[PATCH 0/3] Update some topology matching for newer laptops

[PATCH 0/3] Update some topology matching for newer laptops

[Charles Keepax]

Now function topologies are being used more routinely we are hitting a
couple of issues on some new laptops. Firstly, several laptops ship with
devices in the ACPI which don't physically exist in the laptop, this
confuses the function topology code as it tries to add those into the
card. Add some quirks to ignore these. Secondly, remove some unnecessary
matches and quirk the ones that remain so they only match the laptops
that require them.

Thanks,
Charles

Charles Keepax (3):
  ASoC: soc-acpi-intel-ptl-match: Make Chrome matches conditional
  ASoC: Intel: soc-acpi-intel-ptl-match: Remove unnecessary cs42l43
    match
  soundwire: dmi-quirks: Disable ghost Realtek devices

 drivers/soundwire/dmi-quirks.c                |  35 ++++++
 .../intel/common/soc-acpi-intel-ptl-match.c   | 119 +-----------------
 .../intel/common/soc-acpi-intel-sdca-quirks.c |  16 +++
 .../intel/common/soc-acpi-intel-sdca-quirks.h |   1 +
 4 files changed, 53 insertions(+), 118 deletions(-)

-- 
2.47.3

[PATCH 1/3] ASoC: soc-acpi-intel-ptl-match: Make Chrome matches conditional

[Charles Keepax]

For PTL onwards Cirrus are intending to rely on function
topologies, rather than using a match table for each system
type. Chrome systems tend to have custom magic in the topology
and thus need to load a specific file. This causes problems as
these system can have the same layout as generic laptops causing
the match to apply to other laptops. Add a DMI quirk that forces
these matches to only apply to specific devices.

Signed-off-by: Charles Keepax <ckeepax@opensource.cirrus.com>
---
 .../soc/intel/common/soc-acpi-intel-ptl-match.c  |  1 +
 .../intel/common/soc-acpi-intel-sdca-quirks.c    | 16 ++++++++++++++++
 .../intel/common/soc-acpi-intel-sdca-quirks.h    |  1 +
 3 files changed, 18 insertions(+)

diff --git a/sound/soc/intel/common/soc-acpi-intel-ptl-match.c b/sound/soc/intel/common/soc-acpi-intel-ptl-match.c
index ad3af8834e431..c6bf70e393975 100644
--- a/sound/soc/intel/common/soc-acpi-intel-ptl-match.c
+++ b/sound/soc/intel/common/soc-acpi-intel-ptl-match.c
@@ -632,6 +632,7 @@ struct snd_soc_acpi_mach snd_soc_acpi_intel_ptl_sdw_machines[] = {
 		.link_mask = BIT(2) | BIT(3),
 		.links = ptl_cs42l43_agg_l3_cs35l56_l2,
 		.drv_name = "sof_sdw",
+		.machine_check = snd_soc_acpi_intel_no_function_topology,
 		.sof_tplg_filename = "sof-ptl-cs42l43-agg-l3-cs35l56-l2.tplg",
 	},
 	{
diff --git a/sound/soc/intel/common/soc-acpi-intel-sdca-quirks.c b/sound/soc/intel/common/soc-acpi-intel-sdca-quirks.c
index 3eaa058f84608..7caabc501b163 100644
--- a/sound/soc/intel/common/soc-acpi-intel-sdca-quirks.c
+++ b/sound/soc/intel/common/soc-acpi-intel-sdca-quirks.c
@@ -6,6 +6,7 @@
  *
  */
 
+#include <linux/dmi.h>
 #include <linux/soundwire/sdw_intel.h>
 #include <sound/sdca.h>
 #include <sound/soc-acpi.h>
@@ -37,6 +38,21 @@ bool snd_soc_acpi_intel_sdca_is_device_rt712_vb(void *arg)
 }
 EXPORT_SYMBOL_NS(snd_soc_acpi_intel_sdca_is_device_rt712_vb, "SND_SOC_ACPI_INTEL_SDCA_QUIRKS");
 
