[Xen-devel] [PATCH 1/6] x86/hvm: Fixes to hvmemul_insn_fetch()

[Andrew Cooper <andrew.cooper3@xxxxxxxxxx>]

Force insn_off to a single byte, as offset can wrap around or truncate with
respect to sh_ctxt->insn_buf_eip under a number of normal circumstances.

Furthermore, don't use an ASSERT() for bounds checking the write into
hvmemul_ctxt->insn_buf[].

Signed-off-by: Andrew Cooper <andrew.cooper3@xxxxxxxxxx>
---
CC: Jan Beulich <JBeulich@xxxxxxxx>
CC: Paul Durrant <paul.durrant@xxxxxxxxxx>

For anyone wondering, this is way to explot XSA-186
---
 xen/arch/x86/hvm/emulate.c | 15 +++++++++++++--
 1 file changed, 13 insertions(+), 2 deletions(-)

diff --git a/xen/arch/x86/hvm/emulate.c b/xen/arch/x86/hvm/emulate.c
index 11e4aba..495e312 100644
--- a/xen/arch/x86/hvm/emulate.c
+++ b/xen/arch/x86/hvm/emulate.c
@@ -939,7 +939,8 @@ int hvmemul_insn_fetch(
 {
     struct hvm_emulate_ctxt *hvmemul_ctxt =
         container_of(ctxt, struct hvm_emulate_ctxt, ctxt);
-    unsigned int insn_off = offset - hvmemul_ctxt->insn_buf_eip;
+    /* Careful, as offset can wrap or truncate WRT insn_buf_eip. */
+    uint8_t insn_off = offset - hvmemul_ctxt->insn_buf_eip;
 
     /*
      * Fall back if requested bytes are not in the prefetch cache.
@@ -953,7 +954,17 @@ int hvmemul_insn_fetch(
 
         if ( rc == X86EMUL_OKAY && bytes )
         {
-            ASSERT(insn_off + bytes <= sizeof(hvmemul_ctxt->insn_buf));
+            /*
+             * Will we overflow insn_buf[]?  This shouldn't be able to happen,
+             * which means something went wrong with instruction decoding...
+             */
+            if ( insn_off > sizeof(hvmemul_ctxt->insn_buf) ||
+                 (insn_off + bytes) > sizeof(hvmemul_ctxt->insn_buf) )
+            {
+                ASSERT_UNREACHABLE();
+                return X86EMUL_UNHANDLEABLE;
+            }
+
             memcpy(&hvmemul_ctxt->insn_buf[insn_off], p_data, bytes);
             hvmemul_ctxt->insn_buf_bytes = insn_off + bytes;
         }
-- 
2.1.4


_______________________________________________
Xen-devel mailing list
Xen-devel@xxxxxxxxxxxxx
https://lists.xen.org/xen-devel