Re: [PATCH] coresight: core: Add coresight name support

[Suzuki K Poulose <suzuki.poulose@arm.com>]

On 17/03/2023 05:34, Jinlong Mao wrote:
> 
> On 3/13/2023 5:16 PM, Suzuki K Poulose wrote:
>> Cc: Rob
>>
>> On 01/03/2023 15:11, Jinlong Mao wrote:
>>> Hi Suzuki,
>>>
>>> On 2/9/2023 10:16 AM, Jinlong Mao wrote:
>>>>
>>>> On 2/8/2023 10:26 PM, Suzuki K Poulose wrote:
>>>>> On 08/02/2023 11:07, Mao Jinlong wrote:
>>>>>> Apart from STM and ETM sources, there will be more sources added to
>>>>>> coresight components. For example, there are over 10 TPDM sources.
>>>>>> Add coresight name support for custom names which will be
>>>>>> easy to identify the source.
>>>>>>
>>>>>
>>>>> As we have previously discussed, please don't make this a generic
>>>>> code change. If your device has a "specifici" name, use that for
>>>>> allocating in the driver and leave the core code alone.
>>>>>
>>>>> Suzuki
>>>>>
>>>> Hi Suzuki,
>>>>
>>>> Not only for TPDMs. There could be dozens of CTI devices.
>>>> It is hard for user to know which CTI device it is with current names.
>>>>
>>>> Thanks
>>>> Jinlong Mao
>>>
>>> The coresight name support is applicable to CTI and TPDM devices.
>>> This is a generic change for the source which has dozens of devices.
>>
>> I took a look at the CTI situation and I agree that the situation
>> is a bit tricky. The CTI could be connected to multiple devices,
>> some of them may not be even CoreSight devices. Given there could
>> be numerous of them, we need some way to make the "devices" naming
>> a bit more intuitive.
>>
>> Before we go ahead and add something specific to coresight, I would
>> like to see if there is a generic property. Ideally, the "labels"
>> in the DTS sources would have been an ideal choice, but can't
>> see how that is available in the FDT.
>>
>> Suzuki
> Hi Suzuki,
> 
> Shall we use the full_name of device_node struct as coresight 
> component's name ?
> 
>    struct device_node {
>      const char *name;
>      phandle phandle;
> *    const char *full_name;
> 
> *For component below, the full_name will be "coresight-tpdm-ipcc".
> *
> **coresight-tpdm-ipcc* {

Does that go against the convention of naming the DT nodes ?
I am not sure. Also, we would need a way to solve this for ACPI too.

Suzuki


>      compatible = "qcom,coresight-tpdm", "arm,primecell";
>      reg = <0 0x10c29000 0 0x1000>;
> 
>      clocks = <&aoss_qmp>;
>      clock-names = "apb_pclk";
> 
>      out-ports {
>              port {
>                      tpdm_ipcc_out_tpda_dl_center_27: endpoint {
>                              remote-endpoint =
> <&tpda_dl_center_27_in_tpdm_ipcc>;
>                      };
>              };
>      };
> };*
> 
> *Thanks
> Jinlong Mao
>>
>>
>>>
>>> Thanks
>>> Jinlong Mao
>>>
>>>>>
>>>>>> Signed-off-by: Mao Jinlong <quic_jinlmao@quicinc.com>
>>>>>> ---
>>>>>>   drivers/hwtracing/coresight/coresight-core.c | 34 
>>>>>> +++++++++++---------
>>>>>>   1 file changed, 19 insertions(+), 15 deletions(-)
>>>>>>
>>>>>> diff --git a/drivers/hwtracing/coresight/coresight-core.c 
>>>>>> b/drivers/hwtracing/coresight/coresight-core.c
>>>>>> index d3bf82c0de1d..5e95d9c7f256 100644
>>>>>> --- a/drivers/hwtracing/coresight/coresight-core.c
>>>>>> +++ b/drivers/hwtracing/coresight/coresight-core.c
>>>>>> @@ -1733,28 +1733,32 @@ char *coresight_alloc_device_name(struct 
>>>>>> coresight_dev_list *dict,
>>>>>>   {
>>>>>>       int idx;
>>>>>>       char *name = NULL;
>>>>>> +    const char *coresight_name = NULL;
>>>>>>       struct fwnode_handle **list;
>>>>>> +    struct device_node *node = dev->of_node;
>>>>>>         mutex_lock(&coresight_mutex);
>>>>>>   -    idx = coresight_search_device_idx(dict, dev_fwnode(dev));
>>>>>> -    if (idx < 0) {
>>>>>> -        /* Make space for the new entry */
>>>>>> -        idx = dict->nr_idx;
>>>>>> -        list = krealloc_array(dict->fwnode_list,
>>>>>> -                      idx + 1, sizeof(*dict->fwnode_list),
>>>>>> -                      GFP_KERNEL);
>>>>>> -        if (ZERO_OR_NULL_PTR(list)) {
>>>>>> -            idx = -ENOMEM;
>>>>>> -            goto done;
>>>>>> +    if (!of_property_read_string(node, "coresight-name", 
>>>>>> &coresight_name))
>>>>>> +        name = devm_kasprintf(dev, GFP_KERNEL, "%s", 
>>>>>> coresight_name);
>>>>>> +    else {
>>>>>> +        idx = coresight_search_device_idx(dict, dev_fwnode(dev));
>>>>>> +        if (idx < 0) {
>>>>>> +            /* Make space for the new entry */
>>>>>> +            idx = dict->nr_idx;
>>>>>> +            list = krealloc_array(dict->fwnode_list,
>>>>>> +                          idx + 1, sizeof(*dict->fwnode_list),
>>>>>> +                          GFP_KERNEL);
>>>>>> +            if (ZERO_OR_NULL_PTR(list))
>>>>>> +                goto done;
>>>>>> +
>>>>>> +            list[idx] = dev_fwnode(dev);
>>>>>> +            dict->fwnode_list = list;
>>>>>> +            dict->nr_idx = idx + 1;
>>>>>>           }
>>>>>>   -        list[idx] = dev_fwnode(dev);
>>>>>> -        dict->fwnode_list = list;
>>>>>> -        dict->nr_idx = idx + 1;
>>>>>> +        name = devm_kasprintf(dev, GFP_KERNEL, "%s%d", dict->pfx, 
>>>>>> idx);
>>>>>>       }
>>>>>> -
>>>>>> -    name = devm_kasprintf(dev, GFP_KERNEL, "%s%d", dict->pfx, idx);
>>>>>>   done:
>>>>>>       mutex_unlock(&coresight_mutex);
>>>>>>       return name;
>>>>>
>>>> _______________________________________________
>>>> CoreSight mailing list -- coresight@lists.linaro.org
>>>> To unsubscribe send an email to coresight-leave@lists.linaro.org
>>