@cpp-linter/clang-tidy-darwin-arm64 16.0.3 → 18.1.5

package/lib/clang/{16 → 18}/include/avx512vp2intersectintrin.h

@@ -28,8 +28,9 @@
 #ifndef _AVX512VP2INTERSECT_H
 #define _AVX512VP2INTERSECT_H
 
-#define __DEFAULT_FN_ATTRS \
-  __attribute__((__always_inline__, __nodebug__,  __target__("avx512vp2intersect"), \
+#define __DEFAULT_FN_ATTRS                                                     \
+  __attribute__((__always_inline__, __nodebug__,                               \
+                 __target__("avx512vp2intersect,evex512"),                     \
                  __min_vector_width__(512)))
 
 /// Store, in an even/odd pair of mask registers, the indicators of the

package/lib/clang/{16 → 18}/include/avx512vpopcntdqintrin.h

@@ -17,7 +17,9 @@
 
 /* Define the default attributes for the functions in this file. */
 #define __DEFAULT_FN_ATTRS                                                     \
-  __attribute__((__always_inline__, __nodebug__, __target__("avx512vpopcntdq"), __min_vector_width__(512)))
+  __attribute__((__always_inline__, __nodebug__,                               \
+                 __target__("avx512vpopcntdq,evex512"),                        \
+                 __min_vector_width__(512)))
 
 static __inline__ __m512i __DEFAULT_FN_ATTRS _mm512_popcnt_epi64(__m512i __A) {
   return (__m512i)__builtin_ia32_vpopcntq_512((__v8di)__A);

package/lib/clang/{16 → 18}/include/avx512vpopcntdqvlintrin.h

@@ -17,9 +17,13 @@
 
 /* Define the default attributes for the functions in this file. */
 #define __DEFAULT_FN_ATTRS128                                                  \
-  __attribute__((__always_inline__, __nodebug__, __target__("avx512vpopcntdq,avx512vl"), __min_vector_width__(128)))
+  __attribute__((__always_inline__, __nodebug__,                               \
+                 __target__("avx512vpopcntdq,avx512vl,no-evex512"),            \
+                 __min_vector_width__(128)))
 #define __DEFAULT_FN_ATTRS256                                                  \
-  __attribute__((__always_inline__, __nodebug__, __target__("avx512vpopcntdq,avx512vl"), __min_vector_width__(256)))
+  __attribute__((__always_inline__, __nodebug__,                               \
+                 __target__("avx512vpopcntdq,avx512vl,no-evex512"),            \
+                 __min_vector_width__(256)))
 
 static __inline__ __m128i __DEFAULT_FN_ATTRS128
 _mm_popcnt_epi64(__m128i __A) {

package/lib/clang/{16 → 18}/include/avxintrin.h

@@ -50,8 +50,12 @@ typedef __bf16 __m256bh __attribute__((__vector_size__(32), __aligned__(32)));
 #endif
 
 /* Define the default attributes for the functions in this file. */
-#define __DEFAULT_FN_ATTRS __attribute__((__always_inline__, __nodebug__, __target__("avx"), __min_vector_width__(256)))
-#define __DEFAULT_FN_ATTRS128 __attribute__((__always_inline__, __nodebug__, __target__("avx"), __min_vector_width__(128)))
+#define __DEFAULT_FN_ATTRS                                                     \
+  __attribute__((__always_inline__, __nodebug__, __target__("avx,no-evex512"), \
+                 __min_vector_width__(256)))
+#define __DEFAULT_FN_ATTRS128                                                  \
+  __attribute__((__always_inline__, __nodebug__, __target__("avx,no-evex512"), \
+                 __min_vector_width__(128)))
 
