@nativescript/visionos 8.7.1 → 8.9.0

package/framework/internal/metadata-generator-arm64/bin/lib/clang/{15.0.7 → 17}/include/avx512bf16intrin.h

@@ -10,12 +10,14 @@
 #error "Never use <avx512bf16intrin.h> directly; include <immintrin.h> instead."
 #endif
 
+#ifdef __SSE2__
+
 #ifndef __AVX512BF16INTRIN_H
 #define __AVX512BF16INTRIN_H
 
-typedef short __m512bh __attribute__((__vector_size__(64), __aligned__(64)));
-typedef short __m256bh __attribute__((__vector_size__(32), __aligned__(32)));
-typedef unsigned short __bfloat16;
+typedef __bf16 __v32bf __attribute__((__vector_size__(64), __aligned__(64)));
+typedef __bf16 __m512bh __attribute__((__vector_size__(64), __aligned__(64)));
+typedef __bf16 __bfloat16 __attribute__((deprecated("use __bf16 instead")));
 
 #define __DEFAULT_FN_ATTRS512 \
   __attribute__((__always_inline__, __nodebug__, __target__("avx512bf16"), \
@@ -33,7 +35,7 @@ typedef unsigned short __bfloat16;
 ///    A bfloat data.
 /// \returns A float data whose sign field and exponent field keep unchanged,
 ///    and fraction field is extended to 23 bits.
-static __inline__ float __DEFAULT_FN_ATTRS _mm_cvtsbh_ss(__bfloat16 __A) {
+static __inline__ float __DEFAULT_FN_ATTRS _mm_cvtsbh_ss(__bf16 __A) {
   return __builtin_ia32_cvtsbf162ss_32(__A);
 }
 
@@ -74,9 +76,9 @@ _mm512_cvtne2ps_pbh(__m512 __A, __m512 __B) {
 ///    conversion of __B, and higher 256 bits come from conversion of __A.
 static __inline__ __m512bh __DEFAULT_FN_ATTRS512
 _mm512_mask_cvtne2ps_pbh(__m512bh __W, __mmask32 __U, __m512 __A, __m512 __B) {
-  return (__m512bh)__builtin_ia32_selectw_512((__mmask32)__U,
-                                        (__v32hi)_mm512_cvtne2ps_pbh(__A, __B),
-                                        (__v32hi)__W);
+  return (__m512bh)__builtin_ia32_selectpbf_512((__mmask32)__U,
+                                        (__v32bf)_mm512_cvtne2ps_pbh(__A, __B),
+                                        (__v32bf)__W);
 }
 
 /// Convert Two Packed Single Data to One Packed BF16 Data.
@@ -96,9 +98,9 @@ _mm512_mask_cvtne2ps_pbh(__m512bh __W, __mmask32 __U, __m512 __A, __m512 __B) {
 ///    conversion of __B, and higher 256 bits come from conversion of __A.
 static __inline__ __m512bh __DEFAULT_FN_ATTRS512
 _mm512_maskz_cvtne2ps_pbh(__mmask32 __U, __m512 __A, __m512 __B) {
-  return (__m512bh)__builtin_ia32_selectw_512((__mmask32)__U,
-                                        (__v32hi)_mm512_cvtne2ps_pbh(__A, __B),
-                                        (__v32hi)_mm512_setzero_si512());
+  return (__m512bh)__builtin_ia32_selectpbf_512((__mmask32)__U,
+                                        (__v32bf)_mm512_cvtne2ps_pbh(__A, __B),
+                                        (__v32bf)_mm512_setzero_si512());
 }
 
 /// Convert Packed Single Data to Packed BF16 Data.
@@ -113,7 +115,7 @@ _mm512_maskz_cvtne2ps_pbh(__mmask32 __U, __m512 __A, __m512 __B) {
 static __inline__ __m256bh __DEFAULT_FN_ATTRS512
 _mm512_cvtneps_pbh(__m512 __A) {
   return (__m256bh)__builtin_ia32_cvtneps2bf16_512_mask((__v16sf)__A,
-                                              (__v16hi)_mm256_undefined_si256(),
+                                              (__v16bf)_mm256_undefined_si256(),
                                               (__mmask16)-1);
 }
 
