diff --git a/library/stdarch/crates/core_arch/src/aarch64/neon/generated.rs b/library/stdarch/crates/core_arch/src/aarch64/neon/generated.rs
index d09171a75352..63ffa6765cec 100644
--- a/library/stdarch/crates/core_arch/src/aarch64/neon/generated.rs
+++ b/library/stdarch/crates/core_arch/src/aarch64/neon/generated.rs
@@ -12591,7 +12591,7 @@ pub fn vmul_lane_f64<const LANE: i32>(a: float64x1_t, b: float64x1_t) -> float64
 #[cfg(not(target_arch = "arm64ec"))]
 pub fn vmul_laneq_f16<const LANE: i32>(a: float16x4_t, b: float16x8_t) -> float16x4_t {
     static_assert_uimm_bits!(LANE, 3);
-    unsafe { simd_mul(a, simd_shuffle!(b, b, [LANE as u32; 4])) }
+    unsafe { simd_mul(a, vdup_laneq_f16::<LANE>(b)) }
 }
 #[doc = "Floating-point multiply"]
 #[doc = "[Arm's documentation](https://developer.arm.com/architectures/instruction-sets/intrinsics/vmulq_laneq_f16)"]
@@ -12603,7 +12603,7 @@ pub fn vmul_laneq_f16<const LANE: i32>(a: float16x4_t, b: float16x8_t) -> float1
 #[cfg(not(target_arch = "arm64ec"))]
 pub fn vmulq_laneq_f16<const LANE: i32>(a: float16x8_t, b: float16x8_t) -> float16x8_t {
     static_assert_uimm_bits!(LANE, 3);
-    unsafe { simd_mul(a, simd_shuffle!(b, b, [LANE as u32; 8])) }
+    unsafe { simd_mul(a, vdupq_laneq_f16::<LANE>(b)) }
 }
 #[doc = "Floating-point multiply"]
 #[doc = "[Arm's documentation](https://developer.arm.com/architectures/instruction-sets/intrinsics/vmul_laneq_f64)"]
@@ -12917,7 +12917,7 @@ pub fn vmull_p64(a: p64, b: p64) -> p128 {
 #[stable(feature = "neon_intrinsics", since = "1.59.0")]
 pub fn vmulq_lane_f64<const LANE: i32>(a: float64x2_t, b: float64x1_t) -> float64x2_t {
     static_assert!(LANE == 0);
-    unsafe { simd_mul(a, simd_shuffle!(b, b, [LANE as u32; 2])) }
+    unsafe { simd_mul(a, vdupq_lane_f64::<LANE>(b)) }
 }
 #[doc = "Floating-point multiply"]
 #[doc = "[Arm's documentation](https://developer.arm.com/architectures/instruction-sets/intrinsics/vmulq_laneq_f64)"]
@@ -12928,7 +12928,7 @@ pub fn vmulq_lane_f64<const LANE: i32>(a: float64x2_t, b: float64x1_t) -> float6
 #[stable(feature = "neon_intrinsics", since = "1.59.0")]
 pub fn vmulq_laneq_f64<const LANE: i32>(a: float64x2_t, b: float64x2_t) -> float64x2_t {
     static_assert_uimm_bits!(LANE, 1);
-    unsafe { simd_mul(a, simd_shuffle!(b, b, [LANE as u32; 2])) }
+    unsafe { simd_mul(a, vdupq_laneq_f64::<LANE>(b)) }
 }
 #[doc = "Floating-point multiply"]
 #[doc = "[Arm's documentation](https://developer.arm.com/architectures/instruction-sets/intrinsics/vmuls_lane_f32)"]
diff --git a/library/stdarch/crates/core_arch/src/arm_shared/neon/generated.rs b/library/stdarch/crates/core_arch/src/arm_shared/neon/generated.rs
index 6c0dd2af94e0..2e3a24ee957e 100644
--- a/library/stdarch/crates/core_arch/src/arm_shared/neon/generated.rs
+++ b/library/stdarch/crates/core_arch/src/arm_shared/neon/generated.rs
@@ -29763,12 +29763,7 @@ pub fn vmulq_f32(a: float32x4_t, b: float32x4_t) -> float32x4_t {
