Fix for issue #28012: pow overflow inconsistency

Overflows in integer pow() computations would be missed if they
preceded a 0 bit of the exponent being processed. This made
calls such as 2i32.pow(1024) not trigger an overflow.

src/libcore/num/mod.rs

@@ -537,25 +537,21 @@ pub fn pow(self, mut exp: u32) -> Self {
             let mut base = self;
             let mut acc = Self::one();
 
-            let mut prev_base = self;
-            let mut base_oflo = false;
-            while exp > 0 {
+            while exp > 1 {
                 if (exp & 1) == 1 {
-                    if base_oflo {
-                        // ensure overflow occurs in the same manner it
-                        // would have otherwise (i.e. signal any exception
-                        // it would have otherwise).
-                        acc = acc * (prev_base * prev_base);
-                    } else {
-                        acc = acc * base;
-                    }
+                    acc = acc * base;
                 }
-                prev_base = base;
-                let (new_base, new_base_oflo) = base.overflowing_mul(base);
-                base = new_base;
-                base_oflo = new_base_oflo;
                 exp /= 2;
+                base = base * base;
+            }
+
+            // Deal with the final bit of the exponent separately, since
+            // squaring the base afterwards is not necessary and may cause a
+            // needless overflow.
+            if exp == 1 {
+                acc = acc * base;
             }
+
             acc
         }