Re: [RFC PATCH] thermal: rzg3e: Propagate reset assertion failure from suspend

[claudiu beznea <claudiu.beznea@tuxon.dev>]

Hi,

On 9/28/26 19:36, Daniel Lezcano wrote:
> On 8/28/26 11:21, Pengpeng Hou wrote:
>> rzg3e_thermal_suspend() powers off an active sensor and then ignores
>> failure to assert its reset.  System sleep can therefore continue
>> without the clean reset state expected by resume.
>>
>> Return the reset error and, when the sensor had been active, try to
>> restore power before failing suspend.  This remains an RFC because a
>> reset assertion failure can represent partial hardware state and the
>> preferred recovery policy needs platform confirmation.
>>
>> The issue was identified via static analysis and manually reviewed.
>>
>> Assisted-by: LLM
>> Signed-off-by: Pengpeng Hou <pengpeng@iscas.ac.cn>
>> ---
>>   drivers/thermal/renesas/rzg3e_thermal.c | 10 +++++++---
>>   1 file changed, 7 insertions(+), 3 deletions(-)
>>
>> diff --git a/drivers/thermal/renesas/rzg3e_thermal.c b/drivers/thermal/ 
>> renesas/rzg3e_thermal.c
>> index c44f5b8858d0..7686a45341b3 100644
>> --- a/drivers/thermal/renesas/rzg3e_thermal.c
>> +++ b/drivers/thermal/renesas/rzg3e_thermal.c
>> @@ -499,15 +499,19 @@ static int rzg3e_thermal_runtime_resume(struct device *dev)
>>   static int rzg3e_thermal_suspend(struct device *dev)
>>   {
>>       struct rzg3e_thermal_priv *priv = dev_get_drvdata(dev);
>> +    bool active = pm_runtime_active(dev);
>> +    int ret;
>>       /* If device is active, power it off */
>> -    if (pm_runtime_active(dev))
>> +    if (active)
>>           rzg3e_thermal_power_off(priv);
>>       /* Assert reset to ensure clean state after resume */
>> -    reset_control_assert(priv->rstc);
>> +    ret = reset_control_assert(priv->rstc);
>> +    if (ret && active)
>> +        rzg3e_thermal_power_on(priv);
>> -    return 0;
>> +    return ret;
>>   }
>>   static int rzg3e_thermal_resume(struct device *dev)
> 
> Sounds reasonable
> 
> John ?

John is not with Renesas anymore. I'll reply myself to this.

The patch looks OK, I consider it is the best we can do to recover from that state.

The reset driver implementation try to do its best as well to recover the reset 
line in case the assert fails [1]. However, there is no guarantee the reset 
recover will succeed. Still, this looks like the best we can do to recover from 
the reset assert failure state.

You can add:
Reviewed-by: Claudiu Beznea <claudiu.beznea.uj@bp.renesas.com>

Thank you,
Claudiu

[1] 
https://elixir.bootlin.com/linux/v7.3-rc5/source/drivers/clk/renesas/rzv2h-cpg.c#L1347