Patchwork [2/3] hwmon: (lm90) add support to handle irq

login
register
mail settings
Submitter lightning314
Date July 4, 2013, 7:57 a.m.
Message ID <1372924670-16355-3-git-send-email-wni@nvidia.com>
Download mbox | patch
Permalink /patch/256814/
State Not Applicable, archived
Headers show

Comments

lightning314 - July 4, 2013, 7:57 a.m.
Add support to handle irq. When the temperature touch
the limit value, the driver can handle the interrupt.

Signed-off-by: Wei Ni <wni@nvidia.com>
---
 drivers/hwmon/lm90.c |   25 +++++++++++++++++++++++++
 1 file changed, 25 insertions(+)
Stephen Warren - July 5, 2013, 5:35 p.m.
On 07/04/2013 01:57 AM, Wei Ni wrote:
> Add support to handle irq. When the temperature touch
> the limit value, the driver can handle the interrupt.

> +	if (client->irq >= 0) {

0 isn't a valid IRQ, so you can write that as simply if (client->irq).
--
To unsubscribe from this list: send the line "unsubscribe linux-tegra" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Guenter Roeck - July 5, 2013, 7:47 p.m.
On Fri, Jul 05, 2013 at 11:35:05AM -0600, Stephen Warren wrote:
> On 07/04/2013 01:57 AM, Wei Ni wrote:
> > Add support to handle irq. When the temperature touch
> > the limit value, the driver can handle the interrupt.
> 
> > +	if (client->irq >= 0) {
> 
> 0 isn't a valid IRQ, so you can write that as simply if (client->irq).
> 
If I recall correctly, it is valid on some platforms.

Guenter
--
To unsubscribe from this list: send the line "unsubscribe linux-tegra" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Stephen Warren - July 8, 2013, 3:46 p.m.
On 07/05/2013 01:47 PM, Guenter Roeck wrote:
> On Fri, Jul 05, 2013 at 11:35:05AM -0600, Stephen Warren wrote:
>> On 07/04/2013 01:57 AM, Wei Ni wrote:
>>> Add support to handle irq. When the temperature touch
>>> the limit value, the driver can handle the interrupt.
>>
>>> +	if (client->irq >= 0) {
>>
>> 0 isn't a valid IRQ, so you can write that as simply if (client->irq).
>>
> If I recall correctly, it is valid on some platforms.

I thought ARM (just some ARM sub-architectures?) might have been the
last architecture, and even irrespective of that, we were trying not to
introduce any new code that relies on this strangeness, so it doesn't
propagate?
--
To unsubscribe from this list: send the line "unsubscribe linux-tegra" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Guenter Roeck - July 8, 2013, 8:06 p.m.
On Mon, Jul 08, 2013 at 09:46:41AM -0600, Stephen Warren wrote:
> On 07/05/2013 01:47 PM, Guenter Roeck wrote:
> > On Fri, Jul 05, 2013 at 11:35:05AM -0600, Stephen Warren wrote:
> >> On 07/04/2013 01:57 AM, Wei Ni wrote:
> >>> Add support to handle irq. When the temperature touch
> >>> the limit value, the driver can handle the interrupt.
> >>
> >>> +	if (client->irq >= 0) {
> >>
> >> 0 isn't a valid IRQ, so you can write that as simply if (client->irq).
> >>
> > If I recall correctly, it is valid on some platforms.
> 
> I thought ARM (just some ARM sub-architectures?) might have been the
> last architecture, and even irrespective of that, we were trying not to
> introduce any new code that relies on this strangeness, so it doesn't
> propagate?
> 
Sounds good to me. Another problem, though, may be that NO_IRQ is sometimes
defined as -1, sometimes as 0, sometimes as 0xffffffff, and sometimes as INT_MAX.
Which of course is another mess :(.

Guenter


--
To unsubscribe from this list: send the line "unsubscribe linux-tegra" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Thierry Reding - July 9, 2013, 6:15 a.m.
On Thu, Jul 04, 2013 at 03:57:49PM +0800, Wei Ni wrote:
> Add support to handle irq. When the temperature touch

Nit: This first sentence can be dropped. It merely repeats the subject.
And another nit: "irq" -> "IRQ"

