Re: [PATCH V2 5/5] OPP: decouple dt properties in opp_parse_supplies()

From: James Calligeros <jcalligeros99@gmail.com>
To: viresh.kumar@linaro.org
Cc: jcalligeros99@gmail.com, linux-kernel@vger.kernel.org,
	linux-pm@vger.kernel.org, nm@ti.com, rafael@kernel.org,
	sboyd@kernel.org, vincent.guittot@linaro.org, vireshk@kernel.org
Subject: Re: [PATCH V2 5/5] OPP: decouple dt properties in opp_parse_supplies()
Date: Thu,  3 Nov 2022 22:24:25 +1000	[thread overview]
Message-ID: <3519763.iIbC2pHGDl@makoto> (raw)
In-Reply-To: <5acd93462bff6c108d65802fd39f6002dfadd1a0.1667473008.git.viresh.kumar@linaro.org>

On Thursday, 3 November 2022 9:01:08 PM AEST Viresh Kumar wrote:

> @@ -674,7 +677,7 @@ static int opp_parse_supplies(struct dev_pm_opp *opp, struct device *dev,
>  	bool triplet;
>  
>  	microvolt = opp_parse_microvolt(opp, dev, opp_table, &triplet);
> -	if (IS_ERR_OR_NULL(microvolt))
> +	if (IS_ERR(microvolt))
>  		return PTR_ERR(microvolt);

Erroring out here will still block EM registration on platforms without 
the opp-microvolt prop so we're back to square one, which means the 
patch does not do what the description says it does. It behaves
almost identically to the current code.

I assume this is an intentional choice given the comments in
opp_parse_microvolt(), so the commit message should drop 
references to fixing such platforms.

If this is a hard sticking point and opp_parse_supplies() must return
a regulator with opp-microvolt, then I am of the opinion that the parsing
of opp-microwatt should be detangled entirely from the regulator
infrastructure.

Otherwise for the whole series,

Tested-by: James Calligeros <jcalligeros99@gmail.com>

- James