From: Gustav Ekelund <[email protected]>
The force watchdog event bit is not cleared during SW reset in the
mv88e6393x switch. This is a different behavior compared to mv886390 which
clears the force WD event bit as advertised. This causes a force WD event
to be handled over and over again as the SW reset following the event never
clears the force WD event bit.
Explicitly clear the watchdog event register to 0 in irq_action when
handling an event to prevent the switch from sending continuous interrupts.
Marvell aren't aware of any other stuck bits apart from the force WD
bit.
Signed-off-by: Gustav Ekelund <[email protected]>
---
Notes:
v2 Clarify commit message and added a comment
drivers/net/dsa/mv88e6xxx/chip.c | 2 +-
drivers/net/dsa/mv88e6xxx/global2.c | 20 ++++++++++++++++++++
drivers/net/dsa/mv88e6xxx/global2.h | 1 +
3 files changed, 22 insertions(+), 1 deletion(-)
diff --git a/drivers/net/dsa/mv88e6xxx/chip.c b/drivers/net/dsa/mv88e6xxx/chip.c
index 30383c4f8fd0..ee22d4785e9e 100644
--- a/drivers/net/dsa/mv88e6xxx/chip.c
+++ b/drivers/net/dsa/mv88e6xxx/chip.c
@@ -5596,7 +5596,7 @@ static const struct mv88e6xxx_ops mv88e6393x_ops = {
* .port_set_upstream_port method.
*/
.set_egress_port = mv88e6393x_set_egress_port,
- .watchdog_ops = &mv88e6390_watchdog_ops,
+ .watchdog_ops = &mv88e6393x_watchdog_ops,
.mgmt_rsvd2cpu = mv88e6393x_port_mgmt_rsvd2cpu,
.pot_clear = mv88e6xxx_g2_pot_clear,
.reset = mv88e6352_g1_reset,
diff --git a/drivers/net/dsa/mv88e6xxx/global2.c b/drivers/net/dsa/mv88e6xxx/global2.c
index ed3b2f88e783..a7af3cebae97 100644
--- a/drivers/net/dsa/mv88e6xxx/global2.c
+++ b/drivers/net/dsa/mv88e6xxx/global2.c
@@ -943,6 +943,26 @@ const struct mv88e6xxx_irq_ops mv88e6390_watchdog_ops = {
.irq_free = mv88e6390_watchdog_free,
};
+static int mv88e6393x_watchdog_action(struct mv88e6xxx_chip *chip, int irq)
+{
+ mv88e6390_watchdog_action(chip, irq);
+
+ /* Fix for clearing the force WD event bit.
+ * Unreleased erratum on mv88e6393x.
+ */
+ mv88e6xxx_g2_write(chip, MV88E6390_G2_WDOG_CTL,
+ MV88E6390_G2_WDOG_CTL_UPDATE |
+ MV88E6390_G2_WDOG_CTL_PTR_EVENT);
+
+ return IRQ_HANDLED;
+}
+
+const struct mv88e6xxx_irq_ops mv88e6393x_watchdog_ops = {
+ .irq_action = mv88e6393x_watchdog_action,
+ .irq_setup = mv88e6390_watchdog_setup,
+ .irq_free = mv88e6390_watchdog_free,
+};
+
static irqreturn_t mv88e6xxx_g2_watchdog_thread_fn(int irq, void *dev_id)
{
struct mv88e6xxx_chip *chip = dev_id;
diff --git a/drivers/net/dsa/mv88e6xxx/global2.h b/drivers/net/dsa/mv88e6xxx/global2.h
index e973114d6890..7e091965582b 100644
--- a/drivers/net/dsa/mv88e6xxx/global2.h
+++ b/drivers/net/dsa/mv88e6xxx/global2.h
@@ -369,6 +369,7 @@ int mv88e6xxx_g2_device_mapping_write(struct mv88e6xxx_chip *chip, int target,
extern const struct mv88e6xxx_irq_ops mv88e6097_watchdog_ops;
extern const struct mv88e6xxx_irq_ops mv88e6250_watchdog_ops;
extern const struct mv88e6xxx_irq_ops mv88e6390_watchdog_ops;
+extern const struct mv88e6xxx_irq_ops mv88e6393x_watchdog_ops;
extern const struct mv88e6xxx_avb_ops mv88e6165_avb_ops;
extern const struct mv88e6xxx_avb_ops mv88e6352_avb_ops;
--
2.30.2
On Fri, Mar 31, 2023 at 10:40:13AM +0200, Gustav Ekelund wrote:
> From: Gustav Ekelund <[email protected]>
>
> The force watchdog event bit is not cleared during SW reset in the
> mv88e6393x switch. This is a different behavior compared to mv886390 which
> clears the force WD event bit as advertised. This causes a force WD event
> to be handled over and over again as the SW reset following the event never
> clears the force WD event bit.
>
> Explicitly clear the watchdog event register to 0 in irq_action when
> handling an event to prevent the switch from sending continuous interrupts.
> Marvell aren't aware of any other stuck bits apart from the force WD
> bit.
>
> Signed-off-by: Gustav Ekelund <[email protected]>
Reviewed-by: Andrew Lunn <[email protected]>
Andrew
On 3/31/2023 1:40 AM, Gustav Ekelund wrote:
> From: Gustav Ekelund <[email protected]>
>
> The force watchdog event bit is not cleared during SW reset in the
> mv88e6393x switch. This is a different behavior compared to mv886390 which
> clears the force WD event bit as advertised. This causes a force WD event
> to be handled over and over again as the SW reset following the event never
> clears the force WD event bit.
>
> Explicitly clear the watchdog event register to 0 in irq_action when
> handling an event to prevent the switch from sending continuous interrupts.
> Marvell aren't aware of any other stuck bits apart from the force WD
> bit.
>
> Signed-off-by: Gustav Ekelund <[email protected]>
Would that deserve:
Fixes: de776d0d316f ("net: dsa: mv88e6xxx: add support for mv88e6393x
family")
Reviewed-by: Florian Fainelli <[email protected]>
--
Florian
Hello:
This patch was applied to netdev/net.git (main)
by David S. Miller <[email protected]>:
On Fri, 31 Mar 2023 10:40:13 +0200 you wrote:
> From: Gustav Ekelund <[email protected]>
>
> The force watchdog event bit is not cleared during SW reset in the
> mv88e6393x switch. This is a different behavior compared to mv886390 which
> clears the force WD event bit as advertised. This causes a force WD event
> to be handled over and over again as the SW reset following the event never
> clears the force WD event bit.
>
> [...]
Here is the summary with links:
- [net,v2] net: dsa: mv88e6xxx: Reset mv88e6393x force WD event bit
https://git.kernel.org/netdev/net/c/089b91a0155c
You are awesome, thank you!
--
Deet-doot-dot, I am a bot.
https://korg.docs.kernel.org/patchwork/pwbot.html