2011-09-16 13:33:53

by Mohammed Shafi Shajakhan

[permalink] [raw]
Subject: [PATCH] rfkill: properly assign a boolean type

From: Mohammed Shafi Shajakhan <[email protected]>

Signed-off-by: Mohammed Shafi Shajakhan <[email protected]>
---
net/rfkill/core.c | 2 +-
1 files changed, 1 insertions(+), 1 deletions(-)

diff --git a/net/rfkill/core.c b/net/rfkill/core.c
index be90640..5be1957 100644
--- a/net/rfkill/core.c
+++ b/net/rfkill/core.c
@@ -235,7 +235,7 @@ static bool __rfkill_set_hw_state(struct rfkill *rfkill,
else
rfkill->state &= ~RFKILL_BLOCK_HW;
*change = prev != blocked;
- any = rfkill->state & RFKILL_BLOCK_ANY;
+ any = !!(rfkill->state & RFKILL_BLOCK_ANY);
spin_unlock_irqrestore(&rfkill->lock, flags);

rfkill_led_trigger_event(rfkill);
--
1.7.0.4



2011-09-16 14:32:26

by Mohammed Shafi Shajakhan

[permalink] [raw]
Subject: Re: [PATCH] rfkill: properly assign a boolean type

On Friday 16 September 2011 07:38 PM, Johannes Berg wrote:
> On Fri, 2011-09-16 at 19:32 +0530, mohammed wrote:
>> On Friday 16 September 2011 07:31 PM, Johannes Berg wrote:
>>> On Fri, 2011-09-16 at 19:23 +0530, mohammed wrote:
>>>
>>>>>> @@ -235,7 +235,7 @@ static bool __rfkill_set_hw_state(struct rfkill *rfkill,
>>>>>> else
>>>>>> rfkill->state&= ~RFKILL_BLOCK_HW;
>>>>>> *change = prev != blocked;
>>>>>> - any = rfkill->state& RFKILL_BLOCK_ANY;
>>>>>> + any = !!(rfkill->state& RFKILL_BLOCK_ANY);
>>>>>
>>>>> I believe this is not necessary since "any" is a "bool" and as such
>>>>> should cast correctly to 0/1.
>>>
>>>> I agree the older one works perfectly fine. I stumbled upon this when i
>>>> was trying to understand rfkill. but will not this change make it look
>>>> better ? if it looks like a too trivial please drop it. Thanks!
>>>
>>> Sure, whatever, I don't care; we can change it, but I think it'll
>>> generate exactly the same code :)
>>
>> oh ok, thanks.
>
> Sure. I just wanted to clarify that it was to make the coder nicer, not
> to fix a bug or so.

actually, I was trying to figure out why software rfkill overrides
hardware rfkill when the card is not inbuilt with the laptop. have not
tested with all the cards.
i have tested with ath9k and another one, where i can simply unblock
with software unblock command even though the card is hardblocked. this
does not seems to be the case when the card is inbuilt(i tested with in
an inbuilt iwlagn in lenovo, it works properly). not sure its driver
bug, i need to verify ath9k card that comes inbuilt with the laptop. if
you have any thoughts please let me know. i dont have the complete
understanding of rfkill. thought of doing more ground work before
asking. thanks.


>
> johannes
>


2011-09-16 13:53:39

by Mohammed Shafi Shajakhan

[permalink] [raw]
Subject: Re: [PATCH] rfkill: properly assign a boolean type

On Friday 16 September 2011 07:07 PM, Johannes Berg wrote:
> On Fri, 2011-09-16 at 19:03 +0530, Mohammed Shafi Shajakhan wrote:
>> From: Mohammed Shafi Shajakhan<[email protected]>
>>
>> Signed-off-by: Mohammed Shafi Shajakhan<[email protected]>
>> ---
>> net/rfkill/core.c | 2 +-
>> 1 files changed, 1 insertions(+), 1 deletions(-)
>>
>> diff --git a/net/rfkill/core.c b/net/rfkill/core.c
>> index be90640..5be1957 100644
>> --- a/net/rfkill/core.c
>> +++ b/net/rfkill/core.c
>> @@ -235,7 +235,7 @@ static bool __rfkill_set_hw_state(struct rfkill *rfkill,
>> else
>> rfkill->state&= ~RFKILL_BLOCK_HW;
>> *change = prev != blocked;
>> - any = rfkill->state& RFKILL_BLOCK_ANY;
>> + any = !!(rfkill->state& RFKILL_BLOCK_ANY);
>
> I believe this is not necessary since "any" is a "bool" and as such
> should cast correctly to 0/1.

Hi johannes,

I agree the older one works perfectly fine. I stumbled upon this when i
was trying to understand rfkill. but will not this change make it look
better ? if it looks like a too trivial please drop it. Thanks!


>
> johannes
>


2011-09-16 14:11:43

by Mohammed Shafi Shajakhan

[permalink] [raw]
Subject: Re: [PATCH] rfkill: properly assign a boolean type

On Friday 16 September 2011 07:38 PM, Johannes Berg wrote:
> On Fri, 2011-09-16 at 19:32 +0530, mohammed wrote:
>> On Friday 16 September 2011 07:31 PM, Johannes Berg wrote:
>>> On Fri, 2011-09-16 at 19:23 +0530, mohammed wrote:
>>>
>>>>>> @@ -235,7 +235,7 @@ static bool __rfkill_set_hw_state(struct rfkill *rfkill,
>>>>>> else
>>>>>> rfkill->state&= ~RFKILL_BLOCK_HW;
>>>>>> *change = prev != blocked;
>>>>>> - any = rfkill->state& RFKILL_BLOCK_ANY;
>>>>>> + any = !!(rfkill->state& RFKILL_BLOCK_ANY);
>>>>>
>>>>> I believe this is not necessary since "any" is a "bool" and as such
>>>>> should cast correctly to 0/1.
>>>
>>>> I agree the older one works perfectly fine. I stumbled upon this when i
>>>> was trying to understand rfkill. but will not this change make it look
>>>> better ? if it looks like a too trivial please drop it. Thanks!
>>>
>>> Sure, whatever, I don't care; we can change it, but I think it'll
>>> generate exactly the same code :)
>>
>> oh ok, thanks.
>
> Sure. I just wanted to clarify that it was to make the coder nicer, not
> to fix a bug or so.

no it does not fixes any bug :) thanks a lot for reviewing.

