linux-kernel.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: James Calligeros <jcalligeros99@gmail.com>
To: Liam Girdwood <lgirdwood@gmail.com>,
	Mark Brown <broonie@kernel.org>,
	 Jaroslav Kysela <perex@perex.cz>, Takashi Iwai <tiwai@suse.com>,
	 Shenghao Ding <shenghao-ding@ti.com>, Kevin Lu <kevin-lu@ti.com>,
	 Baojun Xu <baojun.xu@ti.com>, Dan Murphy <dmurphy@ti.com>,
	 Rob Herring <robh@kernel.org>,
	Krzysztof Kozlowski <krzk+dt@kernel.org>,
	 Conor Dooley <conor+dt@kernel.org>,
	Shi Fu <shifu0704@thundersoft.com>,
	 Jean Delvare <jdelvare@suse.com>,
	Guenter Roeck <linux@roeck-us.net>
Cc: "Alyssa Rosenzweig" <alyssa@rosenzweig.io>,
	"Martin Povišer" <povik+lin@cutebit.org>,
	"Hector Martin" <marcan@marcan.st>,
	linux-sound@vger.kernel.org, linux-kernel@vger.kernel.org,
	devicetree@vger.kernel.org, asahi@lists.linux.dev,
	linux-hwmon@vger.kernel.org, "Neal Gompa" <neal@gompa.dev>,
	"James Calligeros" <jcalligeros99@gmail.com>
Subject: [PATCH v3 06/20] ASoC: tas2770: Fix and redo I/V sense TDM slot setting logic
Date: Thu, 27 Feb 2025 22:07:33 +1000	[thread overview]
Message-ID: <20250227-apple-codec-changes-v3-6-cbb130030acf@gmail.com> (raw)
In-Reply-To: <20250227-apple-codec-changes-v3-0-cbb130030acf@gmail.com>

From: Martin Povišer <povik+lin@cutebit.org>

The former code sets the V slot from inside set_bitwidth according to
the bitwidth of the PCM format. That's wrong, since:

 * It overrides the V slot parsed from DT binding.

 * The V slot is set shifted behind the I slot by the length of the PCM
   bitwidth, but the PCM bitwidth has no assured relation to the TDM
   slot width.

Replace the former logic by setting up the I/V sense transmission only
in case of both I/V slots being specified in devicetree, and never
override those values. In case the slots are left unspecified, disable
the transmission completely.

There's an improbable case someone is relying on the old behavior, but
if so, that's a setup that only works by accident, and cannot be sanely
supported going forward. There's no indication anyone is consuming the
I/V sense data up to today, so break the former behavior.

Reviewed-by: Neal Gompa <neal@gompa.dev>
Signed-off-by: Martin Povišer <povik+lin@cutebit.org>
Signed-off-by: James Calligeros <jcalligeros99@gmail.com>
---
 sound/soc/codecs/tas2770.c | 21 +++++++++++----------
 1 file changed, 11 insertions(+), 10 deletions(-)

diff --git a/sound/soc/codecs/tas2770.c b/sound/soc/codecs/tas2770.c
index 4a86fcec0c1ecab45b9ebcdfc90c78f3290c6fa6..7f219df8be7046912bf3ef452f75c17b5118bcf6 100644
--- a/sound/soc/codecs/tas2770.c
+++ b/sound/soc/codecs/tas2770.c
@@ -224,19 +224,16 @@ static int tas2770_set_bitwidth(struct tas2770_priv *tas2770, int bitwidth)
 		ret = snd_soc_component_update_bits(component, TAS2770_TDM_CFG_REG2,
 						    TAS2770_TDM_CFG_REG2_RXW_MASK,
 						    TAS2770_TDM_CFG_REG2_RXW_16BITS);
-		tas2770->v_sense_slot = tas2770->i_sense_slot + 2;
 		break;
 	case SNDRV_PCM_FORMAT_S24_LE:
 		ret = snd_soc_component_update_bits(component, TAS2770_TDM_CFG_REG2,
 						    TAS2770_TDM_CFG_REG2_RXW_MASK,
 						    TAS2770_TDM_CFG_REG2_RXW_24BITS);
-		tas2770->v_sense_slot = tas2770->i_sense_slot + 4;
 		break;
 	case SNDRV_PCM_FORMAT_S32_LE:
 		ret = snd_soc_component_update_bits(component, TAS2770_TDM_CFG_REG2,
 						    TAS2770_TDM_CFG_REG2_RXW_MASK,
 						    TAS2770_TDM_CFG_REG2_RXW_32BITS);
-		tas2770->v_sense_slot = tas2770->i_sense_slot + 4;
 		break;
 
