[Intel-gfx] [PATCH v2 1/2] drm: Add retries for dp dual mode read

[Sharma, Shashank]

Regards

Shashank


On 8/24/2017 5:49 PM, Imre Deak wrote:
> On Thu, Aug 24, 2017 at 05:40:32PM +0530, Sharma, Shashank wrote:
>> Regards
>>
>> Shashank
>>
>>
>> On 8/24/2017 5:19 PM, Imre Deak wrote:
>>> On Wed, Aug 23, 2017 at 06:12:51PM +0530, Shashank Sharma wrote:
>>>>   From the CI builds, its been observed that during a driver
>>>> reload/insert, dp dual mode read function sometimes fails to
>>>> read from dual mode devices (like LSPCON) over i2c-over-aux
>>>> channel.
>>>>
>>>> This patch:
>>>> - adds some delay and few retries, allowing a scope for these
>>>>     devices to settle down and respond.
>>>> - changes one error log's level from ERROR->DEBUG as we want
>>>>     to call it an error only after all the retries are exhausted.
>>>>
>>>> V2: Addressed review comments from Jani (for loop for retry)
>>>>
>>>> Cc: Ville Syrjala <ville.syrjala at linux.intel.com>
>>>> Cc: Imre Deak <imre.deak at intel.com>
>>>> Cc: Jani Nikula <jani.nikula at linux.intel.com>
>>>> Signed-off-by: Shashank Sharma <shashank.sharma at intel.com>
>>>> ---
>>>>    drivers/gpu/drm/drm_dp_dual_mode_helper.c | 12 ++++++++++--
>>>>    1 file changed, 10 insertions(+), 2 deletions(-)
>>>>
>>>> diff --git a/drivers/gpu/drm/drm_dp_dual_mode_helper.c b/drivers/gpu/drm/drm_dp_dual_mode_helper.c
>>>> index 80e62f6..09bf962 100644
>>>> --- a/drivers/gpu/drm/drm_dp_dual_mode_helper.c
>>>> +++ b/drivers/gpu/drm/drm_dp_dual_mode_helper.c
>>>> @@ -75,8 +75,16 @@ ssize_t drm_dp_dual_mode_read(struct i2c_adapter *adapter,
>>>>    		},
>>>>    	};
>>>>    	int ret;
>>>> +	int retry;
>>>> +
>>>> +	for (retry = 0; retry < 6; retry++) {
>>>> +		if (retry)
>>>> +			usleep_range(500, 1000);
>>>> +		ret = i2c_transfer(adapter, msgs, ARRAY_SIZE(msgs));
>>>> +		if (ret > 0)
>>> I think the ret == 0 case should be handled with the rest of
>>> partial-read cases (which we check against further down). So the above
>>> should be
>>> 		if (ret >= 0)
>> My thinking for this case was, that, partial error cases also should be
>> given retries, and try again.
> Well, that would've been then
>
> 		if (ret == ARRAY_SIZE(msgs))
> 			break;
>
> then, not the way you have written no?
>
> But I don't think we shold do that. We only retry here to wait for the
> HW to wake up. Any other error afterwards is really unexpected. This is
> also how the __i2c_transfer() retry logic works.
Ok then, I will add this part.
Shashank
>> IN this way, there is a possibility that in the next attempt there would be
>> proper read.
>> but if after the retries too, we see partial read, then we should call it an
>> error.
>>
>> Does it sound like a good thing to do ?
>> - Shashank
>>> With this fixed it looks ok to me, so fwiw:
>>> Reviewed-by: Imre Deak <imre.deak at intel.com>
>>>
>>> Note that ideally we wouldn't retry for error returns like
>>> -ENOTSUPPORTED, -ENOMEM, -E2BIG, but it's not a big deal to do that
>>> either. One possibility to solving that and also making this workaround
>>> more generic is to reuse the existing retry logic in __i2c_transfer() by
>>> making it retry in all relevant error cases (timeout/IO) and setting
>>> i2c_adapter::retries for the LSPCON adaptor. But that can be done as a
>>> follow-up.
>>>
>>>> +			break;
>>>> +	}
>>>> -	ret = i2c_transfer(adapter, msgs, ARRAY_SIZE(msgs));
>>>>    	if (ret < 0)
>>>>    		return ret;
>>>>    	if (ret != ARRAY_SIZE(msgs))
>>>> @@ -420,7 +428,7 @@ int drm_lspcon_get_mode(struct i2c_adapter *adapter,
>>>>    	ret = drm_dp_dual_mode_read(adapter, DP_DUAL_MODE_LSPCON_CURRENT_MODE,
>>>>    				    &data, sizeof(data));
>>>>    	if (ret < 0) {
>>>> -		DRM_ERROR("LSPCON read(0x80, 0x41) failed\n");
>>>> +		DRM_DEBUG_KMS("LSPCON read(0x80, 0x41) failed\n");
>>>>    		return -EFAULT;
>>>>    	}
>>>> -- 
>>>> 2.7.4
>>>>