+static const struct dmi_system_id function_topology_quirk_table[] = {
+	{
+		.matches = {
+			DMI_MATCH(DMI_SYS_VENDOR, "Google"),
+		},
+	},
+	{}
+};
+
+bool snd_soc_acpi_intel_no_function_topology(void *arg)
+{
+	return !!dmi_check_system(function_topology_quirk_table);
+}
+EXPORT_SYMBOL_NS(snd_soc_acpi_intel_no_function_topology, "SND_SOC_ACPI_INTEL_SDCA_QUIRKS");
+
 MODULE_DESCRIPTION("ASoC ACPI Intel SDCA quirks");
 MODULE_LICENSE("GPL");
 MODULE_IMPORT_NS("SND_SOC_SDCA");
diff --git a/sound/soc/intel/common/soc-acpi-intel-sdca-quirks.h b/sound/soc/intel/common/soc-acpi-intel-sdca-quirks.h
index bead5ec6243f9..2ea0a1881c4bd 100644
--- a/sound/soc/intel/common/soc-acpi-intel-sdca-quirks.h
+++ b/sound/soc/intel/common/soc-acpi-intel-sdca-quirks.h
@@ -10,5 +10,6 @@
 #define _SND_SOC_ACPI_INTEL_SDCA_QUIRKS
 
 bool snd_soc_acpi_intel_sdca_is_device_rt712_vb(void *arg);
+bool snd_soc_acpi_intel_no_function_topology(void *arg);
 
 #endif
-- 
2.47.3

[PATCH 2/3] ASoC: Intel: soc-acpi-intel-ptl-match: Remove unnecessary cs42l43 match

[Charles Keepax]

For PTL onwards Cirrus are intending to rely on function topologies,
rather than using a match table for each system type. Remove this
unnecessary match table entry. Having the match entries can
mean that systems match when they should use function topologies
instead, resulting in incorrect audio configurations. Although,
admittedly this is not too likely with this 6x amp configuration
as those are quite rare, but best to follow best practice.

Signed-off-by: Charles Keepax <ckeepax@opensource.cirrus.com>
---
 .../intel/common/soc-acpi-intel-ptl-match.c   | 118 ------------------
 1 file changed, 118 deletions(-)

diff --git a/sound/soc/intel/common/soc-acpi-intel-ptl-match.c b/sound/soc/intel/common/soc-acpi-intel-ptl-match.c
index c6bf70e393975..f7694b2a2b02e 100644
--- a/sound/soc/intel/common/soc-acpi-intel-ptl-match.c
+++ b/sound/soc/intel/common/soc-acpi-intel-ptl-match.c
@@ -92,48 +92,6 @@ static const struct snd_soc_acpi_endpoint spk_r_endpoint = {
 	.group_id = 1,
 };
 
-static const struct snd_soc_acpi_endpoint spk_1_endpoint = {
-	.num = 0,
-	.aggregated = 1,
-	.group_position = 1,
-	.group_id = 1,
-};
-
-static const struct snd_soc_acpi_endpoint spk_2_endpoint = {
-	.num = 0,
-	.aggregated = 1,
-	.group_position = 2,
-	.group_id = 1,
-};
-
-static const struct snd_soc_acpi_endpoint spk_3_endpoint = {
-	.num = 0,
-	.aggregated = 1,
-	.group_position = 3,
-	.group_id = 1,
-};
-
-static const struct snd_soc_acpi_endpoint spk_4_endpoint = {
-	.num = 0,
-	.aggregated = 1,
-	.group_position = 4,
-	.group_id = 1,
-};
-
-static const struct snd_soc_acpi_endpoint spk_5_endpoint = {
-	.num = 0,
-	.aggregated = 1,
-	.group_position = 5,
-	.group_id = 1,
-};
-
-static const struct snd_soc_acpi_endpoint spk_6_endpoint = {
-	.num = 0,
-	.aggregated = 1,
-	.group_position = 6,
-	.group_id = 1,
-};
-
 static const struct snd_soc_acpi_endpoint jack_dmic_endpoints[] = {
 	/* Jack Endpoint */
 	{
@@ -202,15 +160,6 @@ static const struct snd_soc_acpi_endpoint cs42l43_amp_spkagg_endpoints[] = {
 	},
 };
 
-static const struct snd_soc_acpi_adr_device cs42l43_2_adr[] = {
-	{
-		.adr = 0x00023001fa424301ull,
-		.num_endpoints = ARRAY_SIZE(cs42l43_amp_spkagg_endpoints),
-		.endpoints = cs42l43_amp_spkagg_endpoints,
-		.name_prefix = "cs42l43"
-	}
-};
-
 static const struct snd_soc_acpi_adr_device cs42l43_3_agg_adr[] = {
 	{
 		.adr = 0x00033001FA424301ull,
@@ -235,48 +184,6 @@ static const struct snd_soc_acpi_adr_device cs35l56_2_lr_adr[] = {
 	}
 };
 
