Skip to contentSkip to conversation

ALSA: hda - Fix silent output on MacBook Air 1,1

PATCH In review Adrien Vergé 4 participants 8 messages Jan 24, 2014 – Jan 25, 2014 Sound

lore
Share & tools
lore
#0 lore raw

PATCH 1/1 ALSA: hda - Fix silent output on MacBook Air 1,1

Similarly to other Apple products, MBA 1,1 needs a specific quirk. Pin 0x18 must be set to VREF_50 to have sound output. This was no longer done since commit 1a97b7f, resulting in a mute built-in speaker.

This patch corrects the regression by creating a fixup for the MBA 1,1.

---
 sound/pci/hda/patch_realtek.c | 19 ++++++++++++++++++-
 1 file changed, 18 insertions(+), 1 deletion(-)
sound/pci/hda/patch_realtek.c +18 -1
index c564694..723de28 100644
@@ -1780,6 +1780,7 @@ enum {
 	ALC889_FIXUP_DAC_ROUTE,
 	ALC889_FIXUP_MBP_VREF,
 	ALC889_FIXUP_IMAC91_VREF,
+	ALC889_FIXUP_MBA11_VREF,
 	ALC889_FIXUP_MBA21_VREF,
 	ALC882_FIXUP_INV_DMIC,
 	ALC882_FIXUP_NO_PRIMARY_HP,
@@ -1910,6 +1911,16 @@ static void alc889_fixup_imac91_vref(struct hda_codec *codec,
 		alc889_fixup_mac_pins(codec, nids, ARRAY_SIZE(nids));
 }
 
+/* Set VREF on speaker pins on mba11 */
+static void alc889_fixup_mba11_vref(struct hda_codec *codec,
+				    const struct hda_fixup *fix, int action)
+{
+	static hda_nid_t nids[1] = { 0x18 };
+
+	if (action == HDA_FIXUP_ACT_INIT)
+		alc889_fixup_mac_pins(codec, nids, ARRAY_SIZE(nids));
+}
+
 /* Set VREF on speaker pins on mba21 */
 static void alc889_fixup_mba21_vref(struct hda_codec *codec,
 				    const struct hda_fixup *fix, int action)
@@ -2119,6 +2130,12 @@ static const struct hda_fixup alc882_fixups[] = {
 		.chained = true,
 		.chain_id = ALC882_FIXUP_GPIO1,
 	},
+	[ALC889_FIXUP_MBA11_VREF] = {
+		.type = HDA_FIXUP_FUNC,
+		.v.func = alc889_fixup_mba11_vref,
+		.chained = true,
+		.chain_id = ALC889_FIXUP_MBP_VREF,
+	},
 	[ALC889_FIXUP_MBA21_VREF] = {
 		.type = HDA_FIXUP_FUNC,
 		.v.func = alc889_fixup_mba21_vref,
@@ -2194,7 +2211,7 @@ static const struct snd_pci_quirk alc882_fixup_tbl[] = {
 	SND_PCI_QUIRK(0x106b, 0x2c00, "MacbookPro rev3", ALC889_FIXUP_MBP_VREF),
 	SND_PCI_QUIRK(0x106b, 0x3000, "iMac", ALC889_FIXUP_MBP_VREF),
 	SND_PCI_QUIRK(0x106b, 0x3200, "iMac 7,1 Aluminum", ALC882_FIXUP_EAPD),
-	SND_PCI_QUIRK(0x106b, 0x3400, "MacBookAir 1,1", ALC889_FIXUP_MBP_VREF),
+	SND_PCI_QUIRK(0x106b, 0x3400, "MacBookAir 1,1", ALC889_FIXUP_MBA11_VREF),
 	SND_PCI_QUIRK(0x106b, 0x3500, "MacBookAir 2,1", ALC889_FIXUP_MBA21_VREF),
 	SND_PCI_QUIRK(0x106b, 0x3600, "Macbook 3,1", ALC889_FIXUP_MBP_VREF),
 	SND_PCI_QUIRK(0x106b, 0x3800, "MacbookPro 4,1", ALC889_FIXUP_MBP_VREF),
Signature · 1 line
1.8.5.2
#1 lore raw
Quoted 3 lines · Adrien Vergé

Similarly to other Apple products, MBA 1,1 needs a specific quirk. Pin 0x18 must be set to VREF_50 to have sound output. This was no longer done since commit 1a97b7f, resulting in a mute built-in speaker.

