Message ID | 20161214135209.16369-1-hdegoede@redhat.com |
---|---|
State | Not Applicable |
Headers | show |
On Wed, Dec 14, 2016 at 9:52 PM, Hans de Goede <hdegoede@redhat.com> wrote: > The interrupt line of the entire family of axp2xx pmics is active-low, > for devicetree enumerated irqs, this is dealt with in the devicetree. > > ACPI irq resources have a flag field for this too, I tried using this > on my CUBE iwork8 Air tablet, but it does not contain the right data. > > The dstd shows the irq listed as either ActiveLow or ActiveHigh, > depending on the OSID variable, which seems to be set by the > "OS IMAGE ID" in the BIOS/EFI setup screen. > > Since the acpi-resource info is no good, simply pass in IRQF_TRIGGER_LOW > on the axp288. > > Together with the other axp288 fixes in this series, this fixes the axp288 > irq contineously triggering. > > Signed-off-by: Hans de Goede <hdegoede@redhat.com> This patch looks good to me. However, I do not have any AXP288 hardware to test it on, nor am I familiar with the ACPI stuff. Acked-by: Chen-Yu Tsai <wens@csie.org> P.S. I don't think we're handling IRQ trigger types at all. The hardware default for the NMI interrupt in the Allwinner chips just happens to be active low. -- To unsubscribe from this list: send the line "unsubscribe linux-i2c" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
On Wed, 14 Dec 2016, Hans de Goede wrote: > The interrupt line of the entire family of axp2xx pmics is active-low, > for devicetree enumerated irqs, this is dealt with in the devicetree. > > ACPI irq resources have a flag field for this too, I tried using this > on my CUBE iwork8 Air tablet, but it does not contain the right data. > > The dstd shows the irq listed as either ActiveLow or ActiveHigh, > depending on the OSID variable, which seems to be set by the > "OS IMAGE ID" in the BIOS/EFI setup screen. > > Since the acpi-resource info is no good, simply pass in IRQF_TRIGGER_LOW > on the axp288. > > Together with the other axp288 fixes in this series, this fixes the axp288 > irq contineously triggering. > > Signed-off-by: Hans de Goede <hdegoede@redhat.com> > --- > drivers/mfd/axp20x.c | 6 +++--- > include/linux/mfd/axp20x.h | 1 + > 2 files changed, 4 insertions(+), 3 deletions(-) Applied, thanks. > diff --git a/drivers/mfd/axp20x.c b/drivers/mfd/axp20x.c > index ba130be..3d76941 100644 > --- a/drivers/mfd/axp20x.c > +++ b/drivers/mfd/axp20x.c > @@ -800,6 +800,7 @@ int axp20x_match_device(struct axp20x_dev *axp20x) > axp20x->nr_cells = ARRAY_SIZE(axp288_cells); > axp20x->regmap_cfg = &axp288_regmap_config; > axp20x->regmap_irq_chip = &axp288_regmap_irq_chip; > + axp20x->irq_flags = IRQF_TRIGGER_LOW; > break; > case AXP806_ID: > axp20x->nr_cells = ARRAY_SIZE(axp806_cells); > @@ -829,9 +830,8 @@ int axp20x_device_probe(struct axp20x_dev *axp20x) > int ret; > > ret = regmap_add_irq_chip(axp20x->regmap, axp20x->irq, > - IRQF_ONESHOT | IRQF_SHARED, -1, > - axp20x->regmap_irq_chip, > - &axp20x->regmap_irqc); > + IRQF_ONESHOT | IRQF_SHARED | axp20x->irq_flags, > + -1, axp20x->regmap_irq_chip, &axp20x->regmap_irqc); > if (ret) { > dev_err(axp20x->dev, "failed to add irq chip: %d\n", ret); > return ret; > diff --git a/include/linux/mfd/axp20x.h b/include/linux/mfd/axp20x.h > index fec597f..199cce3 100644 > --- a/include/linux/mfd/axp20x.h > +++ b/include/linux/mfd/axp20x.h > @@ -517,6 +517,7 @@ enum axp809_irqs { > struct axp20x_dev { > struct device *dev; > int irq; > + unsigned long irq_flags; > struct regmap *regmap; > struct regmap_irq_chip_data *regmap_irqc; > long variant;
diff --git a/drivers/mfd/axp20x.c b/drivers/mfd/axp20x.c index ba130be..3d76941 100644 --- a/drivers/mfd/axp20x.c +++ b/drivers/mfd/axp20x.c @@ -800,6 +800,7 @@ int axp20x_match_device(struct axp20x_dev *axp20x) axp20x->nr_cells = ARRAY_SIZE(axp288_cells); axp20x->regmap_cfg = &axp288_regmap_config; axp20x->regmap_irq_chip = &axp288_regmap_irq_chip; + axp20x->irq_flags = IRQF_TRIGGER_LOW; break; case AXP806_ID: axp20x->nr_cells = ARRAY_SIZE(axp806_cells); @@ -829,9 +830,8 @@ int axp20x_device_probe(struct axp20x_dev *axp20x) int ret; ret = regmap_add_irq_chip(axp20x->regmap, axp20x->irq, - IRQF_ONESHOT | IRQF_SHARED, -1, - axp20x->regmap_irq_chip, - &axp20x->regmap_irqc); + IRQF_ONESHOT | IRQF_SHARED | axp20x->irq_flags, + -1, axp20x->regmap_irq_chip, &axp20x->regmap_irqc); if (ret) { dev_err(axp20x->dev, "failed to add irq chip: %d\n", ret); return ret; diff --git a/include/linux/mfd/axp20x.h b/include/linux/mfd/axp20x.h index fec597f..199cce3 100644 --- a/include/linux/mfd/axp20x.h +++ b/include/linux/mfd/axp20x.h @@ -517,6 +517,7 @@ enum axp809_irqs { struct axp20x_dev { struct device *dev; int irq; + unsigned long irq_flags; struct regmap *regmap; struct regmap_irq_chip_data *regmap_irqc; long variant;
The interrupt line of the entire family of axp2xx pmics is active-low, for devicetree enumerated irqs, this is dealt with in the devicetree. ACPI irq resources have a flag field for this too, I tried using this on my CUBE iwork8 Air tablet, but it does not contain the right data. The dstd shows the irq listed as either ActiveLow or ActiveHigh, depending on the OSID variable, which seems to be set by the "OS IMAGE ID" in the BIOS/EFI setup screen. Since the acpi-resource info is no good, simply pass in IRQF_TRIGGER_LOW on the axp288. Together with the other axp288 fixes in this series, this fixes the axp288 irq contineously triggering. Signed-off-by: Hans de Goede <hdegoede@redhat.com> --- drivers/mfd/axp20x.c | 6 +++--- include/linux/mfd/axp20x.h | 1 + 2 files changed, 4 insertions(+), 3 deletions(-)