[v11,17/19] drm: bridge: analogix/dp: expand the look time for waiting AUX CH reply
diff mbox

Message ID 1450238257-3274-1-git-send-email-ykk@rock-chips.com
State New
Headers show

Commit Message

Yakir Yang Dec. 16, 2015, 3:57 a.m. UTC
After test on rockchiop platform, i found sometims driver would failed
at reading EDID message. After debugging more, i found that it's okay
to read_a byte from i2c, but it would failed at AUX transcation if we
try to ready multi-bytes from i2c.

Driver just can't received the AUX CH reply command, even no AUX error
code. I may guess that the AUX wait time is not enough, so I try to
expand the AUX wait time, and i do see this could fix the EDID read
failed at rockchip platform.

And I thought that expand the wait time won't hurt Exynos platform too
much, cause Exynos didn't have this problem, then driver would received
the reply command very soon, so no more additional wait time would bring
to Exynos platform.

Signed-off-by: Yakir Yang <ykk@rock-chips.com>
---
Changes in v11: None
Changes in v10: None
Changes in v9: None
Changes in v8: None
Changes in v7: None
Changes in v6: None
Changes in v5: None
Changes in v4: None
Changes in v3: None
Changes in v2: None

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

Comments

Jingoo Han Dec. 22, 2015, 12:26 p.m. UTC | #1
On Wednesday, December 16, 2015 12:58 PM, Yakir Yang wrote:
> 
> After test on rockchiop platform, i found sometims driver would failed
> at reading EDID message. After debugging more, i found that it's okay
> to read_a byte from i2c, but it would failed at AUX transcation if we
> try to ready multi-bytes from i2c.
> 
> Driver just can't received the AUX CH reply command, even no AUX error
> code. I may guess that the AUX wait time is not enough, so I try to
> expand the AUX wait time, and i do see this could fix the EDID read
> failed at rockchip platform.
> 
> And I thought that expand the wait time won't hurt Exynos platform too
> much, cause Exynos didn't have this problem, then driver would received
> the reply command very soon, so no more additional wait time would bring
> to Exynos platform.
> 
> Signed-off-by: Yakir Yang <ykk@rock-chips.com>
> ---
> Changes in v11: None
> Changes in v10: None
> Changes in v9: None
> Changes in v8: None
> Changes in v7: None
> Changes in v6: None
> Changes in v5: None
> Changes in v4: None
> Changes in v3: None
> Changes in v2: None
> 
>  drivers/gpu/drm/bridge/analogix/analogix_dp_reg.c | 2 +-
>  1 file changed, 1 insertion(+), 1 deletion(-)
> 
> diff --git a/drivers/gpu/drm/bridge/analogix/analogix_dp_reg.c
> b/drivers/gpu/drm/bridge/analogix/analogix_dp_reg.c
> index c7e2959..dc376bd 100644
> --- a/drivers/gpu/drm/bridge/analogix/analogix_dp_reg.c
> +++ b/drivers/gpu/drm/bridge/analogix/analogix_dp_reg.c
> @@ -482,7 +482,7 @@ int analogix_dp_start_aux_transaction(struct analogix_dp_device *dp)
>  	reg = readl(dp->reg_base + ANALOGIX_DP_INT_STA);
>  	while (!(reg & RPLY_RECEIV)) {
>  		timeout_loop++;
> -		if (DP_TIMEOUT_LOOP_COUNT < timeout_loop) {
> +		if (DP_TIMEOUT_LOOP_COUNT * 10 < timeout_loop) {

No, I hate this coding.
analogix_dp_reg.c is the common code that can be shared by various SoCs.
Please, find another way.

