diff --git a/rust-version b/rust-version
index 45cd53ac6c1..e7e504a38a7 100644
--- a/rust-version
+++ b/rust-version
@@ -1 +1 @@
-c9c4b5d7276297679387189d96a952f2b760e7ad
+5bc98076f37dd8c1476de4bbe0515c55a65332b7
diff --git a/src/shims/intrinsics.rs b/src/shims/intrinsics.rs
index f2f046a3ada..d684b41ed81 100644
--- a/src/shims/intrinsics.rs
+++ b/src/shims/intrinsics.rs
@@ -305,6 +305,32 @@ fn call_intrinsic(
                 this.write_scalar(res, dest)?;
             }
 
+            // SIMD operations
+            "simd_add" | "simd_sub" | "simd_mul" | "simd_div" => {
+                let &[ref left, ref right] = check_arg_count(args)?;
+                let (left, left_len) = this.operand_to_simd(left)?;
+                let (right, right_len) = this.operand_to_simd(right)?;
+                let (dest, dest_len) = this.place_to_simd(dest)?;
+
+                assert_eq!(dest_len, left_len);
+                assert_eq!(dest_len, right_len);
+
+                let op = match intrinsic_name {
+                    "simd_add" => mir::BinOp::Add,
+                    "simd_sub" => mir::BinOp::Sub,
+                    "simd_mul" => mir::BinOp::Mul,
+                    "simd_div" => mir::BinOp::Div,
+                    _ => unreachable!(),
+                };
+
+                for i in 0..dest_len {
+                    let left = this.read_immediate(&this.mplace_index(&left, i)?.into())?;
+                    let right = this.read_immediate(&this.mplace_index(&right, i)?.into())?;
+                    let dest = this.mplace_index(&dest, i)?.into();
+                    this.binop_ignore_overflow(op, &left, &right, &dest)?;
+                }
+            }
+
             // Atomic operations
             "atomic_load" => this.atomic_load(args, dest, AtomicReadOp::SeqCst)?,
             "atomic_load_relaxed" => this.atomic_load(args, dest, AtomicReadOp::Relaxed)?,
diff --git a/tests/run-pass/portable-simd.rs b/tests/run-pass/portable-simd.rs
new file mode 100644
index 00000000000..42a6befd868
--- /dev/null
+++ b/tests/run-pass/portable-simd.rs
@@ -0,0 +1,27 @@
+#![feature(portable_simd)]
+use std::simd::*;
+
+fn simd_ops_f32() {
+    let a = f32x4::splat(10.0);
+    let b = f32x4::from_array([1.0, 2.0, 3.0, 4.0]);
+    assert_eq!(a + b, f32x4::from_array([11.0, 12.0, 13.0, 14.0]));
+    assert_eq!(a - b, f32x4::from_array([9.0, 8.0, 7.0, 6.0]));
+    assert_eq!(a * b, f32x4::from_array([10.0, 20.0, 30.0, 40.0]));
+    assert_eq!(b / a, f32x4::from_array([0.1, 0.2, 0.3, 0.4]));
+    assert_eq!(a / 2.0, f32x4::splat(5.0));
+}
+
+fn simd_ops_i32() {
+    let a = i32x4::splat(10);
+    let b = i32x4::from_array([1, 2, 3, 4]);
+    assert_eq!(a + b, i32x4::from_array([11, 12, 13, 14]));
+    assert_eq!(a - b, i32x4::from_array([9, 8, 7, 6]));
+    assert_eq!(a * b, i32x4::from_array([10, 20, 30, 40]));
+    assert_eq!(a / b, i32x4::from_array([10, 5, 3, 2]));
+    assert_eq!(a / 2, i32x4::splat(5));
+}
+
+fn main() {
+    simd_ops_f32();
+    simd_ops_i32();
+}