 /* Arithmetic */
 /// Adds two 256-bit vectors of [4 x double].
@@ -3017,8 +3021,11 @@ _mm256_zeroupper(void)
 static __inline __m128 __DEFAULT_FN_ATTRS128
 _mm_broadcast_ss(float const *__a)
 {
-  float __f = *__a;
-  return __extension__ (__m128)(__v4sf){ __f, __f, __f, __f };
+  struct __mm_broadcast_ss_struct {
+    float __f;
+  } __attribute__((__packed__, __may_alias__));
+  float __f = ((const struct __mm_broadcast_ss_struct*)__a)->__f;
+  return __extension__ (__m128){ __f, __f, __f, __f };
 }
 
 /// Loads a scalar double-precision floating point value from the
@@ -3036,7 +3043,10 @@ _mm_broadcast_ss(float const *__a)
 static __inline __m256d __DEFAULT_FN_ATTRS
 _mm256_broadcast_sd(double const *__a)
 {
-  double __d = *__a;
+  struct __mm256_broadcast_sd_struct {
+    double __d;
+  } __attribute__((__packed__, __may_alias__));
+  double __d = ((const struct __mm256_broadcast_sd_struct*)__a)->__d;
   return __extension__ (__m256d)(__v4df){ __d, __d, __d, __d };
 }
 
@@ -3055,7 +3065,10 @@ _mm256_broadcast_sd(double const *__a)
 static __inline __m256 __DEFAULT_FN_ATTRS
 _mm256_broadcast_ss(float const *__a)
 {
-  float __f = *__a;
+  struct __mm256_broadcast_ss_struct {
+    float __f;
+  } __attribute__((__packed__, __may_alias__));
+  float __f = ((const struct __mm256_broadcast_ss_struct*)__a)->__f;
   return __extension__ (__m256)(__v8sf){ __f, __f, __f, __f, __f, __f, __f, __f };
 }
 
@@ -3554,7 +3567,7 @@ _mm_maskstore_ps(float *__p, __m128i __m, __m128 __a)
 /// \param __b
 ///    A 256-bit integer vector containing the values to be moved.
 static __inline void __DEFAULT_FN_ATTRS
-_mm256_stream_si256(__m256i *__a, __m256i __b)
+_mm256_stream_si256(void *__a, __m256i __b)
 {
   typedef __v4di __v4di_aligned __attribute__((aligned(32)));
   __builtin_nontemporal_store((__v4di_aligned)__b, (__v4di_aligned*)__a);
@@ -3574,7 +3587,7 @@ _mm256_stream_si256(__m256i *__a, __m256i __b)
 /// \param __b
 ///    A 256-bit vector of [4 x double] containing the values to be moved.
 static __inline void __DEFAULT_FN_ATTRS
-_mm256_stream_pd(double *__a, __m256d __b)
+_mm256_stream_pd(void *__a, __m256d __b)
 {
   typedef __v4df __v4df_aligned __attribute__((aligned(32)));
   __builtin_nontemporal_store((__v4df_aligned)__b, (__v4df_aligned*)__a);
@@ -3595,7 +3608,7 @@ _mm256_stream_pd(double *__a, __m256d __b)
 /// \param __a
 ///    A 256-bit vector of [8 x float] containing the values to be moved.
 static __inline void __DEFAULT_FN_ATTRS
-_mm256_stream_ps(float *__p, __m256 __a)
+_mm256_stream_ps(void *__p, __m256 __a)
 {
   typedef __v8sf __v8sf_aligned __attribute__((aligned(32)));
   __builtin_nontemporal_store((__v8sf_aligned)__a, (__v8sf_aligned*)__p);
@@ -4499,7 +4512,8 @@ _mm256_castsi256_si128(__m256i __a)
 static __inline __m256d __DEFAULT_FN_ATTRS
 _mm256_castpd128_pd256(__m128d __a)
 {
-  return __builtin_shufflevector((__v2df)__a, (__v2df)__a, 0, 1, -1, -1);
+  return __builtin_shufflevector(
+      (__v2df)__a, (__v2df)__builtin_nondeterministic_value(__a), 0, 1, 2, 3);
 }
 
