zig cc: update intrinsic headers to LLVM 21

lib/include/__clang_spirv_builtins.h

@@ -0,0 +1,217 @@
+/*===---- spirv_builtin_vars.h - SPIR-V built-in ---------------------------===
+ *
+ * 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 __SPIRV_BUILTIN_VARS_H
+#define __SPIRV_BUILTIN_VARS_H
+
+#if __cplusplus >= 201103L
+#define __SPIRV_NOEXCEPT noexcept
+#else
+#define __SPIRV_NOEXCEPT
+#endif
+
+#pragma push_macro("__size_t")
+#pragma push_macro("__uint32_t")
+#pragma push_macro("__uint64_t")
+#define __size_t __SIZE_TYPE__
+#define __uint32_t __UINT32_TYPE__
+
+#define __SPIRV_overloadable __attribute__((overloadable))
+#define __SPIRV_convergent __attribute__((convergent))
+#define __SPIRV_inline __attribute__((always_inline))
+
+#define __global __attribute__((opencl_global))
+#define __local __attribute__((opencl_local))
+#define __private __attribute__((opencl_private))
+#define __constant __attribute__((opencl_constant))
+#ifdef __SYCL_DEVICE_ONLY__
+#define __generic
+#else
+#define __generic __attribute__((opencl_generic))
+#endif
+
+// Check if SPIR-V builtins are supported.
+// As the translator doesn't use the LLVM intrinsics (which would be emitted if
+// we use the SPIR-V builtins) we can't rely on the SPIRV32/SPIRV64 etc macros
+// to establish if we can use the builtin alias. We disable builtin altogether
+// if we do not intent to use the backend. So instead of use target macros, rely
+// on a __has_builtin test.
+#if (__has_builtin(__builtin_spirv_num_workgroups))
+#define __SPIRV_BUILTIN_ALIAS(builtin)                                         \
+  __attribute__((clang_builtin_alias(builtin)))
+#else
+#define __SPIRV_BUILTIN_ALIAS(builtin)
+#endif
+
+// Builtin IDs and sizes
+
+extern __SPIRV_BUILTIN_ALIAS(__builtin_spirv_num_workgroups) __size_t
+    __spirv_NumWorkgroups(int);
+extern __SPIRV_BUILTIN_ALIAS(__builtin_spirv_workgroup_size) __size_t
+    __spirv_WorkgroupSize(int);
+extern __SPIRV_BUILTIN_ALIAS(__builtin_spirv_workgroup_id) __size_t
+    __spirv_WorkgroupId(int);
+extern __SPIRV_BUILTIN_ALIAS(__builtin_spirv_local_invocation_id) __size_t
+    __spirv_LocalInvocationId(int);
+extern __SPIRV_BUILTIN_ALIAS(__builtin_spirv_global_invocation_id) __size_t
+    __spirv_GlobalInvocationId(int);
+
+extern __SPIRV_BUILTIN_ALIAS(__builtin_spirv_global_size) __size_t
+    __spirv_GlobalSize(int);
+extern __SPIRV_BUILTIN_ALIAS(__builtin_spirv_global_offset) __size_t
+    __spirv_GlobalOffset(int);
+extern __SPIRV_BUILTIN_ALIAS(__builtin_spirv_subgroup_size) __uint32_t
+    __spirv_SubgroupSize();
+extern __SPIRV_BUILTIN_ALIAS(__builtin_spirv_subgroup_max_size) __uint32_t
+    __spirv_SubgroupMaxSize();
+extern __SPIRV_BUILTIN_ALIAS(__builtin_spirv_num_subgroups) __uint32_t
+    __spirv_NumSubgroups();
+extern __SPIRV_BUILTIN_ALIAS(__builtin_spirv_subgroup_id) __uint32_t
+    __spirv_SubgroupId();
+extern __SPIRV_BUILTIN_ALIAS(__builtin_spirv_subgroup_local_invocation_id)
+    __uint32_t __spirv_SubgroupLocalInvocationId();
+
+// OpGenericCastToPtrExplicit
+
+extern __SPIRV_overloadable
+__SPIRV_BUILTIN_ALIAS(__builtin_spirv_generic_cast_to_ptr_explicit)
+__global void *__spirv_GenericCastToPtrExplicit_ToGlobal(__generic void *,
+                                                         int) __SPIRV_NOEXCEPT;
+extern __SPIRV_overloadable
+__SPIRV_BUILTIN_ALIAS(__builtin_spirv_generic_cast_to_ptr_explicit)
+__global const void *
+__spirv_GenericCastToPtrExplicit_ToGlobal(__generic const void *,
+                                          int) __SPIRV_NOEXCEPT;
+extern __SPIRV_overloadable
+__SPIRV_BUILTIN_ALIAS(__builtin_spirv_generic_cast_to_ptr_explicit)
+__global volatile void *
+__spirv_GenericCastToPtrExplicit_ToGlobal(__generic volatile void *,
+                                          int) __SPIRV_NOEXCEPT;
+extern __SPIRV_overloadable
+__SPIRV_BUILTIN_ALIAS(__builtin_spirv_generic_cast_to_ptr_explicit)
+__global const volatile void *
+__spirv_GenericCastToPtrExplicit_ToGlobal(__generic const volatile void *,
+                                          int) __SPIRV_NOEXCEPT;
+extern __SPIRV_overloadable
+__SPIRV_BUILTIN_ALIAS(__builtin_spirv_generic_cast_to_ptr_explicit)
+__local void *__spirv_GenericCastToPtrExplicit_ToLocal(__generic void *,
+                                                       int) __SPIRV_NOEXCEPT;
+extern __SPIRV_overloadable
+__SPIRV_BUILTIN_ALIAS(__builtin_spirv_generic_cast_to_ptr_explicit)
+__local const void *
+__spirv_GenericCastToPtrExplicit_ToLocal(__generic const void *,
+                                         int) __SPIRV_NOEXCEPT;
+extern __SPIRV_overloadable
+__SPIRV_BUILTIN_ALIAS(__builtin_spirv_generic_cast_to_ptr_explicit)
+__local volatile void *
+__spirv_GenericCastToPtrExplicit_ToLocal(__generic volatile void *,
+                                         int) __SPIRV_NOEXCEPT;
+extern __SPIRV_overloadable
+__SPIRV_BUILTIN_ALIAS(__builtin_spirv_generic_cast_to_ptr_explicit)
+__local const volatile void *
+__spirv_GenericCastToPtrExplicit_ToLocal(__generic const volatile void *,
+                                         int) __SPIRV_NOEXCEPT;
+extern __SPIRV_overloadable
+__SPIRV_BUILTIN_ALIAS(__builtin_spirv_generic_cast_to_ptr_explicit)
+__private void *
+__spirv_GenericCastToPtrExplicit_ToPrivate(__generic void *,
+                                           int) __SPIRV_NOEXCEPT;
+extern __SPIRV_overloadable
+__SPIRV_BUILTIN_ALIAS(__builtin_spirv_generic_cast_to_ptr_explicit)
+__private const void *
+__spirv_GenericCastToPtrExplicit_ToPrivate(__generic const void *,
+                                           int) __SPIRV_NOEXCEPT;
+extern __SPIRV_overloadable
+__SPIRV_BUILTIN_ALIAS(__builtin_spirv_generic_cast_to_ptr_explicit)
+__private volatile void *
+__spirv_GenericCastToPtrExplicit_ToPrivate(__generic volatile void *,
+                                           int) __SPIRV_NOEXCEPT;
+extern __SPIRV_overloadable
+__SPIRV_BUILTIN_ALIAS(__builtin_spirv_generic_cast_to_ptr_explicit)
+__private const volatile void *
+__spirv_GenericCastToPtrExplicit_ToPrivate(__generic const volatile void *,
+                                           int) __SPIRV_NOEXCEPT;
+
+// OpGenericCastToPtr
+
+static __SPIRV_overloadable __SPIRV_inline __global void *
+__spirv_GenericCastToPtr_ToGlobal(__generic void *p, int) __SPIRV_NOEXCEPT {
+  return (__global void *)p;
+}
+static __SPIRV_overloadable __SPIRV_inline __global const void *
+__spirv_GenericCastToPtr_ToGlobal(__generic const void *p,
+                                  int) __SPIRV_NOEXCEPT {
+  return (__global const void *)p;
+}
+static __SPIRV_overloadable __SPIRV_inline __global volatile void *
+__spirv_GenericCastToPtr_ToGlobal(__generic volatile void *p,
+                                  int) __SPIRV_NOEXCEPT {
+  return (__global volatile void *)p;
+}
+static __SPIRV_overloadable __SPIRV_inline __global const volatile void *
+__spirv_GenericCastToPtr_ToGlobal(__generic const volatile void *p,
+                                  int) __SPIRV_NOEXCEPT {
+  return (__global const volatile void *)p;
+}
+static __SPIRV_overloadable __SPIRV_inline __local void *
+__spirv_GenericCastToPtr_ToLocal(__generic void *p, int) __SPIRV_NOEXCEPT {
+  return (__local void *)p;
+}
+static __SPIRV_overloadable __SPIRV_inline __local const void *
+__spirv_GenericCastToPtr_ToLocal(__generic const void *p,
+                                 int) __SPIRV_NOEXCEPT {
+  return (__local const void *)p;
+}
+static __SPIRV_overloadable __SPIRV_inline __local volatile void *
+__spirv_GenericCastToPtr_ToLocal(__generic volatile void *p,
+                                 int) __SPIRV_NOEXCEPT {
+  return (__local volatile void *)p;
+}
+static __SPIRV_overloadable __SPIRV_inline __local const volatile void *
+__spirv_GenericCastToPtr_ToLocal(__generic const volatile void *p,
+                                 int) __SPIRV_NOEXCEPT {
+  return (__local const volatile void *)p;
+}
+static __SPIRV_overloadable __SPIRV_inline __private void *
+__spirv_GenericCastToPtr_ToPrivate(__generic void *p, int) __SPIRV_NOEXCEPT {
+  return (__private void *)p;
+}
+static __SPIRV_overloadable __SPIRV_inline __private const void *
+__spirv_GenericCastToPtr_ToPrivate(__generic const void *p,
+                                   int) __SPIRV_NOEXCEPT {
+  return (__private const void *)p;
+}
+static __SPIRV_overloadable __SPIRV_inline __private volatile void *
+__spirv_GenericCastToPtr_ToPrivate(__generic volatile void *p,
+                                   int) __SPIRV_NOEXCEPT {
+  return (__private volatile void *)p;
+}
+static __SPIRV_overloadable __SPIRV_inline __private const volatile void *
+__spirv_GenericCastToPtr_ToPrivate(__generic const volatile void *p,
+                                   int) __SPIRV_NOEXCEPT {
+  return (__private const volatile void *)p;
+}
+
+#pragma pop_macro("__size_t")
+#pragma pop_macro("__uint32_t")
+#pragma pop_macro("__uint64_t")
+
+#undef __SPIRV_overloadable
+#undef __SPIRV_convergent
+#undef __SPIRV_inline
+
+#undef __global
+#undef __local
+#undef __constant
+#undef __generic
+
+#undef __SPIRV_BUILTIN_ALIAS
+#undef __SPIRV_NOEXCEPT
+
+#endif /* __SPIRV_BUILTIN_VARS_H */

lib/include/__stdarg_va_arg.h

@@ -10,8 +10,8 @@
 #ifndef va_arg
 
 #if defined(__STDC_VERSION__) && __STDC_VERSION__ >= 202311L
-/* C23 does not require the second parameter for va_start. */
-#define va_start(ap, ...) __builtin_va_start(ap, 0)
+/* C23 uses a special builtin. */
+#define va_start(...) __builtin_c23_va_start(__VA_ARGS__)
 #else
 /* Versions before C23 do require the second parameter. */
 #define va_start(ap, param) __builtin_va_start(ap, param)

lib/include/altivec.h

@@ -17525,70 +17525,73 @@ vec_bperm(vector unsigned long long __a, vector unsigned char __b) {
 
 /* vec_reve */
 
-static inline __ATTRS_o_ai vector bool char vec_reve(vector bool char __a) {
+static __inline__ __ATTRS_o_ai vector bool char vec_reve(vector bool char __a) {
   return __builtin_shufflevector(__a, __a, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6,
                                  5, 4, 3, 2, 1, 0);
 }
 
-static inline __ATTRS_o_ai vector signed char vec_reve(vector signed char __a) {
+static __inline__ __ATTRS_o_ai vector signed char
+vec_reve(vector signed char __a) {
   return __builtin_shufflevector(__a, __a, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6,
                                  5, 4, 3, 2, 1, 0);
 }
 
-static inline __ATTRS_o_ai vector unsigned char
+static __inline__ __ATTRS_o_ai vector unsigned char
 vec_reve(vector unsigned char __a) {
   return __builtin_shufflevector(__a, __a, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6,
                                  5, 4, 3, 2, 1, 0);
 }
 
-static inline __ATTRS_o_ai vector bool int vec_reve(vector bool int __a) {
+static __inline__ __ATTRS_o_ai vector bool int vec_reve(vector bool int __a) {
   return __builtin_shufflevector(__a, __a, 3, 2, 1, 0);
 }
 
-static inline __ATTRS_o_ai vector signed int vec_reve(vector signed int __a) {
+static __inline__ __ATTRS_o_ai vector signed int
+vec_reve(vector signed int __a) {
   return __builtin_shufflevector(__a, __a, 3, 2, 1, 0);
 }
 
-static inline __ATTRS_o_ai vector unsigned int
+static __inline__ __ATTRS_o_ai vector unsigned int
 vec_reve(vector unsigned int __a) {
   return __builtin_shufflevector(__a, __a, 3, 2, 1, 0);
 }
 
-static inline __ATTRS_o_ai vector bool short vec_reve(vector bool short __a) {
+static __inline__ __ATTRS_o_ai vector bool short
+vec_reve(vector bool short __a) {
   return __builtin_shufflevector(__a, __a, 7, 6, 5, 4, 3, 2, 1, 0);
 }
 
-static inline __ATTRS_o_ai vector signed short
+static __inline__ __ATTRS_o_ai vector signed short
 vec_reve(vector signed short __a) {
   return __builtin_shufflevector(__a, __a, 7, 6, 5, 4, 3, 2, 1, 0);
 }
 
-static inline __ATTRS_o_ai vector unsigned short
+static __inline__ __ATTRS_o_ai vector unsigned short
 vec_reve(vector unsigned short __a) {
   return __builtin_shufflevector(__a, __a, 7, 6, 5, 4, 3, 2, 1, 0);
 }
 
-static inline __ATTRS_o_ai vector float vec_reve(vector float __a) {
+static __inline__ __ATTRS_o_ai vector float vec_reve(vector float __a) {
   return __builtin_shufflevector(__a, __a, 3, 2, 1, 0);
 }
 
 #ifdef __VSX__
-static inline __ATTRS_o_ai vector bool long long
+static __inline__ __ATTRS_o_ai vector bool long long
 vec_reve(vector bool long long __a) {
   return __builtin_shufflevector(__a, __a, 1, 0);
 }
 
-static inline __ATTRS_o_ai vector signed long long
+static __inline__ __ATTRS_o_ai vector signed long long
 vec_reve(vector signed long long __a) {
   return __builtin_shufflevector(__a, __a, 1, 0);
 }
 
-static inline __ATTRS_o_ai vector unsigned long long
+static __inline__ __ATTRS_o_ai vector unsigned long long
 vec_reve(vector unsigned long long __a) {
   return __builtin_shufflevector(__a, __a, 1, 0);
 }
 
-static inline __ATTRS_o_ai vector double vec_reve(vector double __a) {
+static __inline__ __ATTRS_o_ai vector double vec_reve(vector double __a) {
   return __builtin_shufflevector(__a, __a, 1, 0);
 }
 #endif
@@ -17721,42 +17724,42 @@ typedef vector signed int unaligned_vec_sint __attribute__((aligned(1)));
 typedef vector unsigned int unaligned_vec_uint __attribute__((aligned(1)));
 typedef vector float unaligned_vec_float __attribute__((aligned(1)));
 
-static inline __ATTRS_o_ai vector signed char vec_xl(ptrdiff_t __offset,
-                                                     const signed char *__ptr) {
+static __inline__ __ATTRS_o_ai vector signed char
+vec_xl(ptrdiff_t __offset, const signed char *__ptr) {
   return *(unaligned_vec_schar *)(__ptr + __offset);
 }
 
-static inline __ATTRS_o_ai vector unsigned char
+static __inline__ __ATTRS_o_ai vector unsigned char
 vec_xl(ptrdiff_t __offset, const unsigned char *__ptr) {
   return *(unaligned_vec_uchar*)(__ptr + __offset);
 }
 
-static inline __ATTRS_o_ai vector signed short
+static __inline__ __ATTRS_o_ai vector signed short
 vec_xl(ptrdiff_t __offset, const signed short *__ptr) {
   signed char *__addr = (signed char *)__ptr + __offset;
   return *(unaligned_vec_sshort *)__addr;
 }
 
-static inline __ATTRS_o_ai vector unsigned short
+static __inline__ __ATTRS_o_ai vector unsigned short
 vec_xl(ptrdiff_t __offset, const unsigned short *__ptr) {
   signed char *__addr = (signed char *)__ptr + __offset;
   return *(unaligned_vec_ushort *)__addr;
 }
 
-static inline __ATTRS_o_ai vector signed int vec_xl(ptrdiff_t __offset,
-                                                    const signed int *__ptr) {
+static __inline__ __ATTRS_o_ai vector signed int
+vec_xl(ptrdiff_t __offset, const signed int *__ptr) {
   signed char *__addr = (signed char *)__ptr + __offset;
   return *(unaligned_vec_sint *)__addr;
 }
 
-static inline __ATTRS_o_ai vector unsigned int
+static __inline__ __ATTRS_o_ai vector unsigned int
 vec_xl(ptrdiff_t __offset, const unsigned int *__ptr) {
   signed char *__addr = (signed char *)__ptr + __offset;
   return *(unaligned_vec_uint *)__addr;
 }
 
-static inline __ATTRS_o_ai vector float vec_xl(ptrdiff_t __offset,
-                                               const float *__ptr) {
+static __inline__ __ATTRS_o_ai vector float vec_xl(ptrdiff_t __offset,
+                                                   const float *__ptr) {
   signed char *__addr = (signed char *)__ptr + __offset;
   return *(unaligned_vec_float *)__addr;
 }
@@ -17766,20 +17769,20 @@ typedef vector signed long long unaligned_vec_sll __attribute__((aligned(1)));
 typedef vector unsigned long long unaligned_vec_ull __attribute__((aligned(1)));
 typedef vector double unaligned_vec_double __attribute__((aligned(1)));
 
-static inline __ATTRS_o_ai vector signed long long
+static __inline__ __ATTRS_o_ai vector signed long long
 vec_xl(ptrdiff_t __offset, const signed long long *__ptr) {
   signed char *__addr = (signed char *)__ptr + __offset;
   return *(unaligned_vec_sll *)__addr;
 }
 
-static inline __ATTRS_o_ai vector unsigned long long
+static __inline__ __ATTRS_o_ai vector unsigned long long
 vec_xl(ptrdiff_t __offset, const unsigned long long *__ptr) {
   signed char *__addr = (signed char *)__ptr + __offset;
   return *(unaligned_vec_ull *)__addr;
 }
 
-static inline __ATTRS_o_ai vector double vec_xl(ptrdiff_t __offset,
-                                                const double *__ptr) {
+static __inline__ __ATTRS_o_ai vector double vec_xl(ptrdiff_t __offset,
+                                                    const double *__ptr) {
   signed char *__addr = (signed char *)__ptr + __offset;
   return *(unaligned_vec_double *)__addr;
 }
@@ -17790,13 +17793,13 @@ static inline __ATTRS_o_ai vector double vec_xl(ptrdiff_t __offset,
 typedef vector signed __int128 unaligned_vec_si128 __attribute__((aligned(1)));
 typedef vector unsigned __int128 unaligned_vec_ui128
     __attribute__((aligned(1)));
-static inline __ATTRS_o_ai vector signed __int128
+static __inline__ __ATTRS_o_ai vector signed __int128
 vec_xl(ptrdiff_t __offset, const signed __int128 *__ptr) {
   signed char *__addr = (signed char *)__ptr + __offset;
   return *(unaligned_vec_si128 *)__addr;
 }
 
-static inline __ATTRS_o_ai vector unsigned __int128
+static __inline__ __ATTRS_o_ai vector unsigned __int128
 vec_xl(ptrdiff_t __offset, const unsigned __int128 *__ptr) {
   signed char *__addr = (signed char *)__ptr + __offset;
   return *(unaligned_vec_ui128 *)__addr;
@@ -17991,64 +17994,64 @@ vec_load_splats(unsigned long long __offset, const float *__ptr) {
 
 #define vec_xstd2 vec_xst
 #define vec_xstw4 vec_xst
-static inline __ATTRS_o_ai void
+static __inline__ __ATTRS_o_ai void
 vec_xst(vector signed char __vec, ptrdiff_t __offset, signed char *__ptr) {
   *(unaligned_vec_schar *)(__ptr + __offset) = __vec;
 }
 
-static inline __ATTRS_o_ai void
+static __inline__ __ATTRS_o_ai void
 vec_xst(vector unsigned char __vec, ptrdiff_t __offset, unsigned char *__ptr) {
   *(unaligned_vec_uchar *)(__ptr + __offset) = __vec;
 }
 
-static inline __ATTRS_o_ai void
+static __inline__ __ATTRS_o_ai void
 vec_xst(vector signed short __vec, ptrdiff_t __offset, signed short *__ptr) {
   signed char *__addr = (signed char *)__ptr + __offset;
   *(unaligned_vec_sshort *)__addr = __vec;
 }
 
-static inline __ATTRS_o_ai void vec_xst(vector unsigned short __vec,
-                                        ptrdiff_t __offset,
-                                        unsigned short *__ptr) {
+static __inline__ __ATTRS_o_ai void vec_xst(vector unsigned short __vec,
+                                            ptrdiff_t __offset,
+                                            unsigned short *__ptr) {
   signed char *__addr = (signed char *)__ptr + __offset;
   *(unaligned_vec_ushort *)__addr = __vec;
 }
 
-static inline __ATTRS_o_ai void vec_xst(vector signed int __vec,
-                                        ptrdiff_t __offset, signed int *__ptr) {
+static __inline__ __ATTRS_o_ai void
+vec_xst(vector signed int __vec, ptrdiff_t __offset, signed int *__ptr) {
   signed char *__addr = (signed char *)__ptr + __offset;
   *(unaligned_vec_sint *)__addr = __vec;
 }
 
-static inline __ATTRS_o_ai void
+static __inline__ __ATTRS_o_ai void
 vec_xst(vector unsigned int __vec, ptrdiff_t __offset, unsigned int *__ptr) {
   signed char *__addr = (signed char *)__ptr + __offset;
   *(unaligned_vec_uint *)__addr = __vec;
 }
 
-static inline __ATTRS_o_ai void vec_xst(vector float __vec, ptrdiff_t __offset,
-                                        float *__ptr) {
+static __inline__ __ATTRS_o_ai void vec_xst(vector float __vec,
+                                            ptrdiff_t __offset, float *__ptr) {
   signed char *__addr = (signed char *)__ptr + __offset;
   *(unaligned_vec_float *)__addr = __vec;
 }
 
 #ifdef __VSX__
-static inline __ATTRS_o_ai void vec_xst(vector signed long long __vec,
-                                        ptrdiff_t __offset,
-                                        signed long long *__ptr) {
+static __inline__ __ATTRS_o_ai void vec_xst(vector signed long long __vec,
+                                            ptrdiff_t __offset,
+                                            signed long long *__ptr) {
   signed char *__addr = (signed char *)__ptr + __offset;
   *(unaligned_vec_sll *)__addr = __vec;
 }
 
-static inline __ATTRS_o_ai void vec_xst(vector unsigned long long __vec,
-                                        ptrdiff_t __offset,
-                                        unsigned long long *__ptr) {
+static __inline__ __ATTRS_o_ai void vec_xst(vector unsigned long long __vec,
+                                            ptrdiff_t __offset,
+                                            unsigned long long *__ptr) {
   signed char *__addr = (signed char *)__ptr + __offset;
   *(unaligned_vec_ull *)__addr = __vec;
 }
 
-static inline __ATTRS_o_ai void vec_xst(vector double __vec, ptrdiff_t __offset,
-                                        double *__ptr) {
+static __inline__ __ATTRS_o_ai void vec_xst(vector double __vec,
+                                            ptrdiff_t __offset, double *__ptr) {
   signed char *__addr = (signed char *)__ptr + __offset;
   *(unaligned_vec_double *)__addr = __vec;
 }
@@ -18056,16 +18059,16 @@ static inline __ATTRS_o_ai void vec_xst(vector double __vec, ptrdiff_t __offset,
 
 #if defined(__POWER8_VECTOR__) && defined(__powerpc64__) &&                    \
     defined(__SIZEOF_INT128__)
-static inline __ATTRS_o_ai void vec_xst(vector signed __int128 __vec,
-                                        ptrdiff_t __offset,
-                                        signed __int128 *__ptr) {
+static __inline__ __ATTRS_o_ai void vec_xst(vector signed __int128 __vec,
+                                            ptrdiff_t __offset,
+                                            signed __int128 *__ptr) {
   signed char *__addr = (signed char *)__ptr + __offset;
   *(unaligned_vec_si128 *)__addr = __vec;
 }
 
-static inline __ATTRS_o_ai void vec_xst(vector unsigned __int128 __vec,
-                                        ptrdiff_t __offset,
-                                        unsigned __int128 *__ptr) {
+static __inline__ __ATTRS_o_ai void vec_xst(vector unsigned __int128 __vec,
+                                            ptrdiff_t __offset,
+                                            unsigned __int128 *__ptr) {
   signed char *__addr = (signed char *)__ptr + __offset;
   *(unaligned_vec_ui128 *)__addr = __vec;
 }
@@ -18075,51 +18078,51 @@ static inline __ATTRS_o_ai void vec_xst(vector unsigned __int128 __vec,
 
 #if defined(__POWER10_VECTOR__) && defined(__VSX__) &&                         \
     defined(__SIZEOF_INT128__)
-static inline __ATTRS_o_ai void vec_xst_trunc(vector signed __int128 __vec,
-                                              ptrdiff_t __offset,
-                                              signed char *__ptr) {
+static __inline__ __ATTRS_o_ai void vec_xst_trunc(vector signed __int128 __vec,
+                                                  ptrdiff_t __offset,
+                                                  signed char *__ptr) {
   *(__ptr + __offset) = (signed char)__vec[0];
 }
 
-static inline __ATTRS_o_ai void vec_xst_trunc(vector unsigned __int128 __vec,
-                                              ptrdiff_t __offset,
-                                              unsigned char *__ptr) {
+static __inline__ __ATTRS_o_ai void
+vec_xst_trunc(vector unsigned __int128 __vec, ptrdiff_t __offset,
+              unsigned char *__ptr) {
   *(__ptr + __offset) = (unsigned char)__vec[0];
 }
 
-static inline __ATTRS_o_ai void vec_xst_trunc(vector signed __int128 __vec,
-                                              ptrdiff_t __offset,
-                                              signed short *__ptr) {
+static __inline__ __ATTRS_o_ai void vec_xst_trunc(vector signed __int128 __vec,
+                                                  ptrdiff_t __offset,
+                                                  signed short *__ptr) {
   *(__ptr + __offset) = (signed short)__vec[0];
 }
 
-static inline __ATTRS_o_ai void vec_xst_trunc(vector unsigned __int128 __vec,
-                                              ptrdiff_t __offset,
-                                              unsigned short *__ptr) {
+static __inline__ __ATTRS_o_ai void
+vec_xst_trunc(vector unsigned __int128 __vec, ptrdiff_t __offset,
+              unsigned short *__ptr) {
   *(__ptr + __offset) = (unsigned short)__vec[0];
 }
 
-static inline __ATTRS_o_ai void vec_xst_trunc(vector signed __int128 __vec,
-                                              ptrdiff_t __offset,
-                                              signed int *__ptr) {
+static __inline__ __ATTRS_o_ai void vec_xst_trunc(vector signed __int128 __vec,
+                                                  ptrdiff_t __offset,
+                                                  signed int *__ptr) {
   *(__ptr + __offset) = (signed int)__vec[0];
 }
 
-static inline __ATTRS_o_ai void vec_xst_trunc(vector unsigned __int128 __vec,
-                                              ptrdiff_t __offset,
-                                              unsigned int *__ptr) {
+static __inline__ __ATTRS_o_ai void
+vec_xst_trunc(vector unsigned __int128 __vec, ptrdiff_t __offset,
+              unsigned int *__ptr) {
   *(__ptr + __offset) = (unsigned int)__vec[0];
 }
 
-static inline __ATTRS_o_ai void vec_xst_trunc(vector signed __int128 __vec,
-                                              ptrdiff_t __offset,
-                                              signed long long *__ptr) {
+static __inline__ __ATTRS_o_ai void vec_xst_trunc(vector signed __int128 __vec,
+                                                  ptrdiff_t __offset,
+                                                  signed long long *__ptr) {
   *(__ptr + __offset) = (signed long long)__vec[0];
 }
 
-static inline __ATTRS_o_ai void vec_xst_trunc(vector unsigned __int128 __vec,
-                                              ptrdiff_t __offset,
-                                              unsigned long long *__ptr) {
+static __inline__ __ATTRS_o_ai void
+vec_xst_trunc(vector unsigned __int128 __vec, ptrdiff_t __offset,
+              unsigned long long *__ptr) {
   *(__ptr + __offset) = (unsigned long long)__vec[0];
 }
 #endif

lib/include/amxavx512intrin.h

@@ -228,7 +228,7 @@
 ///             dst.byte[i] := a.row[row_index].byte[row_chunk+i]
 /// ENDFOR
 /// \endcode
-#define _tile_movrow(a, b) __builtin_ia32_tilemovrow(a, b)
+#define _tile_movrow(a, b) ((__m512i)__builtin_ia32_tilemovrow(a, b))
 
 /// This is internal intrinsic. C/C++ user should avoid calling it directly.
 

lib/include/amxcomplexintrin.h

@@ -135,9 +135,8 @@ _tile_cmmrlfp16ps_internal(unsigned short m, unsigned short n, unsigned short k,
 ///    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) {
+static __inline__ void __DEFAULT_FN_ATTRS_COMPLEX
+__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);
 }
@@ -158,9 +157,8 @@ static void __tile_cmmimfp16ps(__tile1024i *dst, __tile1024i 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) {
+static __inline__ void __DEFAULT_FN_ATTRS_COMPLEX
+__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);
 }

lib/include/amxmovrstransposeintrin.h

@@ -197,4 +197,4 @@ static void __tile_2rpntlvwz1rst1(__tile1024i *dst0, __tile1024i *dst1,
 
 #undef __DEFAULT_FN_ATTRS
 #endif /* __x86_64__ */
-#endif /* __AMX_MOVRS_TRANSPOSEINTRIN_H */
+#endif /* __AMX_MOVRS_TRANSPOSEINTRIN_H */

lib/include/amxtf32transposeintrin.h

@@ -8,7 +8,7 @@
  */
 #ifndef __IMMINTRIN_H
 #error                                                                         \
-    "Never use <amxtf32tranposeintrin.h> directly; include <immintrin.h> instead."
+    "Never use <amxtf32transposeintrin.h> directly; include <immintrin.h> instead."
 #endif // __IMMINTRIN_H
 
 #ifndef __AMX_TF32TRANSPOSEINTRIN_H

lib/include/andes_vector.h

@@ -0,0 +1,16 @@
+//===----- andes_vector.h - Andes Vector definitions ----------------------===//
+//
+// 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 _ANDES_VECTOR_H_
+#define _ANDES_VECTOR_H_
+
+#include "riscv_vector.h"
+
+#pragma clang riscv intrinsic andes_vector
+
+#endif //_ANDES_VECTOR_H_

lib/include/arm_acle.h

@@ -29,47 +29,16 @@ extern "C" {
 
 /* 7 SYNCHRONIZATION, BARRIER AND HINT INTRINSICS */
 /* 7.3 Memory barriers */
-#if !__has_builtin(__dmb)
-#define __dmb(i) __builtin_arm_dmb(i)
-#endif
-#if !__has_builtin(__dsb)
-#define __dsb(i) __builtin_arm_dsb(i)
-#endif
-#if !__has_builtin(__isb)
-#define __isb(i) __builtin_arm_isb(i)
-#endif
+void __dmb(unsigned int);
+void __dsb(unsigned int);
+void __isb(unsigned int);
 
 /* 7.4 Hints */
-
-#if !__has_builtin(__wfi)
-static __inline__ void __attribute__((__always_inline__, __nodebug__)) __wfi(void) {
-  __builtin_arm_wfi();
-}
-#endif
-
-#if !__has_builtin(__wfe)
-static __inline__ void __attribute__((__always_inline__, __nodebug__)) __wfe(void) {
-  __builtin_arm_wfe();
-}
-#endif
-
-#if !__has_builtin(__sev)
-static __inline__ void __attribute__((__always_inline__, __nodebug__)) __sev(void) {
-  __builtin_arm_sev();
-}
-#endif
-
-#if !__has_builtin(__sevl)
-static __inline__ void __attribute__((__always_inline__, __nodebug__)) __sevl(void) {
-  __builtin_arm_sevl();
-}
-#endif
-
-#if !__has_builtin(__yield)
-static __inline__ void __attribute__((__always_inline__, __nodebug__)) __yield(void) {
-  __builtin_arm_yield();
-}
-#endif
+void __wfi(void);
+void __wfe(void);
+void __sev(void);
+void __sevl(void);
+void __yield(void);
 
 #if defined(__ARM_32BIT_STATE) && __ARM_32BIT_STATE
 #define __dbg(t) __builtin_arm_dbg(t)
@@ -872,8 +841,9 @@ __gcspopm() {
   return __builtin_arm_gcspopm(0);
 }
 
-static __inline__ const void * __attribute__((__always_inline__, __nodebug__, target("gcs")))
-__gcsss(const void *__stack) {
+static __inline__ void *__attribute__((__always_inline__, __nodebug__,
+                                       target("gcs")))
+__gcsss(void *__stack) {
   return __builtin_arm_gcsss(__stack);
 }
 #endif

