On 6/23/26 6:01 PM, Mukesh Ojha wrote:
> On Tue, Jun 23, 2026 at 09:20:15PM +0800, Haoxiang Li wrote:
>> qcom_q6v5_init() acquires the QMP handle before getting the
>> interconnect path. If devm_of_icc_get() fails, the function
>> returns without releasing the QMP handle, leaking the device
>> reference acquired by qmp_get().
>>
>> Release the QMP handle on the interconnect error path before returning.
>>
>> Fixes: 8d9be5c6bdcd ("remoteproc: qcom: q6v5: Add interconnect path proxy
>> vote")
>> Signed-off-by: Haoxiang Li <[email protected]>
>> ---
>> drivers/remoteproc/qcom_q6v5.c | 4 +++-
>> 1 file changed, 3 insertions(+), 1 deletion(-)
>>
>> diff --git a/drivers/remoteproc/qcom_q6v5.c b/drivers/remoteproc/qcom_q6v5.c
>> index 58d5b85e58cd..b57041a37d28 100644
>> --- a/drivers/remoteproc/qcom_q6v5.c
>> +++ b/drivers/remoteproc/qcom_q6v5.c
>> @@ -347,9 +347,11 @@ int qcom_q6v5_init(struct qcom_q6v5 *q6v5, struct
>> platform_device *pdev,
>> }
>>
>> q6v5->path = devm_of_icc_get(&pdev->dev, NULL);
>> - if (IS_ERR(q6v5->path))
>> + if (IS_ERR(q6v5->path)) {
>> + qmp_put(q6v5->qmp);
>> return dev_err_probe(&pdev->dev, PTR_ERR(q6v5->path),
>> "failed to acquire interconnect path\n");
>> + }
>
> I would have preferred to move devm_of_icc_get before qmp_get()
Yeah this saves on cleanup
Konrad
