@nativescript/visionos 8.7.1 → 8.8.1

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

@@ -3202,71 +3202,79 @@ static __inline__ vector double __ATTRS_o_ai vec_cpsgn(vector double __a,
 // the XL-compatible signatures are used for those functions.
 #ifdef __XL_COMPAT_ALTIVEC__
 #define vec_ctf(__a, __b)                                                      \
-  _Generic(                                                                    \
-      (__a), vector int                                                        \
-      : (vector float)__builtin_altivec_vcfsx((vector int)(__a), (__b)),       \
-        vector unsigned int                                                    \
-      : (vector float)__builtin_altivec_vcfux((vector unsigned int)(__a),      \
-                                              (__b)),                          \
-        vector unsigned long long                                              \
-      : (vector float)(__builtin_vsx_xvcvuxdsp(                                \
-                           (vector unsigned long long)(__a)) *                 \
-                       (vector float)(vector unsigned)((0x7f - (__b)) << 23)), \
-        vector signed long long                                                \
-      : (vector float)(__builtin_vsx_xvcvsxdsp(                                \
-                           (vector signed long long)(__a)) *                   \
-                       (vector float)(vector unsigned)((0x7f - (__b)) << 23)))
+  _Generic((__a),                                                              \
+      vector int: (vector float)__builtin_altivec_vcfsx((vector int)(__a),     \
+                                                        ((__b)&0x1F)),         \
+      vector unsigned int: (vector float)__builtin_altivec_vcfux(              \
+               (vector unsigned int)(__a), ((__b)&0x1F)),                      \
+      vector unsigned long long: (                                             \
+               vector float)(__builtin_vsx_xvcvuxdsp(                          \
+                                 (vector unsigned long long)(__a)) *           \
+                             (vector float)(vector unsigned)((0x7f -           \
+                                                              ((__b)&0x1F))    \
+                                                             << 23)),          \
+      vector signed long long: (                                               \
+               vector float)(__builtin_vsx_xvcvsxdsp(                          \
+                                 (vector signed long long)(__a)) *             \
+                             (vector float)(vector unsigned)((0x7f -           \
+                                                              ((__b)&0x1F))    \
+                                                             << 23)))
 #else // __XL_COMPAT_ALTIVEC__
-#define vec_ctf(__a, __b)                                                      \
-  _Generic(                                                                    \
-      (__a), vector int                                                        \
-      : (vector float)__builtin_altivec_vcfsx((vector int)(__a), (__b)),       \
-        vector unsigned int                                                    \
-      : (vector float)__builtin_altivec_vcfux((vector unsigned int)(__a),      \
-                                              (__b)),                          \
-        vector unsigned long long                                              \
-      : (vector float)(__builtin_convertvector(                                \
-                           (vector unsigned long long)(__a), vector double) *  \
-                       (vector double)(vector unsigned long long)((0x3ffULL -  \
-                                                                   (__b))      \
-                                                                  << 52)),     \
-        vector signed long long                                                \
-      : (vector float)(__builtin_convertvector((vector signed long long)(__a), \
-                                               vector double) *                \
-                       (vector double)(vector unsigned long long)((0x3ffULL -  \
-                                                                   (__b))      \
-                                                                  << 52)))
+#define vec_ctf(__a, __b)                                                         \
+  _Generic(                                                                       \
+      (__a),                                                                      \
+      vector int: (vector float)__builtin_altivec_vcfsx((vector int)(__a),        \
+                                                        ((__b)&0x1F)),            \
+      vector unsigned int: (vector float)__builtin_altivec_vcfux(                 \
+          (vector unsigned int)(__a), ((__b)&0x1F)),                              \
+      vector unsigned long long: (                                                \
+          vector float)(__builtin_convertvector(                                  \
+                            (vector unsigned long long)(__a), vector double) *    \
+                        (vector double)(vector unsigned long long)((0x3ffULL -    \
+                                                                    ((__b)&0x1F)) \
+                                                                   << 52)),       \
+      vector signed long long: (                                                  \
+          vector float)(__builtin_convertvector(                                  \
+                            (vector signed long long)(__a), vector double) *      \
+                        (vector double)(vector unsigned long long)((0x3ffULL -    \
+                                                                    ((__b)&0x1F)) \
+                                                                   << 52)))
 #endif // __XL_COMPAT_ALTIVEC__
 #else
 #define vec_ctf(__a, __b)                                                      \
-  _Generic((__a), vector int                                                   \
-           : (vector float)__builtin_altivec_vcfsx((vector int)(__a), (__b)),  \
-             vector unsigned int                                               \
-           : (vector float)__builtin_altivec_vcfux((vector unsigned int)(__a), \
-                                                   (__b)))
+  _Generic((__a),                                                              \
+      vector int: (vector float)__builtin_altivec_vcfsx((vector int)(__a),     \
+                                                        ((__b)&0x1F)),         \
+      vector unsigned int: (vector float)__builtin_altivec_vcfux(              \
+               (vector unsigned int)(__a), ((__b)&0x1F)))
 #endif
 
