Merge branch 'avx-512-cmp' into scatter-512

Author: Daniel Smith

crates/core_arch/src/mod.rs

@@ -57,6 +57,13 @@ pub mod arch {
 
     /// Platform-specific intrinsics for the `wasm32` platform.
     ///
+
+    /// # Availability
+    ///
+    /// Note that intrinsics gated by `target_feature = "atomics"` or `target_feature = "simd128"`
+    /// are only available **when the standard library itself is compiled with the the respective
+    /// target feature**. This version of the standard library is not obtainable via `rustup`,
+    /// but rather will require the standard library to be compiled from source.
     /// See the [module documentation](../index.html) for more details.
     #[cfg(any(target_arch = "wasm32", dox))]
     #[doc(cfg(target_arch = "wasm32"))]

crates/core_arch/src/simd.rs

@@ -198,6 +198,18 @@ simd_ty!(i32x16[i32]:
          | x0, x1, x2, x3, x4, x5, x6, x7,
          x8, x9, x10, x11, x12, x13, x14, x15);
 
+simd_ty!(u32x16[u32]:
+         u32, u32, u32, u32, u32, u32, u32, u32,
+         u32, u32, u32, u32, u32, u32, u32, u32
+         | x0, x1, x2, x3, x4, x5, x6, x7,
+         x8, x9, x10, x11, x12, x13, x14, x15);
+
+simd_ty!(f32x16[f32]:
+         f32, f32, f32, f32, f32, f32, f32, f32,
+         f32, f32, f32, f32, f32, f32, f32, f32
+         | x0, x1, x2, x3, x4, x5, x6, x7,
+         x8, x9, x10, x11, x12, x13, x14, x15);
+
 simd_ty!(i64x8[i64]:
          i64, i64, i64, i64, i64, i64, i64, i64
          | x0, x1, x2, x3, x4, x5, x6, x7);

crates/core_arch/src/x86/avx2.rs

@@ -3743,9 +3743,9 @@ pub unsafe fn _mm256_xor_si256(a: __m256i, b: __m256i) -> __m256i {
 // This intrinsic has no corresponding instruction.
 #[rustc_args_required_const(1)]
 #[stable(feature = "simd_x86", since = "1.27.0")]
-pub unsafe fn _mm256_extract_epi8(a: __m256i, imm8: i32) -> i8 {
+pub unsafe fn _mm256_extract_epi8(a: __m256i, imm8: i32) -> i32 {
     let imm8 = (imm8 & 31) as u32;
-    simd_extract(a.as_i8x32(), imm8)
+    simd_extract::<_, u8>(a.as_u8x32(), imm8) as i32
 }
 
 /// Extracts a 16-bit integer from `a`, selected with `imm8`. Returns a 32-bit
@@ -3759,9 +3759,9 @@ pub unsafe fn _mm256_extract_epi8(a: __m256i, imm8: i32) -> i8 {
 // This intrinsic has no corresponding instruction.
 #[rustc_args_required_const(1)]
 #[stable(feature = "simd_x86", since = "1.27.0")]
-pub unsafe fn _mm256_extract_epi16(a: __m256i, imm8: i32) -> i16 {
+pub unsafe fn _mm256_extract_epi16(a: __m256i, imm8: i32) -> i32 {
     let imm8 = (imm8 & 15) as u32;
-    simd_extract(a.as_i16x16(), imm8)
+    simd_extract::<_, u16>(a.as_u16x16(), imm8) as i32
 }
 
 /// Extracts a 32-bit integer from `a`, selected with `imm8`.
@@ -6115,7 +6115,7 @@ mod tests {
         );
         let r1 = _mm256_extract_epi8(a, 0);
         let r2 = _mm256_extract_epi8(a, 35);
-        assert_eq!(r1, -1);
+        assert_eq!(r1, 0xFF);
         assert_eq!(r2, 3);
     }
 
@@ -6128,7 +6128,7 @@ mod tests {
         );
         let r1 = _mm256_extract_epi16(a, 0);
         let r2 = _mm256_extract_epi16(a, 19);
-        assert_eq!(r1, -1);
+        assert_eq!(r1, 0xFFFF);
         assert_eq!(r2, 3);
     }