[PATCH] ALSA: hda/realtek: Match Yoga Pro 9 16IAH10 quirk by codec SSID

[PATCH] ALSA: hda/realtek: Match Yoga Pro 9 16IAH10 quirk by codec SSID

[Matt Barr]

The PCI SSID quirk added for the Lenovo Yoga Pro 9 16IAH10 (PCI SSID
17aa:3846, codec SSID 17aa:3920) also matches every other machine that
shares PCI SSID 17aa:3846. snd_hda_pick_fixup() tries all PCI SSID entries
before it falls back to the codec SSID, so those machines now get
ALC287_FIXUP_YOGA9_SPEAKER2_TO_DAC1 and never reach their own
codec-SSID quirk.

The Lenovo Legion Pro 7 16IRX8H (product 82WQ) is one such machine: PCI
SSID 17aa:3846, codec SSID 17aa:3884. Before this quirk it picked
ALC287_FIXUP_TAS2781_I2C via the 17aa:3884 codec SSID fallback and bound
its TAS2781 amplifier. With it, the codec never binds the amplifier and
the internal speakers are silent:

  snd_hda_codec_alc269 hdaudioC0D0: ALC287: picked fixup  for PCI SSID 17aa:3846

versus, on a kernel without the quirk:

  snd_hda_codec_alc269 hdaudioC0D0: ALC287: picked fixup  for codec SSID 17aa:3884
  snd_hda_codec_alc269 hdaudioC0D0: bound i2c-TIAS2781:00 (ops tas2781_hda_comp_ops ...)

Match the Yoga Pro 9 16IAH10 by its codec SSID instead, as was done for
the Yoga Pro 7 14IMH9 / Legion 7 16ACHG6 and Yoga Slim 7 14AKP10 / Legion
Slim 7 16IRH8 collisions. The Yoga Pro 9 16IAH10 still matches, through
codec SSID 17aa:3920. That codec SSID is also used by the Yoga S990-16
entry further down, which will now get the same DAC routing fix. That
looks like the same product, but I could not verify it.

Fixes: 41d60cbfde10 ("ALSA: hda/realtek: Fix bass speaker DAC routing for Lenovo Yoga Pro 9 16IAH10")
Link: https://bugzilla.kernel.org/show_bug.cgi?id=220540
Signed-off-by: Matt Barr <matthewjaybarr@gmail.com>
---
--- a/sound/hda/codecs/realtek/alc269.c
+++ b/sound/hda/codecs/realtek/alc269.c
@@ -8164,7 +8164,10 @@
 	SND_PCI_QUIRK(0x17aa, 0x3834, "Lenovo IdeaPad Slim 9i 14ITL5", ALC287_FIXUP_YOGA7_14ITL_SPEAKERS),
 	SND_PCI_QUIRK(0x17aa, 0x383d, "Legion Y9000X 2019", ALC285_FIXUP_LEGION_Y9000X_SPEAKERS),
 	SND_PCI_QUIRK(0x17aa, 0x3843, "Lenovo Yoga 9i / Yoga Book 9i", ALC287_FIXUP_LENOVO_YOGA_BOOK_9I),
-	SND_PCI_QUIRK(0x17aa, 0x3846, "Lenovo Yoga Pro 9 16IAH10", ALC287_FIXUP_YOGA9_SPEAKER2_TO_DAC1),
+	/* Legion Pro 7 16IRX8H shares PCI SSID 17aa:3846 with Yoga Pro 9 16IAH10;
+	 * use codec SSID to distinguish them
+	 */
+	HDA_CODEC_QUIRK(0x17aa, 0x3920, "Lenovo Yoga Pro 9 16IAH10", ALC287_FIXUP_YOGA9_SPEAKER2_TO_DAC1),
 	/* Yoga Pro 7 14IMH9 shares PCI SSID 17aa:3847 with Legion 7 16ACHG6;
 	 * use codec SSID to distinguish them
 	 */

Re: [PATCH] ALSA: hda/realtek: Match Yoga Pro 9 16IAH10 quirk by codec SSID

[Zhang Heng]

Thanks for the patch.
I agree this needs to be fixed, since it affects other sound cards.

I think it might be cleaner to reuse the existing quirk and drop the 
separate entry, something like this:

--- a/sound/hda/codecs/realtek/alc269.c
+++ b/sound/hda/codecs/realtek/alc269.c
@@ -8214,7 +8214,6 @@ static const struct hda_quirk alc269_fixup_tbl[] = {
         SND_PCI_QUIRK(0x17aa, 0x3834, "Lenovo IdeaPad Slim 9i 14ITL5", 
ALC287_FIXUP_YOGA7_14ITL_SPEAKERS),
         SND_PCI_QUIRK(0x17aa, 0x383d, "Legion Y9000X 2019", 
ALC285_FIXUP_LEGION_Y9000X_SPEAKERS),
         SND_PCI_QUIRK(0x17aa, 0x3843, "Lenovo Yoga 9i / Yoga Book 9i", 
ALC287_FIXUP_LENOVO_YOGA_BOOK_9I),
-       SND_PCI_QUIRK(0x17aa, 0x3846, "Lenovo Yoga Pro 9 16IAH10", 
ALC287_FIXUP_YOGA9_SPEAKER2_TO_DAC1),
         /* Yoga Pro 7 14IMH9 shares PCI SSID 17aa:3847 with Legion 7 
16ACHG6;
          * use codec SSID to distinguish them
          */
@@ -8308,7 +8307,7 @@ static const struct hda_quirk alc269_fixup_tbl[] = {
         SND_PCI_QUIRK(0x17aa, 0x3912, "Lenovo Xiaoxin 14 GT", 
ALC287_FIXUP_YOGA9_14IAP7_BASS_SPK_PIN),
         SND_PCI_QUIRK(0x17aa, 0x3913, "Lenovo 145", 
ALC236_FIXUP_LENOVO_INV_DMIC),
         SND_PCI_QUIRK(0x17aa, 0x391f, "Yoga S990-16 pro Quad YC Quad", 
ALC287_FIXUP_TXNW2781_I2C),
-       SND_PCI_QUIRK(0x17aa, 0x3920, "Yoga S990-16 pro Quad VECO Quad", 
ALC287_FIXUP_TXNW2781_I2C),
+       SND_PCI_QUIRK(0x17aa, 0x3920, "Yoga S990-16 pro Quad VECO Quad", 
ALC287_FIXUP_YOGA9_SPEAKER2_TO_DAC1),
         SND_PCI_QUIRK(0x17aa, 0x3929, "Thinkbook 13x Gen 5", 
ALC287_FIXUP_MG_RTKC_CSAMP_CS35L41_I2C_THINKPAD),
         SND_PCI_QUIRK(0x17aa, 0x392b, "Thinkbook 13x Gen 5", 
ALC287_FIXUP_MG_RTKC_CSAMP_CS35L41_I2C_THINKPAD),
         HDA_CODEC_QUIRK(0x17aa, 0x3936, "Legion R9000P ADR10H", 
ALC287_FIXUP_LENOVO_LEGION_AW88399),

Since both entries use the same external amplifier,
reusing the existing quirk seems more consistent.

Does that look reasonable to you?