@@ -134,7 +136,7 @@ _mm512_cvtneps_pbh(__m512 __A) {
 static __inline__ __m256bh __DEFAULT_FN_ATTRS512
 _mm512_mask_cvtneps_pbh(__m256bh __W, __mmask16 __U, __m512 __A) {
   return (__m256bh)__builtin_ia32_cvtneps2bf16_512_mask((__v16sf)__A,
-                                                        (__v16hi)__W,
+                                                        (__v16bf)__W,
                                                         (__mmask16)__U);
 }
 
@@ -153,7 +155,7 @@ _mm512_mask_cvtneps_pbh(__m256bh __W, __mmask16 __U, __m512 __A) {
 static __inline__ __m256bh __DEFAULT_FN_ATTRS512
 _mm512_maskz_cvtneps_pbh(__mmask16 __U, __m512 __A) {
   return (__m256bh)__builtin_ia32_cvtneps2bf16_512_mask((__v16sf)__A,
-                                                (__v16hi)_mm256_setzero_si256(),
+                                                (__v16bf)_mm256_setzero_si256(),
                                                 (__mmask16)__U);
 }
 
@@ -174,8 +176,8 @@ _mm512_maskz_cvtneps_pbh(__mmask16 __U, __m512 __A) {
 static __inline__ __m512 __DEFAULT_FN_ATTRS512
 _mm512_dpbf16_ps(__m512 __D, __m512bh __A, __m512bh __B) {
   return (__m512)__builtin_ia32_dpbf16ps_512((__v16sf) __D,
-                                             (__v16si) __A,
-                                             (__v16si) __B);
+                                             (__v32bf) __A,
+                                             (__v32bf) __B);
 }
 
 /// Dot Product of BF16 Pairs Accumulated into Packed Single Precision.
@@ -277,3 +279,4 @@ _mm512_mask_cvtpbh_ps(__m512 __S, __mmask16 __U, __m256bh __A) {
 #undef __DEFAULT_FN_ATTRS512
 
 #endif
+#endif

package/framework/internal/{metadata-generator-x86_64/bin/lib/clang/15.0.7 → metadata-generator-arm64/bin/lib/clang/17}/include/avx512fintrin.h

@@ -256,8 +256,8 @@ _mm512_maskz_broadcastq_epi64 (__mmask8 __M, __m128i __A)
 static __inline __m512 __DEFAULT_FN_ATTRS512
 _mm512_setzero_ps(void)
 {
-  return __extension__ (__m512){ 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0,
-                                 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0 };
+  return __extension__ (__m512){ 0.0f, 0.0f, 0.0f, 0.0f, 0.0f, 0.0f, 0.0f, 0.0f,
+                                 0.0f, 0.0f, 0.0f, 0.0f, 0.0f, 0.0f, 0.0f, 0.0f };
 }
 
 #define _mm512_setzero _mm512_setzero_ps
@@ -397,14 +397,15 @@ _mm512_broadcastsd_pd(__m128d __A)
 static __inline __m512d __DEFAULT_FN_ATTRS512
 _mm512_castpd256_pd512(__m256d __a)
 {
-  return __builtin_shufflevector(__a, __a, 0, 1, 2, 3, -1, -1, -1, -1);
+  return __builtin_shufflevector(__a, __builtin_nondeterministic_value(__a), 0,
+                                 1, 2, 3, 4, 5, 6, 7);
 }
 
 static __inline __m512 __DEFAULT_FN_ATTRS512
 _mm512_castps256_ps512(__m256 __a)
 {
-  return __builtin_shufflevector(__a, __a, 0,  1,  2,  3,  4,  5,  6,  7,
-                                          -1, -1, -1, -1, -1, -1, -1, -1);
+  return __builtin_shufflevector(__a, __builtin_nondeterministic_value(__a), 0,
+                                 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15);
 }
 