-static const struct snd_soc_acpi_adr_device cs35l56_1_3amp_adr[] = {
-	{
-		.adr = 0x00013001fa355601ull,
-		.num_endpoints = 1,
-		.endpoints = &spk_1_endpoint,
-		.name_prefix = "AMP1"
-	},
-	{
-		.adr = 0x00013101fa355601ull,
-		.num_endpoints = 1,
-		.endpoints = &spk_2_endpoint,
-		.name_prefix = "AMP2"
-	},
-	{
-		.adr = 0x00013201fa355601ull,
-		.num_endpoints = 1,
-		.endpoints = &spk_3_endpoint,
-		.name_prefix = "AMP3"
-	}
-};
-
-static const struct snd_soc_acpi_adr_device cs35l56_3_3amp_adr[] = {
-	{
-		.adr = 0x00033301fa355601ull,
-		.num_endpoints = 1,
-		.endpoints = &spk_4_endpoint,
-		.name_prefix = "AMP4"
-	},
-	{
-		.adr = 0x00033401fa355601ull,
-		.num_endpoints = 1,
-		.endpoints = &spk_5_endpoint,
-		.name_prefix = "AMP5"
-	},
-	{
-		.adr = 0x00033501fa355601ull,
-		.num_endpoints = 1,
-		.endpoints = &spk_6_endpoint,
-		.name_prefix = "AMP6"
-	}
-};
-
 static const struct snd_soc_acpi_adr_device rt711_sdca_0_adr[] = {
 	{
 		.adr = 0x000030025D071101ull,
@@ -408,25 +315,6 @@ static const struct snd_soc_acpi_link_adr ptl_cs42l43_agg_l3_cs35l56_l2[] = {
 	{}
 };
 
-static const struct snd_soc_acpi_link_adr ptl_cs42l43_l2_cs35l56x6_l13[] = {
-	{
-		.mask = BIT(2),
-		.num_adr = ARRAY_SIZE(cs42l43_2_adr),
-		.adr_d = cs42l43_2_adr,
-	},
-	{
-		.mask = BIT(1),
-		.num_adr = ARRAY_SIZE(cs35l56_1_3amp_adr),
-		.adr_d = cs35l56_1_3amp_adr,
-	},
-	{
-		.mask = BIT(3),
-		.num_adr = ARRAY_SIZE(cs35l56_3_3amp_adr),
-		.adr_d = cs35l56_3_3amp_adr,
-	},
-	{}
-};
-
 static const struct snd_soc_acpi_link_adr ptl_rt722_l0_rt1320_l23[] = {
 	{
 		.mask = BIT(0),
@@ -599,12 +487,6 @@ struct snd_soc_acpi_mach snd_soc_acpi_intel_ptl_sdw_machines[] = {
 		.sof_tplg_filename = "sof-ptl-rt713-l3-rt1320-l1.tplg",
 		.get_function_tplg_files = sof_sdw_get_tplg_files,
 	},
-	{
-		.link_mask = BIT(1) | BIT(2) | BIT(3),
-		.links = ptl_cs42l43_l2_cs35l56x6_l13,
-		.drv_name = "sof_sdw",
-		.sof_tplg_filename = "sof-ptl-cs42l43-l2-cs35l56x6-l13.tplg",
-	},
 	{
 		.link_mask = BIT(0) | BIT(2) | BIT(3),
 		.links = ptl_rt722_l0_rt1320_l23,
-- 
2.47.3

[PATCH 3/3] soundwire: dmi-quirks: Disable ghost Realtek devices

[Charles Keepax]

Many systems ship with a Realtek audio codec in the ACPI that doesn't
physically exist in the system. This confuses the newer function
topology system that creates the soundcard, as it builds the card based
on the ACPI information.

Whilst we are working with the laptop vendors to try and stop this
happening there are quite a few systems where this has shipped. Add a
quirk to disable this "ghost" device.

