[PATCH] iio: adc: qcom-spmi-adc5-gen3: Allocate unique ADC_TM aux device IDs

[PATCH] iio: adc: qcom-spmi-adc5-gen3: Allocate unique ADC_TM aux device IDs

[Shawn Guo]

adc5_gen3_add_aux_tm_device() never sets the auxiliary device id, so
every ADC5 Gen3 instance with at least one qcom,adc-tm channel tries
to register the same device name. With ADC_TM channels on more than
one PMIC, the second instance to probe fails:

  sysfs: cannot create duplicate filename '/bus/auxiliary/devices/qcom_spmi_adc5_gen3.adc5_tm_gen3.0'
  ...
  auxiliary qcom_spmi_adc5_gen3.adc5_tm_gen3.0: adding auxiliary device failed!: -17
  qcom-spmi-adc5-gen3 c400000.spmi:pmic@8:adc@7200: error -EEXIST: Failed to add auxiliary TM device

The ADC probe then fails as a whole, so that PMIC loses both its IIO
channels and its thermal zones. This shows up on Nord, which has
thermistor channels on two PMAU0102 PMICs.

Allocate the id from a driver-wide IDA and release it from the
auxiliary device release callback. The ADC_TM driver matches on the
device name only, so the instance number does not affect binding.

Assisted-by: LLM
Signed-off-by: Shawn Guo <shengchao.guo@oss.qualcomm.com>
---
 drivers/iio/adc/qcom-spmi-adc5-gen3.c | 21 ++++++++++++++++++---
 1 file changed, 18 insertions(+), 3 deletions(-)

diff --git a/drivers/iio/adc/qcom-spmi-adc5-gen3.c b/drivers/iio/adc/qcom-spmi-adc5-gen3.c
index c68c6c5f6aca..f0b2544c1632 100644
--- a/drivers/iio/adc/qcom-spmi-adc5-gen3.c
+++ b/drivers/iio/adc/qcom-spmi-adc5-gen3.c
@@ -15,6 +15,7 @@
 #include <linux/dev_printk.h>
 #include <linux/err.h>
 #include <linux/export.h>
+#include <linux/idr.h>
 #include <linux/iio/adc/qcom-adc5-gen3-common.h>
 #include <linux/iio/iio.h>
 #include <linux/interrupt.h>
@@ -608,7 +609,14 @@ static void adc5_gen3_delete_aux(void *data)
 	auxiliary_device_delete(data);
 }
 
-static void adc5_gen3_aux_device_release(struct device *dev) {}
+static DEFINE_IDA(adc5_gen3_aux_ida);
+
+static void adc5_gen3_aux_device_release(struct device *dev)
+{
+	struct auxiliary_device *adev = to_auxiliary_dev(dev);
+
+	ida_free(&adc5_gen3_aux_ida, adev->id);
+}
 
 static int adc5_gen3_add_aux_tm_device(struct adc5_chip *adc)
 {
@@ -642,9 +650,16 @@ static int adc5_gen3_add_aux_tm_device(struct adc5_chip *adc)
 
 	aux_device->n_tm_channels = adc->n_tm_channels;
 
-	ret = auxiliary_device_init(&aux_device->aux_dev);
-	if (ret)
+	ret = ida_alloc(&adc5_gen3_aux_ida, GFP_KERNEL);
+	if (ret < 0)
 		return ret;
+	aux_device->aux_dev.id = ret;
+
+	ret = auxiliary_device_init(&aux_device->aux_dev);
+	if (ret) {
+		ida_free(&adc5_gen3_aux_ida, aux_device->aux_dev.id);
+		return ret;
+	}
 
 	ret = devm_add_action_or_reset(adc->dev, adc5_gen3_uninit_aux,
 				       &aux_device->aux_dev);
-- 
2.43.0

Re: [PATCH] iio: adc: qcom-spmi-adc5-gen3: Allocate unique ADC_TM aux device IDs

[Bartosz Golaszewski]

