Message ID | 20201105080014.45410-5-hdegoede@redhat.com |
---|---|
State | New |
Headers | show |
Series | [RFC,1/4] i2c: core: Allow i2c_board_info to specify that the core should not try to find an IRQ | expand |
On Thu, Nov 5, 2020 at 10:00 AM Hans de Goede <hdegoede@redhat.com> wrote: > > Most I2C-drivers treat IRQs as optional and in some cases ACPI > devices managed by i2c-multi-instantiate.c may have a GpioInt resource > specified on some systems, while it is not there on others. GpioInt() > Add a new IRQ_RESOURCE_GPIO_OPTIONAL IRQ type, which still tries to get > a GpioInt IRQ, but does not consider it a fatal error when this fails. GpioInt() ... > -#define IRQ_RESOURCE_TYPE GENMASK(1, 0) > -#define IRQ_RESOURCE_NONE 0 > -#define IRQ_RESOURCE_GPIO 1 > -#define IRQ_RESOURCE_APIC 2 > +#define IRQ_RESOURCE_TYPE GENMASK(1, 0) > +#define IRQ_RESOURCE_NONE 0 > +#define IRQ_RESOURCE_GPIO 1 > +#define IRQ_RESOURCE_GPIO_OPTIONAL 2 > +#define IRQ_RESOURCE_APIC 3 I think flag is cleaner: #define IRQ_RESOURCE_OPTIONAL BIT(2) ... > case IRQ_RESOURCE_GPIO: > ret = acpi_dev_gpio_irq_get(adev, inst_data[i].irq_idx); > - if (ret < 0) { > + if (ret < 0 && (!irq_optional || ret != -ENOENT)) { ret < 0 && !((inst_data[i].flags & IRQ_RESOURCE_OPTIONAL) && ret == -ENOENT) > dev_err(dev, "Error requesting irq at index %d: %d\n", > inst_data[i].irq_idx, ret); > goto error;
diff --git a/drivers/platform/x86/i2c-multi-instantiate.c b/drivers/platform/x86/i2c-multi-instantiate.c index cbccfcbed44c..55c6d6e8d576 100644 --- a/drivers/platform/x86/i2c-multi-instantiate.c +++ b/drivers/platform/x86/i2c-multi-instantiate.c @@ -15,10 +15,11 @@ #include <linux/platform_device.h> #include <linux/types.h> -#define IRQ_RESOURCE_TYPE GENMASK(1, 0) -#define IRQ_RESOURCE_NONE 0 -#define IRQ_RESOURCE_GPIO 1 -#define IRQ_RESOURCE_APIC 2 +#define IRQ_RESOURCE_TYPE GENMASK(1, 0) +#define IRQ_RESOURCE_NONE 0 +#define IRQ_RESOURCE_GPIO 1 +#define IRQ_RESOURCE_GPIO_OPTIONAL 2 +#define IRQ_RESOURCE_APIC 3 struct i2c_inst_data { const char *type; @@ -64,6 +65,7 @@ static int i2c_multi_inst_probe(struct platform_device *pdev) struct i2c_board_info board_info = {}; struct device *dev = &pdev->dev; struct acpi_device *adev; + bool irq_optional; char name[32]; int i, ret; @@ -94,10 +96,14 @@ static int i2c_multi_inst_probe(struct platform_device *pdev) inst_data[i].type, i); board_info.dev_name = name; board_info.fwnode = dev->fwnode; + irq_optional = false; switch (inst_data[i].flags & IRQ_RESOURCE_TYPE) { + case IRQ_RESOURCE_GPIO_OPTIONAL: + irq_optional = true; + fallthrough; case IRQ_RESOURCE_GPIO: ret = acpi_dev_gpio_irq_get(adev, inst_data[i].irq_idx); - if (ret < 0) { + if (ret < 0 && (!irq_optional || ret != -ENOENT)) { dev_err(dev, "Error requesting irq at index %d: %d\n", inst_data[i].irq_idx, ret); goto error;
Most I2C-drivers treat IRQs as optional and in some cases ACPI devices managed by i2c-multi-instantiate.c may have a GpioInt resource specified on some systems, while it is not there on others. Add a new IRQ_RESOURCE_GPIO_OPTIONAL IRQ type, which still tries to get a GpioInt IRQ, but does not consider it a fatal error when this fails. Signed-off-by: Hans de Goede <hdegoede@redhat.com> --- drivers/platform/x86/i2c-multi-instantiate.c | 16 +++++++++++----- 1 file changed, 11 insertions(+), 5 deletions(-)