 #[cfg(not(target_arch = "arm64ec"))]
 pub fn vmul_lane_f16<const LANE: i32>(a: float16x4_t, v: float16x4_t) -> float16x4_t {
     static_assert_uimm_bits!(LANE, 2);
-    unsafe {
-        simd_mul(
-            a,
-            simd_shuffle!(v, v, [LANE as u32, LANE as u32, LANE as u32, LANE as u32]),
-        )
-    }
+    unsafe { simd_mul(a, vdup_lane_f16::<LANE>(v)) }
 }
 #[doc = "Multiply"]
 #[doc = "[Arm's documentation](https://developer.arm.com/architectures/instruction-sets/intrinsics/vmulq_lane_f16)"]
@@ -29792,25 +29787,7 @@ pub fn vmul_lane_f16<const LANE: i32>(a: float16x4_t, v: float16x4_t) -> float16
 #[cfg(not(target_arch = "arm64ec"))]
 pub fn vmulq_lane_f16<const LANE: i32>(a: float16x8_t, v: float16x4_t) -> float16x8_t {
     static_assert_uimm_bits!(LANE, 2);
-    unsafe {
-        simd_mul(
-            a,
-            simd_shuffle!(
-                v,
-                v,
-                [
-                    LANE as u32,
-                    LANE as u32,
-                    LANE as u32,
-                    LANE as u32,
-                    LANE as u32,
-                    LANE as u32,
-                    LANE as u32,
-                    LANE as u32
-                ]
-            ),
-        )
-    }
+    unsafe { simd_mul(a, vdupq_lane_f16::<LANE>(v)) }
 }
 #[doc = "Floating-point multiply"]
 #[doc = "[Arm's documentation](https://developer.arm.com/architectures/instruction-sets/intrinsics/vmul_lane_f32)"]
@@ -29833,7 +29810,7 @@ pub fn vmulq_lane_f16<const LANE: i32>(a: float16x8_t, v: float16x4_t) -> float1
 )]
 pub fn vmul_lane_f32<const LANE: i32>(a: float32x2_t, b: float32x2_t) -> float32x2_t {
     static_assert_uimm_bits!(LANE, 1);
-    unsafe { simd_mul(a, simd_shuffle!(b, b, [LANE as u32, LANE as u32])) }
+    unsafe { simd_mul(a, vdup_lane_f32::<LANE>(b)) }
 }
 #[doc = "Floating-point multiply"]
 #[doc = "[Arm's documentation](https://developer.arm.com/architectures/instruction-sets/intrinsics/vmul_laneq_f32)"]
@@ -29856,7 +29833,7 @@ pub fn vmul_lane_f32<const LANE: i32>(a: float32x2_t, b: float32x2_t) -> float32
 )]
 pub fn vmul_laneq_f32<const LANE: i32>(a: float32x2_t, b: float32x4_t) -> float32x2_t {
     static_assert_uimm_bits!(LANE, 2);
-    unsafe { simd_mul(a, simd_shuffle!(b, b, [LANE as u32, LANE as u32])) }
+    unsafe { simd_mul(a, vdup_laneq_f32::<LANE>(b)) }
 }
 #[doc = "Floating-point multiply"]
 #[doc = "[Arm's documentation](https://developer.arm.com/architectures/instruction-sets/intrinsics/vmulq_lane_f32)"]
@@ -29879,12 +29856,7 @@ pub fn vmul_laneq_f32<const LANE: i32>(a: float32x2_t, b: float32x4_t) -> float3
 )]
 pub fn vmulq_lane_f32<const LANE: i32>(a: float32x4_t, b: float32x2_t) -> float32x4_t {
     static_assert_uimm_bits!(LANE, 1);
-    unsafe {
-        simd_mul(
-            a,
-            simd_shuffle!(b, b, [LANE as u32, LANE as u32, LANE as u32, LANE as u32]),
-        )
-    }
+    unsafe { simd_mul(a, vdupq_lane_f32::<LANE>(b)) }
 }
