diff mbox

[for-4.8] x86/emul: Reject LGDT/LIDT attempts with non-canonical base addresses

Message ID 1478098337-13933-1-git-send-email-andrew.cooper3@citrix.com (mailing list archive)
State New, archived
Headers show

Commit Message

Andrew Cooper Nov. 2, 2016, 2:52 p.m. UTC
No sane OS would deliberately try this, but make Xen's emulation match real
hardware by delivering #GP(0), rather than suffering a VMEntry failure.

Signed-off-by: Andrew Cooper <andrew.cooper3@citrix.com>
---
CC: Jan Beulich <JBeulich@suse.com>
CC: Wei Liu <wei.liu2@citrix.com>
---
 xen/arch/x86/x86_emulate/x86_emulate.c | 2 ++
 1 file changed, 2 insertions(+)

Comments

Jan Beulich Nov. 2, 2016, 3:42 p.m. UTC | #1
>>> On 02.11.16 at 15:52, <andrew.cooper3@citrix.com> wrote:
> No sane OS would deliberately try this, but make Xen's emulation match real
> hardware by delivering #GP(0), rather than suffering a VMEntry failure.
> 
> Signed-off-by: Andrew Cooper <andrew.cooper3@citrix.com>

Reviewed-by: Jan Beulich <jbeulich@suse.com>
albeit ...

> --- a/xen/arch/x86/x86_emulate/x86_emulate.c
> +++ b/xen/arch/x86/x86_emulate/x86_emulate.c
> @@ -4443,6 +4443,8 @@ x86_emulate(
>                   (rc = read_ulong(ea.mem.seg, ea.mem.off+2,
>                                    &base, mode_64bit() ? 8 : 4, ctxt, ops)) )
>                  goto done;
> +            generate_exception_if(mode_64bit() &&
> +                                  !is_canonical_address(base), EXC_GP, 0);

... I don't think the mode_64bit() part is needed.

Jan
Andrew Cooper Nov. 2, 2016, 5:39 p.m. UTC | #2
On 02/11/16 15:42, Jan Beulich wrote:
>>>> On 02.11.16 at 15:52, <andrew.cooper3@citrix.com> wrote:
>> No sane OS would deliberately try this, but make Xen's emulation match real
>> hardware by delivering #GP(0), rather than suffering a VMEntry failure.
>>
>> Signed-off-by: Andrew Cooper <andrew.cooper3@citrix.com>
> Reviewed-by: Jan Beulich <jbeulich@suse.com>
> albeit ...
>
>> --- a/xen/arch/x86/x86_emulate/x86_emulate.c
>> +++ b/xen/arch/x86/x86_emulate/x86_emulate.c
>> @@ -4443,6 +4443,8 @@ x86_emulate(
>>                   (rc = read_ulong(ea.mem.seg, ea.mem.off+2,
>>                                    &base, mode_64bit() ? 8 : 4, ctxt, ops)) )
>>                  goto done;
>> +            generate_exception_if(mode_64bit() &&
>> +                                  !is_canonical_address(base), EXC_GP, 0);
> ... I don't think the mode_64bit() part is needed.

True - as we explicitly zero sreg to start with, the canonical base
check is safe in all modes.  I will drop it from v2.

~Andrew
Wei Liu Nov. 3, 2016, 10:43 a.m. UTC | #3
On Wed, Nov 02, 2016 at 09:42:33AM -0600, Jan Beulich wrote:
> >>> On 02.11.16 at 15:52, <andrew.cooper3@citrix.com> wrote:
> > No sane OS would deliberately try this, but make Xen's emulation match real
> > hardware by delivering #GP(0), rather than suffering a VMEntry failure.
> > 
> > Signed-off-by: Andrew Cooper <andrew.cooper3@citrix.com>
> 
> Reviewed-by: Jan Beulich <jbeulich@suse.com>

Release-acked-by: Wei Liu <wei.liu2@citrix.com>
diff mbox

Patch

diff --git a/xen/arch/x86/x86_emulate/x86_emulate.c b/xen/arch/x86/x86_emulate/x86_emulate.c
index 7a707dc..6b940c4 100644
--- a/xen/arch/x86/x86_emulate/x86_emulate.c
+++ b/xen/arch/x86/x86_emulate/x86_emulate.c
@@ -4443,6 +4443,8 @@  x86_emulate(
                  (rc = read_ulong(ea.mem.seg, ea.mem.off+2,
                                   &base, mode_64bit() ? 8 : 4, ctxt, ops)) )
                 goto done;
+            generate_exception_if(mode_64bit() &&
+                                  !is_canonical_address(base), EXC_GP, 0);
             sreg.base = base;
             sreg.limit = limit;
             if ( op_bytes == 2 )