When an external device generated a level based interrupt then the
interrupt controller could miss the interrupt. The reason is that the
interrupt controller can detect only link changes.
In the following example, if there is a PHY that generates an interrupt
then the following would be happen. The GPIO detected that the interrupt
line changed, and then the 'ocelot_irq_handler' will be called. Here it
detects which GPIO line seen the change and for that will call the
following:
1. irq_mask
2. phy interrupt routine
3. irq_eoi
4. irq_unmask
And this works fine for simple cases, but if the PHY generates many
interrupts, for example when doing PTP timestamping, then the following
could happen. Again the function 'ocelot_irq_handler' will be called
and then from here the following could happen:
1. irq_mask
2. phy interrupt routine
3. irq_eoi
4. irq_unmask
Right before step 3(irq_eoi), the PHY will generate another interrupt.
Now the interrupt controller will acknowledge the change in the
interrupt line. So we miss the interrupt.
A solution will be to use 'handle_level_irq' instead of
'handle_fasteoi_irq', because for this will change routine order of
handling the interrupt.
1. irq_mask
2. irq_ack
3. phy interrupt routine
4. irq_unmask
And now if the PHY will generate a new interrupt before irq_unmask, the
interrupt controller will detect this because it already acknowledge the
change in interrupt line at step 2(irq_ack).
But this is not the full solution because there is another issue. In
case there are 2 PHYs that shared the interrupt line. For example phy1
generates an interruot, then the following can happen:
1.irq_mask
2.irq_ack
3.phy0 interrupt routine
4.phy1 interrupt routine
5.irq_unmask
In case phy0 will generate an interrupt while clearing the interrupt
source in phy1, then the interrupt line will be kept down by phy0. So
the interrupt controller will not see any changes in the interrupt line.
The solution here is to update 'irq_unmask' such that it can detect if
the interrupt line is still active or not. And if it is active then call
again the procedure to clear the interrupts. But we don't want to do it
every time, only if we know that the interrupt controller have not seen
already that the interrupt line has changed.
While at this, add support also for IRQ_TYPE_LEVEL_LOW.
Fixes: be36abb71d878f ("pinctrl: ocelot: add support for interrupt controller")
Signed-off-by: Horatiu Vultur <[email protected]>
---
drivers/pinctrl/pinctrl-ocelot.c | 93 ++++++++++++++++++++++++++++----
1 file changed, 84 insertions(+), 9 deletions(-)
diff --git a/drivers/pinctrl/pinctrl-ocelot.c b/drivers/pinctrl/pinctrl-ocelot.c
index c5fd154990c8..ae734345c606 100644
--- a/drivers/pinctrl/pinctrl-ocelot.c
+++ b/drivers/pinctrl/pinctrl-ocelot.c
@@ -338,6 +338,11 @@ struct ocelot_match_data {
struct ocelot_pincfg_data pincfg_data;
};
+struct ocelot_irq_work {
+ struct work_struct irq_work;
+ struct irq_desc *irq_desc;
+};
+
#define LUTON_P(p, f0, f1) \
static struct ocelot_pin_caps luton_pin_##p = { \
.pin = p, \
@@ -1813,6 +1818,77 @@ static void ocelot_irq_mask(struct irq_data *data)
gpiochip_disable_irq(chip, gpio);
}
+static void ocelot_irq_work(struct work_struct *work)
+{
+ struct ocelot_irq_work *w = container_of(work, struct ocelot_irq_work, irq_work);
+ struct irq_chip *parent_chip = irq_desc_get_chip(w->irq_desc);
+ struct gpio_chip *chip = irq_desc_get_chip_data(w->irq_desc);
+ struct irq_data *data = irq_desc_get_irq_data(w->irq_desc);
+ unsigned int gpio = irqd_to_hwirq(data);
+
+ local_irq_disable();
+ chained_irq_enter(parent_chip, w->irq_desc);
+ generic_handle_domain_irq(chip->irq.domain, gpio);
+ chained_irq_exit(parent_chip, w->irq_desc);
+ local_irq_enable();
+
+ kfree(w);
+}
+
+static void ocelot_irq_unmask_level(struct irq_data *data)
+{
+ struct gpio_chip *chip = irq_data_get_irq_chip_data(data);
+ struct ocelot_pinctrl *info = gpiochip_get_data(chip);
+ struct irq_desc *desc = irq_data_to_desc(data);
+ unsigned int gpio = irqd_to_hwirq(data);
+ bool ack = false, active = false;
+ u8 trigger_level = IRQ_TYPE_NONE;
+ int val;
+
+ /*
+ * It is enough to read only one action because the trigger level is the
+ * same for all of them.