Currently this patch should cover:
 - Asus UX5406AA
 - Lenovo Yoga Pro 9i (83SF)
 - Lenovo Yoga Slim 7 Ultra (83QK)

Signed-off-by: Charles Keepax <ckeepax@opensource.cirrus.com>
---
 drivers/soundwire/dmi-quirks.c | 35 ++++++++++++++++++++++++++++++++++
 1 file changed, 35 insertions(+)

diff --git a/drivers/soundwire/dmi-quirks.c b/drivers/soundwire/dmi-quirks.c
index 5854218e1a274..32a46a2d90f7c 100644
--- a/drivers/soundwire/dmi-quirks.c
+++ b/drivers/soundwire/dmi-quirks.c
@@ -90,6 +90,19 @@ static const struct adr_remap intel_rooks_county[] = {
 	{}
 };
 
+/*
+ * Many platforms have ghost realtek devices in the ACPI that don't physically
+ * exist, remove those devices.
+ */
+static const struct adr_remap ghost_realtek[] = {
+	/* rt722 on link3 */
+	{
+		0x000330025d072201ull,
+		0x0000000000000000ull
+	},
+	{}
+};
+
 static const struct dmi_system_id adr_remap_quirk_table[] = {
 	/* TGL devices */
 	{
@@ -164,6 +177,28 @@ static const struct dmi_system_id adr_remap_quirk_table[] = {
 		},
 		.driver_data = (void *)hp_omen_16,
 	},
+	/* PTL devices */
+	{
+		.matches = {
+			DMI_MATCH(DMI_SYS_VENDOR, "ASUS"),
+			DMI_MATCH(DMI_BOARD_NAME, "UX5406AA"),
+		},
+		.driver_data = (void *)ghost_realtek,
+	},
+	{
+		.matches = {
+			DMI_MATCH(DMI_SYS_VENDOR, "LENOVO"),
+			DMI_MATCH(DMI_PRODUCT_NAME, "83QK"),
+		},
+		.driver_data = (void *)ghost_realtek,
+	},
+	{
+		.matches = {
+			DMI_MATCH(DMI_SYS_VENDOR, "LENOVO"),
+			DMI_MATCH(DMI_PRODUCT_NAME, "83SF"),
+		},
+		.driver_data = (void *)ghost_realtek,
+	},
 	{}
 };
 
-- 
2.47.3

Re: [PATCH 3/3] soundwire: dmi-quirks: Disable ghost Realtek devices

[Pierre-Louis Bossart]

On 5/20/26 18:36, Charles Keepax wrote:
> Many systems ship with a Realtek audio codec in the ACPI that doesn't
> physically exist in the system. This confuses the newer function
> topology system that creates the soundcard, as it builds the card based
> on the ACPI information.
> 
> Whilst we are working with the laptop vendors to try and stop this
> happening there are quite a few systems where this has shipped. Add a
> quirk to disable this "ghost" device.
> 
> Currently this patch should cover:
>  - Asus UX5406AA
>  - Lenovo Yoga Pro 9i (83SF)
>  - Lenovo Yoga Slim 7 Ultra (83QK)

if this broken ACPI stuff works for Windows then it'll be an uphill battle...

Could we filter instead those 'ghost' devices if they fail to appear in the new ACPI table that describes all the endpoints and card topology?

RE: [PATCH 3/3] soundwire: dmi-quirks: Disable ghost Realtek devices

[Liao, Bard]

> -----Original Message-----
> From: Pierre-Louis Bossart <pierre-louis.bossart@linux.dev>
> Sent: Thursday, May 21, 2026 1:17 AM
> To: Charles Keepax <ckeepax@opensource.cirrus.com>; broonie@kernel.org;
> vkoul@kernel.org
> Cc: lgirdwood@gmail.com; yung-chuan.liao@linux.intel.com;
> peter.ujfalusi@linux.intel.com; oder_chiou@realtek.com;
> jack.yu@realtek.com; shumingf@realtek.com; linux-sound@vger.kernel.org;
> linux-kernel@vger.kernel.org; patches@opensource.cirrus.com
> Subject: Re: [PATCH 3/3] soundwire: dmi-quirks: Disable ghost Realtek devices
> 
> On 5/20/26 18:36, Charles Keepax wrote:
> > Many systems ship with a Realtek audio codec in the ACPI that doesn't
> > physically exist in the system. This confuses the newer function
> > topology system that creates the soundcard, as it builds the card based
> > on the ACPI information.
> >
> > Whilst we are working with the laptop vendors to try and stop this
> > happening there are quite a few systems where this has shipped. Add a
> > quirk to disable this "ghost" device.
> >
> > Currently this patch should cover:
> >  - Asus UX5406AA
> >  - Lenovo Yoga Pro 9i (83SF)
> >  - Lenovo Yoga Slim 7 Ultra (83QK)
> 
> if this broken ACPI stuff works for Windows then it'll be an uphill battle...