Best regards,
Jingoo Han


>  			dev_err(dp->dev, "AUX CH command reply failed!\n");
>  			return -ETIMEDOUT;
>  		}
> --
> 1.9.1
Yakir Yang Dec. 23, 2015, 4:24 a.m. UTC | #2
Hi Jingoo,

On 12/22/2015 08:26 PM, Jingoo Han wrote:
> On Wednesday, December 16, 2015 12:58 PM, Yakir Yang wrote:
>> After test on rockchiop platform, i found sometims driver would failed
>> at reading EDID message. After debugging more, i found that it's okay
>> to read_a byte from i2c, but it would failed at AUX transcation if we
>> try to ready multi-bytes from i2c.
>>
>> Driver just can't received the AUX CH reply command, even no AUX error
>> code. I may guess that the AUX wait time is not enough, so I try to
>> expand the AUX wait time, and i do see this could fix the EDID read
>> failed at rockchip platform.
>>
>> And I thought that expand the wait time won't hurt Exynos platform too
>> much, cause Exynos didn't have this problem, then driver would received
>> the reply command very soon, so no more additional wait time would bring
>> to Exynos platform.
>>
>> Signed-off-by: Yakir Yang <ykk@rock-chips.com>
>> ---
>> Changes in v11: None
>> Changes in v10: None
>> Changes in v9: None
>> Changes in v8: None
>> Changes in v7: None
>> Changes in v6: None
>> Changes in v5: None
>> Changes in v4: None
>> Changes in v3: None
>> Changes in v2: None
>>
>>   drivers/gpu/drm/bridge/analogix/analogix_dp_reg.c | 2 +-
>>   1 file changed, 1 insertion(+), 1 deletion(-)
>>
>> diff --git a/drivers/gpu/drm/bridge/analogix/analogix_dp_reg.c
>> b/drivers/gpu/drm/bridge/analogix/analogix_dp_reg.c
>> index c7e2959..dc376bd 100644
>> --- a/drivers/gpu/drm/bridge/analogix/analogix_dp_reg.c
>> +++ b/drivers/gpu/drm/bridge/analogix/analogix_dp_reg.c
>> @@ -482,7 +482,7 @@ int analogix_dp_start_aux_transaction(struct analogix_dp_device *dp)
>>   	reg = readl(dp->reg_base + ANALOGIX_DP_INT_STA);
>>   	while (!(reg & RPLY_RECEIV)) {
>>   		timeout_loop++;
>> -		if (DP_TIMEOUT_LOOP_COUNT < timeout_loop) {
>> +		if (DP_TIMEOUT_LOOP_COUNT * 10 < timeout_loop) {
> No, I hate this coding.
> analogix_dp_reg.c is the common code that can be shared by various SoCs.
> Please, find another way.

Okay, I have double checked that i do have this problem in my side. 
Hmmm.....
I thought it's okay for you if I expand the "DP_TIMEOUT_LOOP_COUNT" 
directly,
it won't hurt Exynos platform too much, cause Exynos didn't have this 
problem,
then driver would received,the reply command very soon, so no more 
additional
wait time would bring to Exynos platform.

And actually the datasheet haven't described the spec of aux reply delay 
time.

Thanks,
- Yakir

> Best regards,
> Jingoo Han
>
>
>>   			dev_err(dp->dev, "AUX CH command reply failed!\n");
>>   			return -ETIMEDOUT;
>>   		}
>> --
>> 1.9.1
>
>
>
>
Yakir Yang Dec. 23, 2015, 6 a.m. UTC | #3
Hi Jingoo,