 /// Constructs a 256-bit floating-point vector of [8 x float] from a
@@ -4520,7 +4534,9 @@ _mm256_castpd128_pd256(__m128d __a)
 static __inline __m256 __DEFAULT_FN_ATTRS
 _mm256_castps128_ps256(__m128 __a)
 {
-  return __builtin_shufflevector((__v4sf)__a, (__v4sf)__a, 0, 1, 2, 3, -1, -1, -1, -1);
+  return __builtin_shufflevector((__v4sf)__a,
+                                 (__v4sf)__builtin_nondeterministic_value(__a),
+                                 0, 1, 2, 3, 4, 5, 6, 7);
 }
 
 /// Constructs a 256-bit integer vector from a 128-bit integer vector.
@@ -4539,7 +4555,8 @@ _mm256_castps128_ps256(__m128 __a)
 static __inline __m256i __DEFAULT_FN_ATTRS
 _mm256_castsi128_si256(__m128i __a)
 {
-  return __builtin_shufflevector((__v2di)__a, (__v2di)__a, 0, 1, -1, -1);
+  return __builtin_shufflevector(
+      (__v2di)__a, (__v2di)__builtin_nondeterministic_value(__a), 0, 1, 2, 3);
 }
 
 /// Constructs a 256-bit floating-point vector of [4 x double] from a

package/lib/clang/18/include/avxvnniint16intrin.h

@@ -0,0 +1,473 @@
+/*===----------- avxvnniint16intrin.h - AVXVNNIINT16 intrinsics-------------===
+ *
+ * Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+ * See https://llvm.org/LICENSE.txt for license information.
+ * SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+ *
+ *===-----------------------------------------------------------------------===
+ */
+
+#ifndef __IMMINTRIN_H
+#error                                                                         \
+    "Never use <avxvnniint16intrin.h> directly; include <immintrin.h> instead."
+#endif // __IMMINTRIN_H
+
+#ifndef __AVXVNNIINT16INTRIN_H
+#define __AVXVNNIINT16INTRIN_H
+
+/* Define the default attributes for the functions in this file. */
+#define __DEFAULT_FN_ATTRS128                                                  \
+  __attribute__((__always_inline__, __nodebug__, __target__("avxvnniint16"),   \
+                 __min_vector_width__(128)))
+#define __DEFAULT_FN_ATTRS256                                                  \
+  __attribute__((__always_inline__, __nodebug__, __target__("avxvnniint16"),   \
+                 __min_vector_width__(256)))
+
+/// Multiply groups of 2 adjacent pairs of signed 16-bit integers in \a __A with
+///    corresponding unsigned 16-bit integers in \a __B, producing 2 intermediate
+///    signed 16-bit results. Sum these 2 results with the corresponding
+///    32-bit integer in \a __W, and store the packed 32-bit results in \a dst.
+///
+/// \headerfile <immintrin.h>
+///
+/// \code
+/// __m128i _mm_dpwsud_epi32(__m128i __W, __m128i __A, __m128i __B)
+/// \endcode
+///
+/// This intrinsic corresponds to the \c VPDPWSUD instruction.
+///
+/// \param __W
+///    A 128-bit vector of [4 x int].
+/// \param __A
+///    A 128-bit vector of [8 x short].
+/// \param __B
+///    A 128-bit vector of [8 x unsigned short].
+/// \returns
+///    A 128-bit vector of [4 x int].
+///
+/// \code{.operation}
+/// FOR j := 0 to 3
+/// 	tmp1.dword := SignExtend32(__A.word[2*j]) * ZeroExtend32(__B.word[2*j])
+/// 	tmp2.dword := SignExtend32(__A.word[2*j+1]) * ZeroExtend32(__B.word[2*j+1])
+/// 	dst.dword[j] := __W.dword[j] + tmp1 + tmp2
+/// ENDFOR
+/// dst[MAX:128] := 0
+/// \endcode