 #[doc = "Floating-point multiply"]
 #[doc = "[Arm's documentation](https://developer.arm.com/architectures/instruction-sets/intrinsics/vmulq_laneq_f32)"]
@@ -29907,12 +29879,7 @@ pub fn vmulq_lane_f32<const LANE: i32>(a: float32x4_t, b: float32x2_t) -> float3
 )]
 pub fn vmulq_laneq_f32<const LANE: i32>(a: float32x4_t, b: float32x4_t) -> float32x4_t {
     static_assert_uimm_bits!(LANE, 2);
-    unsafe {
-        simd_mul(
-            a,
-            simd_shuffle!(b, b, [LANE as u32, LANE as u32, LANE as u32, LANE as u32]),
-        )
-    }
+    unsafe { simd_mul(a, vdupq_laneq_f32::<LANE>(b)) }
 }
 #[doc = "Multiply"]
 #[doc = "[Arm's documentation](https://developer.arm.com/architectures/instruction-sets/intrinsics/vmul_lane_s16)"]
@@ -29935,12 +29902,7 @@ pub fn vmulq_laneq_f32<const LANE: i32>(a: float32x4_t, b: float32x4_t) -> float
 )]
 pub fn vmul_lane_s16<const LANE: i32>(a: int16x4_t, b: int16x4_t) -> int16x4_t {
     static_assert_uimm_bits!(LANE, 2);
-    unsafe {
-        simd_mul(
-            a,
-            simd_shuffle!(b, b, [LANE as u32, LANE as u32, LANE as u32, LANE as u32]),
-        )
-    }
+    unsafe { simd_mul(a, vdup_lane_s16::<LANE>(b)) }
 }
 #[doc = "Multiply"]
 #[doc = "[Arm's documentation](https://developer.arm.com/architectures/instruction-sets/intrinsics/vmulq_lane_s16)"]
@@ -29963,25 +29925,7 @@ pub fn vmul_lane_s16<const LANE: i32>(a: int16x4_t, b: int16x4_t) -> int16x4_t {
 )]
 pub fn vmulq_lane_s16<const LANE: i32>(a: int16x8_t, b: int16x4_t) -> int16x8_t {
     static_assert_uimm_bits!(LANE, 2);
-    unsafe {
-        simd_mul(
-            a,
-            simd_shuffle!(
-                b,
-                b,
-                [
-                    LANE as u32,
-                    LANE as u32,
-                    LANE as u32,
-                    LANE as u32,
-                    LANE as u32,
-                    LANE as u32,
-                    LANE as u32,
-                    LANE as u32
-                ]
-            ),
-        )
-    }
+    unsafe { simd_mul(a, vdupq_lane_s16::<LANE>(b)) }
 }
 #[doc = "Multiply"]
 #[doc = "[Arm's documentation](https://developer.arm.com/architectures/instruction-sets/intrinsics/vmul_lane_s32)"]
@@ -30004,7 +29948,7 @@ pub fn vmulq_lane_s16<const LANE: i32>(a: int16x8_t, b: int16x4_t) -> int16x8_t
 )]
 pub fn vmul_lane_s32<const LANE: i32>(a: int32x2_t, b: int32x2_t) -> int32x2_t {
     static_assert_uimm_bits!(LANE, 1);
-    unsafe { simd_mul(a, simd_shuffle!(b, b, [LANE as u32, LANE as u32])) }
+    unsafe { simd_mul(a, vdup_lane_s32::<LANE>(b)) }
 }
 #[doc = "Multiply"]
 #[doc = "[Arm's documentation](https://developer.arm.com/architectures/instruction-sets/intrinsics/vmulq_lane_s32)"]
@@ -30027,12 +29971,7 @@ pub fn vmul_lane_s32<const LANE: i32>(a: int32x2_t, b: int32x2_t) -> int32x2_t {
 )]
 pub fn vmulq_lane_s32<const LANE: i32>(a: int32x4_t, b: int32x2_t) -> int32x4_t {
     static_assert_uimm_bits!(LANE, 1);
-    unsafe {