 /* vec_ctd */
 #ifdef __VSX__
 #define vec_ctd(__a, __b)                                                      \
-  _Generic((__a), vector signed int                                            \
-           : (vec_doublee((vector signed int)(__a)) *                          \
-              (vector double)(vector unsigned long long)((0x3ffULL - (__b))    \
-                                                         << 52)),              \
-             vector unsigned int                                               \
-           : (vec_doublee((vector unsigned int)(__a)) *                        \
-              (vector double)(vector unsigned long long)((0x3ffULL - (__b))    \
-                                                         << 52)),              \
-             vector unsigned long long                                         \
-           : (__builtin_convertvector((vector unsigned long long)(__a),        \
-                                      vector double) *                         \
-              (vector double)(vector unsigned long long)((0x3ffULL - (__b))    \
-                                                         << 52)),              \
-             vector signed long long                                           \
-           : (__builtin_convertvector((vector signed long long)(__a),          \
-                                      vector double) *                         \
-              (vector double)(vector unsigned long long)((0x3ffULL - (__b))    \
-                                                         << 52)))
+  _Generic((__a),                                                              \
+      vector signed int: (                                                     \
+               vec_doublee((vector signed int)(__a)) *                         \
+               (vector double)(vector unsigned long long)((0x3ffULL -          \
+                                                           ((__b)&0x1F))       \
+                                                          << 52)),             \
+      vector unsigned int: (                                                   \
+               vec_doublee((vector unsigned int)(__a)) *                       \
+               (vector double)(vector unsigned long long)((0x3ffULL -          \
+                                                           ((__b)&0x1F))       \
+                                                          << 52)),             \
+      vector unsigned long long: (                                             \
+               __builtin_convertvector((vector unsigned long long)(__a),       \
+                                       vector double) *                        \
+               (vector double)(vector unsigned long long)((0x3ffULL -          \
+                                                           ((__b)&0x1F))       \
+                                                          << 52)),             \
+      vector signed long long: (                                               \
+               __builtin_convertvector((vector signed long long)(__a),         \
+                                       vector double) *                        \
+               (vector double)(vector unsigned long long)((0x3ffULL -          \
+                                                           ((__b)&0x1F))       \
+                                                          << 52)))
 #endif // __VSX__
 
