Skip to contentSkip to conversation

Clean some logs on rockchip hdmi audio

PATCH In review Detlev Casanova 3 participants 8 messages Jul 22, 2025 – Jul 23, 2025

Subsystems (6)
DRM drivers Sound DRM misc ASoC +2 more subsystems
lore
Share & tools
lore
#0 lore raw

v2 cover Clean some logs on rockchip hdmi audio

HDMI audio is showing lots of logs, particularly when HDMI is disconnected.

It doubled on RK3588 with the addition of the second HDMI TX port, so clean it up to avoid missing important log lines.

Changes since v1:
- Add comment about clock requirements
- Add Fixes tags to the drm/bridge commits
- Link to v1: https://lore.kernel.org/all/[email protected]/

Detlev Casanova (3):
drm/bridge: dw-hdmi-qp: Return 0 in audio prepare when disconnected
ASoC: hdac_hdmi: Use dev_info on invalid ELD version drm/bridge: synopsys: Do not warn about audio params computation

 drivers/gpu/drm/bridge/synopsys/dw-hdmi-qp.c | 13 ++++++++++---
 sound/soc/codecs/hdac_hdmi.c                 |  2 +-
 2 files changed, 11 insertions(+), 4 deletions(-)
Signature · 1 line
2.50.1
#1 lore raw

PATCH v2 1/3 drm/bridge: dw-hdmi-qp: Return 0 in audio prepare when disconnected

To configure audio registers, the clock of the video port in use must be enabled.
As those clocks are managed by the VOP driver, they can't be enabled here to write the registers even when the HDMI cable is disconnected.

Furthermore, the registers values are computed from the TMDS char rate, which is not available when disconnected.

Returning -ENODEV seemed reasonable at first, but ASoC will log an error multiple times if dw_hdmi_qp_audio_prepare() return an error. Userspace might also retry multiple times, filling the kernel log with:

hdmi-audio-codec hdmi-audio-codec.0.auto: ASoC error (-19): at snd_soc_dai_prepare() on i2s-hifi

This has become even worse with the support of the second HDMI TX port.

Activating the clocks to write fake data (fake because the TMDS char rate is unavailable) would require API changes to communicate between VOP and HDMI, which doesn't really make sense.

Using a cached regmap to be dumped when a cable is connected won't work because writing order is important and some data needs to be retrieved from registers to write others.

Returning 0 to silently fail sounds like the best and simplest solution.

Fixes fd0141d1a8a2 ("drm/bridge: synopsys: Add audio support for dw-hdmi-qp")
---
 drivers/gpu/drm/bridge/synopsys/dw-hdmi-qp.c | 10 +++++++++-
 1 file changed, 9 insertions(+), 1 deletion(-)
drivers/gpu/drm/bridge/synopsys/dw-hdmi-qp.c +9 -1
index 5e5f8c2f95be1..9b9d43c02e3a5 100644
@@ -458,8 +458,16 @@ static int dw_hdmi_qp_audio_prepare(struct drm_connector *connector,
 	struct dw_hdmi_qp *hdmi = dw_hdmi_qp_from_bridge(bridge);
 	bool ref2stream = false;
 
+	/*
+	 * Silently return if tmds_char_rate is not set.
+	 *
+	 * Writing audio registers requires that the clock of the Video Port currently in
+	 * use by the VOP (dclk_vp<id>) is enabled.
+	 * That clock is guaranteed to be enabled when hdmi->tmds_char_rate is set, so we
+	 * only configure audio when it is set.
+	 */
 	if (!hdmi->tmds_char_rate)
-		return -ENODEV;
+		return 0;
 
 	if (fmt->bit_clk_provider | fmt->frame_clk_provider) {
 		dev_err(hdmi->dev, "unsupported clock settings\n");
Signature · 1 line
2.50.1
#2 lore raw

Re: [PATCH v2 1/3] drm/bridge: dw-hdmi-qp: Return 0 in audio prepare when disconnected

Quoted 41 lines · Detlev Casanova

To configure audio registers, the clock of the video port in use must be enabled.
As those clocks are managed by the VOP driver, they can't be enabled here to write the registers even when the HDMI cable is disconnected.

