[PATCH] Fix wrong 'no interrupt' handling in of_i2c

[PATCH] Fix wrong 'no interrupt' handling in of_i2c

[Wolfram Sang]

If an I2C device node does not specify an interrupt, the .irq member of the
board_info struct was set to -1. This caused crashes on following
irq_dispose_mappings. Leave it NO_IRQ as returned from irq_of_parse_and_map.
(Suggesting -1 as 'i2c-no-irq' used to be a bug in linux/i2c.h.)

Signed-off-by: Wolfram Sang <w.sang@pengutronix.de>
---

 drivers/of/of_i2c.c |    2 --
 1 files changed, 0 insertions(+), 2 deletions(-)

diff --git a/drivers/of/of_i2c.c b/drivers/of/of_i2c.c
index b2ccdcb..500cb1e 100644
--- a/drivers/of/of_i2c.c
+++ b/drivers/of/of_i2c.c
@@ -90,8 +90,6 @@ void of_register_i2c_devices(struct i2c_adapter *adap,
 		}
 
 		info.irq = irq_of_parse_and_map(node, 0);
-		if (info.irq == NO_IRQ)
-			info.irq = -1;
 
 		if (of_find_i2c_driver(node, &info) < 0) {
 			irq_dispose_mapping(info.irq);

Re: [PATCH] Fix wrong 'no interrupt' handling in of_i2c

[Grant Likely]

On Sat, Jun 28, 2008 at 12:31 PM, Wolfram Sang <w.sang@pengutronix.de> wrote:
> If an I2C device node does not specify an interrupt, the .irq member of the
> board_info struct was set to -1. This caused crashes on following
> irq_dispose_mappings. Leave it NO_IRQ as returned from irq_of_parse_and_map.
> (Suggesting -1 as 'i2c-no-irq' used to be a bug in linux/i2c.h.)
>
> Signed-off-by: Wolfram Sang <w.sang@pengutronix.de>

Acked-by: Grant Likely <grant.likely@secretlab.ca>

-- 
Grant Likely, B.Sc., P.Eng.
Secret Lab Technologies Ltd.

Re: [PATCH] Fix wrong 'no interrupt' handling in of_i2c

[Sean MacLennan]

On Sat, 28 Jun 2008 13:16:16 -0600
"Grant Likely" <grant.likely@secretlab.ca> wrote:

> On Sat, Jun 28, 2008 at 12:31 PM, Wolfram Sang
> <w.sang@pengutronix.de> wrote:
> > If an I2C device node does not specify an interrupt, the .irq
> > member of the board_info struct was set to -1. This caused crashes
> > on following irq_dispose_mappings. Leave it NO_IRQ as returned from
> > irq_of_parse_and_map. (Suggesting -1 as 'i2c-no-irq' used to be a
> > bug in linux/i2c.h.)
> >
> > Signed-off-by: Wolfram Sang <w.sang@pengutronix.de>
> 
> Acked-by: Grant Likely <grant.likely@secretlab.ca>
> 

I just hit this while trying to get the at24 driver working. So:

Acked-by: Sean MacLennan <smaclennan@pikatech.com>

Cheers,
   Sean

Re: [PATCH] Fix wrong 'no interrupt' handling in of_i2c

[Grant Likely]

On Wed, Jul 16, 2008 at 4:15 PM, Sean MacLennan <smaclennan@pikatech.com> wrote:
> On Sat, 28 Jun 2008 13:16:16 -0600
> "Grant Likely" <grant.likely@secretlab.ca> wrote:
>
>> On Sat, Jun 28, 2008 at 12:31 PM, Wolfram Sang
>> <w.sang@pengutronix.de> wrote:
>> > If an I2C device node does not specify an interrupt, the .irq
>> > member of the board_info struct was set to -1. This caused crashes
>> > on following irq_dispose_mappings. Leave it NO_IRQ as returned from
>> > irq_of_parse_and_map. (Suggesting -1 as 'i2c-no-irq' used to be a
>> > bug in linux/i2c.h.)
>> >
>> > Signed-off-by: Wolfram Sang <w.sang@pengutronix.de>
>>
>> Acked-by: Grant Likely <grant.likely@secretlab.ca>
>>
>
> I just hit this while trying to get the at24 driver working. So:
>
> Acked-by: Sean MacLennan <smaclennan@pikatech.com>

Okay, I'll pick this one up.

g.

-- 
Grant Likely, B.Sc., P.Eng.
Secret Lab Technologies Ltd.