lib/include/arm_fp16.h

@@ -34,408 +34,408 @@ typedef __fp16 float16_t;
   float16_t __ret; \
   float16_t __s0 = __p0; \
   float16_t __s1 = __p1; \
-  __ret = (float16_t) __builtin_neon_vabdh_f16(__s0, __s1); \
+  __ret = __builtin_bit_cast(float16_t, __builtin_neon_vabdh_f16(__s0, __s1)); \
   __ret; \
 })
 #define vabsh_f16(__p0) __extension__ ({ \
   float16_t __ret; \
   float16_t __s0 = __p0; \
-  __ret = (float16_t) __builtin_neon_vabsh_f16(__s0); \
+  __ret = __builtin_bit_cast(float16_t, __builtin_neon_vabsh_f16(__s0)); \
   __ret; \
 })
 #define vaddh_f16(__p0, __p1) __extension__ ({ \
   float16_t __ret; \
   float16_t __s0 = __p0; \
   float16_t __s1 = __p1; \
-  __ret = (float16_t) __builtin_neon_vaddh_f16(__s0, __s1); \
+  __ret = __builtin_bit_cast(float16_t, __builtin_neon_vaddh_f16(__s0, __s1)); \
   __ret; \
 })
 #define vcageh_f16(__p0, __p1) __extension__ ({ \
   uint16_t __ret; \
   float16_t __s0 = __p0; \
   float16_t __s1 = __p1; \
-  __ret = (uint16_t) __builtin_neon_vcageh_f16(__s0, __s1); \
+  __ret = __builtin_bit_cast(uint16_t, __builtin_neon_vcageh_f16(__s0, __s1)); \
   __ret; \
 })
 #define vcagth_f16(__p0, __p1) __extension__ ({ \
   uint16_t __ret; \
   float16_t __s0 = __p0; \
   float16_t __s1 = __p1; \
-  __ret = (uint16_t) __builtin_neon_vcagth_f16(__s0, __s1); \
+  __ret = __builtin_bit_cast(uint16_t, __builtin_neon_vcagth_f16(__s0, __s1)); \
   __ret; \
 })
 #define vcaleh_f16(__p0, __p1) __extension__ ({ \
   uint16_t __ret; \
   float16_t __s0 = __p0; \
   float16_t __s1 = __p1; \
-  __ret = (uint16_t) __builtin_neon_vcaleh_f16(__s0, __s1); \
+  __ret = __builtin_bit_cast(uint16_t, __builtin_neon_vcaleh_f16(__s0, __s1)); \
   __ret; \
 })
 #define vcalth_f16(__p0, __p1) __extension__ ({ \
   uint16_t __ret; \
   float16_t __s0 = __p0; \
   float16_t __s1 = __p1; \
-  __ret = (uint16_t) __builtin_neon_vcalth_f16(__s0, __s1); \
+  __ret = __builtin_bit_cast(uint16_t, __builtin_neon_vcalth_f16(__s0, __s1)); \
   __ret; \
 })
 #define vceqh_f16(__p0, __p1) __extension__ ({ \
   uint16_t __ret; \
   float16_t __s0 = __p0; \
   float16_t __s1 = __p1; \
-  __ret = (uint16_t) __builtin_neon_vceqh_f16(__s0, __s1); \
+  __ret = __builtin_bit_cast(uint16_t, __builtin_neon_vceqh_f16(__s0, __s1)); \
   __ret; \
 })
 #define vceqzh_f16(__p0) __extension__ ({ \
   uint16_t __ret; \
   float16_t __s0 = __p0; \
-  __ret = (uint16_t) __builtin_neon_vceqzh_f16(__s0); \
+  __ret = __builtin_bit_cast(uint16_t, __builtin_neon_vceqzh_f16(__s0)); \
   __ret; \
 })
 #define vcgeh_f16(__p0, __p1) __extension__ ({ \
   uint16_t __ret; \
   float16_t __s0 = __p0; \
   float16_t __s1 = __p1; \
-  __ret = (uint16_t) __builtin_neon_vcgeh_f16(__s0, __s1); \
+  __ret = __builtin_bit_cast(uint16_t, __builtin_neon_vcgeh_f16(__s0, __s1)); \
   __ret; \
 })
 #define vcgezh_f16(__p0) __extension__ ({ \
   uint16_t __ret; \
   float16_t __s0 = __p0; \
-  __ret = (uint16_t) __builtin_neon_vcgezh_f16(__s0); \
+  __ret = __builtin_bit_cast(uint16_t, __builtin_neon_vcgezh_f16(__s0)); \
   __ret; \
 })
 #define vcgth_f16(__p0, __p1) __extension__ ({ \
   uint16_t __ret; \
   float16_t __s0 = __p0; \
   float16_t __s1 = __p1; \
-  __ret = (uint16_t) __builtin_neon_vcgth_f16(__s0, __s1); \
+  __ret = __builtin_bit_cast(uint16_t, __builtin_neon_vcgth_f16(__s0, __s1)); \
   __ret; \
 })
 #define vcgtzh_f16(__p0) __extension__ ({ \
   uint16_t __ret; \
   float16_t __s0 = __p0; \
-  __ret = (uint16_t) __builtin_neon_vcgtzh_f16(__s0); \
+  __ret = __builtin_bit_cast(uint16_t, __builtin_neon_vcgtzh_f16(__s0)); \
   __ret; \
 })
 #define vcleh_f16(__p0, __p1) __extension__ ({ \
   uint16_t __ret; \
   float16_t __s0 = __p0; \
   float16_t __s1 = __p1; \
-  __ret = (uint16_t) __builtin_neon_vcleh_f16(__s0, __s1); \
+  __ret = __builtin_bit_cast(uint16_t, __builtin_neon_vcleh_f16(__s0, __s1)); \
   __ret; \
 })
 #define vclezh_f16(__p0) __extension__ ({ \
   uint16_t __ret; \
   float16_t __s0 = __p0; \
-  __ret = (uint16_t) __builtin_neon_vclezh_f16(__s0); \
+  __ret = __builtin_bit_cast(uint16_t, __builtin_neon_vclezh_f16(__s0)); \
   __ret; \
 })
 #define vclth_f16(__p0, __p1) __extension__ ({ \
   uint16_t __ret; \
   float16_t __s0 = __p0; \
   float16_t __s1 = __p1; \
-  __ret = (uint16_t) __builtin_neon_vclth_f16(__s0, __s1); \
+  __ret = __builtin_bit_cast(uint16_t, __builtin_neon_vclth_f16(__s0, __s1)); \
   __ret; \
 })
 #define vcltzh_f16(__p0) __extension__ ({ \
   uint16_t __ret; \
   float16_t __s0 = __p0; \
-  __ret = (uint16_t) __builtin_neon_vcltzh_f16(__s0); \
+  __ret = __builtin_bit_cast(uint16_t, __builtin_neon_vcltzh_f16(__s0)); \
   __ret; \
 })
 #define vcvth_n_s16_f16(__p0, __p1) __extension__ ({ \
   int16_t __ret; \
   float16_t __s0 = __p0; \
-  __ret = (int16_t) __builtin_neon_vcvth_n_s16_f16(__s0, __p1); \
+  __ret = __builtin_bit_cast(int16_t, __builtin_neon_vcvth_n_s16_f16(__s0, __p1)); \
   __ret; \
 })
 #define vcvth_n_s32_f16(__p0, __p1) __extension__ ({ \
   int32_t __ret; \
   float16_t __s0 = __p0; \
-  __ret = (int32_t) __builtin_neon_vcvth_n_s32_f16(__s0, __p1); \
+  __ret = __builtin_bit_cast(int32_t, __builtin_neon_vcvth_n_s32_f16(__s0, __p1)); \
   __ret; \
 })
 #define vcvth_n_s64_f16(__p0, __p1) __extension__ ({ \
   int64_t __ret; \
   float16_t __s0 = __p0; \
-  __ret = (int64_t) __builtin_neon_vcvth_n_s64_f16(__s0, __p1); \
+  __ret = __builtin_bit_cast(int64_t, __builtin_neon_vcvth_n_s64_f16(__s0, __p1)); \
   __ret; \
 })
 #define vcvth_n_u16_f16(__p0, __p1) __extension__ ({ \
   uint16_t __ret; \
   float16_t __s0 = __p0; \
-  __ret = (uint16_t) __builtin_neon_vcvth_n_u16_f16(__s0, __p1); \
+  __ret = __builtin_bit_cast(uint16_t, __builtin_neon_vcvth_n_u16_f16(__s0, __p1)); \
   __ret; \
 })
 #define vcvth_n_u32_f16(__p0, __p1) __extension__ ({ \
   uint32_t __ret; \
   float16_t __s0 = __p0; \
-  __ret = (uint32_t) __builtin_neon_vcvth_n_u32_f16(__s0, __p1); \
+  __ret = __builtin_bit_cast(uint32_t, __builtin_neon_vcvth_n_u32_f16(__s0, __p1)); \
   __ret; \
 })
 #define vcvth_n_u64_f16(__p0, __p1) __extension__ ({ \
   uint64_t __ret; \
   float16_t __s0 = __p0; \
-  __ret = (uint64_t) __builtin_neon_vcvth_n_u64_f16(__s0, __p1); \
+  __ret = __builtin_bit_cast(uint64_t, __builtin_neon_vcvth_n_u64_f16(__s0, __p1)); \
   __ret; \
 })
 #define vcvth_s16_f16(__p0) __extension__ ({ \
   int16_t __ret; \
   float16_t __s0 = __p0; \
-  __ret = (int16_t) __builtin_neon_vcvth_s16_f16(__s0); \
+  __ret = __builtin_bit_cast(int16_t, __builtin_neon_vcvth_s16_f16(__s0)); \
   __ret; \
 })
 #define vcvth_s32_f16(__p0) __extension__ ({ \
   int32_t __ret; \
   float16_t __s0 = __p0; \
-  __ret = (int32_t) __builtin_neon_vcvth_s32_f16(__s0); \
+  __ret = __builtin_bit_cast(int32_t, __builtin_neon_vcvth_s32_f16(__s0)); \
   __ret; \
 })
 #define vcvth_s64_f16(__p0) __extension__ ({ \
   int64_t __ret; \
   float16_t __s0 = __p0; \
-  __ret = (int64_t) __builtin_neon_vcvth_s64_f16(__s0); \
+  __ret = __builtin_bit_cast(int64_t, __builtin_neon_vcvth_s64_f16(__s0)); \
   __ret; \
 })
 #define vcvth_u16_f16(__p0) __extension__ ({ \
   uint16_t __ret; \
   float16_t __s0 = __p0; \
-  __ret = (uint16_t) __builtin_neon_vcvth_u16_f16(__s0); \
+  __ret = __builtin_bit_cast(uint16_t, __builtin_neon_vcvth_u16_f16(__s0)); \
   __ret; \
 })
 #define vcvth_u32_f16(__p0) __extension__ ({ \
   uint32_t __ret; \
   float16_t __s0 = __p0; \
-  __ret = (uint32_t) __builtin_neon_vcvth_u32_f16(__s0); \
+  __ret = __builtin_bit_cast(uint32_t, __builtin_neon_vcvth_u32_f16(__s0)); \
   __ret; \
 })
 #define vcvth_u64_f16(__p0) __extension__ ({ \
   uint64_t __ret; \
   float16_t __s0 = __p0; \
-  __ret = (uint64_t) __builtin_neon_vcvth_u64_f16(__s0); \
+  __ret = __builtin_bit_cast(uint64_t, __builtin_neon_vcvth_u64_f16(__s0)); \
   __ret; \
 })
 #define vcvtah_s16_f16(__p0) __extension__ ({ \
   int16_t __ret; \
   float16_t __s0 = __p0; \
-  __ret = (int16_t) __builtin_neon_vcvtah_s16_f16(__s0); \
+  __ret = __builtin_bit_cast(int16_t, __builtin_neon_vcvtah_s16_f16(__s0)); \
   __ret; \
 })
 #define vcvtah_s32_f16(__p0) __extension__ ({ \
   int32_t __ret; \
   float16_t __s0 = __p0; \
-  __ret = (int32_t) __builtin_neon_vcvtah_s32_f16(__s0); \
+  __ret = __builtin_bit_cast(int32_t, __builtin_neon_vcvtah_s32_f16(__s0)); \
   __ret; \
 })
 #define vcvtah_s64_f16(__p0) __extension__ ({ \
   int64_t __ret; \
   float16_t __s0 = __p0; \
-  __ret = (int64_t) __builtin_neon_vcvtah_s64_f16(__s0); \
+  __ret = __builtin_bit_cast(int64_t, __builtin_neon_vcvtah_s64_f16(__s0)); \
   __ret; \
 })
 #define vcvtah_u16_f16(__p0) __extension__ ({ \
   uint16_t __ret; \
   float16_t __s0 = __p0; \
-  __ret = (uint16_t) __builtin_neon_vcvtah_u16_f16(__s0); \
+  __ret = __builtin_bit_cast(uint16_t, __builtin_neon_vcvtah_u16_f16(__s0)); \
   __ret; \
 })
 #define vcvtah_u32_f16(__p0) __extension__ ({ \
   uint32_t __ret; \
   float16_t __s0 = __p0; \
-  __ret = (uint32_t) __builtin_neon_vcvtah_u32_f16(__s0); \
+  __ret = __builtin_bit_cast(uint32_t, __builtin_neon_vcvtah_u32_f16(__s0)); \
   __ret; \
 })
 #define vcvtah_u64_f16(__p0) __extension__ ({ \
   uint64_t __ret; \
   float16_t __s0 = __p0; \
-  __ret = (uint64_t) __builtin_neon_vcvtah_u64_f16(__s0); \
+  __ret = __builtin_bit_cast(uint64_t, __builtin_neon_vcvtah_u64_f16(__s0)); \
   __ret; \
 })
 #define vcvth_f16_u16(__p0) __extension__ ({ \
   float16_t __ret; \
   uint16_t __s0 = __p0; \
-  __ret = (float16_t) __builtin_neon_vcvth_f16_u16(__s0); \
+  __ret = __builtin_bit_cast(float16_t, __builtin_neon_vcvth_f16_u16(__s0)); \
   __ret; \
 })
 #define vcvth_f16_s16(__p0) __extension__ ({ \
   float16_t __ret; \
   int16_t __s0 = __p0; \
-  __ret = (float16_t) __builtin_neon_vcvth_f16_s16(__s0); \
+  __ret = __builtin_bit_cast(float16_t, __builtin_neon_vcvth_f16_s16(__s0)); \
   __ret; \
 })
 #define vcvth_f16_u32(__p0) __extension__ ({ \
   float16_t __ret; \
   uint32_t __s0 = __p0; \
-  __ret = (float16_t) __builtin_neon_vcvth_f16_u32(__s0); \
+  __ret = __builtin_bit_cast(float16_t, __builtin_neon_vcvth_f16_u32(__s0)); \
   __ret; \
 })
 #define vcvth_f16_s32(__p0) __extension__ ({ \
   float16_t __ret; \
   int32_t __s0 = __p0; \
-  __ret = (float16_t) __builtin_neon_vcvth_f16_s32(__s0); \
+  __ret = __builtin_bit_cast(float16_t, __builtin_neon_vcvth_f16_s32(__s0)); \
   __ret; \
 })
 #define vcvth_f16_u64(__p0) __extension__ ({ \
   float16_t __ret; \
   uint64_t __s0 = __p0; \
-  __ret = (float16_t) __builtin_neon_vcvth_f16_u64(__s0); \
+  __ret = __builtin_bit_cast(float16_t, __builtin_neon_vcvth_f16_u64(__s0)); \
   __ret; \
 })
 #define vcvth_f16_s64(__p0) __extension__ ({ \
   float16_t __ret; \
   int64_t __s0 = __p0; \
-  __ret = (float16_t) __builtin_neon_vcvth_f16_s64(__s0); \
+  __ret = __builtin_bit_cast(float16_t, __builtin_neon_vcvth_f16_s64(__s0)); \
   __ret; \
 })
 #define vcvth_n_f16_u32(__p0, __p1) __extension__ ({ \
   float16_t __ret; \
   uint32_t __s0 = __p0; \
-  __ret = (float16_t) __builtin_neon_vcvth_n_f16_u32(__s0, __p1); \
+  __ret = __builtin_bit_cast(float16_t, __builtin_neon_vcvth_n_f16_u32(__s0, __p1)); \
   __ret; \
 })
 #define vcvth_n_f16_s32(__p0, __p1) __extension__ ({ \
   float16_t __ret; \
   int32_t __s0 = __p0; \
-  __ret = (float16_t) __builtin_neon_vcvth_n_f16_s32(__s0, __p1); \
+  __ret = __builtin_bit_cast(float16_t, __builtin_neon_vcvth_n_f16_s32(__s0, __p1)); \
   __ret; \
 })
 #define vcvth_n_f16_u64(__p0, __p1) __extension__ ({ \
   float16_t __ret; \
   uint64_t __s0 = __p0; \
-  __ret = (float16_t) __builtin_neon_vcvth_n_f16_u64(__s0, __p1); \
+  __ret = __builtin_bit_cast(float16_t, __builtin_neon_vcvth_n_f16_u64(__s0, __p1)); \
   __ret; \
 })
 #define vcvth_n_f16_s64(__p0, __p1) __extension__ ({ \
   float16_t __ret; \
   int64_t __s0 = __p0; \
-  __ret = (float16_t) __builtin_neon_vcvth_n_f16_s64(__s0, __p1); \
+  __ret = __builtin_bit_cast(float16_t, __builtin_neon_vcvth_n_f16_s64(__s0, __p1)); \
   __ret; \
 })
 #define vcvth_n_f16_u16(__p0, __p1) __extension__ ({ \
   float16_t __ret; \
   uint16_t __s0 = __p0; \
-  __ret = (float16_t) __builtin_neon_vcvth_n_f16_u16(__s0, __p1); \
+  __ret = __builtin_bit_cast(float16_t, __builtin_neon_vcvth_n_f16_u16(__s0, __p1)); \
   __ret; \
 })
 #define vcvth_n_f16_s16(__p0, __p1) __extension__ ({ \
   float16_t __ret; \
   int16_t __s0 = __p0; \
-  __ret = (float16_t) __builtin_neon_vcvth_n_f16_s16(__s0, __p1); \
+  __ret = __builtin_bit_cast(float16_t, __builtin_neon_vcvth_n_f16_s16(__s0, __p1)); \
   __ret; \
 })
 #define vcvtmh_s16_f16(__p0) __extension__ ({ \
   int16_t __ret; \
   float16_t __s0 = __p0; \
-  __ret = (int16_t) __builtin_neon_vcvtmh_s16_f16(__s0); \
+  __ret = __builtin_bit_cast(int16_t, __builtin_neon_vcvtmh_s16_f16(__s0)); \
   __ret; \
 })
 #define vcvtmh_s32_f16(__p0) __extension__ ({ \
   int32_t __ret; \
   float16_t __s0 = __p0; \
-  __ret = (int32_t) __builtin_neon_vcvtmh_s32_f16(__s0); \
+  __ret = __builtin_bit_cast(int32_t, __builtin_neon_vcvtmh_s32_f16(__s0)); \
   __ret; \
 })
 #define vcvtmh_s64_f16(__p0) __extension__ ({ \
   int64_t __ret; \
   float16_t __s0 = __p0; \
-  __ret = (int64_t) __builtin_neon_vcvtmh_s64_f16(__s0); \
+  __ret = __builtin_bit_cast(int64_t, __builtin_neon_vcvtmh_s64_f16(__s0)); \
   __ret; \
 })
 #define vcvtmh_u16_f16(__p0) __extension__ ({ \
   uint16_t __ret; \
   float16_t __s0 = __p0; \
-  __ret = (uint16_t) __builtin_neon_vcvtmh_u16_f16(__s0); \
+  __ret = __builtin_bit_cast(uint16_t, __builtin_neon_vcvtmh_u16_f16(__s0)); \
   __ret; \
 })
 #define vcvtmh_u32_f16(__p0) __extension__ ({ \
   uint32_t __ret; \
   float16_t __s0 = __p0; \
-  __ret = (uint32_t) __builtin_neon_vcvtmh_u32_f16(__s0); \
+  __ret = __builtin_bit_cast(uint32_t, __builtin_neon_vcvtmh_u32_f16(__s0)); \
   __ret; \
 })
 #define vcvtmh_u64_f16(__p0) __extension__ ({ \
   uint64_t __ret; \
   float16_t __s0 = __p0; \
-  __ret = (uint64_t) __builtin_neon_vcvtmh_u64_f16(__s0); \
+  __ret = __builtin_bit_cast(uint64_t, __builtin_neon_vcvtmh_u64_f16(__s0)); \
   __ret; \
 })
 #define vcvtnh_s16_f16(__p0) __extension__ ({ \
   int16_t __ret; \
   float16_t __s0 = __p0; \
-  __ret = (int16_t) __builtin_neon_vcvtnh_s16_f16(__s0); \
+  __ret = __builtin_bit_cast(int16_t, __builtin_neon_vcvtnh_s16_f16(__s0)); \
   __ret; \
 })
 #define vcvtnh_s32_f16(__p0) __extension__ ({ \
   int32_t __ret; \
   float16_t __s0 = __p0; \
-  __ret = (int32_t) __builtin_neon_vcvtnh_s32_f16(__s0); \
+  __ret = __builtin_bit_cast(int32_t, __builtin_neon_vcvtnh_s32_f16(__s0)); \
   __ret; \
 })
 #define vcvtnh_s64_f16(__p0) __extension__ ({ \
   int64_t __ret; \
   float16_t __s0 = __p0; \
-  __ret = (int64_t) __builtin_neon_vcvtnh_s64_f16(__s0); \
+  __ret = __builtin_bit_cast(int64_t, __builtin_neon_vcvtnh_s64_f16(__s0)); \
   __ret; \
 })
 #define vcvtnh_u16_f16(__p0) __extension__ ({ \
   uint16_t __ret; \
   float16_t __s0 = __p0; \
-  __ret = (uint16_t) __builtin_neon_vcvtnh_u16_f16(__s0); \
+  __ret = __builtin_bit_cast(uint16_t, __builtin_neon_vcvtnh_u16_f16(__s0)); \
   __ret; \
 })
 #define vcvtnh_u32_f16(__p0) __extension__ ({ \
   uint32_t __ret; \
   float16_t __s0 = __p0; \
-  __ret = (uint32_t) __builtin_neon_vcvtnh_u32_f16(__s0); \
+  __ret = __builtin_bit_cast(uint32_t, __builtin_neon_vcvtnh_u32_f16(__s0)); \
   __ret; \
 })
 #define vcvtnh_u64_f16(__p0) __extension__ ({ \
   uint64_t __ret; \
   float16_t __s0 = __p0; \
-  __ret = (uint64_t) __builtin_neon_vcvtnh_u64_f16(__s0); \
+  __ret = __builtin_bit_cast(uint64_t, __builtin_neon_vcvtnh_u64_f16(__s0)); \
   __ret; \
 })
 #define vcvtph_s16_f16(__p0) __extension__ ({ \
   int16_t __ret; \
   float16_t __s0 = __p0; \
-  __ret = (int16_t) __builtin_neon_vcvtph_s16_f16(__s0); \
+  __ret = __builtin_bit_cast(int16_t, __builtin_neon_vcvtph_s16_f16(__s0)); \
   __ret; \
 })
 #define vcvtph_s32_f16(__p0) __extension__ ({ \
   int32_t __ret; \
   float16_t __s0 = __p0; \
-  __ret = (int32_t) __builtin_neon_vcvtph_s32_f16(__s0); \
+  __ret = __builtin_bit_cast(int32_t, __builtin_neon_vcvtph_s32_f16(__s0)); \
   __ret; \
 })
 #define vcvtph_s64_f16(__p0) __extension__ ({ \
   int64_t __ret; \
   float16_t __s0 = __p0; \
-  __ret = (int64_t) __builtin_neon_vcvtph_s64_f16(__s0); \
+  __ret = __builtin_bit_cast(int64_t, __builtin_neon_vcvtph_s64_f16(__s0)); \
   __ret; \
 })
 #define vcvtph_u16_f16(__p0) __extension__ ({ \
   uint16_t __ret; \
   float16_t __s0 = __p0; \
-  __ret = (uint16_t) __builtin_neon_vcvtph_u16_f16(__s0); \
+  __ret = __builtin_bit_cast(uint16_t, __builtin_neon_vcvtph_u16_f16(__s0)); \
   __ret; \
 })
 #define vcvtph_u32_f16(__p0) __extension__ ({ \
   uint32_t __ret; \
   float16_t __s0 = __p0; \
-  __ret = (uint32_t) __builtin_neon_vcvtph_u32_f16(__s0); \
+  __ret = __builtin_bit_cast(uint32_t, __builtin_neon_vcvtph_u32_f16(__s0)); \
   __ret; \
 })
 #define vcvtph_u64_f16(__p0) __extension__ ({ \
   uint64_t __ret; \
   float16_t __s0 = __p0; \
-  __ret = (uint64_t) __builtin_neon_vcvtph_u64_f16(__s0); \
+  __ret = __builtin_bit_cast(uint64_t, __builtin_neon_vcvtph_u64_f16(__s0)); \
   __ret; \
 })
 #define vdivh_f16(__p0, __p1) __extension__ ({ \
   float16_t __ret; \
   float16_t __s0 = __p0; \
   float16_t __s1 = __p1; \
-  __ret = (float16_t) __builtin_neon_vdivh_f16(__s0, __s1); \
+  __ret = __builtin_bit_cast(float16_t, __builtin_neon_vdivh_f16(__s0, __s1)); \
   __ret; \
 })
 #define vfmah_f16(__p0, __p1, __p2) __extension__ ({ \
@@ -443,7 +443,7 @@ typedef __fp16 float16_t;
   float16_t __s0 = __p0; \
   float16_t __s1 = __p1; \
   float16_t __s2 = __p2; \
-  __ret = (float16_t) __builtin_neon_vfmah_f16(__s0, __s1, __s2); \
+  __ret = __builtin_bit_cast(float16_t, __builtin_neon_vfmah_f16(__s0, __s1, __s2)); \
   __ret; \
 })
 #define vfmsh_f16(__p0, __p1, __p2) __extension__ ({ \
@@ -451,142 +451,142 @@ typedef __fp16 float16_t;
   float16_t __s0 = __p0; \
   float16_t __s1 = __p1; \
   float16_t __s2 = __p2; \
-  __ret = (float16_t) __builtin_neon_vfmsh_f16(__s0, __s1, __s2); \
+  __ret = __builtin_bit_cast(float16_t, __builtin_neon_vfmsh_f16(__s0, __s1, __s2)); \
   __ret; \
 })
 #define vmaxh_f16(__p0, __p1) __extension__ ({ \
   float16_t __ret; \
   float16_t __s0 = __p0; \
   float16_t __s1 = __p1; \
-  __ret = (float16_t) __builtin_neon_vmaxh_f16(__s0, __s1); \
+  __ret = __builtin_bit_cast(float16_t, __builtin_neon_vmaxh_f16(__s0, __s1)); \
   __ret; \
 })
 #define vmaxnmh_f16(__p0, __p1) __extension__ ({ \
   float16_t __ret; \
   float16_t __s0 = __p0; \
   float16_t __s1 = __p1; \
-  __ret = (float16_t) __builtin_neon_vmaxnmh_f16(__s0, __s1); \
+  __ret = __builtin_bit_cast(float16_t, __builtin_neon_vmaxnmh_f16(__s0, __s1)); \
   __ret; \
 })
 #define vminh_f16(__p0, __p1) __extension__ ({ \
   float16_t __ret; \
   float16_t __s0 = __p0; \
   float16_t __s1 = __p1; \
-  __ret = (float16_t) __builtin_neon_vminh_f16(__s0, __s1); \
+  __ret = __builtin_bit_cast(float16_t, __builtin_neon_vminh_f16(__s0, __s1)); \
   __ret; \
 })
 #define vminnmh_f16(__p0, __p1) __extension__ ({ \
   float16_t __ret; \
   float16_t __s0 = __p0; \
   float16_t __s1 = __p1; \
-  __ret = (float16_t) __builtin_neon_vminnmh_f16(__s0, __s1); \
+  __ret = __builtin_bit_cast(float16_t, __builtin_neon_vminnmh_f16(__s0, __s1)); \
   __ret; \
 })
 #define vmulh_f16(__p0, __p1) __extension__ ({ \
   float16_t __ret; \
   float16_t __s0 = __p0; \
   float16_t __s1 = __p1; \
-  __ret = (float16_t) __builtin_neon_vmulh_f16(__s0, __s1); \
+  __ret = __builtin_bit_cast(float16_t, __builtin_neon_vmulh_f16(__s0, __s1)); \
   __ret; \
 })
 #define vmulxh_f16(__p0, __p1) __extension__ ({ \
   float16_t __ret; \
   float16_t __s0 = __p0; \
   float16_t __s1 = __p1; \
-  __ret = (float16_t) __builtin_neon_vmulxh_f16(__s0, __s1); \
+  __ret = __builtin_bit_cast(float16_t, __builtin_neon_vmulxh_f16(__s0, __s1)); \
   __ret; \
 })
 #define vnegh_f16(__p0) __extension__ ({ \
   float16_t __ret; \
   float16_t __s0 = __p0; \
-  __ret = (float16_t) __builtin_neon_vnegh_f16(__s0); \
+  __ret = __builtin_bit_cast(float16_t, __builtin_neon_vnegh_f16(__s0)); \
   __ret; \
 })
 #define vrecpeh_f16(__p0) __extension__ ({ \
   float16_t __ret; \
   float16_t __s0 = __p0; \
-  __ret = (float16_t) __builtin_neon_vrecpeh_f16(__s0); \
+  __ret = __builtin_bit_cast(float16_t, __builtin_neon_vrecpeh_f16(__s0)); \
   __ret; \
 })
 #define vrecpsh_f16(__p0, __p1) __extension__ ({ \
   float16_t __ret; \
   float16_t __s0 = __p0; \
   float16_t __s1 = __p1; \
-  __ret = (float16_t) __builtin_neon_vrecpsh_f16(__s0, __s1); \
+  __ret = __builtin_bit_cast(float16_t, __builtin_neon_vrecpsh_f16(__s0, __s1)); \
   __ret; \
 })
 #define vrecpxh_f16(__p0) __extension__ ({ \
   float16_t __ret; \
   float16_t __s0 = __p0; \
-  __ret = (float16_t) __builtin_neon_vrecpxh_f16(__s0); \
+  __ret = __builtin_bit_cast(float16_t, __builtin_neon_vrecpxh_f16(__s0)); \
   __ret; \
 })
 #define vrndh_f16(__p0) __extension__ ({ \
   float16_t __ret; \
   float16_t __s0 = __p0; \
-  __ret = (float16_t) __builtin_neon_vrndh_f16(__s0); \
+  __ret = __builtin_bit_cast(float16_t, __builtin_neon_vrndh_f16(__s0)); \
   __ret; \
 })
 #define vrndah_f16(__p0) __extension__ ({ \
   float16_t __ret; \
   float16_t __s0 = __p0; \
-  __ret = (float16_t) __builtin_neon_vrndah_f16(__s0); \
+  __ret = __builtin_bit_cast(float16_t, __builtin_neon_vrndah_f16(__s0)); \
   __ret; \
 })
 #define vrndih_f16(__p0) __extension__ ({ \
   float16_t __ret; \
   float16_t __s0 = __p0; \
-  __ret = (float16_t) __builtin_neon_vrndih_f16(__s0); \
+  __ret = __builtin_bit_cast(float16_t, __builtin_neon_vrndih_f16(__s0)); \
   __ret; \
 })
 #define vrndmh_f16(__p0) __extension__ ({ \
   float16_t __ret; \
   float16_t __s0 = __p0; \
-  __ret = (float16_t) __builtin_neon_vrndmh_f16(__s0); \
+  __ret = __builtin_bit_cast(float16_t, __builtin_neon_vrndmh_f16(__s0)); \
   __ret; \
 })
 #define vrndnh_f16(__p0) __extension__ ({ \
   float16_t __ret; \
   float16_t __s0 = __p0; \
-  __ret = (float16_t) __builtin_neon_vrndnh_f16(__s0); \
+  __ret = __builtin_bit_cast(float16_t, __builtin_neon_vrndnh_f16(__s0)); \
   __ret; \
 })
 #define vrndph_f16(__p0) __extension__ ({ \
   float16_t __ret; \
   float16_t __s0 = __p0; \
-  __ret = (float16_t) __builtin_neon_vrndph_f16(__s0); \
+  __ret = __builtin_bit_cast(float16_t, __builtin_neon_vrndph_f16(__s0)); \
   __ret; \
 })
 #define vrndxh_f16(__p0) __extension__ ({ \
   float16_t __ret; \
   float16_t __s0 = __p0; \
-  __ret = (float16_t) __builtin_neon_vrndxh_f16(__s0); \
+  __ret = __builtin_bit_cast(float16_t, __builtin_neon_vrndxh_f16(__s0)); \
   __ret; \
 })
 #define vrsqrteh_f16(__p0) __extension__ ({ \
   float16_t __ret; \
   float16_t __s0 = __p0; \
-  __ret = (float16_t) __builtin_neon_vrsqrteh_f16(__s0); \
+  __ret = __builtin_bit_cast(float16_t, __builtin_neon_vrsqrteh_f16(__s0)); \
   __ret; \
 })
 #define vrsqrtsh_f16(__p0, __p1) __extension__ ({ \
   float16_t __ret; \
   float16_t __s0 = __p0; \
   float16_t __s1 = __p1; \
-  __ret = (float16_t) __builtin_neon_vrsqrtsh_f16(__s0, __s1); \
+  __ret = __builtin_bit_cast(float16_t, __builtin_neon_vrsqrtsh_f16(__s0, __s1)); \
   __ret; \
 })
 #define vsqrth_f16(__p0) __extension__ ({ \
   float16_t __ret; \
   float16_t __s0 = __p0; \
-  __ret = (float16_t) __builtin_neon_vsqrth_f16(__s0); \
+  __ret = __builtin_bit_cast(float16_t, __builtin_neon_vsqrth_f16(__s0)); \
   __ret; \
 })
 #define vsubh_f16(__p0, __p1) __extension__ ({ \
   float16_t __ret; \
   float16_t __s0 = __p0; \
   float16_t __s1 = __p1; \
-  __ret = (float16_t) __builtin_neon_vsubh_f16(__s0, __s1); \
+  __ret = __builtin_bit_cast(float16_t, __builtin_neon_vsubh_f16(__s0, __s1)); \
   __ret; \
 })
 #endif