Furthermore, the registers values are computed from the TMDS char rate, which is not available when disconnected.

Returning -ENODEV seemed reasonable at first, but ASoC will log an error multiple times if dw_hdmi_qp_audio_prepare() return an error. Userspace might also retry multiple times, filling the kernel log with:

hdmi-audio-codec hdmi-audio-codec.0.auto: ASoC error (-19): at snd_soc_dai_prepare() on i2s-hifi

This has become even worse with the support of the second HDMI TX port.

Activating the clocks to write fake data (fake because the TMDS char rate is unavailable) would require API changes to communicate between VOP and HDMI, which doesn't really make sense.

Using a cached regmap to be dumped when a cable is connected won't work because writing order is important and some data needs to be retrieved from registers to write others.

Returning 0 to silently fail sounds like the best and simplest solution.

Fixes fd0141d1a8a2 ("drm/bridge: synopsys: Add audio support for dw-hdmi-qp")
---
 drivers/gpu/drm/bridge/synopsys/dw-hdmi-qp.c | 10 +++++++++-
 1 file changed, 9 insertions(+), 1 deletion(-)
drivers/gpu/drm/bridge/synopsys/dw-hdmi-qp.c +1 -0
index 5e5f8c2f95be1..9b9d43c02e3a5 100644
@@ -458,8 +458,16 @@ static int dw_hdmi_qp_audio_prepare(struct drm_connector *connector,
 	struct dw_hdmi_qp *hdmi = dw_hdmi_qp_from_bridge(bridge);
 	bool ref2stream = false;
 
+	/*

drivers/gpu/drm/bridge/synopsys/dw-hdmi-qp.c static int dw_hdmi_qp_audio_prepare(struct drm_connector *connector, · line 470

Quoted 10 lines
drivers/gpu/drm/bridge/synopsys/dw-hdmi-qp.c +8 -1
+	 * Silently return if tmds_char_rate is not set.
+	 *
+	 * Writing audio registers requires that the clock of the Video Port currently in
+	 * use by the VOP (dclk_vp<id>) is enabled.
+	 * That clock is guaranteed to be enabled when hdmi->tmds_char_rate is set, so we
+	 * only configure audio when it is set.
+	 */
 	if (!hdmi->tmds_char_rate)
-		return -ENODEV;
+		return 0;

What if the cable gets diconnected _while_ this function is running?

Quoted 6 lines
drivers/gpu/drm/bridge/synopsys/dw-hdmi-qp.c +0 -0
 	if (fmt->bit_clk_provider | fmt->frame_clk_provider) {
 		dev_err(hdmi->dev, "unsupported clock settings\n");
Signature · 1 line
2.50.1
Signature · 2 lines
With best wishes
Dmitry
#3 lore raw

Re: [PATCH v2 1/3] drm/bridge: dw-hdmi-qp: Return 0 in audio prepare when disconnected

Hi Dmitry

Quoted 46 lines · On Wednesday, 23 July 2025 07:41:43 EDT Dmitry Baryshkov wrote:
Quoted 45 lines · Detlev Casanova

To configure audio registers, the clock of the video port in use must be enabled.
As those clocks are managed by the VOP driver, they can't be enabled here to write the registers even when the HDMI cable is disconnected.

Furthermore, the registers values are computed from the TMDS char rate, which is not available when disconnected.

Returning -ENODEV seemed reasonable at first, but ASoC will log an error multiple times if dw_hdmi_qp_audio_prepare() return an error. Userspace might also retry multiple times, filling the kernel log with:

hdmi-audio-codec hdmi-audio-codec.0.auto: ASoC error (-19): at snd_soc_dai_prepare() on i2s-hifi

This has become even worse with the support of the second HDMI TX port.

Activating the clocks to write fake data (fake because the TMDS char rate is unavailable) would require API changes to communicate between VOP and HDMI, which doesn't really make sense.

