[PATCH v1] media: synopsys: hdmirx: Detect broken interrupt

Dmitry Osipenko dmitry.osipenko at collabora.com
Thu Oct 2 06:56:31 PDT 2025


On 10/2/25 16:36, Dmitry Osipenko wrote:
> Hi,
> 
> On 10/2/25 16:18, Sebastian Reichel wrote:
>>> +	*val = status & PHYCREG_CR_PARA_RD_DATA_MASK;
>>> +
>>> +	return 0;
>>> +}
>> Do you expect this to be used in other places in the driver? In that
>> case there should probably be some locking, since the hardware interface
>> obviously cannot handle concurrency. Otherwise maybe add a comment,
>> that the function may not be called if concurrency is possible?
> 
> Don't expect this function to be used in other places and haven't added
> locking on purpose to keep the code cleaner. Will add the comment.
> 
> Thanks for all suggestions.
> 

On a second thought, this would be a case where the guard() thingy would
be useful and help keeping code clean, will use it in v2.

-- 
Best regards,
Dmitry



More information about the Linux-rockchip mailing list