On Thu, 8 Oct 2026 05:11:54 +0200, Shawn Guo
<shengchao.guo@oss.qualcomm.com> said:
> adc5_gen3_add_aux_tm_device() never sets the auxiliary device id, so
> every ADC5 Gen3 instance with at least one qcom,adc-tm channel tries
> to register the same device name. With ADC_TM channels on more than
> one PMIC, the second instance to probe fails:
>
>   sysfs: cannot create duplicate filename '/bus/auxiliary/devices/qcom_spmi_adc5_gen3.adc5_tm_gen3.0'
>   ...
>   auxiliary qcom_spmi_adc5_gen3.adc5_tm_gen3.0: adding auxiliary device failed!: -17
>   qcom-spmi-adc5-gen3 c400000.spmi:pmic@8:adc@7200: error -EEXIST: Failed to add auxiliary TM device
>
> The ADC probe then fails as a whole, so that PMIC loses both its IIO
> channels and its thermal zones. This shows up on Nord, which has
> thermistor channels on two PMAU0102 PMICs.
>
> Allocate the id from a driver-wide IDA and release it from the
> auxiliary device release callback. The ADC_TM driver matches on the
> device name only, so the instance number does not affect binding.
>
> Assisted-by: LLM
> Signed-off-by: Shawn Guo <shengchao.guo@oss.qualcomm.com>
> ---

Reviewed-by: Bartosz Golaszewski <bartosz.golaszewski@oss.qualcomm.com>

Re: [PATCH] iio: adc: qcom-spmi-adc5-gen3: Allocate unique ADC_TM aux device IDs

[Jishnu Prakash]

On 10/8/2026 8:41 AM, Shawn Guo wrote:
> adc5_gen3_add_aux_tm_device() never sets the auxiliary device id, so
> every ADC5 Gen3 instance with at least one qcom,adc-tm channel tries
> to register the same device name. With ADC_TM channels on more than
> one PMIC, the second instance to probe fails:
> 
>   sysfs: cannot create duplicate filename '/bus/auxiliary/devices/qcom_spmi_adc5_gen3.adc5_tm_gen3.0'
>   ...
>   auxiliary qcom_spmi_adc5_gen3.adc5_tm_gen3.0: adding auxiliary device failed!: -17
>   qcom-spmi-adc5-gen3 c400000.spmi:pmic@8:adc@7200: error -EEXIST: Failed to add auxiliary TM device
> 
> The ADC probe then fails as a whole, so that PMIC loses both its IIO
> channels and its thermal zones. This shows up on Nord, which has
> thermistor channels on two PMAU0102 PMICs.
> 
> Allocate the id from a driver-wide IDA and release it from the
> auxiliary device release callback. The ADC_TM driver matches on the
> device name only, so the instance number does not affect binding.
> 
> Assisted-by: LLM
> Signed-off-by: Shawn Guo <shengchao.guo@oss.qualcomm.com>
> ---
>  drivers/iio/adc/qcom-spmi-adc5-gen3.c | 21 ++++++++++++++++++---
>  1 file changed, 18 insertions(+), 3 deletions(-)
> 
> diff --git a/drivers/iio/adc/qcom-spmi-adc5-gen3.c b/drivers/iio/adc/qcom-spmi-adc5-gen3.c
> index c68c6c5f6aca..f0b2544c1632 100644
> --- a/drivers/iio/adc/qcom-spmi-adc5-gen3.c
> +++ b/drivers/iio/adc/qcom-spmi-adc5-gen3.c
> @@ -15,6 +15,7 @@
>  #include <linux/dev_printk.h>
>  #include <linux/err.h>
>  #include <linux/export.h>
> +#include <linux/idr.h>
>  #include <linux/iio/adc/qcom-adc5-gen3-common.h>
>  #include <linux/iio/iio.h>
>  #include <linux/interrupt.h>
> @@ -608,7 +609,14 @@ static void adc5_gen3_delete_aux(void *data)
>  	auxiliary_device_delete(data);
>  }
>  
> -static void adc5_gen3_aux_device_release(struct device *dev) {}
> +static DEFINE_IDA(adc5_gen3_aux_ida);
> +
> +static void adc5_gen3_aux_device_release(struct device *dev)
> +{
> +	struct auxiliary_device *adev = to_auxiliary_dev(dev);
> +
> +	ida_free(&adc5_gen3_aux_ida, adev->id);
> +}
>  
>  static int adc5_gen3_add_aux_tm_device(struct adc5_chip *adc)
>  {
> @@ -642,9 +650,16 @@ static int adc5_gen3_add_aux_tm_device(struct adc5_chip *adc)
>  
>  	aux_device->n_tm_channels = adc->n_tm_channels;
>  
> -	ret = auxiliary_device_init(&aux_device->aux_dev);
> -	if (ret)
> +	ret = ida_alloc(&adc5_gen3_aux_ida, GFP_KERNEL);
> +	if (ret < 0)
>  		return ret;
> +	aux_device->aux_dev.id = ret;
> +
> +	ret = auxiliary_device_init(&aux_device->aux_dev);
> +	if (ret) {
> +		ida_free(&adc5_gen3_aux_ida, aux_device->aux_dev.id);
> +		return ret;
> +	}

