Quasi-Minimum and Quasi-Maximum instructions

Author: Maratyszcza

proposals/simd/BinarySIMD.md

@@ -189,3 +189,7 @@ The `v8x16.shuffle` instruction has 16 bytes after `simdop`.
 | `i64x2.load32x2_s`         |    `0xd6`| m:memarg           |
 | `i64x2.load32x2_u`         |    `0xd7`| m:memarg           |
 | `v128.andnot`              |    `0xd8`| -                  |
+| `f32x4.qmin`               |    `0xd9`| -                  |
+| `f32x4.qmax`               |    `0xda`| -                  |
+| `f64x2.qmin`               |    `0xdb`| -                  |
+| `f64x2.qmax`               |    `0xdc`| -                  |

proposals/simd/SIMD.md

@@ -743,6 +743,18 @@ Lane-wise minimum value, propagating NaNs.
 
 Lane-wise maximum value, propagating NaNs.
 
+### Quasi-minimum
+* `f32x4.qmin(a: v128, b: v128) -> v128`
+* `f64x2.qmin(a: v128, b: v128) -> v128`
+
+Lane-wise minimum value, defined as `b < a ? b : a`.
+
+### Quasi-maximum
+* `f32x4.qmax(a: v128, b: v128) -> v128`
+* `f64x2.qmax(a: v128, b: v128) -> v128`
+
+Lane-wise maximum value, defined as `a < b ? b : a`.
+
 ## Floating-point arithmetic
 
 The floating-point arithmetic operations are all lane-wise versions of the