 /* vec_vcfsx */
@@ -3281,27 +3289,27 @@ static __inline__ vector double __ATTRS_o_ai vec_cpsgn(vector double __a,
 #ifdef __VSX__
 #ifdef __XL_COMPAT_ALTIVEC__
 #define vec_cts(__a, __b)                                                      \
-  _Generic((__a), vector float                                                 \
-           : (vector signed int)__builtin_altivec_vctsxs((vector float)(__a),  \
-                                                         (__b)),               \
-             vector double                                                     \
-           : __extension__({                                                   \
+  _Generic((__a),                                                              \
+      vector float: (vector signed int)__builtin_altivec_vctsxs(               \
+               (vector float)(__a), ((__b)&0x1F)),                             \
+      vector double: __extension__({                                           \
              vector double __ret =                                             \
                  (vector double)(__a) *                                        \
-                 (vector double)(vector unsigned long long)((0x3ffULL + (__b)) \
+                 (vector double)(vector unsigned long long)((0x3ffULL +        \
+                                                             ((__b)&0x1F))     \
                                                             << 52);            \
              (vector signed long long)__builtin_vsx_xvcvdpsxws(__ret);         \
            }))
 #else // __XL_COMPAT_ALTIVEC__
 #define vec_cts(__a, __b)                                                      \
-  _Generic((__a), vector float                                                 \
-           : (vector signed int)__builtin_altivec_vctsxs((vector float)(__a),  \
-                                                         (__b)),               \
-             vector double                                                     \
-           : __extension__({                                                   \
+  _Generic((__a),                                                              \
+      vector float: (vector signed int)__builtin_altivec_vctsxs(               \
+               (vector float)(__a), ((__b)&0x1F)),                             \
+      vector double: __extension__({                                           \
              vector double __ret =                                             \
                  (vector double)(__a) *                                        \
-                 (vector double)(vector unsigned long long)((0x3ffULL + (__b)) \
+                 (vector double)(vector unsigned long long)((0x3ffULL +        \
+                                                             ((__b)&0x1F))     \
                                                             << 52);            \
              (vector signed long long)__builtin_convertvector(                 \
                  __ret, vector signed long long);                              \
@@ -3320,27 +3328,27 @@ static __inline__ vector double __ATTRS_o_ai vec_cpsgn(vector double __a,
 #ifdef __VSX__
 #ifdef __XL_COMPAT_ALTIVEC__
 #define vec_ctu(__a, __b)                                                      \
-  _Generic((__a), vector float                                                 \
-           : (vector unsigned int)__builtin_altivec_vctuxs(                    \
-                 (vector float)(__a), (__b)),                                  \
-             vector double                                                     \
-           : __extension__({                                                   \
+  _Generic((__a),                                                              \
+      vector float: (vector unsigned int)__builtin_altivec_vctuxs(             \
+               (vector float)(__a), ((__b)&0x1F)),                             \
+      vector double: __extension__({                                           \
              vector double __ret =                                             \
                  (vector double)(__a) *                                        \
-                 (vector double)(vector unsigned long long)((0x3ffULL + __b)   \
+                 (vector double)(vector unsigned long long)((0x3ffULL +        \
+                                                             ((__b)&0x1F))     \
                                                             << 52);            \
              (vector unsigned long long)__builtin_vsx_xvcvdpuxws(__ret);       \
            }))
 #else // __XL_COMPAT_ALTIVEC__
 #define vec_ctu(__a, __b)                                                      \
-  _Generic((__a), vector float                                                 \
-           : (vector unsigned int)__builtin_altivec_vctuxs(                    \
-                 (vector float)(__a), (__b)),                                  \
-             vector double                                                     \
-           : __extension__({                                                   \
+  _Generic((__a),                                                              \
+      vector float: (vector unsigned int)__builtin_altivec_vctuxs(             \
+               (vector float)(__a), ((__b)&0x1F)),                             \
+      vector double: __extension__({                                           \
              vector double __ret =                                             \
                  (vector double)(__a) *                                        \
-                 (vector double)(vector unsigned long long)((0x3ffULL + __b)   \
+                 (vector double)(vector unsigned long long)((0x3ffULL +        \
+                                                             ((__b)&0x1F))     \
                                                             << 52);            \
              (vector unsigned long long)__builtin_convertvector(               \
                  __ret, vector unsigned long long);                            \
@@ -3355,60 +3363,62 @@ static __inline__ vector double __ATTRS_o_ai vec_cpsgn(vector double __a,
 
 #ifdef __VSX__
 #define vec_ctsl(__a, __b)                                                     \
-  _Generic((__a), vector float                                                 \
-           : __extension__({                                                   \
-               vector float __ret =                                            \
-                   (vector float)(__a) *                                       \
-                   (vector float)(vector unsigned)((0x7f + (__b)) << 23);      \
-               __builtin_vsx_xvcvspsxds(                                       \
-                   __builtin_vsx_xxsldwi(__ret, __ret, 1));                    \
-             }),                                                               \
-             vector double                                                     \
-           : __extension__({                                                   \
-             vector double __ret =                                             \
-                 (vector double)(__a) *                                        \
-                 (vector double)(vector unsigned long long)((0x3ffULL + __b)   \
-                                                            << 52);            \
-             __builtin_convertvector(__ret, vector signed long long);          \
-           }))
+  _Generic(                                                                    \
+      (__a), vector float                                                      \
+      : __extension__({                                                        \
+          vector float __ret =                                                 \
+              (vector float)(__a) *                                            \
+              (vector float)(vector unsigned)((0x7f + ((__b)&0x1F)) << 23);    \
+          __builtin_vsx_xvcvspsxds(__builtin_vsx_xxsldwi(__ret, __ret, 1));    \
+        }),                                                                    \
+        vector double                                                          \
+      : __extension__({                                                        \
+        vector double __ret =                                                  \
+            (vector double)(__a) *                                             \
+            (vector double)(vector unsigned long long)((0x3ffULL +             \
+                                                        ((__b)&0x1F))          \
+                                                       << 52);                 \
+        __builtin_convertvector(__ret, vector signed long long);               \
+      }))
 
