On 11/20/2012 02:40 PM, Peter Chen wrote:
> On Tue, Nov 20, 2012 at 12:41:03PM +0100, Marc Kleine-Budde wrote:
>> On 11/20/2012 02:33 AM, Peter Chen wrote:
>>> On Mon, Nov 19, 2012 at 05:05:16PM +0100, Marc Kleine-Budde wrote:
>>>>> +int ci_hdrc_otg_init(struct ci13xxx *ci)
>>>>> +{
>>>>> + /* Useless at current */
>>>>> + ci->otg.set_peripheral = ci_otg_set_peripheral;
>>>>> + ci->otg.set_host = ci_otg_set_host;
>>>>> + if (!IS_ERR_OR_NULL(ci->transceiver))
>>>> ^^^^^^^^^^^^^^
>>>>
>>>> Is considered harmful by some people. Can we test against NULL here?
>>>
>>> Any stories that it is harmful?
>>
>> Your use-case isn't but there others are:
>> http://www.mail-archive.com/linux-omap@xxxxxxxxxxxxxxx/msg78030.html
>
> Thanks, marc. It said below code is error if the foo is really null.
> if (IS_ERR_OR_NULL(foo))
> return PTR_ERR(foo);
Yes. So my next comment is nitpicking: :)
> For this case, the ci->transceiver may be null or errno pointer, it
> needs IS_ERR_OR_NULL.
Yes, but you can set ci->transceiver to NULL in case of an error when
assigning it, to avoid IS_ERR_OR_NULL here.
Marc
--
Pengutronix e.K. | Marc Kleine-Budde |
Industrial Linux Solutions | Phone: +49-231-2826-924 |
Vertretung West/Dortmund | Fax: +49-5121-206917-5555 |
Amtsgericht Hildesheim, HRA 2686 | http://www.pengutronix.de |
Attachment:
signature.asc
Description: OpenPGP digital signature
