| Message ID | 20210707062004.782787-1-msp@baylibre.com (mailing list archive) |
|---|---|
| State | New, archived |
| Headers | show |
| Series | irqchip: mtk-sysirq: Skip setting irq-wake | expand |
Hi Markus, On Wed, 07 Jul 2021 07:20:04 +0100, Markus Schneider-Pargmann <msp@baylibre.com> wrote: > > mtk-sysirq doesn't require specific logic to work with wakeup IRQs. To > allow registered IRQs to be used as a wakeup-source, add the flag > IRQCHIP_SKIP_SET_WAKE. > > Signed-off-by: Markus Schneider-Pargmann <msp@baylibre.com> > --- > drivers/irqchip/irq-mtk-sysirq.c | 1 + > 1 file changed, 1 insertion(+) > > diff --git a/drivers/irqchip/irq-mtk-sysirq.c b/drivers/irqchip/irq-mtk-sysirq.c > index 6ff98b87e5c0..586e52d5442b 100644 > --- a/drivers/irqchip/irq-mtk-sysirq.c > +++ b/drivers/irqchip/irq-mtk-sysirq.c > @@ -65,6 +65,7 @@ static struct irq_chip mtk_sysirq_chip = { > .irq_set_type = mtk_sysirq_set_type, > .irq_retrigger = irq_chip_retrigger_hierarchy, > .irq_set_affinity = irq_chip_set_affinity_parent, > + .flags = IRQCHIP_SKIP_SET_WAKE, > }; > > static int mtk_sysirq_domain_translate(struct irq_domain *d, Is this a fix? If so, please provide a Fixes: tag matching the commit it addresses. Add a Cc stable if this requires backporting. Thanks, M.
Hi Marc, On Wed, Jul 07, 2021 at 11:19:19AM +0100, Marc Zyngier wrote: > Hi Markus, > > On Wed, 07 Jul 2021 07:20:04 +0100, > Markus Schneider-Pargmann <msp@baylibre.com> wrote: > > > > mtk-sysirq doesn't require specific logic to work with wakeup IRQs. To > > allow registered IRQs to be used as a wakeup-source, add the flag > > IRQCHIP_SKIP_SET_WAKE. > > > > Signed-off-by: Markus Schneider-Pargmann <msp@baylibre.com> > > --- > > drivers/irqchip/irq-mtk-sysirq.c | 1 + > > 1 file changed, 1 insertion(+) > > > > diff --git a/drivers/irqchip/irq-mtk-sysirq.c b/drivers/irqchip/irq-mtk-sysirq.c > > index 6ff98b87e5c0..586e52d5442b 100644 > > --- a/drivers/irqchip/irq-mtk-sysirq.c > > +++ b/drivers/irqchip/irq-mtk-sysirq.c > > @@ -65,6 +65,7 @@ static struct irq_chip mtk_sysirq_chip = { > > .irq_set_type = mtk_sysirq_set_type, > > .irq_retrigger = irq_chip_retrigger_hierarchy, > > .irq_set_affinity = irq_chip_set_affinity_parent, > > + .flags = IRQCHIP_SKIP_SET_WAKE, > > }; > > > > static int mtk_sysirq_domain_translate(struct irq_domain *d, > > Is this a fix? If so, please provide a Fixes: tag matching the commit > it addresses. Add a Cc stable if this requires backporting. At least from my point of view I would like to prepare the mtk-sysirq for powermanagement stuff, so not a fix. There may be some components trying to use this already but I would consider this more of a feature as it has probably never worked before. Currently calls to enable_irq_wake() using IRQs of sysirq will result in -ENXIO. Thanks, Markus > > Thanks, > > M. > > -- > Without deviation from the norm, progress is not possible.
On Wed, 07 Jul 2021 13:49:32 +0100, Markus Schneider-Pargmann <msp@baylibre.com> wrote: > > Hi Marc, > > On Wed, Jul 07, 2021 at 11:19:19AM +0100, Marc Zyngier wrote: > > Hi Markus, > > > > On Wed, 07 Jul 2021 07:20:04 +0100, > > Markus Schneider-Pargmann <msp@baylibre.com> wrote: > > > > > > mtk-sysirq doesn't require specific logic to work with wakeup IRQs. To > > > allow registered IRQs to be used as a wakeup-source, add the flag > > > IRQCHIP_SKIP_SET_WAKE. > > > > > > Signed-off-by: Markus Schneider-Pargmann <msp@baylibre.com> > > > --- > > > drivers/irqchip/irq-mtk-sysirq.c | 1 + > > > 1 file changed, 1 insertion(+) > > > > > > diff --git a/drivers/irqchip/irq-mtk-sysirq.c b/drivers/irqchip/irq-mtk-sysirq.c > > > index 6ff98b87e5c0..586e52d5442b 100644 > > > --- a/drivers/irqchip/irq-mtk-sysirq.c > > > +++ b/drivers/irqchip/irq-mtk-sysirq.c > > > @@ -65,6 +65,7 @@ static struct irq_chip mtk_sysirq_chip = { > > > .irq_set_type = mtk_sysirq_set_type, > > > .irq_retrigger = irq_chip_retrigger_hierarchy, > > > .irq_set_affinity = irq_chip_set_affinity_parent, > > > + .flags = IRQCHIP_SKIP_SET_WAKE, > > > }; > > > > > > static int mtk_sysirq_domain_translate(struct irq_domain *d, > > > > Is this a fix? If so, please provide a Fixes: tag matching the commit > > it addresses. Add a Cc stable if this requires backporting. > > At least from my point of view I would like to prepare the mtk-sysirq > for powermanagement stuff, so not a fix. There may be some components > trying to use this already but I would consider this more of a feature > as it has probably never worked before. Fair enough. I'll keep that for 5.15 then. Thanks, M.
diff --git a/drivers/irqchip/irq-mtk-sysirq.c b/drivers/irqchip/irq-mtk-sysirq.c index 6ff98b87e5c0..586e52d5442b 100644 --- a/drivers/irqchip/irq-mtk-sysirq.c +++ b/drivers/irqchip/irq-mtk-sysirq.c @@ -65,6 +65,7 @@ static struct irq_chip mtk_sysirq_chip = { .irq_set_type = mtk_sysirq_set_type, .irq_retrigger = irq_chip_retrigger_hierarchy, .irq_set_affinity = irq_chip_set_affinity_parent, + .flags = IRQCHIP_SKIP_SET_WAKE, }; static int mtk_sysirq_domain_translate(struct irq_domain *d,
mtk-sysirq doesn't require specific logic to work with wakeup IRQs. To allow registered IRQs to be used as a wakeup-source, add the flag IRQCHIP_SKIP_SET_WAKE. Signed-off-by: Markus Schneider-Pargmann <msp@baylibre.com> --- drivers/irqchip/irq-mtk-sysirq.c | 1 + 1 file changed, 1 insertion(+)