coresight: Add coresight name support

  Add coresight name support for custom names which will be
easy to identify the device by the name.

Signed-off-by: Mao Jinlong <quic_jinlmao@quicinc.com>
---
 .../hwtracing/coresight/coresight-cti-core.c  | 20 ++++++++------
 drivers/hwtracing/coresight/coresight-dummy.c | 10 ++++---
 .../hwtracing/coresight/coresight-platform.c  | 27 +++++++++++++++++++
 drivers/hwtracing/coresight/coresight-tpdm.c  | 10 ++++---
 include/linux/coresight.h                     |  1 +
 5 files changed, 53 insertions(+), 15 deletions(-)
  

On 28/12/2023 09:33, Mao Jinlong wrote:
> Add coresight name support for custom names which will be
> easy to identify the device by the name.
> 

I suppose this is more of a V2 because the subject is the same as the
one sent earlier this year. But it looks like the discussion on the
previous one wasn't resolved.

With the main issues to solve being:

 * It would be nice to use the existing root node name instead of adding
   a new property. But at the same time DT nodes are supposed to have
   generic names.

 * This only works for DT and not ACPI

To me it seems like adding the new property is just a "cheat" to get
around not being allowed to have a specific name for the root node. But
if we admit that we need a name I don't see the benefit of not putting
the name where the node is already named.

Using the root node name at this point would also undo the hard coded
per-cpu naming of the CTI and ETM devices, so maybe it would be nice,
but it's just too late. That means that a new field is necessary.
Although that field could be a boolean like "use-root-name-for-display"
or something like that. In the end it probably doesn't really make a
difference whether it's that or a name string.

And maybe the answer to the ACPI question is just that if anyone needs
it, they can add it in the future. It doesn't seem like it would
conflict with anything we do here.

> Signed-off-by: Mao Jinlong <quic_jinlmao@quicinc.com>
> ---
>  .../hwtracing/coresight/coresight-cti-core.c  | 20 ++++++++------
>  drivers/hwtracing/coresight/coresight-dummy.c | 10 ++++---
>  .../hwtracing/coresight/coresight-platform.c  | 27 +++++++++++++++++++
>  drivers/hwtracing/coresight/coresight-tpdm.c  | 10 ++++---
>  include/linux/coresight.h                     |  1 +
>  5 files changed, 53 insertions(+), 15 deletions(-)
> 
> diff --git a/drivers/hwtracing/coresight/coresight-cti-core.c b/drivers/hwtracing/coresight/coresight-cti-core.c
> index 3999d0a2cb60..60a1e76064a9 100644
> --- a/drivers/hwtracing/coresight/coresight-cti-core.c
> +++ b/drivers/hwtracing/coresight/coresight-cti-core.c
> @@ -902,14 +902,18 @@ static int cti_probe(struct amba_device *adev, const struct amba_id *id)
>  	/* default to powered - could change on PM notifications */
>  	drvdata->config.hw_powered = true;
>  
> -	/* set up device name - will depend if cpu bound or otherwise */
> -	if (drvdata->ctidev.cpu >= 0)
> -		cti_desc.name = devm_kasprintf(dev, GFP_KERNEL, "cti_cpu%d",
> -					       drvdata->ctidev.cpu);
> -	else
> -		cti_desc.name = coresight_alloc_device_name(&cti_sys_devs, dev);

Can we put the new name stuff inside coresight_alloc_device_name()? Then
it happens by default for every device.

I know Suzuki said previously to do it per-device, but the new DT
property is just "coresight-name", so it's generic. Rather than being
specific like "cti-name". So I don't see the benefit of duplicating the
code at this point if we do decide to do it.

> -	if (!cti_desc.name)
> -		return -ENOMEM;
> +	cti_desc.name = coresight_get_device_name(dev);
> +	if (!cti_desc.name) {
> +		/* set up device name - will depend if cpu bound or otherwise */
> +		if (drvdata->ctidev.cpu >= 0)
> +			cti_desc.name = devm_kasprintf(dev, GFP_KERNEL, "cti_cpu%d",
> +						       drvdata->ctidev.cpu);
> +		else {
> +			cti_desc.name = coresight_alloc_device_name(&cti_sys_devs, dev);
> +			if (!cti_desc.name)
> +				return -ENOMEM;
> +		}
> +	}

>  
>  	/* setup CPU power management handling for CPU bound CTI devices. */
>  	ret = cti_pm_setup(drvdata);
> diff --git a/drivers/hwtracing/coresight/coresight-dummy.c b/drivers/hwtracing/coresight/coresight-dummy.c
> index e4deafae7bc2..b19cd400df79 100644
> --- a/drivers/hwtracing/coresight/coresight-dummy.c
> +++ b/drivers/hwtracing/coresight/coresight-dummy.c
> @@ -76,10 +76,12 @@ static int dummy_probe(struct platform_device *pdev)
>  	struct coresight_desc desc = { 0 };
>  
>  	if (of_device_is_compatible(node, "arm,coresight-dummy-source")) {
> -
> -		desc.name = coresight_alloc_device_name(&source_devs, dev);
> -		if (!desc.name)
> -			return -ENOMEM;
> +		desc.name = coresight_get_device_name(dev);
> +		if (!desc.name) {
> +			desc.name = coresight_alloc_device_name(&source_devs, dev);
> +			if (!desc.name)
> +				return -ENOMEM;
> +		}
>  
>  		desc.type = CORESIGHT_DEV_TYPE_SOURCE;
>  		desc.subtype.source_subtype =
> diff --git a/drivers/hwtracing/coresight/coresight-platform.c b/drivers/hwtracing/coresight/coresight-platform.c
> index 9d550f5697fa..284aa22a06b7 100644
> --- a/drivers/hwtracing/coresight/coresight-platform.c
> +++ b/drivers/hwtracing/coresight/coresight-platform.c
> @@ -183,6 +183,18 @@ static int of_coresight_get_cpu(struct device *dev)
>  	return cpu;
>  }
>  
> +static const char *of_coresight_get_device_name(struct device *dev)
> +{
> +	const char *name = NULL;
> +
> +	if (!dev->of_node)
> +		return NULL;
> +
> +	of_property_read_string(dev->of_node, "coresight-name", &name);

Do you need to update the binding docs with this new property?

Also a minor nit: Maybe "display-name" is better? "Coresight" is
implied, and the node is already named, although that node name isn't
used for display purposes, but this one is.

Thanks
James
  

As James mentions this is clearly a V2 of a previous patch - please
mark as such in future.