-        simd_mul(
-            a,
-            simd_shuffle!(b, b, [LANE as u32, LANE as u32, LANE as u32, LANE as u32]),
-        )
-    }
+    unsafe { simd_mul(a, vdupq_lane_s32::<LANE>(b)) }
 }
 #[doc = "Multiply"]
 #[doc = "[Arm's documentation](https://developer.arm.com/architectures/instruction-sets/intrinsics/vmul_lane_u16)"]
@@ -30055,12 +29994,7 @@ pub fn vmulq_lane_s32<const LANE: i32>(a: int32x4_t, b: int32x2_t) -> int32x4_t
 )]
 pub fn vmul_lane_u16<const LANE: i32>(a: uint16x4_t, b: uint16x4_t) -> uint16x4_t {
     static_assert_uimm_bits!(LANE, 2);
-    unsafe {
-        simd_mul(
-            a,
-            simd_shuffle!(b, b, [LANE as u32, LANE as u32, LANE as u32, LANE as u32]),
-        )
-    }
+    unsafe { simd_mul(a, vdup_lane_u16::<LANE>(b)) }
 }
 #[doc = "Multiply"]
 #[doc = "[Arm's documentation](https://developer.arm.com/architectures/instruction-sets/intrinsics/vmulq_lane_u16)"]
@@ -30083,25 +30017,7 @@ pub fn vmul_lane_u16<const LANE: i32>(a: uint16x4_t, b: uint16x4_t) -> uint16x4_
 )]
 pub fn vmulq_lane_u16<const LANE: i32>(a: uint16x8_t, b: uint16x4_t) -> uint16x8_t {
     static_assert_uimm_bits!(LANE, 2);
-    unsafe {
-        simd_mul(
-            a,
-            simd_shuffle!(
-                b,
-                b,
-                [
-                    LANE as u32,
-                    LANE as u32,
-                    LANE as u32,
-                    LANE as u32,
-                    LANE as u32,
-                    LANE as u32,
-                    LANE as u32,
-                    LANE as u32
-                ]
-            ),
-        )
-    }
+    unsafe { simd_mul(a, vdupq_lane_u16::<LANE>(b)) }
 }
 #[doc = "Multiply"]
 #[doc = "[Arm's documentation](https://developer.arm.com/architectures/instruction-sets/intrinsics/vmul_lane_u32)"]
@@ -30124,7 +30040,7 @@ pub fn vmulq_lane_u16<const LANE: i32>(a: uint16x8_t, b: uint16x4_t) -> uint16x8
 )]
 pub fn vmul_lane_u32<const LANE: i32>(a: uint32x2_t, b: uint32x2_t) -> uint32x2_t {
     static_assert_uimm_bits!(LANE, 1);
-    unsafe { simd_mul(a, simd_shuffle!(b, b, [LANE as u32, LANE as u32])) }
+    unsafe { simd_mul(a, vdup_lane_u32::<LANE>(b)) }
 }
 #[doc = "Multiply"]
 #[doc = "[Arm's documentation](https://developer.arm.com/architectures/instruction-sets/intrinsics/vmulq_lane_u32)"]
@@ -30147,12 +30063,7 @@ pub fn vmul_lane_u32<const LANE: i32>(a: uint32x2_t, b: uint32x2_t) -> uint32x2_
 )]
 pub fn vmulq_lane_u32<const LANE: i32>(a: uint32x4_t, b: uint32x2_t) -> uint32x4_t {
     static_assert_uimm_bits!(LANE, 1);
-    unsafe {
-        simd_mul(
-            a,
-            simd_shuffle!(b, b, [LANE as u32, LANE as u32, LANE as u32, LANE as u32]),
-        )
-    }
+    unsafe { simd_mul(a, vdupq_lane_u32::<LANE>(b)) }
 }
 #[doc = "Multiply"]
 #[doc = "[Arm's documentation](https://developer.arm.com/architectures/instruction-sets/intrinsics/vmul_laneq_s16)"]
