[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [PATCH 15/16] x86/entry: Adjust guest paths to be shadow stack compatible

  • To: Jan Beulich <jbeulich@xxxxxxxx>, Xen-devel <xen-devel@xxxxxxxxxxxxxxxxxxxx>
  • From: Andrew Cooper <andrew.cooper3@xxxxxxxxxx>
  • Date: Thu, 7 May 2020 16:50:26 +0100
  • Authentication-results: esa5.hc3370-68.iphmx.com; dkim=none (message not signed) header.i=none; spf=None smtp.pra=andrew.cooper3@xxxxxxxxxx; spf=Pass smtp.mailfrom=Andrew.Cooper3@xxxxxxxxxx; spf=None smtp.helo=postmaster@xxxxxxxxxxxxxxx
  • Cc: Wei Liu <wl@xxxxxxx>, Roger Pau Monné <roger.pau@xxxxxxxxxx>
  • Delivery-date: Thu, 07 May 2020 15:50:44 +0000
  • Ironport-sdr: iFNkUIlYTvERwQDDESQtbzzPgDl/Qi+1+4Vrv83R7ryE58arRLw/vo9dRbseKkQa3FehAcaqX+ 9tcastObtS3ym4RMoVpMMEnxbbV+byx9sqH5L28FG8zwGITZKFStzLcG4CkEpU81dRj6AbnOng JIOKbiRO75Sa9s5+wfImQrR/ionsQRlVTQt7yoI2VB4AFG2iK0ou4mbVlnJQdSLEJQtrA0Co0Y 8f/wmoQxB5+SOLruUm6hyXgDaoun3JHXOaknQVWdkRKMex0qW4lCyghQuv72oG3xfeW78NKQBL O0o=
  • List-id: Xen developer discussion <xen-devel.lists.xenproject.org>

On 07/05/2020 15:12, Jan Beulich wrote:
> On 02.05.2020 00:58, Andrew Cooper wrote:
>> The SYSCALL/SYSEXIT paths need to use {SET,CLR}SSBSY.
> I take it you mean SYSRET, not SYSEXIT.

I do, sorry.

> I do think though that you
> also need to deal with the SYSENTER entry point we have.

Oh - so we do.

>> --- a/xen/arch/x86/x86_64/compat/entry.S
>> +++ b/xen/arch/x86/x86_64/compat/entry.S
>> @@ -198,7 +198,7 @@ ENTRY(cr4_pv32_restore)
>>  /* See lstar_enter for entry register state. */
>>  ENTRY(cstar_enter)
>> -        /* sti could live here when we don't switch page tables below. */
>> +        ALTERNATIVE "", "setssbsy", X86_FEATURE_XEN_SHSTK
> I don't see why you delete the comment here (or elsewhere). While
> I recall you not really wanting them there, I still think they're
> useful to have, and they shouldn't be deleted as a side effect of
> an entirely unrelated change. Of course they need to live after
> your insertions then.

Do you not remember Juergen performance testing results concerning this
comment?  The results were provably worse.

It is a useless comment.  Sure, its technically accurate, but so are an
arbitrarily large number of other comments about how we could permute
the code.

It has already been concluded that we won't be making the suggested
change.  Having a /* TODO - doing X will make the system slower */ isn't
something we should have adding to the complexity of the code, and
tricking people into thinking that something should be done.

>> --- a/xen/arch/x86/x86_64/entry.S
>> +++ b/xen/arch/x86/x86_64/entry.S
>> @@ -194,6 +194,15 @@ restore_all_guest:
>>          movq  8(%rsp),%rcx            # RIP
>>          ja    iret_exit_to_guest
>> +        /* Clear the supervisor shadow stack token busy bit. */
>> +.macro rag_clrssbsy
>> +        push %rax
>> +        rdsspq %rax
>> +        clrssbsy (%rax)
>> +        pop %rax
>> +.endm
>> +        ALTERNATIVE "", rag_clrssbsy, X86_FEATURE_XEN_SHSTK
> In principle you could get away without spilling %rax:
>         cmpl  $1,%ecx
>         ja    iret_exit_to_guest
>         /* Clear the supervisor shadow stack token busy bit. */
> .macro rag_clrssbsy
>         rdsspq %rcx
>         clrssbsy (%rcx)
> .endm
>         ALTERNATIVE "", rag_clrssbsy, X86_FEATURE_XEN_SHSTK
>         movq  8(%rsp),%rcx            # RIP
>         cmpw  $FLAT_USER_CS32,16(%rsp)# CS
>         movq  32(%rsp),%rsp           # RSP
>         je    1f
>         sysretq
> 1:      sysretl
>         ALIGN
> /* No special register assumptions. */
> iret_exit_to_guest:
>         movq  8(%rsp),%rcx            # RIP
>         andl  $~(X86_EFLAGS_IOPL|X86_EFLAGS_NT|X86_EFLAGS_VM),24(%rsp)
>         ...
> Also - what about CLRSSBSY failing? It would seem easier to diagnose
> this right here than when getting presumably #DF upon next entry into
> Xen. At the very least I think it deserves a comment if an error case
> does not get handled.

I did consider this, but ultimately decided against it.

You can't have an unlikely block inside a alternative block because the
jmp's displacement doesn't get fixed up.  Keeping everything inline puts
an incorrect statically-predicted branch in program flow.

Most importantly however, is that the SYSRET path is vastly less common
than the IRET path.  There is no easy way to proactively spot problems
in the IRET path, which means that conditions leading to a problem are
already far more likely to manifest as #DF, so there is very little
value in adding complexity to the SYSRET path in the first place.

> Somewhat similar for SETSSBSY, except there things get complicated by
> it raising #CP instead of setting EFLAGS.CF: Aiui it would require us
> to handle #CP on an IST stack in order to avoid #DF there.

Right, but having #CP as IST gives us far worse problems.

Being able to spot #CP vs #DF doesn't help usefully.  Its still some
arbitrary period of time after the damage was done.

Any nesting of #CP (including fault on IRET out) results in losing
program state and entering an infinite loop.

The cases which end up as #DF are properly fatal to the system, and we
at least get a clean crash out it.

>> @@ -877,6 +886,14 @@ handle_ist_exception:
>>          movl  $UREGS_kernel_sizeof/8,%ecx
>>          movq  %rdi,%rsp
>>          rep   movsq
>> +
>> +        /* Switch Shadow Stacks */
>> +.macro ist_switch_shstk
>> +        rdsspq %rdi
>> +        clrssbsy (%rdi)
>> +        setssbsy
>> +.endm
> Could you extend the comment to mention the caveat that you point
> out in the description?





Lists.xenproject.org is hosted with RackSpace, monitoring our
servers 24x7x365 and backed by RackSpace's Fanatical Support®.