Adding to what James has already said:-

1) Mapping between the canonical names used in the drivers and the
information as to the precise device is as easy as running 'ls' on
/sys/bus/coresight/devices:-

root@linaro-developer:/home/linaro/cs-mods# ls -al /sys/bus/coresight/devices/
total 0
drwxr-xr-x 2 root root 0 Jan  2 11:27 .
drwxr-xr-x 4 root root 0 Jan  2 11:27 ..
lrwxrwxrwx 1 root root 0 Jan  2 11:27 cti_cpu0 ->
../../../devices/platform/soc@0/858000.cti/cti_cpu0
lrwxrwxrwx 1 root root 0 Jan  2 11:27 cti_cpu1 ->
../../../devices/platform/soc@0/859000.cti/cti_cpu1
lrwxrwxrwx 1 root root 0 Jan  2 11:27 cti_cpu2 ->
../../../devices/platform/soc@0/85a000.cti/cti_cpu2
lrwxrwxrwx 1 root root 0 Jan  2 11:27 cti_cpu3 ->
../../../devices/platform/soc@0/85b000.cti/cti_cpu3
lrwxrwxrwx 1 root root 0 Jan  2 11:27 cti_sys0 ->
../../../devices/platform/soc@0/810000.cti/cti_sys0
lrwxrwxrwx 1 root root 0 Jan  2 11:27 cti_sys1 ->
../../../devices/platform/soc@0/811000.cti/cti_sys1
lrwxrwxrwx 1 root root 0 Jan  2 11:27 etm0 ->
../../../devices/platform/soc@0/85c000.etm/etm0
lrwxrwxrwx 1 root root 0 Jan  2 11:27 etm1 ->
../../../devices/platform/soc@0/85d000.etm/etm1
lrwxrwxrwx 1 root root 0 Jan  2 11:27 etm2 ->
../../../devices/platform/soc@0/85e000.etm/etm2
lrwxrwxrwx 1 root root 0 Jan  2 11:27 etm3 ->
../../../devices/platform/soc@0/85f000.etm/etm3
lrwxrwxrwx 1 root root 0 Jan  2 11:42 funnel0 ->
../../../devices/platform/soc@0/821000.funnel/funnel0
lrwxrwxrwx 1 root root 0 Jan  2 11:42 funnel1 ->
../../../devices/platform/soc@0/841000.funnel/funnel1
lrwxrwxrwx 1 root root 0 Jan  2 11:42 replicator0 ->
../../../devices/platform/soc@0/824000.replicator/replicator0
lrwxrwxrwx 1 root root 0 Jan  2 11:42 tmc_etf0 ->
../../../devices/platform/soc@0/825000.etf/tmc_etf0
lrwxrwxrwx 1 root root 0 Jan  2 11:42 tmc_etr0 ->
../../../devices/platform/soc@0/826000.etr/tmc_etr0


2) The patch set must contain the usage and specification in the .yaml
 file(s) of the property used.

However, there was a standard property called 'name' which is
deprecated - see
https://devicetree-specification.readthedocs.io/en/latest/chapter2-devicetree-basics.html
section 2.3.11. I do not believe that adding another 'name' property
would be accepted by the DT maintainers.

3) the 'device_node' structure has a 'name' field that contains the
node name in the DT approved "node-name@unit-address" format.This
contains whatever node names you used in the dt.  Why not use this if
a change has to be made and find some conditional to activate it.

However, given point 1) above, the problem is solved and the patch
adds no new information not already available.

Regards

Mike

On Thu, 28 Dec 2023 at 11:26, James Clark <james.clark@arm.com> wrote:
>
>
>
> On 28/12/2023 09:33, Mao Jinlong wrote:
> > Add coresight name support for custom names which will be
> > easy to identify the device by the name.
> >
>
> I suppose this is more of a V2 because the subject is the same as the
> one sent earlier this year. But it looks like the discussion on the
> previous one wasn't resolved.
>
> With the main issues to solve being:
>
>  * It would be nice to use the existing root node name instead of adding
>    a new property. But at the same time DT nodes are supposed to have
>    generic names.
>
>  * This only works for DT and not ACPI
>
> To me it seems like adding the new property is just a "cheat" to get
> around not being allowed to have a specific name for the root node. But
> if we admit that we need a name I don't see the benefit of not putting
> the name where the node is already named.
>
> Using the root node name at this point would also undo the hard coded
> per-cpu naming of the CTI and ETM devices, so maybe it would be nice,
> but it's just too late. That means that a new field is necessary.
> Although that field could be a boolean like "use-root-name-for-display"
> or something like that. In the end it probably doesn't really make a
> difference whether it's that or a name string.
>
> And maybe the answer to the ACPI question is just that if anyone needs
> it, they can add it in the future. It doesn't seem like it would
> conflict with anything we do here.
>
> > Signed-off-by: Mao Jinlong <quic_jinlmao@quicinc.com>
> > ---
> >  .../hwtracing/coresight/coresight-cti-core.c  | 20 ++++++++------
> >  drivers/hwtracing/coresight/coresight-dummy.c | 10 ++++---
> >  .../hwtracing/coresight/coresight-platform.c  | 27 +++++++++++++++++++
> >  drivers/hwtracing/coresight/coresight-tpdm.c  | 10 ++++---
> >  include/linux/coresight.h                     |  1 +
> >  5 files changed, 53 insertions(+), 15 deletions(-)
> >
> > diff --git a/drivers/hwtracing/coresight/coresight-cti-core.c b/drivers/hwtracing/coresight/coresight-cti-core.c
> > index 3999d0a2cb60..60a1e76064a9 100644
> > --- a/drivers/hwtracing/coresight/coresight-cti-core.c
> > +++ b/drivers/hwtracing/coresight/coresight-cti-core.c
> > @@ -902,14 +902,18 @@ static int cti_probe(struct amba_device *adev, const struct amba_id *id)
> >       /* default to powered - could change on PM notifications */
> >       drvdata->config.hw_powered = true;
> >
> > -     /* set up device name - will depend if cpu bound or otherwise */
> > -     if (drvdata->ctidev.cpu >= 0)
> > -             cti_desc.name = devm_kasprintf(dev, GFP_KERNEL, "cti_cpu%d",
> > -                                            drvdata->ctidev.cpu);
> > -     else
> > -             cti_desc.name = coresight_alloc_device_name(&cti_sys_devs, dev);
>
> Can we put the new name stuff inside coresight_alloc_device_name()? Then
> it happens by default for every device.
>
> I know Suzuki said previously to do it per-device, but the new DT
> property is just "coresight-name", so it's generic. Rather than being
> specific like "cti-name". So I don't see the benefit of duplicating the
> code at this point if we do decide to do it.
>
> > -     if (!cti_desc.name)
> > -             return -ENOMEM;
> > +     cti_desc.name = coresight_get_device_name(dev);
> > +     if (!cti_desc.name) {
> > +             /* set up device name - will depend if cpu bound or otherwise */
> > +             if (drvdata->ctidev.cpu >= 0)
> > +                     cti_desc.name = devm_kasprintf(dev, GFP_KERNEL, "cti_cpu%d",
> > +                                                    drvdata->ctidev.cpu);
> > +             else {
> > +                     cti_desc.name = coresight_alloc_device_name(&cti_sys_devs, dev);
> > +                     if (!cti_desc.name)
> > +                             return -ENOMEM;
> > +             }
> > +     }
>
> >
> >       /* setup CPU power management handling for CPU bound CTI devices. */
> >       ret = cti_pm_setup(drvdata);
> > diff --git a/drivers/hwtracing/coresight/coresight-dummy.c b/drivers/hwtracing/coresight/coresight-dummy.c
> > index e4deafae7bc2..b19cd400df79 100644
> > --- a/drivers/hwtracing/coresight/coresight-dummy.c
> > +++ b/drivers/hwtracing/coresight/coresight-dummy.c
> > @@ -76,10 +76,12 @@ static int dummy_probe(struct platform_device *pdev)
> >       struct coresight_desc desc = { 0 };
> >
> >       if (of_device_is_compatible(node, "arm,coresight-dummy-source")) {
> > -
> > -             desc.name = coresight_alloc_device_name(&source_devs, dev);
> > -             if (!desc.name)
> > -                     return -ENOMEM;
> > +             desc.name = coresight_get_device_name(dev);
> > +             if (!desc.name) {
> > +                     desc.name = coresight_alloc_device_name(&source_devs, dev);
> > +                     if (!desc.name)
> > +                             return -ENOMEM;
> > +             }
> >
> >               desc.type = CORESIGHT_DEV_TYPE_SOURCE;
> >               desc.subtype.source_subtype =
> > diff --git a/drivers/hwtracing/coresight/coresight-platform.c b/drivers/hwtracing/coresight/coresight-platform.c
> > index 9d550f5697fa..284aa22a06b7 100644
> > --- a/drivers/hwtracing/coresight/coresight-platform.c
> > +++ b/drivers/hwtracing/coresight/coresight-platform.c
> > @@ -183,6 +183,18 @@ static int of_coresight_get_cpu(struct device *dev)
> >       return cpu;
> >  }
> >
> > +static const char *of_coresight_get_device_name(struct device *dev)
> > +{
> > +     const char *name = NULL;
> > +
> > +     if (!dev->of_node)
> > +             return NULL;
> > +
> > +     of_property_read_string(dev->of_node, "coresight-name", &name);
>
> Do you need to update the binding docs with this new property?
>
> Also a minor nit: Maybe "display-name" is better? "Coresight" is
> implied, and the node is already named, although that node name isn't
> used for display purposes, but this one is.
>
> Thanks
> James
  