+ */
+ if (desc->action)
+ trigger_level = desc->action->flags & IRQ_TYPE_SENSE_MASK;
+
+ /* Check if the interrupt line is still active */
+ regmap_read(info->map, REG(OCELOT_GPIO_IN, info, gpio), &val);
+ if ((!(val & BIT(gpio % 32)) && trigger_level == IRQ_TYPE_LEVEL_LOW) ||
+ (val & BIT(gpio % 32) && trigger_level == IRQ_TYPE_LEVEL_HIGH))
+ active = true;
+
+ /*
+ * Check if the interrupt controller has seen any changes in the
+ * interrupt line
+ */
+ regmap_read(info->map, REG(OCELOT_GPIO_INTR, info, gpio), &val);
+ if (val & BIT(gpio % 32))
+ ack = true;
+
+ /* Enable the interrupt now */
+ gpiochip_enable_irq(chip, gpio);
+ regmap_update_bits(info->map, REG(OCELOT_GPIO_INTR_ENA, info, gpio),
+ BIT(gpio % 32), BIT(gpio % 32));
+
+ /*
+ * In case the interrupt line is still active and the interrupt
+ * controller has not seen any changes in the interrupt line, then it
+ * means that there happen another interrupt while the line was active.
+ * So we missed that one, so we need to kick again the interrupt handler
+ */
+ if (active && !ack) {
+ struct ocelot_irq_work *work = kmalloc(sizeof(*work), GFP_ATOMIC);
+
+ if (!work)
+ return;
+
+ work->irq_desc = desc;
+ INIT_WORK(&work->irq_work, ocelot_irq_work);
+ queue_work(system_wq, &work->irq_work);
+ }
+}
+
static void ocelot_irq_unmask(struct irq_data *data)
{
struct gpio_chip *chip = irq_data_get_irq_chip_data(data);
@@ -1836,13 +1912,12 @@ static void ocelot_irq_ack(struct irq_data *data)
static int ocelot_irq_set_type(struct irq_data *data, unsigned int type);
-static struct irq_chip ocelot_eoi_irqchip = {
+static struct irq_chip ocelot_level_irqchip = {
.name = "gpio",
.irq_mask = ocelot_irq_mask,
- .irq_eoi = ocelot_irq_ack,
- .irq_unmask = ocelot_irq_unmask,
- .flags = IRQCHIP_EOI_THREADED | IRQCHIP_EOI_IF_HANDLED |
- IRQCHIP_IMMUTABLE,
+ .irq_ack = ocelot_irq_ack,
+ .irq_unmask = ocelot_irq_unmask_level,
+ .flags = IRQCHIP_IMMUTABLE,
.irq_set_type = ocelot_irq_set_type,
GPIOCHIP_IRQ_RESOURCE_HELPERS
};
@@ -1861,12 +1936,12 @@ static int ocelot_irq_set_type(struct irq_data *data, unsigned int type)
{
type &= IRQ_TYPE_SENSE_MASK;
- if (!(type & (IRQ_TYPE_EDGE_BOTH | IRQ_TYPE_LEVEL_HIGH)))
+ if (type == IRQ_TYPE_NONE)
return -EINVAL;
- if (type & IRQ_TYPE_LEVEL_HIGH)
- irq_set_chip_handler_name_locked(data, &ocelot_eoi_irqchip,
- handle_fasteoi_irq, NULL);
+ if (type & (IRQ_TYPE_LEVEL_HIGH | IRQ_TYPE_LEVEL_LOW))
+ irq_set_chip_handler_name_locked(data, &ocelot_level_irqchip,
+ handle_level_irq, NULL);
if (type & IRQ_TYPE_EDGE_BOTH)
irq_set_chip_handler_name_locked(data, &ocelot_irqchip,
handle_edge_irq, NULL);
--
2.33.0
On Fri, Sep 2, 2022 at 3:40 PM Horatiu Vultur
<[email protected]> wrote:
>
> When an external device generated a level based interrupt then the
> interrupt controller could miss the interrupt. The reason is that the
> interrupt controller can detect only link changes.
>
> In the following example, if there is a PHY that generates an interrupt
> then the following would be happen. The GPIO detected that the interrupt
would happen
> line changed, and then the 'ocelot_irq_handler' will be called. Here it
> detects which GPIO line seen the change and for that will call the
> following:
> 1. irq_mask
> 2. phy interrupt routine
> 3. irq_eoi
> 4. irq_unmask
>
> And this works fine for simple cases, but if the PHY generates many
> interrupts, for example when doing PTP timestamping, then the following
> could happen. Again the function 'ocelot_irq_handler' will be called
> and then from here the following could happen:
> 1. irq_mask
> 2. phy interrupt routine
> 3. irq_eoi
> 4. irq_unmask
>
> Right before step 3(irq_eoi), the PHY will generate another interrupt.