 static __inline __m128d __DEFAULT_FN_ATTRS512
@@ -446,7 +447,10 @@ _mm512_castpd_si512 (__m512d __A)
 static __inline__ __m512d __DEFAULT_FN_ATTRS512
 _mm512_castpd128_pd512 (__m128d __A)
 {
-  return __builtin_shufflevector( __A, __A, 0, 1, -1, -1, -1, -1, -1, -1);
+  __m256d __B = __builtin_nondeterministic_value(__B);
+  return __builtin_shufflevector(
+      __builtin_shufflevector(__A, __builtin_nondeterministic_value(__A), 0, 1, 2, 3),
+      __B, 0, 1, 2, 3, 4, 5, 6, 7);
 }
 
 static __inline __m512d __DEFAULT_FN_ATTRS512
@@ -464,19 +468,25 @@ _mm512_castps_si512 (__m512 __A)
 static __inline__ __m512 __DEFAULT_FN_ATTRS512
 _mm512_castps128_ps512 (__m128 __A)
 {
-    return  __builtin_shufflevector( __A, __A, 0, 1, 2, 3, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1);
+  __m256 __B = __builtin_nondeterministic_value(__B);
+  return __builtin_shufflevector(
+      __builtin_shufflevector(__A, __builtin_nondeterministic_value(__A), 0, 1, 2, 3, 4, 5, 6, 7),
+      __B, 0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15);
 }
 
 static __inline__ __m512i __DEFAULT_FN_ATTRS512
 _mm512_castsi128_si512 (__m128i __A)
 {
-   return  __builtin_shufflevector( __A, __A, 0, 1, -1, -1, -1, -1, -1, -1);
+  __m256i __B = __builtin_nondeterministic_value(__B);
+  return __builtin_shufflevector(
+      __builtin_shufflevector(__A, __builtin_nondeterministic_value(__A), 0, 1, 2, 3),
+      __B, 0, 1, 2, 3, 4, 5, 6, 7);
 }
 
 static __inline__ __m512i __DEFAULT_FN_ATTRS512
 _mm512_castsi256_si512 (__m256i __A)
 {
-   return  __builtin_shufflevector( __A, __A, 0, 1, 2, 3, -1, -1, -1, -1);
+   return  __builtin_shufflevector( __A, __builtin_nondeterministic_value(__A), 0, 1, 2, 3, 4, 5, 6, 7);
 }
 
 static __inline __m512 __DEFAULT_FN_ATTRS512

package/framework/internal/metadata-generator-arm64/bin/lib/clang/{15.0.7 → 17}/include/avx512fp16intrin.h

@@ -10,6 +10,8 @@
 #error "Never use <avx512fp16intrin.h> directly; include <immintrin.h> instead."
 #endif
 
+#ifdef __SSE2__
+
 #ifndef __AVX512FP16INTRIN_H
 #define __AVX512FP16INTRIN_H
 
@@ -17,12 +19,6 @@
 typedef _Float16 __v32hf __attribute__((__vector_size__(64), __aligned__(64)));
 typedef _Float16 __m512h __attribute__((__vector_size__(64), __aligned__(64)));
 typedef _Float16 __m512h_u __attribute__((__vector_size__(64), __aligned__(1)));
-typedef _Float16 __v8hf __attribute__((__vector_size__(16), __aligned__(16)));
-typedef _Float16 __m128h __attribute__((__vector_size__(16), __aligned__(16)));
-typedef _Float16 __m128h_u __attribute__((__vector_size__(16), __aligned__(1)));
-typedef _Float16 __v16hf __attribute__((__vector_size__(32), __aligned__(32)));
-typedef _Float16 __m256h __attribute__((__vector_size__(32), __aligned__(32)));
-typedef _Float16 __m256h_u __attribute__((__vector_size__(32), __aligned__(1)));
 