>
> johannes
>


2011-09-16 13:37:32

by Johannes Berg

[permalink] [raw]
Subject: Re: [PATCH] rfkill: properly assign a boolean type

On Fri, 2011-09-16 at 19:03 +0530, Mohammed Shafi Shajakhan wrote:
> From: Mohammed Shafi Shajakhan <[email protected]>
>
> Signed-off-by: Mohammed Shafi Shajakhan <[email protected]>
> ---
> net/rfkill/core.c | 2 +-
> 1 files changed, 1 insertions(+), 1 deletions(-)
>
> diff --git a/net/rfkill/core.c b/net/rfkill/core.c
> index be90640..5be1957 100644
> --- a/net/rfkill/core.c
> +++ b/net/rfkill/core.c
> @@ -235,7 +235,7 @@ static bool __rfkill_set_hw_state(struct rfkill *rfkill,
> else
> rfkill->state &= ~RFKILL_BLOCK_HW;
> *change = prev != blocked;
> - any = rfkill->state & RFKILL_BLOCK_ANY;
> + any = !!(rfkill->state & RFKILL_BLOCK_ANY);

I believe this is not necessary since "any" is a "bool" and as such
should cast correctly to 0/1.

johannes


2011-09-16 14:08:37

by Johannes Berg

[permalink] [raw]
Subject: Re: [PATCH] rfkill: properly assign a boolean type

On Fri, 2011-09-16 at 19:32 +0530, mohammed wrote:
> On Friday 16 September 2011 07:31 PM, Johannes Berg wrote:
> > On Fri, 2011-09-16 at 19:23 +0530, mohammed wrote:
> >
> >>>> @@ -235,7 +235,7 @@ static bool __rfkill_set_hw_state(struct rfkill *rfkill,
> >>>> else
> >>>> rfkill->state&= ~RFKILL_BLOCK_HW;
> >>>> *change = prev != blocked;
> >>>> - any = rfkill->state& RFKILL_BLOCK_ANY;
> >>>> + any = !!(rfkill->state& RFKILL_BLOCK_ANY);
> >>>
> >>> I believe this is not necessary since "any" is a "bool" and as such
> >>> should cast correctly to 0/1.
> >
> >> I agree the older one works perfectly fine. I stumbled upon this when i
> >> was trying to understand rfkill. but will not this change make it look
> >> better ? if it looks like a too trivial please drop it. Thanks!
> >
> > Sure, whatever, I don't care; we can change it, but I think it'll
> > generate exactly the same code :)
>
> oh ok, thanks.

Sure. I just wanted to clarify that it was to make the coder nicer, not
to fix a bug or so.

johannes


2011-09-16 14:01:28

by Johannes Berg

[permalink] [raw]
Subject: Re: [PATCH] rfkill: properly assign a boolean type

On Fri, 2011-09-16 at 19:23 +0530, mohammed wrote:

> >> @@ -235,7 +235,7 @@ static bool __rfkill_set_hw_state(struct rfkill *rfkill,
> >> else
> >> rfkill->state&= ~RFKILL_BLOCK_HW;
> >> *change = prev != blocked;
> >> - any = rfkill->state& RFKILL_BLOCK_ANY;
> >> + any = !!(rfkill->state& RFKILL_BLOCK_ANY);
> >
> > I believe this is not necessary since "any" is a "bool" and as such
> > should cast correctly to 0/1.

> I agree the older one works perfectly fine. I stumbled upon this when i
> was trying to understand rfkill. but will not this change make it look
> better ? if it looks like a too trivial please drop it. Thanks!

Sure, whatever, I don't care; we can change it, but I think it'll
generate exactly the same code :)

johannes


2011-09-16 14:02:46

by Mohammed Shafi Shajakhan

[permalink] [raw]
Subject: Re: [PATCH] rfkill: properly assign a boolean type

On Friday 16 September 2011 07:31 PM, Johannes Berg wrote:
> On Fri, 2011-09-16 at 19:23 +0530, mohammed wrote:
>
>>>> @@ -235,7 +235,7 @@ static bool __rfkill_set_hw_state(struct rfkill *rfkill,
>>>> else
>>>> rfkill->state&= ~RFKILL_BLOCK_HW;
>>>> *change = prev != blocked;
>>>> - any = rfkill->state& RFKILL_BLOCK_ANY;
>>>> + any = !!(rfkill->state& RFKILL_BLOCK_ANY);
>>>
>>> I believe this is not necessary since "any" is a "bool" and as such
>>> should cast correctly to 0/1.
>
>> I agree the older one works perfectly fine. I stumbled upon this when i
>> was trying to understand rfkill. but will not this change make it look
>> better ? if it looks like a too trivial please drop it. Thanks!
>
> Sure, whatever, I don't care; we can change it, but I think it'll
> generate exactly the same code :)

oh ok, thanks.

>
> johannes
>