+static __inline__ __m128i __DEFAULT_FN_ATTRS128 _mm_dpwsud_epi32(__m128i __W,
+                                                                 __m128i __A,
+                                                                 __m128i __B) {
+  return (__m128i)__builtin_ia32_vpdpwsud128((__v4si)__W, (__v4si)__A,
+                                             (__v4si)__B);
+}
+
+/// Multiply groups of 2 adjacent pairs of signed 16-bit integers in \a __A with
+///    corresponding unsigned 16-bit integers in \a __B, producing 2 intermediate
+///    signed 16-bit results. Sum these 2 results with the corresponding
+///    32-bit integer in \a __W, and store the packed 32-bit results in \a dst.
+///
+/// \headerfile <immintrin.h>
+///
+/// \code
+/// __m256i _mm256_dpwsud_epi32(__m256i __W, __m256i __A, __m256i __B)
+/// \endcode
+///
+/// This intrinsic corresponds to the \c VPDPWSUD instruction.
+///
+/// \param __W
+///    A 256-bit vector of [8 x int].
+/// \param __A
+///    A 256-bit vector of [16 x short].
+/// \param __B
+///    A 256-bit vector of [16 x unsigned short].
+/// \returns
+///    A 256-bit vector of [8 x int].
+///
+/// \code{.operation}
+/// FOR j := 0 to 7
+/// 	tmp1.dword := SignExtend32(__A.word[2*j]) * ZeroExtend32(__B.word[2*j])
+/// 	tmp2.dword := SignExtend32(__A.word[2*j+1]) * ZeroExtend32(__B.word[2*j+1])
+/// 	dst.dword[j] := __W.dword[j] + tmp1 + tmp2
+/// ENDFOR
+/// dst[MAX:256] := 0
+/// \endcode
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
+_mm256_dpwsud_epi32(__m256i __W, __m256i __A, __m256i __B) {
+  return (__m256i)__builtin_ia32_vpdpwsud256((__v8si)__W, (__v8si)__A,
+                                             (__v8si)__B);
+}
+
+/// Multiply groups of 2 adjacent pairs of signed 16-bit integers in \a __A with
+///    corresponding unsigned 16-bit integers in \a __B, producing 2 intermediate
+///    signed 16-bit results. Sum these 2 results with the corresponding
+///    32-bit integer in \a __W with signed saturation, and store the packed
+///    32-bit results in \a dst.
+///
+/// \headerfile <immintrin.h>
+///
+/// \code
+/// __m128i _mm_dpwsuds_epi32(__m128i __W, __m128i __A, __m128i __B)
+/// \endcode
+///
+/// This intrinsic corresponds to the \c VPDPWSUDS instruction.
+///
+/// \param __W
+///    A 128-bit vector of [4 x int].
+/// \param __A
+///    A 128-bit vector of [8 x short].
+/// \param __B
+///    A 128-bit vector of [8 x unsigned short].
+/// \returns
+///    A 128-bit vector of [4 x int].
+///
+/// \code{.operation}
+/// FOR j := 0 to 3
+/// 	tmp1.dword := SignExtend32(__A.word[2*j]) * ZeroExtend32(__B.word[2*j])
+/// 	tmp2.dword := SignExtend32(__A.word[2*j+1]) * ZeroExtend32(__B.word[2*j+1])
+/// 	dst.dword[j] := SIGNED_DWORD_SATURATE(__W.dword[j] + tmp1 + tmp2)
+/// ENDFOR
+/// dst[MAX:128] := 0
+/// \endcode
+static __inline__ __m128i __DEFAULT_FN_ATTRS128 _mm_dpwsuds_epi32(__m128i __W,
+                                                                  __m128i __A,
+                                                                  __m128i __B) {
+  return (__m128i)__builtin_ia32_vpdpwsuds128((__v4si)__W, (__v4si)__A,
+                                              (__v4si)__B);
+}
+
