On Wed, Apr 20, 2022 at 11:51:33AM +0300, Dan Carpenter wrote:
> On Wed, Apr 20, 2022 at 09:50:42AM +0200, Paul Kocialkowski wrote:
> > Hi Dan,
> >
> > On Wed 20 Apr 22, 10:42, Dan Carpenter wrote:
> > > I ran Smatch on this patch.
> >
> > Thanks for doing this!
> >
> > > On Fri, Apr 15, 2022 at 05:37:07PM +0200, Paul Kocialkowski wrote:
> > > > +void sun6i_isp_capture_configure(struct sun6i_isp_device *isp_dev)
> > > > +{
> > > > + unsigned int width, height;
> > > > + unsigned int stride_luma, stride_chroma = 0;
> > > > + unsigned int stride_luma_div4, stride_chroma_div4;
> > > > + const struct sun6i_isp_capture_format *format;
> > > > + const struct v4l2_format_info *info;
> > > > + u32 pixelformat;
> > > > +
> > > > + sun6i_isp_capture_dimensions(isp_dev, &width, &height);
> > > > + sun6i_isp_capture_format(isp_dev, &pixelformat);
> > > > +
> > > > + format = sun6i_isp_capture_format_find(pixelformat);
> > > > + if (WARN_ON(!format))
> > > > + return;
> > > > +
> > > > + sun6i_isp_load_write(isp_dev, SUN6I_ISP_MCH_SIZE_CFG_REG,
> > > > + SUN6I_ISP_MCH_SIZE_CFG_WIDTH(width) |
> > > > + SUN6I_ISP_MCH_SIZE_CFG_HEIGHT(height));
> > > > +
> > > > + info = v4l2_format_info(pixelformat);
> > > > + if (WARN_ON(!info))
> > > > + return;
> > > > +
> > > > + stride_luma = width * info->bpp[0];
> > > > + stride_luma_div4 = DIV_ROUND_UP(stride_luma, 4);
> > > > +
> > > > + if (info->comp_planes > 1) {
> > > > + stride_chroma = width * info->bpp[1] / info->hdiv;
> > > > + stride_chroma_div4 = DIV_ROUND_UP(stride_chroma, 4);
> > >
> > > stride_chroma_div4 is not intialized on the else path.
> >
> > One could say it's not an issue to put an uninitialized value in this situation
> > since the hardware won't be taking it into account but I'll initialize the value
> > early in the next iteration.
> >
>
> My understanding is that it will trigger a KASAN warning at run time.
Or is it KMEMsan? One of those. The one that complains about
uninitialized memory.
regards,
dan carpenter
