Re: [PATCH 3.13 106/162] KVM: x86: Fix far-jump to non-canonical check

[Kamal Mostafa <kamal@canonical.com>]

On Wed, 2014-12-31 at 00:27 +0200, Nadav Amit wrote:
> Vinson Lee <vlee@twopensource.com> wrote:
> 
> > On Thu, Nov 6, 2014 at 2:36 PM, Kamal Mostafa <kamal@canonical.com> wrote:
> >> 3.13.11.11 -stable review patch.  If anyone has any objections, please let me know.
> >> 
> >> ------------------
> >> 
> >> From: Nadav Amit <namit@cs.technion.ac.il>
> >> 
> >> commit 7e46dddd6f6cd5dbf3c7bd04a7e75d19475ac9f2 upstream.
> >> 
> >> Commit d1442d85cc30 ("KVM: x86: Handle errors when RIP is set during far
> >> jumps") introduced a bug that caused the fix to be incomplete.  Due to
> >> incorrect evaluation, far jump to segment with L bit cleared (i.e., 32-bit
> >> segment) and RIP with any of the high bits set (i.e, RIP[63:32] != 0) set may
> >> not trigger #GP.  As we know, this imposes a security problem.
> >> 
> >> In addition, the condition for two warnings was incorrect.
> >> 
> >> Fixes: d1442d85cc30ea75f7d399474ca738e0bc96f715
> >> Reported-by: Dan Carpenter <dan.carpenter@oracle.com>
> >> Signed-off-by: Nadav Amit <namit@cs.technion.ac.il>
> >> [Add #ifdef CONFIG_X86_64 to avoid complaints of undefined behavior. - Paolo]
> >> Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
> >> [ kamal: backport to 3.13-stable: omitted WARN_ON fixes (not in 3.13) ]
> >> Signed-off-by: Kamal Mostafa <kamal@canonical.com>
> >> ---
> >> arch/x86/kvm/emulate.c | 4 +++-
> >> 1 file changed, 3 insertions(+), 1 deletion(-)
> >> 
> >> diff --git a/arch/x86/kvm/emulate.c b/arch/x86/kvm/emulate.c
> >> index a440bea..4ae37e7 100644
> >> --- a/arch/x86/kvm/emulate.c
> >> +++ b/arch/x86/kvm/emulate.c
> >> @@ -581,12 +581,14 @@ static inline int assign_eip_far(struct x86_emulate_ctxt *ctxt, ulong dst,
> >>        case 4:
> >>                ctxt->_eip = (u32)dst;
> >>                break;
> >> +#ifdef CONFIG_X86_64
> >>        case 8:
> >>                if ((cs_l && is_noncanonical_address(dst)) ||
> >> -                   (!cs_l && (dst & ~(u32)-1)))
> >> +                   (!cs_l && (dst >> 32) != 0))
> >>                        return emulate_gp(ctxt, 0);
> >>                ctxt->_eip = dst;
> >>                break;
> >> +#endif
> >>        default:
> >>                WARN(1, "unsupported eip assignment size\n");
> >>        }
> >> --
> >> 1.9.1
> >> 
> >> --
> >> To unsubscribe from this list: send the line "unsubscribe stable" in
> >> the body of a message to majordomo@vger.kernel.org
> >> More majordomo info at  http://vger.kernel.org/majordomo-info.html
> > 
> > 
> > Hi.
> > 
> > Should the WARN_ON fixes have been included as well in 3.13.11.11?

Yes, they should have been.  I wrongly omitted the WARN_ON fixes from
the 3.13-stable backport due to some patch application order confusion.
Thanks very much for catching this, Vinson!

> > WARN_ON hunks were added with the backport of "KVM: x86: Handle errors
> > when RIP is set during far jumps" in 3.13.11.11 commit
> > b8ba339d86fb627d54fea929492114d45f6835c2.
> 
> If it was added, it should be fixed. The WARN_ON mistakenly has
> double-negation.
> 
> Nadav

I'll queue up the WARN_ON fixes for the next 3.13-stable.

Thanks again, Vinson and Nadav.

 -Kamal