lib/include/arm_sme.h

@@ -146,6 +146,8 @@ __ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svread_hor_za128_s32
 svint32_t svread_hor_za128_s32_m(svint32_t, svbool_t, uint64_t, uint32_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svread_hor_za128_s64_m)))
 svint64_t svread_hor_za128_s64_m(svint64_t, svbool_t, uint64_t, uint32_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svread_hor_za128_mf8_m)))
+svmfloat8_t svread_hor_za128_mf8_m(svmfloat8_t, svbool_t, uint64_t, uint32_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svread_hor_za128_s16_m)))
 svint16_t svread_hor_za128_s16_m(svint16_t, svbool_t, uint64_t, uint32_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svread_hor_za16_u16_m)))
@@ -172,6 +174,8 @@ __ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svread_hor_za8_u8_m)
 svuint8_t svread_hor_za8_u8_m(svuint8_t, svbool_t, uint64_t, uint32_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svread_hor_za8_s8_m)))
 svint8_t svread_hor_za8_s8_m(svint8_t, svbool_t, uint64_t, uint32_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svread_hor_za8_mf8_m)))
+svmfloat8_t svread_hor_za8_mf8_m(svmfloat8_t, svbool_t, uint64_t, uint32_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svread_ver_za128_u8_m)))
 svuint8_t svread_ver_za128_u8_m(svuint8_t, svbool_t, uint64_t, uint32_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svread_ver_za128_u32_m)))
@@ -194,6 +198,8 @@ __ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svread_ver_za128_s32
 svint32_t svread_ver_za128_s32_m(svint32_t, svbool_t, uint64_t, uint32_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svread_ver_za128_s64_m)))
 svint64_t svread_ver_za128_s64_m(svint64_t, svbool_t, uint64_t, uint32_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svread_ver_za128_mf8_m)))
+svmfloat8_t svread_ver_za128_mf8_m(svmfloat8_t, svbool_t, uint64_t, uint32_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svread_ver_za128_s16_m)))
 svint16_t svread_ver_za128_s16_m(svint16_t, svbool_t, uint64_t, uint32_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svread_ver_za16_u16_m)))
@@ -220,6 +226,8 @@ __ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svread_ver_za8_u8_m)
 svuint8_t svread_ver_za8_u8_m(svuint8_t, svbool_t, uint64_t, uint32_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svread_ver_za8_s8_m)))
 svint8_t svread_ver_za8_s8_m(svint8_t, svbool_t, uint64_t, uint32_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svread_ver_za8_mf8_m)))
+svmfloat8_t svread_ver_za8_mf8_m(svmfloat8_t, svbool_t, uint64_t, uint32_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svst1_hor_vnum_za128)))
 void svst1_hor_vnum_za128(uint64_t, uint32_t, svbool_t, void *, int64_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svst1_hor_vnum_za16)))
@@ -294,6 +302,8 @@ __ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svwrite_hor_za128_s3
 void svwrite_hor_za128_s32_m(uint64_t, uint32_t, svbool_t, svint32_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svwrite_hor_za128_s64_m)))
 void svwrite_hor_za128_s64_m(uint64_t, uint32_t, svbool_t, svint64_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svwrite_hor_za128_mf8_m)))
+void svwrite_hor_za128_mf8_m(uint64_t, uint32_t, svbool_t, svmfloat8_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svwrite_hor_za128_s16_m)))
 void svwrite_hor_za128_s16_m(uint64_t, uint32_t, svbool_t, svint16_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svwrite_hor_za16_u16_m)))
@@ -320,6 +330,8 @@ __ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svwrite_hor_za8_u8_m
 void svwrite_hor_za8_u8_m(uint64_t, uint32_t, svbool_t, svuint8_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svwrite_hor_za8_s8_m)))
 void svwrite_hor_za8_s8_m(uint64_t, uint32_t, svbool_t, svint8_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svwrite_hor_za8_mf8_m)))
+void svwrite_hor_za8_mf8_m(uint64_t, uint32_t, svbool_t, svmfloat8_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svwrite_ver_za128_u8_m)))
 void svwrite_ver_za128_u8_m(uint64_t, uint32_t, svbool_t, svuint8_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svwrite_ver_za128_u32_m)))
@@ -342,6 +354,8 @@ __ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svwrite_ver_za128_s3
 void svwrite_ver_za128_s32_m(uint64_t, uint32_t, svbool_t, svint32_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svwrite_ver_za128_s64_m)))
 void svwrite_ver_za128_s64_m(uint64_t, uint32_t, svbool_t, svint64_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svwrite_ver_za128_mf8_m)))
+void svwrite_ver_za128_mf8_m(uint64_t, uint32_t, svbool_t, svmfloat8_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svwrite_ver_za128_s16_m)))
 void svwrite_ver_za128_s16_m(uint64_t, uint32_t, svbool_t, svint16_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svwrite_ver_za16_u16_m)))
@@ -368,6 +382,8 @@ __ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svwrite_ver_za8_u8_m
 void svwrite_ver_za8_u8_m(uint64_t, uint32_t, svbool_t, svuint8_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svwrite_ver_za8_s8_m)))
 void svwrite_ver_za8_s8_m(uint64_t, uint32_t, svbool_t, svint8_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svwrite_ver_za8_mf8_m)))
+void svwrite_ver_za8_mf8_m(uint64_t, uint32_t, svbool_t, svmfloat8_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svzero_mask_za)))
 void svzero_mask_za(uint64_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svzero_za)))
@@ -422,6 +438,8 @@ __aio __attribute__((__clang_arm_builtin_alias(__builtin_sme_svread_hor_za128_s3
 svint32_t svread_hor_za128_m(svint32_t, svbool_t, uint64_t, uint32_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sme_svread_hor_za128_s64_m)))
 svint64_t svread_hor_za128_m(svint64_t, svbool_t, uint64_t, uint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sme_svread_hor_za128_mf8_m)))
+svmfloat8_t svread_hor_za128_m(svmfloat8_t, svbool_t, uint64_t, uint32_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sme_svread_hor_za128_s16_m)))
 svint16_t svread_hor_za128_m(svint16_t, svbool_t, uint64_t, uint32_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sme_svread_hor_za16_u16_m)))
@@ -448,6 +466,8 @@ __aio __attribute__((__clang_arm_builtin_alias(__builtin_sme_svread_hor_za8_u8_m
 svuint8_t svread_hor_za8_m(svuint8_t, svbool_t, uint64_t, uint32_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sme_svread_hor_za8_s8_m)))
 svint8_t svread_hor_za8_m(svint8_t, svbool_t, uint64_t, uint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sme_svread_hor_za8_mf8_m)))
+svmfloat8_t svread_hor_za8_m(svmfloat8_t, svbool_t, uint64_t, uint32_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sme_svread_ver_za128_u8_m)))
 svuint8_t svread_ver_za128_m(svuint8_t, svbool_t, uint64_t, uint32_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sme_svread_ver_za128_u32_m)))
@@ -470,6 +490,8 @@ __aio __attribute__((__clang_arm_builtin_alias(__builtin_sme_svread_ver_za128_s3
 svint32_t svread_ver_za128_m(svint32_t, svbool_t, uint64_t, uint32_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sme_svread_ver_za128_s64_m)))
 svint64_t svread_ver_za128_m(svint64_t, svbool_t, uint64_t, uint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sme_svread_ver_za128_mf8_m)))
+svmfloat8_t svread_ver_za128_m(svmfloat8_t, svbool_t, uint64_t, uint32_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sme_svread_ver_za128_s16_m)))
 svint16_t svread_ver_za128_m(svint16_t, svbool_t, uint64_t, uint32_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sme_svread_ver_za16_u16_m)))
@@ -496,6 +518,8 @@ __aio __attribute__((__clang_arm_builtin_alias(__builtin_sme_svread_ver_za8_u8_m
 svuint8_t svread_ver_za8_m(svuint8_t, svbool_t, uint64_t, uint32_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sme_svread_ver_za8_s8_m)))
 svint8_t svread_ver_za8_m(svint8_t, svbool_t, uint64_t, uint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sme_svread_ver_za8_mf8_m)))
+svmfloat8_t svread_ver_za8_m(svmfloat8_t, svbool_t, uint64_t, uint32_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sme_svsumopa_za32_s8_m)))
 void svsumopa_za32_m(uint64_t, svbool_t, svbool_t, svint8_t, svuint8_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sme_svsumops_za32_s8_m)))
@@ -526,6 +550,8 @@ __aio __attribute__((__clang_arm_builtin_alias(__builtin_sme_svwrite_hor_za128_s
 void svwrite_hor_za128_m(uint64_t, uint32_t, svbool_t, svint32_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sme_svwrite_hor_za128_s64_m)))
 void svwrite_hor_za128_m(uint64_t, uint32_t, svbool_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sme_svwrite_hor_za128_mf8_m)))
+void svwrite_hor_za128_m(uint64_t, uint32_t, svbool_t, svmfloat8_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sme_svwrite_hor_za128_s16_m)))
 void svwrite_hor_za128_m(uint64_t, uint32_t, svbool_t, svint16_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sme_svwrite_hor_za16_u16_m)))
@@ -552,6 +578,8 @@ __aio __attribute__((__clang_arm_builtin_alias(__builtin_sme_svwrite_hor_za8_u8_
 void svwrite_hor_za8_m(uint64_t, uint32_t, svbool_t, svuint8_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sme_svwrite_hor_za8_s8_m)))
 void svwrite_hor_za8_m(uint64_t, uint32_t, svbool_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sme_svwrite_hor_za8_mf8_m)))
+void svwrite_hor_za8_m(uint64_t, uint32_t, svbool_t, svmfloat8_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sme_svwrite_ver_za128_u8_m)))
 void svwrite_ver_za128_m(uint64_t, uint32_t, svbool_t, svuint8_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sme_svwrite_ver_za128_u32_m)))
@@ -574,6 +602,8 @@ __aio __attribute__((__clang_arm_builtin_alias(__builtin_sme_svwrite_ver_za128_s
 void svwrite_ver_za128_m(uint64_t, uint32_t, svbool_t, svint32_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sme_svwrite_ver_za128_s64_m)))
 void svwrite_ver_za128_m(uint64_t, uint32_t, svbool_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sme_svwrite_ver_za128_mf8_m)))
+void svwrite_ver_za128_m(uint64_t, uint32_t, svbool_t, svmfloat8_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sme_svwrite_ver_za128_s16_m)))
 void svwrite_ver_za128_m(uint64_t, uint32_t, svbool_t, svint16_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sme_svwrite_ver_za16_u16_m)))
@@ -600,6 +630,8 @@ __aio __attribute__((__clang_arm_builtin_alias(__builtin_sme_svwrite_ver_za8_u8_
 void svwrite_ver_za8_m(uint64_t, uint32_t, svbool_t, svuint8_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sme_svwrite_ver_za8_s8_m)))
 void svwrite_ver_za8_m(uint64_t, uint32_t, svbool_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sme_svwrite_ver_za8_mf8_m)))
+void svwrite_ver_za8_m(uint64_t, uint32_t, svbool_t, svmfloat8_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svadd_za16_f16_vg1x2)))
 void svadd_za16_f16_vg1x2(uint32_t, svfloat16x2_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svadd_za16_f16_vg1x4)))
@@ -1158,6 +1190,8 @@ __ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svluti2_lane_zt_f16)
 svfloat16_t svluti2_lane_zt_f16(uint64_t, svuint8_t, uint64_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svluti2_lane_zt_s32)))
 svint32_t svluti2_lane_zt_s32(uint64_t, svuint8_t, uint64_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svluti2_lane_zt_mf8)))
+svmfloat8_t svluti2_lane_zt_mf8(uint64_t, svuint8_t, uint64_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svluti2_lane_zt_s16)))
 svint16_t svluti2_lane_zt_s16(uint64_t, svuint8_t, uint64_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svluti2_lane_zt_u8_x2)))
@@ -1176,6 +1210,8 @@ __ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svluti2_lane_zt_f16_
 svfloat16x2_t svluti2_lane_zt_f16_x2(uint64_t, svuint8_t, uint64_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svluti2_lane_zt_s32_x2)))
 svint32x2_t svluti2_lane_zt_s32_x2(uint64_t, svuint8_t, uint64_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svluti2_lane_zt_mf8_x2)))
+svmfloat8x2_t svluti2_lane_zt_mf8_x2(uint64_t, svuint8_t, uint64_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svluti2_lane_zt_s16_x2)))
 svint16x2_t svluti2_lane_zt_s16_x2(uint64_t, svuint8_t, uint64_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svluti2_lane_zt_u8_x4)))
@@ -1194,6 +1230,8 @@ __ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svluti2_lane_zt_f16_
 svfloat16x4_t svluti2_lane_zt_f16_x4(uint64_t, svuint8_t, uint64_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svluti2_lane_zt_s32_x4)))
 svint32x4_t svluti2_lane_zt_s32_x4(uint64_t, svuint8_t, uint64_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svluti2_lane_zt_mf8_x4)))
+svmfloat8x4_t svluti2_lane_zt_mf8_x4(uint64_t, svuint8_t, uint64_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svluti2_lane_zt_s16_x4)))
 svint16x4_t svluti2_lane_zt_s16_x4(uint64_t, svuint8_t, uint64_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svluti4_lane_zt_u8)))
@@ -1212,6 +1250,8 @@ __ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svluti4_lane_zt_f16)
 svfloat16_t svluti4_lane_zt_f16(uint64_t, svuint8_t, uint64_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svluti4_lane_zt_s32)))
 svint32_t svluti4_lane_zt_s32(uint64_t, svuint8_t, uint64_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svluti4_lane_zt_mf8)))
+svmfloat8_t svluti4_lane_zt_mf8(uint64_t, svuint8_t, uint64_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svluti4_lane_zt_s16)))
 svint16_t svluti4_lane_zt_s16(uint64_t, svuint8_t, uint64_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svluti4_lane_zt_u8_x2)))
@@ -1230,6 +1270,8 @@ __ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svluti4_lane_zt_f16_
 svfloat16x2_t svluti4_lane_zt_f16_x2(uint64_t, svuint8_t, uint64_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svluti4_lane_zt_s32_x2)))
 svint32x2_t svluti4_lane_zt_s32_x2(uint64_t, svuint8_t, uint64_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svluti4_lane_zt_mf8_x2)))
+svmfloat8x2_t svluti4_lane_zt_mf8_x2(uint64_t, svuint8_t, uint64_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svluti4_lane_zt_s16_x2)))
 svint16x2_t svluti4_lane_zt_s16_x2(uint64_t, svuint8_t, uint64_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svluti4_lane_zt_u32_x4)))
@@ -1514,10 +1556,14 @@ __ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svread_hor_za8_u8_vg
 svuint8x2_t svread_hor_za8_u8_vg2(uint64_t, uint32_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svread_hor_za8_s8_vg2)))
 svint8x2_t svread_hor_za8_s8_vg2(uint64_t, uint32_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svread_hor_za8_mf8_vg2)))
+svmfloat8x2_t svread_hor_za8_mf8_vg2(uint64_t, uint32_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svread_hor_za8_u8_vg4)))
 svuint8x4_t svread_hor_za8_u8_vg4(uint64_t, uint32_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svread_hor_za8_s8_vg4)))
 svint8x4_t svread_hor_za8_s8_vg4(uint64_t, uint32_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svread_hor_za8_mf8_vg4)))
+svmfloat8x4_t svread_hor_za8_mf8_vg4(uint64_t, uint32_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svread_ver_za16_u16_vg2)))
 svuint16x2_t svread_ver_za16_u16_vg2(uint64_t, uint32_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svread_ver_za16_bf16_vg2)))
@@ -1562,10 +1608,14 @@ __ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svread_ver_za8_u8_vg
 svuint8x2_t svread_ver_za8_u8_vg2(uint64_t, uint32_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svread_ver_za8_s8_vg2)))
 svint8x2_t svread_ver_za8_s8_vg2(uint64_t, uint32_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svread_ver_za8_mf8_vg2)))
+svmfloat8x2_t svread_ver_za8_mf8_vg2(uint64_t, uint32_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svread_ver_za8_u8_vg4)))
 svuint8x4_t svread_ver_za8_u8_vg4(uint64_t, uint32_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svread_ver_za8_s8_vg4)))
 svint8x4_t svread_ver_za8_s8_vg4(uint64_t, uint32_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svread_ver_za8_mf8_vg4)))
+svmfloat8x4_t svread_ver_za8_mf8_vg4(uint64_t, uint32_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svread_za16_u16_vg1x2)))
 svuint16x2_t svread_za16_u16_vg1x2(uint32_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svread_za16_bf16_vg1x2)))
@@ -1610,10 +1660,14 @@ __ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svread_za8_u8_vg1x2)
 svuint8x2_t svread_za8_u8_vg1x2(uint32_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svread_za8_s8_vg1x2)))
 svint8x2_t svread_za8_s8_vg1x2(uint32_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svread_za8_mf8_vg1x2)))
+svmfloat8x2_t svread_za8_mf8_vg1x2(uint32_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svread_za8_u8_vg1x4)))
 svuint8x4_t svread_za8_u8_vg1x4(uint32_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svread_za8_s8_vg1x4)))
 svint8x4_t svread_za8_s8_vg1x4(uint32_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svread_za8_mf8_vg1x4)))
+svmfloat8x4_t svread_za8_mf8_vg1x4(uint32_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svstr_zt)))
 void svstr_zt(uint64_t, void *);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svsub_write_single_za32_u32_vg1x2)))
@@ -1760,10 +1814,14 @@ __ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svwrite_hor_za8_u8_v
 void svwrite_hor_za8_u8_vg2(uint64_t, uint32_t, svuint8x2_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svwrite_hor_za8_s8_vg2)))
 void svwrite_hor_za8_s8_vg2(uint64_t, uint32_t, svint8x2_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svwrite_hor_za8_mf8_vg2)))
+void svwrite_hor_za8_mf8_vg2(uint64_t, uint32_t, svmfloat8x2_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svwrite_hor_za8_u8_vg4)))
 void svwrite_hor_za8_u8_vg4(uint64_t, uint32_t, svuint8x4_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svwrite_hor_za8_s8_vg4)))
 void svwrite_hor_za8_s8_vg4(uint64_t, uint32_t, svint8x4_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svwrite_hor_za8_mf8_vg4)))
+void svwrite_hor_za8_mf8_vg4(uint64_t, uint32_t, svmfloat8x4_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svwrite_ver_za16_u16_vg2)))
 void svwrite_ver_za16_u16_vg2(uint64_t, uint32_t, svuint16x2_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svwrite_ver_za16_bf16_vg2)))
@@ -1808,10 +1866,14 @@ __ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svwrite_ver_za8_u8_v
 void svwrite_ver_za8_u8_vg2(uint64_t, uint32_t, svuint8x2_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svwrite_ver_za8_s8_vg2)))
 void svwrite_ver_za8_s8_vg2(uint64_t, uint32_t, svint8x2_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svwrite_ver_za8_mf8_vg2)))
+void svwrite_ver_za8_mf8_vg2(uint64_t, uint32_t, svmfloat8x2_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svwrite_ver_za8_u8_vg4)))
 void svwrite_ver_za8_u8_vg4(uint64_t, uint32_t, svuint8x4_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svwrite_ver_za8_s8_vg4)))
 void svwrite_ver_za8_s8_vg4(uint64_t, uint32_t, svint8x4_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svwrite_ver_za8_mf8_vg4)))
+void svwrite_ver_za8_mf8_vg4(uint64_t, uint32_t, svmfloat8x4_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svwrite_za16_u16_vg1x2)))
 void svwrite_za16_u16_vg1x2(uint32_t, svuint16x2_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svwrite_za16_bf16_vg1x2)))
@@ -1856,10 +1918,14 @@ __ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svwrite_za8_u8_vg1x2
 void svwrite_za8_u8_vg1x2(uint32_t, svuint8x2_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svwrite_za8_s8_vg1x2)))
 void svwrite_za8_s8_vg1x2(uint32_t, svint8x2_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svwrite_za8_mf8_vg1x2)))
+void svwrite_za8_mf8_vg1x2(uint32_t, svmfloat8x2_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svwrite_za8_u8_vg1x4)))
 void svwrite_za8_u8_vg1x4(uint32_t, svuint8x4_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svwrite_za8_s8_vg1x4)))
 void svwrite_za8_s8_vg1x4(uint32_t, svint8x4_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svwrite_za8_mf8_vg1x4)))
+void svwrite_za8_mf8_vg1x4(uint32_t, svmfloat8x4_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svzero_zt)))
 void svzero_zt(uint64_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sme_svadd_write_single_za32_u32_vg1x2)))
@@ -2338,10 +2404,14 @@ __aio __attribute__((__clang_arm_builtin_alias(__builtin_sme_svwrite_hor_za8_u8_
 void svwrite_hor_za8_vg2(uint64_t, uint32_t, svuint8x2_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sme_svwrite_hor_za8_s8_vg2)))
 void svwrite_hor_za8_vg2(uint64_t, uint32_t, svint8x2_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sme_svwrite_hor_za8_mf8_vg2)))
+void svwrite_hor_za8_vg2(uint64_t, uint32_t, svmfloat8x2_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sme_svwrite_hor_za8_u8_vg4)))
 void svwrite_hor_za8_vg4(uint64_t, uint32_t, svuint8x4_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sme_svwrite_hor_za8_s8_vg4)))
 void svwrite_hor_za8_vg4(uint64_t, uint32_t, svint8x4_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sme_svwrite_hor_za8_mf8_vg4)))
+void svwrite_hor_za8_vg4(uint64_t, uint32_t, svmfloat8x4_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sme_svwrite_ver_za16_u16_vg2)))
 void svwrite_ver_za16_vg2(uint64_t, uint32_t, svuint16x2_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sme_svwrite_ver_za16_bf16_vg2)))
@@ -2386,10 +2456,14 @@ __aio __attribute__((__clang_arm_builtin_alias(__builtin_sme_svwrite_ver_za8_u8_
 void svwrite_ver_za8_vg2(uint64_t, uint32_t, svuint8x2_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sme_svwrite_ver_za8_s8_vg2)))
 void svwrite_ver_za8_vg2(uint64_t, uint32_t, svint8x2_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sme_svwrite_ver_za8_mf8_vg2)))
+void svwrite_ver_za8_vg2(uint64_t, uint32_t, svmfloat8x2_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sme_svwrite_ver_za8_u8_vg4)))
 void svwrite_ver_za8_vg4(uint64_t, uint32_t, svuint8x4_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sme_svwrite_ver_za8_s8_vg4)))
 void svwrite_ver_za8_vg4(uint64_t, uint32_t, svint8x4_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sme_svwrite_ver_za8_mf8_vg4)))
+void svwrite_ver_za8_vg4(uint64_t, uint32_t, svmfloat8x4_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sme_svwrite_za16_u16_vg1x2)))
 void svwrite_za16_vg1x2(uint32_t, svuint16x2_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sme_svwrite_za16_bf16_vg1x2)))
@@ -2434,10 +2508,14 @@ __aio __attribute__((__clang_arm_builtin_alias(__builtin_sme_svwrite_za8_u8_vg1x
 void svwrite_za8_vg1x2(uint32_t, svuint8x2_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sme_svwrite_za8_s8_vg1x2)))
 void svwrite_za8_vg1x2(uint32_t, svint8x2_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sme_svwrite_za8_mf8_vg1x2)))
+void svwrite_za8_vg1x2(uint32_t, svmfloat8x2_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sme_svwrite_za8_u8_vg1x4)))
 void svwrite_za8_vg1x4(uint32_t, svuint8x4_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sme_svwrite_za8_s8_vg1x4)))
 void svwrite_za8_vg1x4(uint32_t, svint8x4_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sme_svwrite_za8_mf8_vg1x4)))
+void svwrite_za8_vg1x4(uint32_t, svmfloat8x4_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svadd_za64_f64_vg1x2)))
 void svadd_za64_f64_vg1x2(uint32_t, svfloat64x2_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svadd_za64_f64_vg1x4)))
@@ -2782,6 +2860,602 @@ __aio __attribute__((__clang_arm_builtin_alias(__builtin_sme_svvdot_lane_za64_s1
 void svvdot_lane_za64_vg1x4(uint32_t, svint16x4_t, svint16_t, uint64_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sme_svvdot_lane_za64_u16_vg1x4)))
 void svvdot_lane_za64_vg1x4(uint32_t, svuint16x4_t, svuint16_t, uint64_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4a_1x1_za32_s8_u8)))
+void svmop4a_1x1_za32_s8_u8(uint64_t, svint8_t, svuint8_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4a_1x1_za32_u8_s8)))
+void svmop4a_1x1_za32_u8_s8(uint64_t, svuint8_t, svint8_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4a_1x1_za32_bf16_bf16)))
+void svmop4a_1x1_za32_bf16_bf16(uint64_t, svbfloat16_t, svbfloat16_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4a_1x1_za32_f16_f16)))
+void svmop4a_1x1_za32_f16_f16(uint64_t, svfloat16_t, svfloat16_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4a_1x1_za32_f32_f32)))
+void svmop4a_1x1_za32_f32_f32(uint64_t, svfloat32_t, svfloat32_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4a_1x1_za32_s8_s8)))
+void svmop4a_1x1_za32_s8_s8(uint64_t, svint8_t, svint8_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4a_1x1_za32_s16_s16)))
+void svmop4a_1x1_za32_s16_s16(uint64_t, svint16_t, svint16_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4a_1x1_za32_u8_u8)))
+void svmop4a_1x1_za32_u8_u8(uint64_t, svuint8_t, svuint8_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4a_1x1_za32_u16_u16)))
+void svmop4a_1x1_za32_u16_u16(uint64_t, svuint16_t, svuint16_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4a_1x2_za32_s8_u8)))
+void svmop4a_1x2_za32_s8_u8(uint64_t, svint8_t, svuint8x2_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4a_1x2_za32_u8_s8)))
+void svmop4a_1x2_za32_u8_s8(uint64_t, svuint8_t, svint8x2_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4a_1x2_za32_bf16_bf16)))
+void svmop4a_1x2_za32_bf16_bf16(uint64_t, svbfloat16_t, svbfloat16x2_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4a_1x2_za32_f16_f16)))
+void svmop4a_1x2_za32_f16_f16(uint64_t, svfloat16_t, svfloat16x2_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4a_1x2_za32_f32_f32)))
+void svmop4a_1x2_za32_f32_f32(uint64_t, svfloat32_t, svfloat32x2_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4a_1x2_za32_s8_s8)))
+void svmop4a_1x2_za32_s8_s8(uint64_t, svint8_t, svint8x2_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4a_1x2_za32_s16_s16)))
+void svmop4a_1x2_za32_s16_s16(uint64_t, svint16_t, svint16x2_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4a_1x2_za32_u8_u8)))
+void svmop4a_1x2_za32_u8_u8(uint64_t, svuint8_t, svuint8x2_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4a_1x2_za32_u16_u16)))
+void svmop4a_1x2_za32_u16_u16(uint64_t, svuint16_t, svuint16x2_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4a_2x1_za32_s8_u8)))
+void svmop4a_2x1_za32_s8_u8(uint64_t, svint8x2_t, svuint8_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4a_2x1_za32_u8_s8)))
+void svmop4a_2x1_za32_u8_s8(uint64_t, svuint8x2_t, svint8_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4a_2x1_za32_bf16_bf16)))
+void svmop4a_2x1_za32_bf16_bf16(uint64_t, svbfloat16x2_t, svbfloat16_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4a_2x1_za32_f16_f16)))
+void svmop4a_2x1_za32_f16_f16(uint64_t, svfloat16x2_t, svfloat16_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4a_2x1_za32_f32_f32)))
+void svmop4a_2x1_za32_f32_f32(uint64_t, svfloat32x2_t, svfloat32_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4a_2x1_za32_s8_s8)))
+void svmop4a_2x1_za32_s8_s8(uint64_t, svint8x2_t, svint8_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4a_2x1_za32_s16_s16)))
+void svmop4a_2x1_za32_s16_s16(uint64_t, svint16x2_t, svint16_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4a_2x1_za32_u8_u8)))
+void svmop4a_2x1_za32_u8_u8(uint64_t, svuint8x2_t, svuint8_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4a_2x1_za32_u16_u16)))
+void svmop4a_2x1_za32_u16_u16(uint64_t, svuint16x2_t, svuint16_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4a_2x2_za32_s8_u8)))
+void svmop4a_2x2_za32_s8_u8(uint64_t, svint8x2_t, svuint8x2_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4a_2x2_za32_u8_s8)))
+void svmop4a_2x2_za32_u8_s8(uint64_t, svuint8x2_t, svint8x2_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4a_2x2_za32_bf16_bf16)))
+void svmop4a_2x2_za32_bf16_bf16(uint64_t, svbfloat16x2_t, svbfloat16x2_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4a_2x2_za32_f16_f16)))
+void svmop4a_2x2_za32_f16_f16(uint64_t, svfloat16x2_t, svfloat16x2_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4a_2x2_za32_f32_f32)))
+void svmop4a_2x2_za32_f32_f32(uint64_t, svfloat32x2_t, svfloat32x2_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4a_2x2_za32_s8_s8)))
+void svmop4a_2x2_za32_s8_s8(uint64_t, svint8x2_t, svint8x2_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4a_2x2_za32_s16_s16)))
+void svmop4a_2x2_za32_s16_s16(uint64_t, svint16x2_t, svint16x2_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4a_2x2_za32_u8_u8)))
+void svmop4a_2x2_za32_u8_u8(uint64_t, svuint8x2_t, svuint8x2_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4a_2x2_za32_u16_u16)))
+void svmop4a_2x2_za32_u16_u16(uint64_t, svuint16x2_t, svuint16x2_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4s_1x1_za32_s8_u8)))
+void svmop4s_1x1_za32_s8_u8(uint64_t, svint8_t, svuint8_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4s_1x1_za32_u8_s8)))
+void svmop4s_1x1_za32_u8_s8(uint64_t, svuint8_t, svint8_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4s_1x1_za32_bf16_bf16)))
+void svmop4s_1x1_za32_bf16_bf16(uint64_t, svbfloat16_t, svbfloat16_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4s_1x1_za32_f16_f16)))
+void svmop4s_1x1_za32_f16_f16(uint64_t, svfloat16_t, svfloat16_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4s_1x1_za32_f32_f32)))
+void svmop4s_1x1_za32_f32_f32(uint64_t, svfloat32_t, svfloat32_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4s_1x1_za32_s8_s8)))
+void svmop4s_1x1_za32_s8_s8(uint64_t, svint8_t, svint8_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4s_1x1_za32_s16_s16)))
+void svmop4s_1x1_za32_s16_s16(uint64_t, svint16_t, svint16_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4s_1x1_za32_u8_u8)))
+void svmop4s_1x1_za32_u8_u8(uint64_t, svuint8_t, svuint8_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4s_1x1_za32_u16_u16)))
+void svmop4s_1x1_za32_u16_u16(uint64_t, svuint16_t, svuint16_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4s_1x2_za32_s8_u8)))
+void svmop4s_1x2_za32_s8_u8(uint64_t, svint8_t, svuint8x2_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4s_1x2_za32_u8_s8)))
+void svmop4s_1x2_za32_u8_s8(uint64_t, svuint8_t, svint8x2_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4s_1x2_za32_bf16_bf16)))
+void svmop4s_1x2_za32_bf16_bf16(uint64_t, svbfloat16_t, svbfloat16x2_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4s_1x2_za32_f16_f16)))
+void svmop4s_1x2_za32_f16_f16(uint64_t, svfloat16_t, svfloat16x2_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4s_1x2_za32_f32_f32)))
+void svmop4s_1x2_za32_f32_f32(uint64_t, svfloat32_t, svfloat32x2_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4s_1x2_za32_s8_s8)))
+void svmop4s_1x2_za32_s8_s8(uint64_t, svint8_t, svint8x2_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4s_1x2_za32_s16_s16)))
+void svmop4s_1x2_za32_s16_s16(uint64_t, svint16_t, svint16x2_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4s_1x2_za32_u8_u8)))
+void svmop4s_1x2_za32_u8_u8(uint64_t, svuint8_t, svuint8x2_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4s_1x2_za32_u16_u16)))
+void svmop4s_1x2_za32_u16_u16(uint64_t, svuint16_t, svuint16x2_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4s_2x1_za32_s8_u8)))
+void svmop4s_2x1_za32_s8_u8(uint64_t, svint8x2_t, svuint8_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4s_2x1_za32_u8_s8)))
+void svmop4s_2x1_za32_u8_s8(uint64_t, svuint8x2_t, svint8_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4s_2x1_za32_bf16_bf16)))
+void svmop4s_2x1_za32_bf16_bf16(uint64_t, svbfloat16x2_t, svbfloat16_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4s_2x1_za32_f16_f16)))
+void svmop4s_2x1_za32_f16_f16(uint64_t, svfloat16x2_t, svfloat16_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4s_2x1_za32_f32_f32)))
+void svmop4s_2x1_za32_f32_f32(uint64_t, svfloat32x2_t, svfloat32_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4s_2x1_za32_s8_s8)))
+void svmop4s_2x1_za32_s8_s8(uint64_t, svint8x2_t, svint8_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4s_2x1_za32_s16_s16)))
+void svmop4s_2x1_za32_s16_s16(uint64_t, svint16x2_t, svint16_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4s_2x1_za32_u8_u8)))
+void svmop4s_2x1_za32_u8_u8(uint64_t, svuint8x2_t, svuint8_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4s_2x1_za32_u16_u16)))
+void svmop4s_2x1_za32_u16_u16(uint64_t, svuint16x2_t, svuint16_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4s_2x2_za32_s8_u8)))
+void svmop4s_2x2_za32_s8_u8(uint64_t, svint8x2_t, svuint8x2_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4s_2x2_za32_u8_s8)))
+void svmop4s_2x2_za32_u8_s8(uint64_t, svuint8x2_t, svint8x2_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4s_2x2_za32_bf16_bf16)))
+void svmop4s_2x2_za32_bf16_bf16(uint64_t, svbfloat16x2_t, svbfloat16x2_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4s_2x2_za32_f16_f16)))
+void svmop4s_2x2_za32_f16_f16(uint64_t, svfloat16x2_t, svfloat16x2_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4s_2x2_za32_f32_f32)))
+void svmop4s_2x2_za32_f32_f32(uint64_t, svfloat32x2_t, svfloat32x2_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4s_2x2_za32_s8_s8)))
+void svmop4s_2x2_za32_s8_s8(uint64_t, svint8x2_t, svint8x2_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4s_2x2_za32_s16_s16)))
+void svmop4s_2x2_za32_s16_s16(uint64_t, svint16x2_t, svint16x2_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4s_2x2_za32_u8_u8)))
+void svmop4s_2x2_za32_u8_u8(uint64_t, svuint8x2_t, svuint8x2_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4s_2x2_za32_u16_u16)))
+void svmop4s_2x2_za32_u16_u16(uint64_t, svuint16x2_t, svuint16x2_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4a_1x1_za32_s8_u8)))
+void svmop4a_za32(uint64_t, svint8_t, svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4a_1x1_za32_u8_s8)))
+void svmop4a_za32(uint64_t, svuint8_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4a_1x1_za32_bf16_bf16)))
+void svmop4a_za32(uint64_t, svbfloat16_t, svbfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4a_1x1_za32_f16_f16)))
+void svmop4a_za32(uint64_t, svfloat16_t, svfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4a_1x1_za32_f32_f32)))
+void svmop4a_za32(uint64_t, svfloat32_t, svfloat32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4a_1x1_za32_s8_s8)))
+void svmop4a_za32(uint64_t, svint8_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4a_1x1_za32_s16_s16)))
+void svmop4a_za32(uint64_t, svint16_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4a_1x1_za32_u8_u8)))