 /* Define the default attributes for the functions in this file. */
 #define __DEFAULT_FN_ATTRS512                                                  \
@@ -196,22 +192,26 @@ _mm512_castph512_ph256(__m512h __a) {
 
 static __inline__ __m256h __DEFAULT_FN_ATTRS256
 _mm256_castph128_ph256(__m128h __a) {
-  return __builtin_shufflevector(__a, __a, 0, 1, 2, 3, 4, 5, 6, 7, -1, -1, -1,
-                                 -1, -1, -1, -1, -1);
+  return __builtin_shufflevector(__a, __builtin_nondeterministic_value(__a),
+                                  0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15);
 }
 
 static __inline__ __m512h __DEFAULT_FN_ATTRS512
 _mm512_castph128_ph512(__m128h __a) {
-  return __builtin_shufflevector(__a, __a, 0, 1, 2, 3, 4, 5, 6, 7, -1, -1, -1,
-                                 -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1,
-                                 -1, -1, -1, -1, -1, -1, -1, -1, -1);
+  __m256h __b = __builtin_nondeterministic_value(__b);
+  return __builtin_shufflevector(
+      __builtin_shufflevector(__a, __builtin_nondeterministic_value(__a),
+                              0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15),
+      __b, 0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19,
+      20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31);
 }
 
 static __inline__ __m512h __DEFAULT_FN_ATTRS512
 _mm512_castph256_ph512(__m256h __a) {
-  return __builtin_shufflevector(__a, __a, 0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11,
-                                 12, 13, 14, 15, -1, -1, -1, -1, -1, -1, -1, -1,
-                                 -1, -1, -1, -1, -1, -1, -1, -1);
+  return __builtin_shufflevector(__a, __builtin_nondeterministic_value(__a), 0,
+                                 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14,
+                                 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26,
+                                 27, 28, 29, 30, 31);
 }
 
 /// Constructs a 256-bit floating-point vector of [16 x half] from a
@@ -829,7 +829,7 @@ static __inline__ __m128h __DEFAULT_FN_ATTRS128 _mm_load_sh(void const *__dp) {
   struct __mm_load_sh_struct {
     _Float16 __u;
   } __attribute__((__packed__, __may_alias__));
-  _Float16 __u = ((struct __mm_load_sh_struct *)__dp)->__u;
+  _Float16 __u = ((const struct __mm_load_sh_struct *)__dp)->__u;
   return (__m128h){__u, 0, 0, 0, 0, 0, 0, 0};
 }
 
@@ -838,13 +838,13 @@ _mm_mask_load_sh(__m128h __W, __mmask8 __U, const void *__A) {
   __m128h src = (__v8hf)__builtin_shufflevector(
       (__v8hf)__W, (__v8hf)_mm_setzero_ph(), 0, 8, 8, 8, 8, 8, 8, 8);
 
-  return (__m128h)__builtin_ia32_loadsh128_mask((__v8hf *)__A, src, __U & 1);
+  return (__m128h)__builtin_ia32_loadsh128_mask((const __v8hf *)__A, src, __U & 1);
 }
 
 static __inline__ __m128h __DEFAULT_FN_ATTRS128
 _mm_maskz_load_sh(__mmask8 __U, const void *__A) {
   return (__m128h)__builtin_ia32_loadsh128_mask(
-      (__v8hf *)__A, (__v8hf)_mm_setzero_ph(), __U & 1);
+      (const __v8hf *)__A, (__v8hf)_mm_setzero_ph(), __U & 1);
 }
 
 static __inline__ __m512h __DEFAULT_FN_ATTRS512
@@ -3347,3 +3347,4 @@ _mm512_permutexvar_ph(__m512i __A, __m512h __B) {
 #undef __DEFAULT_FN_ATTRS512
 
 #endif
+#endif

package/framework/internal/{metadata-generator-x86_64/bin/lib/clang/15.0.7 → metadata-generator-arm64/bin/lib/clang/17}/include/avx512ifmavlintrin.h

