Re: [PATCH] mtd: rawnand: mark expected switch fall-throughs

[Gustavo A. R. Silva]

Miquel, Boris,

On 1/28/19 3:13 AM, Miquel Raynal wrote:
> Hi Boris,
> 
> Boris Brezillon <bbrezillon@xxxxxxxxxx> wrote on Sat, 26 Jan 2019
> 17:54:29 +0100:
> 
>> On Sat, 26 Jan 2019 07:48:50 -0600
>> "Gustavo A. R. Silva" <gustavo@xxxxxxxxxxxxxx> wrote:
>>
>>> Hey Boris,
>>>
>>> On 1/26/19 3:52 AM, Boris Brezillon wrote:  
>>>> On Fri, 25 Jan 2019 15:09:50 -0600
>>>> "Gustavo A. R. Silva" <gustavo@xxxxxxxxxxxxxx> wrote:
>>>>     
>>>>> diff --git a/drivers/mtd/nand/raw/nandsim.c b/drivers/mtd/nand/raw/nandsim.c
>>>>> index 933d1a629c51..d33e15dc4cdc 100644
>>>>> --- a/drivers/mtd/nand/raw/nandsim.c
>>>>> +++ b/drivers/mtd/nand/raw/nandsim.c
>>>>> @@ -2251,9 +2251,10 @@ static int __init ns_init_module(void)
>>>>>  
>>>>>  	switch (bbt) {
>>>>>  	case 2:
>>>>> -		 chip->bbt_options |= NAND_BBT_NO_OOB;
>>>>> +		chip->bbt_options |= NAND_BBT_NO_OOB;
>>>>> +		/* fall through */
>>>>>  	case 1:
>>>>> -		 chip->bbt_options |= NAND_BBT_USE_FLASH;
>>>>> +		chip->bbt_options |= NAND_BBT_USE_FLASH;    
>>>>
>>>> You miss a '/* fall through */' here.
>>>>     
>>>
>>> Not really.  Notice that in this case the code falls through
>>> to a break statement.  
>>
>> Still find it weird to mandate fall through comments in all cases but
>> this one...
> 
> Yes please, even if there is no GCC warning I think you can add one
> here.
> 

Yep. I get your point.

I've just sent v2: https://lore.kernel.org/patchwork/patch/1036251/

Thanks for the feedback.
--
Gustavo