@@ -30175,12 +30086,7 @@ pub fn vmulq_lane_u32<const LANE: i32>(a: uint32x4_t, b: uint32x2_t) -> uint32x4
 )]
 pub fn vmul_laneq_s16<const LANE: i32>(a: int16x4_t, b: int16x8_t) -> int16x4_t {
     static_assert_uimm_bits!(LANE, 3);
-    unsafe {
-        simd_mul(
-            a,
-            simd_shuffle!(b, b, [LANE as u32, LANE as u32, LANE as u32, LANE as u32]),
-        )
-    }
+    unsafe { simd_mul(a, vdup_laneq_s16::<LANE>(b)) }
 }
 #[doc = "Multiply"]
 #[doc = "[Arm's documentation](https://developer.arm.com/architectures/instruction-sets/intrinsics/vmulq_laneq_s16)"]
@@ -30203,25 +30109,7 @@ pub fn vmul_laneq_s16<const LANE: i32>(a: int16x4_t, b: int16x8_t) -> int16x4_t
 )]
 pub fn vmulq_laneq_s16<const LANE: i32>(a: int16x8_t, b: int16x8_t) -> int16x8_t {
     static_assert_uimm_bits!(LANE, 3);
-    unsafe {
-        simd_mul(
-            a,
-            simd_shuffle!(
-                b,
-                b,
-                [
-                    LANE as u32,
-                    LANE as u32,
-                    LANE as u32,
-                    LANE as u32,
-                    LANE as u32,
-                    LANE as u32,
-                    LANE as u32,
-                    LANE as u32
-                ]
-            ),
-        )
-    }
+    unsafe { simd_mul(a, vdupq_laneq_s16::<LANE>(b)) }
 }
 #[doc = "Multiply"]
 #[doc = "[Arm's documentation](https://developer.arm.com/architectures/instruction-sets/intrinsics/vmul_laneq_s32)"]
@@ -30244,7 +30132,7 @@ pub fn vmulq_laneq_s16<const LANE: i32>(a: int16x8_t, b: int16x8_t) -> int16x8_t
 )]
 pub fn vmul_laneq_s32<const LANE: i32>(a: int32x2_t, b: int32x4_t) -> int32x2_t {
     static_assert_uimm_bits!(LANE, 2);
-    unsafe { simd_mul(a, simd_shuffle!(b, b, [LANE as u32, LANE as u32])) }
+    unsafe { simd_mul(a, vdup_laneq_s32::<LANE>(b)) }
 }
 #[doc = "Multiply"]
 #[doc = "[Arm's documentation](https://developer.arm.com/architectures/instruction-sets/intrinsics/vmulq_laneq_s32)"]
@@ -30267,12 +30155,7 @@ pub fn vmul_laneq_s32<const LANE: i32>(a: int32x2_t, b: int32x4_t) -> int32x2_t
 )]
 pub fn vmulq_laneq_s32<const LANE: i32>(a: int32x4_t, b: int32x4_t) -> int32x4_t {
     static_assert_uimm_bits!(LANE, 2);
-    unsafe {
-        simd_mul(
-            a,
-            simd_shuffle!(b, b, [LANE as u32, LANE as u32, LANE as u32, LANE as u32]),
-        )
-    }
+    unsafe { simd_mul(a, vdupq_laneq_s32::<LANE>(b)) }
 }
 #[doc = "Multiply"]
 #[doc = "[Arm's documentation](https://developer.arm.com/architectures/instruction-sets/intrinsics/vmul_laneq_u16)"]
@@ -30295,12 +30178,7 @@ pub fn vmulq_laneq_s32<const LANE: i32>(a: int32x4_t, b: int32x4_t) -> int32x4_t
 )]
 pub fn vmul_laneq_u16<const LANE: i32>(a: uint16x4_t, b: uint16x8_t) -> uint16x4_t {
     static_assert_uimm_bits!(LANE, 3);
-    unsafe {
-        simd_mul(
-            a,
-            simd_shuffle!(b, b, [LANE as u32, LANE as u32, LANE as u32, LANE as u32]),
-        )
-    }
+    unsafe { simd_mul(a, vdup_laneq_u16::<LANE>(b)) }
