HDA probe failure is observed on Tegra194 based platforms and this
happens due to reset failure. This series fixes the problem by
skipping the failing reset and DT bindings are updated accordingly.
Changelog
=========
v2 -> v3:
---------
* Use reset bulk APIs in HDA driver as suggested by Dmitry.
v1 -> v2:
---------
* Updated HDA driver patch to skip the failing reset instead of
skipping resets in general for BPMP devices as per comment from
Dmitry.
* Used a better strucure name for SoC data as per comment from
Thierry.
* Dropped 'Fixes' tag in binding doc patch as per comment from
Dmitry.
Sameer Pujar (3):
ALSA: hda/tegra: Fix Tegra194 HDA reset failure
dt-bindings: sound: tegra: Update HDA resets
arm64: tegra: Remove non existent Tegra194 reset
.../bindings/sound/nvidia,tegra30-hda.yaml | 13 +++++--
arch/arm64/boot/dts/nvidia/tegra194.dtsi | 5 +--
sound/pci/hda/hda_tegra.c | 45 +++++++++++++++++-----
3 files changed, 47 insertions(+), 16 deletions(-)
--
2.7.4
HDA regression is recently reported on Tegra194 based platforms.
This happens because "hda2codec_2x" reset does not really exist
in Tegra194 and it causes probe failure. All the HDA based audio
tests fail at the moment. This underlying issue is exposed by
commit c045ceb5a145 ("reset: tegra-bpmp: Handle errors in BPMP
response") which now checks return code of BPMP command response.
Fix this issue by skipping unavailable reset on Tegra194.
Signed-off-by: Sameer Pujar <[email protected]>
Cc: [email protected]
Depends-on: 87f0e46e7559 ("ALSA: hda/tegra: Reset hardware")
---
sound/pci/hda/hda_tegra.c | 45 ++++++++++++++++++++++++++++++++++++---------
1 file changed, 36 insertions(+), 9 deletions(-)
diff --git a/sound/pci/hda/hda_tegra.c b/sound/pci/hda/hda_tegra.c
index ea700395..7c3df54 100644
--- a/sound/pci/hda/hda_tegra.c
+++ b/sound/pci/hda/hda_tegra.c
@@ -68,14 +68,20 @@
*/
#define TEGRA194_NUM_SDO_LINES 4
+struct hda_tegra_soc {
+ bool has_hda2codec_2x_reset;
+};
+
struct hda_tegra {
struct azx chip;
struct device *dev;
- struct reset_control *reset;
+ struct reset_control_bulk_data resets[3];
struct clk_bulk_data clocks[3];
+ unsigned int nresets;
unsigned int nclocks;
void __iomem *regs;
struct work_struct probe_work;
+ const struct hda_tegra_soc *data;
};
#ifdef CONFIG_PM
@@ -170,7 +176,7 @@ static int __maybe_unused hda_tegra_runtime_resume(struct device *dev)
int rc;
if (!chip->running) {
- rc = reset_control_assert(hda->reset);
+ rc = reset_control_bulk_assert(hda->nresets, hda->resets);
if (rc)
return rc;
}
@@ -187,7 +193,7 @@ static int __maybe_unused hda_tegra_runtime_resume(struct device *dev)
} else {
usleep_range(10, 100);
- rc = reset_control_deassert(hda->reset);
+ rc = reset_control_bulk_deassert(hda->nresets, hda->resets);
if (rc)
return rc;
}
@@ -427,9 +433,17 @@ static int hda_tegra_create(struct snd_card *card,
return 0;
}
+static const struct hda_tegra_soc tegra30_data = {
+ .has_hda2codec_2x_reset = true,
+};
+
+static const struct hda_tegra_soc tegra194_data = {
+ .has_hda2codec_2x_reset = false,
+};
+
static const struct of_device_id hda_tegra_match[] = {
- { .compatible = "nvidia,tegra30-hda" },
- { .compatible = "nvidia,tegra194-hda" },
+ { .compatible = "nvidia,tegra30-hda", .data = &tegra30_data },
+ { .compatible = "nvidia,tegra194-hda", .data = &tegra194_data },
{},
};
MODULE_DEVICE_TABLE(of, hda_tegra_match);
@@ -449,6 +463,10 @@ static int hda_tegra_probe(struct platform_device *pdev)
hda->dev = &pdev->dev;
chip = &hda->chip;
+ hda->data = of_device_get_match_data(&pdev->dev);
+ if (!hda->data)
+ return -EINVAL;
+
err = snd_card_new(&pdev->dev, SNDRV_DEFAULT_IDX1, SNDRV_DEFAULT_STR1,
THIS_MODULE, 0, &card);
if (err < 0) {
@@ -456,11 +474,20 @@ static int hda_tegra_probe(struct platform_device *pdev)
return err;
}
- hda->reset = devm_reset_control_array_get_exclusive(&pdev->dev);
- if (IS_ERR(hda->reset)) {
- err = PTR_ERR(hda->reset);
+ hda->resets[hda->nresets++].id = "hda";
+ hda->resets[hda->nresets++].id = "hda2hdmi";
+ /*
+ * "hda2codec_2x" reset is not present on Tegra194. Though DT would
+ * be updated to reflect this, but to have backward compatibility
+ * below is necessary.
