Re: [PATCH v15 03/15] FWU: Add FWU metadata access driver for GPT partitioned block devices

[Ilias Apalodimas <ilias.apalodimas@linaro.org>]

On Fri, Oct 21, 2022 at 09:57:45PM +0530, Sughosh Ganu wrote:
> hi Ilias,
> 
> On Fri, 21 Oct 2022 at 20:33, Ilias Apalodimas
> <ilias.apalodimas@linaro.org> wrote:
> >
> > Hi Sughosh
> >
> > > +{
> > > +     int ret;
> > > +     u32 len, blk_start, blkcnt;
> > > +     struct disk_partition info;
> > > +
> > > +     ALLOC_CACHE_ALIGN_BUFFER_PAD(struct fwu_mdata, mdata_aligned, 1,
> > > +                                  desc->blksz);
> > > +
> > > +     if (!mdata)
> > > +             return -ENOMEM;
> >
> > ENOMEM is usually for allocation failures this is an -EINVAL
> >
> > > +
> > > +     ret = gpt_get_mdata_disk_part(desc, &info, part_num);
> > > +     if (ret < 0) {
> > > +             printf("Unable to get the FWU metadata partition\n");
> > > +             return -ENOENT;
> > > +     }
> > > +
> > > +     len = sizeof(*mdata);
> > > +     blkcnt = BLOCK_CNT(len, desc);
> > > +     if (blkcnt > info.size) {
> > > +             log_debug("Block count exceeds FWU metadata partition size\n");
> > > +             return -ERANGE;
> > > +     }
> > > +
> > > +     blk_start = info.start;
> > > +     if (access == MDATA_READ) {
> > > +             if (blk_dread(desc, blk_start, blkcnt, mdata_aligned) != blkcnt) {
> > > +                     log_debug("Error reading FWU metadata from the device\n");
> > > +                     return -EIO;
> > > +             }
> > > +             memcpy(mdata, mdata_aligned, sizeof(struct fwu_mdata));
> > > +     } else {
> >
> > else if ?
> >
> > > +             if (blk_dwrite(desc, blk_start, blkcnt, mdata) != blkcnt) {
> > > +                     log_debug("Error writing FWU metadata to the device\n");
> > > +                     return -EIO;
> > > +             }
> > > +     }
> > > +
> > > +     return 0;
> > > +}
> > > +
> > > +static int fwu_gpt_update_mdata(struct udevice *dev, struct fwu_mdata *mdata)
> > > +{
> > > +     int ret;
> > > +     struct blk_desc *desc;
> > > +     uint mdata_parts[2];
> > > +     struct fwu_mdata_gpt_blk_priv *priv = dev_get_priv(dev);
> > > +
> > > +     desc = dev_get_uclass_plat(priv->blk_dev);
> >
> > dev_get_uclass_plat might return NULL,  gpt_read_write_mdata() doesn't check
> > against NULL and then it ends up calling gpt_get_mdata_disk_part() which
> > then calls part_get_info().  I don't think anyone checks for the desc ptr
> >
> > And I think this is a problem  overall in all the callbacks belowe that
> > invoke dev_get_uclass_plat()
> 
> dev_get_uclass_plat() returns NULL only when the dev parameter passed
> to the function is NULL. And that won't happen since that would mean
> that the mdata_dev is NULL, and in such a scenario the driver's probe
> function would fail. So these functions would not get called with desc
> being NULL.
> 
> -sughosh

Ok makes sense.

Acked-by: Ilias Apalodimas <ilias.apalodimas@linaro.org>

Thanks
/Ilias


> 
> >
[...]