> Now the interrupt controller will acknowledge the change in the
> interrupt line. So we miss the interrupt.
>
> A solution will be to use 'handle_level_irq' instead of
> 'handle_fasteoi_irq', because for this will change routine order of
> handling the interrupt.
> 1. irq_mask
> 2. irq_ack
> 3. phy interrupt routine
> 4. irq_unmask
>
> And now if the PHY will generate a new interrupt before irq_unmask, the
> interrupt controller will detect this because it already acknowledge the
> change in interrupt line at step 2(irq_ack).
>
> But this is not the full solution because there is another issue. In
> case there are 2 PHYs that shared the interrupt line. For example phy1
share
> generates an interruot, then the following can happen:
interrupt
> 1.irq_mask
> 2.irq_ack
> 3.phy0 interrupt routine
> 4.phy1 interrupt routine
> 5.irq_unmask
>
> In case phy0 will generate an interrupt while clearing the interrupt
> source in phy1, then the interrupt line will be kept down by phy0. So
> the interrupt controller will not see any changes in the interrupt line.
> The solution here is to update 'irq_unmask' such that it can detect if
> the interrupt line is still active or not. And if it is active then call
> again the procedure to clear the interrupts. But we don't want to do it
> every time, only if we know that the interrupt controller have not seen
> already that the interrupt line has changed.
>
> While at this, add support also for IRQ_TYPE_LEVEL_LOW.
...
> + /*
> + * It is enough to read only one action because the trigger level is the
> + * same for all of them.
> + */
Hmm... this is interesting. How is the hardware supposed to work if
the user asks for two contradictory levels for two different IRQs?
...
> + /*
> + * Check if the interrupt controller has seen any changes in the
> + * interrupt line
Missed period.
> + */
...
> + /*
> + * In case the interrupt line is still active and the interrupt
> + * controller has not seen any changes in the interrupt line, then it
> + * means that there happen another interrupt while the line was active.
> + * So we missed that one, so we need to kick again the interrupt handler
> + */
Ditto.
...
> + struct ocelot_irq_work *work = kmalloc(sizeof(*work), GFP_ATOMIC);
It's more visible what's going on if you split this to definition and
assignment and move assignment closer to its first user.
> + if (!work)
> + return;
...
> type &= IRQ_TYPE_SENSE_MASK;
This seems redundant, see below.
> - if (!(type & (IRQ_TYPE_EDGE_BOTH | IRQ_TYPE_LEVEL_HIGH)))
> + if (type == IRQ_TYPE_NONE)
> return -EINVAL;
Is it ever possible? IIRC the IRQ chip code, the set->type won't be
called at all in such a case. Also type is already limited to the
sense mask, no?
--
With Best Regards,
Andy Shevchenko
The 09/02/2022 17:51, Andy Shevchenko wrote:
Hi Andy,
>
> > + /*
> > + * It is enough to read only one action because the trigger level is the
> > + * same for all of them.
> > + */
>
> Hmm... this is interesting. How is the hardware supposed to work if
> the user asks for two contradictory levels for two different IRQs?
The HW can detect the changes in line for each pin on which the IRQ is.
And each pin will have a different irq_desc with different actions.
Or maybe I missunderstood the question?
Also maybe a better way to get trigger level is to use
irqd_get_trigger_type.
...
> > + struct ocelot_irq_work *work = kmalloc(sizeof(*work), GFP_ATOMIC);
>
> It's more visible what's going on if you split this to definition and
> assignment and move assignment closer to its first user.
>
> > + if (!work)
> > + return;
So you would like something like this:
---
struct ocelot_irq_work *work;
work = kmalloc(sizeof(*work), GFP_ATOMIC);
if (!work)
return;
...
---
>
> ...
>
> > type &= IRQ_TYPE_SENSE_MASK;
>
> This seems redundant, see below.
>
>
> > - if (!(type & (IRQ_TYPE_EDGE_BOTH | IRQ_TYPE_LEVEL_HIGH)))
> > + if (type == IRQ_TYPE_NONE)
> > return -EINVAL;
>
> Is it ever possible? IIRC the IRQ chip code, the set->type won't be
> called at all in such a case. Also type is already limited to the
> sense mask, no?
It is not possible. From what I have seen on the callstack, the type is
already anded with IRQ_TYPE_SENSE_MASK, and it would not call
ocelot_irq_set_type for IRQ_TYPE_NONE.
Therefor I will remove these.
>
> --
> With Best Regards,
> Andy Shevchenko
--
/Horatiu