diff --git a/gcc/ChangeLog b/gcc/ChangeLog
index 0c100df55aa9..3f74243c00b9 100644
--- a/gcc/ChangeLog
+++ b/gcc/ChangeLog
@@ -1,3 +1,9 @@
+2012-08-17  Marc Glisse  <marc.glisse@inria.fr>
+
+	* simplify-rtx.c (simplify_binary_operation_1): Optimize shuffle of
+	a concatenation.
+
+
 2012-08-17  H.J. Lu  <hongjiu.lu@intel.com>
 
 	* stor-layout.c (compute_record_mode): Replace
diff --git a/gcc/simplify-rtx.c b/gcc/simplify-rtx.c
index 16dbd8a03eaf..a878048ac979 100644
--- a/gcc/simplify-rtx.c
+++ b/gcc/simplify-rtx.c
@@ -3253,6 +3253,23 @@ simplify_binary_operation_1 (enum rtx_code code, enum machine_mode mode,
 	      subop0 = XEXP (XEXP (trueop0, i0 / 2), i0 % 2);
 	      subop1 = XEXP (XEXP (trueop0, i1 / 2), i1 % 2);
 
+	      return simplify_gen_binary (VEC_CONCAT, mode, subop0, subop1);
+	    }
+
+	  if (XVECLEN (trueop1, 0) == 2
+	      && CONST_INT_P (XVECEXP (trueop1, 0, 0))
+	      && CONST_INT_P (XVECEXP (trueop1, 0, 1))
+	      && GET_CODE (trueop0) == VEC_CONCAT
+	      && GET_MODE (trueop0) == mode)
+	    {
+	      unsigned int i0 = INTVAL (XVECEXP (trueop1, 0, 0));
+	      unsigned int i1 = INTVAL (XVECEXP (trueop1, 0, 1));
+	      rtx subop0, subop1;
+
+	      gcc_assert (i0 < 2 && i1 < 2);
+	      subop0 = XEXP (trueop0, i0);
+	      subop1 = XEXP (trueop0, i1);
+
 	      return simplify_gen_binary (VEC_CONCAT, mode, subop0, subop1);
 	    }
 	}
diff --git a/gcc/testsuite/ChangeLog b/gcc/testsuite/ChangeLog
index 3936642d60f6..6f561b860395 100644
--- a/gcc/testsuite/ChangeLog
+++ b/gcc/testsuite/ChangeLog
@@ -1,3 +1,7 @@
+2012-08-17  Marc Glisse  <marc.glisse@inria.fr>
+
+	* gcc.target/i386/perm-concat.c: New test.
+
 2012-08-17  Julian Brown  <julian@codesourcery.com>
 
 	* gcc.target/arm/div64-unwinding.c: New test.
diff --git a/gcc/testsuite/gcc.target/i386/perm-concat.c b/gcc/testsuite/gcc.target/i386/perm-concat.c
new file mode 100644
index 000000000000..10955c2073e4
--- /dev/null
+++ b/gcc/testsuite/gcc.target/i386/perm-concat.c
@@ -0,0 +1,13 @@
+/* { dg-do compile } */
+/* { dg-options "-O -mavx -mfpmath=sse" } */
+
+typedef double v2df __attribute__ ((__vector_size__ (16)));
+
+v2df
+f (double d)
+{
+  v2df x = {-d, d};
+  return __builtin_ia32_vpermilpd (x, 1);
+}
+
+/* { dg-final { scan-assembler-not "\tvpermilpd\[ \t\]" } } */