[BUG] iio: imu: st_lsm6dsx: read error when resuming from suspend

[BUG] iio: imu: st_lsm6dsx: read error when resuming from suspend

[Sean Nyekjaer]

Hi Lorenzo,

I'm seeing this error sometimes when resuming:
st_lsm6dsx_i2c 1-006a: failed to read fifo status (err=-13)

I doesn't matter if it's the ism330 or another wake source.
I have not enabled fifo readings.

I have tried the "[PATCH 00/13] various st_lsm6dsx fixes and missing 
bits​" same results.

The regmap_read in st_lsm6dsx_handler_thread for the event seems to be 
reading something as the event is reported correctly.

Later events are reported correctly...

/Sean

Re: [BUG] iio: imu: st_lsm6dsx: read error when resuming from suspend

[Lorenzo Bianconi]

[-- Attachment #1: Type: text/plain, Size: 809 bytes --]

> Hi Lorenzo,
> 

Hi Sean,

> I'm seeing this error sometimes when resuming:
> st_lsm6dsx_i2c 1-006a: failed to read fifo status (err=-13)

it seems the error is directly reported by regmap/i2c driver you are running,
it is not related to the last series I sent.

> 
> I doesn't matter if it's the ism330 or another wake source.
> I have not enabled fifo readings.
> 
> I have tried the "[PATCH 00/13] various st_lsm6dsx fixes and missing bits​"
> same results.
> 
> The regmap_read in st_lsm6dsx_handler_thread for the event seems to be
> reading something as the event is reported correctly.
> 
> Later events are reported correctly...

So IIUC the system just reports a sporadic error sometimes during resume but it
is able to recover, right?

Regards,
Lorenzo

> 
> /Sean

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 228 bytes --]

Re: [BUG] iio: imu: st_lsm6dsx: read error when resuming from suspend

[Sean Nyekjaer]

On 07/10/2019 12.34, Lorenzo Bianconi wrote:
>> Hi Lorenzo,
>>
> 
> Hi Sean,
> 
>> I'm seeing this error sometimes when resuming:
>> st_lsm6dsx_i2c 1-006a: failed to read fifo status (err=-13)
> 
> it seems the error is directly reported by regmap/i2c driver you are running,
> it is not related to the last series I sent.
> 
No not related :) The same error with or without the series ;)
>>
>> I doesn't matter if it's the ism330 or another wake source.
>> I have not enabled fifo readings.
>>
>> I have tried the "[PATCH 00/13] various st_lsm6dsx fixes and missing bits​"
>> same results.
>>
>> The regmap_read in st_lsm6dsx_handler_thread for the event seems to be
>> reading something as the event is reported correctly.
>>
>> Later events are reported correctly...
> 
> So IIUC the system just reports a sporadic error sometimes during resume but it
> is able to recover, right?
> Yes, it's sporadic and suspend/resume sometimes does this... and we are 
able to recover.
I just find it weird that regmap_read goes well and the next call to 
regmap_bulk_read (st_lsm6dsx_read_locked) gets "permission denied".

So all in all, just a heads up. Maybe we can ignore it for now :)

/Sean