linux-kernel - Re: [PATCH] hwmon: (lm75) Add a basic interrupt handler

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Date:   Sun, 16 Jul 2017 21:51:24 +0000
From:   Tony O'Brien <Tony.OBrien@...iedtelesis.co.nz>
To:     Guenter Roeck <linux@...ck-us.net>,
        "jdelvare@...e.com" <jdelvare@...e.com>,
        "linux-hwmon@...r.kernel.org" <linux-hwmon@...r.kernel.org>,
        "linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>
Subject: Re: [PATCH] hwmon: (lm75) Add a basic interrupt handler

Hi Guenter -

The interrupt line the LM75 sits on is shared by other devices and therefore cannot be disabled.  It was added in hardware as a possible feature can cannot be disconnected.

Cheers,

Tony O'Brien
Senior Software/Hardware Engineer
Allied Telesis Labs
27 Nazareth Avenue
Christchurch 8024
New Zealand
Ph: +64-3-339 3000
DDI: +64-3-339 9210
Web: http://AlliedTelesis.com



________________________________________
From: Guenter Roeck <linux@...ck-us.net>
Sent: Sunday, 16 July 2017 12:42 p.m.
To: Tony O'Brien; jdelvare@...e.com; linux-hwmon@...r.kernel.org; linux-kernel@...r.kernel.org
Subject: Re: [PATCH] hwmon: (lm75) Add a basic interrupt handler

On 07/11/2017 07:52 PM, Tony O'Brien wrote:
> The LM75 interrupt cannot be masked in the device so an over-temperature
> event can cause an interrupt that cannot be cleared.
> Add an interrupt handler if an interrupt node exists in the DTS for an
> LM75.  The handler simply reads a device register to clear the interrupt
> and returns.
>

Is this an actual problem ? If no interrupt handler is registered, the interrupt
should be disabled by the interrupt controller, and it should not matter if
the interrupt line from the chip is active or not.

Thanks,
Guenter

> Signed-off-by: Tony O'Brien <tony.obrien@...iedtelesis.co.nz>
> Reviewed-by: Chris Packham <chris.packham@...iedtelesis.co.nz>
> ---
>   drivers/hwmon/lm75.c | 28 ++++++++++++++++++++++++++++
>   drivers/hwmon/lm75.h |  1 +
>   2 files changed, 29 insertions(+)
>
> diff --git a/drivers/hwmon/lm75.c b/drivers/hwmon/lm75.c
> index 005ffb5ffa92..08a833e960aa 100644
> --- a/drivers/hwmon/lm75.c
> +++ b/drivers/hwmon/lm75.c
> @@ -22,6 +22,7 @@
>   #include <linux/init.h>
>   #include <linux/slab.h>
>   #include <linux/jiffies.h>
> +#include <linux/interrupt.h>
>   #include <linux/i2c.h>
>   #include <linux/hwmon.h>
>   #include <linux/hwmon-sysfs.h>
> @@ -265,6 +266,17 @@ static void lm75_remove(void *data)
>       i2c_smbus_write_byte_data(client, LM75_REG_CONF, lm75->orig_conf);
>   }
>
> +static irqreturn_t lm75_process_interrupt(int irq, void *data)
> +{
> +     struct i2c_client *client = (struct i2c_client *)data;
> +     int val;
> +
> +     /* Do a read to clear the interrupt */
> +     val = i2c_smbus_read_byte_data(client, LM75_REG_CONF);
> +
> +     return IRQ_HANDLED;
> +}
> +
>   static int
>   lm75_probe(struct i2c_client *client, const struct i2c_device_id *id)
>   {
> @@ -275,6 +287,7 @@ lm75_probe(struct i2c_client *client, const struct i2c_device_id *id)
>       u8 set_mask, clr_mask;
>       int new;
>       enum lm75_type kind;
> +     int ret;
>
>       if (client->dev.of_node)
>               kind = (enum lm75_type)of_device_get_match_data(&client->dev);
> @@ -375,6 +388,21 @@ lm75_probe(struct i2c_client *client, const struct i2c_device_id *id)
>               break;
>       }
>
> +     if (client->irq) {
> +             /* If requesting an interrupt then set to interrupt mode */
> +             set_mask |= LM75_MODE_INTERRUPT;
> +             ret = devm_request_threaded_irq(dev, client->irq, NULL,
> +                                             lm75_process_interrupt,
> +                                             IRQF_TRIGGER_LOW | IRQF_ONESHOT |
> +                                                     IRQF_SHARED,
> +                                             dev_name(dev),
> +                                             client);
> +             if (ret) {
> +                     dev_err(dev, "Error requesting irq\n");
> +                     return ret;
> +             }
> +     }
> +
>       /* configure as specified */
>       status = i2c_smbus_read_byte_data(client, LM75_REG_CONF);
>       if (status < 0) {
> diff --git a/drivers/hwmon/lm75.h b/drivers/hwmon/lm75.h
> index 5cde94e56f17..82a71d870331 100644
> --- a/drivers/hwmon/lm75.h
> +++ b/drivers/hwmon/lm75.h
> @@ -31,6 +31,7 @@
>   #define LM75_TEMP_MIN (-55000)
>   #define LM75_TEMP_MAX 125000
>   #define LM75_SHUTDOWN 0x01
> +#define LM75_MODE_INTERRUPT 0x02
>
>   /* TEMP: 0.001C/bit (-55C to +125C)
>      REG: (0.5C/bit, two's complement) << 7 */
>