@@ -18,14 +18,21 @@
 #define __DEFAULT_FN_ATTRS128 __attribute__((__always_inline__, __nodebug__, __target__("avx512ifma,avx512vl"), __min_vector_width__(128)))
 #define __DEFAULT_FN_ATTRS256 __attribute__((__always_inline__, __nodebug__, __target__("avx512ifma,avx512vl"), __min_vector_width__(256)))
 
+#define _mm_madd52hi_epu64(X, Y, Z)                                            \
+  ((__m128i)__builtin_ia32_vpmadd52huq128((__v2di)(X), (__v2di)(Y),            \
+                                          (__v2di)(Z)))
 
+#define _mm256_madd52hi_epu64(X, Y, Z)                                         \
+  ((__m256i)__builtin_ia32_vpmadd52huq256((__v4di)(X), (__v4di)(Y),            \
+                                          (__v4di)(Z)))
 
-static __inline__ __m128i __DEFAULT_FN_ATTRS128
-_mm_madd52hi_epu64 (__m128i __X, __m128i __Y, __m128i __Z)
-{
-  return (__m128i)__builtin_ia32_vpmadd52huq128((__v2di) __X, (__v2di) __Y,
-                                                (__v2di) __Z);
-}
+#define _mm_madd52lo_epu64(X, Y, Z)                                            \
+  ((__m128i)__builtin_ia32_vpmadd52luq128((__v2di)(X), (__v2di)(Y),            \
+                                          (__v2di)(Z)))
+
+#define _mm256_madd52lo_epu64(X, Y, Z)                                         \
+  ((__m256i)__builtin_ia32_vpmadd52luq256((__v4di)(X), (__v4di)(Y),            \
+                                          (__v4di)(Z)))
 
 static __inline__ __m128i __DEFAULT_FN_ATTRS128
 _mm_mask_madd52hi_epu64 (__m128i __W, __mmask8 __M, __m128i __X, __m128i __Y)
@@ -43,13 +50,6 @@ _mm_maskz_madd52hi_epu64 (__mmask8 __M, __m128i __X, __m128i __Y, __m128i __Z)
                                       (__v2di)_mm_setzero_si128());
 }
 
-static __inline__ __m256i __DEFAULT_FN_ATTRS256
-_mm256_madd52hi_epu64 (__m256i __X, __m256i __Y, __m256i __Z)
-{
-  return (__m256i)__builtin_ia32_vpmadd52huq256((__v4di)__X, (__v4di)__Y,
-                                                (__v4di)__Z);
-}
-
 static __inline__ __m256i __DEFAULT_FN_ATTRS256
 _mm256_mask_madd52hi_epu64 (__m256i __W, __mmask8 __M, __m256i __X, __m256i __Y)
 {
@@ -66,13 +66,6 @@ _mm256_maskz_madd52hi_epu64 (__mmask8 __M, __m256i __X, __m256i __Y, __m256i __Z
                                    (__v4di)_mm256_setzero_si256());
 }
 
-static __inline__ __m128i __DEFAULT_FN_ATTRS128
-_mm_madd52lo_epu64 (__m128i __X, __m128i __Y, __m128i __Z)
-{
-  return (__m128i)__builtin_ia32_vpmadd52luq128((__v2di)__X, (__v2di)__Y,
-                                                (__v2di)__Z);
-}
-
 static __inline__ __m128i __DEFAULT_FN_ATTRS128
 _mm_mask_madd52lo_epu64 (__m128i __W, __mmask8 __M, __m128i __X, __m128i __Y)
 {
@@ -89,13 +82,6 @@ _mm_maskz_madd52lo_epu64 (__mmask8 __M, __m128i __X, __m128i __Y, __m128i __Z)
                                       (__v2di)_mm_setzero_si128());
 }
 