+/// Multiply groups of 2 adjacent pairs of signed 16-bit integers in \a __A with
+///    corresponding unsigned 16-bit integers in \a __B, producing 2 intermediate
+///    signed 16-bit results. Sum these 2 results with the corresponding
+///    32-bit integer in \a __W with signed saturation, and store the packed
+///    32-bit results in \a dst.
+///
+/// \headerfile <immintrin.h>
+///
+/// \code
+/// __m256i _mm256_dpwsuds_epi32(__m256i __W, __m256i __A, __m256i __B)
+/// \endcode
+///
+/// This intrinsic corresponds to the \c VPDPWSUDS instruction.
+///
+/// \param __W
+///    A 256-bit vector of [8 x int].
+/// \param __A
+///    A 256-bit vector of [16 x short].
+/// \param __B
+///    A 256-bit vector of [16 x unsigned short].
+/// \returns
+///    A 256-bit vector of [8 x int].
+///
+/// \code{.operation}
+/// FOR j := 0 to 7
+/// 	tmp1.dword := SignExtend32(__A.word[2*j]) * ZeroExtend32(__B.word[2*j])
+/// 	tmp2.dword := SignExtend32(__A.word[2*j+1]) * ZeroExtend32(__B.word[2*j+1])
+/// 	dst.dword[j] := SIGNED_DWORD_SATURATE(__W.dword[j] + tmp1 + tmp2)
+/// ENDFOR
+/// dst[MAX:256] := 0
+/// \endcode
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
+_mm256_dpwsuds_epi32(__m256i __W, __m256i __A, __m256i __B) {
+  return (__m256i)__builtin_ia32_vpdpwsuds256((__v8si)__W, (__v8si)__A,
+                                              (__v8si)__B);
+}
+
+/// Multiply groups of 2 adjacent pairs of unsigned 16-bit integers in \a __A with
+///    corresponding signed 16-bit integers in \a __B, producing 2 intermediate
+///    signed 16-bit results. Sum these 2 results with the corresponding
+///    32-bit integer in \a __W, and store the packed 32-bit results in \a dst.
+///
+/// \headerfile <immintrin.h>
+///
+/// \code
+/// __m128i _mm_dpbusd_epi32(__m128i __W, __m128i __A, __m128i __B)
+/// \endcode
+///
+/// This intrinsic corresponds to the \c VPDPWUSD instruction.
+///
+/// \param __W
+///    A 128-bit vector of [4 x int].
+/// \param __A
+///    A 128-bit vector of [8 x unsigned short].
+/// \param __B
+///    A 128-bit vector of [8 x short].
+/// \returns
+///    A 128-bit vector of [4 x int].
+///
+/// \code{.operation}
+/// FOR j := 0 to 3
+/// 	tmp1.dword := ZeroExtend32(__A.word[2*j]) * SignExtend32(__B.word[2*j])
+/// 	tmp2.dword := ZeroExtend32(__A.word[2*j+1]) * SignExtend32(__B.word[2*j+1])
+/// 	dst.dword[j] := __W.dword[j] + tmp1 + tmp2
+/// ENDFOR
+/// dst[MAX:128] := 0
+/// \endcode
+static __inline__ __m128i __DEFAULT_FN_ATTRS128 _mm_dpwusd_epi32(__m128i __W,
+                                                                 __m128i __A,
+                                                                 __m128i __B) {
+  return (__m128i)__builtin_ia32_vpdpwusd128((__v4si)__W, (__v4si)__A,
+                                             (__v4si)__B);
+}
+
+/// Multiply groups of 2 adjacent pairs of unsigned 16-bit integers in \a __A with
+///    corresponding signed 16-bit integers in \a __B, producing 2 intermediate
+///    signed 16-bit results. Sum these 2 results with the corresponding
+///    32-bit integer in \a __W, and store the packed 32-bit results in \a dst.
+///
+/// \headerfile <immintrin.h>
+///
+/// \code
+/// __m256i _mm256_dpwusd_epi32(__m256i __W, __m256i __A, __m256i __B)
