Re: [PATCH] libelf: improve PVH elfnote parsing

[Roger Pau Monné <roger.pau@xxxxxxxxxx>]

On Fri, May 14, 2021 at 11:11:14AM -0400, Jason Andryuk wrote:
> On Fri, May 14, 2021 at 9:50 AM Roger Pau Monne <roger.pau@xxxxxxxxxx> wrote:
> >
> > Pass an hvm boolean parameter to the elf note parsing and checking
> > routines, so that better checking can be done in case libelf is
> > dealing with an hvm container.
> >
> > elf_xen_note_check shouldn't return early unless PHYS32_ENTRY is set
> > and the container is of type HVM, or else the loader and version
> > checks would be avoided for kernels intended to be booted as PV but
> > that also have PHYS32_ENTRY set.
> >
> > Adjust elf_xen_addr_calc_check so that the virtual addresses are
> > actually physical ones (by setting virt_base and elf_paddr_offset to
> > zero) when the container is of type HVM, as that container is always
> > started with paging disabled.
> 
> Should elf_xen_addr_calc_check be changed so that PV operates on
> virtual addresses and HVM operates on physical addresses?

Right... I was aiming with getting away with something simpler and
just assume phys == virt on HVM in order to avoid more complicated
changes and the need to introduce new fields on the structure.

> I worked on some patches for this a while back, but lost track when
> other work pulled me away.  I'll send out what I had, but I think I
> had not tested many of the cases.  Also, I had other questions about
> the approach.  Fundamentally, what notes and limits need to be checked
> for PVH vs. PV?

Those are only sanity checks to assert that the image is kind of fine,
libelf also has checks when loading stuff to make sure a malicious elf
payload cannot fool the loader.

I'm unlikely to be able to do much work on this aside from this
current patch.

Thanks, Roger.