Re: [PATCH] power: supply: bq24190: Fix NTC status decoding for BQ24193

[Patrick Fischer]

Hi Sebastian,

On Mon, Nov 09, 2026 at 01:08:13AM +0100, Sebastian Reichel wrote:
> Looks like bq24190_charger_get_ntc_status can just be updated to
> handle a single temperature sensor as warm/cool and dual-sensor as
> overheat/cold?

Thanks for looking at this. If I understand you correctly, codes 1-4
(one sensor out of range) would become warm/cool and codes 5/6 (both
sensors) cold/overheat, for all chips.

I don't think that is right for the BQ24190, BQ24192 and BQ24192I. Their
datasheet (SLUSAW5B) says for both TS pins "Charge suspends when either
TS pin is out of range", so a single hot or cold sensor stops charging
there. Reporting that as Warm or Cool would understate it.

The BQ24193 is different in that its TS1 and TS2 must be connected
together and its codes are a JEITA profile (SLUSBG7A): 010 is Warm and
011 is Cool, and charging continues in both states.

That is why I used the per-chip .get_ntc_status hook, like the BQ24296.
I can make it smaller, though, by handling only the two JEITA states and
falling back to the existing function for everything else:

static int bq24193_charger_get_ntc_status(u8 value)
{
	switch (value >> BQ24190_REG_F_NTC_FAULT_SHIFT & 0x7) {
	case 0x2: /* Warm */
		return POWER_SUPPLY_HEALTH_WARM;
	case 0x3: /* Cool */
		return POWER_SUPPLY_HEALTH_COOL;
	default:
		return bq24190_charger_get_ntc_status(value);
	}
}

Would that work for you, or did you have something else in mind? I'll
send a v2 once I know which way you prefer.

Greetings,

Patrick