 	default:
@@ -246,11 +243,6 @@ static int tas2770_set_bitwidth(struct tas2770_priv *tas2770, int bitwidth)
 	if (ret < 0)
 		return ret;
 
-	ret = tas2770_set_ivsense_transmit(tas2770, tas2770->i_sense_slot,
-					   tas2770->v_sense_slot);
-	if (ret < 0)
-		return ret;
-
 	return 0;
 }
 
@@ -505,6 +497,7 @@ static int tas2770_codec_probe(struct snd_soc_component *component)
 {
 	struct tas2770_priv *tas2770 =
 			snd_soc_component_get_drvdata(component);
+	int ret;
 
 	tas2770->component = component;
 
@@ -516,6 +509,14 @@ static int tas2770_codec_probe(struct snd_soc_component *component)
 	tas2770_reset(tas2770);
 	regmap_reinit_cache(tas2770->regmap, &tas2770_i2c_regmap);
 
+	if (tas2770->i_sense_slot != -1 && tas2770->v_sense_slot != -1) {
+		ret = tas2770_set_ivsense_transmit(tas2770, tas2770->i_sense_slot,
+						   tas2770->v_sense_slot);
+
+		if (ret < 0)
+			return ret;
+	}
+
 	return 0;
 }
 
@@ -643,7 +644,7 @@ static int tas2770_parse_dt(struct device *dev, struct tas2770_priv *tas2770)
 		dev_info(tas2770->dev, "Property %s is missing setting default slot\n",
 			 "ti,imon-slot-no");
 
-		tas2770->i_sense_slot = 0;
+		tas2770->i_sense_slot = -1;
 	}
 
 	rc = fwnode_property_read_u32(dev->fwnode, "ti,vmon-slot-no",
@@ -652,7 +653,7 @@ static int tas2770_parse_dt(struct device *dev, struct tas2770_priv *tas2770)
 		dev_info(tas2770->dev, "Property %s is missing setting default slot\n",
 			 "ti,vmon-slot-no");
 
-		tas2770->v_sense_slot = 2;
+		tas2770->v_sense_slot = -1;
 	}
 
 	tas2770->sdz_gpio = devm_gpiod_get_optional(dev, "shutdown", GPIOD_OUT_HIGH);

-- 
2.48.1


  parent reply	other threads:[~2025-02-27 12:09 UTC|newest]

