
- u32 isr, isr_lo, v, v_lo;
-
- isr = in_be32(priv->reg + TALITOS_ISR);
- isr_lo = in_be32(priv->reg + TALITOS_ISR_LO);
+ u32 v, v_lo;
for (ch = 0; ch < priv->num_channels; ch++) {
/* skip channels without errors */
@@ -566,7 +562,7 @@ static irqreturn_t talitos_interrupt(int irq, void *data)
out_be32(priv->reg + TALITOS_ICR_LO, isr_lo);
if (unlikely((isr & ~TALITOS_ISR_CHDONE) || isr_lo))
- talitos_error((unsigned long)data);
+ talitos_error((unsigned long)data, isr, isr_lo);
else
if (likely(isr & TALITOS_ISR_CHDONE))
tasklet_schedule(&priv->done_task);
@@ -1391,7 +1387,6 @@ static int __devexit talitos_remove(struct of_device *ofdev)
}
tasklet_kill(&priv->done_task);
- tasklet_kill(&priv->error_task);
iounmap(priv->reg);
@@ -1452,7 +1447,6 @@ static int talitos_probe(struct of_device *ofdev,
priv->ofdev = ofdev;
tasklet_init(&priv->done_task, talitos_done, (unsigned long)dev);
- tasklet_init(&priv->error_task, talitos_error, (unsigned long)dev);
priv->irq = irq_of_parse_and_map(np, 0);
--
1.5.6.2
On Mon, 06 Oct 2008 15:09:00 -0500
Lee Nipper <[email protected]> wrote:
> From: Kim Phillips <[email protected]>
>
> since we ack early, the re-read interrupt status in talitos_error
> may be already updated with a new value. Pass the error ISR value
> directly in order to report and handle the error based on the correct
> error status.
>
> Also remove unused error tasklet.
>
> Signed-off-by: Kim Phillips <[email protected]>
> Signed-off-by: Lee Nipper <[email protected]>
> ---
Herbert, I noticed your crypto update pull request was missing this
patch series (and the off-by-one fix from Suresh). I realize they were
posted a bit late, but are they under consideration for 2.6.28?
Kim
On Fri, Oct 10, 2008 at 05:58:48PM -0500, Kim Phillips wrote:
>
> Herbert, I noticed your crypto update pull request was missing this
> patch series (and the off-by-one fix from Suresh). I realize they were
> posted a bit late, but are they under consideration for 2.6.28?
I'm going to run them through cryptodev-2.6 before pushing upstream
if the schedule works out.
Cheers,
--
Visit Openswan at http://www.openswan.org/
Email: Herbert Xu ~{PmV>HI~} <[email protected]>
Home Page: http://gondor.apana.org.au/~herbert/
PGP Key: http://gondor.apana.org.au/~herbert/pubkey.txt
Lee Nipper <[email protected]> wrote:
> From: Kim Phillips <[email protected]>
>
> since we ack early, the re-read interrupt status in talitos_error
> may be already updated with a new value. Pass the error ISR value
> directly in order to report and handle the error based on the correct
> error status.
>
> Also remove unused error tasklet.
>
> Signed-off-by: Kim Phillips <[email protected]>
> Signed-off-by: Lee Nipper <[email protected]>
I've applied all three patches. However, I had to fix up the
bits around INIT_LIST_HEAD in 1 and 3. So please check the
cryptodev-2.6 tree and let me know if it's wrong.
Thanks,
--
Visit Openswan at http://www.openswan.org/
Email: Herbert Xu ~{PmV>HI~} <[email protected]>
Home Page: http://gondor.apana.org.au/~herbert/
PGP Key: http://gondor.apana.org.au/~herbert/pubkey.txt
On Sun, 2008-10-12 at 05:34 -0700, Herbert Xu wrote:
> I've applied all three patches. However, I had to fix up the
> bits around INIT_LIST_HEAD in 1 and 3. So please check the
> cryptodev-2.6 tree and let me know if it's wrong.
>
I checked cryptodev-2.6 and it looks good. Thank you!
I thought I had double checked the 3 patches against mainline
but obviously I had not. I must have gotten skewed a bit.
I will be more careful.
Thanks for the fix up.
Lee