On 28/12/2023 11:26, James Clark wrote:
> 
> 
> On 28/12/2023 09:33, Mao Jinlong wrote:
>> Add coresight name support for custom names which will be
>> easy to identify the device by the name.
>>
> 
> I suppose this is more of a V2 because the subject is the same as the
> one sent earlier this year. But it looks like the discussion on the
> previous one wasn't resolved.
> 
> With the main issues to solve being:
> 
>   * It would be nice to use the existing root node name instead of adding
>     a new property. But at the same time DT nodes are supposed to have
>     generic names.
> 
>   * This only works for DT and not ACPI
> 
> To me it seems like adding the new property is just a "cheat" to get
> around not being allowed to have a specific name for the root node. But
> if we admit that we need a name I don't see the benefit of not putting
> the name where the node is already named.
> 
> Using the root node name at this point would also undo the hard coded
> per-cpu naming of the CTI and ETM devices, so maybe it would be nice,
> but it's just too late. That means that a new field is necessary.

The CTI and ETM can be handled as special cases, like they are
already done and fall back to the nodename for the rest ?
But, I thought the node names must be generic (e.g, cti) and doesn't
really solve the naming requirements for naming CTIs. (e.g,
<device>_tpda, etr_cti). Is there something I missed ?

> Although that field could be a boolean like "use-root-name-for-display"
> or something like that. In the end it probably doesn't really make a
> difference whether it's that or a name string. >
> And maybe the answer to the ACPI question is just that if anyone needs
> it, they can add it in the future. It doesn't seem like it would
> conflict with anything we do here.
> 
>> Signed-off-by: Mao Jinlong <quic_jinlmao@quicinc.com>
>> ---
>>   .../hwtracing/coresight/coresight-cti-core.c  | 20 ++++++++------
>>   drivers/hwtracing/coresight/coresight-dummy.c | 10 ++++---
>>   .../hwtracing/coresight/coresight-platform.c  | 27 +++++++++++++++++++
>>   drivers/hwtracing/coresight/coresight-tpdm.c  | 10 ++++---
>>   include/linux/coresight.h                     |  1 +
>>   5 files changed, 53 insertions(+), 15 deletions(-)
>>
>> diff --git a/drivers/hwtracing/coresight/coresight-cti-core.c b/drivers/hwtracing/coresight/coresight-cti-core.c
>> index 3999d0a2cb60..60a1e76064a9 100644
>> --- a/drivers/hwtracing/coresight/coresight-cti-core.c
>> +++ b/drivers/hwtracing/coresight/coresight-cti-core.c
>> @@ -902,14 +902,18 @@ static int cti_probe(struct amba_device *adev, const struct amba_id *id)
>>   	/* default to powered - could change on PM notifications */
>>   	drvdata->config.hw_powered = true;
>>   
>> -	/* set up device name - will depend if cpu bound or otherwise */
>> -	if (drvdata->ctidev.cpu >= 0)
>> -		cti_desc.name = devm_kasprintf(dev, GFP_KERNEL, "cti_cpu%d",
>> -					       drvdata->ctidev.cpu);
>> -	else
>> -		cti_desc.name = coresight_alloc_device_name(&cti_sys_devs, dev);
> 
> Can we put the new name stuff inside coresight_alloc_device_name()? Then
> it happens by default for every device.

+1

> 
> I know Suzuki said previously to do it per-device, but the new DT
> property is just "coresight-name", so it's generic. Rather than being
> specific like "cti-name". So I don't see the benefit of duplicating the
> code at this point if we do decide to do it.

My suggestion was to name the device based on the specific device rather
than following a generic rule for all device. e.g., A TPDM connected to 
modem, could be named as such based on the platform information. It
could be any means, for e.g., tpdm nodes are always children nodes of
the devices they are connected to ? or could have a phandle to point to 
the device they are monitoring etc. And the name could be created from
the "monitoring device name" + tpdm. Also, we do this for CPU bound CTI
and ETMs already, where we name them based on the CPU.

But then the "nodename" is something we explored and it looks like
may not be an option.

> 
>> -	if (!cti_desc.name)
>> -		return -ENOMEM;
>> +	cti_desc.name = coresight_get_device_name(dev);
>> +	if (!cti_desc.name) {
>> +		/* set up device name - will depend if cpu bound or otherwise */
>> +		if (drvdata->ctidev.cpu >= 0)
>> +			cti_desc.name = devm_kasprintf(dev, GFP_KERNEL, "cti_cpu%d",
>> +						       drvdata->ctidev.cpu);
>> +		else {
>> +			cti_desc.name = coresight_alloc_device_name(&cti_sys_devs, dev);
>> +			if (!cti_desc.name)
>> +				return -ENOMEM;
>> +		}
>> +	}

For these special cases, i.e., CPU bound, we should handle them with 
priority.

if (drvdata->ctidev.cpu >= 0)
	name = devm_kasprintf(... "cti_cpu%d", .. cpu);
else
         name = coresight_alloc_device_name(...);

> 
>>   
>>   	/* setup CPU power management handling for CPU bound CTI devices. */
>>   	ret = cti_pm_setup(drvdata);
>> diff --git a/drivers/hwtracing/coresight/coresight-dummy.c b/drivers/hwtracing/coresight/coresight-dummy.c
>> index e4deafae7bc2..b19cd400df79 100644
>> --- a/drivers/hwtracing/coresight/coresight-dummy.c
>> +++ b/drivers/hwtracing/coresight/coresight-dummy.c
>> @@ -76,10 +76,12 @@ static int dummy_probe(struct platform_device *pdev)
>>   	struct coresight_desc desc = { 0 };
>>   
>>   	if (of_device_is_compatible(node, "arm,coresight-dummy-source")) {
>> -
>> -		desc.name = coresight_alloc_device_name(&source_devs, dev);
>> -		if (!desc.name)
>> -			return -ENOMEM;
>> +		desc.name = coresight_get_device_name(dev);
>> +		if (!desc.name) {
>> +			desc.name = coresight_alloc_device_name(&source_devs, dev);
>> +			if (!desc.name)
>> +				return -ENOMEM;
>> +		}
>>   
>>   		desc.type = CORESIGHT_DEV_TYPE_SOURCE;
>>   		desc.subtype.source_subtype =
>> diff --git a/drivers/hwtracing/coresight/coresight-platform.c b/drivers/hwtracing/coresight/coresight-platform.c
>> index 9d550f5697fa..284aa22a06b7 100644
>> --- a/drivers/hwtracing/coresight/coresight-platform.c
>> +++ b/drivers/hwtracing/coresight/coresight-platform.c
>> @@ -183,6 +183,18 @@ static int of_coresight_get_cpu(struct device *dev)
>>   	return cpu;
>>   }
>>   
>> +static const char *of_coresight_get_device_name(struct device *dev)
>> +{
>> +	const char *name = NULL;
>> +
>> +	if (!dev->of_node)
>> +		return NULL;
>> +
>> +	of_property_read_string(dev->of_node, "coresight-name", &name);
> 
> Do you need to update the binding docs with this new property?
> 
> Also a minor nit: Maybe "display-name" is better? "Coresight" is
> implied, and the node is already named, although that node name isn't
> used for display purposes, but this one is.

