Hi

Am 28.07.21 um 22:11 schrieb Sam Ravnborg:
> Hi Dan,
> 
>>>
>>> I think I got it - we need to set irq_enabled to true.
>>> The documentation says so:
>>> "
>>>            * @irq_enabled:
>>>            *
>>>            * Indicates that interrupt handling is enabled, specifically vblank
>>>            * handling. Drivers which don't use drm_irq_install() need to set this
>>>            * to true manually.
>>> "
>>>
>>> Can you try to add the following line:
>>>
>>>
>>> +static int atmel_hlcdc_dc_irq_install(struct drm_device *dev, unsigned int irq)
>>> +{
>>> +       int ret;
>>> +
>>> +       if (irq == IRQ_NOTCONNECTED)
>>> +               return -ENOTCONN;
>>> +
>>>
>>>           dev->irq_enabled = true;                <= THIS LINE
>>>
>>>
>>> +       atmel_hlcdc_dc_irq_disable(dev);
>>> +
>>> +       ret = request_irq(irq, atmel_hlcdc_dc_irq_handler, 0, dev->driver->name, dev);
>>> +       if (ret)
>>> +               return ret;
>>>
>>> I hope this fixes it.
>>
>> It does!  With the irq_enabled line added everything is looking good.

Are you sure, you're testing with the latest drm-misc-next or drm-tip? 
Because using irq_enabled is deprecated and the flag was recently 
replaced by commit 1e4cd78ed493 ("drm: Don't test for IRQ support in 
VBLANK ioctls").

Best regards
Thomas

> 
> Great, thanks for testing.
> 
> Thomas - I assume you will do a re-spin and there is likely some fixes
> for the applied IRQ conversions too.
> 
> Note - irq_enabled must be cleared if request_irq fails. I did not
> include this in the testing here.
> 
> 	Sam
> 

-- 
Thomas Zimmermann
Graphics Driver Developer
SUSE Software Solutions Germany GmbH
Maxfeldstr. 5, 90409 Nürnberg, Germany
(HRB 36809, AG Nürnberg)
Geschäftsführer: Felix Imendörffer