Hi Stefan On Tue, 14 Apr 2020 at 17:49, Stefan Wahren <stefan.wahren@xxxxxxxx> wrote: > > Hi, > > Am 14.04.20 um 17:33 schrieb coverity-bot: > > Hello! > > > > This is an experimental semi-automated report about issues detected by > > Coverity from a scan of next-20200414 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: > > > > Sun Mar 29 14:44:58 2020 +0200 > > 1a59532382a6 ("staging: bcm2835-camera: Move video component setup in its own function") > > > > Coverity reported the following: > > > > *** CID 1492591: Code maintainability issues (UNUSED_VALUE) > > /drivers/staging/vc04_services/bcm2835-camera/bcm2835-camera.c: 1014 in mmal_setup_video_component() > > 1008 if (overlay_enabled) { > > 1009 /* Need to disable the overlay before we can update > > 1010 * the resolution > > 1011 */ > > 1012 ret = vchiq_mmal_port_disable(dev->instance, preview_port); > > 1013 if (!ret) { > > vvv CID 1492591: Code maintainability issues (UNUSED_VALUE) > > vvv Assigning value from "vchiq_mmal_port_connect_tunnel(dev->instance, preview_port, NULL)" to "ret" here, but that stored value is overwritten before it can be used. > > 1014 ret = vchiq_mmal_port_connect_tunnel(dev->instance, > > 1015 preview_port, > > 1016 NULL); > > 1017 } > > 1018 } > > 1019 preview_port->es.video.width = f->fmt.pix.width; > > > > 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): > > thanks for the report. The finding is correct, but the issue already > exists before. The intention of my patch was to increase readibility, > not to change the behavior. > > My problem is that i'm not aware how to handle the error case here. > > @Dave Should we bail out or ignore the error? Neither vchiq_mmal_port_disable nor vchiq_mmal_port_connect_tunnel(inst, src, NULL) should fail - that would imply something seriously wrong within the firmware. I'd suggest ignore the error - if something is seriously wrong then it'll fail and bail later on. Looking at vchiq_mmal_port_connect_tunnel, if dst is NULL (as in this case) then the first thing it does is call port_disable, same as vchiq_mmal_port_disable. The call to vchiq_mmal_port_disable can therefore be removed. Same in vidioc_overlay. Dave > Best regards > Stefan > > > > > Reported-by: coverity-bot <keescook+coverity-bot@xxxxxxxxxxxx> > > Addresses-Coverity-ID: 1492591 ("Code maintainability issues") > > Fixes: 1a59532382a6 ("staging: bcm2835-camera: Move video component setup in its own function") > > > > Thanks for your attention! > > >