On that front, the name is used as a "device" name and not simply 
display. So, even "device-name" sounds more appropriate.

Suzuki

> 
> Thanks
> James
  

On Tue, Jan 2, 2024 at 5:05 AM Mike Leach <mike.leach@linaro.org> wrote:
>
> As James mentions this is clearly a V2 of a previous patch - please
> mark as such in future.
>
> Adding to what James has already said:-
>
> 1) Mapping between the canonical names used in the drivers and the
> information as to the precise device is as easy as running 'ls' on
> /sys/bus/coresight/devices:-
>
> root@linaro-developer:/home/linaro/cs-mods# ls -al /sys/bus/coresight/devices/
> total 0
> drwxr-xr-x 2 root root 0 Jan  2 11:27 .
> drwxr-xr-x 4 root root 0 Jan  2 11:27 ..
> lrwxrwxrwx 1 root root 0 Jan  2 11:27 cti_cpu0 ->
> ../../../devices/platform/soc@0/858000.cti/cti_cpu0
> lrwxrwxrwx 1 root root 0 Jan  2 11:27 cti_cpu1 ->
> ../../../devices/platform/soc@0/859000.cti/cti_cpu1
> lrwxrwxrwx 1 root root 0 Jan  2 11:27 cti_cpu2 ->
> ../../../devices/platform/soc@0/85a000.cti/cti_cpu2
> lrwxrwxrwx 1 root root 0 Jan  2 11:27 cti_cpu3 ->
> ../../../devices/platform/soc@0/85b000.cti/cti_cpu3
> lrwxrwxrwx 1 root root 0 Jan  2 11:27 cti_sys0 ->
> ../../../devices/platform/soc@0/810000.cti/cti_sys0
> lrwxrwxrwx 1 root root 0 Jan  2 11:27 cti_sys1 ->
> ../../../devices/platform/soc@0/811000.cti/cti_sys1
> lrwxrwxrwx 1 root root 0 Jan  2 11:27 etm0 ->
> ../../../devices/platform/soc@0/85c000.etm/etm0
> lrwxrwxrwx 1 root root 0 Jan  2 11:27 etm1 ->
> ../../../devices/platform/soc@0/85d000.etm/etm1
> lrwxrwxrwx 1 root root 0 Jan  2 11:27 etm2 ->
> ../../../devices/platform/soc@0/85e000.etm/etm2
> lrwxrwxrwx 1 root root 0 Jan  2 11:27 etm3 ->
> ../../../devices/platform/soc@0/85f000.etm/etm3
> lrwxrwxrwx 1 root root 0 Jan  2 11:42 funnel0 ->
> ../../../devices/platform/soc@0/821000.funnel/funnel0
> lrwxrwxrwx 1 root root 0 Jan  2 11:42 funnel1 ->
> ../../../devices/platform/soc@0/841000.funnel/funnel1
> lrwxrwxrwx 1 root root 0 Jan  2 11:42 replicator0 ->
> ../../../devices/platform/soc@0/824000.replicator/replicator0
> lrwxrwxrwx 1 root root 0 Jan  2 11:42 tmc_etf0 ->
> ../../../devices/platform/soc@0/825000.etf/tmc_etf0
> lrwxrwxrwx 1 root root 0 Jan  2 11:42 tmc_etr0 ->
> ../../../devices/platform/soc@0/826000.etr/tmc_etr0
>
>
> 2) The patch set must contain the usage and specification in the .yaml
>  file(s) of the property used.

For the record, I don't like "coresight-name". I don't have another
suggestion because "easy" is not sufficient reasoning for why this is
needed.

> However, there was a standard property called 'name' which is
> deprecated - see
> https://devicetree-specification.readthedocs.io/en/latest/chapter2-devicetree-basics.html
> section 2.3.11. I do not believe that adding another 'name' property
> would be accepted by the DT maintainers.

"name" is just the node name for anything in the last 15 years. They
used to be separate, but would still mostly be the same. The only case
I found with them different was old PowerPC Macs.

> 3) the 'device_node' structure has a 'name' field that contains the
> node name in the DT approved "node-name@unit-address" format.

Actually, it is without the unit-address. full_name is with the unit-address.

> This
> contains whatever node names you used in the dt.  Why not use this if
> a change has to be made and find some conditional to activate it.

Don't go accessing "name" or "full_name" directly. I intend to get rid
of "name" and generate it from full_name. So use the accessors and
printk specifiers if you need node names.

Rob
  

On 12/28/2023 7:26 PM, James Clark wrote:
> 
> 
> On 28/12/2023 09:33, Mao Jinlong wrote:
>> Add coresight name support for custom names which will be
>> easy to identify the device by the name.
>>
> 
> I suppose this is more of a V2 because the subject is the same as the
> one sent earlier this year. But it looks like the discussion on the
> previous one wasn't resolved.

