On Fri, Jan 18, 2013 at 9:13 PM, Russell King - ARM Linux
<li...@arm.linux.org.uk> wrote:
> On Fri, Jan 18, 2013 at 04:08:48PM +0100, Linus Walleij wrote:
>> - Abstain from using IS_ERR_OR_NULL() in the driver core,
>> Russell recently sent a patch to remove it. Handle the
>> NULL case explicitly even though it's a bogus case.
>
> Hmmmmmmmm... I assume you're talking about this:
>
>> + p = create_pinctrl(dev);
>> + if (p && !IS_ERR(p))
>> + kref_init(&p->users);
>> + return p;
>
> which is... interesting. In mainline currently, the version I see there
> never returns NULL - it either returns an err pointer or a real pointer.
Actually it was this thing:
+ dpi->p = devm_pinctrl_get(dev);
+ if (!dpi->p) {
+ /* Shouldn't happen, but handle this anyway */
+ dev_warn(dev, "devm_pinctrl_get() returned NULL\n");
+ devm_kfree(dev, dpi);
+ dev->pins = NULL;
+ return 0;
+ }
I will delete both NULL checks and spin a v4.
> Next thing though is...
>
> create_pinctrl adds the struct pinctrl onto the global list, and then you
> intialize it here, which is an interesting sequence. It's safe at the
> moment because everything here is operating under a mutex, but it's a
> little obscure.
>
> Is there a reason not to initialize p->users inside create_pinctrl() ?
Not really. I'll fix it...
Yours,
Linus Walleij
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majord...@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