 /* vec_ctul */
 
 #define vec_ctul(__a, __b)                                                     \
-  _Generic((__a), vector float                                                 \
-           : __extension__({                                                   \
-               vector float __ret =                                            \
-                   (vector float)(__a) *                                       \
-                   (vector float)(vector unsigned)((0x7f + (__b)) << 23);      \
-               __builtin_vsx_xvcvspuxds(                                       \
-                   __builtin_vsx_xxsldwi(__ret, __ret, 1));                    \
-             }),                                                               \
-             vector double                                                     \
-           : __extension__({                                                   \
-             vector double __ret =                                             \
-                 (vector double)(__a) *                                        \
-                 (vector double)(vector unsigned long long)((0x3ffULL + __b)   \
-                                                            << 52);            \
-             __builtin_convertvector(__ret, vector unsigned long long);        \
-           }))
+  _Generic(                                                                    \
+      (__a), vector float                                                      \
+      : __extension__({                                                        \
+          vector float __ret =                                                 \
+              (vector float)(__a) *                                            \
+              (vector float)(vector unsigned)((0x7f + ((__b)&0x1F)) << 23);    \
+          __builtin_vsx_xvcvspuxds(__builtin_vsx_xxsldwi(__ret, __ret, 1));    \
+        }),                                                                    \
+        vector double                                                          \
+      : __extension__({                                                        \
+        vector double __ret =                                                  \
+            (vector double)(__a) *                                             \
+            (vector double)(vector unsigned long long)((0x3ffULL +             \
+                                                        ((__b)&0x1F))          \
+                                                       << 52);                 \
+        __builtin_convertvector(__ret, vector unsigned long long);             \
+      }))
 #endif
 #else // __LITTLE_ENDIAN__
 /* vec_ctsl */
 
 #ifdef __VSX__
 #define vec_ctsl(__a, __b)                                                     \
-  _Generic((__a), vector float                                                 \
-           : __extension__({                                                   \
-               vector float __ret =                                            \
-                   (vector float)(__a) *                                       \
-                   (vector float)(vector unsigned)((0x7f + (__b)) << 23);      \
-               __builtin_vsx_xvcvspsxds(__ret);                                \
-             }),                                                               \
-             vector double                                                     \
-           : __extension__({                                                   \
+  _Generic((__a),                                                              \
+      vector float: __extension__({                                            \
+             vector float __ret =                                              \
+                 (vector float)(__a) *                                         \
+                 (vector float)(vector unsigned)((0x7f + ((__b)&0x1F)) << 23); \
+             __builtin_vsx_xvcvspsxds(__ret);                                  \
+           }),                                                                 \
+      vector double: __extension__({                                           \
              vector double __ret =                                             \
                  (vector double)(__a) *                                        \
-                 (vector double)(vector unsigned long long)((0x3ffULL + __b)   \
+                 (vector double)(vector unsigned long long)((0x3ffULL +        \
+                                                             ((__b)&0x1F))     \
                                                             << 52);            \
              __builtin_convertvector(__ret, vector signed long long);          \
            }))
@@ -3420,14 +3430,16 @@ static __inline__ vector double __ATTRS_o_ai vec_cpsgn(vector double __a,
            : __extension__({                                                   \
                vector float __ret =                                            \
                    (vector float)(__a) *                                       \
-                   (vector float)(vector unsigned)((0x7f + (__b)) << 23);      \
+                   (vector float)(vector unsigned)((0x7f + ((__b)&0x1F))       \
+                                                   << 23);                     \
                __builtin_vsx_xvcvspuxds(__ret);                                \
              }),                                                               \
              vector double                                                     \
            : __extension__({                                                   \
              vector double __ret =                                             \
                  (vector double)(__a) *                                        \
-                 (vector double)(vector unsigned long long)((0x3ffULL + __b)   \
+                 (vector double)(vector unsigned long long)((0x3ffULL +        \
+                                                             ((__b)&0x1F))     \
                                                             << 52);            \
              __builtin_convertvector(__ret, vector unsigned long long);        \
            }))
@@ -17323,32 +17335,32 @@ provided.
 #define vec_ncipherlast_be __builtin_altivec_crypto_vncipherlast
 