+void svmop4a_za32(uint64_t, svuint8_t, svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4a_1x1_za32_u16_u16)))
+void svmop4a_za32(uint64_t, svuint16_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4a_1x2_za32_s8_u8)))
+void svmop4a_za32(uint64_t, svint8_t, svuint8x2_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4a_1x2_za32_u8_s8)))
+void svmop4a_za32(uint64_t, svuint8_t, svint8x2_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4a_1x2_za32_bf16_bf16)))
+void svmop4a_za32(uint64_t, svbfloat16_t, svbfloat16x2_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4a_1x2_za32_f16_f16)))
+void svmop4a_za32(uint64_t, svfloat16_t, svfloat16x2_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4a_1x2_za32_f32_f32)))
+void svmop4a_za32(uint64_t, svfloat32_t, svfloat32x2_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4a_1x2_za32_s8_s8)))
+void svmop4a_za32(uint64_t, svint8_t, svint8x2_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4a_1x2_za32_s16_s16)))
+void svmop4a_za32(uint64_t, svint16_t, svint16x2_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4a_1x2_za32_u8_u8)))
+void svmop4a_za32(uint64_t, svuint8_t, svuint8x2_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4a_1x2_za32_u16_u16)))
+void svmop4a_za32(uint64_t, svuint16_t, svuint16x2_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4a_2x1_za32_s8_u8)))
+void svmop4a_za32(uint64_t, svint8x2_t, svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4a_2x1_za32_u8_s8)))
+void svmop4a_za32(uint64_t, svuint8x2_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4a_2x1_za32_bf16_bf16)))
+void svmop4a_za32(uint64_t, svbfloat16x2_t, svbfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4a_2x1_za32_f16_f16)))
+void svmop4a_za32(uint64_t, svfloat16x2_t, svfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4a_2x1_za32_f32_f32)))
+void svmop4a_za32(uint64_t, svfloat32x2_t, svfloat32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4a_2x1_za32_s8_s8)))
+void svmop4a_za32(uint64_t, svint8x2_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4a_2x1_za32_s16_s16)))
+void svmop4a_za32(uint64_t, svint16x2_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4a_2x1_za32_u8_u8)))
+void svmop4a_za32(uint64_t, svuint8x2_t, svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4a_2x1_za32_u16_u16)))
+void svmop4a_za32(uint64_t, svuint16x2_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4a_2x2_za32_s8_u8)))
+void svmop4a_za32(uint64_t, svint8x2_t, svuint8x2_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4a_2x2_za32_u8_s8)))
+void svmop4a_za32(uint64_t, svuint8x2_t, svint8x2_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4a_2x2_za32_bf16_bf16)))
+void svmop4a_za32(uint64_t, svbfloat16x2_t, svbfloat16x2_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4a_2x2_za32_f16_f16)))
+void svmop4a_za32(uint64_t, svfloat16x2_t, svfloat16x2_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4a_2x2_za32_f32_f32)))
+void svmop4a_za32(uint64_t, svfloat32x2_t, svfloat32x2_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4a_2x2_za32_s8_s8)))
+void svmop4a_za32(uint64_t, svint8x2_t, svint8x2_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4a_2x2_za32_s16_s16)))
+void svmop4a_za32(uint64_t, svint16x2_t, svint16x2_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4a_2x2_za32_u8_u8)))
+void svmop4a_za32(uint64_t, svuint8x2_t, svuint8x2_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4a_2x2_za32_u16_u16)))
+void svmop4a_za32(uint64_t, svuint16x2_t, svuint16x2_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4s_1x1_za32_s8_u8)))
+void svmop4s_za32(uint64_t, svint8_t, svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4s_1x1_za32_u8_s8)))
+void svmop4s_za32(uint64_t, svuint8_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4s_1x1_za32_bf16_bf16)))
+void svmop4s_za32(uint64_t, svbfloat16_t, svbfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4s_1x1_za32_f16_f16)))
+void svmop4s_za32(uint64_t, svfloat16_t, svfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4s_1x1_za32_f32_f32)))
+void svmop4s_za32(uint64_t, svfloat32_t, svfloat32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4s_1x1_za32_s8_s8)))
+void svmop4s_za32(uint64_t, svint8_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4s_1x1_za32_s16_s16)))
+void svmop4s_za32(uint64_t, svint16_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4s_1x1_za32_u8_u8)))
+void svmop4s_za32(uint64_t, svuint8_t, svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4s_1x1_za32_u16_u16)))
+void svmop4s_za32(uint64_t, svuint16_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4s_1x2_za32_s8_u8)))
+void svmop4s_za32(uint64_t, svint8_t, svuint8x2_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4s_1x2_za32_u8_s8)))
+void svmop4s_za32(uint64_t, svuint8_t, svint8x2_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4s_1x2_za32_bf16_bf16)))
+void svmop4s_za32(uint64_t, svbfloat16_t, svbfloat16x2_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4s_1x2_za32_f16_f16)))
+void svmop4s_za32(uint64_t, svfloat16_t, svfloat16x2_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4s_1x2_za32_f32_f32)))
+void svmop4s_za32(uint64_t, svfloat32_t, svfloat32x2_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4s_1x2_za32_s8_s8)))
+void svmop4s_za32(uint64_t, svint8_t, svint8x2_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4s_1x2_za32_s16_s16)))
+void svmop4s_za32(uint64_t, svint16_t, svint16x2_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4s_1x2_za32_u8_u8)))
+void svmop4s_za32(uint64_t, svuint8_t, svuint8x2_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4s_1x2_za32_u16_u16)))
+void svmop4s_za32(uint64_t, svuint16_t, svuint16x2_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4s_2x1_za32_s8_u8)))
+void svmop4s_za32(uint64_t, svint8x2_t, svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4s_2x1_za32_u8_s8)))
+void svmop4s_za32(uint64_t, svuint8x2_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4s_2x1_za32_bf16_bf16)))
+void svmop4s_za32(uint64_t, svbfloat16x2_t, svbfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4s_2x1_za32_f16_f16)))
+void svmop4s_za32(uint64_t, svfloat16x2_t, svfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4s_2x1_za32_f32_f32)))
+void svmop4s_za32(uint64_t, svfloat32x2_t, svfloat32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4s_2x1_za32_s8_s8)))
+void svmop4s_za32(uint64_t, svint8x2_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4s_2x1_za32_s16_s16)))
+void svmop4s_za32(uint64_t, svint16x2_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4s_2x1_za32_u8_u8)))
+void svmop4s_za32(uint64_t, svuint8x2_t, svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4s_2x1_za32_u16_u16)))
+void svmop4s_za32(uint64_t, svuint16x2_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4s_2x2_za32_s8_u8)))
+void svmop4s_za32(uint64_t, svint8x2_t, svuint8x2_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4s_2x2_za32_u8_s8)))
+void svmop4s_za32(uint64_t, svuint8x2_t, svint8x2_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4s_2x2_za32_bf16_bf16)))
+void svmop4s_za32(uint64_t, svbfloat16x2_t, svbfloat16x2_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4s_2x2_za32_f16_f16)))
+void svmop4s_za32(uint64_t, svfloat16x2_t, svfloat16x2_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4s_2x2_za32_f32_f32)))
+void svmop4s_za32(uint64_t, svfloat32x2_t, svfloat32x2_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4s_2x2_za32_s8_s8)))
+void svmop4s_za32(uint64_t, svint8x2_t, svint8x2_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4s_2x2_za32_s16_s16)))
+void svmop4s_za32(uint64_t, svint16x2_t, svint16x2_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4s_2x2_za32_u8_u8)))
+void svmop4s_za32(uint64_t, svuint8x2_t, svuint8x2_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4s_2x2_za32_u16_u16)))
+void svmop4s_za32(uint64_t, svuint16x2_t, svuint16x2_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4a_1x1_za16_bf16_bf16)))
+void svmop4a_1x1_za16_bf16_bf16(uint64_t, svbfloat16_t, svbfloat16_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4a_1x2_za16_bf16_bf16)))
+void svmop4a_1x2_za16_bf16_bf16(uint64_t, svbfloat16_t, svbfloat16x2_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4a_2x1_za16_bf16_bf16)))
+void svmop4a_2x1_za16_bf16_bf16(uint64_t, svbfloat16x2_t, svbfloat16_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4a_2x2_za16_bf16_bf16)))
+void svmop4a_2x2_za16_bf16_bf16(uint64_t, svbfloat16x2_t, svbfloat16x2_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4s_1x1_za16_bf16_bf16)))
+void svmop4s_1x1_za16_bf16_bf16(uint64_t, svbfloat16_t, svbfloat16_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4s_1x2_za16_bf16_bf16)))
+void svmop4s_1x2_za16_bf16_bf16(uint64_t, svbfloat16_t, svbfloat16x2_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4s_2x1_za16_bf16_bf16)))
+void svmop4s_2x1_za16_bf16_bf16(uint64_t, svbfloat16x2_t, svbfloat16_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4s_2x2_za16_bf16_bf16)))
+void svmop4s_2x2_za16_bf16_bf16(uint64_t, svbfloat16x2_t, svbfloat16x2_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4a_1x1_za16_bf16_bf16)))
+void svmop4a_za16(uint64_t, svbfloat16_t, svbfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4a_1x2_za16_bf16_bf16)))
+void svmop4a_za16(uint64_t, svbfloat16_t, svbfloat16x2_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4a_2x1_za16_bf16_bf16)))
+void svmop4a_za16(uint64_t, svbfloat16x2_t, svbfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4a_2x2_za16_bf16_bf16)))
+void svmop4a_za16(uint64_t, svbfloat16x2_t, svbfloat16x2_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4s_1x1_za16_bf16_bf16)))
+void svmop4s_za16(uint64_t, svbfloat16_t, svbfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4s_1x2_za16_bf16_bf16)))
+void svmop4s_za16(uint64_t, svbfloat16_t, svbfloat16x2_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4s_2x1_za16_bf16_bf16)))
+void svmop4s_za16(uint64_t, svbfloat16x2_t, svbfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4s_2x2_za16_bf16_bf16)))
+void svmop4s_za16(uint64_t, svbfloat16x2_t, svbfloat16x2_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4a_1x1_za16_f16_f16)))
+void svmop4a_1x1_za16_f16_f16(uint64_t, svfloat16_t, svfloat16_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4a_1x2_za16_f16_f16)))
+void svmop4a_1x2_za16_f16_f16(uint64_t, svfloat16_t, svfloat16x2_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4a_2x1_za16_f16_f16)))
+void svmop4a_2x1_za16_f16_f16(uint64_t, svfloat16x2_t, svfloat16_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4a_2x2_za16_f16_f16)))
+void svmop4a_2x2_za16_f16_f16(uint64_t, svfloat16x2_t, svfloat16x2_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4s_1x1_za16_f16_f16)))
+void svmop4s_1x1_za16_f16_f16(uint64_t, svfloat16_t, svfloat16_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4s_1x2_za16_f16_f16)))
+void svmop4s_1x2_za16_f16_f16(uint64_t, svfloat16_t, svfloat16x2_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4s_2x1_za16_f16_f16)))
+void svmop4s_2x1_za16_f16_f16(uint64_t, svfloat16x2_t, svfloat16_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4s_2x2_za16_f16_f16)))
+void svmop4s_2x2_za16_f16_f16(uint64_t, svfloat16x2_t, svfloat16x2_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4a_1x1_za16_f16_f16)))
+void svmop4a_za16(uint64_t, svfloat16_t, svfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4a_1x2_za16_f16_f16)))
+void svmop4a_za16(uint64_t, svfloat16_t, svfloat16x2_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4a_2x1_za16_f16_f16)))
+void svmop4a_za16(uint64_t, svfloat16x2_t, svfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4a_2x2_za16_f16_f16)))
+void svmop4a_za16(uint64_t, svfloat16x2_t, svfloat16x2_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4s_1x1_za16_f16_f16)))
+void svmop4s_za16(uint64_t, svfloat16_t, svfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4s_1x2_za16_f16_f16)))
+void svmop4s_za16(uint64_t, svfloat16_t, svfloat16x2_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4s_2x1_za16_f16_f16)))
+void svmop4s_za16(uint64_t, svfloat16x2_t, svfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4s_2x2_za16_f16_f16)))
+void svmop4s_za16(uint64_t, svfloat16x2_t, svfloat16x2_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4a_1x1_za64_f64_f64)))
+void svmop4a_1x1_za64_f64_f64(uint64_t, svfloat64_t, svfloat64_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4a_1x2_za64_f64_f64)))
+void svmop4a_1x2_za64_f64_f64(uint64_t, svfloat64_t, svfloat64x2_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4a_2x1_za64_f64_f64)))
+void svmop4a_2x1_za64_f64_f64(uint64_t, svfloat64x2_t, svfloat64_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4a_2x2_za64_f64_f64)))
+void svmop4a_2x2_za64_f64_f64(uint64_t, svfloat64x2_t, svfloat64x2_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4s_1x1_za64_f64_f64)))
+void svmop4s_1x1_za64_f64_f64(uint64_t, svfloat64_t, svfloat64_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4s_1x2_za64_f64_f64)))
+void svmop4s_1x2_za64_f64_f64(uint64_t, svfloat64_t, svfloat64x2_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4s_2x1_za64_f64_f64)))
+void svmop4s_2x1_za64_f64_f64(uint64_t, svfloat64x2_t, svfloat64_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4s_2x2_za64_f64_f64)))
+void svmop4s_2x2_za64_f64_f64(uint64_t, svfloat64x2_t, svfloat64x2_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4a_1x1_za64_f64_f64)))
+void svmop4a_za64(uint64_t, svfloat64_t, svfloat64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4a_1x2_za64_f64_f64)))
+void svmop4a_za64(uint64_t, svfloat64_t, svfloat64x2_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4a_2x1_za64_f64_f64)))
+void svmop4a_za64(uint64_t, svfloat64x2_t, svfloat64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4a_2x2_za64_f64_f64)))
+void svmop4a_za64(uint64_t, svfloat64x2_t, svfloat64x2_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4s_1x1_za64_f64_f64)))
+void svmop4s_za64(uint64_t, svfloat64_t, svfloat64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4s_1x2_za64_f64_f64)))
+void svmop4s_za64(uint64_t, svfloat64_t, svfloat64x2_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4s_2x1_za64_f64_f64)))
+void svmop4s_za64(uint64_t, svfloat64x2_t, svfloat64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4s_2x2_za64_f64_f64)))
+void svmop4s_za64(uint64_t, svfloat64x2_t, svfloat64x2_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4a_1x1_za16_mf8_mf8_fpm)))
+void svmop4a_1x1_za16_mf8_mf8_fpm(uint64_t, svmfloat8_t, svmfloat8_t, fpm_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4a_1x2_za16_mf8_mf8_fpm)))
+void svmop4a_1x2_za16_mf8_mf8_fpm(uint64_t, svmfloat8_t, svmfloat8x2_t, fpm_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4a_2x1_za16_mf8_mf8_fpm)))
+void svmop4a_2x1_za16_mf8_mf8_fpm(uint64_t, svmfloat8x2_t, svmfloat8_t, fpm_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4a_2x2_za16_mf8_mf8_fpm)))
+void svmop4a_2x2_za16_mf8_mf8_fpm(uint64_t, svmfloat8x2_t, svmfloat8x2_t, fpm_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4a_1x1_za16_mf8_mf8_fpm)))
+void svmop4a_za16_fpm(uint64_t, svmfloat8_t, svmfloat8_t, fpm_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4a_1x2_za16_mf8_mf8_fpm)))
+void svmop4a_za16_fpm(uint64_t, svmfloat8_t, svmfloat8x2_t, fpm_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4a_2x1_za16_mf8_mf8_fpm)))
+void svmop4a_za16_fpm(uint64_t, svmfloat8x2_t, svmfloat8_t, fpm_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4a_2x2_za16_mf8_mf8_fpm)))
+void svmop4a_za16_fpm(uint64_t, svmfloat8x2_t, svmfloat8x2_t, fpm_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4a_1x1_za32_mf8_mf8_fpm)))
+void svmop4a_1x1_za32_mf8_mf8_fpm(uint64_t, svmfloat8_t, svmfloat8_t, fpm_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4a_1x2_za32_mf8_mf8_fpm)))
+void svmop4a_1x2_za32_mf8_mf8_fpm(uint64_t, svmfloat8_t, svmfloat8x2_t, fpm_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4a_2x1_za32_mf8_mf8_fpm)))
+void svmop4a_2x1_za32_mf8_mf8_fpm(uint64_t, svmfloat8x2_t, svmfloat8_t, fpm_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4a_2x2_za32_mf8_mf8_fpm)))
+void svmop4a_2x2_za32_mf8_mf8_fpm(uint64_t, svmfloat8x2_t, svmfloat8x2_t, fpm_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4a_1x1_za32_mf8_mf8_fpm)))
+void svmop4a_za32_fpm(uint64_t, svmfloat8_t, svmfloat8_t, fpm_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4a_1x2_za32_mf8_mf8_fpm)))
+void svmop4a_za32_fpm(uint64_t, svmfloat8_t, svmfloat8x2_t, fpm_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4a_2x1_za32_mf8_mf8_fpm)))
+void svmop4a_za32_fpm(uint64_t, svmfloat8x2_t, svmfloat8_t, fpm_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4a_2x2_za32_mf8_mf8_fpm)))
+void svmop4a_za32_fpm(uint64_t, svmfloat8x2_t, svmfloat8x2_t, fpm_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4a_1x1_za64_s16_u16)))
+void svmop4a_1x1_za64_s16_u16(uint64_t, svint16_t, svuint16_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4a_1x1_za64_u16_s16)))
+void svmop4a_1x1_za64_u16_s16(uint64_t, svuint16_t, svint16_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4a_1x1_za64_s16_s16)))
+void svmop4a_1x1_za64_s16_s16(uint64_t, svint16_t, svint16_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4a_1x1_za64_u16_u16)))
+void svmop4a_1x1_za64_u16_u16(uint64_t, svuint16_t, svuint16_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4a_1x2_za64_s16_u16)))
+void svmop4a_1x2_za64_s16_u16(uint64_t, svint16_t, svuint16x2_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4a_1x2_za64_u16_s16)))
+void svmop4a_1x2_za64_u16_s16(uint64_t, svuint16_t, svint16x2_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4a_1x2_za64_s16_s16)))
+void svmop4a_1x2_za64_s16_s16(uint64_t, svint16_t, svint16x2_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4a_1x2_za64_u16_u16)))
+void svmop4a_1x2_za64_u16_u16(uint64_t, svuint16_t, svuint16x2_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4a_2x1_za64_s16_u16)))
+void svmop4a_2x1_za64_s16_u16(uint64_t, svint16x2_t, svuint16_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4a_2x1_za64_u16_s16)))
+void svmop4a_2x1_za64_u16_s16(uint64_t, svuint16x2_t, svint16_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4a_2x1_za64_s16_s16)))
+void svmop4a_2x1_za64_s16_s16(uint64_t, svint16x2_t, svint16_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4a_2x1_za64_u16_u16)))
+void svmop4a_2x1_za64_u16_u16(uint64_t, svuint16x2_t, svuint16_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4a_2x2_za64_s16_u16)))
+void svmop4a_2x2_za64_s16_u16(uint64_t, svint16x2_t, svuint16x2_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4a_2x2_za64_u16_s16)))
+void svmop4a_2x2_za64_u16_s16(uint64_t, svuint16x2_t, svint16x2_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4a_2x2_za64_s16_s16)))
+void svmop4a_2x2_za64_s16_s16(uint64_t, svint16x2_t, svint16x2_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4a_2x2_za64_u16_u16)))
+void svmop4a_2x2_za64_u16_u16(uint64_t, svuint16x2_t, svuint16x2_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4s_1x1_za64_s16_u16)))
+void svmop4s_1x1_za64_s16_u16(uint64_t, svint16_t, svuint16_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4s_1x1_za64_u16_s16)))
+void svmop4s_1x1_za64_u16_s16(uint64_t, svuint16_t, svint16_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4s_1x1_za64_s16_s16)))
+void svmop4s_1x1_za64_s16_s16(uint64_t, svint16_t, svint16_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4s_1x1_za64_u16_u16)))
+void svmop4s_1x1_za64_u16_u16(uint64_t, svuint16_t, svuint16_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4s_1x2_za64_s16_u16)))
+void svmop4s_1x2_za64_s16_u16(uint64_t, svint16_t, svuint16x2_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4s_1x2_za64_u16_s16)))
+void svmop4s_1x2_za64_u16_s16(uint64_t, svuint16_t, svint16x2_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4s_1x2_za64_s16_s16)))
+void svmop4s_1x2_za64_s16_s16(uint64_t, svint16_t, svint16x2_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4s_1x2_za64_u16_u16)))
+void svmop4s_1x2_za64_u16_u16(uint64_t, svuint16_t, svuint16x2_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4s_2x1_za64_s16_u16)))
+void svmop4s_2x1_za64_s16_u16(uint64_t, svint16x2_t, svuint16_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4s_2x1_za64_u16_s16)))
+void svmop4s_2x1_za64_u16_s16(uint64_t, svuint16x2_t, svint16_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4s_2x1_za64_s16_s16)))
+void svmop4s_2x1_za64_s16_s16(uint64_t, svint16x2_t, svint16_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4s_2x1_za64_u16_u16)))
+void svmop4s_2x1_za64_u16_u16(uint64_t, svuint16x2_t, svuint16_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4s_2x2_za64_s16_u16)))
+void svmop4s_2x2_za64_s16_u16(uint64_t, svint16x2_t, svuint16x2_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4s_2x2_za64_u16_s16)))
+void svmop4s_2x2_za64_u16_s16(uint64_t, svuint16x2_t, svint16x2_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4s_2x2_za64_s16_s16)))
+void svmop4s_2x2_za64_s16_s16(uint64_t, svint16x2_t, svint16x2_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4s_2x2_za64_u16_u16)))
+void svmop4s_2x2_za64_u16_u16(uint64_t, svuint16x2_t, svuint16x2_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4a_1x1_za64_s16_u16)))
+void svmop4a_za64(uint64_t, svint16_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4a_1x1_za64_u16_s16)))
+void svmop4a_za64(uint64_t, svuint16_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4a_1x1_za64_s16_s16)))
+void svmop4a_za64(uint64_t, svint16_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4a_1x1_za64_u16_u16)))
+void svmop4a_za64(uint64_t, svuint16_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4a_1x2_za64_s16_u16)))
+void svmop4a_za64(uint64_t, svint16_t, svuint16x2_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4a_1x2_za64_u16_s16)))
+void svmop4a_za64(uint64_t, svuint16_t, svint16x2_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4a_1x2_za64_s16_s16)))
+void svmop4a_za64(uint64_t, svint16_t, svint16x2_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4a_1x2_za64_u16_u16)))
+void svmop4a_za64(uint64_t, svuint16_t, svuint16x2_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4a_2x1_za64_s16_u16)))
+void svmop4a_za64(uint64_t, svint16x2_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4a_2x1_za64_u16_s16)))
+void svmop4a_za64(uint64_t, svuint16x2_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4a_2x1_za64_s16_s16)))
+void svmop4a_za64(uint64_t, svint16x2_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4a_2x1_za64_u16_u16)))
+void svmop4a_za64(uint64_t, svuint16x2_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4a_2x2_za64_s16_u16)))
+void svmop4a_za64(uint64_t, svint16x2_t, svuint16x2_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4a_2x2_za64_u16_s16)))
+void svmop4a_za64(uint64_t, svuint16x2_t, svint16x2_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4a_2x2_za64_s16_s16)))
+void svmop4a_za64(uint64_t, svint16x2_t, svint16x2_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4a_2x2_za64_u16_u16)))
+void svmop4a_za64(uint64_t, svuint16x2_t, svuint16x2_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4s_1x1_za64_s16_u16)))
+void svmop4s_za64(uint64_t, svint16_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4s_1x1_za64_u16_s16)))
+void svmop4s_za64(uint64_t, svuint16_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4s_1x1_za64_s16_s16)))
+void svmop4s_za64(uint64_t, svint16_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4s_1x1_za64_u16_u16)))
+void svmop4s_za64(uint64_t, svuint16_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4s_1x2_za64_s16_u16)))
+void svmop4s_za64(uint64_t, svint16_t, svuint16x2_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4s_1x2_za64_u16_s16)))
+void svmop4s_za64(uint64_t, svuint16_t, svint16x2_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4s_1x2_za64_s16_s16)))
+void svmop4s_za64(uint64_t, svint16_t, svint16x2_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4s_1x2_za64_u16_u16)))
+void svmop4s_za64(uint64_t, svuint16_t, svuint16x2_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4s_2x1_za64_s16_u16)))
+void svmop4s_za64(uint64_t, svint16x2_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4s_2x1_za64_u16_s16)))
+void svmop4s_za64(uint64_t, svuint16x2_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4s_2x1_za64_s16_s16)))
+void svmop4s_za64(uint64_t, svint16x2_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4s_2x1_za64_u16_u16)))
+void svmop4s_za64(uint64_t, svuint16x2_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4s_2x2_za64_s16_u16)))
+void svmop4s_za64(uint64_t, svint16x2_t, svuint16x2_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4s_2x2_za64_u16_s16)))
+void svmop4s_za64(uint64_t, svuint16x2_t, svint16x2_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4s_2x2_za64_s16_s16)))
+void svmop4s_za64(uint64_t, svint16x2_t, svint16x2_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sme_svmop4s_2x2_za64_u16_u16)))
+void svmop4s_za64(uint64_t, svuint16x2_t, svuint16x2_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svtmopa_lane_za32_s8_u8)))
+void svtmopa_lane_za32_s8_u8(uint64_t, svint8x2_t, svuint8_t, svuint8_t, uint64_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svtmopa_lane_za32_u8_s8)))
+void svtmopa_lane_za32_u8_s8(uint64_t, svuint8x2_t, svint8_t, svuint8_t, uint64_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svtmopa_lane_za32_s8_s8)))
+void svtmopa_lane_za32_s8_s8(uint64_t, svint8x2_t, svint8_t, svuint8_t, uint64_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svtmopa_lane_za32_s16_s16)))
+void svtmopa_lane_za32_s16_s16(uint64_t, svint16x2_t, svint16_t, svuint8_t, uint64_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svtmopa_lane_za32_bf16_bf16)))
+void svtmopa_lane_za32_bf16_bf16(uint64_t, svbfloat16x2_t, svbfloat16_t, svuint8_t, uint64_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svtmopa_lane_za32_f32_f32)))
+void svtmopa_lane_za32_f32_f32(uint64_t, svfloat32x2_t, svfloat32_t, svuint8_t, uint64_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svtmopa_lane_za32_f16_f16)))
+void svtmopa_lane_za32_f16_f16(uint64_t, svfloat16x2_t, svfloat16_t, svuint8_t, uint64_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svtmopa_lane_za32_u8_u8)))
+void svtmopa_lane_za32_u8_u8(uint64_t, svuint8x2_t, svuint8_t, svuint8_t, uint64_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svtmopa_lane_za32_u16_u16)))
+void svtmopa_lane_za32_u16_u16(uint64_t, svuint16x2_t, svuint16_t, svuint8_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sme_svtmopa_lane_za32_s8_u8)))
+void svtmopa_lane_za32(uint64_t, svint8x2_t, svuint8_t, svuint8_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sme_svtmopa_lane_za32_u8_s8)))
+void svtmopa_lane_za32(uint64_t, svuint8x2_t, svint8_t, svuint8_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sme_svtmopa_lane_za32_s8_s8)))
+void svtmopa_lane_za32(uint64_t, svint8x2_t, svint8_t, svuint8_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sme_svtmopa_lane_za32_s16_s16)))
+void svtmopa_lane_za32(uint64_t, svint16x2_t, svint16_t, svuint8_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sme_svtmopa_lane_za32_bf16_bf16)))
+void svtmopa_lane_za32(uint64_t, svbfloat16x2_t, svbfloat16_t, svuint8_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sme_svtmopa_lane_za32_f32_f32)))
+void svtmopa_lane_za32(uint64_t, svfloat32x2_t, svfloat32_t, svuint8_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sme_svtmopa_lane_za32_f16_f16)))
+void svtmopa_lane_za32(uint64_t, svfloat16x2_t, svfloat16_t, svuint8_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sme_svtmopa_lane_za32_u8_u8)))
+void svtmopa_lane_za32(uint64_t, svuint8x2_t, svuint8_t, svuint8_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sme_svtmopa_lane_za32_u16_u16)))
+void svtmopa_lane_za32(uint64_t, svuint16x2_t, svuint16_t, svuint8_t, uint64_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svtmopa_lane_za16_bf16_bf16)))
+void svtmopa_lane_za16_bf16_bf16(uint64_t, svbfloat16x2_t, svbfloat16_t, svuint8_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sme_svtmopa_lane_za16_bf16_bf16)))
+void svtmopa_lane_za16(uint64_t, svbfloat16x2_t, svbfloat16_t, svuint8_t, uint64_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svtmopa_lane_za16_f16_f16)))
+void svtmopa_lane_za16_f16_f16(uint64_t, svfloat16x2_t, svfloat16_t, svuint8_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sme_svtmopa_lane_za16_f16_f16)))
+void svtmopa_lane_za16(uint64_t, svfloat16x2_t, svfloat16_t, svuint8_t, uint64_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svtmopa_lane_za16_mf8_mf8_fpm)))
+void svtmopa_lane_za16_mf8_mf8_fpm(uint64_t, svmfloat8x2_t, svmfloat8_t, svuint8_t, uint64_t, fpm_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sme_svtmopa_lane_za16_mf8_mf8_fpm)))
+void svtmopa_lane_za16_fpm(uint64_t, svmfloat8x2_t, svmfloat8_t, svuint8_t, uint64_t, fpm_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svtmopa_lane_za32_mf8_mf8_fpm)))
+void svtmopa_lane_za32_mf8_mf8_fpm(uint64_t, svmfloat8x2_t, svmfloat8_t, svuint8_t, uint64_t, fpm_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sme_svtmopa_lane_za32_mf8_mf8_fpm)))
+void svtmopa_lane_za32_fpm(uint64_t, svmfloat8x2_t, svmfloat8_t, svuint8_t, uint64_t, fpm_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svreadz_hor_za128_u8)))
 svuint8_t svreadz_hor_za128_u8(uint64_t, uint32_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svreadz_hor_za128_u32)))
@@ -2804,6 +3478,8 @@ __ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svreadz_hor_za128_s3
 svint32_t svreadz_hor_za128_s32(uint64_t, uint32_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svreadz_hor_za128_s64)))
 svint64_t svreadz_hor_za128_s64(uint64_t, uint32_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svreadz_hor_za128_mf8)))
+svmfloat8_t svreadz_hor_za128_mf8(uint64_t, uint32_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svreadz_hor_za128_s16)))
 svint16_t svreadz_hor_za128_s16(uint64_t, uint32_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svreadz_hor_za16_u16)))
@@ -2870,14 +3546,20 @@ __ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svreadz_hor_za8_u8))
 svuint8_t svreadz_hor_za8_u8(uint64_t, uint32_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svreadz_hor_za8_s8)))
 svint8_t svreadz_hor_za8_s8(uint64_t, uint32_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svreadz_hor_za8_mf8)))
+svmfloat8_t svreadz_hor_za8_mf8(uint64_t, uint32_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svreadz_hor_za8_u8_vg2)))
 svuint8x2_t svreadz_hor_za8_u8_vg2(uint64_t, uint32_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svreadz_hor_za8_s8_vg2)))
 svint8x2_t svreadz_hor_za8_s8_vg2(uint64_t, uint32_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svreadz_hor_za8_mf8_vg2)))
+svmfloat8x2_t svreadz_hor_za8_mf8_vg2(uint64_t, uint32_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svreadz_hor_za8_u8_vg4)))
 svuint8x4_t svreadz_hor_za8_u8_vg4(uint64_t, uint32_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svreadz_hor_za8_s8_vg4)))
 svint8x4_t svreadz_hor_za8_s8_vg4(uint64_t, uint32_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svreadz_hor_za8_mf8_vg4)))
+svmfloat8x4_t svreadz_hor_za8_mf8_vg4(uint64_t, uint32_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svreadz_ver_za128_u8)))
 svuint8_t svreadz_ver_za128_u8(uint64_t, uint32_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svreadz_ver_za128_u32)))
@@ -2900,6 +3582,8 @@ __ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svreadz_ver_za128_s3
 svint32_t svreadz_ver_za128_s32(uint64_t, uint32_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svreadz_ver_za128_s64)))
 svint64_t svreadz_ver_za128_s64(uint64_t, uint32_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svreadz_ver_za128_mf8)))
+svmfloat8_t svreadz_ver_za128_mf8(uint64_t, uint32_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svreadz_ver_za128_s16)))
 svint16_t svreadz_ver_za128_s16(uint64_t, uint32_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svreadz_ver_za16_u16)))
@@ -2966,14 +3650,20 @@ __ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svreadz_ver_za8_u8))
 svuint8_t svreadz_ver_za8_u8(uint64_t, uint32_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svreadz_ver_za8_s8)))
 svint8_t svreadz_ver_za8_s8(uint64_t, uint32_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svreadz_ver_za8_mf8)))
+svmfloat8_t svreadz_ver_za8_mf8(uint64_t, uint32_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svreadz_ver_za8_u8_vg2)))
 svuint8x2_t svreadz_ver_za8_u8_vg2(uint64_t, uint32_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svreadz_ver_za8_s8_vg2)))
 svint8x2_t svreadz_ver_za8_s8_vg2(uint64_t, uint32_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svreadz_ver_za8_mf8_vg2)))
+svmfloat8x2_t svreadz_ver_za8_mf8_vg2(uint64_t, uint32_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svreadz_ver_za8_u8_vg4)))
 svuint8x4_t svreadz_ver_za8_u8_vg4(uint64_t, uint32_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svreadz_ver_za8_s8_vg4)))
 svint8x4_t svreadz_ver_za8_s8_vg4(uint64_t, uint32_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svreadz_ver_za8_mf8_vg4)))
+svmfloat8x4_t svreadz_ver_za8_mf8_vg4(uint64_t, uint32_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svreadz_za16_u16_vg1x2)))
 svuint16x2_t svreadz_za16_u16_vg1x2(uint32_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svreadz_za16_bf16_vg1x2)))
@@ -3018,10 +3708,14 @@ __ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svreadz_za8_u8_vg1x2
 svuint8x2_t svreadz_za8_u8_vg1x2(uint32_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svreadz_za8_s8_vg1x2)))
 svint8x2_t svreadz_za8_s8_vg1x2(uint32_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svreadz_za8_mf8_vg1x2)))
+svmfloat8x2_t svreadz_za8_mf8_vg1x2(uint32_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svreadz_za8_u8_vg1x4)))
 svuint8x4_t svreadz_za8_u8_vg1x4(uint32_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svreadz_za8_s8_vg1x4)))
 svint8x4_t svreadz_za8_s8_vg1x4(uint32_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svreadz_za8_mf8_vg1x4)))
+svmfloat8x4_t svreadz_za8_mf8_vg1x4(uint32_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svzero_za64_vg1x2)))
 void svzero_za64_vg1x2(uint32_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sme_svzero_za64_vg1x4)))