Thread overview: 40+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2025-02-27 12:07 [PATCH v3 00/20] ASoC: tas27{64,70}: improve support for Apple codec variants James Calligeros
2025-02-27 12:07 ` [PATCH v3 01/20] ASoC: dt-bindings: tas27xx: add compatible for SN012776 James Calligeros
2025-02-27 12:07 ` [PATCH v3 02/20] ASoC: dt-bindings: tas2770: add compatible for TAS5770L James Calligeros
2025-02-27 12:07 ` [PATCH v3 03/20] ASoC: tas2764: Extend driver to SN012776 James Calligeros
2025-02-27 12:07 ` [PATCH v3 04/20] ASoC: tas2764: Add control concerning overcurrent events James Calligeros
2025-02-27 12:07 ` [PATCH v3 05/20] ASoC: tas2770: Factor out set_ivsense_slots James Calligeros
2025-02-27 12:07 ` James Calligeros [this message]
2025-02-27 12:07 ` [PATCH v3 07/20] ASoC: tas2764: Reinit cache on part reset James Calligeros
2025-02-27 14:09   ` Mark Brown
2025-02-27 12:07 ` [PATCH v3 08/20] ASoC: dt-bindings: tas27xx: document ti,sdout-force-zero-mask property James Calligeros
2025-02-28 20:22   ` Rob Herring (Arm)
2025-02-27 12:07 ` [PATCH v3 09/20] ASoC: tas2764: Configure zeroing of SDOUT slots James Calligeros
2025-02-27 12:07 ` [PATCH v3 10/20] ASoC: tas2764: Apply Apple quirks James Calligeros
2025-02-27 12:07 ` [PATCH v3 11/20] ASoC: tas2764: Raise regmap range maximum James Calligeros
2025-02-27 12:07 ` [PATCH v3 12/20] ASoC: tas2770: expose die temp to hwmon James Calligeros
2025-02-27 12:45   ` Guenter Roeck
2025-02-27 14:04     ` Mark Brown
2025-02-27 12:07 ` [PATCH v3 13/20] ASoC: tas2764: " James Calligeros
2025-02-27 12:07 ` [PATCH v3 14/20] ASoC: tas2764: Crop SDOUT zero-out mask based on BCLK ratio James Calligeros
2025-02-27 16:22   ` Mark Brown
2025-02-27 12:07 ` [PATCH v3 15/20] ASoC: tas2764: Enable main IRQs James Calligeros
2025-02-27 16:23   ` Mark Brown
2025-02-27 21:58     ` James Calligeros
2025-02-28 13:53       ` Mark Brown
2025-02-27 12:07 ` [PATCH v3 16/20] ASoC: tas2770: Power cycle amp on ISENSE/VSENSE change James Calligeros
2025-02-27 12:07 ` [PATCH v3 17/20] ASoC: dt-bindings: tas2770: add flags for SDOUT pulldown and zero-fill James Calligeros
2025-03-04 13:50   ` Rob Herring
2025-03-05  1:19     ` James Calligeros
2025-03-05 13:22       ` Rob Herring
2025-03-07  6:18         ` James Calligeros
2025-03-07 20:51           ` Rob Herring
2025-03-10  9:30             ` James Calligeros
2025-03-12 12:58               ` Rob Herring
2025-03-13  0:49                 ` James Calligeros
2025-03-14 13:23                 ` Martin Povišer
2025-02-27 12:07 ` [PATCH v3 18/20] ASoC: tas2770: Add zero-fill and pull-down controls James Calligeros
2025-02-27 12:07 ` [PATCH v3 19/20] ASoC: tas2770: Support setting the PDM TX slot James Calligeros
2025-02-27 12:07 ` [PATCH v3 20/20] ASoC: tas2770: Set the SDOUT polarity correctly James Calligeros
2025-03-14 15:52 ` (subset) [PATCH v3 00/20] ASoC: tas27{64,70}: improve support for Apple codec variants Mark Brown
2025-04-14 13:56 ` Mark Brown

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20250227-apple-codec-changes-v3-6-cbb130030acf@gmail.com \
    --to=jcalligeros99@gmail.com \
    --cc=alyssa@rosenzweig.io \
    --cc=asahi@lists.linux.dev \
    --cc=baojun.xu@ti.com \
    --cc=broonie@kernel.org \
    --cc=conor+dt@kernel.org \
    --cc=devicetree@vger.kernel.org \
    --cc=dmurphy@ti.com \
    --cc=jdelvare@suse.com \
    --cc=kevin-lu@ti.com \
    --cc=krzk+dt@kernel.org \
    --cc=lgirdwood@gmail.com \
    --cc=linux-hwmon@vger.kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-sound@vger.kernel.org \
    --cc=linux@roeck-us.net \
    --cc=marcan@marcan.st \
    --cc=neal@gompa.dev \
    --cc=perex@perex.cz \
    --cc=povik+lin@cutebit.org \
    --cc=robh@kernel.org \
    --cc=shenghao-ding@ti.com \
    --cc=shifu0704@thundersoft.com \
    --cc=tiwai@suse.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).