On Sat, Jan 12, 2019 at 01:34:42AM -0600, Segher Boessenkool wrote:
> On Sat, Jan 12, 2019 at 08:44:26AM +0300, Dan Carpenter wrote:
> > On Sat, Jan 12, 2019 at 11:30:35AM +1100, Balbir Singh wrote:
> > > On Wed, Jan 09, 2019 at 01:23:29PM +0300, Dan Carpenter wrote:
> > > > There is a typo so we accidentally allocate enough memory for a pointer
> > > > when we wanted to allocate enough for a struct.
> > > >
> > > > Fixes: 0bd971676e68 ("powerpc/powernv/npu: Add compound IOMMU groups")
> > > > Signed-off-by: Dan Carpenter <dan.carpenter@xxxxxxxxxx>
> > > > ---
> > > > arch/powerpc/platforms/powernv/npu-dma.c | 2 +-
> > > > 1 file changed, 1 insertion(+), 1 deletion(-)
> > > >
> > > > diff --git a/arch/powerpc/platforms/powernv/npu-dma.c b/arch/powerpc/platforms/powernv/npu-dma.c
> > > > index d7f742ed48ba..3f58c7dbd581 100644
> > > > --- a/arch/powerpc/platforms/powernv/npu-dma.c
> > > > +++ b/arch/powerpc/platforms/powernv/npu-dma.c
> > > > @@ -564,7 +564,7 @@ struct iommu_table_group *pnv_try_setup_npu_table_group(struct pnv_ioda_pe *pe)
> > > > }
> > > > } else {
> > > > /* Create a group for 1 GPU and attached NPUs for POWER8 */
> > > > - pe->npucomp = kzalloc(sizeof(pe->npucomp), GFP_KERNEL);
> > > > + pe->npucomp = kzalloc(sizeof(*pe->npucomp), GFP_KERNEL);
> > >
> > > To avoid these in the future, I wonder if instead of sizeof(pe->npucomp), we insist on
> > > sizeof structure
> > >
> > > pe->npucomp = kzalloc(sizeof(struct npucomp), GFP_KERNEL);
> > >
> >
> > The latest kernel fashion is sizeof(*ptr). It can go wrong either way.
> > I don't have strong feelings about it. These sorts of bugs don't last
> > long because they're caught in testing or with static analysis.
>
> And it is easy to see someone forgot the * in "sizeof *ptr", and with
> experience it will just automatically look wrong if it is forgotten; but
> it isn't obvious at all if the wrong struct is used, which cannot happen
> with the *ptr form, but happens frequently with the "sizeof(struct x)"
> form.
It doesn't happen very frequently. I look for this with Smatch. The
results I see are when the first few members of a struct are a header
and the actual size can vary.
hdr = alloc(sizeof(struct larger_struct));
regards,
dan carpenter