 #ifdef __VSX__
-static __inline__ vector unsigned long long __attribute__((__always_inline__))
-__builtin_crypto_vsbox(vector unsigned long long __a) {
+static __inline__ vector unsigned char __attribute__((__always_inline__))
+__builtin_crypto_vsbox(vector unsigned char __a) {
   return __builtin_altivec_crypto_vsbox(__a);
 }
 
-static __inline__ vector unsigned long long __attribute__((__always_inline__))
-__builtin_crypto_vcipher(vector unsigned long long __a,
-                         vector unsigned long long __b) {
+static __inline__ vector unsigned char __attribute__((__always_inline__))
+__builtin_crypto_vcipher(vector unsigned char __a,
+                         vector unsigned char __b) {
   return __builtin_altivec_crypto_vcipher(__a, __b);
 }
 
-static __inline__ vector unsigned long long __attribute__((__always_inline__))
-__builtin_crypto_vcipherlast(vector unsigned long long __a,
-                             vector unsigned long long __b) {
+static __inline__ vector unsigned char __attribute__((__always_inline__))
+__builtin_crypto_vcipherlast(vector unsigned char __a,
+                             vector unsigned char __b) {
   return __builtin_altivec_crypto_vcipherlast(__a, __b);
 }
 
-static __inline__ vector unsigned long long __attribute__((__always_inline__))
-__builtin_crypto_vncipher(vector unsigned long long __a,
-                          vector unsigned long long __b) {
+static __inline__ vector unsigned char __attribute__((__always_inline__))
+__builtin_crypto_vncipher(vector unsigned char __a,
+                          vector unsigned char __b) {
   return __builtin_altivec_crypto_vncipher(__a, __b);
 }
 
-static __inline__ vector unsigned long long __attribute__((__always_inline__))
-__builtin_crypto_vncipherlast(vector unsigned long long __a,
-                              vector unsigned long long __b) {
+static __inline__ vector unsigned char  __attribute__((__always_inline__))
+__builtin_crypto_vncipherlast(vector unsigned char __a,
+                              vector unsigned char __b) {
   return __builtin_altivec_crypto_vncipherlast(__a, __b);
 }
 #endif /* __VSX__ */

package/framework/internal/metadata-generator-arm64/bin/lib/clang/17/include/amxcomplexintrin.h

