[v4] net: mac802154: comparision issue of type cast, finding by EXTRA_CFLAGS=-W

Need change ~(u8)0 to 0xff, or in mac802154_wpan_xmit(), "u8 chan"
will never == MAC802154_CHAN_NONE.

The related warning:
  net/mac802154/wpan.c:325:2: warning: comparison is always false
  due to limited range of data type [-Wtype-limits]

The related unit test:
  cc -Wall -O2 -o test test.c
  ./test
  will output 'different'

  (after use '0xff' instead of '~(u8)0', it will output 'equal')

-----------------------test code begin--------------------------------

#include <stdio.h>

typedef unsigned char u8;

int main()
{
	u8 uc = 0xff;

	if (uc == ~(u8)0)
		printf("\nequal\n");
	else
		printf("\ndifferent\n");

	return 0;
}

-----------------------test code end----------------------------------


Signed-off-by: Chen Gang <gang.chen@asianux.com>
---
 net/mac802154/mac802154.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

On 04/25/2013 11:40 AM, Chen Gang wrote:
> Need change ~(u8)0 to 0xff, or in mac802154_wpan_xmit(), "u8 chan"
> will never == MAC802154_CHAN_NONE.
>
> The related warning:
>    net/mac802154/wpan.c:325:2: warning: comparison is always false
>    due to limited range of data type [-Wtype-limits]

Hm, do you really need to include this huge pile of code below into the
commit message?

Couldn't you make the commit message simpler resp. more compact, e.g.:

Change MAC802154_CHAN_NONE from ~(u8)0 to 0xff, otherwise the comparison
in mac802154_wpan_xmit() for ``chan == MAC802154_CHAN_NONE'' will not succeed.
This bug can be boiled down to ``u8 foo = 0xff; if (foo == ~(u8)0) [...]
else [...]'' where the condition will always take the else branch.

.... something like that maybe?

> The related unit test:
>    cc -Wall -O2 -o test test.c
>    ./test
>    will output 'different'
>
>    (after use '0xff' instead of '~(u8)0', it will output 'equal')
>
> -----------------------test code begin--------------------------------
>
> #include <stdio.h>
>
> typedef unsigned char u8;
>
> int main()
> {
> 	u8 uc = 0xff;
>
> 	if (uc == ~(u8)0)
> 		printf("\nequal\n");
> 	else
> 		printf("\ndifferent\n");
>
> 	return 0;
> }
>
> -----------------------test code end----------------------------------
>
>
> Signed-off-by: Chen Gang <gang.chen@asianux.com>
> ---
>   net/mac802154/mac802154.h | 2 +-
>   1 file changed, 1 insertion(+), 1 deletion(-)
>
> diff --git a/net/mac802154/mac802154.h b/net/mac802154/mac802154.h
> index a4dcaf1..703c121 100644
> --- a/net/mac802154/mac802154.h
> +++ b/net/mac802154/mac802154.h
> @@ -90,7 +90,7 @@ struct mac802154_sub_if_data {
>
>   #define MAC802154_MAX_XMIT_ATTEMPTS	3
>
> -#define MAC802154_CHAN_NONE		(~(u8)0) /* No channel is assigned */
> +#define MAC802154_CHAN_NONE		0xff /* No channel is assigned */
>
>   extern struct ieee802154_reduced_mlme_ops mac802154_mlme_reduced;
>   extern struct ieee802154_mlme_ops mac802154_mlme_wpan;
>
--
To unsubscribe from this list: send the line "unsubscribe netdev" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

On 2013年04月25日 17:59, Daniel Borkmann wrote:
> Change MAC802154_CHAN_NONE from ~(u8)0 to 0xff, otherwise the comparison
> in mac802154_wpan_xmit() for ``chan == MAC802154_CHAN_NONE'' will not
> succeed.
> This bug can be boiled down to ``u8 foo = 0xff; if (foo == ~(u8)0) [...]
> else [...]'' where the condition will always take the else branch.
> 
> ... something like that maybe?

That sounds good !!

I prefer to send patch v5 for that.

:-)