diff mbox

[v3,12/14] drm: bridge/analogix_dp: expand the delay time for hpd detect

Message ID 1439995929-18719-1-git-send-email-ykk@rock-chips.com (mailing list archive)
State New, archived
Headers show

Commit Message

Yakir Yang Aug. 19, 2015, 2:52 p.m. UTC
Some edp screen with no hpd signal would need some delay time
to ensure that screen would be ready for work, so we can expand
the delay time in hpd detect function, it works prefectly on my
rk3288 sdk board.

Signed-off-by: Yakir Yang <ykk@rock-chips.com>
---
Changes in v3: None
Changes in v2: None

 drivers/gpu/drm/bridge/analogix_dp_core.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

Comments

Han Jingoo Aug. 20, 2015, 6:11 a.m. UTC | #1
On 2015. 8. 19., at PM 11:52, Yakir Yang <ykk@rock-chips.com> wrote:
> 
> Some edp screen with no hpd signal would need some delay time
> to ensure that screen would be ready for work, so we can expand
> the delay time in hpd detect function, it works prefectly on my
> rk3288 sdk board.

Then, this delay has a dependency on the "rk3288 sdk" board.
Also, if the delay time is expanded, the booting time of some Exybos boards will be increased unnecessarily. :-(

So, please add new DT property such as 'hpd-delay' that can be added to board DT files.

If there is not that DT property in DT files, the default value '10' will written to a variable such as 'unsigned int hpd_delay'.
If there is the DT property in DT files, the delay value will written to the variable when parsing DT values
and will be used in analogix_dp_detect_hpd().

What I want to say is that there should not be harmful effect on the existing Exynos boards, due to unrelated reasons.

Best regards,
Jingoo Han

> Signed-off-by: Yakir Yang <ykk@rock-chips.com>
> ---
> Changes in v3: None
> Changes in v2: None
> 
> drivers/gpu/drm/bridge/analogix_dp_core.c | 2 +-
> 1 file changed, 1 insertion(+), 1 deletion(-)
> 
> diff --git a/drivers/gpu/drm/bridge/analogix_dp_core.c b/drivers/gpu/drm/bridge/analogix_dp_core.c
> index 99870f7..75dd44a 100644
> --- a/drivers/gpu/drm/bridge/analogix_dp_core.c
> +++ b/drivers/gpu/drm/bridge/analogix_dp_core.c
> @@ -68,7 +68,7 @@ static int analogix_dp_detect_hpd(struct analogix_dp_device *dp)
>            return 0;
> 
>        timeout_loop++;
> -        usleep_range(10, 11);
> +        usleep_range(100, 110);
>    }
> 
>    /*
> -- 
> 1.9.1
> 
>
Yakir Yang Aug. 20, 2015, 8:02 a.m. UTC | #2
Hi Jingoo,

On 08/20/2015 01:11 AM, Jingoo Han wrote:
> On 2015. 8. 19., at PM 11:52, Yakir Yang <ykk@rock-chips.com> wrote:
>> Some edp screen with no hpd signal would need some delay time
>> to ensure that screen would be ready for work, so we can expand
>> the delay time in hpd detect function, it works prefectly on my
>> rk3288 sdk board.
> Then, this delay has a dependency on the "rk3288 sdk" board.
> Also, if the delay time is expanded, the booting time of some Exybos boards will be increased unnecessarily. :-(
>
> So, please add new DT property such as 'hpd-delay' that can be added to board DT files.
>
> If there is not that DT property in DT files, the default value '10' will written to a variable such as 'unsigned int hpd_delay'.
> If there is the DT property in DT files, the delay value will written to the variable when parsing DT values
> and will be used in analogix_dp_detect_hpd().
>
> What I want to say is that there should not be harmful effect on the existing Exynos boards, due to unrelated reasons.

Yeah, you are right, I made an mistake here.

And I want to put this delay to "need-force-hpd" code, cause
this property is for the no-hpd-signal eDP screen.

But strangely, with my this series, I don't need the expand delay
any more, I am not sure which change improved this, I guess
those delay time should come from drm core ?

Whatever seems we don't need this delay for now, and if I can
find the exact reason and realize I still need this delay, I prefer
to add those delay in "need-force-hpd" code.

Thanks,
- Yakir
> Best regards,
> Jingoo Han
>
>> Signed-off-by: Yakir Yang <ykk@rock-chips.com>
>> ---
>> Changes in v3: None
>> Changes in v2: None
>>
>> drivers/gpu/drm/bridge/analogix_dp_core.c | 2 +-
>> 1 file changed, 1 insertion(+), 1 deletion(-)
>>
>> diff --git a/drivers/gpu/drm/bridge/analogix_dp_core.c b/drivers/gpu/drm/bridge/analogix_dp_core.c
>> index 99870f7..75dd44a 100644
>> --- a/drivers/gpu/drm/bridge/analogix_dp_core.c
>> +++ b/drivers/gpu/drm/bridge/analogix_dp_core.c
>> @@ -68,7 +68,7 @@ static int analogix_dp_detect_hpd(struct analogix_dp_device *dp)
>>             return 0;
>>
>>         timeout_loop++;
>> -        usleep_range(10, 11);
>> +        usleep_range(100, 110);
>>     }
>>
>>     /*
>> -- 
>> 1.9.1
>>
>>
>
>
diff mbox

Patch

diff --git a/drivers/gpu/drm/bridge/analogix_dp_core.c b/drivers/gpu/drm/bridge/analogix_dp_core.c
index 99870f7..75dd44a 100644
--- a/drivers/gpu/drm/bridge/analogix_dp_core.c
+++ b/drivers/gpu/drm/bridge/analogix_dp_core.c
@@ -68,7 +68,7 @@  static int analogix_dp_detect_hpd(struct analogix_dp_device *dp)
 			return 0;
 
 		timeout_loop++;
-		usleep_range(10, 11);
+		usleep_range(100, 110);
 	}
 
 	/*