[PATCH] ASoC: ES8389: convert to devm_clk_get_optional() to get clock

[PATCH] ASoC: ES8389: convert to devm_clk_get_optional() to get clock

[Li Jian]

When enabling ES8390 via ACPI description, es8389 would fail to
obtain a clock source, causing the driver to fail to initialize.
This was not an issue with older kernels, but since commit
abae8e57e49a ("clk: generalize devm_clk_get() a bit"),
devm_clk_get() would return an error pointer when a clock source
was not detected (instead of falling back to a static clock),
causing the driver to fail early.

Use devm_clk_get_optional() instead to return to the previous
behaviour, allowing the use of a static clock source.

Cc: stable@vger.kernel.org
Fixes: abae8e57e49a ("clk: generalize devm_clk_get() a bit")
Signed-off-by: Li Jian <lazycat-xiao@foxmail.com>
---
 sound/soc/codecs/es8389.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/sound/soc/codecs/es8389.c b/sound/soc/codecs/es8389.c
index 8d418cae3..449d9574b 100644
--- a/sound/soc/codecs/es8389.c
+++ b/sound/soc/codecs/es8389.c
@@ -892,7 +892,7 @@ static int es8389_probe(struct snd_soc_component *component)
 		return ret;
 	}
 
-	es8389->mclk = devm_clk_get(component->dev, "mclk");
+	es8389->mclk = devm_clk_get_optional(component->dev, "mclk");
 	if (IS_ERR(es8389->mclk))
 		return dev_err_probe(component->dev, PTR_ERR(es8389->mclk),
 			"ES8389 is unable to get mclk\n");
-- 
2.47.3

Re: [PATCH] ASoC: ES8389: convert to devm_clk_get_optional() to get clock

[Uwe Kleine-König]

[-- Attachment #1: Type: text/plain, Size: 926 bytes --]

Hello,

On Fri, Apr 17, 2026 at 06:53:14PM +0800, Li Jian wrote:
> When enabling ES8390 via ACPI description, es8389 would fail to
> obtain a clock source, causing the driver to fail to initialize.
> This was not an issue with older kernels, but since commit
> abae8e57e49a ("clk: generalize devm_clk_get() a bit"),
> devm_clk_get() would return an error pointer when a clock source
> was not detected (instead of falling back to a static clock),
> causing the driver to fail early.
> 
> Use devm_clk_get_optional() instead to return to the previous
> behaviour, allowing the use of a static clock source.
> 
> Cc: stable@vger.kernel.org
> Fixes: abae8e57e49a ("clk: generalize devm_clk_get() a bit")

Are you sure you identified the breaking commit correctly? I intended
the patch not to introduce any semantic change, and even with your claim
I don't spot the issue in abae8e57e49a.

Best regards
Uwe

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 488 bytes --]

Re: [PATCH] ASoC: ES8389: convert to devm_clk_get_optional() to get clock

[Li Jian]

Hi Uwe

在 2026/4/17 21:34, Uwe Kleine-König 写道:

> Hello,
> 
> On Fri, Apr 17, 2026 at 06:53:14PM +0800, Li Jian wrote:
>> When enabling ES8390 via ACPI description, es8389 would fail to
>> obtain a clock source, causing the driver to fail to initialize.
>> This was not an issue with older kernels, but since commit
>> abae8e57e49a ("clk: generalize devm_clk_get() a bit"),
>> devm_clk_get() would return an error pointer when a clock source
>> was not detected (instead of falling back to a static clock),
>> causing the driver to fail early.
>>
>> Use devm_clk_get_optional() instead to return to the previous
>> behaviour, allowing the use of a static clock source.
>>
>> Cc: stable@vger.kernel.org
>> Fixes: abae8e57e49a ("clk: generalize devm_clk_get() a bit")
> 
> Are you sure you identified the breaking commit correctly? I intended
> the patch not to introduce any semantic change, and even with your claim
> I don't spot the issue in abae8e57e49a.
> 


There was a misunderstanding on how the Fixes: tag should be used - I 
meant to say that your commit changed a behaviour, not that it was 
broken. I should have pointed to a commit to this driver instead.