lib/include/arm_sve.h

@@ -3359,6 +3359,8 @@ __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsel_s32_x2)))
 svint32x2_t svsel_s32_x2(svcount_t, svint32x2_t, svint32x2_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsel_s64_x2)))
 svint64x2_t svsel_s64_x2(svcount_t, svint64x2_t, svint64x2_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsel_mf8_x2)))
+svmfloat8x2_t svsel_mf8_x2(svcount_t, svmfloat8x2_t, svmfloat8x2_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsel_s16_x2)))
 svint16x2_t svsel_s16_x2(svcount_t, svint16x2_t, svint16x2_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsel_u8_x4)))
@@ -3383,6 +3385,8 @@ __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsel_s32_x4)))
 svint32x4_t svsel_s32_x4(svcount_t, svint32x4_t, svint32x4_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsel_s64_x4)))
 svint64x4_t svsel_s64_x4(svcount_t, svint64x4_t, svint64x4_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsel_mf8_x4)))
+svmfloat8x4_t svsel_mf8_x4(svcount_t, svmfloat8x4_t, svmfloat8x4_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsel_s16_x4)))
 svint16x4_t svsel_s16_x4(svcount_t, svint16x4_t, svint16x4_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svunpk_s32_s16_x2)))
@@ -3431,6 +3435,8 @@ __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzp_s32_x2)))
 svint32x2_t svuzp_s32_x2(svint32x2_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzp_s64_x2)))
 svint64x2_t svuzp_s64_x2(svint64x2_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzp_mf8_x2)))
+svmfloat8x2_t svuzp_mf8_x2(svmfloat8x2_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzp_s16_x2)))
 svint16x2_t svuzp_s16_x2(svint16x2_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzp_u8_x4)))
@@ -3455,6 +3461,8 @@ __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzp_s32_x4)))
 svint32x4_t svuzp_s32_x4(svint32x4_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzp_s64_x4)))
 svint64x4_t svuzp_s64_x4(svint64x4_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzp_mf8_x4)))
+svmfloat8x4_t svuzp_mf8_x4(svmfloat8x4_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzp_s16_x4)))
 svint16x4_t svuzp_s16_x4(svint16x4_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzpq_u8_x2)))
@@ -3479,6 +3487,8 @@ __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzpq_s32_x2)))
 svint32x2_t svuzpq_s32_x2(svint32x2_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzpq_s64_x2)))
 svint64x2_t svuzpq_s64_x2(svint64x2_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzpq_mf8_x2)))
+svmfloat8x2_t svuzpq_mf8_x2(svmfloat8x2_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzpq_s16_x2)))
 svint16x2_t svuzpq_s16_x2(svint16x2_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzpq_u8_x4)))
@@ -3503,6 +3513,8 @@ __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzpq_s32_x4)))
 svint32x4_t svuzpq_s32_x4(svint32x4_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzpq_s64_x4)))
 svint64x4_t svuzpq_s64_x4(svint64x4_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzpq_mf8_x4)))
+svmfloat8x4_t svuzpq_mf8_x4(svmfloat8x4_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzpq_s16_x4)))
 svint16x4_t svuzpq_s16_x4(svint16x4_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzip_u8_x2)))
@@ -3527,6 +3539,8 @@ __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzip_s32_x2)))
 svint32x2_t svzip_s32_x2(svint32x2_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzip_s64_x2)))
 svint64x2_t svzip_s64_x2(svint64x2_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzip_mf8_x2)))
+svmfloat8x2_t svzip_mf8_x2(svmfloat8x2_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzip_s16_x2)))
 svint16x2_t svzip_s16_x2(svint16x2_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzip_u8_x4)))
@@ -3551,6 +3565,8 @@ __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzip_s32_x4)))
 svint32x4_t svzip_s32_x4(svint32x4_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzip_s64_x4)))
 svint64x4_t svzip_s64_x4(svint64x4_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzip_mf8_x4)))
+svmfloat8x4_t svzip_mf8_x4(svmfloat8x4_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzip_s16_x4)))
 svint16x4_t svzip_s16_x4(svint16x4_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzipq_u8_x2)))
@@ -3575,6 +3591,8 @@ __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzipq_s32_x2)))
 svint32x2_t svzipq_s32_x2(svint32x2_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzipq_s64_x2)))
 svint64x2_t svzipq_s64_x2(svint64x2_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzipq_mf8_x2)))
+svmfloat8x2_t svzipq_mf8_x2(svmfloat8x2_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzipq_s16_x2)))
 svint16x2_t svzipq_s16_x2(svint16x2_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzipq_u8_x4)))
@@ -3599,6 +3617,8 @@ __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzipq_s32_x4)))
 svint32x4_t svzipq_s32_x4(svint32x4_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzipq_s64_x4)))
 svint64x4_t svzipq_s64_x4(svint64x4_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzipq_mf8_x4)))
+svmfloat8x4_t svzipq_mf8_x4(svmfloat8x4_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzipq_s16_x4)))
 svint16x4_t svzipq_s16_x4(svint16x4_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svadd_single_u8_x2)))
@@ -4119,6 +4139,8 @@ __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsel_s32_x2)))
 svint32x2_t svsel(svcount_t, svint32x2_t, svint32x2_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsel_s64_x2)))
 svint64x2_t svsel(svcount_t, svint64x2_t, svint64x2_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsel_mf8_x2)))
+svmfloat8x2_t svsel(svcount_t, svmfloat8x2_t, svmfloat8x2_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsel_s16_x2)))
 svint16x2_t svsel(svcount_t, svint16x2_t, svint16x2_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsel_u8_x4)))
@@ -4143,6 +4165,8 @@ __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsel_s32_x4)))
 svint32x4_t svsel(svcount_t, svint32x4_t, svint32x4_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsel_s64_x4)))
 svint64x4_t svsel(svcount_t, svint64x4_t, svint64x4_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsel_mf8_x4)))
+svmfloat8x4_t svsel(svcount_t, svmfloat8x4_t, svmfloat8x4_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsel_s16_x4)))
 svint16x4_t svsel(svcount_t, svint16x4_t, svint16x4_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svunpk_s32_s16_x2)))
@@ -4191,6 +4215,8 @@ __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzp_s32_x2)))
 svint32x2_t svuzp(svint32x2_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzp_s64_x2)))
 svint64x2_t svuzp(svint64x2_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzp_mf8_x2)))
+svmfloat8x2_t svuzp(svmfloat8x2_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzp_s16_x2)))
 svint16x2_t svuzp(svint16x2_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzp_u8_x4)))
@@ -4215,6 +4241,8 @@ __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzp_s32_x4)))
 svint32x4_t svuzp(svint32x4_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzp_s64_x4)))
 svint64x4_t svuzp(svint64x4_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzp_mf8_x4)))
+svmfloat8x4_t svuzp(svmfloat8x4_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzp_s16_x4)))
 svint16x4_t svuzp(svint16x4_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzpq_u8_x2)))
@@ -4239,6 +4267,8 @@ __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzpq_s32_x2)))
 svint32x2_t svuzpq(svint32x2_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzpq_s64_x2)))
 svint64x2_t svuzpq(svint64x2_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzpq_mf8_x2)))
+svmfloat8x2_t svuzpq(svmfloat8x2_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzpq_s16_x2)))
 svint16x2_t svuzpq(svint16x2_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzpq_u8_x4)))
@@ -4263,6 +4293,8 @@ __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzpq_s32_x4)))
 svint32x4_t svuzpq(svint32x4_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzpq_s64_x4)))
 svint64x4_t svuzpq(svint64x4_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzpq_mf8_x4)))
+svmfloat8x4_t svuzpq(svmfloat8x4_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzpq_s16_x4)))
 svint16x4_t svuzpq(svint16x4_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzip_u8_x2)))
@@ -4287,6 +4319,8 @@ __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzip_s32_x2)))
 svint32x2_t svzip(svint32x2_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzip_s64_x2)))
 svint64x2_t svzip(svint64x2_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzip_mf8_x2)))
+svmfloat8x2_t svzip(svmfloat8x2_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzip_s16_x2)))
 svint16x2_t svzip(svint16x2_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzip_u8_x4)))
@@ -4311,6 +4345,8 @@ __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzip_s32_x4)))
 svint32x4_t svzip(svint32x4_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzip_s64_x4)))
 svint64x4_t svzip(svint64x4_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzip_mf8_x4)))
+svmfloat8x4_t svzip(svmfloat8x4_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzip_s16_x4)))
 svint16x4_t svzip(svint16x4_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzipq_u8_x2)))
@@ -4335,6 +4371,8 @@ __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzipq_s32_x2)))
 svint32x2_t svzipq(svint32x2_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzipq_s64_x2)))
 svint64x2_t svzipq(svint64x2_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzipq_mf8_x2)))
+svmfloat8x2_t svzipq(svmfloat8x2_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzipq_s16_x2)))
 svint16x2_t svzipq(svint16x2_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzipq_u8_x4)))
@@ -4359,6 +4397,8 @@ __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzipq_s32_x4)))
 svint32x4_t svzipq(svint32x4_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzipq_s64_x4)))
 svint64x4_t svzipq(svint64x4_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzipq_mf8_x4)))
+svmfloat8x4_t svzipq(svmfloat8x4_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzipq_s16_x4)))
 svint16x4_t svzipq(svint16x4_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svamax_f64_x2)))
@@ -4627,12 +4667,6 @@ __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcompact_s32)))
 svint32_t svcompact_s32(svbool_t, svint32_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcompact_s64)))
 svint64_t svcompact_s64(svbool_t, svint64_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svexpa_f64)))
-svfloat64_t svexpa_f64(svuint64_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svexpa_f32)))
-svfloat32_t svexpa_f32(svuint32_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svexpa_f16)))
-svfloat16_t svexpa_f16(svuint16_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1_gather_u32base_index_u32)))
 svuint32_t svld1_gather_u32base_index_u32(svbool_t, svuint32_t, int64_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1_gather_u64base_index_u64)))
@@ -4957,6 +4991,8 @@ __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldff1_u64)))
 svuint64_t svldff1_u64(svbool_t, uint64_t const *);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldff1_u16)))
 svuint16_t svldff1_u16(svbool_t, uint16_t const *);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldff1_bf16)))
+svbfloat16_t svldff1_bf16(svbool_t, bfloat16_t const *);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldff1_s8)))
 svint8_t svldff1_s8(svbool_t, int8_t const *);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldff1_f64)))
@@ -4969,6 +5005,8 @@ __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldff1_s32)))
 svint32_t svldff1_s32(svbool_t, int32_t const *);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldff1_s64)))
 svint64_t svldff1_s64(svbool_t, int64_t const *);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldff1_mf8)))
+svmfloat8_t svldff1_mf8(svbool_t, mfloat8_t const *);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldff1_s16)))
 svint16_t svldff1_s16(svbool_t, int16_t const *);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldff1_gather_u32base_index_u32)))
@@ -5063,6 +5101,8 @@ __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldff1_vnum_u64)))
 svuint64_t svldff1_vnum_u64(svbool_t, uint64_t const *, int64_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldff1_vnum_u16)))
 svuint16_t svldff1_vnum_u16(svbool_t, uint16_t const *, int64_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldff1_vnum_bf16)))
+svbfloat16_t svldff1_vnum_bf16(svbool_t, bfloat16_t const *, int64_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldff1_vnum_s8)))
 svint8_t svldff1_vnum_s8(svbool_t, int8_t const *, int64_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldff1_vnum_f64)))
@@ -5075,6 +5115,8 @@ __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldff1_vnum_s32)))
 svint32_t svldff1_vnum_s32(svbool_t, int32_t const *, int64_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldff1_vnum_s64)))
 svint64_t svldff1_vnum_s64(svbool_t, int64_t const *, int64_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldff1_vnum_mf8)))
+svmfloat8_t svldff1_vnum_mf8(svbool_t, mfloat8_t const *, int64_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldff1_vnum_s16)))
 svint16_t svldff1_vnum_s16(svbool_t, int16_t const *, int64_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldff1sb_gather_u32base_offset_u32)))
@@ -5413,6 +5455,8 @@ __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnf1_u64)))
 svuint64_t svldnf1_u64(svbool_t, uint64_t const *);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnf1_u16)))
 svuint16_t svldnf1_u16(svbool_t, uint16_t const *);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnf1_bf16)))
+svbfloat16_t svldnf1_bf16(svbool_t, bfloat16_t const *);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnf1_s8)))
 svint8_t svldnf1_s8(svbool_t, int8_t const *);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnf1_f64)))
@@ -5425,6 +5469,8 @@ __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnf1_s32)))
 svint32_t svldnf1_s32(svbool_t, int32_t const *);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnf1_s64)))
 svint64_t svldnf1_s64(svbool_t, int64_t const *);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnf1_mf8)))
+svmfloat8_t svldnf1_mf8(svbool_t, mfloat8_t const *);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnf1_s16)))
 svint16_t svldnf1_s16(svbool_t, int16_t const *);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnf1_vnum_u8)))
@@ -5435,6 +5481,8 @@ __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnf1_vnum_u64)))
 svuint64_t svldnf1_vnum_u64(svbool_t, uint64_t const *, int64_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnf1_vnum_u16)))
 svuint16_t svldnf1_vnum_u16(svbool_t, uint16_t const *, int64_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnf1_vnum_bf16)))
+svbfloat16_t svldnf1_vnum_bf16(svbool_t, bfloat16_t const *, int64_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnf1_vnum_s8)))
 svint8_t svldnf1_vnum_s8(svbool_t, int8_t const *, int64_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnf1_vnum_f64)))
@@ -5447,6 +5495,8 @@ __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnf1_vnum_s32)))
 svint32_t svldnf1_vnum_s32(svbool_t, int32_t const *, int64_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnf1_vnum_s64)))
 svint64_t svldnf1_vnum_s64(svbool_t, int64_t const *, int64_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnf1_vnum_mf8)))
+svmfloat8_t svldnf1_vnum_mf8(svbool_t, mfloat8_t const *, int64_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnf1_vnum_s16)))
 svint16_t svldnf1_vnum_s16(svbool_t, int16_t const *, int64_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnf1sb_vnum_u32)))
@@ -5885,12 +5935,6 @@ __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcompact_s32)))
 svint32_t svcompact(svbool_t, svint32_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcompact_s64)))
 svint64_t svcompact(svbool_t, svint64_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svexpa_f64)))
-svfloat64_t svexpa(svuint64_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svexpa_f32)))
-svfloat32_t svexpa(svuint32_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svexpa_f16)))
-svfloat16_t svexpa(svuint16_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1_gather_u32base_index_u32)))
 svuint32_t svld1_gather_index_u32(svbool_t, svuint32_t, int64_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1_gather_u64base_index_u64)))
@@ -6215,6 +6259,8 @@ __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldff1_u64)))
 svuint64_t svldff1(svbool_t, uint64_t const *);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldff1_u16)))
 svuint16_t svldff1(svbool_t, uint16_t const *);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldff1_bf16)))
+svbfloat16_t svldff1(svbool_t, bfloat16_t const *);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldff1_s8)))
 svint8_t svldff1(svbool_t, int8_t const *);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldff1_f64)))
@@ -6227,6 +6273,8 @@ __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldff1_s32)))
 svint32_t svldff1(svbool_t, int32_t const *);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldff1_s64)))
 svint64_t svldff1(svbool_t, int64_t const *);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldff1_mf8)))
+svmfloat8_t svldff1(svbool_t, mfloat8_t const *);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldff1_s16)))
 svint16_t svldff1(svbool_t, int16_t const *);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldff1_gather_u32base_index_u32)))
@@ -6321,6 +6369,8 @@ __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldff1_vnum_u64)))
 svuint64_t svldff1_vnum(svbool_t, uint64_t const *, int64_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldff1_vnum_u16)))
 svuint16_t svldff1_vnum(svbool_t, uint16_t const *, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldff1_vnum_bf16)))
+svbfloat16_t svldff1_vnum(svbool_t, bfloat16_t const *, int64_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldff1_vnum_s8)))
 svint8_t svldff1_vnum(svbool_t, int8_t const *, int64_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldff1_vnum_f64)))
@@ -6333,6 +6383,8 @@ __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldff1_vnum_s32)))
 svint32_t svldff1_vnum(svbool_t, int32_t const *, int64_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldff1_vnum_s64)))
 svint64_t svldff1_vnum(svbool_t, int64_t const *, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldff1_vnum_mf8)))
+svmfloat8_t svldff1_vnum(svbool_t, mfloat8_t const *, int64_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldff1_vnum_s16)))
 svint16_t svldff1_vnum(svbool_t, int16_t const *, int64_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldff1sb_gather_u32base_offset_u32)))
@@ -6575,6 +6627,8 @@ __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnf1_u64)))
 svuint64_t svldnf1(svbool_t, uint64_t const *);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnf1_u16)))
 svuint16_t svldnf1(svbool_t, uint16_t const *);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnf1_bf16)))
+svbfloat16_t svldnf1(svbool_t, bfloat16_t const *);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnf1_s8)))
 svint8_t svldnf1(svbool_t, int8_t const *);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnf1_f64)))
@@ -6587,6 +6641,8 @@ __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnf1_s32)))
 svint32_t svldnf1(svbool_t, int32_t const *);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnf1_s64)))
 svint64_t svldnf1(svbool_t, int64_t const *);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnf1_mf8)))
+svmfloat8_t svldnf1(svbool_t, mfloat8_t const *);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnf1_s16)))
 svint16_t svldnf1(svbool_t, int16_t const *);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnf1_vnum_u8)))
@@ -6597,6 +6653,8 @@ __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnf1_vnum_u64)))
 svuint64_t svldnf1_vnum(svbool_t, uint64_t const *, int64_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnf1_vnum_u16)))
 svuint16_t svldnf1_vnum(svbool_t, uint16_t const *, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnf1_vnum_bf16)))
+svbfloat16_t svldnf1_vnum(svbool_t, bfloat16_t const *, int64_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnf1_vnum_s8)))
 svint8_t svldnf1_vnum(svbool_t, int8_t const *, int64_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnf1_vnum_f64)))
@@ -6609,6 +6667,8 @@ __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnf1_vnum_s32)))
 svint32_t svldnf1_vnum(svbool_t, int32_t const *, int64_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnf1_vnum_s64)))
 svint64_t svldnf1_vnum(svbool_t, int64_t const *, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnf1_vnum_mf8)))
+svmfloat8_t svldnf1_vnum(svbool_t, mfloat8_t const *, int64_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnf1_vnum_s16)))
 svint16_t svldnf1_vnum(svbool_t, int16_t const *, int64_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svprfb_gather_u32base)))
@@ -6893,50 +6953,1834 @@ __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtssel_f32)))
 svfloat32_t svtssel(svfloat32_t, svuint32_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtssel_f16)))
 svfloat16_t svtssel(svfloat16_t, svuint16_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svclamp_s8)))
+svint8_t svclamp_s8(svint8_t, svint8_t, svint8_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svclamp_s32)))
+svint32_t svclamp_s32(svint32_t, svint32_t, svint32_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svclamp_s64)))
+svint64_t svclamp_s64(svint64_t, svint64_t, svint64_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svclamp_s16)))
+svint16_t svclamp_s16(svint16_t, svint16_t, svint16_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svclamp_u8)))
+svuint8_t svclamp_u8(svuint8_t, svuint8_t, svuint8_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svclamp_u32)))
+svuint32_t svclamp_u32(svuint32_t, svuint32_t, svuint32_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svclamp_u64)))
+svuint64_t svclamp_u64(svuint64_t, svuint64_t, svuint64_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svclamp_u16)))
+svuint16_t svclamp_u16(svuint16_t, svuint16_t, svuint16_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svpsel_lane_b16)))
+svbool_t svpsel_lane_b16(svbool_t, svbool_t, uint32_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svpsel_lane_b32)))
+svbool_t svpsel_lane_b32(svbool_t, svbool_t, uint32_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svpsel_lane_b64)))
+svbool_t svpsel_lane_b64(svbool_t, svbool_t, uint32_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svpsel_lane_b8)))
+svbool_t svpsel_lane_b8(svbool_t, svbool_t, uint32_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrevd_u8_m)))
+svuint8_t svrevd_u8_m(svuint8_t, svbool_t, svuint8_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrevd_u32_m)))
+svuint32_t svrevd_u32_m(svuint32_t, svbool_t, svuint32_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrevd_u64_m)))
+svuint64_t svrevd_u64_m(svuint64_t, svbool_t, svuint64_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrevd_u16_m)))
+svuint16_t svrevd_u16_m(svuint16_t, svbool_t, svuint16_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrevd_bf16_m)))
+svbfloat16_t svrevd_bf16_m(svbfloat16_t, svbool_t, svbfloat16_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrevd_s8_m)))
+svint8_t svrevd_s8_m(svint8_t, svbool_t, svint8_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrevd_f64_m)))
+svfloat64_t svrevd_f64_m(svfloat64_t, svbool_t, svfloat64_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrevd_f32_m)))
+svfloat32_t svrevd_f32_m(svfloat32_t, svbool_t, svfloat32_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrevd_f16_m)))
+svfloat16_t svrevd_f16_m(svfloat16_t, svbool_t, svfloat16_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrevd_s32_m)))
+svint32_t svrevd_s32_m(svint32_t, svbool_t, svint32_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrevd_s64_m)))
+svint64_t svrevd_s64_m(svint64_t, svbool_t, svint64_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrevd_mf8_m)))
+svmfloat8_t svrevd_mf8_m(svmfloat8_t, svbool_t, svmfloat8_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrevd_s16_m)))
+svint16_t svrevd_s16_m(svint16_t, svbool_t, svint16_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrevd_u8_x)))
+svuint8_t svrevd_u8_x(svbool_t, svuint8_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrevd_u32_x)))
+svuint32_t svrevd_u32_x(svbool_t, svuint32_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrevd_u64_x)))
+svuint64_t svrevd_u64_x(svbool_t, svuint64_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrevd_u16_x)))
+svuint16_t svrevd_u16_x(svbool_t, svuint16_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrevd_bf16_x)))
+svbfloat16_t svrevd_bf16_x(svbool_t, svbfloat16_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrevd_s8_x)))
+svint8_t svrevd_s8_x(svbool_t, svint8_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrevd_f64_x)))
+svfloat64_t svrevd_f64_x(svbool_t, svfloat64_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrevd_f32_x)))
+svfloat32_t svrevd_f32_x(svbool_t, svfloat32_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrevd_f16_x)))
+svfloat16_t svrevd_f16_x(svbool_t, svfloat16_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrevd_s32_x)))
+svint32_t svrevd_s32_x(svbool_t, svint32_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrevd_s64_x)))
+svint64_t svrevd_s64_x(svbool_t, svint64_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrevd_mf8_x)))
+svmfloat8_t svrevd_mf8_x(svbool_t, svmfloat8_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrevd_s16_x)))
+svint16_t svrevd_s16_x(svbool_t, svint16_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrevd_u8_z)))
+svuint8_t svrevd_u8_z(svbool_t, svuint8_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrevd_u32_z)))
+svuint32_t svrevd_u32_z(svbool_t, svuint32_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrevd_u64_z)))
+svuint64_t svrevd_u64_z(svbool_t, svuint64_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrevd_u16_z)))
+svuint16_t svrevd_u16_z(svbool_t, svuint16_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrevd_bf16_z)))
+svbfloat16_t svrevd_bf16_z(svbool_t, svbfloat16_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrevd_s8_z)))
+svint8_t svrevd_s8_z(svbool_t, svint8_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrevd_f64_z)))
+svfloat64_t svrevd_f64_z(svbool_t, svfloat64_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrevd_f32_z)))
+svfloat32_t svrevd_f32_z(svbool_t, svfloat32_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrevd_f16_z)))
+svfloat16_t svrevd_f16_z(svbool_t, svfloat16_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrevd_s32_z)))
+svint32_t svrevd_s32_z(svbool_t, svint32_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrevd_s64_z)))
+svint64_t svrevd_s64_z(svbool_t, svint64_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrevd_mf8_z)))
+svmfloat8_t svrevd_mf8_z(svbool_t, svmfloat8_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrevd_s16_z)))
+svint16_t svrevd_s16_z(svbool_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svclamp_s8)))
+svint8_t svclamp(svint8_t, svint8_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svclamp_s32)))
+svint32_t svclamp(svint32_t, svint32_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svclamp_s64)))
+svint64_t svclamp(svint64_t, svint64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svclamp_s16)))
+svint16_t svclamp(svint16_t, svint16_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svclamp_u8)))
+svuint8_t svclamp(svuint8_t, svuint8_t, svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svclamp_u32)))
+svuint32_t svclamp(svuint32_t, svuint32_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svclamp_u64)))
+svuint64_t svclamp(svuint64_t, svuint64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svclamp_u16)))
+svuint16_t svclamp(svuint16_t, svuint16_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrevd_u8_m)))
+svuint8_t svrevd_m(svuint8_t, svbool_t, svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrevd_u32_m)))
+svuint32_t svrevd_m(svuint32_t, svbool_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrevd_u64_m)))
+svuint64_t svrevd_m(svuint64_t, svbool_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrevd_u16_m)))
+svuint16_t svrevd_m(svuint16_t, svbool_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrevd_bf16_m)))
+svbfloat16_t svrevd_m(svbfloat16_t, svbool_t, svbfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrevd_s8_m)))
+svint8_t svrevd_m(svint8_t, svbool_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrevd_f64_m)))
+svfloat64_t svrevd_m(svfloat64_t, svbool_t, svfloat64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrevd_f32_m)))
+svfloat32_t svrevd_m(svfloat32_t, svbool_t, svfloat32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrevd_f16_m)))
+svfloat16_t svrevd_m(svfloat16_t, svbool_t, svfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrevd_s32_m)))
+svint32_t svrevd_m(svint32_t, svbool_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrevd_s64_m)))
+svint64_t svrevd_m(svint64_t, svbool_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrevd_mf8_m)))
+svmfloat8_t svrevd_m(svmfloat8_t, svbool_t, svmfloat8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrevd_s16_m)))
+svint16_t svrevd_m(svint16_t, svbool_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrevd_u8_x)))
+svuint8_t svrevd_x(svbool_t, svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrevd_u32_x)))
+svuint32_t svrevd_x(svbool_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrevd_u64_x)))
+svuint64_t svrevd_x(svbool_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrevd_u16_x)))
+svuint16_t svrevd_x(svbool_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrevd_bf16_x)))
+svbfloat16_t svrevd_x(svbool_t, svbfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrevd_s8_x)))
+svint8_t svrevd_x(svbool_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrevd_f64_x)))
+svfloat64_t svrevd_x(svbool_t, svfloat64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrevd_f32_x)))
+svfloat32_t svrevd_x(svbool_t, svfloat32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrevd_f16_x)))
+svfloat16_t svrevd_x(svbool_t, svfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrevd_s32_x)))
+svint32_t svrevd_x(svbool_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrevd_s64_x)))
+svint64_t svrevd_x(svbool_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrevd_mf8_x)))
+svmfloat8_t svrevd_x(svbool_t, svmfloat8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrevd_s16_x)))
+svint16_t svrevd_x(svbool_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrevd_u8_z)))
+svuint8_t svrevd_z(svbool_t, svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrevd_u32_z)))
+svuint32_t svrevd_z(svbool_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrevd_u64_z)))
+svuint64_t svrevd_z(svbool_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrevd_u16_z)))
+svuint16_t svrevd_z(svbool_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrevd_bf16_z)))
+svbfloat16_t svrevd_z(svbool_t, svbfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrevd_s8_z)))
+svint8_t svrevd_z(svbool_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrevd_f64_z)))
+svfloat64_t svrevd_z(svbool_t, svfloat64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrevd_f32_z)))
+svfloat32_t svrevd_z(svbool_t, svfloat32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrevd_f16_z)))
+svfloat16_t svrevd_z(svbool_t, svfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrevd_s32_z)))
+svint32_t svrevd_z(svbool_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrevd_s64_z)))
+svint64_t svrevd_z(svbool_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrevd_mf8_z)))
+svmfloat8_t svrevd_z(svbool_t, svmfloat8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrevd_s16_z)))
+svint16_t svrevd_z(svbool_t, svint16_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbfmlslb_f32)))
+svfloat32_t svbfmlslb_f32(svfloat32_t, svbfloat16_t, svbfloat16_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbfmlslb_lane_f32)))
+svfloat32_t svbfmlslb_lane_f32(svfloat32_t, svbfloat16_t, svbfloat16_t, uint64_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbfmlslt_f32)))
+svfloat32_t svbfmlslt_f32(svfloat32_t, svbfloat16_t, svbfloat16_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbfmlslt_lane_f32)))
+svfloat32_t svbfmlslt_lane_f32(svfloat32_t, svbfloat16_t, svbfloat16_t, uint64_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svclamp_f64)))
+svfloat64_t svclamp_f64(svfloat64_t, svfloat64_t, svfloat64_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svclamp_f32)))
+svfloat32_t svclamp_f32(svfloat32_t, svfloat32_t, svfloat32_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svclamp_f16)))
+svfloat16_t svclamp_f16(svfloat16_t, svfloat16_t, svfloat16_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcreate2_b)))
+svboolx2_t svcreate2_b(svbool_t, svbool_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcreate4_b)))
+svboolx4_t svcreate4_b(svbool_t, svbool_t, svbool_t, svbool_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdot_f32_f16)))
+svfloat32_t svdot_f32_f16(svfloat32_t, svfloat16_t, svfloat16_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdot_s32_s16)))
+svint32_t svdot_s32_s16(svint32_t, svint16_t, svint16_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdot_u32_u16)))
+svuint32_t svdot_u32_u16(svuint32_t, svuint16_t, svuint16_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdot_lane_f32_f16)))
+svfloat32_t svdot_lane_f32_f16(svfloat32_t, svfloat16_t, svfloat16_t, uint64_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdot_lane_s32_s16)))
+svint32_t svdot_lane_s32_s16(svint32_t, svint16_t, svint16_t, uint64_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdot_lane_u32_u16)))
+svuint32_t svdot_lane_u32_u16(svuint32_t, svuint16_t, svuint16_t, uint64_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svget2_b)))
+svbool_t svget2_b(svboolx2_t, uint64_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svget4_b)))
+svbool_t svget4_b(svboolx4_t, uint64_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqcvtn_s16_s32_x2)))
+svint16_t svqcvtn_s16_s32_x2(svint32x2_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqcvtn_u16_s32_x2)))
+svuint16_t svqcvtn_u16_s32_x2(svint32x2_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqcvtn_u16_u32_x2)))
+svuint16_t svqcvtn_u16_u32_x2(svuint32x2_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshrn_n_s16_s32_x2)))
+svint16_t svqrshrn_n_s16_s32_x2(svint32x2_t, uint64_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshrn_n_u16_u32_x2)))
+svuint16_t svqrshrn_n_u16_u32_x2(svuint32x2_t, uint64_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshrun_n_u16_s32_x2)))
+svuint16_t svqrshrun_n_u16_s32_x2(svint32x2_t, uint64_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svset2_b)))
+svboolx2_t svset2_b(svboolx2_t, uint64_t, svbool_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svset4_b)))
+svboolx4_t svset4_b(svboolx4_t, uint64_t, svbool_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svundef2_b)))
+svboolx2_t svundef2_b();
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svundef4_b)))
+svboolx4_t svundef4_b();
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilege_b8_s64_x2)))
+svboolx2_t svwhilege_b8_s64_x2(int64_t, int64_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilege_b32_s64_x2)))
+svboolx2_t svwhilege_b32_s64_x2(int64_t, int64_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilege_b64_s64_x2)))
+svboolx2_t svwhilege_b64_s64_x2(int64_t, int64_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilege_b16_s64_x2)))
+svboolx2_t svwhilege_b16_s64_x2(int64_t, int64_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilege_b8_u64_x2)))
+svboolx2_t svwhilege_b8_u64_x2(uint64_t, uint64_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilege_b32_u64_x2)))
+svboolx2_t svwhilege_b32_u64_x2(uint64_t, uint64_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilege_b64_u64_x2)))
+svboolx2_t svwhilege_b64_u64_x2(uint64_t, uint64_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilege_b16_u64_x2)))
+svboolx2_t svwhilege_b16_u64_x2(uint64_t, uint64_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilegt_b8_s64_x2)))
+svboolx2_t svwhilegt_b8_s64_x2(int64_t, int64_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilegt_b32_s64_x2)))
+svboolx2_t svwhilegt_b32_s64_x2(int64_t, int64_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilegt_b64_s64_x2)))
+svboolx2_t svwhilegt_b64_s64_x2(int64_t, int64_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilegt_b16_s64_x2)))
+svboolx2_t svwhilegt_b16_s64_x2(int64_t, int64_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilegt_b8_u64_x2)))
+svboolx2_t svwhilegt_b8_u64_x2(uint64_t, uint64_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilegt_b32_u64_x2)))
+svboolx2_t svwhilegt_b32_u64_x2(uint64_t, uint64_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilegt_b64_u64_x2)))
+svboolx2_t svwhilegt_b64_u64_x2(uint64_t, uint64_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilegt_b16_u64_x2)))
+svboolx2_t svwhilegt_b16_u64_x2(uint64_t, uint64_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilele_b8_s64_x2)))
+svboolx2_t svwhilele_b8_s64_x2(int64_t, int64_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilele_b32_s64_x2)))
+svboolx2_t svwhilele_b32_s64_x2(int64_t, int64_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilele_b64_s64_x2)))
+svboolx2_t svwhilele_b64_s64_x2(int64_t, int64_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilele_b16_s64_x2)))
+svboolx2_t svwhilele_b16_s64_x2(int64_t, int64_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilele_b8_u64_x2)))
+svboolx2_t svwhilele_b8_u64_x2(uint64_t, uint64_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilele_b32_u64_x2)))
+svboolx2_t svwhilele_b32_u64_x2(uint64_t, uint64_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilele_b64_u64_x2)))
+svboolx2_t svwhilele_b64_u64_x2(uint64_t, uint64_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilele_b16_u64_x2)))
+svboolx2_t svwhilele_b16_u64_x2(uint64_t, uint64_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilelt_b8_u64_x2)))
+svboolx2_t svwhilelt_b8_u64_x2(uint64_t, uint64_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilelt_b32_u64_x2)))
+svboolx2_t svwhilelt_b32_u64_x2(uint64_t, uint64_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilelt_b64_u64_x2)))
+svboolx2_t svwhilelt_b64_u64_x2(uint64_t, uint64_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilelt_b16_u64_x2)))
+svboolx2_t svwhilelt_b16_u64_x2(uint64_t, uint64_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilelt_b8_s64_x2)))
+svboolx2_t svwhilelt_b8_s64_x2(int64_t, int64_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilelt_b32_s64_x2)))
+svboolx2_t svwhilelt_b32_s64_x2(int64_t, int64_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilelt_b64_s64_x2)))
+svboolx2_t svwhilelt_b64_s64_x2(int64_t, int64_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilelt_b16_s64_x2)))
+svboolx2_t svwhilelt_b16_s64_x2(int64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbfmlslb_f32)))
+svfloat32_t svbfmlslb(svfloat32_t, svbfloat16_t, svbfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbfmlslb_lane_f32)))
+svfloat32_t svbfmlslb_lane(svfloat32_t, svbfloat16_t, svbfloat16_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbfmlslt_f32)))
+svfloat32_t svbfmlslt(svfloat32_t, svbfloat16_t, svbfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbfmlslt_lane_f32)))
+svfloat32_t svbfmlslt_lane(svfloat32_t, svbfloat16_t, svbfloat16_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svclamp_f64)))
+svfloat64_t svclamp(svfloat64_t, svfloat64_t, svfloat64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svclamp_f32)))
+svfloat32_t svclamp(svfloat32_t, svfloat32_t, svfloat32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svclamp_f16)))
+svfloat16_t svclamp(svfloat16_t, svfloat16_t, svfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcreate2_b)))
+svboolx2_t svcreate2(svbool_t, svbool_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcreate4_b)))
+svboolx4_t svcreate4(svbool_t, svbool_t, svbool_t, svbool_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdot_f32_f16)))
+svfloat32_t svdot(svfloat32_t, svfloat16_t, svfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdot_s32_s16)))
+svint32_t svdot(svint32_t, svint16_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdot_u32_u16)))
+svuint32_t svdot(svuint32_t, svuint16_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdot_lane_f32_f16)))
+svfloat32_t svdot_lane(svfloat32_t, svfloat16_t, svfloat16_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdot_lane_s32_s16)))
+svint32_t svdot_lane(svint32_t, svint16_t, svint16_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdot_lane_u32_u16)))
+svuint32_t svdot_lane(svuint32_t, svuint16_t, svuint16_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svget2_b)))
+svbool_t svget2(svboolx2_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svget4_b)))
+svbool_t svget4(svboolx4_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqcvtn_s16_s32_x2)))
+svint16_t svqcvtn_s16(svint32x2_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqcvtn_u16_s32_x2)))
+svuint16_t svqcvtn_u16(svint32x2_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqcvtn_u16_u32_x2)))
+svuint16_t svqcvtn_u16(svuint32x2_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshrn_n_s16_s32_x2)))
+svint16_t svqrshrn_s16(svint32x2_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshrn_n_u16_u32_x2)))
+svuint16_t svqrshrn_u16(svuint32x2_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshrun_n_u16_s32_x2)))
+svuint16_t svqrshrun_u16(svint32x2_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svset2_b)))
+svboolx2_t svset2(svboolx2_t, uint64_t, svbool_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svset4_b)))
+svboolx4_t svset4(svboolx4_t, uint64_t, svbool_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilege_b8_s64_x2)))
+svboolx2_t svwhilege_b8_x2(int64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilege_b32_s64_x2)))
+svboolx2_t svwhilege_b32_x2(int64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilege_b64_s64_x2)))
+svboolx2_t svwhilege_b64_x2(int64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilege_b16_s64_x2)))
+svboolx2_t svwhilege_b16_x2(int64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilege_b8_u64_x2)))
+svboolx2_t svwhilege_b8_x2(uint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilege_b32_u64_x2)))
+svboolx2_t svwhilege_b32_x2(uint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilege_b64_u64_x2)))
+svboolx2_t svwhilege_b64_x2(uint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilege_b16_u64_x2)))
+svboolx2_t svwhilege_b16_x2(uint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilegt_b8_s64_x2)))
+svboolx2_t svwhilegt_b8_x2(int64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilegt_b32_s64_x2)))
+svboolx2_t svwhilegt_b32_x2(int64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilegt_b64_s64_x2)))
+svboolx2_t svwhilegt_b64_x2(int64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilegt_b16_s64_x2)))
+svboolx2_t svwhilegt_b16_x2(int64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilegt_b8_u64_x2)))
+svboolx2_t svwhilegt_b8_x2(uint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilegt_b32_u64_x2)))
+svboolx2_t svwhilegt_b32_x2(uint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilegt_b64_u64_x2)))
+svboolx2_t svwhilegt_b64_x2(uint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilegt_b16_u64_x2)))
+svboolx2_t svwhilegt_b16_x2(uint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilele_b8_s64_x2)))
+svboolx2_t svwhilele_b8_x2(int64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilele_b32_s64_x2)))
+svboolx2_t svwhilele_b32_x2(int64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilele_b64_s64_x2)))
+svboolx2_t svwhilele_b64_x2(int64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilele_b16_s64_x2)))
+svboolx2_t svwhilele_b16_x2(int64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilele_b8_u64_x2)))
+svboolx2_t svwhilele_b8_x2(uint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilele_b32_u64_x2)))
+svboolx2_t svwhilele_b32_x2(uint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilele_b64_u64_x2)))
+svboolx2_t svwhilele_b64_x2(uint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilele_b16_u64_x2)))
+svboolx2_t svwhilele_b16_x2(uint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilelt_b8_u64_x2)))
+svboolx2_t svwhilelt_b8_x2(uint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilelt_b32_u64_x2)))
+svboolx2_t svwhilelt_b32_x2(uint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilelt_b64_u64_x2)))
+svboolx2_t svwhilelt_b64_x2(uint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilelt_b16_u64_x2)))
+svboolx2_t svwhilelt_b16_x2(uint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilelt_b8_s64_x2)))
+svboolx2_t svwhilelt_b8_x2(int64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilelt_b32_s64_x2)))
+svboolx2_t svwhilelt_b32_x2(int64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilelt_b64_s64_x2)))
+svboolx2_t svwhilelt_b64_x2(int64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilelt_b16_s64_x2)))
+svboolx2_t svwhilelt_b16_x2(int64_t, int64_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddqv_u8)))
+uint8x16_t svaddqv_u8(svbool_t, svuint8_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddqv_u32)))
+uint32x4_t svaddqv_u32(svbool_t, svuint32_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddqv_u64)))
+uint64x2_t svaddqv_u64(svbool_t, svuint64_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddqv_u16)))
+uint16x8_t svaddqv_u16(svbool_t, svuint16_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddqv_s8)))
+int8x16_t svaddqv_s8(svbool_t, svint8_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddqv_s32)))
+int32x4_t svaddqv_s32(svbool_t, svint32_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddqv_s64)))
+int64x2_t svaddqv_s64(svbool_t, svint64_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddqv_s16)))
+int16x8_t svaddqv_s16(svbool_t, svint16_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddqv_f64)))
+float64x2_t svaddqv_f64(svbool_t, svfloat64_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddqv_f32)))
+float32x4_t svaddqv_f32(svbool_t, svfloat32_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddqv_f16)))
+float16x8_t svaddqv_f16(svbool_t, svfloat16_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svandqv_u8)))
+uint8x16_t svandqv_u8(svbool_t, svuint8_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svandqv_u32)))
+uint32x4_t svandqv_u32(svbool_t, svuint32_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svandqv_u64)))
+uint64x2_t svandqv_u64(svbool_t, svuint64_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svandqv_u16)))
+uint16x8_t svandqv_u16(svbool_t, svuint16_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svandqv_s8)))
+int8x16_t svandqv_s8(svbool_t, svint8_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svandqv_s32)))
+int32x4_t svandqv_s32(svbool_t, svint32_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svandqv_s64)))
+int64x2_t svandqv_s64(svbool_t, svint64_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svandqv_s16)))
+int16x8_t svandqv_s16(svbool_t, svint16_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdup_laneq_u8)))
+svuint8_t svdup_laneq_u8(svuint8_t, uint64_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdup_laneq_s8)))
+svint8_t svdup_laneq_s8(svint8_t, uint64_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdup_laneq_mf8)))
+svmfloat8_t svdup_laneq_mf8(svmfloat8_t, uint64_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdup_laneq_u64)))
+svuint64_t svdup_laneq_u64(svuint64_t, uint64_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdup_laneq_f64)))
+svfloat64_t svdup_laneq_f64(svfloat64_t, uint64_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdup_laneq_s64)))
+svint64_t svdup_laneq_s64(svint64_t, uint64_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdup_laneq_u16)))
+svuint16_t svdup_laneq_u16(svuint16_t, uint64_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdup_laneq_bf16)))
+svbfloat16_t svdup_laneq_bf16(svbfloat16_t, uint64_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdup_laneq_f16)))
+svfloat16_t svdup_laneq_f16(svfloat16_t, uint64_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdup_laneq_s16)))
+svint16_t svdup_laneq_s16(svint16_t, uint64_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdup_laneq_u32)))
+svuint32_t svdup_laneq_u32(svuint32_t, uint64_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdup_laneq_f32)))
+svfloat32_t svdup_laneq_f32(svfloat32_t, uint64_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdup_laneq_s32)))
+svint32_t svdup_laneq_s32(svint32_t, uint64_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveorqv_u8)))
+uint8x16_t sveorqv_u8(svbool_t, svuint8_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveorqv_u32)))
+uint32x4_t sveorqv_u32(svbool_t, svuint32_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveorqv_u64)))
+uint64x2_t sveorqv_u64(svbool_t, svuint64_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveorqv_u16)))
+uint16x8_t sveorqv_u16(svbool_t, svuint16_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveorqv_s8)))
+int8x16_t sveorqv_s8(svbool_t, svint8_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveorqv_s32)))
+int32x4_t sveorqv_s32(svbool_t, svint32_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveorqv_s64)))
+int64x2_t sveorqv_s64(svbool_t, svint64_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveorqv_s16)))
+int16x8_t sveorqv_s16(svbool_t, svint16_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svextq_u8)))
+svuint8_t svextq_u8(svuint8_t, svuint8_t, int32_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svextq_u32)))
+svuint32_t svextq_u32(svuint32_t, svuint32_t, int32_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svextq_u64)))
+svuint64_t svextq_u64(svuint64_t, svuint64_t, int32_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svextq_u16)))
+svuint16_t svextq_u16(svuint16_t, svuint16_t, int32_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svextq_bf16)))
+svbfloat16_t svextq_bf16(svbfloat16_t, svbfloat16_t, int32_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svextq_s8)))
+svint8_t svextq_s8(svint8_t, svint8_t, int32_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svextq_f64)))
+svfloat64_t svextq_f64(svfloat64_t, svfloat64_t, int32_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svextq_f32)))
+svfloat32_t svextq_f32(svfloat32_t, svfloat32_t, int32_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svextq_f16)))
+svfloat16_t svextq_f16(svfloat16_t, svfloat16_t, int32_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svextq_s32)))
+svint32_t svextq_s32(svint32_t, svint32_t, int32_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svextq_s64)))
+svint64_t svextq_s64(svint64_t, svint64_t, int32_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svextq_mf8)))
+svmfloat8_t svextq_mf8(svmfloat8_t, svmfloat8_t, int32_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svextq_s16)))
+svint16_t svextq_s16(svint16_t, svint16_t, int32_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld2q_u8)))
+svuint8x2_t svld2q_u8(svbool_t, uint8_t const *);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld2q_u32)))
+svuint32x2_t svld2q_u32(svbool_t, uint32_t const *);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld2q_u64)))
+svuint64x2_t svld2q_u64(svbool_t, uint64_t const *);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld2q_u16)))
+svuint16x2_t svld2q_u16(svbool_t, uint16_t const *);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld2q_bf16)))
+svbfloat16x2_t svld2q_bf16(svbool_t, bfloat16_t const *);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld2q_s8)))
+svint8x2_t svld2q_s8(svbool_t, int8_t const *);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld2q_f64)))
+svfloat64x2_t svld2q_f64(svbool_t, float64_t const *);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld2q_f32)))
+svfloat32x2_t svld2q_f32(svbool_t, float32_t const *);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld2q_f16)))
+svfloat16x2_t svld2q_f16(svbool_t, float16_t const *);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld2q_s32)))
+svint32x2_t svld2q_s32(svbool_t, int32_t const *);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld2q_s64)))
+svint64x2_t svld2q_s64(svbool_t, int64_t const *);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld2q_mf8)))
+svmfloat8x2_t svld2q_mf8(svbool_t, mfloat8_t const *);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld2q_s16)))
+svint16x2_t svld2q_s16(svbool_t, int16_t const *);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld2q_vnum_u8)))
+svuint8x2_t svld2q_vnum_u8(svbool_t, uint8_t const *, int64_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld2q_vnum_u32)))
+svuint32x2_t svld2q_vnum_u32(svbool_t, uint32_t const *, int64_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld2q_vnum_u64)))
+svuint64x2_t svld2q_vnum_u64(svbool_t, uint64_t const *, int64_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld2q_vnum_u16)))
+svuint16x2_t svld2q_vnum_u16(svbool_t, uint16_t const *, int64_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld2q_vnum_bf16)))
+svbfloat16x2_t svld2q_vnum_bf16(svbool_t, bfloat16_t const *, int64_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld2q_vnum_s8)))
+svint8x2_t svld2q_vnum_s8(svbool_t, int8_t const *, int64_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld2q_vnum_f64)))
+svfloat64x2_t svld2q_vnum_f64(svbool_t, float64_t const *, int64_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld2q_vnum_f32)))
+svfloat32x2_t svld2q_vnum_f32(svbool_t, float32_t const *, int64_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld2q_vnum_f16)))
+svfloat16x2_t svld2q_vnum_f16(svbool_t, float16_t const *, int64_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld2q_vnum_s32)))
+svint32x2_t svld2q_vnum_s32(svbool_t, int32_t const *, int64_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld2q_vnum_s64)))
+svint64x2_t svld2q_vnum_s64(svbool_t, int64_t const *, int64_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld2q_vnum_mf8)))
+svmfloat8x2_t svld2q_vnum_mf8(svbool_t, mfloat8_t const *, int64_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld2q_vnum_s16)))
+svint16x2_t svld2q_vnum_s16(svbool_t, int16_t const *, int64_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld3q_u8)))
+svuint8x3_t svld3q_u8(svbool_t, uint8_t const *);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld3q_u32)))
+svuint32x3_t svld3q_u32(svbool_t, uint32_t const *);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld3q_u64)))
+svuint64x3_t svld3q_u64(svbool_t, uint64_t const *);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld3q_u16)))
+svuint16x3_t svld3q_u16(svbool_t, uint16_t const *);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld3q_bf16)))
+svbfloat16x3_t svld3q_bf16(svbool_t, bfloat16_t const *);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld3q_s8)))
+svint8x3_t svld3q_s8(svbool_t, int8_t const *);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld3q_f64)))
+svfloat64x3_t svld3q_f64(svbool_t, float64_t const *);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld3q_f32)))
+svfloat32x3_t svld3q_f32(svbool_t, float32_t const *);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld3q_f16)))