 }
 #[doc = "Multiply"]
 #[doc = "[Arm's documentation](https://developer.arm.com/architectures/instruction-sets/intrinsics/vmulq_laneq_u16)"]
@@ -30323,25 +30201,7 @@ pub fn vmul_laneq_u16<const LANE: i32>(a: uint16x4_t, b: uint16x8_t) -> uint16x4
 )]
 pub fn vmulq_laneq_u16<const LANE: i32>(a: uint16x8_t, b: uint16x8_t) -> uint16x8_t {
     static_assert_uimm_bits!(LANE, 3);
-    unsafe {
-        simd_mul(
-            a,
-            simd_shuffle!(
-                b,
-                b,
-                [
-                    LANE as u32,
-                    LANE as u32,
-                    LANE as u32,
-                    LANE as u32,
-                    LANE as u32,
-                    LANE as u32,
-                    LANE as u32,
-                    LANE as u32
-                ]
-            ),
-        )
-    }
+    unsafe { simd_mul(a, vdupq_laneq_u16::<LANE>(b)) }
 }
 #[doc = "Multiply"]
 #[doc = "[Arm's documentation](https://developer.arm.com/architectures/instruction-sets/intrinsics/vmul_laneq_u32)"]
@@ -30364,7 +30224,7 @@ pub fn vmulq_laneq_u16<const LANE: i32>(a: uint16x8_t, b: uint16x8_t) -> uint16x
 )]
 pub fn vmul_laneq_u32<const LANE: i32>(a: uint32x2_t, b: uint32x4_t) -> uint32x2_t {
     static_assert_uimm_bits!(LANE, 2);
-    unsafe { simd_mul(a, simd_shuffle!(b, b, [LANE as u32, LANE as u32])) }
+    unsafe { simd_mul(a, vdup_laneq_u32::<LANE>(b)) }
 }
 #[doc = "Multiply"]
 #[doc = "[Arm's documentation](https://developer.arm.com/architectures/instruction-sets/intrinsics/vmulq_laneq_u32)"]
@@ -30387,12 +30247,7 @@ pub fn vmul_laneq_u32<const LANE: i32>(a: uint32x2_t, b: uint32x4_t) -> uint32x2
 )]
 pub fn vmulq_laneq_u32<const LANE: i32>(a: uint32x4_t, b: uint32x4_t) -> uint32x4_t {
     static_assert_uimm_bits!(LANE, 2);
-    unsafe {
-        simd_mul(
-            a,
-            simd_shuffle!(b, b, [LANE as u32, LANE as u32, LANE as u32, LANE as u32]),
-        )
-    }
+    unsafe { simd_mul(a, vdupq_laneq_u32::<LANE>(b)) }
 }
 #[doc = "Vector multiply by scalar"]
 #[doc = "[Arm's documentation](https://developer.arm.com/architectures/instruction-sets/intrinsics/vmul_n_f16)"]
diff --git a/library/stdarch/crates/stdarch-gen-arm/spec/neon/aarch64.spec.yml b/library/stdarch/crates/stdarch-gen-arm/spec/neon/aarch64.spec.yml
index 912eff7ce4cc..7664bf3c10fa 100644
--- a/library/stdarch/crates/stdarch-gen-arm/spec/neon/aarch64.spec.yml
+++ b/library/stdarch/crates/stdarch-gen-arm/spec/neon/aarch64.spec.yml
@@ -11006,7 +11006,7 @@ intrinsics:
       - FnCall:
           - simd_mul
           - - a
-            - FnCall: ["simd_shuffle!", [b, b, '[LANE as u32; 2]']]
+            - FnCall: [vdupq_lane_f64, [b], [LANE]]
 
   - name: "vmuld_lane_f64"
     doc: "Floating-point multiply"
@@ -11063,7 +11063,7 @@ intrinsics:
       - FnCall:
           - simd_mul
           - - a
-            - FnCall: [simd_shuffle!, [b, b, '[LANE as u32; 2]']]
