Re: acpi_pcihp_eject_slot() bug if passed 'slots == 0'

[Michael S. Tsirkin]

On Thu, Mar 26, 2020 at 01:50:41PM +0100, Igor Mammedov wrote:
> On Thu, 26 Mar 2020 13:29:01 +0100
> Igor Mammedov <address@hidden> wrote:
> 
> > On Thu, 26 Mar 2020 11:52:36 +0000
> > Peter Maydell <address@hidden> wrote:
> > 
> > > Hi; Coverity spots that if hw/acpi/pcihp.c:acpi_pcihp_eject_slot()
> > > is passed a zero 'slots' argument then ctz32(slots) will return 32,
> > > and then the code that does '1U << slot' is C undefined behaviour
> > > because it's an oversized shift. (This is CID 1421896.)
> > > 
> > > Since the pci_write() function in this file can call
> > > acpi_pcihp_eject_slot() with an arbitrary value from the guest,
> > > I think we need to handle 'slots == 0' safely. But what should
> > > the behaviour be?  
> > 
> > 0 is not valid value, we should ignore and return early in this case
> > like we do with bsel. I'll post a path shortly.
> well, looking more that is only true for main bus, for bridges it can be
> slot number can be zero,

It can but we don't allow slot zero hotplug with SHPC
so it's easier if we don't allow this with ACPI either.

> then AML left shifts it and writes into B0EJ
> which traps into pci_write(, data) and that is supposed to eject
> slot 0 according to guest(AML).
> 
> Michael,
> what's your take on it?
> 
> > 
> > > 
> > > thanks
> > > -- PMM
> > >   
> > 
> >