On 12/23/2015 12:24 PM, Yakir Yang wrote:
> Hi Jingoo,
>
> On 12/22/2015 08:26 PM, Jingoo Han wrote:
>> On Wednesday, December 16, 2015 12:58 PM, Yakir Yang wrote:
>>> After test on rockchiop platform, i found sometims driver would failed
>>> at reading EDID message. After debugging more, i found that it's okay
>>> to read_a byte from i2c, but it would failed at AUX transcation if we
>>> try to ready multi-bytes from i2c.
>>>
>>> Driver just can't received the AUX CH reply command, even no AUX error
>>> code. I may guess that the AUX wait time is not enough, so I try to
>>> expand the AUX wait time, and i do see this could fix the EDID read
>>> failed at rockchip platform.
>>>
>>> And I thought that expand the wait time won't hurt Exynos platform too
>>> much, cause Exynos didn't have this problem, then driver would received
>>> the reply command very soon, so no more additional wait time would 
>>> bring
>>> to Exynos platform.
>>>
>>> Signed-off-by: Yakir Yang <ykk@rock-chips.com>
>>> ---
>>> Changes in v11: None
>>> Changes in v10: None
>>> Changes in v9: None
>>> Changes in v8: None
>>> Changes in v7: None
>>> Changes in v6: None
>>> Changes in v5: None
>>> Changes in v4: None
>>> Changes in v3: None
>>> Changes in v2: None
>>>
>>>   drivers/gpu/drm/bridge/analogix/analogix_dp_reg.c | 2 +-
>>>   1 file changed, 1 insertion(+), 1 deletion(-)
>>>
>>> diff --git a/drivers/gpu/drm/bridge/analogix/analogix_dp_reg.c
>>> b/drivers/gpu/drm/bridge/analogix/analogix_dp_reg.c
>>> index c7e2959..dc376bd 100644
>>> --- a/drivers/gpu/drm/bridge/analogix/analogix_dp_reg.c
>>> +++ b/drivers/gpu/drm/bridge/analogix/analogix_dp_reg.c
>>> @@ -482,7 +482,7 @@ int analogix_dp_start_aux_transaction(struct 
>>> analogix_dp_device *dp)
>>>       reg = readl(dp->reg_base + ANALOGIX_DP_INT_STA);
>>>       while (!(reg & RPLY_RECEIV)) {
>>>           timeout_loop++;
>>> -        if (DP_TIMEOUT_LOOP_COUNT < timeout_loop) {
>>> +        if (DP_TIMEOUT_LOOP_COUNT * 10 < timeout_loop) {
>> No, I hate this coding.
>> analogix_dp_reg.c is the common code that can be shared by various SoCs.
>> Please, find another way.
>
> Okay, I have double checked that i do have this problem in my side. 
> Hmmm.....
> I thought it's okay for you if I expand the "DP_TIMEOUT_LOOP_COUNT" 
> directly,
> it won't hurt Exynos platform too much, cause Exynos didn't have this 
> problem,
> then driver would received,the reply command very soon, so no more 
> additional
> wait time would bring to Exynos platform.
>

Oh, sorry, little mistaken, I mean, is it okay for you to expand the 
"DP_TIMEOUT_LOOP_COUNT" directly ?

