Re: [PATCH] coresight: etm4x: Ensure valid drvdata and clock before clk_put()

[James Clark]

On 11/08/2023 10:22, Anshuman Khandual wrote:
> 
> 
> On 8/11/23 14:39, Suzuki K Poulose wrote:
>> On 11/08/2023 09:39, James Clark wrote:
>>>
>>>
>>> On 11/08/2023 07:27, Anshuman Khandual wrote:
>>>> This validates 'drvdata' and 'drvdata->pclk' clock before calling clk_put()
>>>> in etm4_remove_platform_dev(). The problem was detected using Smatch static
>>>> checker as reported.
>>>>
>>>> Cc: Suzuki K Poulose <suzuki.poulose@xxxxxxx>
>>>> Cc: Mike Leach <mike.leach@xxxxxxxxxx>
>>>> Cc: James Clark <james.clark@xxxxxxx>
>>>> Cc: coresight@xxxxxxxxxxxxxxxx
>>>> Cc: linux-arm-kernel@xxxxxxxxxxxxxxxxxxx
>>>> Cc: linux-kernel@xxxxxxxxxxxxxxx
>>>> Reported-by: Dan Carpenter <dan.carpenter@xxxxxxxxxx>
>>>> Closes: https://lists.linaro.org/archives/list/coresight@xxxxxxxxxxxxxxxx/thread/G4N6P4OXELPLLQSNU3GU2MR4LOLRXRMJ/
>>>> Signed-off-by: Anshuman Khandual <anshuman.khandual@xxxxxxx>
>>>> ---
>>>> This applies on coresight-next
>>>>
>>>>   drivers/hwtracing/coresight/coresight-etm4x-core.c | 2 +-
>>>>   1 file changed, 1 insertion(+), 1 deletion(-)
>>>>
>>>> diff --git a/drivers/hwtracing/coresight/coresight-etm4x-core.c b/drivers/hwtracing/coresight/coresight-etm4x-core.c
>>>> index 703b6fcbb6a5..eb412ce302cc 100644
>>>> --- a/drivers/hwtracing/coresight/coresight-etm4x-core.c
>>>> +++ b/drivers/hwtracing/coresight/coresight-etm4x-core.c
>>>> @@ -2269,7 +2269,7 @@ static int __exit etm4_remove_platform_dev(struct platform_device *pdev)
>>>>           etm4_remove_dev(drvdata);
>>>>       pm_runtime_disable(&pdev->dev);
>>>>   -    if (drvdata->pclk)
>>>> +    if (drvdata && drvdata->pclk && !IS_ERR(drvdata->pclk))
>>>>           clk_put(drvdata->pclk);
>>>>         return 0;
>>>
>>> It could be !IS_ERR_OR_NULL(drvdata->pclk), but I wouldn't bother
>>> changing it at this point.
>>
>> +1, please could we have that. Someone else will run a code scanner and
>> send a patch later. Given this is straight and easy change, lets do it
>> in the first place.
> 
> But we already have a drvdata->pclk validation check before IS_ERR().
> Would not _OR_NULL be redundant ?

I meant that it could be replaced with the single check:

  if (drvdata && !IS_ERR_OR_NULL(drvdata->pclk))
      clk_put(drvdata->pclk);

As Dan mentions it can't be an error pointer anyway, but leaving it like
this could just be considered defensive coding.