[RFC 02/11] i2c: add quirk checks to core

Sergei Shtylyov sergei.shtylyov at cogentembedded.com
Fri Jan 9 13:05:07 PST 2015


Hello.

On 01/09/2015 11:45 PM, Wolfram Sang wrote:

>>> Let the core do the checks if HW quirks prevent a transfer. Saves code
>> >from drivers and adds consistency.

>>> Signed-off-by: Wolfram Sang <wsa at the-dreams.de>
>>> ---
>>>   drivers/i2c/i2c-core.c | 53 ++++++++++++++++++++++++++++++++++++++++++++++++++
>>>   1 file changed, 53 insertions(+)
>>>
>>> diff --git a/drivers/i2c/i2c-core.c b/drivers/i2c/i2c-core.c
>>> index 39d25a8cb1ad..7b10a19abf5b 100644
>>> --- a/drivers/i2c/i2c-core.c
>>> +++ b/drivers/i2c/i2c-core.c
>>> @@ -2063,6 +2063,56 @@ module_exit(i2c_exit);
>>>    * ----------------------------------------------------
>>>    */
>>>
>>> +/* Check if val is exceeding the quirk IFF quirk is non 0 */
>>> +#define i2c_quirk_exceeded(val, quirk) ((quirk) && ((val) > (quirk)))
>>> +
>>> +static int i2c_quirk_error(struct i2c_adapter *adap, struct i2c_msg *msg, char *err_msg)
>>> +{
>>> +	dev_err(&adap->dev, "quirk: %s (addr 0x%04x, size %u)\n", err_msg, msg->addr, msg->len);
>>> +	return -EOPNOTSUPP;
>>> +}

>>     Always returning the same value doesn't make much sense. Are you trying
>> to save space on the call sites?

> Please elaborate. I think it does. If a quirk matches, we report that we
> don't support this transfer.

    OK, but what's the point of having this function return *int* if it always 
returns the same value? AFAIU, you're trying to save the code space on the 
call sites of this function by not having *return* -EOPNOTSUPP there each time?

>> [...]
>>> @@ -2080,6 +2130,9 @@ int __i2c_transfer(struct i2c_adapter *adap, struct i2c_msg *msgs, int num)
>>>   	unsigned long orig_jiffies;
>>>   	int ret, try;
>>>
>>> +	if (adap->quirks && i2c_check_for_quirks(adap, msgs, num))

>>     So, you only check for non-zero result of this function? Perhaps it makes
>> sense to return true/false instead?

> Could be done, but what would be the advantage? A lot of functions
> return errno or 0.

    It would have been OK if you were actually caring about the result, e.g. 
returning it from __i2c_transfer(). Since you don't, IMO it would make more 
sense to return true from i2c_check_for_quirks() (making it *bool*) iff it did 
find/apply a quirk.

WBR, Sergei




More information about the linux-arm-kernel mailing list