+/// \endcode
+///
+/// This intrinsic corresponds to the \c VPDPWUSD instruction.
+///
+/// \param __W
+///    A 256-bit vector of [8 x int].
+/// \param __A
+///    A 256-bit vector of [16 x unsigned short].
+/// \param __B
+///    A 256-bit vector of [16 x short].
+/// \returns
+///    A 256-bit vector of [8 x int].
+///
+/// \code{.operation}
+/// FOR j := 0 to 7
+/// 	tmp1.dword := ZeroExtend32(__A.word[2*j]) * SignExtend32(__B.word[2*j])
+/// 	tmp2.dword := ZeroExtend32(__A.word[2*j+1]) * SignExtend32(__B.word[2*j+1])
+/// 	dst.dword[j] := __W.dword[j] + tmp1 + tmp2
+/// ENDFOR
+/// dst[MAX:256] := 0
+/// \endcode
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
+_mm256_dpwusd_epi32(__m256i __W, __m256i __A, __m256i __B) {
+  return (__m256i)__builtin_ia32_vpdpwusd256((__v8si)__W, (__v8si)__A,
+                                             (__v8si)__B);
+}
+
+/// Multiply groups of 2 adjacent pairs of unsigned 16-bit integers in \a __A with
+///    corresponding signed 16-bit integers in \a __B, producing 2 intermediate
+///    signed 16-bit results. Sum these 2 results with the corresponding
+///    32-bit integer in \a __W with signed saturation, and store the packed
+///    32-bit results in \a dst.
+///
+/// \headerfile <immintrin.h>
+///
+/// \code
+/// __m128i _mm_dpwusds_epi32(__m128i __W, __m128i __A, __m128i __B)
+/// \endcode
+///
+/// This intrinsic corresponds to the \c VPDPWSUDS instruction.
+///
+/// \param __W
+///    A 128-bit vector of [4 x int].
+/// \param __A
+///    A 128-bit vector of [8 x unsigned short].
+/// \param __B
+///    A 128-bit vector of [8 x short].
+/// \returns
+///    A 128-bit vector of [4 x int].
+///
+/// \code{.operation}
+/// FOR j := 0 to 3
+/// 	tmp1.dword := ZeroExtend32(__A.word[2*j]) * SignExtend32(__B.word[2*j])
+/// 	tmp2.dword := ZeroExtend32(__A.word[2*j+1]) * SignExtend32(__B.word[2*j+1])
+/// 	dst.dword[j] := SIGNED_DWORD_SATURATE(__W.dword[j] + tmp1 + tmp2)
+/// ENDFOR
+/// dst[MAX:128] := 0
+/// \endcode
+static __inline__ __m128i __DEFAULT_FN_ATTRS128 _mm_dpwusds_epi32(__m128i __W,
+                                                                  __m128i __A,
+                                                                  __m128i __B) {
+  return (__m128i)__builtin_ia32_vpdpwusds128((__v4si)__W, (__v4si)__A,
+                                              (__v4si)__B);
+}
+
+/// Multiply groups of 2 adjacent pairs of unsigned 16-bit integers in \a __A with
+///    corresponding signed 16-bit integers in \a __B, producing 2 intermediate
+///    signed 16-bit results. Sum these 2 results with the corresponding
+///    32-bit integer in \a __W with signed saturation, and store the packed
+///    32-bit results in \a dst.
+///
+/// \headerfile <immintrin.h>
+///
+/// \code
+/// __m256i _mm256_dpwsuds_epi32(__m256i __W, __m256i __A, __m256i __B)
+/// \endcode
+///
+/// This intrinsic corresponds to the \c VPDPWSUDS instruction.
+///
+/// \param __W
+///    A 256-bit vector of [8 x int].
+/// \param __A
+///    A 256-bit vector of [16 x unsigned short].
+/// \param __B
+///    A 256-bit vector of [16 x short].
+/// \returns
+///    A 256-bit vector of [8 x int].
+///