Thanks, this looks right.

Reviewed-by: Jishnu Prakash <jishnu.prakash@oss.qualcomm.com>

>  
>  	ret = devm_add_action_or_reset(adc->dev, adc5_gen3_uninit_aux,
>  				       &aux_device->aux_dev);

Re: [PATCH] iio: adc: qcom-spmi-adc5-gen3: Allocate unique ADC_TM aux device IDs

[Jonathan Cameron]

On Thu,  8 Oct 2026 11:11:54 +0800
Shawn Guo <shengchao.guo@oss.qualcomm.com> wrote:

> adc5_gen3_add_aux_tm_device() never sets the auxiliary device id, so
> every ADC5 Gen3 instance with at least one qcom,adc-tm channel tries
> to register the same device name. With ADC_TM channels on more than
> one PMIC, the second instance to probe fails:
> 
>   sysfs: cannot create duplicate filename '/bus/auxiliary/devices/qcom_spmi_adc5_gen3.adc5_tm_gen3.0'
>   ...
>   auxiliary qcom_spmi_adc5_gen3.adc5_tm_gen3.0: adding auxiliary device failed!: -17
>   qcom-spmi-adc5-gen3 c400000.spmi:pmic@8:adc@7200: error -EEXIST: Failed to add auxiliary TM device
> 
> The ADC probe then fails as a whole, so that PMIC loses both its IIO
> channels and its thermal zones. This shows up on Nord, which has
> thermistor channels on two PMAU0102 PMICs.

Fixes tag?  Just replying to this thread is fine rather than sending
a v2 just for that.

> 
> Allocate the id from a driver-wide IDA and release it from the
> auxiliary device release callback. The ADC_TM driver matches on the
> device name only, so the instance number does not affect binding.
> 
> Assisted-by: LLM
> Signed-off-by: Shawn Guo <shengchao.guo@oss.qualcomm.com>
> ---
>  drivers/iio/adc/qcom-spmi-adc5-gen3.c | 21 ++++++++++++++++++---
>  1 file changed, 18 insertions(+), 3 deletions(-)
> 
> diff --git a/drivers/iio/adc/qcom-spmi-adc5-gen3.c b/drivers/iio/adc/qcom-spmi-adc5-gen3.c
> index c68c6c5f6aca..f0b2544c1632 100644
> --- a/drivers/iio/adc/qcom-spmi-adc5-gen3.c
> +++ b/drivers/iio/adc/qcom-spmi-adc5-gen3.c
> @@ -15,6 +15,7 @@
>  #include <linux/dev_printk.h>
>  #include <linux/err.h>
>  #include <linux/export.h>
> +#include <linux/idr.h>
>  #include <linux/iio/adc/qcom-adc5-gen3-common.h>
>  #include <linux/iio/iio.h>
>  #include <linux/interrupt.h>
> @@ -608,7 +609,14 @@ static void adc5_gen3_delete_aux(void *data)
>  	auxiliary_device_delete(data);
>  }
>  
> -static void adc5_gen3_aux_device_release(struct device *dev) {}
> +static DEFINE_IDA(adc5_gen3_aux_ida);
> +
> +static void adc5_gen3_aux_device_release(struct device *dev)
> +{
> +	struct auxiliary_device *adev = to_auxiliary_dev(dev);
> +
> +	ida_free(&adc5_gen3_aux_ida, adev->id);
> +}
>  
>  static int adc5_gen3_add_aux_tm_device(struct adc5_chip *adc)
>  {
> @@ -642,9 +650,16 @@ static int adc5_gen3_add_aux_tm_device(struct adc5_chip *adc)
>  
>  	aux_device->n_tm_channels = adc->n_tm_channels;
>  
> -	ret = auxiliary_device_init(&aux_device->aux_dev);
> -	if (ret)
> +	ret = ida_alloc(&adc5_gen3_aux_ida, GFP_KERNEL);
> +	if (ret < 0)
>  		return ret;
> +	aux_device->aux_dev.id = ret;
> +
> +	ret = auxiliary_device_init(&aux_device->aux_dev);
> +	if (ret) {
> +		ida_free(&adc5_gen3_aux_ida, aux_device->aux_dev.id);
> +		return ret;
> +	}
>  
>  	ret = devm_add_action_or_reset(adc->dev, adc5_gen3_uninit_aux,
>  				       &aux_device->aux_dev);

