On Tue, 7 Aug 2018, Dave Hansen wrote:
> On 08/07/2018 10:29 AM, Sean Christopherson wrote:
> > if (unlikely(fault_in_kernel_space(address))) {
> > + /*
> > + * We should never encounter a protection keys fault on a
> > + * kernel address as kernel address are always mapped with
> > + * _PAGE_USER=0, i.e. PKRU isn't enforced.
> > + */
> > + if (WARN_ON_ONCE(error_code & X86_PF_PK))
> > + goto bad_kernel_address;
>
> I just realized one more thing: the vsyscall page can bite us here.
> It's at a fault_in_kernel_space() address and we *can* trigger a pkey
> fault on it if we jump to an instruction that reads from a
> pkey-protected area.
>
> We can make a gadget out of unaligned vsyscall instructions that does
> that. See:
>
> 0xffffffffff600002: shlb $0x0,0x0(%rax)
>
> Then, we turn off access to all pkeys, including pkey-0, then jump to
> the unaligned vsyscall instruction, which reads %rax, which is a kernel
> address:
>
> asm("movl $0xffffffff, %eax;\
> movl $0x00000000, %ecx;\
> movl $0x00000000, %edx;\
> wrpkru;\
> movq $0xffffffffff600000, %rax;\
> movq $0xffffffffff600002, %rbx;\
> jmpq *%rbx;");
>
> So, my bad. It was not a good suggestion to do a WARN_ON(). But, the
> other funny thing is I would have expected spurious_fault() to get us
> into a fault loop, which it doesn't. It's definitely getting *called*
> with my little test program (I see it in ftrace) but it's not quite
> doing what I expect.
>
> I need to dig a bit more.
Given the time span you should be close to ground water with your digging
by now.
Thanks,
tglx
