[PATCH] Revert "ALSA: hda: Add ASRock X670E Taichi to denylist"

[PATCH] Revert "ALSA: hda: Add ASRock X670E Taichi to denylist"

[Mario Limonciello (AMD)]

On a motherboard with an AMD Granite Ridge CPU there is a report
that 3.5mm microphone and headphones aren't working.  In the
log it's observed:

snd_hda_intel 0000:02:00.6: Skipping the device on the denylist

This was because of commit df42ee7e22f03 ("ALSA: hda: Add ASRock
X670E Taichi to denylist").  Reverting this commit allows the
microphone and headphones to work again. As at least some combinations
of this motherboard do have applicable devices, revert so that they
can be probed.

Cc: Richard Gong <richard.gong@amd.com>
Cc: Juan Martinez <juan.martinez@amd.com>
Signed-off-by: Mario Limonciello (AMD) <superm1@kernel.org>
---
 sound/hda/controllers/intel.c | 1 -
 1 file changed, 1 deletion(-)

diff --git a/sound/hda/controllers/intel.c b/sound/hda/controllers/intel.c
index fcf67e97a546a..1bb3ff55b1151 100644
--- a/sound/hda/controllers/intel.c
+++ b/sound/hda/controllers/intel.c
@@ -2077,7 +2077,6 @@ static const struct pci_device_id driver_denylist[] = {
 	{ PCI_DEVICE_SUB(0x1022, 0x1487, 0x1043, 0x874f) }, /* ASUS ROG Zenith II / Strix */
 	{ PCI_DEVICE_SUB(0x1022, 0x1487, 0x1462, 0xcb59) }, /* MSI TRX40 Creator */
 	{ PCI_DEVICE_SUB(0x1022, 0x1487, 0x1462, 0xcb60) }, /* MSI TRX40 */
-	{ PCI_DEVICE_SUB(0x1022, 0x15e3, 0x1022, 0xd601) }, /* ASRock X670E Taichi */
 	{}
 };
 
-- 
2.43.0

Re: [PATCH] Revert "ALSA: hda: Add ASRock X670E Taichi to denylist"

[Takashi Iwai]

On Wed, 13 Aug 2025 16:03:08 +0200,
Mario Limonciello (AMD) wrote:
> 
> On a motherboard with an AMD Granite Ridge CPU there is a report
> that 3.5mm microphone and headphones aren't working.  In the
> log it's observed:
> 
> snd_hda_intel 0000:02:00.6: Skipping the device on the denylist
> 
> This was because of commit df42ee7e22f03 ("ALSA: hda: Add ASRock
> X670E Taichi to denylist").  Reverting this commit allows the
> microphone and headphones to work again. As at least some combinations
> of this motherboard do have applicable devices, revert so that they
> can be probed.
> 
> Cc: Richard Gong <richard.gong@amd.com>
> Cc: Juan Martinez <juan.martinez@amd.com>
> Signed-off-by: Mario Limonciello (AMD) <superm1@kernel.org>
> ---
>  sound/hda/controllers/intel.c | 1 -
>  1 file changed, 1 deletion(-)
> 
> diff --git a/sound/hda/controllers/intel.c b/sound/hda/controllers/intel.c
> index fcf67e97a546a..1bb3ff55b1151 100644
> --- a/sound/hda/controllers/intel.c
> +++ b/sound/hda/controllers/intel.c
> @@ -2077,7 +2077,6 @@ static const struct pci_device_id driver_denylist[] = {
>  	{ PCI_DEVICE_SUB(0x1022, 0x1487, 0x1043, 0x874f) }, /* ASUS ROG Zenith II / Strix */
>  	{ PCI_DEVICE_SUB(0x1022, 0x1487, 0x1462, 0xcb59) }, /* MSI TRX40 Creator */
>  	{ PCI_DEVICE_SUB(0x1022, 0x1487, 0x1462, 0xcb60) }, /* MSI TRX40 */
> -	{ PCI_DEVICE_SUB(0x1022, 0x15e3, 0x1022, 0xd601) }, /* ASRock X670E Taichi */

Hmm, it's a bit awkward situation.  I believe this will make the
problem of the original fix back.

But, the original bug was merely error messages and memory wastes, and
the PCI SSID looks also not specific to the mobo vendor, so we can
start as a revert, but we'd need yet another fix to cover the original
problem.
Maybe some additional DMI check.  Oh well.  Such mobo vendors should
be cursed if they don't assign the proper PCI SSIDs...


Takashi

Re: [PATCH] Revert "ALSA: hda: Add ASRock X670E Taichi to denylist"

[Mario Limonciello]

On 8/13/25 9:15 AM, Takashi Iwai wrote:
> On Wed, 13 Aug 2025 16:03:08 +0200,
> Mario Limonciello (AMD) wrote:
>>
>> On a motherboard with an AMD Granite Ridge CPU there is a report
>> that 3.5mm microphone and headphones aren't working.  In the
>> log it's observed:
>>
>> snd_hda_intel 0000:02:00.6: Skipping the device on the denylist
>>
>> This was because of commit df42ee7e22f03 ("ALSA: hda: Add ASRock
>> X670E Taichi to denylist").  Reverting this commit allows the
>> microphone and headphones to work again. As at least some combinations
>> of this motherboard do have applicable devices, revert so that they
>> can be probed.
>>
>> Cc: Richard Gong <richard.gong@amd.com>
>> Cc: Juan Martinez <juan.martinez@amd.com>
>> Signed-off-by: Mario Limonciello (AMD) <superm1@kernel.org>
>> ---
>>   sound/hda/controllers/intel.c | 1 -
>>   1 file changed, 1 deletion(-)
>>
>> diff --git a/sound/hda/controllers/intel.c b/sound/hda/controllers/intel.c
>> index fcf67e97a546a..1bb3ff55b1151 100644
>> --- a/sound/hda/controllers/intel.c
>> +++ b/sound/hda/controllers/intel.c
>> @@ -2077,7 +2077,6 @@ static const struct pci_device_id driver_denylist[] = {
>>   	{ PCI_DEVICE_SUB(0x1022, 0x1487, 0x1043, 0x874f) }, /* ASUS ROG Zenith II / Strix */
>>   	{ PCI_DEVICE_SUB(0x1022, 0x1487, 0x1462, 0xcb59) }, /* MSI TRX40 Creator */
>>   	{ PCI_DEVICE_SUB(0x1022, 0x1487, 0x1462, 0xcb60) }, /* MSI TRX40 */
>> -	{ PCI_DEVICE_SUB(0x1022, 0x15e3, 0x1022, 0xd601) }, /* ASRock X670E Taichi */
> 
> Hmm, it's a bit awkward situation.  I believe this will make the
> problem of the original fix back.
> 
> But, the original bug was merely error messages and memory wastes, and
> the PCI SSID looks also not specific to the mobo vendor, so we can
> start as a revert, but we'd need yet another fix to cover the original
> problem.
> Maybe some additional DMI check.  Oh well.  Such mobo vendors should
> be cursed if they don't assign the proper PCI SSIDs...
> 
> 
> Takashi

Yeah I thought about it too but it's fortunate original issue was just 
messages and needless probing so this seemed like a pragmatic approach.

Thanks!