Re: [PATCH] PCI: Check value of resource alignment before using __ffs

[Amey Narkhede]

On 21/04/22 01:41PM, Leon Romanovsky wrote:
> On Thu, Apr 22, 2021 at 03:13:23PM +0530, Amey Narkhede wrote:
> > On 21/04/22 09:43AM, Leon Romanovsky wrote:
> > > On Thu, Apr 22, 2021 at 12:17:47AM +0530, Amey Narkhede wrote:
> > > > Return value of __ffs is undefined if no set bit exists in
> > > > its argument. This indicates that the associated BAR has
> > > > invalid alignment.
> > > >
> > > > Signed-off-by: Amey Narkhede <ameynarkhede03@xxxxxxxxx>
> > > > ---
> > > >  drivers/pci/setup-bus.c | 5 +++++
> > > >  1 file changed, 5 insertions(+)
> > > >
> > > > diff --git a/drivers/pci/setup-bus.c b/drivers/pci/setup-bus.c
> > > > index 2ce636937c6e..44e8449418ae 100644
> > > > --- a/drivers/pci/setup-bus.c
> > > > +++ b/drivers/pci/setup-bus.c
> > > > @@ -1044,6 +1044,11 @@ static int pbus_size_mem(struct pci_bus *bus, unsigned long mask,
> > > >  			 * resources.
> > > >  			 */
> > > >  			align = pci_resource_alignment(dev, r);
> > > > +			if (!align) {
> > > > +				pci_warn(dev, "BAR %d: %pR has bogus alignment\n",
> > > > +					 i, r);
> > > > +				continue;
> > > > +			}
> > >
> > > I see that you copied it from pdev_sort_resources(), but it is
> > > incorrect change, see how negative order is handled and later
> > > ARRAY_SIZE() check.
> > >
> > > Thanks
> > >
> > Is it guaranteed that it will return value which will result
> > in negative value or >= ARRAY_SIZE? Comment on __ffs says value
> > is undefined for 0 that means it could be anything or am I missing
> > something?
>
> diff --git a/drivers/pci/setup-bus.c b/drivers/pci/setup-bus.c
> index 2ce636937c6e..ce5380bdd2fd 100644
> --- a/drivers/pci/setup-bus.c
> +++ b/drivers/pci/setup-bus.c
> @@ -1044,10 +1044,11 @@ static int pbus_size_mem(struct pci_bus *bus, unsigned long mask,
>                          * resources.
>                          */
>                         align = pci_resource_alignment(dev, r);
> -                       order = __ffs(align) - 20;
> -                       if (order < 0)
> -                               order = 0;
> -                       if (order >= ARRAY_SIZE(aligns)) {
> +                       if (align) {
> +                               order = __ffs(align) - 20;
> +                               order = (order < 0) ? 0 : order;
> +                       }
> +                       if (!align || order >= ARRAY_SIZE(aligns)) {
>                                 pci_warn(dev, "disabling BAR %d: %pR (bad alignment %#llx)\n",
>                                          i, r, (unsigned long long) align);
>                                 r->flags = 0;
>
>
Oh I see. Thanks. I'll correct this in v2.

Thanks,
Amey