-static __inline__ __m256i __DEFAULT_FN_ATTRS256
-_mm256_madd52lo_epu64 (__m256i __X, __m256i __Y, __m256i __Z)
-{
-  return (__m256i)__builtin_ia32_vpmadd52luq256((__v4di)__X, (__v4di)__Y,
-                                                (__v4di)__Z);
-}
-
 static __inline__ __m256i __DEFAULT_FN_ATTRS256
 _mm256_mask_madd52lo_epu64 (__m256i __W, __mmask8 __M, __m256i __X, __m256i __Y)
 {

package/framework/internal/metadata-generator-arm64/bin/lib/clang/{15.0.7 → 17}/include/avx512vlbf16intrin.h

@@ -10,11 +10,11 @@
 #error "Never use <avx512vlbf16intrin.h> directly; include <immintrin.h> instead."
 #endif
 
+#ifdef __SSE2__
+
 #ifndef __AVX512VLBF16INTRIN_H
 #define __AVX512VLBF16INTRIN_H
 
-typedef short __m128bh __attribute__((__vector_size__(16), __aligned__(16)));
-
 #define __DEFAULT_FN_ATTRS128 \
   __attribute__((__always_inline__, __nodebug__, \
                  __target__("avx512vl, avx512bf16"), __min_vector_width__(128)))
@@ -59,9 +59,9 @@ _mm_cvtne2ps_pbh(__m128 __A, __m128 __B) {
 ///    conversion of __B, and higher 64 bits come from conversion of __A.
 static __inline__ __m128bh __DEFAULT_FN_ATTRS128
 _mm_mask_cvtne2ps_pbh(__m128bh __W, __mmask8 __U, __m128 __A, __m128 __B) {
-  return (__m128bh)__builtin_ia32_selectw_128((__mmask8)__U,
-                                             (__v8hi)_mm_cvtne2ps_pbh(__A, __B),
-                                             (__v8hi)__W);
+  return (__m128bh)__builtin_ia32_selectpbf_128((__mmask8)__U,
+                                             (__v8bf)_mm_cvtne2ps_pbh(__A, __B),
+                                             (__v8bf)__W);
 }
 
 /// Convert Two Packed Single Data to One Packed BF16 Data.
@@ -81,9 +81,9 @@ _mm_mask_cvtne2ps_pbh(__m128bh __W, __mmask8 __U, __m128 __A, __m128 __B) {
 ///    conversion of __B, and higher 64 bits come from conversion of __A.
 static __inline__ __m128bh __DEFAULT_FN_ATTRS128
 _mm_maskz_cvtne2ps_pbh(__mmask8 __U, __m128 __A, __m128 __B) {
-  return (__m128bh)__builtin_ia32_selectw_128((__mmask8)__U,
-                                             (__v8hi)_mm_cvtne2ps_pbh(__A, __B),
-                                             (__v8hi)_mm_setzero_si128());
+  return (__m128bh)__builtin_ia32_selectpbf_128((__mmask8)__U,
+                                             (__v8bf)_mm_cvtne2ps_pbh(__A, __B),
+                                             (__v8bf)_mm_setzero_si128());
 }
 
 /// Convert Two Packed Single Data to One Packed BF16 Data.
@@ -123,9 +123,9 @@ _mm256_cvtne2ps_pbh(__m256 __A, __m256 __B) {
 ///    conversion of __B, and higher 128 bits come from conversion of __A.
 static __inline__ __m256bh __DEFAULT_FN_ATTRS256
 _mm256_mask_cvtne2ps_pbh(__m256bh __W, __mmask16 __U, __m256 __A, __m256 __B) {
-  return (__m256bh)__builtin_ia32_selectw_256((__mmask16)__U,
-                                         (__v16hi)_mm256_cvtne2ps_pbh(__A, __B),
-                                         (__v16hi)__W);
+  return (__m256bh)__builtin_ia32_selectpbf_256((__mmask16)__U,
+                                         (__v16bf)_mm256_cvtne2ps_pbh(__A, __B),
+                                         (__v16bf)__W);
 }
 