> the limit value, the driver can handle the interrupt.
> 
> Signed-off-by: Wei Ni <wni@nvidia.com>
> ---
>  drivers/hwmon/lm90.c |   25 +++++++++++++++++++++++++
>  1 file changed, 25 insertions(+)
> 
> diff --git a/drivers/hwmon/lm90.c b/drivers/hwmon/lm90.c
> index 2cb7f8e..fce9dfa 100644
> --- a/drivers/hwmon/lm90.c
> +++ b/drivers/hwmon/lm90.c
> @@ -89,6 +89,7 @@
>  #include <linux/err.h>
>  #include <linux/mutex.h>
>  #include <linux/sysfs.h>
> +#include <linux/interrupt.h>
>  
>  /*
>   * Addresses to scan
> @@ -1423,6 +1424,18 @@ static void lm90_init_client(struct i2c_client *client)
>  		i2c_smbus_write_byte_data(client, LM90_REG_W_CONFIG1, config);
>  }
>  
> +static void lm90_alert(struct i2c_client *client, unsigned int flag);

Any reason why the implementation of lm90_alert() can't be moved here.
Granted, it'll make the diff larger but at least it'll allow us to get
rid of the forward declaration.

> +static irqreturn_t lm90_irq_thread(int irq, void *dev_id)
> +{
> +	struct lm90_data *data = dev_id;
> +	struct i2c_client *client = to_i2c_client(data->hwmon_dev->parent);
> +
> +	lm90_alert(client, 0);
> +
> +	return IRQ_HANDLED;
> +}

Isn't this potentially dangerous, since if the IRQ was shared, IRQ
processing will always stop after this handler. Shouldn't you check
whether the device actually triggered an interrupt (by reading the
interrupt status register)?

> +
>  static int lm90_probe(struct i2c_client *client,
>  		      const struct i2c_device_id *id)
>  {
> @@ -1499,6 +1512,18 @@ static int lm90_probe(struct i2c_client *client,
>  		goto exit_remove_files;
>  	}
>  
> +	if (client->irq >= 0) {
> +		dev_dbg(dev, "lm90 irq: %d\n", client->irq);

Nit: "irq" -> "IRQ"

Thierry
Jean Delvare - July 9, 2013, 6:49 a.m.
Hi Thierry,

Thanks for the review, it is greatly appreciated. A couple additions...

On Tue, 9 Jul 2013 08:15:15 +0200, Thierry Reding wrote:
> On Thu, Jul 04, 2013 at 03:57:49PM +0800, Wei Ni wrote:
> > Add support to handle irq. When the temperature touch
> 
> Nit: This first sentence can be dropped. It merely repeats the subject.

Note that I have no problem with that.

> > (...)
> > @@ -1423,6 +1424,18 @@ static void lm90_init_client(struct i2c_client *client)
> >  		i2c_smbus_write_byte_data(client, LM90_REG_W_CONFIG1, config);
> >  }
> >  
> > +static void lm90_alert(struct i2c_client *client, unsigned int flag);
> 
> Any reason why the implementation of lm90_alert() can't be moved here.
> Granted, it'll make the diff larger but at least it'll allow us to get
> rid of the forward declaration.

Yes, please avoid forward declarations as much as possible.
lightning314 - July 9, 2013, 7:54 a.m.
On 07/09/2013 04:06 AM, Guenter Roeck wrote:
> On Mon, Jul 08, 2013 at 09:46:41AM -0600, Stephen Warren wrote:
>> On 07/05/2013 01:47 PM, Guenter Roeck wrote:
>>> On Fri, Jul 05, 2013 at 11:35:05AM -0600, Stephen Warren wrote:
>>>> On 07/04/2013 01:57 AM, Wei Ni wrote:
>>>>> Add support to handle irq. When the temperature touch
>>>>> the limit value, the driver can handle the interrupt.
>>>>
>>>>> +	if (client->irq >= 0) {
>>>>
>>>> 0 isn't a valid IRQ, so you can write that as simply if (client->irq).
>>>>
>>> If I recall correctly, it is valid on some platforms.
>>
>> I thought ARM (just some ARM sub-architectures?) might have been the
>> last architecture, and even irrespective of that, we were trying not to
>> introduce any new code that relies on this strangeness, so it doesn't
>> propagate?
>>
> Sounds good to me. Another problem, though, may be that NO_IRQ is sometimes
> defined as -1, sometimes as 0, sometimes as 0xffffffff, and sometimes as INT_MAX.
> Which of course is another mess :(.
> 
> Guenter
> 

Ok, so I will use the "if (client->irq)".

Thanks.
Wei.

> 

--
To unsubscribe from this list: send the line "unsubscribe linux-tegra" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
lightning314 - July 9, 2013, 7:58 a.m.
On 07/09/2013 02:15 PM, Thierry Reding wrote:
> * PGP Signed by an unknown key
> 
> On Thu, Jul 04, 2013 at 03:57:49PM +0800, Wei Ni wrote:
>> Add support to handle irq. When the temperature touch
> 
> Nit: This first sentence can be dropped. It merely repeats the subject.
> And another nit: "irq" -> "IRQ"

Ok, I will update in my next version.