+/// \code{.operation}
+/// FOR j := 0 to 7
+/// 	tmp1.dword := ZeroExtend32(__A.word[2*j]) * SignExtend32(__B.word[2*j])
+/// 	tmp2.dword := ZeroExtend32(__A.word[2*j+1]) * SignExtend32(__B.word[2*j+1])
+/// 	dst.dword[j] := SIGNED_DWORD_SATURATE(__W.dword[j] + tmp1 + tmp2)
+/// ENDFOR
+/// dst[MAX:256] := 0
+/// \endcode
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
+_mm256_dpwusds_epi32(__m256i __W, __m256i __A, __m256i __B) {
+  return (__m256i)__builtin_ia32_vpdpwusds256((__v8si)__W, (__v8si)__A,
+                                              (__v8si)__B);
+}
+
+/// Multiply groups of 2 adjacent pairs of unsigned 16-bit integers in \a __A with
+///    corresponding unsigned 16-bit integers in \a __B, producing 2 intermediate
+///    signed 16-bit results. Sum these 2 results with the corresponding
+///    32-bit integer in \a __W, and store the packed 32-bit results in \a dst.
+///
+/// \headerfile <immintrin.h>
+///
+/// \code
+/// __m128i _mm_dpwuud_epi32(__m128i __W, __m128i __A, __m128i __B)
+/// \endcode
+///
+/// This intrinsic corresponds to the \c VPDPWUUD instruction.
+///
+/// \param __W
+///    A 128-bit vector of [4 x unsigned int].
+/// \param __A
+///    A 128-bit vector of [8 x unsigned short].
+/// \param __B
+///    A 128-bit vector of [8 x unsigned short].
+/// \returns
+///    A 128-bit vector of [4 x unsigned int].
+///
+/// \code{.operation}
+/// FOR j := 0 to 3
+/// 	tmp1.dword := ZeroExtend32(__A.word[2*j]) * ZeroExtend32(__B.word[2*j])
+/// 	tmp2.dword := ZeroExtend32(__A.word[2*j+1]) * ZeroExtend32(__B.word[2*j+1])
+/// 	dst.dword[j] := __W.dword[j] + tmp1 + tmp2
+/// ENDFOR
+/// dst[MAX:128] := 0
+/// \endcode
+static __inline__ __m128i __DEFAULT_FN_ATTRS128 _mm_dpwuud_epi32(__m128i __W,
+                                                                 __m128i __A,
+                                                                 __m128i __B) {
+  return (__m128i)__builtin_ia32_vpdpwuud128((__v4si)__W, (__v4si)__A,
+                                             (__v4si)__B);
+}
+
+/// Multiply groups of 2 adjacent pairs of unsigned 16-bit integers in \a __A with
+///    corresponding unsigned 16-bit integers in \a __B, producing 2 intermediate
+///    signed 16-bit results. Sum these 2 results with the corresponding
+///    32-bit integer in \a __W, and store the packed 32-bit results in \a dst.
+///
+/// \headerfile <immintrin.h>
+///
+/// \code
+/// __m256i _mm256_dpwuud_epi32(__m256i __W, __m256i __A, __m256i __B)
+/// \endcode
+///
+/// This intrinsic corresponds to the \c VPDPWUUD instruction.
+///
+/// \param __W
+///    A 256-bit vector of [8 x unsigned int].
+/// \param __A
+///    A 256-bit vector of [16 x unsigned short].
+/// \param __B
+///    A 256-bit vector of [16 x unsigned short].
+/// \returns
+///    A 256-bit vector of [8 x unsigned int].
+///
+/// \code{.operation}
+/// FOR j := 0 to 7
+/// 	tmp1.dword := ZeroExtend32(__A.word[2*j]) * ZeroExtend32(__B.word[2*j])
+/// 	tmp2.dword := ZeroExtend32(__A.word[2*j+1]) * ZeroExtend32(__B.word[2*j+1])
+/// 	dst.dword[j] := __W.dword[j] + tmp1 + tmp2
+/// ENDFOR
+/// dst[MAX:256] := 0
+/// \endcode
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