+ */
+ if (hda->data->has_hda2codec_2x_reset)
+ hda->resets[hda->nresets++].id = "hda2codec_2x";
+
+ err = devm_reset_control_bulk_get_exclusive(&pdev->dev, hda->nresets,
+ hda->resets);
+ if (err)
goto out_free;
- }
hda->clocks[hda->nclocks++].id = "hda";
hda->clocks[hda->nclocks++].id = "hda2hdmi";
--
2.7.4
Tegra194 HDA has only two resets unlike the previous generations of
Tegra SoCs. Hence update the reset list accordingly.
Signed-off-by: Sameer Pujar <[email protected]>
Acked-by: Thierry Reding <[email protected]>
---
.../devicetree/bindings/sound/nvidia,tegra30-hda.yaml | 13 +++++++++----
1 file changed, 9 insertions(+), 4 deletions(-)
diff --git a/Documentation/devicetree/bindings/sound/nvidia,tegra30-hda.yaml b/Documentation/devicetree/bindings/sound/nvidia,tegra30-hda.yaml
index b55775e..70dbdff5 100644
--- a/Documentation/devicetree/bindings/sound/nvidia,tegra30-hda.yaml
+++ b/Documentation/devicetree/bindings/sound/nvidia,tegra30-hda.yaml
@@ -50,13 +50,18 @@ properties:
- const: hda2codec_2x
resets:
+ minItems: 2
maxItems: 3
reset-names:
- items:
- - const: hda
- - const: hda2hdmi
- - const: hda2codec_2x
+ oneOf:
+ - items:
+ - const: hda
+ - const: hda2hdmi
+ - const: hda2codec_2x
+ - items:
+ - const: hda
+ - const: hda2hdmi
power-domains:
maxItems: 1
--
2.7.4
Tegra194 does not really have "hda2codec_2x" related reset. Hence drop
this entry to reflect actual HW.
Fixes: 4878cc0c9fab ("arm64: tegra: Add HDA controller on Tegra194")
Signed-off-by: Sameer Pujar <[email protected]>
---
arch/arm64/boot/dts/nvidia/tegra194.dtsi | 5 ++---
1 file changed, 2 insertions(+), 3 deletions(-)
diff --git a/arch/arm64/boot/dts/nvidia/tegra194.dtsi b/arch/arm64/boot/dts/nvidia/tegra194.dtsi
index 8d29b7f..6a1d896 100644
--- a/arch/arm64/boot/dts/nvidia/tegra194.dtsi
+++ b/arch/arm64/boot/dts/nvidia/tegra194.dtsi
@@ -976,9 +976,8 @@
<&bpmp TEGRA194_CLK_HDA2CODEC_2X>;
clock-names = "hda", "hda2hdmi", "hda2codec_2x";
resets = <&bpmp TEGRA194_RESET_HDA>,
- <&bpmp TEGRA194_RESET_HDA2HDMICODEC>,
- <&bpmp TEGRA194_RESET_HDA2CODEC_2X>;
- reset-names = "hda", "hda2hdmi", "hda2codec_2x";
+ <&bpmp TEGRA194_RESET_HDA2HDMICODEC>;
+ reset-names = "hda", "hda2hdmi";
power-domains = <&bpmp TEGRA194_POWER_DOMAIN_DISP>;
interconnects = <&mc TEGRA194_MEMORY_CLIENT_HDAR &emc>,
<&mc TEGRA194_MEMORY_CLIENT_HDAW &emc>;
--
2.7.4
22.12.2021 07:35, Sameer Pujar пишет:
> HDA regression is recently reported on Tegra194 based platforms.
