Re: [PATCH v38 21/24] x86/vdso: Implement a vDSO for Intel SGX enclave call

[Andy Lutomirski]

On Mon, Sep 28, 2020 at 11:08 AM H.J. Lu <hjl.tools@xxxxxxxxx> wrote:
>
> On Mon, Sep 28, 2020 at 9:44 AM Andrew Cooper <andrew.cooper3@xxxxxxxxxx> wrote:
> >
> > On 28/09/2020 01:58, Jarkko Sakkinen wrote:
> > > On Fri, Sep 25, 2020 at 07:23:59PM +0100, Andrew Cooper wrote:
> > >> On 15/09/2020 12:28, Jarkko Sakkinen wrote:
> > >>> diff --git a/arch/x86/entry/vdso/vsgx_enter_enclave.S b/arch/x86/entry/vdso/vsgx_enter_enclave.S
> > >>> new file mode 100644
> > >>> index 000000000000..adbd59d41517
> > >>> --- /dev/null
> > >>> +++ b/arch/x86/entry/vdso/vsgx_enter_enclave.S
> > >>> @@ -0,0 +1,157 @@
> > >>> +SYM_FUNC_START(__vdso_sgx_enter_enclave)
> > >>> <snip>
> > >>> +.Lretpoline:
> > >>> +   call    2f
> > >>> +1: pause
> > >>> +   lfence
> > >>> +   jmp     1b
> > >>> +2: mov     %rax, (%rsp)
> > >>> +   ret
> > >> I hate to throw further spanners in the work, but this is not compatible
> > >> with CET, and the user shadow stack work in progress.
> > > CET goes beyond my expertise. Can you describe, at least rudimentary,
> > > how this code is not compatible?
> >
> > CET Shadow Stacks detect attacks which modify the return address on the
> > stack.
> >
> > Retpoline *is* a ROP gadget.  It really does modify the return address
> > on the stack, even if its purpose is defensive (vs Spectre v2) rather
> > than malicious.
> >
> > >> Whichever of these two large series lands first is going to inflict
> > >> fixing this problem on the other.
> > >>
> > >> As the vdso text is global (to a first approximation), it must not be a
> > >> retpoline if any other process is liable to want to use CET-SS.
> > > Why is that?
> >
> > Because when CET-SS is enabled, the ret will suffer a #CP exception
> > (return address on the stack not matching the one recorded in the shadow
> > stack), which I presume/hope is wired into SIGSEGV.
> >
>
> Here is the CET compatible retpoline:
>
> endbr64
> /* Check if shadow stack is in use.  NB: R11 is the only usable
>    scratch register for function calls.  */
> xorl %r11d, %r11d
> rdsspq %r11
> testq %r11, %r11
> jnz 3f
> call 2f
> 1:
> pause
> lfence
> jmp 1b
> 2:
> mov %rax, (%rsp)
> ret
> 3:
> /* Shadow stack is in use.  Make the indirect call.  */
> call *%rax
> ret

What do we expect user programs to do on CET systems?  It would be
nice if we could instead ALTERNATIVE this out if X86_FEATURE_SHSTK.

--Andy