@@ -0,0 +1,169 @@
+/*===--------- amxcomplexintrin.h - AMXCOMPLEX intrinsics -*- C++ -*---------===
+ *
+ * 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 <amxcomplexintrin.h> directly; include <immintrin.h> instead."
+#endif // __IMMINTRIN_H
+
+#ifndef __AMX_COMPLEXINTRIN_H
+#define __AMX_COMPLEXINTRIN_H
+#ifdef __x86_64__
+
+#define __DEFAULT_FN_ATTRS_COMPLEX                                             \
+  __attribute__((__always_inline__, __nodebug__, __target__("amx-complex")))
+
+/// Perform matrix multiplication of two tiles containing complex elements and
+///    accumulate the results into a packed single precision tile. Each dword
+///    element in input tiles \a a and \a b is interpreted as a complex number
+///    with FP16 real part and FP16 imaginary part.
+/// Calculates the imaginary part of the result. For each possible combination
+///    of (row of \a a, column of \a b), it performs a set of multiplication
+///    and accumulations on all corresponding complex numbers (one from \a a
+///    and one from \a b). The imaginary part of the \a a element is multiplied
+///    with the real part of the corresponding \a b element, and the real part
+///    of the \a a element is multiplied with the imaginary part of the
+///    corresponding \a b elements. The two accumulated results are added, and
+///    then accumulated into the corresponding row and column of \a dst.
+///
+/// \headerfile <x86intrin.h>
+///
+/// \code
+/// void _tile_cmmimfp16ps(__tile dst, __tile a, __tile b);
+/// \endcode
+///
+/// \code{.operation}
+/// FOR m := 0 TO dst.rows - 1
+///	tmp := dst.row[m]
+///	FOR k := 0 TO (a.colsb / 4) - 1
+///		FOR n := 0 TO (dst.colsb / 4) - 1
+///			tmp.fp32[n] += FP32(a.row[m].fp16[2*k+0]) * FP32(b.row[k].fp16[2*n+1])
+///			tmp.fp32[n] += FP32(a.row[m].fp16[2*k+1]) * FP32(b.row[k].fp16[2*n+0])
+///		ENDFOR
+///	ENDFOR
+///	write_row_and_zero(dst, m, tmp, dst.colsb)
+/// ENDFOR
+/// zero_upper_rows(dst, dst.rows)
+/// zero_tileconfig_start()
+/// \endcode
+///
+/// This intrinsic corresponds to the \c TCMMIMFP16PS instruction.
+///
+/// \param dst
+///    The destination tile. Max size is 1024 Bytes.
+/// \param a
+///    The 1st source tile. Max size is 1024 Bytes.
+/// \param b
+///    The 2nd source tile. Max size is 1024 Bytes.
+#define _tile_cmmimfp16ps(dst, a, b) __builtin_ia32_tcmmimfp16ps(dst, a, b)
+
+/// Perform matrix multiplication of two tiles containing complex elements and
+///    accumulate the results into a packed single precision tile. Each dword
+///    element in input tiles \a a and \a b is interpreted as a complex number
+///    with FP16 real part and FP16 imaginary part.
+/// Calculates the real part of the result. For each possible combination
+///    of (row of \a a, column of \a b), it performs a set of multiplication
+///    and accumulations on all corresponding complex numbers (one from \a a
+///    and one from \a b). The real part of the \a a element is multiplied
+///    with the real part of the corresponding \a b element, and the negated
+///    imaginary part of the \a a element is multiplied with the imaginary
+///    part of the corresponding \a b elements. The two accumulated results
+///    are added, and then accumulated into the corresponding row and column
+///    of \a dst.
+///
+/// \headerfile <x86intrin.h>
+///
+/// \code
+/// void _tile_cmmrlfp16ps(__tile dst, __tile a, __tile b);
+/// \endcode
+///
+/// \code{.operation}
+/// FOR m := 0 TO dst.rows - 1
+///	tmp := dst.row[m]
+///	FOR k := 0 TO (a.colsb / 4) - 1
+///		FOR n := 0 TO (dst.colsb / 4) - 1
+///			tmp.fp32[n] += FP32(a.row[m].fp16[2*k+0]) * FP32(b.row[k].fp16[2*n+0])
+///			tmp.fp32[n] += FP32(-a.row[m].fp16[2*k+1]) * FP32(b.row[k].fp16[2*n+1])
+///		ENDFOR
+///	ENDFOR
+///	write_row_and_zero(dst, m, tmp, dst.colsb)
+/// ENDFOR
+/// zero_upper_rows(dst, dst.rows)
+/// zero_tileconfig_start()
+/// \endcode
+///
+/// This intrinsic corresponds to the \c TCMMIMFP16PS instruction.
+///
+/// \param dst
+///    The destination tile. Max size is 1024 Bytes.
+/// \param a
+///    The 1st source tile. Max size is 1024 Bytes.
+/// \param b
+///    The 2nd source tile. Max size is 1024 Bytes.
+#define _tile_cmmrlfp16ps(dst, a, b) __builtin_ia32_tcmmrlfp16ps(dst, a, b)
+
+static __inline__ _tile1024i __DEFAULT_FN_ATTRS_COMPLEX
+_tile_cmmimfp16ps_internal(unsigned short m, unsigned short n, unsigned short k,
+                           _tile1024i dst, _tile1024i src1, _tile1024i src2) {
+  return __builtin_ia32_tcmmimfp16ps_internal(m, n, k, dst, src1, src2);
+}
+
+static __inline__ _tile1024i __DEFAULT_FN_ATTRS_COMPLEX
+_tile_cmmrlfp16ps_internal(unsigned short m, unsigned short n, unsigned short k,
+                           _tile1024i dst, _tile1024i src1, _tile1024i src2) {
+  return __builtin_ia32_tcmmrlfp16ps_internal(m, n, k, dst, src1, src2);
+}
+
+/// Perform matrix multiplication of two tiles containing complex elements and
+/// accumulate the results into a packed single precision tile. Each dword
+/// element in input tiles src0 and src1 is interpreted as a complex number with
+/// FP16 real part and FP16 imaginary part.
+/// This function calculates the imaginary part of the result.
+///
+/// \headerfile <immintrin.h>
+///
+/// This intrinsic corresponds to the <c> TCMMIMFP16PS </c> instruction.
+///
+/// \param dst
+///    The destination tile. Max size is 1024 Bytes.
+/// \param src0
+///    The 1st source tile. Max size is 1024 Bytes.
+/// \param src1
+///    The 2nd source tile. Max size is 1024 Bytes.
+__DEFAULT_FN_ATTRS_COMPLEX
+static void __tile_cmmimfp16ps(__tile1024i *dst, __tile1024i src0,
+                               __tile1024i src1) {
+  dst->tile = _tile_cmmimfp16ps_internal(src0.row, src1.col, src0.col,
+                                         dst->tile, src0.tile, src1.tile);
+}
+
+/// Perform matrix multiplication of two tiles containing complex elements and
+/// accumulate the results into a packed single precision tile. Each dword
+/// element in input tiles src0 and src1 is interpreted as a complex number with
+/// FP16 real part and FP16 imaginary part.
+/// This function calculates the real part of the result.
+///
+/// \headerfile <immintrin.h>
+///
+/// This intrinsic corresponds to the <c> TCMMRLFP16PS </c> instruction.
+///
+/// \param dst
+///    The destination tile. Max size is 1024 Bytes.
+/// \param src0
+///    The 1st source tile. Max size is 1024 Bytes.
+/// \param src1
+///    The 2nd source tile. Max size is 1024 Bytes.
+__DEFAULT_FN_ATTRS_COMPLEX
+static void __tile_cmmrlfp16ps(__tile1024i *dst, __tile1024i src0,
+                               __tile1024i src1) {
+  dst->tile = _tile_cmmrlfp16ps_internal(src0.row, src1.col, src0.col,
+                                         dst->tile, src0.tile, src1.tile);
+}
+
+#endif // __x86_64__
+#endif // __AMX_COMPLEXINTRIN_H