- Yakir


> And actually the datasheet haven't described the spec of aux reply 
> delay time.
>
> Thanks,
> - Yakir
>
>> Best regards,
>> Jingoo Han
>>
>>
>>>               dev_err(dp->dev, "AUX CH command reply failed!\n");
>>>               return -ETIMEDOUT;
>>>           }
>>> -- 
>>> 1.9.1
>>
>>
>>
>>
>
>
>
> _______________________________________________
> Linux-rockchip mailing list
> Linux-rockchip@lists.infradead.org
> http://lists.infradead.org/mailman/listinfo/linux-rockchip
>
>
>
Jingoo Han Dec. 23, 2015, 3:15 p.m. UTC | #4
On Wednesday, December 23, 2015 3:01 PM, Yakir Yang wrote:
> 
> Hi Jingoo,
> 
> On 12/23/2015 12:24 PM, Yakir Yang wrote:
> > Hi Jingoo,
> >
> > On 12/22/2015 08:26 PM, Jingoo Han wrote:
> >> On Wednesday, December 16, 2015 12:58 PM, Yakir Yang wrote:
> >>> After test on rockchiop platform, i found sometims driver would failed
> >>> at reading EDID message. After debugging more, i found that it's okay
> >>> to read_a byte from i2c, but it would failed at AUX transcation if we
> >>> try to ready multi-bytes from i2c.
> >>>
> >>> Driver just can't received the AUX CH reply command, even no AUX error
> >>> code. I may guess that the AUX wait time is not enough, so I try to
> >>> expand the AUX wait time, and i do see this could fix the EDID read
> >>> failed at rockchip platform.
> >>>
> >>> And I thought that expand the wait time won't hurt Exynos platform too
> >>> much, cause Exynos didn't have this problem, then driver would received
> >>> the reply command very soon, so no more additional wait time would
> >>> bring
> >>> to Exynos platform.
> >>>
> >>> Signed-off-by: Yakir Yang <ykk@rock-chips.com>
> >>> ---
> >>> Changes in v11: None
> >>> Changes in v10: None
> >>> Changes in v9: None
> >>> Changes in v8: None
> >>> Changes in v7: None
> >>> Changes in v6: None
> >>> Changes in v5: None
> >>> Changes in v4: None
> >>> Changes in v3: None
> >>> Changes in v2: None
> >>>
> >>>   drivers/gpu/drm/bridge/analogix/analogix_dp_reg.c | 2 +-
> >>>   1 file changed, 1 insertion(+), 1 deletion(-)
> >>>
> >>> diff --git a/drivers/gpu/drm/bridge/analogix/analogix_dp_reg.c
> >>> b/drivers/gpu/drm/bridge/analogix/analogix_dp_reg.c
> >>> index c7e2959..dc376bd 100644
> >>> --- a/drivers/gpu/drm/bridge/analogix/analogix_dp_reg.c
> >>> +++ b/drivers/gpu/drm/bridge/analogix/analogix_dp_reg.c
> >>> @@ -482,7 +482,7 @@ int analogix_dp_start_aux_transaction(struct
> >>> analogix_dp_device *dp)
> >>>       reg = readl(dp->reg_base + ANALOGIX_DP_INT_STA);
> >>>       while (!(reg & RPLY_RECEIV)) {
> >>>           timeout_loop++;
> >>> -        if (DP_TIMEOUT_LOOP_COUNT < timeout_loop) {
> >>> +        if (DP_TIMEOUT_LOOP_COUNT * 10 < timeout_loop) {
> >> No, I hate this coding.
> >> analogix_dp_reg.c is the common code that can be shared by various SoCs.
> >> Please, find another way.
> >
> > Okay, I have double checked that i do have this problem in my side.
> > Hmmm.....
> > I thought it's okay for you if I expand the "DP_TIMEOUT_LOOP_COUNT"
> > directly,
> > it won't hurt Exynos platform too much, cause Exynos didn't have this
> > problem,
> > then driver would received,the reply command very soon, so no more
> > additional
> > wait time would bring to Exynos platform.
> >
> 
> Oh, sorry, little mistaken, I mean, is it okay for you to expand the
> "DP_TIMEOUT_LOOP_COUNT" directly ?

NO.

Don't change DP_TIMEOUT_LOOP_COUNT.
Just, change wait time by adding new DT property.

> 
> - Yakir
> 
> 
> > And actually the datasheet haven't described the spec of aux reply
> > delay time.

Right.
However, the problem is specific for Rockchip platform.
It is the hardware bug of Rockchip platform.
Your delay time is too long to apply to other SoCs platform.

Best regards,
Jingoo Han

> >
> > Thanks,
> > - Yakir
> >
> >> Best regards,
> >> Jingoo Han
> >>
> >>
> >>>               dev_err(dp->dev, "AUX CH command reply failed!\n");
> >>>               return -ETIMEDOUT;
> >>>           }
> >>> --
> >>> 1.9.1
> >>
> >>
> >>
> >>
> >
> >
> >
> > _______________________________________________
> > Linux-rockchip mailing list
> > Linux-rockchip@lists.infradead.org
> > http://lists.infradead.org/mailman/listinfo/linux-rockchip
> >
> >
> >

Patch
diff mbox

diff --git a/drivers/gpu/drm/bridge/analogix/analogix_dp_reg.c b/drivers/gpu/drm/bridge/analogix/analogix_dp_reg.c
index c7e2959..dc376bd 100644
--- a/drivers/gpu/drm/bridge/analogix/analogix_dp_reg.c
+++ b/drivers/gpu/drm/bridge/analogix/analogix_dp_reg.c
@@ -482,7 +482,7 @@  int analogix_dp_start_aux_transaction(struct analogix_dp_device *dp)
 	reg = readl(dp->reg_base + ANALOGIX_DP_INT_STA);
 	while (!(reg & RPLY_RECEIV)) {
 		timeout_loop++;
-		if (DP_TIMEOUT_LOOP_COUNT < timeout_loop) {
+		if (DP_TIMEOUT_LOOP_COUNT * 10 < timeout_loop) {
 			dev_err(dp->dev, "AUX CH command reply failed!\n");
 			return -ETIMEDOUT;
 		}