On 08/18/2014 02:58 PM, Andrzej Hajda wrote:
> On 08/18/2014 02:38 PM, Shawn Guo wrote:
>> Hi Andrzej,
>>
>> On Mon, Aug 18, 2014 at 09:43:19AM +0200, Andrzej Hajda wrote:
>>>> diff --git a/drivers/staging/imx-drm/imx-drm-core.c b/drivers/staging/imx-drm/imx-drm-core.c
>>>> index def8280d7ee6..ab41152089a3 100644
>>>> --- a/drivers/staging/imx-drm/imx-drm-core.c
>>>> +++ b/drivers/staging/imx-drm/imx-drm-core.c
>>>> @@ -696,6 +696,29 @@ static int imx_drm_platform_remove(struct platform_device *pdev)
>>>> return 0;
>>>> }
>>>>
>>>> +#ifdef CONFIG_PM_SLEEP
>>>> +static int imx_drm_suspend(struct device *dev)
>>>> +{
>>>> + struct drm_device *drm_dev = dev_get_drvdata(dev);
>>>> +
>>>> + drm_kms_helper_poll_disable(drm_dev);
>>> drm_dev can be NULL here. You should add check before.
>> The drvdata is set to drm_dev in drm_driver .load() hook. So are you
>> saying that .suspend() hook could possibly be called before the .load()
>> hook gets called?
> .load hook is called after all components are attached to the master.
> So if suspend happen after probe of the master and before attaching the last
> component you will have NULL here, I guess.
Probably you can test it by disabling driver for one of drm components
and putting board in sleep mode.
Moreover I guess drvdata should be cleaned in .unload callback,
otherwise if for some reason one of components will be detached and suspend
happens drvdata will contain invalid pointer.
Regards
Andrzej
>
> Regards
> Andrzej
>
>> Shawn
>>
>>>> +
>>>> + return 0;
>>>> +}
>>>> +
>>>> +static int imx_drm_resume(struct device *dev)
>>>> +{
>>>> + struct drm_device *drm_dev = dev_get_drvdata(dev);
>>>> +
>>>> + drm_helper_resume_force_mode(drm_dev);
>>> ditto
_______________________________________________
dri-devel mailing list
dri-devel@xxxxxxxxxxxxxxxxxxxxx
http://lists.freedesktop.org/mailman/listinfo/dri-devel