+_mm256_dpwuud_epi32(__m256i __W, __m256i __A, __m256i __B) {
+  return (__m256i)__builtin_ia32_vpdpwuud256((__v8si)__W, (__v8si)__A,
+                                             (__v8si)__B);
+}
+
+/// Multiply groups of 2 adjacent pairs of unsigned 16-bit integers in \a __A with
+///    corresponding unsigned 16-bit integers in \a __B, producing 2 intermediate
+///    signed 16-bit results. Sum these 2 results with the corresponding
+///    32-bit integer in \a __W with signed saturation, and store the packed
+///    32-bit results in \a dst.
+///
+/// \headerfile <immintrin.h>
+///
+/// \code
+/// __m128i _mm_dpwsuds_epi32(__m128i __W, __m128i __A, __m128i __B)
+/// \endcode
+///
+/// This intrinsic corresponds to the \c VPDPWSUDS instruction.
+///
+/// \param __W
+///    A 128-bit vector of [4 x unsigned int].
+/// \param __A
+///    A 128-bit vector of [8 x unsigned short].
+/// \param __B
+///    A 128-bit vector of [8 x unsigned short].
+/// \returns
+///    A 128-bit vector of [4 x unsigned int].
+///
+/// \code{.operation}
+/// FOR j := 0 to 3
+/// 	tmp1.dword := ZeroExtend32(__A.word[2*j]) * ZeroExtend32(__B.word[2*j])
+/// 	tmp2.dword := ZeroExtend32(__A.word[2*j+1]) * ZeroExtend32(__B.word[2*j+1])
+/// 	dst.dword[j] := UNSIGNED_DWORD_SATURATE(__W.dword[j] + tmp1 + tmp2)
+/// ENDFOR
+/// dst[MAX:128] := 0
+/// \endcode
+static __inline__ __m128i __DEFAULT_FN_ATTRS128 _mm_dpwuuds_epi32(__m128i __W,
+                                                                  __m128i __A,
+                                                                  __m128i __B) {
+  return (__m128i)__builtin_ia32_vpdpwuuds128((__v4si)__W, (__v4si)__A,
+                                              (__v4si)__B);
+}
+
+/// Multiply groups of 2 adjacent pairs of unsigned 16-bit integers in \a __A with
+///    corresponding unsigned 16-bit integers in \a __B, producing 2 intermediate
+///    signed 16-bit results. Sum these 2 results with the corresponding
+///    32-bit integer in \a __W with signed saturation, and store the packed
+///    32-bit results in \a dst.
+///
+/// \headerfile <immintrin.h>
+///
+/// \code
+/// __m256i _mm256_dpwuuds_epi32(__m256i __W, __m256i __A, __m256i __B)
+/// \endcode
+///
+/// This intrinsic corresponds to the \c VPDPWSUDS instruction.
+///
+/// \param __W
+///    A 256-bit vector of [8 x unsigned int].
+/// \param __A
+///    A 256-bit vector of [16 x unsigned short].
+/// \param __B
+///    A 256-bit vector of [16 x unsigned short].
+/// \returns
+///    A 256-bit vector of [8 x unsigned int].
+///
+/// \code{.operation}
+/// FOR j := 0 to 7
+/// 	tmp1.dword := ZeroExtend32(__A.word[2*j]) * ZeroExtend32(__B.word[2*j])
+/// 	tmp2.dword := ZeroExtend32(__A.word[2*j+1]) * ZeroExtend32(__B.word[2*j+1])
+/// 	dst.dword[j] := UNSIGNED_DWORD_SATURATE(__W.dword[j] + tmp1 + tmp2)
+/// ENDFOR
+/// dst[MAX:256] := 0
+/// \endcode
+static __inline__ __m256i __DEFAULT_FN_ATTRS256
+_mm256_dpwuuds_epi32(__m256i __W, __m256i __A, __m256i __B) {
+  return (__m256i)__builtin_ia32_vpdpwuuds256((__v8si)__W, (__v8si)__A,
+                                              (__v8si)__B);
+}
+
+#undef __DEFAULT_FN_ATTRS128
+#undef __DEFAULT_FN_ATTRS256
+
+#endif // __AVXVNNIINT16INTRIN_H