Re: [PATCH] drivers/scsi/lpfc/lpfc_hw.h: Some minor cleanup.

[Richard Knutsson <ricknu-0@student.ltu.se>]

Denys Vlasenko wrote:
> On Tuesday 30 October 2007 10:54, Richard Knutsson wrote:
>   
>> Signed-off-by: Richard Knutsson <ricknu-0@student.ltu.se>
>> ---
>> Diffed against linus-git
>> Checked with script/checkpatch.pl
>>
>>
>> diff --git a/drivers/scsi/lpfc/lpfc_hw.h b/drivers/scsi/lpfc/lpfc_hw.h
>> index 451accd..6f56528 100644
>> --- a/drivers/scsi/lpfc/lpfc_hw.h
>> +++ b/drivers/scsi/lpfc/lpfc_hw.h
>> @@ -3158,31 +3158,30 @@ struct lpfc_sli2_slim {
>>   *
>>   * Parameters:
>>   * device : struct pci_dev 's device field
>> - *
>> - * return 1 => TRUE
>> - *        0 => FALSE
>>   */
>> -static inline int
>> +static inline bool
>>  lpfc_is_LC_HBA(unsigned short device)
>>  {
>> -	if ((device == PCI_DEVICE_ID_TFLY) ||
>> -	    (device == PCI_DEVICE_ID_PFLY) ||
>> -	    (device == PCI_DEVICE_ID_LP101) ||
>> -	    (device == PCI_DEVICE_ID_BMID) ||
>> -	    (device == PCI_DEVICE_ID_BSMB) ||
>> -	    (device == PCI_DEVICE_ID_ZMID) ||
>> -	    (device == PCI_DEVICE_ID_ZSMB) ||
>> -	    (device == PCI_DEVICE_ID_RFLY))
>> -		return 1;
>> -	else
>> -		return 0;
>> +	switch (device) {
>> +	case PCI_DEVICE_ID_TFLY:
>> +	case PCI_DEVICE_ID_PFLY:
>> +	case PCI_DEVICE_ID_LP101:
>> +	case PCI_DEVICE_ID_BMID:
>> +	case PCI_DEVICE_ID_BSMB:
>> +	case PCI_DEVICE_ID_ZMID:
>> +	case PCI_DEVICE_ID_ZSMB:
>> +	case PCI_DEVICE_ID_RFLY:
>> +		return true;
>> +	}
>> +
>> +	return false;
>>  }
>>     
>
> Why is this patch useful? 
Just simpler to read, no?
> I'd rather do this instead:
>
> -static inline int
> +static int
>
> (this function has three callsites, thus de-inlining will
> make code smaller)
>   
It is returning (and the results are used as) a boolean so why should it 
be an integer? But I have no objection to de-inline it.
Alright with:
-static inline int
+static bool
?

Richard Knutsson