In my case, since the device was described in ACPI and it does not 
export a clock to the operating system, it was then necessary to utilize 
a fallback. Before your commit, missing clocks returned a NULL pointer. 
However, your commit correctly makes it return an error pointer instead 
- now, since the driver initialization checks for error pointers, so I 
switched to devm_clk_get_optional(), which falls back to a static clock 
source to avoid this issue on ACPI platforms.

Again, your commit is correct, but I was not using the Fixes: tag 
correct - indeed, a missing clock should return an error pointer, not 
NULL. Sorry about that and I will submit v2 to fix this.

Best regards,
Li Jian

 > Best regards
 > Uwe

Re: [PATCH] ASoC: ES8389: convert to devm_clk_get_optional() to get clock

[Uwe Kleine-König]

[-- Attachment #1: Type: text/plain, Size: 2076 bytes --]

Hello,

[Fixed address of loongarch mailing list]

On Sat, Apr 18, 2026 at 04:18:10PM +0800, Li Jian wrote:
> 在 2026/4/17 21:34, Uwe Kleine-König 写道:
> > On Fri, Apr 17, 2026 at 06:53:14PM +0800, Li Jian wrote:
> > > When enabling ES8390 via ACPI description, es8389 would fail to
> > > obtain a clock source, causing the driver to fail to initialize.
> > > This was not an issue with older kernels, but since commit
> > > abae8e57e49a ("clk: generalize devm_clk_get() a bit"),
> > > devm_clk_get() would return an error pointer when a clock source
> > > was not detected (instead of falling back to a static clock),
> > > causing the driver to fail early.
> > > 
> > > Use devm_clk_get_optional() instead to return to the previous
> > > behaviour, allowing the use of a static clock source.
> > > 
> > > Cc: stable@vger.kernel.org
> > > Fixes: abae8e57e49a ("clk: generalize devm_clk_get() a bit")
> > 
> > Are you sure you identified the breaking commit correctly? I intended
> > the patch not to introduce any semantic change, and even with your claim
> > I don't spot the issue in abae8e57e49a.
> 
> There was a misunderstanding on how the Fixes: tag should be used - I meant
> to say that your commit changed a behaviour, not that it was broken. I

In principle the usage is fine: *If* a commit changed behaviour and
failed to adapt a user relying on the old behaviour, that warrants a
Fixes line.

*But* I claim the commit didn't change behaviour. devm_clk_get()
returned and returns whatever clk_get() returned; with and without
abae8e57e49a. And clk_get() wasn't touched in my commit.

> should have pointed to a commit to this driver instead.
> 
> In my case, since the device was described in ACPI and it does not export a
> clock to the operating system, it was then necessary to utilize a fallback.
> Before your commit, missing clocks returned a NULL pointer.

So I'm pretty sure you're wrong here. Did you try reverting abae8e57e49a
on a broken state and confirm that fixes things for you?

Best regards
Uwe

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 488 bytes --]

[PATCH] ASoC: ES8389: convert to devm_clk_get_optional() to get clock

[Li Jian]

When enabling ES8390 via ACPI description, es8389 would fail to
obtain a clock source, causing the driver to fail to initialize.
This was not an issue with older kernels, but since commit
abae8e57e49a ("clk: generalize devm_clk_get() a bit"),
devm_clk_get() would return an error pointer when a clock source
was not detected (instead of falling back to a static clock),
causing the driver to fail early.

Use devm_clk_get_optional() instead to return to the previous
behaviour, allowing the use of a static clock source.

Cc: stable@vger.kernel.org
Fixes: abae8e57e49a ("clk: generalize devm_clk_get() a bit")
Signed-off-by: Li Jian <lazycat-xiao@foxmail.com>
---
 sound/soc/codecs/es8389.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/sound/soc/codecs/es8389.c b/sound/soc/codecs/es8389.c
index 8d418cae3..449d9574b 100644
--- a/sound/soc/codecs/es8389.c
+++ b/sound/soc/codecs/es8389.c
@@ -892,7 +892,7 @@ static int es8389_probe(struct snd_soc_component *component)
 		return ret;
 	}
 
-	es8389->mclk = devm_clk_get(component->dev, "mclk");
+	es8389->mclk = devm_clk_get_optional(component->dev, "mclk");
 	if (IS_ERR(es8389->mclk))
 		return dev_err_probe(component->dev, PTR_ERR(es8389->mclk),
 			"ES8389 is unable to get mclk\n");
-- 
2.47.3