Skip to contentSkip to conversation

Revert "ALSA: hda - mask buggy stream DMA0 for Broadwell display controller"

PATCH In review [email protected] 2 participants 4 messages Jun 9, 2014 Sound

lore
Share & tools
lore
#0 lore raw

PATCH 1/2 Revert "ALSA: hda - mask buggy stream DMA0 for Broadwell display controller"

From: Libin Yang <[email protected]>

This reverts commit 7189eb9b8f7962474956196c301676470542f253.

It will use LPIB to get the DMA position on Broadwell HDMI Audio.

---
 sound/pci/hda/hda_intel.c | 6 ------
 1 file changed, 6 deletions(-)
sound/pci/hda/hda_intel.c +0 -6
index cd77b9b..f93f338 100644
@@ -1367,12 +1367,6 @@ static int azx_first_init(struct azx *chip)
 	/* initialize streams */
 	azx_init_stream(chip);
 
-	/* workaround for Broadwell HDMI: the first stream is broken,
-	 * so mask it by keeping it as if opened
-	 */
-	if (pci->vendor == 0x8086 && pci->device == 0x160c)
-		chip->azx_dev[0].opened = 1;
-
 	/* initialize chip */
 	azx_init_pci(chip);
 	azx_init_chip(chip, (probe_only[dev] & 2) == 0);
Signature · 1 line
1.8.3.2
#1 lore raw

PATCH 2/2 ALSA: hda - using POS_FIX_LPIB on Broadwell HDMI Audio

From: Libin Yang <[email protected]>

Broadwell HDMI can't use position buffer reliably, force to use LPIB

---
 sound/pci/hda/hda_intel.c | 8 +++++++-
 1 file changed, 7 insertions(+), 1 deletion(-)
sound/pci/hda/hda_intel.c +7 -1
index f93f338..bb65a124 100644
@@ -237,6 +237,12 @@ enum {
 	 AZX_DCAPS_COUNT_LPIB_DELAY | AZX_DCAPS_PM_RUNTIME | \
 	 AZX_DCAPS_I915_POWERWELL)
 
+/* Broadwell HDMI can't use position buffer reliably, force to use LPIB */
+#define AZX_DCAPS_INTEL_BROADWELL \
+	(AZX_DCAPS_SCH_SNOOP | AZX_DCAPS_ALIGN_BUFSIZE | \
+	 AZX_DCAPS_POSFIX_LPIB | AZX_DCAPS_PM_RUNTIME | \
+	 AZX_DCAPS_I915_POWERWELL)
+
 /* quirks for ATI SB / AMD Hudson */
 #define AZX_DCAPS_PRESET_ATI_SB \
 	(AZX_DCAPS_ATI_SNOOP | AZX_DCAPS_NO_TCSEL | \
@@ -1763,7 +1769,7 @@ static const struct pci_device_id azx_ids[] = {
 	  .driver_data = AZX_DRIVER_HDMI | AZX_DCAPS_INTEL_HASWELL },
 	/* Broadwell */
 	{ PCI_DEVICE(0x8086, 0x160c),
-	  .driver_data = AZX_DRIVER_HDMI | AZX_DCAPS_INTEL_HASWELL },
+	  .driver_data = AZX_DRIVER_HDMI | AZX_DCAPS_INTEL_BROADWELL },
 	/* 5 Series/3400 */
 	{ PCI_DEVICE(0x8086, 0x3b56),
 	  .driver_data = AZX_DRIVER_SCH | AZX_DCAPS_INTEL_PCH_NOPM },
Signature · 1 line
1.8.3.2
#2 lore raw
In reply to [email protected]

At Mon, 9 Jun 2014 15:28:58 +0800,

Quoted 8 lines · [email protected] wrote:

From: Libin Yang <[email protected]>

This reverts commit 7189eb9b8f7962474956196c301676470542f253.

It will use LPIB to get the DMA position on Broadwell HDMI Audio.

Thanks, applied this with the next patch, but in a different order. In general, you shouldn't break things in the middle, especially for bisection. The second patch doesn't conflict with the DMA0 disablement, so we can apply it first, then revert DMA0 disablement.

Takashi

Quoted 24 lines
---
 sound/pci/hda/hda_intel.c | 6 ------
 1 file changed, 6 deletions(-)
sound/pci/hda/hda_intel.c +0 -6
index cd77b9b..f93f338 100644
@@ -1367,12 +1367,6 @@ static int azx_first_init(struct azx *chip)
 	/* initialize streams */
 	azx_init_stream(chip);
 
-	/* workaround for Broadwell HDMI: the first stream is broken,
-	 * so mask it by keeping it as if opened
-	 */
-	if (pci->vendor == 0x8086 && pci->device == 0x160c)
-		chip->azx_dev[0].opened = 1;
-
 	/* initialize chip */
 	azx_init_pci(chip);
 	azx_init_chip(chip, (probe_only[dev] & 2) == 0);
Signature · 1 line
1.8.3.2
#3 lore raw

Hi Takashi,

Got it. Thanks.

Regards,
Libin

Quoted 51 lines

-----Original Message-----
From: Takashi Iwai [mailto:[email protected]]
Sent: Monday, June 9, 2014 3:35 PM
To: Yang, Libin
Cc: [email protected]; Lin, Mengdong
Subject: Re: [PATCH 1/2] Revert "ALSA: hda - mask buggy stream DMA0 for Broadwell display controller"

At Mon, 9 Jun 2014 15:28:58 +0800,

Quoted 8 lines · [email protected] wrote:

From: Libin Yang <[email protected]>

This reverts commit 7189eb9b8f7962474956196c301676470542f253.

It will use LPIB to get the DMA position on Broadwell HDMI Audio.

Thanks, applied this with the next patch, but in a different order. In general, you shouldn't break things in the middle, especially for bisection. The second patch doesn't conflict with the DMA0 disablement, so we can apply it first, then revert DMA0 disablement.

Takashi

Quoted 24 lines
---
 sound/pci/hda/hda_intel.c | 6 ------
 1 file changed, 6 deletions(-)
sound/pci/hda/hda_intel.c +0 -7
index cd77b9b..f93f338 100644
@@ -1367,12 +1367,6 @@ static int azx_first_init(struct azx *chip)
 	/* initialize streams */
 	azx_init_stream(chip);
 
-	/* workaround for Broadwell HDMI: the first stream is broken,
-	 * so mask it by keeping it as if opened
-	 */
-	if (pci->vendor == 0x8086 && pci->device == 0x160c)
-		chip->azx_dev[0].opened = 1;
-
 	/* initialize chip */
 	azx_init_pci(chip);
 	azx_init_chip(chip, (probe_only[dev] & 2) == 0);
--

1.8.3.2