Commit 1a97b7f ("ALSA: hda/realtek - Remove the last static quirks for ALC882") was included in v3.4. Did no-one notice this for 9 releases?

Quoted 4 lines

This patch corrects the regression by creating a fixup for the MBA 1,1.

Paul Bolle

#2 lore raw

I think very few people use Linux on their MBA 1,1. Moreover, part of them remained on v2.6.

2014/1/24 Paul Bolle <[email protected]>:

Quoted 16 lines · Adrien Vergé
Quoted 3 lines

Similarly to other Apple products, MBA 1,1 needs a specific quirk. Pin 0x18 must be set to VREF_50 to have sound output. This was no longer done since commit 1a97b7f, resulting in a mute built-in speaker.

Commit 1a97b7f ("ALSA: hda/realtek - Remove the last static quirks for ALC882") was included in v3.4. Did no-one notice this for 9 releases?

Quoted 4 lines

This patch corrects the regression by creating a fixup for the MBA 1,1.

Paul Bolle

#3 lore raw
Quoted 2 lines · Adrien Vergé

I think very few people use Linux on their MBA 1,1. Moreover, part of them remained on v2.6.

I see. Well, if your analysis is correct I think you're supposed to add

Fixes 1a97b7f22774 ("ALSA: hda/realtek - Remove the last static quirks for ALC882")

to your commit explanation. Not sure, since this tag is not documented, as far as I can see.

Paul Bolle

#4 lore raw

At Fri, 24 Jan 2014 22:08:05 +0100,

Quoted 10 lines · Paul Bolle
Quoted 2 lines · Adrien Vergé

I think very few people use Linux on their MBA 1,1. Moreover, part of them remained on v2.6.

I see. Well, if your analysis is correct I think you're supposed to add

Fixes 1a97b7f22774 ("ALSA: hda/realtek - Remove the last static quirks for ALC882")

to your commit explanation. Not sure, since this tag is not documented, as far as I can see.

I applied with fixes and cc-to-stable tags now.

thanks,

Takashi

#5 lore raw

I can confirm this happens on my machine too: does anybody have a hint on why changing VREF on 0x18 mutes/unmutes speakers and hp, which are on 0x14 and 0x15 ?

Quoted 18 lines · Takashi Iwai

At Fri, 24 Jan 2014 22:08:05 +0100,

Quoted 10 lines · Paul Bolle
Quoted 2 lines · Adrien Vergé

I think very few people use Linux on their MBA 1,1. Moreover, part of them remained on v2.6.

I see. Well, if your analysis is correct I think you're supposed to add

Fixes 1a97b7f22774 ("ALSA: hda/realtek - Remove the last static quirks for ALC882")

to your commit explanation. Not sure, since this tag is not documented, as far as I can see.

I applied with fixes and cc-to-stable tags now.

thanks,

Takashi

#6 lore raw

2014-01-24 Paul Bolle <[email protected]>:

Quoted 2 lines

I see. Well, if your analysis is correct I think you're supposed to add

Fixes 1a97b7f22774 ("ALSA: hda/realtek - Remove the last static quirks for ALC882")

Sorry for that. Thank you for correcting.

2014-01-25 Nicolo' <[email protected]>:

Quoted 3 lines

I can confirm this happens on my machine too: does anybody have a hint on why changing VREF on 0x18 mutes/unmutes speakers and hp, which are on 0x14 and 0x15 ?

Very strange indeed. However Nicolo', on my machine (with v3.13), playing with pin 0x18 only mutes/unmutes speakers, not headphones.

#7 lore raw

I stand corrected, also on mine (3.2) VREF on 0x18 mutes only speakers

Quoted 13 lines · Adrien Vergé

2014-01-24 Paul Bolle <[email protected]>:

Quoted 2 lines

I see. Well, if your analysis is correct I think you're supposed to add

Fixes 1a97b7f22774 ("ALSA: hda/realtek - Remove the last static quirks for ALC882")

Sorry for that. Thank you for correcting.

2014-01-25 Nicolo' <[email protected]>:

Quoted 3 lines

I can confirm this happens on my machine too: does anybody have a hint on why changing VREF on 0x18 mutes/unmutes speakers and hp, which are on 0x14 and 0x15 ?

Very strange indeed. However Nicolo', on my machine (with v3.13), playing with pin 0x18 only mutes/unmutes speakers, not headphones.