> This happens because "hda2codec_2x" reset does not really exist
> in Tegra194 and it causes probe failure. All the HDA based audio
> tests fail at the moment. This underlying issue is exposed by
> commit c045ceb5a145 ("reset: tegra-bpmp: Handle errors in BPMP
> response") which now checks return code of BPMP command response.
> Fix this issue by skipping unavailable reset on Tegra194.
>
> Signed-off-by: Sameer Pujar <[email protected]>
> Cc: [email protected]
> Depends-on: 87f0e46e7559 ("ALSA: hda/tegra: Reset hardware")
Is "Depends-on" a valid tag? I can't find it in Documentation/.
> ---
> sound/pci/hda/hda_tegra.c | 45 ++++++++++++++++++++++++++++++++++++---------
> 1 file changed, 36 insertions(+), 9 deletions(-)
>
> diff --git a/sound/pci/hda/hda_tegra.c b/sound/pci/hda/hda_tegra.c
> index ea700395..7c3df54 100644
> --- a/sound/pci/hda/hda_tegra.c
> +++ b/sound/pci/hda/hda_tegra.c
> @@ -68,14 +68,20 @@
> */
> #define TEGRA194_NUM_SDO_LINES 4
>
> +struct hda_tegra_soc {
> + bool has_hda2codec_2x_reset;
> +};
> +
> struct hda_tegra {
> struct azx chip;
> struct device *dev;
> - struct reset_control *reset;
> + struct reset_control_bulk_data resets[3];
> struct clk_bulk_data clocks[3];
> + unsigned int nresets;
> unsigned int nclocks;
> void __iomem *regs;
> struct work_struct probe_work;
> + const struct hda_tegra_soc *data;
> };
>
> #ifdef CONFIG_PM
> @@ -170,7 +176,7 @@ static int __maybe_unused hda_tegra_runtime_resume(struct device *dev)
> int rc;
>
> if (!chip->running) {
> - rc = reset_control_assert(hda->reset);
> + rc = reset_control_bulk_assert(hda->nresets, hda->resets);
> if (rc)
> return rc;
> }
> @@ -187,7 +193,7 @@ static int __maybe_unused hda_tegra_runtime_resume(struct device *dev)
> } else {
> usleep_range(10, 100);
>
> - rc = reset_control_deassert(hda->reset);
> + rc = reset_control_bulk_deassert(hda->nresets, hda->resets);
> if (rc)
> return rc;
> }
> @@ -427,9 +433,17 @@ static int hda_tegra_create(struct snd_card *card,
> return 0;
> }
>
> +static const struct hda_tegra_soc tegra30_data = {
> + .has_hda2codec_2x_reset = true,
> +};
> +
> +static const struct hda_tegra_soc tegra194_data = {
> + .has_hda2codec_2x_reset = false,
> +};
> +
> static const struct of_device_id hda_tegra_match[] = {
> - { .compatible = "nvidia,tegra30-hda" },
> - { .compatible = "nvidia,tegra194-hda" },
> + { .compatible = "nvidia,tegra30-hda", .data = &tegra30_data },
> + { .compatible = "nvidia,tegra194-hda", .data = &tegra194_data },
> {},
> };
> MODULE_DEVICE_TABLE(of, hda_tegra_match);
> @@ -449,6 +463,10 @@ static int hda_tegra_probe(struct platform_device *pdev)
> hda->dev = &pdev->dev;
> chip = &hda->chip;
>
> + hda->data = of_device_get_match_data(&pdev->dev);
> + if (!hda->data)
> + return -EINVAL;
hda->data can't ever be NULL because all hda_tegra_match[] compatibles
above have .data assigned. Technically this check is redundant.