I will make next patch as V2

> 
> With the main issues to solve being:
> 
>   * It would be nice to use the existing root node name instead of adding
>     a new property. But at the same time DT nodes are supposed to have
>     generic names.
> 
>   * This only works for DT and not ACPI
> 
> To me it seems like adding the new property is just a "cheat" to get
> around not being allowed to have a specific name for the root node. But
> if we admit that we need a name I don't see the benefit of not putting
> the name where the node is already named.
> 
> Using the root node name at this point would also undo the hard coded
> per-cpu naming of the CTI and ETM devices, so maybe it would be nice,
> but it's just too late. That means that a new field is necessary.
> Although that field could be a boolean like "use-root-name-for-display"
> or something like that. In the end it probably doesn't really make a
> difference whether it's that or a name string.
> 
> And maybe the answer to the ACPI question is just that if anyone needs
> it, they can add it in the future. It doesn't seem like it would
> conflict with anything we do here.
> 
>> Signed-off-by: Mao Jinlong <quic_jinlmao@quicinc.com>
>> ---
>>   .../hwtracing/coresight/coresight-cti-core.c  | 20 ++++++++------
>>   drivers/hwtracing/coresight/coresight-dummy.c | 10 ++++---
>>   .../hwtracing/coresight/coresight-platform.c  | 27 +++++++++++++++++++
>>   drivers/hwtracing/coresight/coresight-tpdm.c  | 10 ++++---
>>   include/linux/coresight.h                     |  1 +
>>   5 files changed, 53 insertions(+), 15 deletions(-)
>>
>> diff --git a/drivers/hwtracing/coresight/coresight-cti-core.c b/drivers/hwtracing/coresight/coresight-cti-core.c
>> index 3999d0a2cb60..60a1e76064a9 100644
>> --- a/drivers/hwtracing/coresight/coresight-cti-core.c
>> +++ b/drivers/hwtracing/coresight/coresight-cti-core.c
>> @@ -902,14 +902,18 @@ static int cti_probe(struct amba_device *adev, const struct amba_id *id)
>>   	/* default to powered - could change on PM notifications */
>>   	drvdata->config.hw_powered = true;
>>   
>> -	/* set up device name - will depend if cpu bound or otherwise */
>> -	if (drvdata->ctidev.cpu >= 0)
>> -		cti_desc.name = devm_kasprintf(dev, GFP_KERNEL, "cti_cpu%d",
>> -					       drvdata->ctidev.cpu);
>> -	else
>> -		cti_desc.name = coresight_alloc_device_name(&cti_sys_devs, dev);
> 
> Can we put the new name stuff inside coresight_alloc_device_name()? Then
> it happens by default for every device.
> 
> I know Suzuki said previously to do it per-device, but the new DT
> property is just "coresight-name", so it's generic. Rather than being
> specific like "cti-name". So I don't see the benefit of duplicating the
> code at this point if we do decide to do it.
I will add it inside the coresight_alloc_device_name.
> 
>> -	if (!cti_desc.name)
>> -		return -ENOMEM;
>> +	cti_desc.name = coresight_get_device_name(dev);
>> +	if (!cti_desc.name) {
>> +		/* set up device name - will depend if cpu bound or otherwise */
>> +		if (drvdata->ctidev.cpu >= 0)
>> +			cti_desc.name = devm_kasprintf(dev, GFP_KERNEL, "cti_cpu%d",
>> +						       drvdata->ctidev.cpu);
>> +		else {
>> +			cti_desc.name = coresight_alloc_device_name(&cti_sys_devs, dev);
>> +			if (!cti_desc.name)
>> +				return -ENOMEM;
>> +		}
>> +	}
> 
>>   
>>   	/* setup CPU power management handling for CPU bound CTI devices. */
>>   	ret = cti_pm_setup(drvdata);
>> diff --git a/drivers/hwtracing/coresight/coresight-dummy.c b/drivers/hwtracing/coresight/coresight-dummy.c
>> index e4deafae7bc2..b19cd400df79 100644
>> --- a/drivers/hwtracing/coresight/coresight-dummy.c
>> +++ b/drivers/hwtracing/coresight/coresight-dummy.c
>> @@ -76,10 +76,12 @@ static int dummy_probe(struct platform_device *pdev)
>>   	struct coresight_desc desc = { 0 };
>>   
>>   	if (of_device_is_compatible(node, "arm,coresight-dummy-source")) {
>> -
>> -		desc.name = coresight_alloc_device_name(&source_devs, dev);
>> -		if (!desc.name)
>> -			return -ENOMEM;
>> +		desc.name = coresight_get_device_name(dev);
>> +		if (!desc.name) {
>> +			desc.name = coresight_alloc_device_name(&source_devs, dev);
>> +			if (!desc.name)
>> +				return -ENOMEM;
>> +		}
>>   
>>   		desc.type = CORESIGHT_DEV_TYPE_SOURCE;
>>   		desc.subtype.source_subtype =
>> diff --git a/drivers/hwtracing/coresight/coresight-platform.c b/drivers/hwtracing/coresight/coresight-platform.c
>> index 9d550f5697fa..284aa22a06b7 100644
>> --- a/drivers/hwtracing/coresight/coresight-platform.c
>> +++ b/drivers/hwtracing/coresight/coresight-platform.c
>> @@ -183,6 +183,18 @@ static int of_coresight_get_cpu(struct device *dev)
>>   	return cpu;
>>   }
>>   
>> +static const char *of_coresight_get_device_name(struct device *dev)
>> +{
>> +	const char *name = NULL;
>> +
>> +	if (!dev->of_node)
>> +		return NULL;
>> +
>> +	of_property_read_string(dev->of_node, "coresight-name", &name);
> 
> Do you need to update the binding docs with this new property?
Yes. I will update the binding doc.
> 
> Also a minor nit: Maybe "display-name" is better? "Coresight" is
> implied, and the node is already named, although that node name isn't
> used for display purposes, but this one is.
 From Suzuki's comments, "device-name" should be better. What do you think ?

Thanks
Jinlong Mao
  

On 1/2/2024 8:04 PM, Mike Leach wrote:
> As James mentions this is clearly a V2 of a previous patch - please
> mark as such in future.
> 
> Adding to what James has already said:-
> 
> 1) Mapping between the canonical names used in the drivers and the
> information as to the precise device is as easy as running 'ls' on
> /sys/bus/coresight/devices:-

For the components bounded with CPU, we can easily identify them by the 
number in the name. But for other components, we can only get the 
component type and registers address of it. We can't identify which 
system it belongs to from current name.

lrwxrwxrwx    1 root     0                0 Jan  1 00:01 cti_sys0 -> 
./../../devices/platform/soc@0/138f0000.cti/cti_sys0
lrwxrwxrwx    1 root     0                0 Jan  1 00:01 cti_sys1 -> 
./../../devices/platform/soc@0/13900000.cti/cti_sys1
lrwxrwxrwx    1 root     0                0 Jan  1 00:01 tpdm0 -> 
./../../devices/platform/soc@0/10b0d000.tpdm/tpdm0
lrwxrwxrwx    1 root     0                0 Jan  1 00:01 tpdm1 -> 
./../../devices/platform/soc@0/10c28000.tpdm/tpdm1
lrwxrwxrwx    1 root     0                0 Jan  1 00:01 tpdm2 -> 
./../../devices/platform/soc@0/10c29000.tpdm/tpdm2

> 
> root@linaro-developer:/home/linaro/cs-mods# ls -al /sys/bus/coresight/devices/
> total 0
> drwxr-xr-x 2 root root 0 Jan  2 11:27 .
> drwxr-xr-x 4 root root 0 Jan  2 11:27 ..
> lrwxrwxrwx 1 root root 0 Jan  2 11:27 cti_cpu0 ->
> ../../../devices/platform/soc@0/858000.cti/cti_cpu0
> lrwxrwxrwx 1 root root 0 Jan  2 11:27 cti_cpu1 ->
> ../../../devices/platform/soc@0/859000.cti/cti_cpu1
> lrwxrwxrwx 1 root root 0 Jan  2 11:27 cti_cpu2 ->
> ../../../devices/platform/soc@0/85a000.cti/cti_cpu2
> lrwxrwxrwx 1 root root 0 Jan  2 11:27 cti_cpu3 ->
> ../../../devices/platform/soc@0/85b000.cti/cti_cpu3
> lrwxrwxrwx 1 root root 0 Jan  2 11:27 cti_sys0 ->
> ../../../devices/platform/soc@0/810000.cti/cti_sys0
> lrwxrwxrwx 1 root root 0 Jan  2 11:27 cti_sys1 ->
> ../../../devices/platform/soc@0/811000.cti/cti_sys1
> lrwxrwxrwx 1 root root 0 Jan  2 11:27 etm0 ->
> ../../../devices/platform/soc@0/85c000.etm/etm0
> lrwxrwxrwx 1 root root 0 Jan  2 11:27 etm1 ->
> ../../../devices/platform/soc@0/85d000.etm/etm1
> lrwxrwxrwx 1 root root 0 Jan  2 11:27 etm2 ->
> ../../../devices/platform/soc@0/85e000.etm/etm2
> lrwxrwxrwx 1 root root 0 Jan  2 11:27 etm3 ->
> ../../../devices/platform/soc@0/85f000.etm/etm3
> lrwxrwxrwx 1 root root 0 Jan  2 11:42 funnel0 ->
> ../../../devices/platform/soc@0/821000.funnel/funnel0
> lrwxrwxrwx 1 root root 0 Jan  2 11:42 funnel1 ->
> ../../../devices/platform/soc@0/841000.funnel/funnel1
> lrwxrwxrwx 1 root root 0 Jan  2 11:42 replicator0 ->
> ../../../devices/platform/soc@0/824000.replicator/replicator0
> lrwxrwxrwx 1 root root 0 Jan  2 11:42 tmc_etf0 ->
> ../../../devices/platform/soc@0/825000.etf/tmc_etf0
> lrwxrwxrwx 1 root root 0 Jan  2 11:42 tmc_etr0 ->
> ../../../devices/platform/soc@0/826000.etr/tmc_etr0
> 
> 
> 2) The patch set must contain the usage and specification in the .yaml
>   file(s) of the property used.
> 
I will add the usage in yaml.