+            - FnCall: [vdupq_laneq_f64, [b], [LANE]]
 
 
   # vmulq_laneq_f16
@@ -11080,14 +11080,14 @@ intrinsics:
     static_defs: ['const LANE: i32']
     safety: safe
     types:
-      - [float16x4_t, float16x8_t, '_lane', "[LANE as u32; 4]"]
-      - [float16x8_t, float16x8_t, 'q_lane', "[LANE as u32; 8]"]
+      - [float16x4_t, float16x8_t, '_lane']
+      - [float16x8_t, float16x8_t, 'q_lane']
     compose:
       - FnCall: [static_assert_uimm_bits!, [LANE, '3']]
       - FnCall:
           - simd_mul
           - - a
-            - FnCall: [simd_shuffle!, [b, b, "{type[3]}"]]
+            - FnCall: ['vdup{neon_type[0].laneq_nox}', [b], [LANE]]
 
 
   - name: "vmul{type[1]}_{type[0]}"
diff --git a/library/stdarch/crates/stdarch-gen-arm/spec/neon/arm_shared.spec.yml b/library/stdarch/crates/stdarch-gen-arm/spec/neon/arm_shared.spec.yml
index f22f503c564c..d20c94a2f8cb 100644
--- a/library/stdarch/crates/stdarch-gen-arm/spec/neon/arm_shared.spec.yml
+++ b/library/stdarch/crates/stdarch-gen-arm/spec/neon/arm_shared.spec.yml
@@ -6300,20 +6300,20 @@ intrinsics:
     static_defs: ["const LANE: i32"]
     safety: safe
     types:
-      - [int16x4_t, int16x4_t, '2', '[LANE as u32, LANE as u32, LANE as u32, LANE as u32]']
-      - [int16x8_t, int16x4_t, '2', '[LANE as u32, LANE as u32, LANE as u32, LANE as u32, LANE as u32, LANE as u32, LANE as u32, LANE as u32]']
-      - [int32x2_t, int32x2_t, '1', '[LANE as u32, LANE as u32]']
-      - [int32x4_t, int32x2_t, '1', '[LANE as u32, LANE as u32, LANE as u32, LANE as u32]']
-      - [uint16x4_t, uint16x4_t, '2', '[LANE as u32, LANE as u32, LANE as u32, LANE as u32]']
-      - [uint16x8_t, uint16x4_t, '2', '[LANE as u32, LANE as u32, LANE as u32, LANE as u32, LANE as u32, LANE as u32, LANE as u32, LANE as u32]']
-      - [uint32x2_t, uint32x2_t, '1', '[LANE as u32, LANE as u32]']
-      - [uint32x4_t, uint32x2_t, '1', '[LANE as u32, LANE as u32, LANE as u32, LANE as u32]']
+      - [int16x4_t, int16x4_t, '2']
+      - [int16x8_t, int16x4_t, '2']
+      - [int32x2_t, int32x2_t, '1']
+      - [int32x4_t, int32x2_t, '1']
+      - [uint16x4_t, uint16x4_t, '2']
+      - [uint16x8_t, uint16x4_t, '2']
+      - [uint32x2_t, uint32x2_t, '1']
+      - [uint32x4_t, uint32x2_t, '1']
     compose:
       - FnCall: ["static_assert_uimm_bits!", [LANE, "{type[2]}"]]
       - FnCall:
           - simd_mul
           - - a
-            - FnCall: ["simd_shuffle!", [b, b, "{type[3]}"]]
+            - FnCall: ["vdup{neon_type[0].lane_nox}", [b], [LANE]]
 
 
   - name: "vmul{neon_type[0].lane_nox}"
@@ -6332,14 +6332,14 @@ intrinsics:
     static_defs: ["const LANE: i32"]
     safety: safe
     types:
-      - [float16x4_t, float16x4_t, '2', '[LANE as u32, LANE as u32, LANE as u32, LANE as u32]']
-      - [float16x8_t, float16x4_t, '2', '[LANE as u32, LANE as u32, LANE as u32, LANE as u32, LANE as u32, LANE as u32, LANE as u32, LANE as u32]']
+      - [float16x4_t, float16x4_t, '2']
