Re: [PATCH v1 1/9] staging: android: binder: Move some of the logic into subfunction

[Serban Constantinescu <Serban.Constantinescu@arm.com>]

On 05/12/13 08:00, Dan Carpenter wrote:
> On Wed, Dec 04, 2013 at 06:09:33PM +0000, Serban Constantinescu wrote:
>> +static void bc_dead_binder_done(struct binder_proc *proc,
>> +		    struct binder_thread *thread, void __user *cookie)
>> +{
>> +	struct binder_work *w;
>> +	struct binder_ref_death *death = NULL;
>> +
>> +	list_for_each_entry(w, &proc->delivered_death, entry) {
>> +		struct binder_ref_death *tmp_death = container_of(w, struct binder_ref_death, work);
>
> Put a blank line after the variable declaration block.  Also break it up
> into two lines instead of having the lines be a million characters long.

I will do that, I have followed the same conventions used by the code 
allready there.

>
> 	list_for_each_entry(w, &proc->delivered_death, entry) {
> 		struct binder_ref_death *tmp_death;
>
> 		tmp_death = container_of(w, struct binder_ref_death, work);
>
>> +		if (tmp_death->cookie == cookie) {
>> +			death = tmp_death;
>> +			return;
>
> Should be a break here.  This function wasn't tested.

There should be a break there! I have tested the driver with 32bit 
Android tests, none of which seem to hit this.

Thanks for your review,
Serban C