> However, there was a standard property called 'name' which is
> deprecated - see
> https://devicetree-specification.readthedocs.io/en/latest/chapter2-devicetree-basics.html
> section 2.3.11. I do not believe that adding another 'name' property
> would be accepted by the DT maintainers.
> 
> 3) the 'device_node' structure has a 'name' field that contains the
> node name in the DT approved "node-name@unit-address" format.This
> contains whatever node names you used in the dt.  Why not use this if
> a change has to be made and find some conditional to activate it.
> 
> However, given point 1) above, the problem is solved and the patch
> adds no new information not already available.
> 
The name in the DT only has the general class of device and registers 
address. It can't describe the subsystem it belongs to.

Thanks
Jinlong Mao

> Regards
> 
> Mike
> 
> On Thu, 28 Dec 2023 at 11:26, James Clark <james.clark@arm.com> wrote:
>>
>>
>>
>> On 28/12/2023 09:33, Mao Jinlong wrote:
>>> Add coresight name support for custom names which will be
>>> easy to identify the device by the name.
>>>
>>
>> I suppose this is more of a V2 because the subject is the same as the
>> one sent earlier this year. But it looks like the discussion on the
>> previous one wasn't resolved.
>>
>> With the main issues to solve being:
>>
>>   * It would be nice to use the existing root node name instead of adding
>>     a new property. But at the same time DT nodes are supposed to have
>>     generic names.
>>
>>   * This only works for DT and not ACPI
>>
>> To me it seems like adding the new property is just a "cheat" to get
>> around not being allowed to have a specific name for the root node. But
>> if we admit that we need a name I don't see the benefit of not putting
>> the name where the node is already named.
>>
>> Using the root node name at this point would also undo the hard coded
>> per-cpu naming of the CTI and ETM devices, so maybe it would be nice,
>> but it's just too late. That means that a new field is necessary.
>> Although that field could be a boolean like "use-root-name-for-display"
>> or something like that. In the end it probably doesn't really make a
>> difference whether it's that or a name string.
>>
>> And maybe the answer to the ACPI question is just that if anyone needs
>> it, they can add it in the future. It doesn't seem like it would
>> conflict with anything we do here.
>>
>>> Signed-off-by: Mao Jinlong <quic_jinlmao@quicinc.com>
>>> ---
>>>   .../hwtracing/coresight/coresight-cti-core.c  | 20 ++++++++------
>>>   drivers/hwtracing/coresight/coresight-dummy.c | 10 ++++---
>>>   .../hwtracing/coresight/coresight-platform.c  | 27 +++++++++++++++++++
>>>   drivers/hwtracing/coresight/coresight-tpdm.c  | 10 ++++---
>>>   include/linux/coresight.h                     |  1 +
>>>   5 files changed, 53 insertions(+), 15 deletions(-)
>>>
>>> diff --git a/drivers/hwtracing/coresight/coresight-cti-core.c b/drivers/hwtracing/coresight/coresight-cti-core.c
>>> index 3999d0a2cb60..60a1e76064a9 100644
>>> --- a/drivers/hwtracing/coresight/coresight-cti-core.c
>>> +++ b/drivers/hwtracing/coresight/coresight-cti-core.c
>>> @@ -902,14 +902,18 @@ static int cti_probe(struct amba_device *adev, const struct amba_id *id)
>>>        /* default to powered - could change on PM notifications */
>>>        drvdata->config.hw_powered = true;
>>>
>>> -     /* set up device name - will depend if cpu bound or otherwise */
>>> -     if (drvdata->ctidev.cpu >= 0)
>>> -             cti_desc.name = devm_kasprintf(dev, GFP_KERNEL, "cti_cpu%d",
>>> -                                            drvdata->ctidev.cpu);
>>> -     else
>>> -             cti_desc.name = coresight_alloc_device_name(&cti_sys_devs, dev);
>>
>> Can we put the new name stuff inside coresight_alloc_device_name()? Then
>> it happens by default for every device.
>>
>> I know Suzuki said previously to do it per-device, but the new DT
>> property is just "coresight-name", so it's generic. Rather than being
>> specific like "cti-name". So I don't see the benefit of duplicating the
>> code at this point if we do decide to do it.
>>
>>> -     if (!cti_desc.name)
>>> -             return -ENOMEM;
>>> +     cti_desc.name = coresight_get_device_name(dev);
>>> +     if (!cti_desc.name) {
>>> +             /* set up device name - will depend if cpu bound or otherwise */
>>> +             if (drvdata->ctidev.cpu >= 0)
>>> +                     cti_desc.name = devm_kasprintf(dev, GFP_KERNEL, "cti_cpu%d",
>>> +                                                    drvdata->ctidev.cpu);
>>> +             else {
>>> +                     cti_desc.name = coresight_alloc_device_name(&cti_sys_devs, dev);
>>> +                     if (!cti_desc.name)
>>> +                             return -ENOMEM;
>>> +             }
>>> +     }
>>
>>>
>>>        /* setup CPU power management handling for CPU bound CTI devices. */
>>>        ret = cti_pm_setup(drvdata);
>>> diff --git a/drivers/hwtracing/coresight/coresight-dummy.c b/drivers/hwtracing/coresight/coresight-dummy.c
>>> index e4deafae7bc2..b19cd400df79 100644
>>> --- a/drivers/hwtracing/coresight/coresight-dummy.c
>>> +++ b/drivers/hwtracing/coresight/coresight-dummy.c
>>> @@ -76,10 +76,12 @@ static int dummy_probe(struct platform_device *pdev)
>>>        struct coresight_desc desc = { 0 };
>>>
>>>        if (of_device_is_compatible(node, "arm,coresight-dummy-source")) {
>>> -
>>> -             desc.name = coresight_alloc_device_name(&source_devs, dev);
>>> -             if (!desc.name)
>>> -                     return -ENOMEM;
>>> +             desc.name = coresight_get_device_name(dev);
>>> +             if (!desc.name) {
>>> +                     desc.name = coresight_alloc_device_name(&source_devs, dev);
>>> +                     if (!desc.name)
>>> +                             return -ENOMEM;
>>> +             }
>>>
>>>                desc.type = CORESIGHT_DEV_TYPE_SOURCE;
>>>                desc.subtype.source_subtype =
>>> diff --git a/drivers/hwtracing/coresight/coresight-platform.c b/drivers/hwtracing/coresight/coresight-platform.c
>>> index 9d550f5697fa..284aa22a06b7 100644
>>> --- a/drivers/hwtracing/coresight/coresight-platform.c
>>> +++ b/drivers/hwtracing/coresight/coresight-platform.c
>>> @@ -183,6 +183,18 @@ static int of_coresight_get_cpu(struct device *dev)
>>>        return cpu;
>>>   }
>>>
>>> +static const char *of_coresight_get_device_name(struct device *dev)
>>> +{
>>> +     const char *name = NULL;
>>> +
>>> +     if (!dev->of_node)
>>> +             return NULL;
>>> +
>>> +     of_property_read_string(dev->of_node, "coresight-name", &name);
>>
>> Do you need to update the binding docs with this new property?
>>
>> Also a minor nit: Maybe "display-name" is better? "Coresight" is
>> implied, and the node is already named, although that node name isn't
>> used for display purposes, but this one is.
>>
>> Thanks
>> James
> 
> 
>
  