Thierry suggested previously to name it "hda->soc", like we usually do
it in other drivers.
> err = snd_card_new(&pdev->dev, SNDRV_DEFAULT_IDX1, SNDRV_DEFAULT_STR1,
> THIS_MODULE, 0, &card);
> if (err < 0) {
> @@ -456,11 +474,20 @@ static int hda_tegra_probe(struct platform_device *pdev)
> return err;
> }
>
> - hda->reset = devm_reset_control_array_get_exclusive(&pdev->dev);
> - if (IS_ERR(hda->reset)) {
> - err = PTR_ERR(hda->reset);
> + hda->resets[hda->nresets++].id = "hda";
> + hda->resets[hda->nresets++].id = "hda2hdmi";
> + /*
> + * "hda2codec_2x" reset is not present on Tegra194. Though DT would
> + * be updated to reflect this, but to have backward compatibility
> + * below is necessary.
> + */
> + if (hda->data->has_hda2codec_2x_reset)
> + hda->resets[hda->nresets++].id = "hda2codec_2x";
> +
> + err = devm_reset_control_bulk_get_exclusive(&pdev->dev, hda->nresets,
> + hda->resets);
> + if (err)
> goto out_free;
> - }
>
> hda->clocks[hda->nclocks++].id = "hda";
> hda->clocks[hda->nclocks++].id = "hda2hdmi";
>
Not sure whether the above nits worth making v4. I'll leave it up to you
and other reviewers to decide.
Overall this patch looks good to me, thank you.
Reviewed-by: Dmitry Osipenko <[email protected]>
On Wed, Dec 22, 2021 at 10:05:50AM +0530, Sameer Pujar wrote:
> Tegra194 HDA has only two resets unlike the previous generations of
> Tegra SoCs. Hence update the reset list accordingly.
>
> Signed-off-by: Sameer Pujar <[email protected]>
> Acked-by: Thierry Reding <[email protected]>
> ---
> .../devicetree/bindings/sound/nvidia,tegra30-hda.yaml | 13 +++++++++----
> 1 file changed, 9 insertions(+), 4 deletions(-)
>
> diff --git a/Documentation/devicetree/bindings/sound/nvidia,tegra30-hda.yaml b/Documentation/devicetree/bindings/sound/nvidia,tegra30-hda.yaml
> index b55775e..70dbdff5 100644
> --- a/Documentation/devicetree/bindings/sound/nvidia,tegra30-hda.yaml
> +++ b/Documentation/devicetree/bindings/sound/nvidia,tegra30-hda.yaml
> @@ -50,13 +50,18 @@ properties:
> - const: hda2codec_2x
>
> resets:
> + minItems: 2
> maxItems: 3
>
> reset-names:
Just add 'minItems: 2' here instead.
> - items:
> - - const: hda
> - - const: hda2hdmi
> - - const: hda2codec_2x
> + oneOf:
> + - items:
> + - const: hda
> + - const: hda2hdmi
> + - const: hda2codec_2x
> + - items:
> + - const: hda
> + - const: hda2hdmi
>
> power-domains:
> maxItems: 1
> --
> 2.7.4
>
>
On 12/23/2021 12:10 AM, Dmitry Osipenko wrote:
> 22.12.2021 07:35, Sameer Pujar пишет:
>> HDA regression is recently reported on Tegra194 based platforms.
>> This happens because "hda2codec_2x" reset does not really exist
>> in Tegra194 and it causes probe failure. All the HDA based audio
>> tests fail at the moment. This underlying issue is exposed by
>> commit c045ceb5a145 ("reset: tegra-bpmp: Handle errors in BPMP
>> response") which now checks return code of BPMP command response.
>> Fix this issue by skipping unavailable reset on Tegra194.
