On Thu, Dec 3, 2020 at 8:24 PM Colin Ian King <colin.king@xxxxxxxxxxxxx> wrote:
> On 03/12/2020 18:10, Andy Shevchenko wrote:
> > On Thu, Dec 3, 2020 at 8:03 PM Colin Ian King <colin.king@xxxxxxxxxxxxx> wrote:
> >
> >> Static analysis on linux-next with Coverity has detected an issue with
> >> the following commit:
> >
> > If you want to fix it properly, see my comments below...
> >
> >> 529 static int ov02a10_s_stream(struct v4l2_subdev *sd, int on)
> >> 530 {
> >> 531 struct ov02a10 *ov02a10 = to_ov02a10(sd);
> >> 532 struct i2c_client *client =
> >> v4l2_get_subdevdata(&ov02a10->subdev);
> >>
> >> 1. var_decl: Declaring variable ret without initializer.
> >>
> >> 533 int ret;
> >> 534
> >> 535 mutex_lock(&ov02a10->mutex);
> >> 536
> >>
> >> 2. Condition ov02a10->streaming == on, taking true branch.
> >>
> >> 537 if (ov02a10->streaming == on)
> >>
> >> 3. Jumping to label unlock_and_return.
> >>
> >> 538 goto unlock_and_return;
> >> 539
> >> 540 if (on) {
> >> 541 ret = pm_runtime_get_sync(&client->dev);
> >> 542 if (ret < 0) {
> >
> >> 543 pm_runtime_put_noidle(&client->dev);
> >> 544 goto unlock_and_return;
> >
> > Instead of two above:
> > goto err_rpm_put;
> >
> >> 545 }
> >> 546
> >> 547 ret = __ov02a10_start_stream(ov02a10);
> >> 548 if (ret) {
> >> 549 __ov02a10_stop_stream(ov02a10);
> >> 550 ov02a10->streaming = !on;
> >> 551 goto err_rpm_put;
> >> 552 }
> >> 553 } else {
> >> 554 __ov02a10_stop_stream(ov02a10);
> >> 555 pm_runtime_put(&client->dev);
> >> 556 }
> >> 557
> >> 558 ov02a10->streaming = on;
> >
> > (1)
> >
> >> 559 mutex_unlock(&ov02a10->mutex);
> >> 560
> >> 561 return 0;
> >> 562
> >> 563 err_rpm_put:
> >> 564 pm_runtime_put(&client->dev);
> >
> >> 565 unlock_and_return:
> >
> > Should be moved to (1).
> >
> >> 566 mutex_unlock(&ov02a10->mutex);
> >> 567
> >>
> >> Uninitialized scalar variable (UNINIT)
> >> 4. uninit_use: Using uninitialized value ret.
> >>
> >> 568 return ret;
> >> 569 }
> >>
> >> Variable ret has not been initialized, so the error return value is a
> >> garbage value. It should be initialized with some appropriate negative
> >> error code, or ret could be removed and the return should return a
> >> literal value of a error code.
> >>
> >> I was unsure what value is appropriate to fix this, so instead I'm
> >> reporting this issue.
> >
> Not sure I fully understand how that fixes it.
If you are not sure and have no means to test, then don't bother. This
is not the priority driver anyway.
> Given that ret is
> currently not initialized when we hit:
>
> if (ov02a10->streaming == on)
> goto unlock_and_return;
>
> either we initialize ret to 0 at the start of the function, or do:
>
> if (ov02a10->streaming == on) {
> ret = 0;
> goto unlock_and_return;
> }
>
> (assuming the intention is to return zero for this specific case).
No, please read my message again.
--
With Best Regards,
Andy Shevchenko