> 
>> the limit value, the driver can handle the interrupt.
>>
>> Signed-off-by: Wei Ni <wni@nvidia.com>
>> ---
>>  drivers/hwmon/lm90.c |   25 +++++++++++++++++++++++++
>>  1 file changed, 25 insertions(+)
>>
>> diff --git a/drivers/hwmon/lm90.c b/drivers/hwmon/lm90.c
>> index 2cb7f8e..fce9dfa 100644
>> --- a/drivers/hwmon/lm90.c
>> +++ b/drivers/hwmon/lm90.c
>> @@ -89,6 +89,7 @@
>>  #include <linux/err.h>
>>  #include <linux/mutex.h>
>>  #include <linux/sysfs.h>
>> +#include <linux/interrupt.h>
>>  
>>  /*
>>   * Addresses to scan
>> @@ -1423,6 +1424,18 @@ static void lm90_init_client(struct i2c_client *client)
>>  		i2c_smbus_write_byte_data(client, LM90_REG_W_CONFIG1, config);
>>  }
>>  
>> +static void lm90_alert(struct i2c_client *client, unsigned int flag);
> 
> Any reason why the implementation of lm90_alert() can't be moved here.
> Granted, it'll make the diff larger but at least it'll allow us to get
> rid of the forward declaration.

Ok, you are right, I will move it.

> 
>> +static irqreturn_t lm90_irq_thread(int irq, void *dev_id)
>> +{
>> +	struct lm90_data *data = dev_id;
>> +	struct i2c_client *client = to_i2c_client(data->hwmon_dev->parent);
>> +
>> +	lm90_alert(client, 0);
>> +
>> +	return IRQ_HANDLED;
>> +}
> 
> Isn't this potentially dangerous, since if the IRQ was shared, IRQ
> processing will always stop after this handler. Shouldn't you check
> whether the device actually triggered an interrupt (by reading the
> interrupt status register)?

Yes, it has potential dangerous, I didn't consider it carefully.
I will check the interrupt status.

> 
>> +
>>  static int lm90_probe(struct i2c_client *client,
>>  		      const struct i2c_device_id *id)
>>  {
>> @@ -1499,6 +1512,18 @@ static int lm90_probe(struct i2c_client *client,
>>  		goto exit_remove_files;
>>  	}
>>  
>> +	if (client->irq >= 0) {
>> +		dev_dbg(dev, "lm90 irq: %d\n", client->irq);
> 
> Nit: "irq" -> "IRQ"

Ok, got it.

> 
> Thierry
> 
> * Unknown Key
> * 0x7F3EB3A1
> 

--
To unsubscribe from this list: send the line "unsubscribe linux-tegra" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Patch

diff --git a/drivers/hwmon/lm90.c b/drivers/hwmon/lm90.c
index 2cb7f8e..fce9dfa 100644
--- a/drivers/hwmon/lm90.c
+++ b/drivers/hwmon/lm90.c
@@ -89,6 +89,7 @@ 
 #include <linux/err.h>
 #include <linux/mutex.h>
 #include <linux/sysfs.h>
+#include <linux/interrupt.h>
 
 /*
  * Addresses to scan
@@ -1423,6 +1424,18 @@  static void lm90_init_client(struct i2c_client *client)
 		i2c_smbus_write_byte_data(client, LM90_REG_W_CONFIG1, config);
 }
 
+static void lm90_alert(struct i2c_client *client, unsigned int flag);
+
+static irqreturn_t lm90_irq_thread(int irq, void *dev_id)
+{
+	struct lm90_data *data = dev_id;
+	struct i2c_client *client = to_i2c_client(data->hwmon_dev->parent);
+
+	lm90_alert(client, 0);
+
+	return IRQ_HANDLED;
+}
+
 static int lm90_probe(struct i2c_client *client,
 		      const struct i2c_device_id *id)
 {
@@ -1499,6 +1512,18 @@  static int lm90_probe(struct i2c_client *client,
 		goto exit_remove_files;
 	}
 
+	if (client->irq >= 0) {
+		dev_dbg(dev, "lm90 irq: %d\n", client->irq);
+		err = devm_request_threaded_irq(dev, client->irq,
+						NULL, lm90_irq_thread,
+						IRQF_TRIGGER_LOW | IRQF_ONESHOT,
+						"lm90", data);
+		if (err < 0) {
+			dev_err(dev, "cannot request interrupt\n");
+			goto exit_remove_files;
+		}
+	}
+
 	return 0;
 
 exit_remove_files: