Re: Coverity: imx290_ctrl_init(): Error handling issues

[Kees Cook]

On Thu, Nov 10, 2022 at 04:41:28PM +0000, Dave Stevenson wrote:
> On Thu, 10 Nov 2022 at 16:31, coverity-bot <keescook@xxxxxxxxxxxx> wrote:
> >
> > Hello!
> >
> > This is an experimental semi-automated report about issues detected by
> > Coverity from a scan of next-20221110 as part of the linux-next scan project:
> > https://scan.coverity.com/projects/linux-next-weekly-scan
> >
> > You're getting this email because you were associated with the identified
> > lines of code (noted below) that were touched by commits:
> >
> >   Thu Oct 27 14:38:02 2022 +0300
> >     4c9c93cf8657 ("media: i2c: imx290: Create controls for fwnode properties")
> >
> > Coverity reported the following:
> >
> > *** CID 1527251:  Error handling issues  (CHECKED_RETURN)
> > drivers/media/i2c/imx290.c:1056 in imx290_ctrl_init()
> > 1050            imx290->vblank = v4l2_ctrl_new_std(&imx290->ctrls, &imx290_ctrl_ops,
> > 1051                                               V4L2_CID_VBLANK, blank, blank, 1,
> > 1052                                               blank);
> > 1053            if (imx290->vblank)
> > 1054                    imx290->vblank->flags |= V4L2_CTRL_FLAG_READ_ONLY;
> > 1055
> > vvv     CID 1527251:  Error handling issues  (CHECKED_RETURN)
> > vvv     Calling "v4l2_ctrl_new_fwnode_properties" without checking return value (as is done elsewhere 9 out of 10 times).
> > 1056            v4l2_ctrl_new_fwnode_properties(&imx290->ctrls, &imx290_ctrl_ops,
> > 1057                                            &props);
> > 1058
> > 1059            imx290->sd.ctrl_handler = &imx290->ctrls;
> > 1060
> > 1061            if (imx290->ctrls.error) {
> >
> > If this is a false positive, please let us know so we can mark it as
> > such, or teach the Coverity rules to be smarter. If not, please make
> > sure fixes get into linux-next. :) For patches fixing this, please
> > include these lines (but double-check the "Fixes" first):
> 
> I looked at this one when the patches were sent to the list.
> 
> On failure, v4l2_ctrl_new_fwnode_properties will have set the error
> flag in struct v4l2_ctrl_handler. This is also what it returns.
> 
> In most of the existing drivers the error flag has already been
> checked before calling v4l2_ctrl_new_fwnode_properties, therefore the
> return value has to be checked explicitly. In this case it is checked
> at line 1061 which is after v4l2_ctrl_new_fwnode_properties has been
> called, and therefore there is no need to check the return value of
> the call.
> 
> IMHO Neither is particularly right or wrong, just slightly different
> approaches. In some regards this new code pattern is nicer as it
> removes a number of error handling paths.

Great! Thanks for double-checking it. :)

-- 
Kees Cook