Using a cached regmap to be dumped when a cable is connected won't work because writing order is important and some data needs to be retrieved from registers to write others.

Returning 0 to silently fail sounds like the best and simplest solution.

Fixes fd0141d1a8a2 ("drm/bridge: synopsys: Add audio support for

dw-hdmi-qp") Signed-off-by: Detlev Casanova
<[email protected]>
---

 drivers/gpu/drm/bridge/synopsys/dw-hdmi-qp.c | 10 +++++++++-
 1 file changed, 9 insertions(+), 1 deletion(-)
(quoted diff) +0 -0

b/drivers/gpu/drm/bridge/synopsys/dw-hdmi-qp.c index 5e5f8c2f95be1..9b9d43c02e3a5 100644

drivers/gpu/drm/bridge/synopsys/dw-hdmi-qp.c +0 -0
@@ -458,8 +458,16 @@ static int dw_hdmi_qp_audio_prepare(struct

drm_connector *connector,>

drivers/gpu/drm/bridge/synopsys/dw-hdmi-qp.c +0 -0
 	struct dw_hdmi_qp *hdmi = dw_hdmi_qp_from_bridge(bridge);
 	bool ref2stream = false;
Quoted 6 lines

drivers/gpu/drm/bridge/synopsys/dw-hdmi-qp.c

Quoted 6 lines
drivers/gpu/drm/bridge/synopsys/dw-hdmi-qp.c +4 -0
+	/*
+	 * Silently return if tmds_char_rate is not set.
+	 *
+	 * Writing audio registers requires that the clock of the Video Port

currently in + * use by the VOP (dclk_vp<id>) is enabled.

drivers/gpu/drm/bridge/synopsys/dw-hdmi-qp.c +1 -0
+	 * That clock is guaranteed to be enabled when hdmi->tmds_char_rate 

is

Quoted 9 lines

drivers/gpu/drm/bridge/synopsys/dw-hdmi-qp.c static int dw_hdmi_qp_audio_prepare(struct drm_connector *connector,

Quoted 7 lines

set, so we + * only configure audio when it is set.

drivers/gpu/drm/bridge/synopsys/dw-hdmi-qp.c +2 -1
+	 */
 
 	if (!hdmi->tmds_char_rate)
 
-		return -ENODEV;
+		return 0;

What if the cable gets diconnected _while_ this function is running?

Unplugging the cable will not deactivate the clocks, that's the job of the VOP's vop2_crtc_atomic_disable() function. My understanding is that that function will not be called until this one has returned, but I could be wrong.

Quoted 3 lines

drivers/gpu/drm/bridge/synopsys/dw-hdmi-qp.c static int dw_hdmi_qp_audio_prepare(struct drm_connector *connector, · line 473

Quoted 3 lines
drivers/gpu/drm/bridge/synopsys/dw-hdmi-qp.c +0 -0
 	if (fmt->bit_clk_provider | fmt->frame_clk_provider) {
 
 		dev_err(hdmi->dev, "unsupported clock settings\n");

Regards,
Detlev.

#4 lore raw

PATCH v2 2/3 ASoC: hdac_hdmi: Use dev_info on invalid ELD version

In reply to Detlev Casanova

When disconnected, the ELD data cannot be read by the display driver, so it just sets the data to 0.

That makes the ELD parsing code read an ELD version of 0, which is invalid. In hdac_hdmi, that is logged with dev_err(), but should be logged with dev_info() instead as it is done in sound/core/pcm_drm_eld.c

This avoids printing multiple messages like:

HDMI: Unknown ELD version 0

in the kernel log when userspace tries to open the sound device.