package/framework/internal/metadata-generator-arm64/bin/lib/clang/17/include/amxfp16intrin.h

@@ -0,0 +1,58 @@
+/*===------------- amxfp16intrin.h - AMX_FP16 intrinsics -*- C++ -*---------===
+ *
+ * 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 <amxfp16intrin.h> directly; use <immintrin.h> instead."
+#endif /* __IMMINTRIN_H */
+
+#ifndef __AMX_FP16INTRIN_H
+#define __AMX_FP16INTRIN_H
+#ifdef __x86_64__
+
+/// Compute dot-product of FP16 (16-bit) floating-point pairs in tiles \a a
+///    and \a b, accumulating the intermediate single-precision (32-bit)
+///    floating-point elements with elements in \a dst, and store the 32-bit
+///    result back to tile \a dst.
+///
+/// \headerfile <immintrin.h>
+///
+/// \code
+/// void _tile_dpfp16ps (__tile dst, __tile a, __tile b)
+/// \endcode
+///
+/// \code{.operation}
+/// FOR m := 0 TO dst.rows - 1
+///	tmp := dst.row[m]
+///	FOR k := 0 TO (a.colsb / 4) - 1
+///		FOR n := 0 TO (dst.colsb / 4) - 1
+///			tmp.fp32[n] += FP32(a.row[m].fp16[2*k+0]) *
+///					FP32(b.row[k].fp16[2*n+0])
+///			tmp.fp32[n] += FP32(a.row[m].fp16[2*k+1]) *
+///					FP32(b.row[k].fp16[2*n+1])
+///		ENDFOR
+///	ENDFOR
+///	write_row_and_zero(dst, m, tmp, dst.colsb)
+/// ENDFOR
+/// zero_upper_rows(dst, dst.rows)
+/// zero_tileconfig_start()
+/// \endcode
+///
+/// This intrinsic corresponds to the \c TDPFP16PS instruction.
+///
+/// \param dst
+///    The destination tile. Max size is 1024 Bytes.
+/// \param a
+///    The 1st source tile. Max size is 1024 Bytes.
+/// \param b
+///    The 2nd source tile. Max size is 1024 Bytes.
+#define _tile_dpfp16ps(dst, a, b)                                \
+  __builtin_ia32_tdpfp16ps(dst, a, b)
+
+#endif /* __x86_64__ */
+#endif /* __AMX_FP16INTRIN_H */