+svfloat16x3_t svld3q_f16(svbool_t, float16_t const *);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld3q_s32)))
+svint32x3_t svld3q_s32(svbool_t, int32_t const *);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld3q_s64)))
+svint64x3_t svld3q_s64(svbool_t, int64_t const *);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld3q_mf8)))
+svmfloat8x3_t svld3q_mf8(svbool_t, mfloat8_t const *);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld3q_s16)))
+svint16x3_t svld3q_s16(svbool_t, int16_t const *);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld3q_vnum_u8)))
+svuint8x3_t svld3q_vnum_u8(svbool_t, uint8_t const *, int64_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld3q_vnum_u32)))
+svuint32x3_t svld3q_vnum_u32(svbool_t, uint32_t const *, int64_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld3q_vnum_u64)))
+svuint64x3_t svld3q_vnum_u64(svbool_t, uint64_t const *, int64_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld3q_vnum_u16)))
+svuint16x3_t svld3q_vnum_u16(svbool_t, uint16_t const *, int64_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld3q_vnum_bf16)))
+svbfloat16x3_t svld3q_vnum_bf16(svbool_t, bfloat16_t const *, int64_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld3q_vnum_s8)))
+svint8x3_t svld3q_vnum_s8(svbool_t, int8_t const *, int64_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld3q_vnum_f64)))
+svfloat64x3_t svld3q_vnum_f64(svbool_t, float64_t const *, int64_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld3q_vnum_f32)))
+svfloat32x3_t svld3q_vnum_f32(svbool_t, float32_t const *, int64_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld3q_vnum_f16)))
+svfloat16x3_t svld3q_vnum_f16(svbool_t, float16_t const *, int64_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld3q_vnum_s32)))
+svint32x3_t svld3q_vnum_s32(svbool_t, int32_t const *, int64_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld3q_vnum_s64)))
+svint64x3_t svld3q_vnum_s64(svbool_t, int64_t const *, int64_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld3q_vnum_mf8)))
+svmfloat8x3_t svld3q_vnum_mf8(svbool_t, mfloat8_t const *, int64_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld3q_vnum_s16)))
+svint16x3_t svld3q_vnum_s16(svbool_t, int16_t const *, int64_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld4q_u8)))
+svuint8x4_t svld4q_u8(svbool_t, uint8_t const *);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld4q_u32)))
+svuint32x4_t svld4q_u32(svbool_t, uint32_t const *);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld4q_u64)))
+svuint64x4_t svld4q_u64(svbool_t, uint64_t const *);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld4q_u16)))
+svuint16x4_t svld4q_u16(svbool_t, uint16_t const *);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld4q_bf16)))
+svbfloat16x4_t svld4q_bf16(svbool_t, bfloat16_t const *);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld4q_s8)))
+svint8x4_t svld4q_s8(svbool_t, int8_t const *);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld4q_f64)))
+svfloat64x4_t svld4q_f64(svbool_t, float64_t const *);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld4q_f32)))
+svfloat32x4_t svld4q_f32(svbool_t, float32_t const *);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld4q_f16)))
+svfloat16x4_t svld4q_f16(svbool_t, float16_t const *);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld4q_s32)))
+svint32x4_t svld4q_s32(svbool_t, int32_t const *);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld4q_s64)))
+svint64x4_t svld4q_s64(svbool_t, int64_t const *);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld4q_mf8)))
+svmfloat8x4_t svld4q_mf8(svbool_t, mfloat8_t const *);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld4q_s16)))
+svint16x4_t svld4q_s16(svbool_t, int16_t const *);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld4q_vnum_u8)))
+svuint8x4_t svld4q_vnum_u8(svbool_t, uint8_t const *, int64_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld4q_vnum_u32)))
+svuint32x4_t svld4q_vnum_u32(svbool_t, uint32_t const *, int64_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld4q_vnum_u64)))
+svuint64x4_t svld4q_vnum_u64(svbool_t, uint64_t const *, int64_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld4q_vnum_u16)))
+svuint16x4_t svld4q_vnum_u16(svbool_t, uint16_t const *, int64_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld4q_vnum_bf16)))
+svbfloat16x4_t svld4q_vnum_bf16(svbool_t, bfloat16_t const *, int64_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld4q_vnum_s8)))
+svint8x4_t svld4q_vnum_s8(svbool_t, int8_t const *, int64_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld4q_vnum_f64)))
+svfloat64x4_t svld4q_vnum_f64(svbool_t, float64_t const *, int64_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld4q_vnum_f32)))
+svfloat32x4_t svld4q_vnum_f32(svbool_t, float32_t const *, int64_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld4q_vnum_f16)))
+svfloat16x4_t svld4q_vnum_f16(svbool_t, float16_t const *, int64_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld4q_vnum_s32)))
+svint32x4_t svld4q_vnum_s32(svbool_t, int32_t const *, int64_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld4q_vnum_s64)))
+svint64x4_t svld4q_vnum_s64(svbool_t, int64_t const *, int64_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld4q_vnum_mf8)))
+svmfloat8x4_t svld4q_vnum_mf8(svbool_t, mfloat8_t const *, int64_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld4q_vnum_s16)))
+svint16x4_t svld4q_vnum_s16(svbool_t, int16_t const *, int64_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmaxnmqv_f64)))
+float64x2_t svmaxnmqv_f64(svbool_t, svfloat64_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmaxnmqv_f32)))
+float32x4_t svmaxnmqv_f32(svbool_t, svfloat32_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmaxnmqv_f16)))
+float16x8_t svmaxnmqv_f16(svbool_t, svfloat16_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmaxqv_f64)))
+float64x2_t svmaxqv_f64(svbool_t, svfloat64_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmaxqv_f32)))
+float32x4_t svmaxqv_f32(svbool_t, svfloat32_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmaxqv_f16)))
+float16x8_t svmaxqv_f16(svbool_t, svfloat16_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmaxqv_s8)))
+int8x16_t svmaxqv_s8(svbool_t, svint8_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmaxqv_s32)))
+int32x4_t svmaxqv_s32(svbool_t, svint32_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmaxqv_s64)))
+int64x2_t svmaxqv_s64(svbool_t, svint64_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmaxqv_s16)))
+int16x8_t svmaxqv_s16(svbool_t, svint16_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmaxqv_u8)))
+uint8x16_t svmaxqv_u8(svbool_t, svuint8_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmaxqv_u32)))
+uint32x4_t svmaxqv_u32(svbool_t, svuint32_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmaxqv_u64)))
+uint64x2_t svmaxqv_u64(svbool_t, svuint64_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmaxqv_u16)))
+uint16x8_t svmaxqv_u16(svbool_t, svuint16_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svminnmqv_f64)))
+float64x2_t svminnmqv_f64(svbool_t, svfloat64_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svminnmqv_f32)))
+float32x4_t svminnmqv_f32(svbool_t, svfloat32_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svminnmqv_f16)))
+float16x8_t svminnmqv_f16(svbool_t, svfloat16_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svminqv_f64)))
+float64x2_t svminqv_f64(svbool_t, svfloat64_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svminqv_f32)))
+float32x4_t svminqv_f32(svbool_t, svfloat32_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svminqv_f16)))
+float16x8_t svminqv_f16(svbool_t, svfloat16_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svminqv_s8)))
+int8x16_t svminqv_s8(svbool_t, svint8_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svminqv_s32)))
+int32x4_t svminqv_s32(svbool_t, svint32_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svminqv_s64)))
+int64x2_t svminqv_s64(svbool_t, svint64_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svminqv_s16)))
+int16x8_t svminqv_s16(svbool_t, svint16_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svminqv_u8)))
+uint8x16_t svminqv_u8(svbool_t, svuint8_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svminqv_u32)))
+uint32x4_t svminqv_u32(svbool_t, svuint32_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svminqv_u64)))
+uint64x2_t svminqv_u64(svbool_t, svuint64_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svminqv_u16)))
+uint16x8_t svminqv_u16(svbool_t, svuint16_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svorqv_u8)))
+uint8x16_t svorqv_u8(svbool_t, svuint8_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svorqv_u32)))
+uint32x4_t svorqv_u32(svbool_t, svuint32_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svorqv_u64)))
+uint64x2_t svorqv_u64(svbool_t, svuint64_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svorqv_u16)))
+uint16x8_t svorqv_u16(svbool_t, svuint16_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svorqv_s8)))
+int8x16_t svorqv_s8(svbool_t, svint8_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svorqv_s32)))
+int32x4_t svorqv_s32(svbool_t, svint32_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svorqv_s64)))
+int64x2_t svorqv_s64(svbool_t, svint64_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svorqv_s16)))
+int16x8_t svorqv_s16(svbool_t, svint16_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svpmov_u8)))
+svbool_t svpmov_u8(svuint8_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svpmov_s8)))
+svbool_t svpmov_s8(svint8_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svpmov_u64)))
+svbool_t svpmov_u64(svuint64_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svpmov_s64)))
+svbool_t svpmov_s64(svint64_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svpmov_u16)))
+svbool_t svpmov_u16(svuint16_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svpmov_s16)))
+svbool_t svpmov_s16(svint16_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svpmov_u32)))
+svbool_t svpmov_u32(svuint32_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svpmov_s32)))
+svbool_t svpmov_s32(svint32_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svpmov_lane_u8)))
+svbool_t svpmov_lane_u8(svuint8_t, uint64_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svpmov_lane_s8)))
+svbool_t svpmov_lane_s8(svint8_t, uint64_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svpmov_lane_u64)))
+svbool_t svpmov_lane_u64(svuint64_t, uint64_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svpmov_lane_s64)))
+svbool_t svpmov_lane_s64(svint64_t, uint64_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svpmov_lane_u16)))
+svbool_t svpmov_lane_u16(svuint16_t, uint64_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svpmov_lane_s16)))
+svbool_t svpmov_lane_s16(svint16_t, uint64_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svpmov_lane_u32)))
+svbool_t svpmov_lane_u32(svuint32_t, uint64_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svpmov_lane_s32)))
+svbool_t svpmov_lane_s32(svint32_t, uint64_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svpmov_lane_u64_m)))
+svuint64_t svpmov_lane_u64_m(svuint64_t, svbool_t, uint64_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svpmov_lane_s64_m)))
+svint64_t svpmov_lane_s64_m(svint64_t, svbool_t, uint64_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svpmov_lane_u16_m)))
+svuint16_t svpmov_lane_u16_m(svuint16_t, svbool_t, uint64_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svpmov_lane_s16_m)))
+svint16_t svpmov_lane_s16_m(svint16_t, svbool_t, uint64_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svpmov_lane_u32_m)))
+svuint32_t svpmov_lane_u32_m(svuint32_t, svbool_t, uint64_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svpmov_lane_s32_m)))
+svint32_t svpmov_lane_s32_m(svint32_t, svbool_t, uint64_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svpmov_u8_z)))
+svuint8_t svpmov_u8_z(svbool_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svpmov_s8_z)))
+svint8_t svpmov_s8_z(svbool_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svpmov_u64_z)))
+svuint64_t svpmov_u64_z(svbool_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svpmov_s64_z)))
+svint64_t svpmov_s64_z(svbool_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svpmov_u16_z)))
+svuint16_t svpmov_u16_z(svbool_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svpmov_s16_z)))
+svint16_t svpmov_s16_z(svbool_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svpmov_u32_z)))
+svuint32_t svpmov_u32_z(svbool_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svpmov_s32_z)))
+svint32_t svpmov_s32_z(svbool_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst2q_u8)))
+void svst2q_u8(svbool_t, uint8_t const *, svuint8x2_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst2q_u32)))
+void svst2q_u32(svbool_t, uint32_t const *, svuint32x2_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst2q_u64)))
+void svst2q_u64(svbool_t, uint64_t const *, svuint64x2_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst2q_u16)))
+void svst2q_u16(svbool_t, uint16_t const *, svuint16x2_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst2q_bf16)))
+void svst2q_bf16(svbool_t, bfloat16_t const *, svbfloat16x2_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst2q_s8)))
+void svst2q_s8(svbool_t, int8_t const *, svint8x2_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst2q_f64)))
+void svst2q_f64(svbool_t, float64_t const *, svfloat64x2_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst2q_f32)))
+void svst2q_f32(svbool_t, float32_t const *, svfloat32x2_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst2q_f16)))
+void svst2q_f16(svbool_t, float16_t const *, svfloat16x2_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst2q_s32)))
+void svst2q_s32(svbool_t, int32_t const *, svint32x2_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst2q_s64)))
+void svst2q_s64(svbool_t, int64_t const *, svint64x2_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst2q_mf8)))
+void svst2q_mf8(svbool_t, mfloat8_t const *, svmfloat8x2_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst2q_s16)))
+void svst2q_s16(svbool_t, int16_t const *, svint16x2_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst2q_vnum_u8)))
+void svst2q_vnum_u8(svbool_t, uint8_t const *, int64_t, svuint8x2_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst2q_vnum_u32)))
+void svst2q_vnum_u32(svbool_t, uint32_t const *, int64_t, svuint32x2_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst2q_vnum_u64)))
+void svst2q_vnum_u64(svbool_t, uint64_t const *, int64_t, svuint64x2_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst2q_vnum_u16)))
+void svst2q_vnum_u16(svbool_t, uint16_t const *, int64_t, svuint16x2_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst2q_vnum_bf16)))
+void svst2q_vnum_bf16(svbool_t, bfloat16_t const *, int64_t, svbfloat16x2_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst2q_vnum_s8)))
+void svst2q_vnum_s8(svbool_t, int8_t const *, int64_t, svint8x2_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst2q_vnum_f64)))
+void svst2q_vnum_f64(svbool_t, float64_t const *, int64_t, svfloat64x2_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst2q_vnum_f32)))
+void svst2q_vnum_f32(svbool_t, float32_t const *, int64_t, svfloat32x2_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst2q_vnum_f16)))
+void svst2q_vnum_f16(svbool_t, float16_t const *, int64_t, svfloat16x2_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst2q_vnum_s32)))
+void svst2q_vnum_s32(svbool_t, int32_t const *, int64_t, svint32x2_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst2q_vnum_s64)))
+void svst2q_vnum_s64(svbool_t, int64_t const *, int64_t, svint64x2_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst2q_vnum_mf8)))
+void svst2q_vnum_mf8(svbool_t, mfloat8_t const *, int64_t, svmfloat8x2_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst2q_vnum_s16)))
+void svst2q_vnum_s16(svbool_t, int16_t const *, int64_t, svint16x2_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst3q_u8)))
+void svst3q_u8(svbool_t, uint8_t const *, svuint8x3_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst3q_u32)))
+void svst3q_u32(svbool_t, uint32_t const *, svuint32x3_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst3q_u64)))
+void svst3q_u64(svbool_t, uint64_t const *, svuint64x3_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst3q_u16)))
+void svst3q_u16(svbool_t, uint16_t const *, svuint16x3_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst3q_bf16)))
+void svst3q_bf16(svbool_t, bfloat16_t const *, svbfloat16x3_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst3q_s8)))
+void svst3q_s8(svbool_t, int8_t const *, svint8x3_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst3q_f64)))
+void svst3q_f64(svbool_t, float64_t const *, svfloat64x3_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst3q_f32)))
+void svst3q_f32(svbool_t, float32_t const *, svfloat32x3_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst3q_f16)))
+void svst3q_f16(svbool_t, float16_t const *, svfloat16x3_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst3q_s32)))
+void svst3q_s32(svbool_t, int32_t const *, svint32x3_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst3q_s64)))
+void svst3q_s64(svbool_t, int64_t const *, svint64x3_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst3q_mf8)))
+void svst3q_mf8(svbool_t, mfloat8_t const *, svmfloat8x3_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst3q_s16)))
+void svst3q_s16(svbool_t, int16_t const *, svint16x3_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst3q_vnum_u8)))
+void svst3q_vnum_u8(svbool_t, uint8_t const *, int64_t, svuint8x3_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst3q_vnum_u32)))
+void svst3q_vnum_u32(svbool_t, uint32_t const *, int64_t, svuint32x3_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst3q_vnum_u64)))
+void svst3q_vnum_u64(svbool_t, uint64_t const *, int64_t, svuint64x3_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst3q_vnum_u16)))
+void svst3q_vnum_u16(svbool_t, uint16_t const *, int64_t, svuint16x3_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst3q_vnum_bf16)))
+void svst3q_vnum_bf16(svbool_t, bfloat16_t const *, int64_t, svbfloat16x3_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst3q_vnum_s8)))
+void svst3q_vnum_s8(svbool_t, int8_t const *, int64_t, svint8x3_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst3q_vnum_f64)))
+void svst3q_vnum_f64(svbool_t, float64_t const *, int64_t, svfloat64x3_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst3q_vnum_f32)))
+void svst3q_vnum_f32(svbool_t, float32_t const *, int64_t, svfloat32x3_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst3q_vnum_f16)))
+void svst3q_vnum_f16(svbool_t, float16_t const *, int64_t, svfloat16x3_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst3q_vnum_s32)))
+void svst3q_vnum_s32(svbool_t, int32_t const *, int64_t, svint32x3_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst3q_vnum_s64)))
+void svst3q_vnum_s64(svbool_t, int64_t const *, int64_t, svint64x3_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst3q_vnum_mf8)))
+void svst3q_vnum_mf8(svbool_t, mfloat8_t const *, int64_t, svmfloat8x3_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst3q_vnum_s16)))
+void svst3q_vnum_s16(svbool_t, int16_t const *, int64_t, svint16x3_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst4q_u8)))
+void svst4q_u8(svbool_t, uint8_t const *, svuint8x4_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst4q_u32)))
+void svst4q_u32(svbool_t, uint32_t const *, svuint32x4_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst4q_u64)))
+void svst4q_u64(svbool_t, uint64_t const *, svuint64x4_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst4q_u16)))
+void svst4q_u16(svbool_t, uint16_t const *, svuint16x4_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst4q_bf16)))
+void svst4q_bf16(svbool_t, bfloat16_t const *, svbfloat16x4_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst4q_s8)))
+void svst4q_s8(svbool_t, int8_t const *, svint8x4_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst4q_f64)))
+void svst4q_f64(svbool_t, float64_t const *, svfloat64x4_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst4q_f32)))
+void svst4q_f32(svbool_t, float32_t const *, svfloat32x4_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst4q_f16)))
+void svst4q_f16(svbool_t, float16_t const *, svfloat16x4_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst4q_s32)))
+void svst4q_s32(svbool_t, int32_t const *, svint32x4_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst4q_s64)))
+void svst4q_s64(svbool_t, int64_t const *, svint64x4_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst4q_mf8)))
+void svst4q_mf8(svbool_t, mfloat8_t const *, svmfloat8x4_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst4q_s16)))
+void svst4q_s16(svbool_t, int16_t const *, svint16x4_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst4q_vnum_u8)))
+void svst4q_vnum_u8(svbool_t, uint8_t const *, int64_t, svuint8x4_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst4q_vnum_u32)))
+void svst4q_vnum_u32(svbool_t, uint32_t const *, int64_t, svuint32x4_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst4q_vnum_u64)))
+void svst4q_vnum_u64(svbool_t, uint64_t const *, int64_t, svuint64x4_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst4q_vnum_u16)))
+void svst4q_vnum_u16(svbool_t, uint16_t const *, int64_t, svuint16x4_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst4q_vnum_bf16)))
+void svst4q_vnum_bf16(svbool_t, bfloat16_t const *, int64_t, svbfloat16x4_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst4q_vnum_s8)))
+void svst4q_vnum_s8(svbool_t, int8_t const *, int64_t, svint8x4_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst4q_vnum_f64)))
+void svst4q_vnum_f64(svbool_t, float64_t const *, int64_t, svfloat64x4_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst4q_vnum_f32)))
+void svst4q_vnum_f32(svbool_t, float32_t const *, int64_t, svfloat32x4_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst4q_vnum_f16)))
+void svst4q_vnum_f16(svbool_t, float16_t const *, int64_t, svfloat16x4_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst4q_vnum_s32)))
+void svst4q_vnum_s32(svbool_t, int32_t const *, int64_t, svint32x4_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst4q_vnum_s64)))
+void svst4q_vnum_s64(svbool_t, int64_t const *, int64_t, svint64x4_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst4q_vnum_mf8)))
+void svst4q_vnum_mf8(svbool_t, mfloat8_t const *, int64_t, svmfloat8x4_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst4q_vnum_s16)))
+void svst4q_vnum_s16(svbool_t, int16_t const *, int64_t, svint16x4_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtblq_u8)))
+svuint8_t svtblq_u8(svuint8_t, svuint8_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtblq_u32)))
+svuint32_t svtblq_u32(svuint32_t, svuint32_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtblq_u64)))
+svuint64_t svtblq_u64(svuint64_t, svuint64_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtblq_u16)))
+svuint16_t svtblq_u16(svuint16_t, svuint16_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtblq_bf16)))
+svbfloat16_t svtblq_bf16(svbfloat16_t, svuint16_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtblq_s8)))
+svint8_t svtblq_s8(svint8_t, svuint8_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtblq_f64)))
+svfloat64_t svtblq_f64(svfloat64_t, svuint64_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtblq_f32)))
+svfloat32_t svtblq_f32(svfloat32_t, svuint32_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtblq_f16)))
+svfloat16_t svtblq_f16(svfloat16_t, svuint16_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtblq_s32)))
+svint32_t svtblq_s32(svint32_t, svuint32_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtblq_s64)))
+svint64_t svtblq_s64(svint64_t, svuint64_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtblq_mf8)))
+svmfloat8_t svtblq_mf8(svmfloat8_t, svuint8_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtblq_s16)))
+svint16_t svtblq_s16(svint16_t, svuint16_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtbxq_u8)))
+svuint8_t svtbxq_u8(svuint8_t, svuint8_t, svuint8_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtbxq_u32)))
+svuint32_t svtbxq_u32(svuint32_t, svuint32_t, svuint32_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtbxq_u64)))
+svuint64_t svtbxq_u64(svuint64_t, svuint64_t, svuint64_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtbxq_u16)))
+svuint16_t svtbxq_u16(svuint16_t, svuint16_t, svuint16_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtbxq_bf16)))
+svbfloat16_t svtbxq_bf16(svbfloat16_t, svbfloat16_t, svuint16_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtbxq_s8)))
+svint8_t svtbxq_s8(svint8_t, svint8_t, svuint8_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtbxq_f64)))
+svfloat64_t svtbxq_f64(svfloat64_t, svfloat64_t, svuint64_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtbxq_f32)))
+svfloat32_t svtbxq_f32(svfloat32_t, svfloat32_t, svuint32_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtbxq_f16)))
+svfloat16_t svtbxq_f16(svfloat16_t, svfloat16_t, svuint16_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtbxq_s32)))
+svint32_t svtbxq_s32(svint32_t, svint32_t, svuint32_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtbxq_s64)))
+svint64_t svtbxq_s64(svint64_t, svint64_t, svuint64_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtbxq_mf8)))
+svmfloat8_t svtbxq_mf8(svmfloat8_t, svmfloat8_t, svuint8_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtbxq_s16)))
+svint16_t svtbxq_s16(svint16_t, svint16_t, svuint16_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzpq1_u8)))
+svuint8_t svuzpq1_u8(svuint8_t, svuint8_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzpq1_u32)))
+svuint32_t svuzpq1_u32(svuint32_t, svuint32_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzpq1_u64)))
+svuint64_t svuzpq1_u64(svuint64_t, svuint64_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzpq1_u16)))
+svuint16_t svuzpq1_u16(svuint16_t, svuint16_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzpq1_bf16)))
+svbfloat16_t svuzpq1_bf16(svbfloat16_t, svbfloat16_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzpq1_s8)))
+svint8_t svuzpq1_s8(svint8_t, svint8_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzpq1_f64)))
+svfloat64_t svuzpq1_f64(svfloat64_t, svfloat64_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzpq1_f32)))
+svfloat32_t svuzpq1_f32(svfloat32_t, svfloat32_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzpq1_f16)))
+svfloat16_t svuzpq1_f16(svfloat16_t, svfloat16_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzpq1_s32)))
+svint32_t svuzpq1_s32(svint32_t, svint32_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzpq1_s64)))
+svint64_t svuzpq1_s64(svint64_t, svint64_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzpq1_mf8)))
+svmfloat8_t svuzpq1_mf8(svmfloat8_t, svmfloat8_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzpq1_s16)))
+svint16_t svuzpq1_s16(svint16_t, svint16_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzpq2_u8)))
+svuint8_t svuzpq2_u8(svuint8_t, svuint8_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzpq2_u32)))
+svuint32_t svuzpq2_u32(svuint32_t, svuint32_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzpq2_u64)))
+svuint64_t svuzpq2_u64(svuint64_t, svuint64_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzpq2_u16)))
+svuint16_t svuzpq2_u16(svuint16_t, svuint16_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzpq2_bf16)))
+svbfloat16_t svuzpq2_bf16(svbfloat16_t, svbfloat16_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzpq2_s8)))
+svint8_t svuzpq2_s8(svint8_t, svint8_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzpq2_f64)))
+svfloat64_t svuzpq2_f64(svfloat64_t, svfloat64_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzpq2_f32)))
+svfloat32_t svuzpq2_f32(svfloat32_t, svfloat32_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzpq2_f16)))
+svfloat16_t svuzpq2_f16(svfloat16_t, svfloat16_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzpq2_s32)))
+svint32_t svuzpq2_s32(svint32_t, svint32_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzpq2_s64)))
+svint64_t svuzpq2_s64(svint64_t, svint64_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzpq2_mf8)))
+svmfloat8_t svuzpq2_mf8(svmfloat8_t, svmfloat8_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzpq2_s16)))
+svint16_t svuzpq2_s16(svint16_t, svint16_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzipq1_u8)))
+svuint8_t svzipq1_u8(svuint8_t, svuint8_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzipq1_u32)))
+svuint32_t svzipq1_u32(svuint32_t, svuint32_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzipq1_u64)))
+svuint64_t svzipq1_u64(svuint64_t, svuint64_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzipq1_u16)))
+svuint16_t svzipq1_u16(svuint16_t, svuint16_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzipq1_bf16)))
+svbfloat16_t svzipq1_bf16(svbfloat16_t, svbfloat16_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzipq1_s8)))
+svint8_t svzipq1_s8(svint8_t, svint8_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzipq1_f64)))
+svfloat64_t svzipq1_f64(svfloat64_t, svfloat64_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzipq1_f32)))
+svfloat32_t svzipq1_f32(svfloat32_t, svfloat32_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzipq1_f16)))
+svfloat16_t svzipq1_f16(svfloat16_t, svfloat16_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzipq1_s32)))
+svint32_t svzipq1_s32(svint32_t, svint32_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzipq1_s64)))
+svint64_t svzipq1_s64(svint64_t, svint64_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzipq1_mf8)))
+svmfloat8_t svzipq1_mf8(svmfloat8_t, svmfloat8_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzipq1_s16)))
+svint16_t svzipq1_s16(svint16_t, svint16_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzipq2_u8)))
+svuint8_t svzipq2_u8(svuint8_t, svuint8_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzipq2_u32)))
+svuint32_t svzipq2_u32(svuint32_t, svuint32_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzipq2_u64)))
+svuint64_t svzipq2_u64(svuint64_t, svuint64_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzipq2_u16)))
+svuint16_t svzipq2_u16(svuint16_t, svuint16_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzipq2_bf16)))
+svbfloat16_t svzipq2_bf16(svbfloat16_t, svbfloat16_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzipq2_s8)))
+svint8_t svzipq2_s8(svint8_t, svint8_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzipq2_f64)))
+svfloat64_t svzipq2_f64(svfloat64_t, svfloat64_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzipq2_f32)))
+svfloat32_t svzipq2_f32(svfloat32_t, svfloat32_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzipq2_f16)))
+svfloat16_t svzipq2_f16(svfloat16_t, svfloat16_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzipq2_s32)))
+svint32_t svzipq2_s32(svint32_t, svint32_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzipq2_s64)))
+svint64_t svzipq2_s64(svint64_t, svint64_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzipq2_mf8)))
+svmfloat8_t svzipq2_mf8(svmfloat8_t, svmfloat8_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzipq2_s16)))
+svint16_t svzipq2_s16(svint16_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddqv_u8)))
+uint8x16_t svaddqv(svbool_t, svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddqv_u32)))
+uint32x4_t svaddqv(svbool_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddqv_u64)))
+uint64x2_t svaddqv(svbool_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddqv_u16)))
+uint16x8_t svaddqv(svbool_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddqv_s8)))
+int8x16_t svaddqv(svbool_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddqv_s32)))
+int32x4_t svaddqv(svbool_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddqv_s64)))
+int64x2_t svaddqv(svbool_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddqv_s16)))
+int16x8_t svaddqv(svbool_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddqv_f64)))
+float64x2_t svaddqv(svbool_t, svfloat64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddqv_f32)))
+float32x4_t svaddqv(svbool_t, svfloat32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddqv_f16)))
+float16x8_t svaddqv(svbool_t, svfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svandqv_u8)))
+uint8x16_t svandqv(svbool_t, svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svandqv_u32)))
+uint32x4_t svandqv(svbool_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svandqv_u64)))
+uint64x2_t svandqv(svbool_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svandqv_u16)))
+uint16x8_t svandqv(svbool_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svandqv_s8)))
+int8x16_t svandqv(svbool_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svandqv_s32)))
+int32x4_t svandqv(svbool_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svandqv_s64)))
+int64x2_t svandqv(svbool_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svandqv_s16)))
+int16x8_t svandqv(svbool_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdup_laneq_u8)))
+svuint8_t svdup_laneq(svuint8_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdup_laneq_s8)))
+svint8_t svdup_laneq(svint8_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdup_laneq_mf8)))
+svmfloat8_t svdup_laneq(svmfloat8_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdup_laneq_u64)))
+svuint64_t svdup_laneq(svuint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdup_laneq_f64)))
+svfloat64_t svdup_laneq(svfloat64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdup_laneq_s64)))
+svint64_t svdup_laneq(svint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdup_laneq_u16)))
+svuint16_t svdup_laneq(svuint16_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdup_laneq_bf16)))
+svbfloat16_t svdup_laneq(svbfloat16_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdup_laneq_f16)))
+svfloat16_t svdup_laneq(svfloat16_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdup_laneq_s16)))
+svint16_t svdup_laneq(svint16_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdup_laneq_u32)))
+svuint32_t svdup_laneq(svuint32_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdup_laneq_f32)))
+svfloat32_t svdup_laneq(svfloat32_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdup_laneq_s32)))
+svint32_t svdup_laneq(svint32_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveorqv_u8)))
+uint8x16_t sveorqv(svbool_t, svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveorqv_u32)))
+uint32x4_t sveorqv(svbool_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveorqv_u64)))
+uint64x2_t sveorqv(svbool_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveorqv_u16)))
+uint16x8_t sveorqv(svbool_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveorqv_s8)))
+int8x16_t sveorqv(svbool_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveorqv_s32)))
+int32x4_t sveorqv(svbool_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveorqv_s64)))
+int64x2_t sveorqv(svbool_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveorqv_s16)))
+int16x8_t sveorqv(svbool_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svextq_u8)))
+svuint8_t svextq(svuint8_t, svuint8_t, int32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svextq_u32)))
+svuint32_t svextq(svuint32_t, svuint32_t, int32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svextq_u64)))
+svuint64_t svextq(svuint64_t, svuint64_t, int32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svextq_u16)))
+svuint16_t svextq(svuint16_t, svuint16_t, int32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svextq_bf16)))
+svbfloat16_t svextq(svbfloat16_t, svbfloat16_t, int32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svextq_s8)))
+svint8_t svextq(svint8_t, svint8_t, int32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svextq_f64)))
+svfloat64_t svextq(svfloat64_t, svfloat64_t, int32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svextq_f32)))
+svfloat32_t svextq(svfloat32_t, svfloat32_t, int32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svextq_f16)))
+svfloat16_t svextq(svfloat16_t, svfloat16_t, int32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svextq_s32)))
+svint32_t svextq(svint32_t, svint32_t, int32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svextq_s64)))
+svint64_t svextq(svint64_t, svint64_t, int32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svextq_mf8)))
+svmfloat8_t svextq(svmfloat8_t, svmfloat8_t, int32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svextq_s16)))
+svint16_t svextq(svint16_t, svint16_t, int32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld2q_u8)))
+svuint8x2_t svld2q(svbool_t, uint8_t const *);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld2q_u32)))
+svuint32x2_t svld2q(svbool_t, uint32_t const *);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld2q_u64)))
+svuint64x2_t svld2q(svbool_t, uint64_t const *);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld2q_u16)))
+svuint16x2_t svld2q(svbool_t, uint16_t const *);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld2q_bf16)))
+svbfloat16x2_t svld2q(svbool_t, bfloat16_t const *);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld2q_s8)))
+svint8x2_t svld2q(svbool_t, int8_t const *);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld2q_f64)))
+svfloat64x2_t svld2q(svbool_t, float64_t const *);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld2q_f32)))
+svfloat32x2_t svld2q(svbool_t, float32_t const *);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld2q_f16)))
+svfloat16x2_t svld2q(svbool_t, float16_t const *);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld2q_s32)))
+svint32x2_t svld2q(svbool_t, int32_t const *);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld2q_s64)))
+svint64x2_t svld2q(svbool_t, int64_t const *);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld2q_mf8)))
+svmfloat8x2_t svld2q(svbool_t, mfloat8_t const *);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld2q_s16)))
+svint16x2_t svld2q(svbool_t, int16_t const *);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld2q_vnum_u8)))
+svuint8x2_t svld2q_vnum(svbool_t, uint8_t const *, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld2q_vnum_u32)))
+svuint32x2_t svld2q_vnum(svbool_t, uint32_t const *, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld2q_vnum_u64)))
+svuint64x2_t svld2q_vnum(svbool_t, uint64_t const *, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld2q_vnum_u16)))
+svuint16x2_t svld2q_vnum(svbool_t, uint16_t const *, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld2q_vnum_bf16)))
+svbfloat16x2_t svld2q_vnum(svbool_t, bfloat16_t const *, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld2q_vnum_s8)))
+svint8x2_t svld2q_vnum(svbool_t, int8_t const *, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld2q_vnum_f64)))
+svfloat64x2_t svld2q_vnum(svbool_t, float64_t const *, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld2q_vnum_f32)))
+svfloat32x2_t svld2q_vnum(svbool_t, float32_t const *, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld2q_vnum_f16)))
+svfloat16x2_t svld2q_vnum(svbool_t, float16_t const *, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld2q_vnum_s32)))
+svint32x2_t svld2q_vnum(svbool_t, int32_t const *, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld2q_vnum_s64)))
+svint64x2_t svld2q_vnum(svbool_t, int64_t const *, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld2q_vnum_mf8)))
+svmfloat8x2_t svld2q_vnum(svbool_t, mfloat8_t const *, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld2q_vnum_s16)))
+svint16x2_t svld2q_vnum(svbool_t, int16_t const *, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld3q_u8)))
+svuint8x3_t svld3q(svbool_t, uint8_t const *);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld3q_u32)))
+svuint32x3_t svld3q(svbool_t, uint32_t const *);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld3q_u64)))
+svuint64x3_t svld3q(svbool_t, uint64_t const *);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld3q_u16)))
+svuint16x3_t svld3q(svbool_t, uint16_t const *);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld3q_bf16)))
+svbfloat16x3_t svld3q(svbool_t, bfloat16_t const *);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld3q_s8)))
+svint8x3_t svld3q(svbool_t, int8_t const *);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld3q_f64)))
+svfloat64x3_t svld3q(svbool_t, float64_t const *);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld3q_f32)))
+svfloat32x3_t svld3q(svbool_t, float32_t const *);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld3q_f16)))
+svfloat16x3_t svld3q(svbool_t, float16_t const *);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld3q_s32)))
+svint32x3_t svld3q(svbool_t, int32_t const *);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld3q_s64)))
+svint64x3_t svld3q(svbool_t, int64_t const *);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld3q_mf8)))
+svmfloat8x3_t svld3q(svbool_t, mfloat8_t const *);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld3q_s16)))
+svint16x3_t svld3q(svbool_t, int16_t const *);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld3q_vnum_u8)))
+svuint8x3_t svld3q_vnum(svbool_t, uint8_t const *, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld3q_vnum_u32)))
+svuint32x3_t svld3q_vnum(svbool_t, uint32_t const *, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld3q_vnum_u64)))
+svuint64x3_t svld3q_vnum(svbool_t, uint64_t const *, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld3q_vnum_u16)))
+svuint16x3_t svld3q_vnum(svbool_t, uint16_t const *, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld3q_vnum_bf16)))
+svbfloat16x3_t svld3q_vnum(svbool_t, bfloat16_t const *, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld3q_vnum_s8)))
+svint8x3_t svld3q_vnum(svbool_t, int8_t const *, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld3q_vnum_f64)))
+svfloat64x3_t svld3q_vnum(svbool_t, float64_t const *, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld3q_vnum_f32)))
+svfloat32x3_t svld3q_vnum(svbool_t, float32_t const *, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld3q_vnum_f16)))
+svfloat16x3_t svld3q_vnum(svbool_t, float16_t const *, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld3q_vnum_s32)))
+svint32x3_t svld3q_vnum(svbool_t, int32_t const *, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld3q_vnum_s64)))
+svint64x3_t svld3q_vnum(svbool_t, int64_t const *, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld3q_vnum_mf8)))
+svmfloat8x3_t svld3q_vnum(svbool_t, mfloat8_t const *, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld3q_vnum_s16)))
+svint16x3_t svld3q_vnum(svbool_t, int16_t const *, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld4q_u8)))
+svuint8x4_t svld4q(svbool_t, uint8_t const *);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld4q_u32)))
+svuint32x4_t svld4q(svbool_t, uint32_t const *);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld4q_u64)))
+svuint64x4_t svld4q(svbool_t, uint64_t const *);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld4q_u16)))
+svuint16x4_t svld4q(svbool_t, uint16_t const *);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld4q_bf16)))
+svbfloat16x4_t svld4q(svbool_t, bfloat16_t const *);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld4q_s8)))
+svint8x4_t svld4q(svbool_t, int8_t const *);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld4q_f64)))
+svfloat64x4_t svld4q(svbool_t, float64_t const *);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld4q_f32)))
+svfloat32x4_t svld4q(svbool_t, float32_t const *);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld4q_f16)))
+svfloat16x4_t svld4q(svbool_t, float16_t const *);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld4q_s32)))
+svint32x4_t svld4q(svbool_t, int32_t const *);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld4q_s64)))
+svint64x4_t svld4q(svbool_t, int64_t const *);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld4q_mf8)))
+svmfloat8x4_t svld4q(svbool_t, mfloat8_t const *);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld4q_s16)))
+svint16x4_t svld4q(svbool_t, int16_t const *);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld4q_vnum_u8)))
+svuint8x4_t svld4q_vnum(svbool_t, uint8_t const *, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld4q_vnum_u32)))
+svuint32x4_t svld4q_vnum(svbool_t, uint32_t const *, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld4q_vnum_u64)))
+svuint64x4_t svld4q_vnum(svbool_t, uint64_t const *, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld4q_vnum_u16)))
+svuint16x4_t svld4q_vnum(svbool_t, uint16_t const *, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld4q_vnum_bf16)))
+svbfloat16x4_t svld4q_vnum(svbool_t, bfloat16_t const *, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld4q_vnum_s8)))
+svint8x4_t svld4q_vnum(svbool_t, int8_t const *, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld4q_vnum_f64)))
+svfloat64x4_t svld4q_vnum(svbool_t, float64_t const *, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld4q_vnum_f32)))
+svfloat32x4_t svld4q_vnum(svbool_t, float32_t const *, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld4q_vnum_f16)))
+svfloat16x4_t svld4q_vnum(svbool_t, float16_t const *, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld4q_vnum_s32)))
+svint32x4_t svld4q_vnum(svbool_t, int32_t const *, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld4q_vnum_s64)))
+svint64x4_t svld4q_vnum(svbool_t, int64_t const *, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld4q_vnum_mf8)))
+svmfloat8x4_t svld4q_vnum(svbool_t, mfloat8_t const *, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld4q_vnum_s16)))
+svint16x4_t svld4q_vnum(svbool_t, int16_t const *, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmaxnmqv_f64)))
+float64x2_t svmaxnmqv(svbool_t, svfloat64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmaxnmqv_f32)))
+float32x4_t svmaxnmqv(svbool_t, svfloat32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmaxnmqv_f16)))
+float16x8_t svmaxnmqv(svbool_t, svfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmaxqv_f64)))
+float64x2_t svmaxqv(svbool_t, svfloat64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmaxqv_f32)))
+float32x4_t svmaxqv(svbool_t, svfloat32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmaxqv_f16)))
+float16x8_t svmaxqv(svbool_t, svfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmaxqv_s8)))
+int8x16_t svmaxqv(svbool_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmaxqv_s32)))
+int32x4_t svmaxqv(svbool_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmaxqv_s64)))
+int64x2_t svmaxqv(svbool_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmaxqv_s16)))
+int16x8_t svmaxqv(svbool_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmaxqv_u8)))
+uint8x16_t svmaxqv(svbool_t, svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmaxqv_u32)))
+uint32x4_t svmaxqv(svbool_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmaxqv_u64)))
+uint64x2_t svmaxqv(svbool_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmaxqv_u16)))
+uint16x8_t svmaxqv(svbool_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svminnmqv_f64)))
+float64x2_t svminnmqv(svbool_t, svfloat64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svminnmqv_f32)))
+float32x4_t svminnmqv(svbool_t, svfloat32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svminnmqv_f16)))
+float16x8_t svminnmqv(svbool_t, svfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svminqv_f64)))
+float64x2_t svminqv(svbool_t, svfloat64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svminqv_f32)))
+float32x4_t svminqv(svbool_t, svfloat32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svminqv_f16)))
+float16x8_t svminqv(svbool_t, svfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svminqv_s8)))
+int8x16_t svminqv(svbool_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svminqv_s32)))
+int32x4_t svminqv(svbool_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svminqv_s64)))
+int64x2_t svminqv(svbool_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svminqv_s16)))
+int16x8_t svminqv(svbool_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svminqv_u8)))
+uint8x16_t svminqv(svbool_t, svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svminqv_u32)))
+uint32x4_t svminqv(svbool_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svminqv_u64)))
+uint64x2_t svminqv(svbool_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svminqv_u16)))
+uint16x8_t svminqv(svbool_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svorqv_u8)))
+uint8x16_t svorqv(svbool_t, svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svorqv_u32)))
+uint32x4_t svorqv(svbool_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svorqv_u64)))
+uint64x2_t svorqv(svbool_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svorqv_u16)))
+uint16x8_t svorqv(svbool_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svorqv_s8)))
+int8x16_t svorqv(svbool_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svorqv_s32)))
+int32x4_t svorqv(svbool_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svorqv_s64)))
+int64x2_t svorqv(svbool_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svorqv_s16)))
+int16x8_t svorqv(svbool_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svpmov_u8)))
+svbool_t svpmov(svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svpmov_s8)))
+svbool_t svpmov(svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svpmov_u64)))
+svbool_t svpmov(svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svpmov_s64)))
+svbool_t svpmov(svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svpmov_u16)))
+svbool_t svpmov(svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svpmov_s16)))
+svbool_t svpmov(svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svpmov_u32)))
+svbool_t svpmov(svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svpmov_s32)))
+svbool_t svpmov(svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svpmov_lane_u8)))
+svbool_t svpmov_lane(svuint8_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svpmov_lane_s8)))
+svbool_t svpmov_lane(svint8_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svpmov_lane_u64)))
+svbool_t svpmov_lane(svuint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svpmov_lane_s64)))
+svbool_t svpmov_lane(svint64_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svpmov_lane_u16)))
+svbool_t svpmov_lane(svuint16_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svpmov_lane_s16)))
+svbool_t svpmov_lane(svint16_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svpmov_lane_u32)))
+svbool_t svpmov_lane(svuint32_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svpmov_lane_s32)))
+svbool_t svpmov_lane(svint32_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svpmov_lane_u64_m)))
+svuint64_t svpmov_lane_m(svuint64_t, svbool_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svpmov_lane_s64_m)))
+svint64_t svpmov_lane_m(svint64_t, svbool_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svpmov_lane_u16_m)))
+svuint16_t svpmov_lane_m(svuint16_t, svbool_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svpmov_lane_s16_m)))
+svint16_t svpmov_lane_m(svint16_t, svbool_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svpmov_lane_u32_m)))
+svuint32_t svpmov_lane_m(svuint32_t, svbool_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svpmov_lane_s32_m)))
+svint32_t svpmov_lane_m(svint32_t, svbool_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst2q_u8)))
+void svst2q(svbool_t, uint8_t const *, svuint8x2_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst2q_u32)))
+void svst2q(svbool_t, uint32_t const *, svuint32x2_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst2q_u64)))
+void svst2q(svbool_t, uint64_t const *, svuint64x2_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst2q_u16)))
+void svst2q(svbool_t, uint16_t const *, svuint16x2_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst2q_bf16)))
+void svst2q(svbool_t, bfloat16_t const *, svbfloat16x2_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst2q_s8)))
+void svst2q(svbool_t, int8_t const *, svint8x2_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst2q_f64)))
+void svst2q(svbool_t, float64_t const *, svfloat64x2_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst2q_f32)))
+void svst2q(svbool_t, float32_t const *, svfloat32x2_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst2q_f16)))
+void svst2q(svbool_t, float16_t const *, svfloat16x2_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst2q_s32)))
+void svst2q(svbool_t, int32_t const *, svint32x2_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst2q_s64)))
+void svst2q(svbool_t, int64_t const *, svint64x2_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst2q_mf8)))
+void svst2q(svbool_t, mfloat8_t const *, svmfloat8x2_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst2q_s16)))
+void svst2q(svbool_t, int16_t const *, svint16x2_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst2q_vnum_u8)))
+void svst2q_vnum(svbool_t, uint8_t const *, int64_t, svuint8x2_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst2q_vnum_u32)))
+void svst2q_vnum(svbool_t, uint32_t const *, int64_t, svuint32x2_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst2q_vnum_u64)))
+void svst2q_vnum(svbool_t, uint64_t const *, int64_t, svuint64x2_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst2q_vnum_u16)))
+void svst2q_vnum(svbool_t, uint16_t const *, int64_t, svuint16x2_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst2q_vnum_bf16)))
+void svst2q_vnum(svbool_t, bfloat16_t const *, int64_t, svbfloat16x2_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst2q_vnum_s8)))
+void svst2q_vnum(svbool_t, int8_t const *, int64_t, svint8x2_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst2q_vnum_f64)))
+void svst2q_vnum(svbool_t, float64_t const *, int64_t, svfloat64x2_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst2q_vnum_f32)))
+void svst2q_vnum(svbool_t, float32_t const *, int64_t, svfloat32x2_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst2q_vnum_f16)))
+void svst2q_vnum(svbool_t, float16_t const *, int64_t, svfloat16x2_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst2q_vnum_s32)))
+void svst2q_vnum(svbool_t, int32_t const *, int64_t, svint32x2_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst2q_vnum_s64)))
+void svst2q_vnum(svbool_t, int64_t const *, int64_t, svint64x2_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst2q_vnum_mf8)))
+void svst2q_vnum(svbool_t, mfloat8_t const *, int64_t, svmfloat8x2_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst2q_vnum_s16)))
+void svst2q_vnum(svbool_t, int16_t const *, int64_t, svint16x2_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst3q_u8)))
+void svst3q(svbool_t, uint8_t const *, svuint8x3_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst3q_u32)))
+void svst3q(svbool_t, uint32_t const *, svuint32x3_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst3q_u64)))
+void svst3q(svbool_t, uint64_t const *, svuint64x3_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst3q_u16)))
+void svst3q(svbool_t, uint16_t const *, svuint16x3_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst3q_bf16)))
+void svst3q(svbool_t, bfloat16_t const *, svbfloat16x3_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst3q_s8)))
+void svst3q(svbool_t, int8_t const *, svint8x3_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst3q_f64)))
+void svst3q(svbool_t, float64_t const *, svfloat64x3_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst3q_f32)))
