Hi Arnd,
On 29 November 2017 at 18:07, Arnd Bergmann <a...@arndb.de> wrote:
> On Mon, Nov 20, 2017 at 7:54 AM, Baolin Wang <baolin.w...@linaro.org> wrote:
>> Some system control registers need hardware spinlock to synchronize
>> between the multiple subsystems, so we should add hardware spinlock
>> support for syscon.
>>
>
>> @@ -87,6 +88,12 @@ static struct syscon *of_syscon_register(struct
>> device_node *np)
>> if (ret)
>> reg_io_width = 4;
>>
>> + ret = of_hwspin_lock_get_id(np, 0);
>> + if (ret > 0) {
>> + syscon_config.hwlock_id = ret;
>> + syscon_config.hwlock_mode = HWLOCK_IRQSTATE;
>> + }
>
> Hi Baolin,
>
> The error handling here seems insufficient, I think we have to treat
> the following cases separately:
>
> - ret>0 (this is fine)
> - no spinlock in DT: of_hwspin_lock_get_id currently returns -EINVAL,
> we probably want to use a different return code (maybe -ESRCH ?)
> here so we can tell the difference beween no spinlock, and an
> invalid spinlock
> - deferred probing: if the spinlock is there but the driver is not
> yet available, we may need to propagate the -EPROBE_DEFER
> here
> - other error: this would probably be a fatal condition here, and
> we should print a warning and clean up.
You are correct and I will add some error handling. Thanks for your comments.
--
Baolin.wang
Best Regards
