Hi Joe,
On Thu, Jun 21, 2012 at 4:48 AM, Joe Perches <joe@xxxxxxxxxxx> wrote:
> On Wed, 2012-06-20 at 16:08 -0700, Greg Kroah-Hartman wrote:
>> On Wed, Jun 20, 2012 at 12:35:42AM +0530, Devendra Naga wrote:
>> > fixed some of the coding style problems reported by checkpatch
> []
>> > @@ -66,11 +69,10 @@ short eprom_r(struct net_device *dev)
>> > {
>> > short bit;
>> >
>> > - bit=(read_nic_byte_E(dev, EPROM_CMD) & (1<<EPROM_R_SHIFT) );
>> > + bit = (read_nic_byte_E(dev, EPROM_CMD) & (1<<EPROM_R_SHIFT));
>> > udelay(EPROM_DELAY);
>> >
>> > - if(bit) return 1;
>> > - return 0;
>> > + return !!bit;
>>
>> Oh come on, really? !! is more "clear" here?
>
> Depends on the reader. !! is pretty common.
>
>> No, please be painfully obvious, that's the only way to write kernel
>> code. Not like this.
>
> I'd just make the return a bool instead.
>
taking another variable and assign it like bool ret = !!bit, and
returning ret?, i think this doesn't look better.
> Also, there are unnecessary parens that could
> be removed to make the code clearer.
>
> (1<<EPROM_R_SHIFT), (1<<EPROM_W_SHIFT) and
> (1<<EPROM_CK_SHIFT) could be new #defines too.
>
>
Will do.
thanks joe.
Devendra.
_______________________________________________
devel mailing list
devel@xxxxxxxxxxxxxxxxxxxxxx
http://driverdev.linuxdriverproject.org/mailman/listinfo/devel