>>
>> Signed-off-by: Sameer Pujar <[email protected]>
>> Cc: [email protected]
>> Depends-on: 87f0e46e7559 ("ALSA: hda/tegra: Reset hardware")
> Is "Depends-on" a valid tag? I can't find it in Documentation/.
I do find the usage of the tag in many commits though there is no
reference of this in doc. I always thought it would act as a reference
when commits get pulled to other branches. If this is not true and it
does not mean anything, I will drop this.
>
>> ---
>> sound/pci/hda/hda_tegra.c | 45 ++++++++++++++++++++++++++++++++++++---------
>> 1 file changed, 36 insertions(+), 9 deletions(-)
>>
>> diff --git a/sound/pci/hda/hda_tegra.c b/sound/pci/hda/hda_tegra.c
>> index ea700395..7c3df54 100644
>> --- a/sound/pci/hda/hda_tegra.c
>> +++ b/sound/pci/hda/hda_tegra.c
>> @@ -68,14 +68,20 @@
>> */
>> #define TEGRA194_NUM_SDO_LINES 4
>>
>> +struct hda_tegra_soc {
>> + bool has_hda2codec_2x_reset;
>> +};
>> +
>> struct hda_tegra {
>> struct azx chip;
>> struct device *dev;
>> - struct reset_control *reset;
>> + struct reset_control_bulk_data resets[3];
>> struct clk_bulk_data clocks[3];
>> + unsigned int nresets;
>> unsigned int nclocks;
>> void __iomem *regs;
>> struct work_struct probe_work;
>> + const struct hda_tegra_soc *data;
>> };
>>
>> #ifdef CONFIG_PM
>> @@ -170,7 +176,7 @@ static int __maybe_unused hda_tegra_runtime_resume(struct device *dev)
>> int rc;
>>
>> if (!chip->running) {
>> - rc = reset_control_assert(hda->reset);
>> + rc = reset_control_bulk_assert(hda->nresets, hda->resets);
>> if (rc)
>> return rc;
>> }
>> @@ -187,7 +193,7 @@ static int __maybe_unused hda_tegra_runtime_resume(struct device *dev)
>> } else {
>> usleep_range(10, 100);
>>
>> - rc = reset_control_deassert(hda->reset);
>> + rc = reset_control_bulk_deassert(hda->nresets, hda->resets);
>> if (rc)
>> return rc;
>> }
>> @@ -427,9 +433,17 @@ static int hda_tegra_create(struct snd_card *card,
>> return 0;
>> }
>>
>> +static const struct hda_tegra_soc tegra30_data = {
>> + .has_hda2codec_2x_reset = true,
>> +};
>> +
>> +static const struct hda_tegra_soc tegra194_data = {
>> + .has_hda2codec_2x_reset = false,
>> +};
>> +
>> static const struct of_device_id hda_tegra_match[] = {
>> - { .compatible = "nvidia,tegra30-hda" },
>> - { .compatible = "nvidia,tegra194-hda" },
>> + { .compatible = "nvidia,tegra30-hda", .data = &tegra30_data },
>> + { .compatible = "nvidia,tegra194-hda", .data = &tegra194_data },
>> {},
>> };
>> MODULE_DEVICE_TABLE(of, hda_tegra_match);
>> @@ -449,6 +463,10 @@ static int hda_tegra_probe(struct platform_device *pdev)
>> hda->dev = &pdev->dev;
>> chip = &hda->chip;
>>
>> + hda->data = of_device_get_match_data(&pdev->dev);
>> + if (!hda->data)
>> + return -EINVAL;
> hda->data can't ever be NULL because all hda_tegra_match[] compatibles
> above have .data assigned. Technically this check is redundant.
Will remove.
>
> Thierry suggested previously to name it "hda->soc", like we usually do
> it in other drivers.
Previously I renamed strcture, but didn't update the member name. Will
update.