+void svst3q(svbool_t, float32_t const *, svfloat32x3_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst3q_f16)))
+void svst3q(svbool_t, float16_t const *, svfloat16x3_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst3q_s32)))
+void svst3q(svbool_t, int32_t const *, svint32x3_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst3q_s64)))
+void svst3q(svbool_t, int64_t const *, svint64x3_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst3q_mf8)))
+void svst3q(svbool_t, mfloat8_t const *, svmfloat8x3_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst3q_s16)))
+void svst3q(svbool_t, int16_t const *, svint16x3_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst3q_vnum_u8)))
+void svst3q_vnum(svbool_t, uint8_t const *, int64_t, svuint8x3_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst3q_vnum_u32)))
+void svst3q_vnum(svbool_t, uint32_t const *, int64_t, svuint32x3_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst3q_vnum_u64)))
+void svst3q_vnum(svbool_t, uint64_t const *, int64_t, svuint64x3_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst3q_vnum_u16)))
+void svst3q_vnum(svbool_t, uint16_t const *, int64_t, svuint16x3_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst3q_vnum_bf16)))
+void svst3q_vnum(svbool_t, bfloat16_t const *, int64_t, svbfloat16x3_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst3q_vnum_s8)))
+void svst3q_vnum(svbool_t, int8_t const *, int64_t, svint8x3_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst3q_vnum_f64)))
+void svst3q_vnum(svbool_t, float64_t const *, int64_t, svfloat64x3_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst3q_vnum_f32)))
+void svst3q_vnum(svbool_t, float32_t const *, int64_t, svfloat32x3_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst3q_vnum_f16)))
+void svst3q_vnum(svbool_t, float16_t const *, int64_t, svfloat16x3_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst3q_vnum_s32)))
+void svst3q_vnum(svbool_t, int32_t const *, int64_t, svint32x3_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst3q_vnum_s64)))
+void svst3q_vnum(svbool_t, int64_t const *, int64_t, svint64x3_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst3q_vnum_mf8)))
+void svst3q_vnum(svbool_t, mfloat8_t const *, int64_t, svmfloat8x3_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst3q_vnum_s16)))
+void svst3q_vnum(svbool_t, int16_t const *, int64_t, svint16x3_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst4q_u8)))
+void svst4q(svbool_t, uint8_t const *, svuint8x4_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst4q_u32)))
+void svst4q(svbool_t, uint32_t const *, svuint32x4_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst4q_u64)))
+void svst4q(svbool_t, uint64_t const *, svuint64x4_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst4q_u16)))
+void svst4q(svbool_t, uint16_t const *, svuint16x4_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst4q_bf16)))
+void svst4q(svbool_t, bfloat16_t const *, svbfloat16x4_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst4q_s8)))
+void svst4q(svbool_t, int8_t const *, svint8x4_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst4q_f64)))
+void svst4q(svbool_t, float64_t const *, svfloat64x4_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst4q_f32)))
+void svst4q(svbool_t, float32_t const *, svfloat32x4_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst4q_f16)))
+void svst4q(svbool_t, float16_t const *, svfloat16x4_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst4q_s32)))
+void svst4q(svbool_t, int32_t const *, svint32x4_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst4q_s64)))
+void svst4q(svbool_t, int64_t const *, svint64x4_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst4q_mf8)))
+void svst4q(svbool_t, mfloat8_t const *, svmfloat8x4_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst4q_s16)))
+void svst4q(svbool_t, int16_t const *, svint16x4_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst4q_vnum_u8)))
+void svst4q_vnum(svbool_t, uint8_t const *, int64_t, svuint8x4_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst4q_vnum_u32)))
+void svst4q_vnum(svbool_t, uint32_t const *, int64_t, svuint32x4_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst4q_vnum_u64)))
+void svst4q_vnum(svbool_t, uint64_t const *, int64_t, svuint64x4_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst4q_vnum_u16)))
+void svst4q_vnum(svbool_t, uint16_t const *, int64_t, svuint16x4_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst4q_vnum_bf16)))
+void svst4q_vnum(svbool_t, bfloat16_t const *, int64_t, svbfloat16x4_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst4q_vnum_s8)))
+void svst4q_vnum(svbool_t, int8_t const *, int64_t, svint8x4_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst4q_vnum_f64)))
+void svst4q_vnum(svbool_t, float64_t const *, int64_t, svfloat64x4_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst4q_vnum_f32)))
+void svst4q_vnum(svbool_t, float32_t const *, int64_t, svfloat32x4_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst4q_vnum_f16)))
+void svst4q_vnum(svbool_t, float16_t const *, int64_t, svfloat16x4_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst4q_vnum_s32)))
+void svst4q_vnum(svbool_t, int32_t const *, int64_t, svint32x4_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst4q_vnum_s64)))
+void svst4q_vnum(svbool_t, int64_t const *, int64_t, svint64x4_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst4q_vnum_mf8)))
+void svst4q_vnum(svbool_t, mfloat8_t const *, int64_t, svmfloat8x4_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst4q_vnum_s16)))
+void svst4q_vnum(svbool_t, int16_t const *, int64_t, svint16x4_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtblq_u8)))
+svuint8_t svtblq(svuint8_t, svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtblq_u32)))
+svuint32_t svtblq(svuint32_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtblq_u64)))
+svuint64_t svtblq(svuint64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtblq_u16)))
+svuint16_t svtblq(svuint16_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtblq_bf16)))
+svbfloat16_t svtblq(svbfloat16_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtblq_s8)))
+svint8_t svtblq(svint8_t, svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtblq_f64)))
+svfloat64_t svtblq(svfloat64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtblq_f32)))
+svfloat32_t svtblq(svfloat32_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtblq_f16)))
+svfloat16_t svtblq(svfloat16_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtblq_s32)))
+svint32_t svtblq(svint32_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtblq_s64)))
+svint64_t svtblq(svint64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtblq_mf8)))
+svmfloat8_t svtblq(svmfloat8_t, svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtblq_s16)))
+svint16_t svtblq(svint16_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtbxq_u8)))
+svuint8_t svtbxq(svuint8_t, svuint8_t, svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtbxq_u32)))
+svuint32_t svtbxq(svuint32_t, svuint32_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtbxq_u64)))
+svuint64_t svtbxq(svuint64_t, svuint64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtbxq_u16)))
+svuint16_t svtbxq(svuint16_t, svuint16_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtbxq_bf16)))
+svbfloat16_t svtbxq(svbfloat16_t, svbfloat16_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtbxq_s8)))
+svint8_t svtbxq(svint8_t, svint8_t, svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtbxq_f64)))
+svfloat64_t svtbxq(svfloat64_t, svfloat64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtbxq_f32)))
+svfloat32_t svtbxq(svfloat32_t, svfloat32_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtbxq_f16)))
+svfloat16_t svtbxq(svfloat16_t, svfloat16_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtbxq_s32)))
+svint32_t svtbxq(svint32_t, svint32_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtbxq_s64)))
+svint64_t svtbxq(svint64_t, svint64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtbxq_mf8)))
+svmfloat8_t svtbxq(svmfloat8_t, svmfloat8_t, svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtbxq_s16)))
+svint16_t svtbxq(svint16_t, svint16_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzpq1_u8)))
+svuint8_t svuzpq1(svuint8_t, svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzpq1_u32)))
+svuint32_t svuzpq1(svuint32_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzpq1_u64)))
+svuint64_t svuzpq1(svuint64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzpq1_u16)))
+svuint16_t svuzpq1(svuint16_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzpq1_bf16)))
+svbfloat16_t svuzpq1(svbfloat16_t, svbfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzpq1_s8)))
+svint8_t svuzpq1(svint8_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzpq1_f64)))
+svfloat64_t svuzpq1(svfloat64_t, svfloat64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzpq1_f32)))
+svfloat32_t svuzpq1(svfloat32_t, svfloat32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzpq1_f16)))
+svfloat16_t svuzpq1(svfloat16_t, svfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzpq1_s32)))
+svint32_t svuzpq1(svint32_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzpq1_s64)))
+svint64_t svuzpq1(svint64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzpq1_mf8)))
+svmfloat8_t svuzpq1(svmfloat8_t, svmfloat8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzpq1_s16)))
+svint16_t svuzpq1(svint16_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzpq2_u8)))
+svuint8_t svuzpq2(svuint8_t, svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzpq2_u32)))
+svuint32_t svuzpq2(svuint32_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzpq2_u64)))
+svuint64_t svuzpq2(svuint64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzpq2_u16)))
+svuint16_t svuzpq2(svuint16_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzpq2_bf16)))
+svbfloat16_t svuzpq2(svbfloat16_t, svbfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzpq2_s8)))
+svint8_t svuzpq2(svint8_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzpq2_f64)))
+svfloat64_t svuzpq2(svfloat64_t, svfloat64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzpq2_f32)))
+svfloat32_t svuzpq2(svfloat32_t, svfloat32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzpq2_f16)))
+svfloat16_t svuzpq2(svfloat16_t, svfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzpq2_s32)))
+svint32_t svuzpq2(svint32_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzpq2_s64)))
+svint64_t svuzpq2(svint64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzpq2_mf8)))
+svmfloat8_t svuzpq2(svmfloat8_t, svmfloat8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzpq2_s16)))
+svint16_t svuzpq2(svint16_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzipq1_u8)))
+svuint8_t svzipq1(svuint8_t, svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzipq1_u32)))
+svuint32_t svzipq1(svuint32_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzipq1_u64)))
+svuint64_t svzipq1(svuint64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzipq1_u16)))
+svuint16_t svzipq1(svuint16_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzipq1_bf16)))
+svbfloat16_t svzipq1(svbfloat16_t, svbfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzipq1_s8)))
+svint8_t svzipq1(svint8_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzipq1_f64)))
+svfloat64_t svzipq1(svfloat64_t, svfloat64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzipq1_f32)))
+svfloat32_t svzipq1(svfloat32_t, svfloat32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzipq1_f16)))
+svfloat16_t svzipq1(svfloat16_t, svfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzipq1_s32)))
+svint32_t svzipq1(svint32_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzipq1_s64)))
+svint64_t svzipq1(svint64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzipq1_mf8)))
+svmfloat8_t svzipq1(svmfloat8_t, svmfloat8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzipq1_s16)))
+svint16_t svzipq1(svint16_t, svint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzipq2_u8)))
+svuint8_t svzipq2(svuint8_t, svuint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzipq2_u32)))
+svuint32_t svzipq2(svuint32_t, svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzipq2_u64)))
+svuint64_t svzipq2(svuint64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzipq2_u16)))
+svuint16_t svzipq2(svuint16_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzipq2_bf16)))
+svbfloat16_t svzipq2(svbfloat16_t, svbfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzipq2_s8)))
+svint8_t svzipq2(svint8_t, svint8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzipq2_f64)))
+svfloat64_t svzipq2(svfloat64_t, svfloat64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzipq2_f32)))
+svfloat32_t svzipq2(svfloat32_t, svfloat32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzipq2_f16)))
+svfloat16_t svzipq2(svfloat16_t, svfloat16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzipq2_s32)))
+svint32_t svzipq2(svint32_t, svint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzipq2_s64)))
+svint64_t svzipq2(svint64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzipq2_mf8)))
+svmfloat8_t svzipq2(svmfloat8_t, svmfloat8_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzipq2_s16)))
+svint16_t svzipq2(svint16_t, svint16_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbfmmla_f32)))
 svfloat32_t svbfmmla_f32(svfloat32_t, svbfloat16_t, svbfloat16_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldff1_bf16)))
-svbfloat16_t svldff1_bf16(svbool_t, bfloat16_t const *);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldff1_vnum_bf16)))
-svbfloat16_t svldff1_vnum_bf16(svbool_t, bfloat16_t const *, int64_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnf1_bf16)))
-svbfloat16_t svldnf1_bf16(svbool_t, bfloat16_t const *);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnf1_vnum_bf16)))
-svbfloat16_t svldnf1_vnum_bf16(svbool_t, bfloat16_t const *, int64_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbfmmla_f32)))
 svfloat32_t svbfmmla(svfloat32_t, svbfloat16_t, svbfloat16_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldff1_bf16)))
-svbfloat16_t svldff1(svbool_t, bfloat16_t const *);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldff1_vnum_bf16)))
-svbfloat16_t svldff1_vnum(svbool_t, bfloat16_t const *, int64_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnf1_bf16)))
-svbfloat16_t svldnf1(svbool_t, bfloat16_t const *);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnf1_vnum_bf16)))
-svbfloat16_t svldnf1_vnum(svbool_t, bfloat16_t const *, int64_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtrn1q_bf16)))
-svbfloat16_t svtrn1q_bf16(svbfloat16_t, svbfloat16_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtrn2q_bf16)))
-svbfloat16_t svtrn2q_bf16(svbfloat16_t, svbfloat16_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzp1q_bf16)))
-svbfloat16_t svuzp1q_bf16(svbfloat16_t, svbfloat16_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzp2q_bf16)))
-svbfloat16_t svuzp2q_bf16(svbfloat16_t, svbfloat16_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzip1q_bf16)))
-svbfloat16_t svzip1q_bf16(svbfloat16_t, svbfloat16_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzip2q_bf16)))
-svbfloat16_t svzip2q_bf16(svbfloat16_t, svbfloat16_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtrn1q_bf16)))
-svbfloat16_t svtrn1q(svbfloat16_t, svbfloat16_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtrn2q_bf16)))
-svbfloat16_t svtrn2q(svbfloat16_t, svbfloat16_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzp1q_bf16)))
-svbfloat16_t svuzp1q(svbfloat16_t, svbfloat16_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzp2q_bf16)))
-svbfloat16_t svuzp2q(svbfloat16_t, svbfloat16_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzip1q_bf16)))
-svbfloat16_t svzip1q(svbfloat16_t, svbfloat16_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzip2q_bf16)))
-svbfloat16_t svzip2q(svbfloat16_t, svbfloat16_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbfdot_n_f32)))
 svfloat32_t svbfdot_n_f32(svfloat32_t, svbfloat16_t, bfloat16_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbfdot_f32)))
@@ -6955,26 +8799,6 @@ __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbfmlalt_f32)))
 svfloat32_t svbfmlalt_f32(svfloat32_t, svbfloat16_t, svbfloat16_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbfmlalt_lane_f32)))
 svfloat32_t svbfmlalt_lane_f32(svfloat32_t, svbfloat16_t, svbfloat16_t, uint64_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svclasta_n_bf16)))
-bfloat16_t svclasta_n_bf16(svbool_t, bfloat16_t, svbfloat16_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svclasta_bf16)))
-svbfloat16_t svclasta_bf16(svbool_t, svbfloat16_t, svbfloat16_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svclastb_n_bf16)))
-bfloat16_t svclastb_n_bf16(svbool_t, bfloat16_t, svbfloat16_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svclastb_bf16)))
-svbfloat16_t svclastb_bf16(svbool_t, svbfloat16_t, svbfloat16_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcnt_bf16_m)))
-svuint16_t svcnt_bf16_m(svuint16_t, svbool_t, svbfloat16_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcnt_bf16_x)))
-svuint16_t svcnt_bf16_x(svbool_t, svbfloat16_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcnt_bf16_z)))
-svuint16_t svcnt_bf16_z(svbool_t, svbfloat16_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcreate2_bf16)))
-svbfloat16x2_t svcreate2_bf16(svbfloat16_t, svbfloat16_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcreate3_bf16)))
-svbfloat16x3_t svcreate3_bf16(svbfloat16_t, svbfloat16_t, svbfloat16_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcreate4_bf16)))
-svbfloat16x4_t svcreate4_bf16(svbfloat16_t, svbfloat16_t, svbfloat16_t, svbfloat16_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcvt_bf16_f32_m)))
 svbfloat16_t svcvt_bf16_f32_m(svbfloat16_t, svbool_t, svfloat32_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcvt_bf16_f32_x)))
@@ -6983,112 +8807,6 @@ __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcvt_bf16_f32_z)))
 svbfloat16_t svcvt_bf16_f32_z(svbool_t, svfloat32_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcvtnt_bf16_f32_m)))
 svbfloat16_t svcvtnt_bf16_f32_m(svbfloat16_t, svbool_t, svfloat32_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdup_n_bf16)))
-svbfloat16_t svdup_n_bf16(bfloat16_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdup_n_bf16_m)))
-svbfloat16_t svdup_n_bf16_m(svbfloat16_t, svbool_t, bfloat16_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdup_n_bf16_x)))
-svbfloat16_t svdup_n_bf16_x(svbool_t, bfloat16_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdup_n_bf16_z)))
-svbfloat16_t svdup_n_bf16_z(svbool_t, bfloat16_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdup_lane_bf16)))
-svbfloat16_t svdup_lane_bf16(svbfloat16_t, uint16_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdupq_n_bf16)))
-svbfloat16_t svdupq_n_bf16(bfloat16_t, bfloat16_t, bfloat16_t, bfloat16_t, bfloat16_t, bfloat16_t, bfloat16_t, bfloat16_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdupq_lane_bf16)))
-svbfloat16_t svdupq_lane_bf16(svbfloat16_t, uint64_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svext_bf16)))
-svbfloat16_t svext_bf16(svbfloat16_t, svbfloat16_t, uint64_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svget2_bf16)))
-svbfloat16_t svget2_bf16(svbfloat16x2_t, uint64_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svget3_bf16)))
-svbfloat16_t svget3_bf16(svbfloat16x3_t, uint64_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svget4_bf16)))
-svbfloat16_t svget4_bf16(svbfloat16x4_t, uint64_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svinsr_n_bf16)))
-svbfloat16_t svinsr_n_bf16(svbfloat16_t, bfloat16_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlasta_bf16)))
-bfloat16_t svlasta_bf16(svbool_t, svbfloat16_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlastb_bf16)))
-bfloat16_t svlastb_bf16(svbool_t, svbfloat16_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1_bf16)))
-svbfloat16_t svld1_bf16(svbool_t, bfloat16_t const *);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1_vnum_bf16)))
-svbfloat16_t svld1_vnum_bf16(svbool_t, bfloat16_t const *, int64_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1rq_bf16)))
-svbfloat16_t svld1rq_bf16(svbool_t, bfloat16_t const *);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld2_bf16)))
-svbfloat16x2_t svld2_bf16(svbool_t, bfloat16_t const *);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld2_vnum_bf16)))
-svbfloat16x2_t svld2_vnum_bf16(svbool_t, bfloat16_t const *, int64_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld3_bf16)))
-svbfloat16x3_t svld3_bf16(svbool_t, bfloat16_t const *);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld3_vnum_bf16)))
-svbfloat16x3_t svld3_vnum_bf16(svbool_t, bfloat16_t const *, int64_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld4_bf16)))
-svbfloat16x4_t svld4_bf16(svbool_t, bfloat16_t const *);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld4_vnum_bf16)))
-svbfloat16x4_t svld4_vnum_bf16(svbool_t, bfloat16_t const *, int64_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_bf16)))
-svbfloat16_t svldnt1_bf16(svbool_t, bfloat16_t const *);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_vnum_bf16)))
-svbfloat16_t svldnt1_vnum_bf16(svbool_t, bfloat16_t const *, int64_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlen_bf16)))
-uint64_t svlen_bf16(svbfloat16_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrev_bf16)))
-svbfloat16_t svrev_bf16(svbfloat16_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsel_bf16)))
-svbfloat16_t svsel_bf16(svbool_t, svbfloat16_t, svbfloat16_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svset2_bf16)))
-svbfloat16x2_t svset2_bf16(svbfloat16x2_t, uint64_t, svbfloat16_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svset3_bf16)))
-svbfloat16x3_t svset3_bf16(svbfloat16x3_t, uint64_t, svbfloat16_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svset4_bf16)))
-svbfloat16x4_t svset4_bf16(svbfloat16x4_t, uint64_t, svbfloat16_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsplice_bf16)))
-svbfloat16_t svsplice_bf16(svbool_t, svbfloat16_t, svbfloat16_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1_bf16)))
-void svst1_bf16(svbool_t, bfloat16_t *, svbfloat16_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1_vnum_bf16)))
-void svst1_vnum_bf16(svbool_t, bfloat16_t *, int64_t, svbfloat16_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst2_bf16)))
-void svst2_bf16(svbool_t, bfloat16_t *, svbfloat16x2_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst2_vnum_bf16)))
-void svst2_vnum_bf16(svbool_t, bfloat16_t *, int64_t, svbfloat16x2_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst3_bf16)))
-void svst3_bf16(svbool_t, bfloat16_t *, svbfloat16x3_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst3_vnum_bf16)))
-void svst3_vnum_bf16(svbool_t, bfloat16_t *, int64_t, svbfloat16x3_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst4_bf16)))
-void svst4_bf16(svbool_t, bfloat16_t *, svbfloat16x4_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst4_vnum_bf16)))
-void svst4_vnum_bf16(svbool_t, bfloat16_t *, int64_t, svbfloat16x4_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_bf16)))
-void svstnt1_bf16(svbool_t, bfloat16_t *, svbfloat16_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_vnum_bf16)))
-void svstnt1_vnum_bf16(svbool_t, bfloat16_t *, int64_t, svbfloat16_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtbl_bf16)))
-svbfloat16_t svtbl_bf16(svbfloat16_t, svuint16_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtrn1_bf16)))
-svbfloat16_t svtrn1_bf16(svbfloat16_t, svbfloat16_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtrn2_bf16)))
-svbfloat16_t svtrn2_bf16(svbfloat16_t, svbfloat16_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svundef2_bf16)))
-svbfloat16x2_t svundef2_bf16(void);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svundef3_bf16)))
-svbfloat16x3_t svundef3_bf16(void);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svundef4_bf16)))
-svbfloat16x4_t svundef4_bf16(void);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svundef_bf16)))
-svbfloat16_t svundef_bf16(void);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzp1_bf16)))
-svbfloat16_t svuzp1_bf16(svbfloat16_t, svbfloat16_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzp2_bf16)))
-svbfloat16_t svuzp2_bf16(svbfloat16_t, svbfloat16_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzip1_bf16)))
-svbfloat16_t svzip1_bf16(svbfloat16_t, svbfloat16_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzip2_bf16)))
-svbfloat16_t svzip2_bf16(svbfloat16_t, svbfloat16_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbfdot_n_f32)))
 svfloat32_t svbfdot(svfloat32_t, svbfloat16_t, bfloat16_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbfdot_f32)))
@@ -7107,26 +8825,6 @@ __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbfmlalt_f32)))
 svfloat32_t svbfmlalt(svfloat32_t, svbfloat16_t, svbfloat16_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbfmlalt_lane_f32)))
 svfloat32_t svbfmlalt_lane(svfloat32_t, svbfloat16_t, svbfloat16_t, uint64_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svclasta_n_bf16)))
-bfloat16_t svclasta(svbool_t, bfloat16_t, svbfloat16_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svclasta_bf16)))
-svbfloat16_t svclasta(svbool_t, svbfloat16_t, svbfloat16_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svclastb_n_bf16)))
-bfloat16_t svclastb(svbool_t, bfloat16_t, svbfloat16_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svclastb_bf16)))
-svbfloat16_t svclastb(svbool_t, svbfloat16_t, svbfloat16_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcnt_bf16_m)))
-svuint16_t svcnt_m(svuint16_t, svbool_t, svbfloat16_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcnt_bf16_x)))
-svuint16_t svcnt_x(svbool_t, svbfloat16_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcnt_bf16_z)))
-svuint16_t svcnt_z(svbool_t, svbfloat16_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcreate2_bf16)))
-svbfloat16x2_t svcreate2(svbfloat16_t, svbfloat16_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcreate3_bf16)))
-svbfloat16x3_t svcreate3(svbfloat16_t, svbfloat16_t, svbfloat16_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcreate4_bf16)))
-svbfloat16x4_t svcreate4(svbfloat16_t, svbfloat16_t, svbfloat16_t, svbfloat16_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcvt_bf16_f32_m)))
 svbfloat16_t svcvt_bf16_m(svbfloat16_t, svbool_t, svfloat32_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcvt_bf16_f32_x)))
@@ -7135,104 +8833,6 @@ __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcvt_bf16_f32_z)))
 svbfloat16_t svcvt_bf16_z(svbool_t, svfloat32_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcvtnt_bf16_f32_m)))
 svbfloat16_t svcvtnt_bf16_m(svbfloat16_t, svbool_t, svfloat32_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdup_n_bf16)))
-svbfloat16_t svdup_bf16(bfloat16_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdup_n_bf16_m)))
-svbfloat16_t svdup_bf16_m(svbfloat16_t, svbool_t, bfloat16_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdup_n_bf16_x)))
-svbfloat16_t svdup_bf16_x(svbool_t, bfloat16_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdup_n_bf16_z)))
-svbfloat16_t svdup_bf16_z(svbool_t, bfloat16_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdup_lane_bf16)))
-svbfloat16_t svdup_lane(svbfloat16_t, uint16_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdupq_n_bf16)))
-svbfloat16_t svdupq_bf16(bfloat16_t, bfloat16_t, bfloat16_t, bfloat16_t, bfloat16_t, bfloat16_t, bfloat16_t, bfloat16_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdupq_lane_bf16)))
-svbfloat16_t svdupq_lane(svbfloat16_t, uint64_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svext_bf16)))
-svbfloat16_t svext(svbfloat16_t, svbfloat16_t, uint64_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svget2_bf16)))
-svbfloat16_t svget2(svbfloat16x2_t, uint64_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svget3_bf16)))
-svbfloat16_t svget3(svbfloat16x3_t, uint64_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svget4_bf16)))
-svbfloat16_t svget4(svbfloat16x4_t, uint64_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svinsr_n_bf16)))
-svbfloat16_t svinsr(svbfloat16_t, bfloat16_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlasta_bf16)))
-bfloat16_t svlasta(svbool_t, svbfloat16_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlastb_bf16)))
-bfloat16_t svlastb(svbool_t, svbfloat16_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1_bf16)))
-svbfloat16_t svld1(svbool_t, bfloat16_t const *);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1_vnum_bf16)))
-svbfloat16_t svld1_vnum(svbool_t, bfloat16_t const *, int64_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1rq_bf16)))
-svbfloat16_t svld1rq(svbool_t, bfloat16_t const *);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld2_bf16)))
-svbfloat16x2_t svld2(svbool_t, bfloat16_t const *);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld2_vnum_bf16)))
-svbfloat16x2_t svld2_vnum(svbool_t, bfloat16_t const *, int64_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld3_bf16)))
-svbfloat16x3_t svld3(svbool_t, bfloat16_t const *);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld3_vnum_bf16)))
-svbfloat16x3_t svld3_vnum(svbool_t, bfloat16_t const *, int64_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld4_bf16)))
-svbfloat16x4_t svld4(svbool_t, bfloat16_t const *);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld4_vnum_bf16)))
-svbfloat16x4_t svld4_vnum(svbool_t, bfloat16_t const *, int64_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_bf16)))
-svbfloat16_t svldnt1(svbool_t, bfloat16_t const *);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_vnum_bf16)))
-svbfloat16_t svldnt1_vnum(svbool_t, bfloat16_t const *, int64_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlen_bf16)))
-uint64_t svlen(svbfloat16_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrev_bf16)))
-svbfloat16_t svrev(svbfloat16_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsel_bf16)))
-svbfloat16_t svsel(svbool_t, svbfloat16_t, svbfloat16_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svset2_bf16)))
-svbfloat16x2_t svset2(svbfloat16x2_t, uint64_t, svbfloat16_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svset3_bf16)))
-svbfloat16x3_t svset3(svbfloat16x3_t, uint64_t, svbfloat16_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svset4_bf16)))
-svbfloat16x4_t svset4(svbfloat16x4_t, uint64_t, svbfloat16_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsplice_bf16)))
-svbfloat16_t svsplice(svbool_t, svbfloat16_t, svbfloat16_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1_bf16)))
-void svst1(svbool_t, bfloat16_t *, svbfloat16_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1_vnum_bf16)))
-void svst1_vnum(svbool_t, bfloat16_t *, int64_t, svbfloat16_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst2_bf16)))
-void svst2(svbool_t, bfloat16_t *, svbfloat16x2_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst2_vnum_bf16)))
-void svst2_vnum(svbool_t, bfloat16_t *, int64_t, svbfloat16x2_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst3_bf16)))
-void svst3(svbool_t, bfloat16_t *, svbfloat16x3_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst3_vnum_bf16)))
-void svst3_vnum(svbool_t, bfloat16_t *, int64_t, svbfloat16x3_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst4_bf16)))
-void svst4(svbool_t, bfloat16_t *, svbfloat16x4_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst4_vnum_bf16)))
-void svst4_vnum(svbool_t, bfloat16_t *, int64_t, svbfloat16x4_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_bf16)))
-void svstnt1(svbool_t, bfloat16_t *, svbfloat16_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_vnum_bf16)))
-void svstnt1_vnum(svbool_t, bfloat16_t *, int64_t, svbfloat16_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtbl_bf16)))
-svbfloat16_t svtbl(svbfloat16_t, svuint16_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtrn1_bf16)))
-svbfloat16_t svtrn1(svbfloat16_t, svbfloat16_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtrn2_bf16)))
-svbfloat16_t svtrn2(svbfloat16_t, svbfloat16_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzp1_bf16)))
-svbfloat16_t svuzp1(svbfloat16_t, svbfloat16_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzp2_bf16)))
-svbfloat16_t svuzp2(svbfloat16_t, svbfloat16_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzip1_bf16)))
-svbfloat16_t svzip1(svbfloat16_t, svbfloat16_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzip2_bf16)))
-svbfloat16_t svzip2(svbfloat16_t, svbfloat16_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmmla_f32)))
 svfloat32_t svmmla_f32(svfloat32_t, svfloat32_t, svfloat32_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmmla_f32)))
@@ -7245,6 +8845,8 @@ __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1ro_u64)))
 svuint64_t svld1ro_u64(svbool_t, uint64_t const *);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1ro_u16)))
 svuint16_t svld1ro_u16(svbool_t, uint16_t const *);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1ro_bf16)))
+svbfloat16_t svld1ro_bf16(svbool_t, bfloat16_t const *);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1ro_s8)))
 svint8_t svld1ro_s8(svbool_t, int8_t const *);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1ro_f64)))
@@ -7257,6 +8859,8 @@ __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1ro_s32)))
 svint32_t svld1ro_s32(svbool_t, int32_t const *);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1ro_s64)))
 svint64_t svld1ro_s64(svbool_t, int64_t const *);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1ro_mf8)))
+svmfloat8_t svld1ro_mf8(svbool_t, mfloat8_t const *);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1ro_s16)))
 svint16_t svld1ro_s16(svbool_t, int16_t const *);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmmla_f64)))
@@ -7269,6 +8873,8 @@ __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtrn1q_u64)))
 svuint64_t svtrn1q_u64(svuint64_t, svuint64_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtrn1q_u16)))
 svuint16_t svtrn1q_u16(svuint16_t, svuint16_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtrn1q_bf16)))
+svbfloat16_t svtrn1q_bf16(svbfloat16_t, svbfloat16_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtrn1q_s8)))
 svint8_t svtrn1q_s8(svint8_t, svint8_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtrn1q_f64)))
@@ -7291,6 +8897,8 @@ __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtrn2q_u64)))
 svuint64_t svtrn2q_u64(svuint64_t, svuint64_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtrn2q_u16)))
 svuint16_t svtrn2q_u16(svuint16_t, svuint16_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtrn2q_bf16)))
+svbfloat16_t svtrn2q_bf16(svbfloat16_t, svbfloat16_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtrn2q_s8)))
 svint8_t svtrn2q_s8(svint8_t, svint8_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtrn2q_f64)))
@@ -7313,6 +8921,8 @@ __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzp1q_u64)))
 svuint64_t svuzp1q_u64(svuint64_t, svuint64_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzp1q_u16)))
 svuint16_t svuzp1q_u16(svuint16_t, svuint16_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzp1q_bf16)))
+svbfloat16_t svuzp1q_bf16(svbfloat16_t, svbfloat16_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzp1q_s8)))
 svint8_t svuzp1q_s8(svint8_t, svint8_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzp1q_f64)))
@@ -7335,6 +8945,8 @@ __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzp2q_u64)))
 svuint64_t svuzp2q_u64(svuint64_t, svuint64_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzp2q_u16)))
 svuint16_t svuzp2q_u16(svuint16_t, svuint16_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzp2q_bf16)))
+svbfloat16_t svuzp2q_bf16(svbfloat16_t, svbfloat16_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzp2q_s8)))
 svint8_t svuzp2q_s8(svint8_t, svint8_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzp2q_f64)))
@@ -7357,6 +8969,8 @@ __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzip1q_u64)))
 svuint64_t svzip1q_u64(svuint64_t, svuint64_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzip1q_u16)))
 svuint16_t svzip1q_u16(svuint16_t, svuint16_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzip1q_bf16)))
+svbfloat16_t svzip1q_bf16(svbfloat16_t, svbfloat16_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzip1q_s8)))
 svint8_t svzip1q_s8(svint8_t, svint8_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzip1q_f64)))
@@ -7379,6 +8993,8 @@ __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzip2q_u64)))
 svuint64_t svzip2q_u64(svuint64_t, svuint64_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzip2q_u16)))
 svuint16_t svzip2q_u16(svuint16_t, svuint16_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzip2q_bf16)))
+svbfloat16_t svzip2q_bf16(svbfloat16_t, svbfloat16_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzip2q_s8)))
 svint8_t svzip2q_s8(svint8_t, svint8_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzip2q_f64)))
@@ -7401,6 +9017,8 @@ __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1ro_u64)))
 svuint64_t svld1ro(svbool_t, uint64_t const *);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1ro_u16)))
 svuint16_t svld1ro(svbool_t, uint16_t const *);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1ro_bf16)))
+svbfloat16_t svld1ro(svbool_t, bfloat16_t const *);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1ro_s8)))
 svint8_t svld1ro(svbool_t, int8_t const *);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1ro_f64)))
@@ -7413,6 +9031,8 @@ __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1ro_s32)))
 svint32_t svld1ro(svbool_t, int32_t const *);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1ro_s64)))
 svint64_t svld1ro(svbool_t, int64_t const *);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1ro_mf8)))
+svmfloat8_t svld1ro(svbool_t, mfloat8_t const *);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1ro_s16)))
 svint16_t svld1ro(svbool_t, int16_t const *);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmmla_f64)))
@@ -7425,6 +9045,8 @@ __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtrn1q_u64)))
 svuint64_t svtrn1q(svuint64_t, svuint64_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtrn1q_u16)))
 svuint16_t svtrn1q(svuint16_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtrn1q_bf16)))
+svbfloat16_t svtrn1q(svbfloat16_t, svbfloat16_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtrn1q_s8)))
 svint8_t svtrn1q(svint8_t, svint8_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtrn1q_f64)))
@@ -7447,6 +9069,8 @@ __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtrn2q_u64)))
 svuint64_t svtrn2q(svuint64_t, svuint64_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtrn2q_u16)))
 svuint16_t svtrn2q(svuint16_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtrn2q_bf16)))
+svbfloat16_t svtrn2q(svbfloat16_t, svbfloat16_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtrn2q_s8)))
 svint8_t svtrn2q(svint8_t, svint8_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtrn2q_f64)))
@@ -7469,6 +9093,8 @@ __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzp1q_u64)))
 svuint64_t svuzp1q(svuint64_t, svuint64_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzp1q_u16)))
 svuint16_t svuzp1q(svuint16_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzp1q_bf16)))
+svbfloat16_t svuzp1q(svbfloat16_t, svbfloat16_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzp1q_s8)))
 svint8_t svuzp1q(svint8_t, svint8_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzp1q_f64)))
@@ -7491,6 +9117,8 @@ __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzp2q_u64)))
 svuint64_t svuzp2q(svuint64_t, svuint64_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzp2q_u16)))
 svuint16_t svuzp2q(svuint16_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzp2q_bf16)))
+svbfloat16_t svuzp2q(svbfloat16_t, svbfloat16_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzp2q_s8)))
 svint8_t svuzp2q(svint8_t, svint8_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzp2q_f64)))
@@ -7513,6 +9141,8 @@ __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzip1q_u64)))
 svuint64_t svzip1q(svuint64_t, svuint64_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzip1q_u16)))
 svuint16_t svzip1q(svuint16_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzip1q_bf16)))
+svbfloat16_t svzip1q(svbfloat16_t, svbfloat16_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzip1q_s8)))
 svint8_t svzip1q(svint8_t, svint8_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzip1q_f64)))
@@ -7535,6 +9165,8 @@ __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzip2q_u64)))
 svuint64_t svzip2q(svuint64_t, svuint64_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzip2q_u16)))
 svuint16_t svzip2q(svuint16_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzip2q_bf16)))
+svbfloat16_t svzip2q(svbfloat16_t, svbfloat16_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzip2q_s8)))
 svint8_t svzip2q(svint8_t, svint8_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzip2q_f64)))
@@ -7549,10 +9181,6 @@ __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzip2q_s64)))
 svint64_t svzip2q(svint64_t, svint64_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzip2q_s16)))
 svint16_t svzip2q(svint16_t, svint16_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1ro_bf16)))
-svbfloat16_t svld1ro_bf16(svbool_t, bfloat16_t const *);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1ro_bf16)))
-svbfloat16_t svld1ro(svbool_t, bfloat16_t const *);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmmla_s32)))
 svint32_t svmmla_s32(svint32_t, svint8_t, svint8_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmmla_u32)))
@@ -7589,6 +9217,14 @@ __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svusdot_s32)))
 svint32_t svusdot(svint32_t, svuint8_t, svint8_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svusdot_lane_s32)))
 svint32_t svusdot_lane(svint32_t, svuint8_t, svint8_t, uint64_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrax1_u64)))
+svuint64_t svrax1_u64(svuint64_t, svuint64_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrax1_s64)))
+svint64_t svrax1_s64(svint64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrax1_u64)))
+svuint64_t svrax1(svuint64_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrax1_s64)))
+svint64_t svrax1(svint64_t, svint64_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhistcnt_u32_z)))
 svuint32_t svhistcnt_u32_z(svbool_t, svuint32_t, svuint32_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svhistcnt_u64_z)))
@@ -8509,22 +10145,6 @@ __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1w_scatter_u6
 void svstnt1w_scatter_offset(svbool_t, int32_t *, svuint64_t, svint64_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1w_scatter_u64offset_u64)))
 void svstnt1w_scatter_offset(svbool_t, uint32_t *, svuint64_t, svuint64_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtbl2_bf16)))
-svbfloat16_t svtbl2_bf16(svbfloat16x2_t, svuint16_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtbx_bf16)))
-svbfloat16_t svtbx_bf16(svbfloat16_t, svbfloat16_t, svuint16_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilerw_bf16)))
-svbool_t svwhilerw_bf16(bfloat16_t const *, bfloat16_t const *);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilewr_bf16)))
-svbool_t svwhilewr_bf16(bfloat16_t const *, bfloat16_t const *);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtbl2_bf16)))
-svbfloat16_t svtbl2(svbfloat16x2_t, svuint16_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtbx_bf16)))
-svbfloat16_t svtbx(svbfloat16_t, svbfloat16_t, svuint16_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilerw_bf16)))
-svbool_t svwhilerw(bfloat16_t const *, bfloat16_t const *);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilewr_bf16)))
-svbool_t svwhilewr(bfloat16_t const *, bfloat16_t const *);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svamax_n_f64_m)))
 svfloat64_t svamax_n_f64_m(svbool_t, svfloat64_t, float64_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svamax_n_f32_m)))
@@ -8813,24 +10433,14 @@ __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcvtnb_mf8_f32_x2_
 svmfloat8_t svcvtnb_mf8_fpm(svfloat32x2_t, fpm_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcvtnt_mf8_f32_x2_fpm)))
 svmfloat8_t svcvtnt_mf8_fpm(svmfloat8_t, svfloat32x2_t, fpm_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svluti2_lane_bf16)))
-svbfloat16_t svluti2_lane_bf16(svbfloat16_t, svuint8_t, uint64_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svluti4_lane_bf16)))
-svbfloat16_t svluti4_lane_bf16(svbfloat16_t, svuint8_t, uint64_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svluti4_lane_bf16_x2)))
-svbfloat16_t svluti4_lane_bf16_x2(svbfloat16x2_t, svuint8_t, uint64_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svluti2_lane_bf16)))
-svbfloat16_t svluti2_lane(svbfloat16_t, svuint8_t, uint64_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svluti4_lane_bf16)))
-svbfloat16_t svluti4_lane(svbfloat16_t, svuint8_t, uint64_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svluti4_lane_bf16_x2)))
-svbfloat16_t svluti4_lane(svbfloat16x2_t, svuint8_t, uint64_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svluti2_lane_u8)))
 svuint8_t svluti2_lane_u8(svuint8_t, svuint8_t, uint64_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svluti2_lane_s8)))
 svint8_t svluti2_lane_s8(svint8_t, svuint8_t, uint64_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svluti2_lane_u16)))
 svuint16_t svluti2_lane_u16(svuint16_t, svuint8_t, uint64_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svluti2_lane_bf16)))
+svbfloat16_t svluti2_lane_bf16(svbfloat16_t, svuint8_t, uint64_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svluti2_lane_f16)))
 svfloat16_t svluti2_lane_f16(svfloat16_t, svuint8_t, uint64_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svluti2_lane_s16)))
@@ -8841,12 +10451,16 @@ __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svluti4_lane_s8)))
 svint8_t svluti4_lane_s8(svint8_t, svuint8_t, uint64_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svluti4_lane_u16)))
 svuint16_t svluti4_lane_u16(svuint16_t, svuint8_t, uint64_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svluti4_lane_bf16)))
+svbfloat16_t svluti4_lane_bf16(svbfloat16_t, svuint8_t, uint64_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svluti4_lane_f16)))
 svfloat16_t svluti4_lane_f16(svfloat16_t, svuint8_t, uint64_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svluti4_lane_s16)))
 svint16_t svluti4_lane_s16(svint16_t, svuint8_t, uint64_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svluti4_lane_u16_x2)))
 svuint16_t svluti4_lane_u16_x2(svuint16x2_t, svuint8_t, uint64_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svluti4_lane_bf16_x2)))
+svbfloat16_t svluti4_lane_bf16_x2(svbfloat16x2_t, svuint8_t, uint64_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svluti4_lane_f16_x2)))
 svfloat16_t svluti4_lane_f16_x2(svfloat16x2_t, svuint8_t, uint64_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svluti4_lane_s16_x2)))
@@ -8857,6 +10471,8 @@ __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svluti2_lane_s8)))
 svint8_t svluti2_lane(svint8_t, svuint8_t, uint64_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svluti2_lane_u16)))
 svuint16_t svluti2_lane(svuint16_t, svuint8_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svluti2_lane_bf16)))
+svbfloat16_t svluti2_lane(svbfloat16_t, svuint8_t, uint64_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svluti2_lane_f16)))
 svfloat16_t svluti2_lane(svfloat16_t, svuint8_t, uint64_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svluti2_lane_s16)))
@@ -8867,12 +10483,16 @@ __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svluti4_lane_s8)))
 svint8_t svluti4_lane(svint8_t, svuint8_t, uint64_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svluti4_lane_u16)))
 svuint16_t svluti4_lane(svuint16_t, svuint8_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svluti4_lane_bf16)))
+svbfloat16_t svluti4_lane(svbfloat16_t, svuint8_t, uint64_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svluti4_lane_f16)))
 svfloat16_t svluti4_lane(svfloat16_t, svuint8_t, uint64_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svluti4_lane_s16)))
 svint16_t svluti4_lane(svint16_t, svuint8_t, uint64_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svluti4_lane_u16_x2)))
 svuint16_t svluti4_lane(svuint16x2_t, svuint8_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svluti4_lane_bf16_x2)))
+svbfloat16_t svluti4_lane(svbfloat16x2_t, svuint8_t, uint64_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svluti4_lane_f16_x2)))
 svfloat16_t svluti4_lane(svfloat16x2_t, svuint8_t, uint64_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svluti4_lane_s16_x2)))
@@ -9237,14 +10857,6 @@ __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbgrp_u64)))
 svuint64_t svbgrp(svuint64_t, svuint64_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbgrp_u16)))
 svuint16_t svbgrp(svuint16_t, svuint16_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrax1_u64)))
-svuint64_t svrax1_u64(svuint64_t, svuint64_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrax1_s64)))
-svint64_t svrax1_s64(svint64_t, svint64_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrax1_u64)))
-svuint64_t svrax1(svuint64_t, svuint64_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrax1_s64)))
-svint64_t svrax1(svint64_t, svint64_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsm4e_u32)))
 svuint32_t svsm4e_u32(svuint32_t, svuint32_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsm4ekey_u32)))
@@ -9253,86 +10865,6 @@ __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsm4e_u32)))
 svuint32_t svsm4e(svuint32_t, svuint32_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsm4ekey_u32)))
 svuint32_t svsm4ekey(svuint32_t, svuint32_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddqv_u8)))
-uint8x16_t svaddqv_u8(svbool_t, svuint8_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddqv_u32)))
-uint32x4_t svaddqv_u32(svbool_t, svuint32_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddqv_u64)))
-uint64x2_t svaddqv_u64(svbool_t, svuint64_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddqv_u16)))
-uint16x8_t svaddqv_u16(svbool_t, svuint16_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddqv_s8)))
-int8x16_t svaddqv_s8(svbool_t, svint8_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddqv_s32)))
-int32x4_t svaddqv_s32(svbool_t, svint32_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddqv_s64)))
-int64x2_t svaddqv_s64(svbool_t, svint64_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddqv_s16)))
-int16x8_t svaddqv_s16(svbool_t, svint16_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddqv_f64)))
-float64x2_t svaddqv_f64(svbool_t, svfloat64_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddqv_f32)))
-float32x4_t svaddqv_f32(svbool_t, svfloat32_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddqv_f16)))
-float16x8_t svaddqv_f16(svbool_t, svfloat16_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svandqv_u8)))
-uint8x16_t svandqv_u8(svbool_t, svuint8_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svandqv_u32)))
-uint32x4_t svandqv_u32(svbool_t, svuint32_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svandqv_u64)))
-uint64x2_t svandqv_u64(svbool_t, svuint64_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svandqv_u16)))
-uint16x8_t svandqv_u16(svbool_t, svuint16_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svandqv_s8)))
-int8x16_t svandqv_s8(svbool_t, svint8_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svandqv_s32)))
-int32x4_t svandqv_s32(svbool_t, svint32_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svandqv_s64)))
-int64x2_t svandqv_s64(svbool_t, svint64_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svandqv_s16)))
-int16x8_t svandqv_s16(svbool_t, svint16_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveorqv_u8)))
-uint8x16_t sveorqv_u8(svbool_t, svuint8_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveorqv_u32)))
-uint32x4_t sveorqv_u32(svbool_t, svuint32_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveorqv_u64)))
-uint64x2_t sveorqv_u64(svbool_t, svuint64_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveorqv_u16)))
-uint16x8_t sveorqv_u16(svbool_t, svuint16_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveorqv_s8)))
-int8x16_t sveorqv_s8(svbool_t, svint8_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveorqv_s32)))
-int32x4_t sveorqv_s32(svbool_t, svint32_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveorqv_s64)))
-int64x2_t sveorqv_s64(svbool_t, svint64_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveorqv_s16)))
-int16x8_t sveorqv_s16(svbool_t, svint16_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svextq_u8)))
-svuint8_t svextq_u8(svuint8_t, svuint8_t, int32_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svextq_u32)))
-svuint32_t svextq_u32(svuint32_t, svuint32_t, int32_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svextq_u64)))
-svuint64_t svextq_u64(svuint64_t, svuint64_t, int32_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svextq_u16)))
-svuint16_t svextq_u16(svuint16_t, svuint16_t, int32_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svextq_bf16)))
-svbfloat16_t svextq_bf16(svbfloat16_t, svbfloat16_t, int32_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svextq_s8)))
-svint8_t svextq_s8(svint8_t, svint8_t, int32_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svextq_f64)))
-svfloat64_t svextq_f64(svfloat64_t, svfloat64_t, int32_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svextq_f32)))
-svfloat32_t svextq_f32(svfloat32_t, svfloat32_t, int32_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svextq_f16)))
-svfloat16_t svextq_f16(svfloat16_t, svfloat16_t, int32_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svextq_s32)))
-svint32_t svextq_s32(svint32_t, svint32_t, int32_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svextq_s64)))
-svint64_t svextq_s64(svint64_t, svint64_t, int32_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svextq_mf8)))
-svmfloat8_t svextq_mf8(svmfloat8_t, svmfloat8_t, int32_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svextq_s16)))
-svint16_t svextq_s16(svint16_t, svint16_t, int32_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1q_gather_u64base_index_u32)))
 svuint32_t svld1q_gather_u64base_index_u32(svbool_t, svuint64_t, int64_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1q_gather_u64base_index_u64)))
@@ -9375,6 +10907,8 @@ __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1q_gather_u64bas
 svint32_t svld1q_gather_u64base_offset_s32(svbool_t, svuint64_t, int64_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1q_gather_u64base_offset_s64)))
 svint64_t svld1q_gather_u64base_offset_s64(svbool_t, svuint64_t, int64_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1q_gather_u64base_offset_mf8)))
+svmfloat8_t svld1q_gather_u64base_offset_mf8(svbool_t, svuint64_t, int64_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1q_gather_u64base_offset_s16)))
 svint16_t svld1q_gather_u64base_offset_s16(svbool_t, svuint64_t, int64_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1q_gather_u64base_u8)))
@@ -9399,6 +10933,8 @@ __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1q_gather_u64bas
 svint32_t svld1q_gather_u64base_s32(svbool_t, svuint64_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1q_gather_u64base_s64)))