We already informed the OEMs. Hope they can fix in the next generation
products.

> 
> Could we filter instead those 'ghost' devices if they fail to appear in the new
> ACPI table that describes all the endpoints and card topology?

Sorry Pierre, I am not sure if I get your question correctly.
The 'ghost' device has a complete DisCo properties description in the
ACPI table. We could check if the device is ever attached. But
we need to set a timeout time and wait until timeout to claim
the device is not present. It seems not be realistic to wait say 3
seconds in each boot.

Re: [PATCH 3/3] soundwire: dmi-quirks: Disable ghost Realtek devices

[Charles Keepax]

On Wed, May 20, 2026 at 07:16:55PM +0200, Pierre-Louis Bossart wrote:
> On 5/20/26 18:36, Charles Keepax wrote:
> Could we filter instead those 'ghost' devices if they fail to
> appear in the new ACPI table that describes all the endpoints
> and card topology?

Eventually perhaps once that makes it out of the standardisation
process (I believe the committee is looking at it), but for now
that is all Microsoft magic that I would rather avoid.

Thanks,
Charles

Re: [PATCH 3/3] soundwire: dmi-quirks: Disable ghost Realtek devices

[Pierre-Louis Bossart]

On 5/21/26 10:17, Charles Keepax wrote:
> On Wed, May 20, 2026 at 07:16:55PM +0200, Pierre-Louis Bossart wrote:
>> On 5/20/26 18:36, Charles Keepax wrote:
>> Could we filter instead those 'ghost' devices if they fail to
>> appear in the new ACPI table that describes all the endpoints
>> and card topology?
> 
> Eventually perhaps once that makes it out of the standardisation
> process (I believe the committee is looking at it), but for now
> that is all Microsoft magic that I would rather avoid.

Nothing prevents developers from reverse-engineering DSDT contents. I remember seeing all the information needed in the DSDT of one of the Dell devices.
The standardization was supposed to happen before Covid, I looked at this in 2020-2021. If it hasn't happened by now it's unlikely it ever will...
This DSDT information is interesting for Intel platforms, it could help probe a generic machine driver for the card but raises other problems with deferred probe,
c.f. https://lore.kernel.org/alsa-devel/20210817190057.255264-2-pierre-louis.bossart@linux.intel.com/

Anyways this patchset isn't wrong either, so 

Reviewed-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.dev>

Re: (subset) [PATCH 0/3] Update some topology matching for newer laptops

[Mark Brown]

On Wed, 20 May 2026 17:36:28 +0100, Charles Keepax wrote:
> Update some topology matching for newer laptops
> 
> Now function topologies are being used more routinely we are hitting a
> couple of issues on some new laptops. Firstly, several laptops ship with
> devices in the ACPI which don't physically exist in the laptop, this
> confuses the function topology code as it tries to add those into the
> card. Add some quirks to ignore these. Secondly, remove some unnecessary
> matches and quirk the ones that remain so they only match the laptops
> that require them.
> 
> [...]

Applied to

   https://git.kernel.org/pub/scm/linux/kernel/git/broonie/sound.git for-7.1

Thanks!

[1/3] ASoC: soc-acpi-intel-ptl-match: Make Chrome matches conditional
      https://git.kernel.org/broonie/sound/c/e0fb794d67f8
[2/3] ASoC: Intel: soc-acpi-intel-ptl-match: Remove unnecessary cs42l43 match
      https://git.kernel.org/broonie/sound/c/45cf24da0a10

All being well this means that it will be integrated into the linux-next
tree (usually sometime in the next 24 hours) and sent to Linus during
the next merge window (or sooner if it is a bug fix), however if
problems are discovered then the patch may be dropped or reverted.

You may get further e-mails resulting from automated or manual testing
and review of the tree, please engage with people reporting problems and
send followup patches addressing any issues that are reported if needed.

If any updates are required or you are submitting further changes they
should be sent as incremental updates against current git, existing
patches will not be replaced.

Please add any relevant lists and maintainers to the CCs when replying
to this mail.

Thanks,
Mark