On 1/3/2024 7:33 PM, Suzuki K Poulose wrote:
> On 28/12/2023 11:26, James Clark wrote:
>>
>>
>> On 28/12/2023 09:33, Mao Jinlong wrote:
>>> Add coresight name support for custom names which will be
>>> easy to identify the device by the name.
>>>
>>
>> I suppose this is more of a V2 because the subject is the same as the
>> one sent earlier this year. But it looks like the discussion on the
>> previous one wasn't resolved.
>>
>> With the main issues to solve being:
>>
>>   * It would be nice to use the existing root node name instead of adding
>>     a new property. But at the same time DT nodes are supposed to have
>>     generic names.
>>
>>   * This only works for DT and not ACPI
>>
>> To me it seems like adding the new property is just a "cheat" to get
>> around not being allowed to have a specific name for the root node. But
>> if we admit that we need a name I don't see the benefit of not putting
>> the name where the node is already named.
>>
>> Using the root node name at this point would also undo the hard coded
>> per-cpu naming of the CTI and ETM devices, so maybe it would be nice,
>> but it's just too late. That means that a new field is necessary.
> 
> The CTI and ETM can be handled as special cases, like they are
> already done and fall back to the nodename for the rest ?
> But, I thought the node names must be generic (e.g, cti) and doesn't
> really solve the naming requirements for naming CTIs. (e.g,
> <device>_tpda, etr_cti). Is there something I missed ?
> 
>> Although that field could be a boolean like "use-root-name-for-display"
>> or something like that. In the end it probably doesn't really make a
>> difference whether it's that or a name string. >
>> And maybe the answer to the ACPI question is just that if anyone needs
>> it, they can add it in the future. It doesn't seem like it would
>> conflict with anything we do here.
>>
>>> Signed-off-by: Mao Jinlong <quic_jinlmao@quicinc.com>
>>> ---
>>>   .../hwtracing/coresight/coresight-cti-core.c  | 20 ++++++++------
>>>   drivers/hwtracing/coresight/coresight-dummy.c | 10 ++++---
>>>   .../hwtracing/coresight/coresight-platform.c  | 27 +++++++++++++++++++
>>>   drivers/hwtracing/coresight/coresight-tpdm.c  | 10 ++++---
>>>   include/linux/coresight.h                     |  1 +
>>>   5 files changed, 53 insertions(+), 15 deletions(-)
>>>
>>> diff --git a/drivers/hwtracing/coresight/coresight-cti-core.c 
>>> b/drivers/hwtracing/coresight/coresight-cti-core.c
>>> index 3999d0a2cb60..60a1e76064a9 100644
>>> --- a/drivers/hwtracing/coresight/coresight-cti-core.c
>>> +++ b/drivers/hwtracing/coresight/coresight-cti-core.c
>>> @@ -902,14 +902,18 @@ static int cti_probe(struct amba_device *adev, 
>>> const struct amba_id *id)
>>>       /* default to powered - could change on PM notifications */
>>>       drvdata->config.hw_powered = true;
>>> -    /* set up device name - will depend if cpu bound or otherwise */
>>> -    if (drvdata->ctidev.cpu >= 0)
>>> -        cti_desc.name = devm_kasprintf(dev, GFP_KERNEL, "cti_cpu%d",
>>> -                           drvdata->ctidev.cpu);
>>> -    else
>>> -        cti_desc.name = coresight_alloc_device_name(&cti_sys_devs, 
>>> dev);
>>
>> Can we put the new name stuff inside coresight_alloc_device_name()? Then
>> it happens by default for every device.
> 
> +1
> 
>>
>> I know Suzuki said previously to do it per-device, but the new DT
>> property is just "coresight-name", so it's generic. Rather than being
>> specific like "cti-name". So I don't see the benefit of duplicating the
>> code at this point if we do decide to do it.
> 
> My suggestion was to name the device based on the specific device rather
> than following a generic rule for all device. e.g., A TPDM connected to 
> modem, could be named as such based on the platform information. It
> could be any means, for e.g., tpdm nodes are always children nodes of
> the devices they are connected to ? or could have a phandle to point to 
> the device they are monitoring etc. And the name could be created from
> the "monitoring device name" + tpdm. Also, we do this for CPU bound CTI
> and ETMs already, where we name them based on the CPU.

TPDM can only connect to the funnel or TPDA. The system TPDM is 
monitoring may not have the device node in DT.

> 
> But then the "nodename" is something we explored and it looks like
> may not be an option.
> 
>>
>>> -    if (!cti_desc.name)
>>> -        return -ENOMEM;
>>> +    cti_desc.name = coresight_get_device_name(dev);
>>> +    if (!cti_desc.name) {
>>> +        /* set up device name - will depend if cpu bound or 
>>> otherwise */
>>> +        if (drvdata->ctidev.cpu >= 0)
>>> +            cti_desc.name = devm_kasprintf(dev, GFP_KERNEL, 
>>> "cti_cpu%d",
>>> +                               drvdata->ctidev.cpu);
>>> +        else {
>>> +            cti_desc.name = 
>>> coresight_alloc_device_name(&cti_sys_devs, dev);
>>> +            if (!cti_desc.name)
>>> +                return -ENOMEM;
>>> +        }
>>> +    }
> 
> For these special cases, i.e., CPU bound, we should handle them with 
> priority.
> 
> if (drvdata->ctidev.cpu >= 0)
>      name = devm_kasprintf(... "cti_cpu%d", .. cpu);
> else
>          name = coresight_alloc_device_name(...);
> 
>>
>>>       /* setup CPU power management handling for CPU bound CTI 
>>> devices. */
>>>       ret = cti_pm_setup(drvdata);
>>> diff --git a/drivers/hwtracing/coresight/coresight-dummy.c 
>>> b/drivers/hwtracing/coresight/coresight-dummy.c
>>> index e4deafae7bc2..b19cd400df79 100644
>>> --- a/drivers/hwtracing/coresight/coresight-dummy.c
>>> +++ b/drivers/hwtracing/coresight/coresight-dummy.c
>>> @@ -76,10 +76,12 @@ static int dummy_probe(struct platform_device *pdev)
>>>       struct coresight_desc desc = { 0 };
>>>       if (of_device_is_compatible(node, "arm,coresight-dummy-source")) {
>>> -
>>> -        desc.name = coresight_alloc_device_name(&source_devs, dev);
>>> -        if (!desc.name)
>>> -            return -ENOMEM;
>>> +        desc.name = coresight_get_device_name(dev);
>>> +        if (!desc.name) {
>>> +            desc.name = coresight_alloc_device_name(&source_devs, dev);
>>> +            if (!desc.name)
>>> +                return -ENOMEM;
>>> +        }
>>>           desc.type = CORESIGHT_DEV_TYPE_SOURCE;
>>>           desc.subtype.source_subtype =
>>> diff --git a/drivers/hwtracing/coresight/coresight-platform.c 
>>> b/drivers/hwtracing/coresight/coresight-platform.c
>>> index 9d550f5697fa..284aa22a06b7 100644
>>> --- a/drivers/hwtracing/coresight/coresight-platform.c
>>> +++ b/drivers/hwtracing/coresight/coresight-platform.c
>>> @@ -183,6 +183,18 @@ static int of_coresight_get_cpu(struct device *dev)
>>>       return cpu;
>>>   }
>>> +static const char *of_coresight_get_device_name(struct device *dev)
>>> +{
>>> +    const char *name = NULL;
>>> +
>>> +    if (!dev->of_node)
>>> +        return NULL;
>>> +
>>> +    of_property_read_string(dev->of_node, "coresight-name", &name);
>>
>> Do you need to update the binding docs with this new property?
>>
>> Also a minor nit: Maybe "display-name" is better? "Coresight" is
>> implied, and the node is already named, although that node name isn't
>> used for display purposes, but this one is.
> 
> On that front, the name is used as a "device" name and not simply 
> display. So, even "device-name" sounds more appropriate.
> 
I will use the device-name.

