support multiplication

src/lib.rs

@@ -41,21 +41,21 @@ impl F8 {
 		Self((e << M_BITS) | m)
 	}
 
-	fn merge_unbias(in_m: u8, in_e: i8) -> Self {
+	fn merge_unbias(in_m: u32, in_e: i32) -> Self {
 		if in_m == 0 {
 			return Self(0);
 		}
 		let base_e = in_m.ilog2() as u8;
 		let off = base_e as i8 - M_BITS as i8;
 		let m = if off >= 0 { in_m >> off } else { in_m << -off };
-		let e = (base_e as i8) + in_e + (E_BIAS as i8);
+		let e = (base_e as i32) + in_e + (E_BIAS as i32);
 		if e < 0 {
 			return Self(0);
 		}
-		if e > E_STORAGE_MAX as i8 {
+		if e > E_STORAGE_MAX as i32 {
 			return Self(0xff);
 		}
-		Self::merge(m & M_STORAGE_MAX, e as u8)
+		Self::merge(m as u8 & M_STORAGE_MAX, e as u8)
 	}
 }
 
@@ -163,7 +163,7 @@ mod tests {
 		assert_eq!(f32::from(F8::merge_unbias(3, 1)), 6.0);
 		assert_eq!(f32::from(F8::merge_unbias(3, -1)), 1.5);
 		assert_eq!(
-			f32::from(F8::merge_unbias(EXACT_INT_MAX, 0)),
+			f32::from(F8::merge_unbias(EXACT_INT_MAX.into(), 0)),
 			EXACT_INT_MAX as f32
 		);
 	}

src/ops.rs

@@ -14,7 +14,7 @@ impl std::ops::Add for F8 {
 		}
 		let e_diff = e1 - e2;
 		let m = m1 + (m2 >> e_diff);
-		Self::merge_unbias(m, e1)
+		Self::merge_unbias(m.into(), e1.into())
 	}
 }
 
@@ -34,7 +34,20 @@ impl std::ops::Sub for F8 {
 		let Some(m) = m1.checked_sub(m2 >> e_diff) else {
 			return Self(0);
 		};
-		Self::merge_unbias(m, e1)
+		Self::merge_unbias(m.into(), e1.into())
+	}
+}
+
+impl std::ops::Mul for F8 {
+	type Output = Self;
+
+	fn mul(self, rhs: Self) -> Self::Output {
+		let (m1, e1) = self.split_unbias();
+		let (m2, e2) = rhs.split_unbias();
+		if self.0 == 0 || rhs.0 == 0 {
+			return Self(0);
+		}
+		Self::merge_unbias(m1 as u32 * m2 as u32, e1 as i32 + e2 as i32)
 	}
 }
 
@@ -63,4 +76,16 @@ mod tests {
 		assert_eq!(F8::from(7) - F8::from(3), F8::from(4));
 		assert_eq!(F8::from(12) - F8::from(1), F8::from(11));
 	}
+
+	#[test]
+	fn test_mul() {
+		assert_eq!(F8::from(3) * F8::from(0), F8::from(0));
+		assert_eq!(F8::from(0) * F8::from(7), F8::from(0));
+		assert_eq!(F8::from(3) * F8::from(5), F8::from(15));
+		assert_eq!(F8::from(5) * F8::from(3), F8::from(15));
+		assert_eq!(F8::from(3) * F8::from(6), F8::from(18));
+		assert_eq!(F8::from(3) * F8::from(7), F8::from(21));
+		assert_eq!(F8::from(12) * F8::from(1), F8::from(12));
+		assert_eq!(F8::from(12) * F8::from(2), F8::from(24));
+	}
 }