>> err = snd_card_new(&pdev->dev, SNDRV_DEFAULT_IDX1, SNDRV_DEFAULT_STR1,
>> THIS_MODULE, 0, &card);
>> if (err < 0) {
>> @@ -456,11 +474,20 @@ static int hda_tegra_probe(struct platform_device *pdev)
>> return err;
>> }
>>
>> - hda->reset = devm_reset_control_array_get_exclusive(&pdev->dev);
>> - if (IS_ERR(hda->reset)) {
>> - err = PTR_ERR(hda->reset);
>> + hda->resets[hda->nresets++].id = "hda";
>> + hda->resets[hda->nresets++].id = "hda2hdmi";
>> + /*
>> + * "hda2codec_2x" reset is not present on Tegra194. Though DT would
>> + * be updated to reflect this, but to have backward compatibility
>> + * below is necessary.
>> + */
>> + if (hda->data->has_hda2codec_2x_reset)
>> + hda->resets[hda->nresets++].id = "hda2codec_2x";
>> +
>> + err = devm_reset_control_bulk_get_exclusive(&pdev->dev, hda->nresets,
>> + hda->resets);
>> + if (err)
>> goto out_free;
>> - }
>>
>> hda->clocks[hda->nclocks++].id = "hda";
>> hda->clocks[hda->nclocks++].id = "hda2hdmi";
>>
> Not sure whether the above nits worth making v4. I'll leave it up to you
> and other reviewers to decide.
>
> Overall this patch looks good to me, thank you.
>
> Reviewed-by: Dmitry Osipenko <[email protected]>
On Thu, Dec 23, 2021 at 10:04:19AM +0530, Sameer Pujar wrote:
>
>
> On 12/23/2021 12:10 AM, Dmitry Osipenko wrote:
> > 22.12.2021 07:35, Sameer Pujar пишет:
> > > HDA regression is recently reported on Tegra194 based platforms.
> > > This happens because "hda2codec_2x" reset does not really exist
> > > in Tegra194 and it causes probe failure. All the HDA based audio
> > > tests fail at the moment. This underlying issue is exposed by
> > > commit c045ceb5a145 ("reset: tegra-bpmp: Handle errors in BPMP
> > > response") which now checks return code of BPMP command response.
> > > Fix this issue by skipping unavailable reset on Tegra194.
> > >
> > > Signed-off-by: Sameer Pujar <[email protected]>
> > > Cc: [email protected]
> > > Depends-on: 87f0e46e7559 ("ALSA: hda/tegra: Reset hardware")
> > Is "Depends-on" a valid tag? I can't find it in Documentation/.
>
> I do find the usage of the tag in many commits though there is no reference
> of this in doc. I always thought it would act as a reference when commits
> get pulled to other branches. If this is not true and it does not mean
> anything, I will drop this.
It is not true at all, please read:
https://www.kernel.org/doc/html/latest/process/stable-kernel-rules.html
for how to do this properly.
On 12/23/2021 12:46 PM, Greg KH wrote:
> On Thu, Dec 23, 2021 at 10:04:19AM +0530, Sameer Pujar wrote:
>>
>> On 12/23/2021 12:10 AM, Dmitry Osipenko wrote:
>>> 22.12.2021 07:35, Sameer Pujar пишет:
>>>> HDA regression is recently reported on Tegra194 based platforms.
>>>> This happens because "hda2codec_2x" reset does not really exist
>>>> in Tegra194 and it causes probe failure. All the HDA based audio
>>>> tests fail at the moment. This underlying issue is exposed by
>>>> commit c045ceb5a145 ("reset: tegra-bpmp: Handle errors in BPMP
>>>> response") which now checks return code of BPMP command response.
>>>> Fix this issue by skipping unavailable reset on Tegra194.
>>>>
>>>> Signed-off-by: Sameer Pujar <[email protected]>
>>>> Cc: [email protected]
>>>> Depends-on: 87f0e46e7559 ("ALSA: hda/tegra: Reset hardware")
>>> Is "Depends-on" a valid tag? I can't find it in Documentation/.
>> I do find the usage of the tag in many commits though there is no reference
>> of this in doc. I always thought it would act as a reference when commits
>> get pulled to other branches. If this is not true and it does not mean
>> anything, I will drop this.
> It is not true at all, please read:
> https://www.kernel.org/doc/html/latest/process/stable-kernel-rules.html
> for how to do this properly.
Thanks Greg for the pointer. I will drop above tag in v4.