Thanks
Jinlong Mao
> Suzuki
> 
>>
>> Thanks
>> James
>
  

On 1/3/2024 11:32 PM, Rob Herring wrote:
> On Tue, Jan 2, 2024 at 5:05 AM Mike Leach <mike.leach@linaro.org> wrote:
>>
>> As James mentions this is clearly a V2 of a previous patch - please
>> mark as such in future.
>>
>> Adding to what James has already said:-
>>
>> 1) Mapping between the canonical names used in the drivers and the
>> information as to the precise device is as easy as running 'ls' on
>> /sys/bus/coresight/devices:-
>>
>> root@linaro-developer:/home/linaro/cs-mods# ls -al /sys/bus/coresight/devices/
>> total 0
>> drwxr-xr-x 2 root root 0 Jan  2 11:27 .
>> drwxr-xr-x 4 root root 0 Jan  2 11:27 ..
>> lrwxrwxrwx 1 root root 0 Jan  2 11:27 cti_cpu0 ->
>> ../../../devices/platform/soc@0/858000.cti/cti_cpu0
>> lrwxrwxrwx 1 root root 0 Jan  2 11:27 cti_cpu1 ->
>> ../../../devices/platform/soc@0/859000.cti/cti_cpu1
>> lrwxrwxrwx 1 root root 0 Jan  2 11:27 cti_cpu2 ->
>> ../../../devices/platform/soc@0/85a000.cti/cti_cpu2
>> lrwxrwxrwx 1 root root 0 Jan  2 11:27 cti_cpu3 ->
>> ../../../devices/platform/soc@0/85b000.cti/cti_cpu3
>> lrwxrwxrwx 1 root root 0 Jan  2 11:27 cti_sys0 ->
>> ../../../devices/platform/soc@0/810000.cti/cti_sys0
>> lrwxrwxrwx 1 root root 0 Jan  2 11:27 cti_sys1 ->
>> ../../../devices/platform/soc@0/811000.cti/cti_sys1
>> lrwxrwxrwx 1 root root 0 Jan  2 11:27 etm0 ->
>> ../../../devices/platform/soc@0/85c000.etm/etm0
>> lrwxrwxrwx 1 root root 0 Jan  2 11:27 etm1 ->
>> ../../../devices/platform/soc@0/85d000.etm/etm1
>> lrwxrwxrwx 1 root root 0 Jan  2 11:27 etm2 ->
>> ../../../devices/platform/soc@0/85e000.etm/etm2
>> lrwxrwxrwx 1 root root 0 Jan  2 11:27 etm3 ->
>> ../../../devices/platform/soc@0/85f000.etm/etm3
>> lrwxrwxrwx 1 root root 0 Jan  2 11:42 funnel0 ->
>> ../../../devices/platform/soc@0/821000.funnel/funnel0
>> lrwxrwxrwx 1 root root 0 Jan  2 11:42 funnel1 ->
>> ../../../devices/platform/soc@0/841000.funnel/funnel1
>> lrwxrwxrwx 1 root root 0 Jan  2 11:42 replicator0 ->
>> ../../../devices/platform/soc@0/824000.replicator/replicator0
>> lrwxrwxrwx 1 root root 0 Jan  2 11:42 tmc_etf0 ->
>> ../../../devices/platform/soc@0/825000.etf/tmc_etf0
>> lrwxrwxrwx 1 root root 0 Jan  2 11:42 tmc_etr0 ->
>> ../../../devices/platform/soc@0/826000.etr/tmc_etr0
>>
>>
>> 2) The patch set must contain the usage and specification in the .yaml
>>   file(s) of the property used.
> 
> For the record, I don't like "coresight-name". I don't have another
> suggestion because "easy" is not sufficient reasoning for why this is
> needed.

For example, if we want to configure the trigger and HW events for 
modem, we can't know which cti or TPDM is for modem from current names.

lrwxrwxrwx    1 root     0                0 Jan  1 00:01 cti_sys0 -> 
./../../devices/platform/soc@0/138f0000.cti/cti_sys0
lrwxrwxrwx    1 root     0                0 Jan  1 00:01 cti_sys1 -> 
./../../devices/platform/soc@0/13900000.cti/cti_sys1
lrwxrwxrwx    1 root     0                0 Jan  1 00:01 tpdm0 -> 
./../../devices/platform/soc@0/10b0d000.tpdm/tpdm0
lrwxrwxrwx    1 root     0                0 Jan  1 00:01 tpdm1 -> 
./../../devices/platform/soc@0/10c28000.tpdm/tpdm1
lrwxrwxrwx    1 root     0                0 Jan  1 00:01 tpdm2 -> 
./../../devices/platform/soc@0/10c29000.tpdm/tpdm2

Thanks
Jinlong Mao
> 
>> However, there was a standard property called 'name' which is
>> deprecated - see
>> https://devicetree-specification.readthedocs.io/en/latest/chapter2-devicetree-basics.html
>> section 2.3.11. I do not believe that adding another 'name' property
>> would be accepted by the DT maintainers.
> 
> "name" is just the node name for anything in the last 15 years. They
> used to be separate, but would still mostly be the same. The only case
> I found with them different was old PowerPC Macs.
> 
>> 3) the 'device_node' structure has a 'name' field that contains the
>> node name in the DT approved "node-name@unit-address" format.
> 
> Actually, it is without the unit-address. full_name is with the unit-address.
> 
>> This
>> contains whatever node names you used in the dt.  Why not use this if
>> a change has to be made and find some conditional to activate it.
> 
> Don't go accessing "name" or "full_name" directly. I intend to get rid
> of "name" and generate it from full_name. So use the accessors and
> printk specifiers if you need node names.
> 
> Rob