Hi,
On 10/14/21 7:38 PM, Jonathan Cameron wrote:
> On Thu, 14 Oct 2021 10:20:34 +0200
> Hans de Goede <hdegoede@xxxxxxxxxx> wrote:
>
>> Hi,
>>
>> On 10/14/21 5:53 AM, Yang Yingliang wrote:
>>> When ACPI type is ACPI_SMO8500, the data->dready_trig will not be set, the
>>> memory allocated by iio_triggered_buffer_setup() will not be freed, and cause
>>> memory leak as follows:
>>>
>>> unreferenced object 0xffff888009551400 (size 512):
>>> comm "i2c-SMO8500-125", pid 911, jiffies 4294911787 (age 83.852s)
>>> hex dump (first 32 bytes):
>>> 02 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 ................
>>> 00 00 00 00 00 00 00 00 20 e2 e5 c0 ff ff ff ff ........ .......
>>> backtrace:
>>> [<0000000041ce75ee>] kmem_cache_alloc_trace+0x16d/0x360
>>> [<000000000aeb17b0>] iio_kfifo_allocate+0x41/0x130 [kfifo_buf]
>>> [<000000004b40c1f5>] iio_triggered_buffer_setup_ext+0x2c/0x210 [industrialio_triggered_buffer]
>>> [<000000004375b15f>] kxcjk1013_probe+0x10c3/0x1d81 [kxcjk_1013]
>>> [<0000000020115b9a>] i2c_device_probe+0xa31/0xbe0
>>> [<00000000d9f581a6>] really_probe+0x299/0xc30
>>> [<00000000c6c16cde>] __driver_probe_device+0x357/0x500
>>> [<00000000909852a1>] driver_probe_device+0x4e/0x140
>>> [<000000008419ba53>] __device_attach_driver+0x257/0x340
>>> [<00000000533bb466>] bus_for_each_drv+0x166/0x1e0
>>> [<000000005bf45d75>] __device_attach+0x272/0x420
>>> [<0000000075220311>] bus_probe_device+0x1eb/0x2a0
>>> [<0000000015587e85>] device_add+0xbf0/0x1f90
>>> [<0000000086901b9e>] i2c_new_client_device+0x622/0xb20
>>> [<000000000865ca18>] new_device_store+0x1fa/0x420
>>> [<0000000059a3d183>] dev_attr_store+0x58/0x80
>>>
>>> Fix it by remove data->dready_trig condition in probe and remove.
>>>
>>> Reported-by: Hulk Robot <hulkci@xxxxxxxxxx>
>>> Fixes: a25691c1f967 ("iio: accel: kxcjk1013: allow using an external trigger")
>>> Signed-off-by: Yang Yingliang <yangyingliang@xxxxxxxxxx>
>>
>> Hmm, wouldn't the right fix be to also move the
>> iio_triggered_buffer_setup() call to inside the:
>>
>> if (client->irq > 0 && data->acpi_type != ACPI_SMO8500) {
>> }
>>
>> block ?
>>
>> Jonathan (jic23) can you take a look at this, to me it seems that having
>> a triggered buffer allocated without any triggers is not useful ?
>
> It can use another trigger not supplied by this particular device.
> e.g. sysfs or hrtimer trigger. This is common for cases where
> we may or may not have an irq wired and the validate_* callbacks are
> not provided (which would indicate we had to use the device's own trigger).
Ok, thank you for clarifying this.
With that resolved, this patch looks good to me:
Reviewed-by: Hans de Goede <hdegoede@xxxxxxxxxx>
Regards,
Hans
>>> ---
>>> drivers/iio/accel/kxcjk-1013.c | 5 ++---
>>> 1 file changed, 2 insertions(+), 3 deletions(-)
>>>
>>> diff --git a/drivers/iio/accel/kxcjk-1013.c b/drivers/iio/accel/kxcjk-1013.c
>>> index a51fdd3c9b5b..24c9387c2968 100644
>>> --- a/drivers/iio/accel/kxcjk-1013.c
>>> +++ b/drivers/iio/accel/kxcjk-1013.c
>>> @@ -1595,8 +1595,7 @@ static int kxcjk1013_probe(struct i2c_client *client,
>>> return 0;
>>>
>>> err_buffer_cleanup:
>>> - if (data->dready_trig)
>>> - iio_triggered_buffer_cleanup(indio_dev);
>>> + iio_triggered_buffer_cleanup(indio_dev);
>>> err_trigger_unregister:
>>> if (data->dready_trig)
>>> iio_trigger_unregister(data->dready_trig);
>>> @@ -1618,8 +1617,8 @@ static int kxcjk1013_remove(struct i2c_client *client)
>>> pm_runtime_disable(&client->dev);
>>> pm_runtime_set_suspended(&client->dev);
>>>
>>> + iio_triggered_buffer_cleanup(indio_dev);
>>> if (data->dready_trig) {
>>> - iio_triggered_buffer_cleanup(indio_dev);
>>> iio_trigger_unregister(data->dready_trig);
>>> iio_trigger_unregister(data->motion_trig);
>>> }
>>>
>>
>