 /// Convert Two Packed Single Data to One Packed BF16 Data.
@@ -145,9 +145,9 @@ _mm256_mask_cvtne2ps_pbh(__m256bh __W, __mmask16 __U, __m256 __A, __m256 __B) {
 ///    conversion of __B, and higher 128 bits come from conversion of __A.
 static __inline__ __m256bh __DEFAULT_FN_ATTRS256
 _mm256_maskz_cvtne2ps_pbh(__mmask16 __U, __m256 __A, __m256 __B) {
-  return (__m256bh)__builtin_ia32_selectw_256((__mmask16)__U,
-                                         (__v16hi)_mm256_cvtne2ps_pbh(__A, __B),
-                                         (__v16hi)_mm256_setzero_si256());
+  return (__m256bh)__builtin_ia32_selectpbf_256((__mmask16)__U,
+                                         (__v16bf)_mm256_cvtne2ps_pbh(__A, __B),
+                                         (__v16bf)_mm256_setzero_si256());
 }
 
 /// Convert Packed Single Data to Packed BF16 Data.
@@ -160,12 +160,8 @@ _mm256_maskz_cvtne2ps_pbh(__mmask16 __U, __m256 __A, __m256 __B) {
 ///    A 128-bit vector of [4 x float].
 /// \returns A 128-bit vector of [8 x bfloat] whose lower 64 bits come from
 ///    conversion of __A, and higher 64 bits are 0.
-static __inline__ __m128bh __DEFAULT_FN_ATTRS128
-_mm_cvtneps_pbh(__m128 __A) {
-  return (__m128bh)__builtin_ia32_cvtneps2bf16_128_mask((__v4sf) __A,
-                                                  (__v8hi)_mm_undefined_si128(),
-                                                  (__mmask8)-1);
-}
+#define _mm_cvtneps_pbh(A)                                                     \
+  ((__m128bh)__builtin_ia32_vcvtneps2bf16128((__v4sf)(A)))
 
 /// Convert Packed Single Data to Packed BF16 Data.
 ///
@@ -185,7 +181,7 @@ _mm_cvtneps_pbh(__m128 __A) {
 static __inline__ __m128bh __DEFAULT_FN_ATTRS128
 _mm_mask_cvtneps_pbh(__m128bh __W, __mmask8 __U, __m128 __A) {
   return (__m128bh)__builtin_ia32_cvtneps2bf16_128_mask((__v4sf) __A,
-                                                        (__v8hi)__W,
+                                                        (__v8bf)__W,
                                                         (__mmask8)__U);
 }
 
@@ -205,7 +201,7 @@ _mm_mask_cvtneps_pbh(__m128bh __W, __mmask8 __U, __m128 __A) {
 static __inline__ __m128bh __DEFAULT_FN_ATTRS128
 _mm_maskz_cvtneps_pbh(__mmask8 __U, __m128 __A) {
   return (__m128bh)__builtin_ia32_cvtneps2bf16_128_mask((__v4sf) __A,
-                                                    (__v8hi)_mm_setzero_si128(),
+                                                    (__v8bf)_mm_setzero_si128(),
                                                     (__mmask8)__U);
 }
 
@@ -218,12 +214,8 @@ _mm_maskz_cvtneps_pbh(__mmask8 __U, __m128 __A) {
 /// \param __A
 ///    A 256-bit vector of [8 x float].
 /// \returns A 128-bit vector of [8 x bfloat] comes from conversion of __A.
-static __inline__ __m128bh __DEFAULT_FN_ATTRS256
-_mm256_cvtneps_pbh(__m256 __A) {
-  return (__m128bh)__builtin_ia32_cvtneps2bf16_256_mask((__v8sf)__A,
-                                                  (__v8hi)_mm_undefined_si128(),
-                                                  (__mmask8)-1);
-}
+#define _mm256_cvtneps_pbh(A)                                                  \
+  ((__m128bh)__builtin_ia32_vcvtneps2bf16256((__v8sf)(A)))
 