---
 sound/soc/codecs/hdac_hdmi.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
sound/soc/codecs/hdac_hdmi.c +1 -1
index 1139a2754ca33..4cc3b7a1062bd 100644
@@ -1232,7 +1232,7 @@ static int hdac_hdmi_parse_eld(struct hdac_device *hdev,
 						>> DRM_ELD_VER_SHIFT;
 
 	if (ver != ELD_VER_CEA_861D && ver != ELD_VER_PARTIAL) {
-		dev_err(&hdev->dev, "HDMI: Unknown ELD version %d\n", ver);
+		dev_info(&hdev->dev, "HDMI: Unknown ELD version %d\n", ver);
 		return -EINVAL;
 	}
 
Signature · 1 line
2.50.1
maintainer
#5 lore raw

Re: [PATCH v2 2/3] ASoC: hdac_hdmi: Use dev_info on invalid ELD version

Quoted 2 lines · Detlev Casanova

When disconnected, the ELD data cannot be read by the display driver, so it just sets the data to 0.

Please don't put patches for different subsystems into the same series if there's no dependencies, it just makes dependencies less obvious and creates hassle merging things.

Quoted 9 lines

That makes the ELD parsing code read an ELD version of 0, which is invalid. In hdac_hdmi, that is logged with dev_err(), but should be logged with dev_info() instead as it is done in sound/core/pcm_drm_eld.c

This avoids printing multiple messages like:

HDMI: Unknown ELD version 0

in the kernel log when userspace tries to open the sound device.

It doesn't, it just lowers the severity of the logs that are printed. If the goal is to lower the number of messages printed you need to use a ratelimited print.

#6 lore raw

Re: [PATCH v2 2/3] ASoC: hdac_hdmi: Use dev_info on invalid ELD version

Hi Mark,

Quoted 7 lines · On Wednesday, 23 July 2025 06:40:13 EDT Mark Brown wrote:
Quoted 2 lines · Detlev Casanova

When disconnected, the ELD data cannot be read by the display driver, so it just sets the data to 0.

Please don't put patches for different subsystems into the same series if there's no dependencies, it just makes dependencies less obvious and creates hassle merging things.

Yes, sorry, I'll send v3 of this patch separately.

Quoted 4 lines
Quoted 4 lines

That makes the ELD parsing code read an ELD version of 0, which is invalid. In hdac_hdmi, that is logged with dev_err(), but should be logged with dev_info() instead as it is done in sound/core/pcm_drm_eld.c

Quoted 8 lines
Quoted 4 lines

This avoids printing multiple messages like:
HDMI: Unknown ELD version 0

in the kernel log when userspace tries to open the sound device.

It doesn't, it just lowers the severity of the logs that are printed. If the goal is to lower the number of messages printed you need to use a ratelimited print.

I see, ratelimited would be good, but it still prints a message about something that is normal behaviour. Maybe this should go further to a dev_dbg(), or is there a specific reason to show this message ?

This could also be a special case:
- version == 0 -> dev_dbg()
- version !=0 && != known_versions -> dev_err()

Detlev.

#7 lore raw

PATCH v2 3/3 drm/bridge: synopsys: Do not warn about audio params computation

In reply to Detlev Casanova

There is no need to warn about non pre-computed values, just change it to dbg.

Fixes fd0141d1a8a2 ("drm/bridge: synopsys: Add audio support for dw-hdmi-qp")
---
 drivers/gpu/drm/bridge/synopsys/dw-hdmi-qp.c | 3 +--
 1 file changed, 1 insertion(+), 2 deletions(-)
drivers/gpu/drm/bridge/synopsys/dw-hdmi-qp.c +1 -2
index 9b9d43c02e3a5..d974bcad8f94a 100644
@@ -276,8 +276,7 @@ static unsigned int dw_hdmi_qp_find_n(struct dw_hdmi_qp *hdmi, unsigned long pix
 	if (n > 0)
 		return n;
 
-	dev_warn(hdmi->dev, "Rate %lu missing; compute N dynamically\n",
-		 pixel_clk);
+	dev_dbg(hdmi->dev, "Rate %lu missing; compute N dynamically\n", pixel_clk);
 
 	return dw_hdmi_qp_compute_n(hdmi, pixel_clk, sample_rate);
 }
Signature · 1 line
2.50.1