Received-SPF: pass (google.com: domain of linux-kernel+bounces-183851-linux.lists.archive=gmail.com@vger.kernel.org designates 147.75.199.223 as permitted sender) client-ip=147.75.199.223;
Date: Mon, 20 May 2024 15:41:17 +0100
From: Mark Brown <broonie@kernel.org>
To: Artur Weber <aweber.kernel@gmail.com>
Cc: Sylwester Nawrocki <s.nawrocki@samsung.com>,
	Krzysztof Kozlowski <krzk+dt@kernel.org>,
	Liam Girdwood <lgirdwood@gmail.com>, Rob Herring <robh@kernel.org>,
	Conor Dooley <conor+dt@kernel.org>,
	Jaroslav Kysela <perex@perex.cz>, Takashi Iwai <tiwai@suse.com>,
	Alim Akhtar <alim.akhtar@samsung.com>, alsa-devel@alsa-project.org,
	linux-sound@vger.kernel.org, devicetree@vger.kernel.org,
	linux-kernel@vger.kernel.org, linux-arm-kernel@lists.infradead.org,
	linux-samsung-soc@vger.kernel.org,
	~postmarketos/upstreaming@lists.sr.ht
Subject: Re: [PATCH v3 4/8] ASoC: samsung: midas_wm1811: Add GPIO-based
 headset jack detection
Message-ID: <2c4c0aeb-2870-4905-bcfc-642ae40c87f1@sirena.org.uk>
References: <20240519-midas-wm1811-gpio-jack-v3-0-0c1736144c0e@gmail.com>
 <20240519-midas-wm1811-gpio-jack-v3-4-0c1736144c0e@gmail.com>
Precedence: bulk
MIME-Version: 1.0
Content-Type: multipart/signed; micalg=pgp-sha512;
	protocol="application/pgp-signature"; boundary="lfTGNXXe0eUJp9Hl"
Content-Disposition: inline
In-Reply-To: <20240519-midas-wm1811-gpio-jack-v3-4-0c1736144c0e@gmail.com>


--lfTGNXXe0eUJp9Hl
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline

On Sun, May 19, 2024 at 10:17:50AM +0200, Artur Weber wrote:

> +	if (priv->reg_headset_mic_bias) {
> +		/*
> +		 * Get state of Headset Mic switch by checking the headset mic
> +		 * bias regulator
> +		 */
> +		bias_already_on = \
> +			regulator_is_enabled(priv->reg_headset_mic_bias);

If you're going to do this you need to use regulator_get_exclusive() to
ensure that nothing else can enable the regulator, otherwise you should
just unconditionally enable the regulator.

> +	/*
> +	 * Revert the headset mic bias supply to its previous state
> +	 * (i.e. if it was disabled before the check, disable it again)
> +	 */
> +	if (priv->reg_headset_mic_bias && bias_already_on == 0) {
> +		ret = regulator_disable(priv->reg_headset_mic_bias);
> +		if (ret)
> +			pr_err("%s: Failed to disable micbias: %d\n",
> +			       __func__, ret);
> +	}

Given that you're just briefly bouncing the regulator on and off it'd be
best to just unconditionally enable and disable here, I can't see what
the enable check gains you other than possible race conditions.

--lfTGNXXe0eUJp9Hl
Content-Type: application/pgp-signature; name="signature.asc"

-----BEGIN PGP SIGNATURE-----

iQEzBAABCgAdFiEEreZoqmdXGLWf4p/qJNaLcl1Uh9AFAmZLYQwACgkQJNaLcl1U
h9Dwjgf/WnkaIdmy+K4GxTBYERrrATUvQAIiQ6PjAPC/Gf9ZYNuIgC/EEe6k4Nlw
WkBAfRhzMebnNCagnadq8ZyPNKp5zUVpGo2/r0QOvrJ3Ajk3RD7e18nKipsWpb7D
DFrusOYqlUdIRUjWCPn/u6WWRojAH9UhnF7XiuL/NSnz9F/g0xvd8wyWa2a5+LjK
NrAKc2Xd0jpBvcTGmTzS+hgFW+xr10HGoG6nutfh9Gm3hi7b9WQ2dl5eo5rGzMqO
kJNvPmQA6kFPcJkCtkLFSgEARq7edL6c+l3rIr42nORS6PtmclS/vRVlkIvXBHyZ
CtJveZ2pcrft4S204mzZk+4vNymYzg==
=EEPv
-----END PGP SIGNATURE-----

--lfTGNXXe0eUJp9Hl--