+      - [float16x8_t, float16x4_t, '2']
     compose:
       - FnCall: ["static_assert_uimm_bits!", [LANE, "{type[2]}"]]
       - FnCall:
           - simd_mul
           - - a
-            - FnCall: ["simd_shuffle!", [v, v, "{type[3]}"]]
+            - FnCall: ["vdup{neon_type[0].lane_nox}", [v], [LANE]]
 
 
   - name: "vmul{neon_type[0].laneq_nox}"
@@ -6356,20 +6356,20 @@ intrinsics:
     static_defs: ["const LANE: i32"]
     safety: safe
     types:
-      - [int16x4_t, int16x8_t, '3', '[LANE as u32, LANE as u32, LANE as u32, LANE as u32]']
-      - [int16x8_t, int16x8_t, '3', '[LANE as u32, LANE as u32, LANE as u32, LANE as u32, LANE as u32, LANE as u32, LANE as u32, LANE as u32]']
-      - [int32x2_t, int32x4_t, '2', '[LANE as u32, LANE as u32]']
-      - [int32x4_t, int32x4_t, '2', '[LANE as u32, LANE as u32, LANE as u32, LANE as u32]']
-      - [uint16x4_t, uint16x8_t, '3', '[LANE as u32, LANE as u32, LANE as u32, LANE as u32]']
-      - [uint16x8_t, uint16x8_t, '3', '[LANE as u32, LANE as u32, LANE as u32, LANE as u32, LANE as u32, LANE as u32, LANE as u32, LANE as u32]']
-      - [uint32x2_t, uint32x4_t, '2', '[LANE as u32, LANE as u32]']
-      - [uint32x4_t, uint32x4_t, '2', '[LANE as u32, LANE as u32, LANE as u32, LANE as u32]']
+      - [int16x4_t, int16x8_t, '3']
+      - [int16x8_t, int16x8_t, '3']
+      - [int32x2_t, int32x4_t, '2']
+      - [int32x4_t, int32x4_t, '2']
+      - [uint16x4_t, uint16x8_t, '3']
+      - [uint16x8_t, uint16x8_t, '3']
+      - [uint32x2_t, uint32x4_t, '2']
+      - [uint32x4_t, uint32x4_t, '2']
     compose:
       - FnCall: ["static_assert_uimm_bits!", [LANE, "{type[2]}"]]
       - FnCall:
           - simd_mul
           - - a
-            - FnCall: ["simd_shuffle!", [b, b, "{type[3]}"]]
+            - FnCall: ["vdup{neon_type[0].laneq_nox}", [b], [LANE]]
 
   - name: "vmull{neon_type[1].no}"
     doc: Signed multiply long
@@ -11204,16 +11204,16 @@ intrinsics:
     static_defs: ['const LANE: i32']
     safety: safe
     types:
-      - [float32x2_t, float32x2_t, '_lane_f32', '1', '[LANE as u32, LANE as u32]']
-      - [float32x2_t, float32x4_t, '_laneq_f32', '2', '[LANE as u32, LANE as u32]']
-      - [float32x4_t, float32x2_t, 'q_lane_f32', '1', '[LANE as u32, LANE as u32, LANE as u32, LANE as u32]']
-      - [float32x4_t, float32x4_t, 'q_laneq_f32', '2', '[LANE as u32, LANE as u32, LANE as u32, LANE as u32]']
+      - [float32x2_t, float32x2_t, '_lane_f32', '1']
+      - [float32x2_t, float32x4_t, '_laneq_f32', '2']
+      - [float32x4_t, float32x2_t, 'q_lane_f32', '1']
+      - [float32x4_t, float32x4_t, 'q_laneq_f32', '2']
     compose:
       - FnCall: [static_assert_uimm_bits!, [LANE, '{type[3]}']]
       - FnCall:
           - simd_mul
           - - a
-            - FnCall: [simd_shuffle!, [b, b, "{type[4]}"]]
+            - FnCall: ['vdup{type[2]}', [b], [LANE]]
 
   - name: "vqrdmulh{type[0]}"
     doc: "Vector rounding saturating doubling multiply high by scalar"