x86emul: tighten CLFLUSH emulation

While for us it's not as bad as it was for Linux, their commit
13e457e0ee ("KVM: x86: Emulator does not decode clflush well", by
Nadav Amit <namit@cs.technion.ac.il>) nevertheless points out two
shortcomings in our code: opcode 0F AE /7 is clflush only when it uses
a memory mode (otherwise it's SFENCE) and when there's no REP prefix
(an operand size prefix is fine, as that's CLFLUSHOPT).

Signed-off-by: Jan Beulich <jbeulich@suse.com>
Acked-by: Andrew Cooper <andrew.cooper3@citrix.com>

diff --git a/xen/arch/x86/x86_emulate/x86_emulate.c b/xen/arch/x86/x86_emulate/x86_emulate.c
index be07ffb264ac4d8164ed3e5195b3a02d70a411e9..f13f07d21413ff1a28f99b383ab86631001c999b 100644 (file)
--- a/xen/arch/x86/x86_emulate/x86_emulate.c
+++ b/xen/arch/x86/x86_emulate/x86_emulate.c
@@ -4410,7 +4410,9 @@ x86_emulate(
     case 0xae: /* Grp15 */
         switch ( modrm_reg & 7 )
         {
-        case 7: /* clflush */
+        case 7: /* clflush{,opt} */
+            fail_if(modrm_mod == 3);
+            fail_if(rep_prefix());
             fail_if(ops->wbinvd == NULL);
             if ( (rc = ops->wbinvd(ctxt)) != 0 )
                 goto done;