 /// Convert Packed Single Data to Packed BF16 Data.
 ///
@@ -242,7 +234,7 @@ _mm256_cvtneps_pbh(__m256 __A) {
 static __inline__ __m128bh __DEFAULT_FN_ATTRS256
 _mm256_mask_cvtneps_pbh(__m128bh __W, __mmask8 __U, __m256 __A) {
   return (__m128bh)__builtin_ia32_cvtneps2bf16_256_mask((__v8sf)__A,
-                                                        (__v8hi)__W,
+                                                        (__v8bf)__W,
                                                         (__mmask8)__U);
 }
 
@@ -261,7 +253,7 @@ _mm256_mask_cvtneps_pbh(__m128bh __W, __mmask8 __U, __m256 __A) {
 static __inline__ __m128bh __DEFAULT_FN_ATTRS256
 _mm256_maskz_cvtneps_pbh(__mmask8 __U, __m256 __A) {
   return (__m128bh)__builtin_ia32_cvtneps2bf16_256_mask((__v8sf)__A,
-                                                    (__v8hi)_mm_setzero_si128(),
+                                                    (__v8bf)_mm_setzero_si128(),
                                                     (__mmask8)__U);
 }
 
@@ -282,8 +274,8 @@ _mm256_maskz_cvtneps_pbh(__mmask8 __U, __m256 __A) {
 static __inline__ __m128 __DEFAULT_FN_ATTRS128
 _mm_dpbf16_ps(__m128 __D, __m128bh __A, __m128bh __B) {
   return (__m128)__builtin_ia32_dpbf16ps_128((__v4sf)__D,
-                                             (__v4si)__A,
-                                             (__v4si)__B);
+                                             (__v8bf)__A,
+                                             (__v8bf)__B);
 }
 
 /// Dot Product of BF16 Pairs Accumulated into Packed Single Precision.
@@ -351,8 +343,8 @@ _mm_maskz_dpbf16_ps(__mmask8 __U, __m128 __D, __m128bh __A, __m128bh __B) {
 static __inline__ __m256 __DEFAULT_FN_ATTRS256
 _mm256_dpbf16_ps(__m256 __D, __m256bh __A, __m256bh __B) {
   return (__m256)__builtin_ia32_dpbf16ps_256((__v8sf)__D,
-                                             (__v8si)__A,
-                                             (__v8si)__B);
+                                             (__v16bf)__A,
+                                             (__v16bf)__B);
 }
 
 /// Dot Product of BF16 Pairs Accumulated into Packed Single Precision.
@@ -413,11 +405,11 @@ _mm256_maskz_dpbf16_ps(__mmask8 __U, __m256 __D, __m256bh __A, __m256bh __B) {
 ///    A float data.
 /// \returns A bf16 data whose sign field and exponent field keep unchanged,
 ///    and fraction field is truncated to 7 bits.
-static __inline__ __bfloat16 __DEFAULT_FN_ATTRS128 _mm_cvtness_sbh(float __A) {
+static __inline__ __bf16 __DEFAULT_FN_ATTRS128 _mm_cvtness_sbh(float __A) {
   __v4sf __V = {__A, 0, 0, 0};
-  __v8hi __R = __builtin_ia32_cvtneps2bf16_128_mask(
-      (__v4sf)__V, (__v8hi)_mm_undefined_si128(), (__mmask8)-1);
-  return (__bfloat16)__R[0];
+  __v8bf __R = __builtin_ia32_cvtneps2bf16_128_mask(
+      (__v4sf)__V, (__v8bf)_mm_undefined_si128(), (__mmask8)-1);
+  return (__bf16)__R[0];
 }
 
 /// Convert Packed BF16 Data to Packed float Data.
@@ -520,3 +512,4 @@ _mm256_mask_cvtpbh_ps(__m256 __S, __mmask8 __U, __m128bh __A) {
 #undef __DEFAULT_FN_ATTRS256
 
 #endif
+#endif