Re: [PATCH] iio: adc: qcom-spmi-adc5-gen3: Allocate unique ADC_TM aux device IDs

[Shawn Guo]

On Thu, Oct 08, 2026 at 11:11:54AM +0800, Shawn Guo wrote:
> adc5_gen3_add_aux_tm_device() never sets the auxiliary device id, so
> every ADC5 Gen3 instance with at least one qcom,adc-tm channel tries
> to register the same device name. With ADC_TM channels on more than
> one PMIC, the second instance to probe fails:
> 
>   sysfs: cannot create duplicate filename '/bus/auxiliary/devices/qcom_spmi_adc5_gen3.adc5_tm_gen3.0'
>   ...
>   auxiliary qcom_spmi_adc5_gen3.adc5_tm_gen3.0: adding auxiliary device failed!: -17
>   qcom-spmi-adc5-gen3 c400000.spmi:pmic@8:adc@7200: error -EEXIST: Failed to add auxiliary TM device
> 
> The ADC probe then fails as a whole, so that PMIC loses both its IIO
> channels and its thermal zones. This shows up on Nord, which has
> thermistor channels on two PMAU0102 PMICs.
> 
> Allocate the id from a driver-wide IDA and release it from the
> auxiliary device release callback. The ADC_TM driver matches on the
> device name only, so the instance number does not affect binding.
> 

Fixes: baff45179e90 ("iio: adc: Add support for QCOM PMIC5 Gen3 ADC")

> Assisted-by: LLM
> Signed-off-by: Shawn Guo <shengchao.guo@oss.qualcomm.com>

Re: [PATCH] iio: adc: qcom-spmi-adc5-gen3: Allocate unique ADC_TM aux device IDs

[Shawn Guo]

On Sun, Oct 18, 2026 at 07:21:51PM +0100, Jonathan Cameron wrote:
> On Thu,  8 Oct 2026 11:11:54 +0800
> Shawn Guo <shengchao.guo@oss.qualcomm.com> wrote:
> 
> > adc5_gen3_add_aux_tm_device() never sets the auxiliary device id, so
> > every ADC5 Gen3 instance with at least one qcom,adc-tm channel tries
> > to register the same device name. With ADC_TM channels on more than
> > one PMIC, the second instance to probe fails:
> > 
> >   sysfs: cannot create duplicate filename '/bus/auxiliary/devices/qcom_spmi_adc5_gen3.adc5_tm_gen3.0'
> >   ...
> >   auxiliary qcom_spmi_adc5_gen3.adc5_tm_gen3.0: adding auxiliary device failed!: -17
> >   qcom-spmi-adc5-gen3 c400000.spmi:pmic@8:adc@7200: error -EEXIST: Failed to add auxiliary TM device
> > 
> > The ADC probe then fails as a whole, so that PMIC loses both its IIO
> > channels and its thermal zones. This shows up on Nord, which has
> > thermistor channels on two PMAU0102 PMICs.
> 
> Fixes tag?  Just replying to this thread is fine rather than sending
> a v2 just for that.

Replied to the patch with a hope that b4 will be able to pick up
the Fixes tag.

Thanks,
Shawn