 svint64_t svld1q_gather_u64base_s64(svbool_t, svuint64_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1q_gather_u64base_mf8)))
+svmfloat8_t svld1q_gather_u64base_mf8(svbool_t, svuint64_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1q_gather_u64base_s16)))
 svint16_t svld1q_gather_u64base_s16(svbool_t, svuint64_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1q_gather_u64index_u32)))
@@ -9443,6 +10979,8 @@ __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1q_gather_u64off
 svint32_t svld1q_gather_u64offset_s32(svbool_t, int32_t const *, svuint64_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1q_gather_u64offset_s64)))
 svint64_t svld1q_gather_u64offset_s64(svbool_t, int64_t const *, svuint64_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1q_gather_u64offset_mf8)))
+svmfloat8_t svld1q_gather_u64offset_mf8(svbool_t, mfloat8_t const *, svuint64_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1q_gather_u64offset_s16)))
 svint16_t svld1q_gather_u64offset_s16(svbool_t, int16_t const *, svuint64_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1udq_u64)))
@@ -9469,282 +11007,6 @@ __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1uwq_vnum_f32)))
 svfloat32_t svld1uwq_vnum_f32(svbool_t, float32_t const *, int64_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1uwq_vnum_s32)))
 svint32_t svld1uwq_vnum_s32(svbool_t, int32_t const *, int64_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld2q_u8)))
-svuint8x2_t svld2q_u8(svbool_t, uint8_t const *);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld2q_u32)))
-svuint32x2_t svld2q_u32(svbool_t, uint32_t const *);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld2q_u64)))
-svuint64x2_t svld2q_u64(svbool_t, uint64_t const *);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld2q_u16)))
-svuint16x2_t svld2q_u16(svbool_t, uint16_t const *);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld2q_s8)))
-svint8x2_t svld2q_s8(svbool_t, int8_t const *);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld2q_f64)))
-svfloat64x2_t svld2q_f64(svbool_t, float64_t const *);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld2q_f32)))
-svfloat32x2_t svld2q_f32(svbool_t, float32_t const *);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld2q_f16)))
-svfloat16x2_t svld2q_f16(svbool_t, float16_t const *);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld2q_s32)))
-svint32x2_t svld2q_s32(svbool_t, int32_t const *);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld2q_s64)))
-svint64x2_t svld2q_s64(svbool_t, int64_t const *);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld2q_s16)))
-svint16x2_t svld2q_s16(svbool_t, int16_t const *);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld2q_bf16)))
-svbfloat16x2_t svld2q_bf16(svbool_t, bfloat16_t const *);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld2q_vnum_u8)))
-svuint8x2_t svld2q_vnum_u8(svbool_t, uint8_t const *, int64_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld2q_vnum_u32)))
-svuint32x2_t svld2q_vnum_u32(svbool_t, uint32_t const *, int64_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld2q_vnum_u64)))
-svuint64x2_t svld2q_vnum_u64(svbool_t, uint64_t const *, int64_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld2q_vnum_u16)))
-svuint16x2_t svld2q_vnum_u16(svbool_t, uint16_t const *, int64_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld2q_vnum_s8)))
-svint8x2_t svld2q_vnum_s8(svbool_t, int8_t const *, int64_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld2q_vnum_f64)))
-svfloat64x2_t svld2q_vnum_f64(svbool_t, float64_t const *, int64_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld2q_vnum_f32)))
-svfloat32x2_t svld2q_vnum_f32(svbool_t, float32_t const *, int64_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld2q_vnum_f16)))
-svfloat16x2_t svld2q_vnum_f16(svbool_t, float16_t const *, int64_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld2q_vnum_s32)))
-svint32x2_t svld2q_vnum_s32(svbool_t, int32_t const *, int64_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld2q_vnum_s64)))
-svint64x2_t svld2q_vnum_s64(svbool_t, int64_t const *, int64_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld2q_vnum_s16)))
-svint16x2_t svld2q_vnum_s16(svbool_t, int16_t const *, int64_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld2q_vnum_bf16)))
-svbfloat16x2_t svld2q_vnum_bf16(svbool_t, bfloat16_t const *, int64_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld3q_u8)))
-svuint8x3_t svld3q_u8(svbool_t, uint8_t const *);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld3q_u32)))
-svuint32x3_t svld3q_u32(svbool_t, uint32_t const *);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld3q_u64)))
-svuint64x3_t svld3q_u64(svbool_t, uint64_t const *);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld3q_u16)))
-svuint16x3_t svld3q_u16(svbool_t, uint16_t const *);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld3q_s8)))
-svint8x3_t svld3q_s8(svbool_t, int8_t const *);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld3q_f64)))
-svfloat64x3_t svld3q_f64(svbool_t, float64_t const *);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld3q_f32)))
-svfloat32x3_t svld3q_f32(svbool_t, float32_t const *);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld3q_f16)))
-svfloat16x3_t svld3q_f16(svbool_t, float16_t const *);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld3q_s32)))
-svint32x3_t svld3q_s32(svbool_t, int32_t const *);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld3q_s64)))
-svint64x3_t svld3q_s64(svbool_t, int64_t const *);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld3q_s16)))
-svint16x3_t svld3q_s16(svbool_t, int16_t const *);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld3q_bf16)))
-svbfloat16x3_t svld3q_bf16(svbool_t, bfloat16_t const *);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld3q_vnum_u8)))
-svuint8x3_t svld3q_vnum_u8(svbool_t, uint8_t const *, int64_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld3q_vnum_u32)))
-svuint32x3_t svld3q_vnum_u32(svbool_t, uint32_t const *, int64_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld3q_vnum_u64)))
-svuint64x3_t svld3q_vnum_u64(svbool_t, uint64_t const *, int64_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld3q_vnum_u16)))
-svuint16x3_t svld3q_vnum_u16(svbool_t, uint16_t const *, int64_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld3q_vnum_s8)))
-svint8x3_t svld3q_vnum_s8(svbool_t, int8_t const *, int64_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld3q_vnum_f64)))
-svfloat64x3_t svld3q_vnum_f64(svbool_t, float64_t const *, int64_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld3q_vnum_f32)))
-svfloat32x3_t svld3q_vnum_f32(svbool_t, float32_t const *, int64_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld3q_vnum_f16)))
-svfloat16x3_t svld3q_vnum_f16(svbool_t, float16_t const *, int64_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld3q_vnum_s32)))
-svint32x3_t svld3q_vnum_s32(svbool_t, int32_t const *, int64_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld3q_vnum_s64)))
-svint64x3_t svld3q_vnum_s64(svbool_t, int64_t const *, int64_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld3q_vnum_s16)))
-svint16x3_t svld3q_vnum_s16(svbool_t, int16_t const *, int64_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld3q_vnum_bf16)))
-svbfloat16x3_t svld3q_vnum_bf16(svbool_t, bfloat16_t const *, int64_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld4q_u8)))
-svuint8x4_t svld4q_u8(svbool_t, uint8_t const *);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld4q_u32)))
-svuint32x4_t svld4q_u32(svbool_t, uint32_t const *);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld4q_u64)))
-svuint64x4_t svld4q_u64(svbool_t, uint64_t const *);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld4q_u16)))
-svuint16x4_t svld4q_u16(svbool_t, uint16_t const *);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld4q_s8)))
-svint8x4_t svld4q_s8(svbool_t, int8_t const *);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld4q_f64)))
-svfloat64x4_t svld4q_f64(svbool_t, float64_t const *);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld4q_f32)))
-svfloat32x4_t svld4q_f32(svbool_t, float32_t const *);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld4q_f16)))
-svfloat16x4_t svld4q_f16(svbool_t, float16_t const *);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld4q_s32)))
-svint32x4_t svld4q_s32(svbool_t, int32_t const *);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld4q_s64)))
-svint64x4_t svld4q_s64(svbool_t, int64_t const *);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld4q_s16)))
-svint16x4_t svld4q_s16(svbool_t, int16_t const *);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld4q_bf16)))
-svbfloat16x4_t svld4q_bf16(svbool_t, bfloat16_t const *);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld4q_vnum_u8)))
-svuint8x4_t svld4q_vnum_u8(svbool_t, uint8_t const *, int64_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld4q_vnum_u32)))
-svuint32x4_t svld4q_vnum_u32(svbool_t, uint32_t const *, int64_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld4q_vnum_u64)))
-svuint64x4_t svld4q_vnum_u64(svbool_t, uint64_t const *, int64_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld4q_vnum_u16)))
-svuint16x4_t svld4q_vnum_u16(svbool_t, uint16_t const *, int64_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld4q_vnum_s8)))
-svint8x4_t svld4q_vnum_s8(svbool_t, int8_t const *, int64_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld4q_vnum_f64)))
-svfloat64x4_t svld4q_vnum_f64(svbool_t, float64_t const *, int64_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld4q_vnum_f32)))
-svfloat32x4_t svld4q_vnum_f32(svbool_t, float32_t const *, int64_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld4q_vnum_f16)))
-svfloat16x4_t svld4q_vnum_f16(svbool_t, float16_t const *, int64_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld4q_vnum_s32)))
-svint32x4_t svld4q_vnum_s32(svbool_t, int32_t const *, int64_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld4q_vnum_s64)))
-svint64x4_t svld4q_vnum_s64(svbool_t, int64_t const *, int64_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld4q_vnum_s16)))
-svint16x4_t svld4q_vnum_s16(svbool_t, int16_t const *, int64_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld4q_vnum_bf16)))
-svbfloat16x4_t svld4q_vnum_bf16(svbool_t, bfloat16_t const *, int64_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmaxnmqv_f64)))
-float64x2_t svmaxnmqv_f64(svbool_t, svfloat64_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmaxnmqv_f32)))
-float32x4_t svmaxnmqv_f32(svbool_t, svfloat32_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmaxnmqv_f16)))
-float16x8_t svmaxnmqv_f16(svbool_t, svfloat16_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmaxqv_f64)))
-float64x2_t svmaxqv_f64(svbool_t, svfloat64_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmaxqv_f32)))
-float32x4_t svmaxqv_f32(svbool_t, svfloat32_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmaxqv_f16)))
-float16x8_t svmaxqv_f16(svbool_t, svfloat16_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmaxqv_s8)))
-int8x16_t svmaxqv_s8(svbool_t, svint8_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmaxqv_s32)))
-int32x4_t svmaxqv_s32(svbool_t, svint32_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmaxqv_s64)))
-int64x2_t svmaxqv_s64(svbool_t, svint64_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmaxqv_s16)))
-int16x8_t svmaxqv_s16(svbool_t, svint16_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmaxqv_u8)))
-uint8x16_t svmaxqv_u8(svbool_t, svuint8_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmaxqv_u32)))
-uint32x4_t svmaxqv_u32(svbool_t, svuint32_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmaxqv_u64)))
-uint64x2_t svmaxqv_u64(svbool_t, svuint64_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmaxqv_u16)))
-uint16x8_t svmaxqv_u16(svbool_t, svuint16_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svminnmqv_f64)))
-float64x2_t svminnmqv_f64(svbool_t, svfloat64_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svminnmqv_f32)))
-float32x4_t svminnmqv_f32(svbool_t, svfloat32_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svminnmqv_f16)))
-float16x8_t svminnmqv_f16(svbool_t, svfloat16_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svminqv_f64)))
-float64x2_t svminqv_f64(svbool_t, svfloat64_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svminqv_f32)))
-float32x4_t svminqv_f32(svbool_t, svfloat32_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svminqv_f16)))
-float16x8_t svminqv_f16(svbool_t, svfloat16_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svminqv_s8)))
-int8x16_t svminqv_s8(svbool_t, svint8_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svminqv_s32)))
-int32x4_t svminqv_s32(svbool_t, svint32_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svminqv_s64)))
-int64x2_t svminqv_s64(svbool_t, svint64_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svminqv_s16)))
-int16x8_t svminqv_s16(svbool_t, svint16_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svminqv_u8)))
-uint8x16_t svminqv_u8(svbool_t, svuint8_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svminqv_u32)))
-uint32x4_t svminqv_u32(svbool_t, svuint32_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svminqv_u64)))
-uint64x2_t svminqv_u64(svbool_t, svuint64_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svminqv_u16)))
-uint16x8_t svminqv_u16(svbool_t, svuint16_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svorqv_u8)))
-uint8x16_t svorqv_u8(svbool_t, svuint8_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svorqv_u32)))
-uint32x4_t svorqv_u32(svbool_t, svuint32_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svorqv_u64)))
-uint64x2_t svorqv_u64(svbool_t, svuint64_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svorqv_u16)))
-uint16x8_t svorqv_u16(svbool_t, svuint16_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svorqv_s8)))
-int8x16_t svorqv_s8(svbool_t, svint8_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svorqv_s32)))
-int32x4_t svorqv_s32(svbool_t, svint32_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svorqv_s64)))
-int64x2_t svorqv_s64(svbool_t, svint64_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svorqv_s16)))
-int16x8_t svorqv_s16(svbool_t, svint16_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svpmov_u8)))
-svbool_t svpmov_u8(svuint8_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svpmov_s8)))
-svbool_t svpmov_s8(svint8_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svpmov_u64)))
-svbool_t svpmov_u64(svuint64_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svpmov_s64)))
-svbool_t svpmov_s64(svint64_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svpmov_u16)))
-svbool_t svpmov_u16(svuint16_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svpmov_s16)))
-svbool_t svpmov_s16(svint16_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svpmov_u32)))
-svbool_t svpmov_u32(svuint32_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svpmov_s32)))
-svbool_t svpmov_s32(svint32_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svpmov_lane_u8)))
-svbool_t svpmov_lane_u8(svuint8_t, uint64_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svpmov_lane_s8)))
-svbool_t svpmov_lane_s8(svint8_t, uint64_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svpmov_lane_u64)))
-svbool_t svpmov_lane_u64(svuint64_t, uint64_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svpmov_lane_s64)))
-svbool_t svpmov_lane_s64(svint64_t, uint64_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svpmov_lane_u16)))
-svbool_t svpmov_lane_u16(svuint16_t, uint64_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svpmov_lane_s16)))
-svbool_t svpmov_lane_s16(svint16_t, uint64_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svpmov_lane_u32)))
-svbool_t svpmov_lane_u32(svuint32_t, uint64_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svpmov_lane_s32)))
-svbool_t svpmov_lane_s32(svint32_t, uint64_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svpmov_lane_u64_m)))
-svuint64_t svpmov_lane_u64_m(svuint64_t, svbool_t, uint64_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svpmov_lane_s64_m)))
-svint64_t svpmov_lane_s64_m(svint64_t, svbool_t, uint64_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svpmov_lane_u16_m)))
-svuint16_t svpmov_lane_u16_m(svuint16_t, svbool_t, uint64_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svpmov_lane_s16_m)))
-svint16_t svpmov_lane_s16_m(svint16_t, svbool_t, uint64_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svpmov_lane_u32_m)))
-svuint32_t svpmov_lane_u32_m(svuint32_t, svbool_t, uint64_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svpmov_lane_s32_m)))
-svint32_t svpmov_lane_s32_m(svint32_t, svbool_t, uint64_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svpmov_u8_z)))
-svuint8_t svpmov_u8_z(svbool_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svpmov_s8_z)))
-svint8_t svpmov_s8_z(svbool_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svpmov_u64_z)))
-svuint64_t svpmov_u64_z(svbool_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svpmov_s64_z)))
-svint64_t svpmov_s64_z(svbool_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svpmov_u16_z)))
-svuint16_t svpmov_u16_z(svbool_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svpmov_s16_z)))
-svint16_t svpmov_s16_z(svbool_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svpmov_u32_z)))
-svuint32_t svpmov_u32_z(svbool_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svpmov_s32_z)))
-svint32_t svpmov_s32_z(svbool_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1dq_u64)))
 void svst1dq_u64(svbool_t, uint64_t *, svuint64_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1dq_f64)))
@@ -9779,6 +11041,8 @@ __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1q_scatter_u64ba
 void svst1q_scatter_u64base_s32(svbool_t, svuint64_t, svint32_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1q_scatter_u64base_s64)))
 void svst1q_scatter_u64base_s64(svbool_t, svuint64_t, svint64_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1q_scatter_u64base_mf8)))
+void svst1q_scatter_u64base_mf8(svbool_t, svuint64_t, svmfloat8_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1q_scatter_u64base_s16)))
 void svst1q_scatter_u64base_s16(svbool_t, svuint64_t, svint16_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1q_scatter_u64base_index_u32)))
@@ -9823,6 +11087,8 @@ __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1q_scatter_u64ba
 void svst1q_scatter_u64base_offset_s32(svbool_t, svuint64_t, int64_t, svint32_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1q_scatter_u64base_offset_s64)))
 void svst1q_scatter_u64base_offset_s64(svbool_t, svuint64_t, int64_t, svint64_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1q_scatter_u64base_offset_mf8)))
+void svst1q_scatter_u64base_offset_mf8(svbool_t, svuint64_t, int64_t, svmfloat8_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1q_scatter_u64base_offset_s16)))
 void svst1q_scatter_u64base_offset_s16(svbool_t, svuint64_t, int64_t, svint16_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1q_scatter_s64index_u32)))
@@ -9887,6 +11153,8 @@ __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1q_scatter_s64of
 void svst1q_scatter_s64offset_s32(svbool_t, int32_t *, svint64_t, svint32_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1q_scatter_s64offset_s64)))
 void svst1q_scatter_s64offset_s64(svbool_t, int64_t *, svint64_t, svint64_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1q_scatter_s64offset_mf8)))
+void svst1q_scatter_s64offset_mf8(svbool_t, mfloat8_t *, svint64_t, svmfloat8_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1q_scatter_s64offset_s16)))
 void svst1q_scatter_s64offset_s16(svbool_t, int16_t *, svint64_t, svint16_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1q_scatter_u64offset_u8)))
@@ -9911,6 +11179,8 @@ __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1q_scatter_u64of
 void svst1q_scatter_u64offset_s32(svbool_t, int32_t *, svuint64_t, svint32_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1q_scatter_u64offset_s64)))
 void svst1q_scatter_u64offset_s64(svbool_t, int64_t *, svuint64_t, svint64_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1q_scatter_u64offset_mf8)))
+void svst1q_scatter_u64offset_mf8(svbool_t, mfloat8_t *, svuint64_t, svmfloat8_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1q_scatter_u64offset_s16)))
 void svst1q_scatter_u64offset_s16(svbool_t, int16_t *, svuint64_t, svint16_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1wq_u32)))
@@ -9925,386 +11195,6 @@ __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1wq_vnum_f32)))
 void svst1wq_vnum_f32(svbool_t, float32_t *, int64_t, svfloat32_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1wq_vnum_s32)))
 void svst1wq_vnum_s32(svbool_t, int32_t *, int64_t, svint32_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst2q_u8)))
-void svst2q_u8(svbool_t, uint8_t const *, svuint8x2_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst2q_u32)))
-void svst2q_u32(svbool_t, uint32_t const *, svuint32x2_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst2q_u64)))
-void svst2q_u64(svbool_t, uint64_t const *, svuint64x2_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst2q_u16)))
-void svst2q_u16(svbool_t, uint16_t const *, svuint16x2_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst2q_s8)))
-void svst2q_s8(svbool_t, int8_t const *, svint8x2_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst2q_f64)))
-void svst2q_f64(svbool_t, float64_t const *, svfloat64x2_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst2q_f32)))
-void svst2q_f32(svbool_t, float32_t const *, svfloat32x2_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst2q_f16)))
-void svst2q_f16(svbool_t, float16_t const *, svfloat16x2_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst2q_s32)))
-void svst2q_s32(svbool_t, int32_t const *, svint32x2_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst2q_s64)))
-void svst2q_s64(svbool_t, int64_t const *, svint64x2_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst2q_s16)))
-void svst2q_s16(svbool_t, int16_t const *, svint16x2_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst2q_bf16)))
-void svst2q_bf16(svbool_t, bfloat16_t const *, svbfloat16x2_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst2q_vnum_u8)))
-void svst2q_vnum_u8(svbool_t, uint8_t const *, int64_t, svuint8x2_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst2q_vnum_u32)))
-void svst2q_vnum_u32(svbool_t, uint32_t const *, int64_t, svuint32x2_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst2q_vnum_u64)))
-void svst2q_vnum_u64(svbool_t, uint64_t const *, int64_t, svuint64x2_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst2q_vnum_u16)))
-void svst2q_vnum_u16(svbool_t, uint16_t const *, int64_t, svuint16x2_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst2q_vnum_s8)))
-void svst2q_vnum_s8(svbool_t, int8_t const *, int64_t, svint8x2_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst2q_vnum_f64)))
-void svst2q_vnum_f64(svbool_t, float64_t const *, int64_t, svfloat64x2_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst2q_vnum_f32)))
-void svst2q_vnum_f32(svbool_t, float32_t const *, int64_t, svfloat32x2_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst2q_vnum_f16)))
-void svst2q_vnum_f16(svbool_t, float16_t const *, int64_t, svfloat16x2_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst2q_vnum_s32)))
-void svst2q_vnum_s32(svbool_t, int32_t const *, int64_t, svint32x2_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst2q_vnum_s64)))
-void svst2q_vnum_s64(svbool_t, int64_t const *, int64_t, svint64x2_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst2q_vnum_s16)))
-void svst2q_vnum_s16(svbool_t, int16_t const *, int64_t, svint16x2_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst2q_vnum_bf16)))
-void svst2q_vnum_bf16(svbool_t, bfloat16_t const *, int64_t, svbfloat16x2_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst3q_u8)))
-void svst3q_u8(svbool_t, uint8_t const *, svuint8x3_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst3q_u32)))
-void svst3q_u32(svbool_t, uint32_t const *, svuint32x3_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst3q_u64)))
-void svst3q_u64(svbool_t, uint64_t const *, svuint64x3_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst3q_u16)))
-void svst3q_u16(svbool_t, uint16_t const *, svuint16x3_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst3q_s8)))
-void svst3q_s8(svbool_t, int8_t const *, svint8x3_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst3q_f64)))
-void svst3q_f64(svbool_t, float64_t const *, svfloat64x3_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst3q_f32)))
-void svst3q_f32(svbool_t, float32_t const *, svfloat32x3_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst3q_f16)))
-void svst3q_f16(svbool_t, float16_t const *, svfloat16x3_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst3q_s32)))
-void svst3q_s32(svbool_t, int32_t const *, svint32x3_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst3q_s64)))
-void svst3q_s64(svbool_t, int64_t const *, svint64x3_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst3q_s16)))
-void svst3q_s16(svbool_t, int16_t const *, svint16x3_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst3q_bf16)))
-void svst3q_bf16(svbool_t, bfloat16_t const *, svbfloat16x3_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst3q_vnum_u8)))
-void svst3q_vnum_u8(svbool_t, uint8_t const *, int64_t, svuint8x3_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst3q_vnum_u32)))
-void svst3q_vnum_u32(svbool_t, uint32_t const *, int64_t, svuint32x3_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst3q_vnum_u64)))
-void svst3q_vnum_u64(svbool_t, uint64_t const *, int64_t, svuint64x3_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst3q_vnum_u16)))
-void svst3q_vnum_u16(svbool_t, uint16_t const *, int64_t, svuint16x3_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst3q_vnum_s8)))
-void svst3q_vnum_s8(svbool_t, int8_t const *, int64_t, svint8x3_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst3q_vnum_f64)))
-void svst3q_vnum_f64(svbool_t, float64_t const *, int64_t, svfloat64x3_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst3q_vnum_f32)))
-void svst3q_vnum_f32(svbool_t, float32_t const *, int64_t, svfloat32x3_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst3q_vnum_f16)))
-void svst3q_vnum_f16(svbool_t, float16_t const *, int64_t, svfloat16x3_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst3q_vnum_s32)))
-void svst3q_vnum_s32(svbool_t, int32_t const *, int64_t, svint32x3_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst3q_vnum_s64)))
-void svst3q_vnum_s64(svbool_t, int64_t const *, int64_t, svint64x3_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst3q_vnum_s16)))
-void svst3q_vnum_s16(svbool_t, int16_t const *, int64_t, svint16x3_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst3q_vnum_bf16)))
-void svst3q_vnum_bf16(svbool_t, bfloat16_t const *, int64_t, svbfloat16x3_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst4q_u8)))
-void svst4q_u8(svbool_t, uint8_t const *, svuint8x4_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst4q_u32)))
-void svst4q_u32(svbool_t, uint32_t const *, svuint32x4_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst4q_u64)))
-void svst4q_u64(svbool_t, uint64_t const *, svuint64x4_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst4q_u16)))
-void svst4q_u16(svbool_t, uint16_t const *, svuint16x4_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst4q_s8)))
-void svst4q_s8(svbool_t, int8_t const *, svint8x4_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst4q_f64)))
-void svst4q_f64(svbool_t, float64_t const *, svfloat64x4_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst4q_f32)))
-void svst4q_f32(svbool_t, float32_t const *, svfloat32x4_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst4q_f16)))
-void svst4q_f16(svbool_t, float16_t const *, svfloat16x4_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst4q_s32)))
-void svst4q_s32(svbool_t, int32_t const *, svint32x4_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst4q_s64)))
-void svst4q_s64(svbool_t, int64_t const *, svint64x4_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst4q_s16)))
-void svst4q_s16(svbool_t, int16_t const *, svint16x4_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst4q_bf16)))
-void svst4q_bf16(svbool_t, bfloat16_t const *, svbfloat16x4_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst4q_vnum_u8)))
-void svst4q_vnum_u8(svbool_t, uint8_t const *, int64_t, svuint8x4_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst4q_vnum_u32)))
-void svst4q_vnum_u32(svbool_t, uint32_t const *, int64_t, svuint32x4_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst4q_vnum_u64)))
-void svst4q_vnum_u64(svbool_t, uint64_t const *, int64_t, svuint64x4_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst4q_vnum_u16)))
-void svst4q_vnum_u16(svbool_t, uint16_t const *, int64_t, svuint16x4_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst4q_vnum_s8)))
-void svst4q_vnum_s8(svbool_t, int8_t const *, int64_t, svint8x4_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst4q_vnum_f64)))
-void svst4q_vnum_f64(svbool_t, float64_t const *, int64_t, svfloat64x4_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst4q_vnum_f32)))
-void svst4q_vnum_f32(svbool_t, float32_t const *, int64_t, svfloat32x4_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst4q_vnum_f16)))
-void svst4q_vnum_f16(svbool_t, float16_t const *, int64_t, svfloat16x4_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst4q_vnum_s32)))
-void svst4q_vnum_s32(svbool_t, int32_t const *, int64_t, svint32x4_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst4q_vnum_s64)))
-void svst4q_vnum_s64(svbool_t, int64_t const *, int64_t, svint64x4_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst4q_vnum_s16)))
-void svst4q_vnum_s16(svbool_t, int16_t const *, int64_t, svint16x4_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst4q_vnum_bf16)))
-void svst4q_vnum_bf16(svbool_t, bfloat16_t const *, int64_t, svbfloat16x4_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtblq_u8)))
-svuint8_t svtblq_u8(svuint8_t, svuint8_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtblq_u32)))
-svuint32_t svtblq_u32(svuint32_t, svuint32_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtblq_u64)))
-svuint64_t svtblq_u64(svuint64_t, svuint64_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtblq_u16)))
-svuint16_t svtblq_u16(svuint16_t, svuint16_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtblq_bf16)))
-svbfloat16_t svtblq_bf16(svbfloat16_t, svuint16_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtblq_s8)))
-svint8_t svtblq_s8(svint8_t, svuint8_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtblq_f64)))
-svfloat64_t svtblq_f64(svfloat64_t, svuint64_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtblq_f32)))
-svfloat32_t svtblq_f32(svfloat32_t, svuint32_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtblq_f16)))
-svfloat16_t svtblq_f16(svfloat16_t, svuint16_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtblq_s32)))
-svint32_t svtblq_s32(svint32_t, svuint32_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtblq_s64)))
-svint64_t svtblq_s64(svint64_t, svuint64_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtblq_mf8)))
-svmfloat8_t svtblq_mf8(svmfloat8_t, svuint8_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtblq_s16)))
-svint16_t svtblq_s16(svint16_t, svuint16_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtbxq_u8)))
-svuint8_t svtbxq_u8(svuint8_t, svuint8_t, svuint8_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtbxq_u32)))
-svuint32_t svtbxq_u32(svuint32_t, svuint32_t, svuint32_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtbxq_u64)))
-svuint64_t svtbxq_u64(svuint64_t, svuint64_t, svuint64_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtbxq_u16)))
-svuint16_t svtbxq_u16(svuint16_t, svuint16_t, svuint16_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtbxq_bf16)))
-svbfloat16_t svtbxq_bf16(svbfloat16_t, svbfloat16_t, svuint16_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtbxq_s8)))
-svint8_t svtbxq_s8(svint8_t, svint8_t, svuint8_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtbxq_f64)))
-svfloat64_t svtbxq_f64(svfloat64_t, svfloat64_t, svuint64_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtbxq_f32)))
-svfloat32_t svtbxq_f32(svfloat32_t, svfloat32_t, svuint32_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtbxq_f16)))
-svfloat16_t svtbxq_f16(svfloat16_t, svfloat16_t, svuint16_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtbxq_s32)))
-svint32_t svtbxq_s32(svint32_t, svint32_t, svuint32_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtbxq_s64)))
-svint64_t svtbxq_s64(svint64_t, svint64_t, svuint64_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtbxq_mf8)))
-svmfloat8_t svtbxq_mf8(svmfloat8_t, svmfloat8_t, svuint8_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtbxq_s16)))
-svint16_t svtbxq_s16(svint16_t, svint16_t, svuint16_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzpq1_u8)))
-svuint8_t svuzpq1_u8(svuint8_t, svuint8_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzpq1_u32)))
-svuint32_t svuzpq1_u32(svuint32_t, svuint32_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzpq1_u64)))
-svuint64_t svuzpq1_u64(svuint64_t, svuint64_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzpq1_u16)))
-svuint16_t svuzpq1_u16(svuint16_t, svuint16_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzpq1_bf16)))
-svbfloat16_t svuzpq1_bf16(svbfloat16_t, svbfloat16_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzpq1_s8)))
-svint8_t svuzpq1_s8(svint8_t, svint8_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzpq1_f64)))
-svfloat64_t svuzpq1_f64(svfloat64_t, svfloat64_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzpq1_f32)))
-svfloat32_t svuzpq1_f32(svfloat32_t, svfloat32_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzpq1_f16)))
-svfloat16_t svuzpq1_f16(svfloat16_t, svfloat16_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzpq1_s32)))
-svint32_t svuzpq1_s32(svint32_t, svint32_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzpq1_s64)))
-svint64_t svuzpq1_s64(svint64_t, svint64_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzpq1_mf8)))
-svmfloat8_t svuzpq1_mf8(svmfloat8_t, svmfloat8_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzpq1_s16)))
-svint16_t svuzpq1_s16(svint16_t, svint16_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzpq2_u8)))
-svuint8_t svuzpq2_u8(svuint8_t, svuint8_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzpq2_u32)))
-svuint32_t svuzpq2_u32(svuint32_t, svuint32_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzpq2_u64)))
-svuint64_t svuzpq2_u64(svuint64_t, svuint64_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzpq2_u16)))
-svuint16_t svuzpq2_u16(svuint16_t, svuint16_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzpq2_bf16)))
-svbfloat16_t svuzpq2_bf16(svbfloat16_t, svbfloat16_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzpq2_s8)))
-svint8_t svuzpq2_s8(svint8_t, svint8_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzpq2_f64)))
-svfloat64_t svuzpq2_f64(svfloat64_t, svfloat64_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzpq2_f32)))
-svfloat32_t svuzpq2_f32(svfloat32_t, svfloat32_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzpq2_f16)))
-svfloat16_t svuzpq2_f16(svfloat16_t, svfloat16_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzpq2_s32)))
-svint32_t svuzpq2_s32(svint32_t, svint32_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzpq2_s64)))
-svint64_t svuzpq2_s64(svint64_t, svint64_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzpq2_mf8)))
-svmfloat8_t svuzpq2_mf8(svmfloat8_t, svmfloat8_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzpq2_s16)))
-svint16_t svuzpq2_s16(svint16_t, svint16_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzipq1_u8)))
-svuint8_t svzipq1_u8(svuint8_t, svuint8_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzipq1_u32)))
-svuint32_t svzipq1_u32(svuint32_t, svuint32_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzipq1_u64)))
-svuint64_t svzipq1_u64(svuint64_t, svuint64_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzipq1_u16)))
-svuint16_t svzipq1_u16(svuint16_t, svuint16_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzipq1_bf16)))
-svbfloat16_t svzipq1_bf16(svbfloat16_t, svbfloat16_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzipq1_s8)))
-svint8_t svzipq1_s8(svint8_t, svint8_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzipq1_f64)))
-svfloat64_t svzipq1_f64(svfloat64_t, svfloat64_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzipq1_f32)))
-svfloat32_t svzipq1_f32(svfloat32_t, svfloat32_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzipq1_f16)))
-svfloat16_t svzipq1_f16(svfloat16_t, svfloat16_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzipq1_s32)))
-svint32_t svzipq1_s32(svint32_t, svint32_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzipq1_s64)))
-svint64_t svzipq1_s64(svint64_t, svint64_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzipq1_mf8)))
-svmfloat8_t svzipq1_mf8(svmfloat8_t, svmfloat8_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzipq1_s16)))
-svint16_t svzipq1_s16(svint16_t, svint16_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzipq2_u8)))
-svuint8_t svzipq2_u8(svuint8_t, svuint8_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzipq2_u32)))
-svuint32_t svzipq2_u32(svuint32_t, svuint32_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzipq2_u64)))
-svuint64_t svzipq2_u64(svuint64_t, svuint64_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzipq2_u16)))
-svuint16_t svzipq2_u16(svuint16_t, svuint16_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzipq2_bf16)))
-svbfloat16_t svzipq2_bf16(svbfloat16_t, svbfloat16_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzipq2_s8)))
-svint8_t svzipq2_s8(svint8_t, svint8_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzipq2_f64)))
-svfloat64_t svzipq2_f64(svfloat64_t, svfloat64_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzipq2_f32)))
-svfloat32_t svzipq2_f32(svfloat32_t, svfloat32_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzipq2_f16)))
-svfloat16_t svzipq2_f16(svfloat16_t, svfloat16_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzipq2_s32)))
-svint32_t svzipq2_s32(svint32_t, svint32_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzipq2_s64)))
-svint64_t svzipq2_s64(svint64_t, svint64_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzipq2_mf8)))
-svmfloat8_t svzipq2_mf8(svmfloat8_t, svmfloat8_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzipq2_s16)))
-svint16_t svzipq2_s16(svint16_t, svint16_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddqv_u8)))
-uint8x16_t svaddqv(svbool_t, svuint8_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddqv_u32)))
-uint32x4_t svaddqv(svbool_t, svuint32_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddqv_u64)))
-uint64x2_t svaddqv(svbool_t, svuint64_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddqv_u16)))
-uint16x8_t svaddqv(svbool_t, svuint16_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddqv_s8)))
-int8x16_t svaddqv(svbool_t, svint8_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddqv_s32)))
-int32x4_t svaddqv(svbool_t, svint32_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddqv_s64)))
-int64x2_t svaddqv(svbool_t, svint64_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddqv_s16)))
-int16x8_t svaddqv(svbool_t, svint16_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddqv_f64)))
-float64x2_t svaddqv(svbool_t, svfloat64_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddqv_f32)))
-float32x4_t svaddqv(svbool_t, svfloat32_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaddqv_f16)))
-float16x8_t svaddqv(svbool_t, svfloat16_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svandqv_u8)))
-uint8x16_t svandqv(svbool_t, svuint8_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svandqv_u32)))
-uint32x4_t svandqv(svbool_t, svuint32_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svandqv_u64)))
-uint64x2_t svandqv(svbool_t, svuint64_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svandqv_u16)))
-uint16x8_t svandqv(svbool_t, svuint16_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svandqv_s8)))
-int8x16_t svandqv(svbool_t, svint8_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svandqv_s32)))
-int32x4_t svandqv(svbool_t, svint32_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svandqv_s64)))
-int64x2_t svandqv(svbool_t, svint64_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svandqv_s16)))
-int16x8_t svandqv(svbool_t, svint16_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveorqv_u8)))
-uint8x16_t sveorqv(svbool_t, svuint8_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveorqv_u32)))
-uint32x4_t sveorqv(svbool_t, svuint32_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveorqv_u64)))
-uint64x2_t sveorqv(svbool_t, svuint64_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveorqv_u16)))
-uint16x8_t sveorqv(svbool_t, svuint16_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveorqv_s8)))
-int8x16_t sveorqv(svbool_t, svint8_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveorqv_s32)))
-int32x4_t sveorqv(svbool_t, svint32_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveorqv_s64)))
-int64x2_t sveorqv(svbool_t, svint64_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_sveorqv_s16)))
-int16x8_t sveorqv(svbool_t, svint16_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svextq_u8)))
-svuint8_t svextq(svuint8_t, svuint8_t, int32_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svextq_u32)))
-svuint32_t svextq(svuint32_t, svuint32_t, int32_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svextq_u64)))
-svuint64_t svextq(svuint64_t, svuint64_t, int32_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svextq_u16)))
-svuint16_t svextq(svuint16_t, svuint16_t, int32_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svextq_bf16)))
-svbfloat16_t svextq(svbfloat16_t, svbfloat16_t, int32_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svextq_s8)))
-svint8_t svextq(svint8_t, svint8_t, int32_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svextq_f64)))
-svfloat64_t svextq(svfloat64_t, svfloat64_t, int32_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svextq_f32)))
-svfloat32_t svextq(svfloat32_t, svfloat32_t, int32_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svextq_f16)))
-svfloat16_t svextq(svfloat16_t, svfloat16_t, int32_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svextq_s32)))
-svint32_t svextq(svint32_t, svint32_t, int32_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svextq_s64)))
-svint64_t svextq(svint64_t, svint64_t, int32_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svextq_mf8)))
-svmfloat8_t svextq(svmfloat8_t, svmfloat8_t, int32_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svextq_s16)))
-svint16_t svextq(svint16_t, svint16_t, int32_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1q_gather_u64base_index_u32)))
 svuint32_t svld1q_gather_index_u32(svbool_t, svuint64_t, int64_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1q_gather_u64base_index_u64)))
@@ -10347,6 +11237,8 @@ __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1q_gather_u64ba
 svint32_t svld1q_gather_offset_s32(svbool_t, svuint64_t, int64_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1q_gather_u64base_offset_s64)))
 svint64_t svld1q_gather_offset_s64(svbool_t, svuint64_t, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1q_gather_u64base_offset_mf8)))
+svmfloat8_t svld1q_gather_offset_mf8(svbool_t, svuint64_t, int64_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1q_gather_u64base_offset_s16)))
 svint16_t svld1q_gather_offset_s16(svbool_t, svuint64_t, int64_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1q_gather_u64base_u8)))
@@ -10371,6 +11263,8 @@ __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1q_gather_u64ba
 svint32_t svld1q_gather_s32(svbool_t, svuint64_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1q_gather_u64base_s64)))
 svint64_t svld1q_gather_s64(svbool_t, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1q_gather_u64base_mf8)))
+svmfloat8_t svld1q_gather_mf8(svbool_t, svuint64_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1q_gather_u64base_s16)))
 svint16_t svld1q_gather_s16(svbool_t, svuint64_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1q_gather_u64index_u32)))
@@ -10415,6 +11309,8 @@ __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1q_gather_u64of
 svint32_t svld1q_gather_offset(svbool_t, int32_t const *, svuint64_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1q_gather_u64offset_s64)))
 svint64_t svld1q_gather_offset(svbool_t, int64_t const *, svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1q_gather_u64offset_mf8)))
+svmfloat8_t svld1q_gather_offset(svbool_t, mfloat8_t const *, svuint64_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1q_gather_u64offset_s16)))
 svint16_t svld1q_gather_offset(svbool_t, int16_t const *, svuint64_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1udq_u64)))
@@ -10441,266 +11337,6 @@ __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1uwq_vnum_f32))
 svfloat32_t svld1uwq_vnum(svbool_t, float32_t const *, int64_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1uwq_vnum_s32)))
 svint32_t svld1uwq_vnum(svbool_t, int32_t const *, int64_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld2q_u8)))
-svuint8x2_t svld2q(svbool_t, uint8_t const *);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld2q_u32)))
-svuint32x2_t svld2q(svbool_t, uint32_t const *);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld2q_u64)))
-svuint64x2_t svld2q(svbool_t, uint64_t const *);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld2q_u16)))
-svuint16x2_t svld2q(svbool_t, uint16_t const *);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld2q_s8)))
-svint8x2_t svld2q(svbool_t, int8_t const *);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld2q_f64)))
-svfloat64x2_t svld2q(svbool_t, float64_t const *);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld2q_f32)))
-svfloat32x2_t svld2q(svbool_t, float32_t const *);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld2q_f16)))
-svfloat16x2_t svld2q(svbool_t, float16_t const *);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld2q_s32)))
-svint32x2_t svld2q(svbool_t, int32_t const *);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld2q_s64)))
-svint64x2_t svld2q(svbool_t, int64_t const *);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld2q_s16)))
-svint16x2_t svld2q(svbool_t, int16_t const *);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld2q_bf16)))
-svbfloat16x2_t svld2q(svbool_t, bfloat16_t const *);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld2q_vnum_u8)))
-svuint8x2_t svld2q_vnum(svbool_t, uint8_t const *, int64_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld2q_vnum_u32)))
-svuint32x2_t svld2q_vnum(svbool_t, uint32_t const *, int64_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld2q_vnum_u64)))
-svuint64x2_t svld2q_vnum(svbool_t, uint64_t const *, int64_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld2q_vnum_u16)))
-svuint16x2_t svld2q_vnum(svbool_t, uint16_t const *, int64_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld2q_vnum_s8)))
-svint8x2_t svld2q_vnum(svbool_t, int8_t const *, int64_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld2q_vnum_f64)))
-svfloat64x2_t svld2q_vnum(svbool_t, float64_t const *, int64_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld2q_vnum_f32)))
-svfloat32x2_t svld2q_vnum(svbool_t, float32_t const *, int64_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld2q_vnum_f16)))
-svfloat16x2_t svld2q_vnum(svbool_t, float16_t const *, int64_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld2q_vnum_s32)))
-svint32x2_t svld2q_vnum(svbool_t, int32_t const *, int64_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld2q_vnum_s64)))
-svint64x2_t svld2q_vnum(svbool_t, int64_t const *, int64_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld2q_vnum_s16)))
-svint16x2_t svld2q_vnum(svbool_t, int16_t const *, int64_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld2q_vnum_bf16)))
-svbfloat16x2_t svld2q_vnum(svbool_t, bfloat16_t const *, int64_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld3q_u8)))
-svuint8x3_t svld3q(svbool_t, uint8_t const *);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld3q_u32)))
-svuint32x3_t svld3q(svbool_t, uint32_t const *);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld3q_u64)))
-svuint64x3_t svld3q(svbool_t, uint64_t const *);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld3q_u16)))
-svuint16x3_t svld3q(svbool_t, uint16_t const *);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld3q_s8)))
-svint8x3_t svld3q(svbool_t, int8_t const *);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld3q_f64)))
-svfloat64x3_t svld3q(svbool_t, float64_t const *);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld3q_f32)))
-svfloat32x3_t svld3q(svbool_t, float32_t const *);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld3q_f16)))
-svfloat16x3_t svld3q(svbool_t, float16_t const *);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld3q_s32)))
-svint32x3_t svld3q(svbool_t, int32_t const *);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld3q_s64)))
-svint64x3_t svld3q(svbool_t, int64_t const *);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld3q_s16)))
-svint16x3_t svld3q(svbool_t, int16_t const *);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld3q_bf16)))
-svbfloat16x3_t svld3q(svbool_t, bfloat16_t const *);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld3q_vnum_u8)))
-svuint8x3_t svld3q_vnum(svbool_t, uint8_t const *, int64_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld3q_vnum_u32)))
-svuint32x3_t svld3q_vnum(svbool_t, uint32_t const *, int64_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld3q_vnum_u64)))
-svuint64x3_t svld3q_vnum(svbool_t, uint64_t const *, int64_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld3q_vnum_u16)))
-svuint16x3_t svld3q_vnum(svbool_t, uint16_t const *, int64_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld3q_vnum_s8)))
-svint8x3_t svld3q_vnum(svbool_t, int8_t const *, int64_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld3q_vnum_f64)))
-svfloat64x3_t svld3q_vnum(svbool_t, float64_t const *, int64_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld3q_vnum_f32)))
-svfloat32x3_t svld3q_vnum(svbool_t, float32_t const *, int64_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld3q_vnum_f16)))
-svfloat16x3_t svld3q_vnum(svbool_t, float16_t const *, int64_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld3q_vnum_s32)))
-svint32x3_t svld3q_vnum(svbool_t, int32_t const *, int64_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld3q_vnum_s64)))
-svint64x3_t svld3q_vnum(svbool_t, int64_t const *, int64_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld3q_vnum_s16)))
-svint16x3_t svld3q_vnum(svbool_t, int16_t const *, int64_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld3q_vnum_bf16)))
-svbfloat16x3_t svld3q_vnum(svbool_t, bfloat16_t const *, int64_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld4q_u8)))
-svuint8x4_t svld4q(svbool_t, uint8_t const *);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld4q_u32)))
-svuint32x4_t svld4q(svbool_t, uint32_t const *);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld4q_u64)))
-svuint64x4_t svld4q(svbool_t, uint64_t const *);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld4q_u16)))
-svuint16x4_t svld4q(svbool_t, uint16_t const *);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld4q_s8)))
-svint8x4_t svld4q(svbool_t, int8_t const *);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld4q_f64)))
-svfloat64x4_t svld4q(svbool_t, float64_t const *);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld4q_f32)))
-svfloat32x4_t svld4q(svbool_t, float32_t const *);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld4q_f16)))
-svfloat16x4_t svld4q(svbool_t, float16_t const *);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld4q_s32)))
-svint32x4_t svld4q(svbool_t, int32_t const *);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld4q_s64)))
-svint64x4_t svld4q(svbool_t, int64_t const *);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld4q_s16)))
-svint16x4_t svld4q(svbool_t, int16_t const *);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld4q_bf16)))
-svbfloat16x4_t svld4q(svbool_t, bfloat16_t const *);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld4q_vnum_u8)))
-svuint8x4_t svld4q_vnum(svbool_t, uint8_t const *, int64_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld4q_vnum_u32)))
-svuint32x4_t svld4q_vnum(svbool_t, uint32_t const *, int64_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld4q_vnum_u64)))
-svuint64x4_t svld4q_vnum(svbool_t, uint64_t const *, int64_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld4q_vnum_u16)))
-svuint16x4_t svld4q_vnum(svbool_t, uint16_t const *, int64_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld4q_vnum_s8)))
-svint8x4_t svld4q_vnum(svbool_t, int8_t const *, int64_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld4q_vnum_f64)))
-svfloat64x4_t svld4q_vnum(svbool_t, float64_t const *, int64_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld4q_vnum_f32)))
-svfloat32x4_t svld4q_vnum(svbool_t, float32_t const *, int64_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld4q_vnum_f16)))
-svfloat16x4_t svld4q_vnum(svbool_t, float16_t const *, int64_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld4q_vnum_s32)))
-svint32x4_t svld4q_vnum(svbool_t, int32_t const *, int64_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld4q_vnum_s64)))
-svint64x4_t svld4q_vnum(svbool_t, int64_t const *, int64_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld4q_vnum_s16)))
-svint16x4_t svld4q_vnum(svbool_t, int16_t const *, int64_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld4q_vnum_bf16)))
-svbfloat16x4_t svld4q_vnum(svbool_t, bfloat16_t const *, int64_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmaxnmqv_f64)))
-float64x2_t svmaxnmqv(svbool_t, svfloat64_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmaxnmqv_f32)))
-float32x4_t svmaxnmqv(svbool_t, svfloat32_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmaxnmqv_f16)))
-float16x8_t svmaxnmqv(svbool_t, svfloat16_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmaxqv_f64)))
-float64x2_t svmaxqv(svbool_t, svfloat64_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmaxqv_f32)))
-float32x4_t svmaxqv(svbool_t, svfloat32_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmaxqv_f16)))
-float16x8_t svmaxqv(svbool_t, svfloat16_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmaxqv_s8)))
-int8x16_t svmaxqv(svbool_t, svint8_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmaxqv_s32)))
-int32x4_t svmaxqv(svbool_t, svint32_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmaxqv_s64)))
-int64x2_t svmaxqv(svbool_t, svint64_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmaxqv_s16)))
-int16x8_t svmaxqv(svbool_t, svint16_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmaxqv_u8)))
-uint8x16_t svmaxqv(svbool_t, svuint8_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmaxqv_u32)))
-uint32x4_t svmaxqv(svbool_t, svuint32_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmaxqv_u64)))
-uint64x2_t svmaxqv(svbool_t, svuint64_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svmaxqv_u16)))
-uint16x8_t svmaxqv(svbool_t, svuint16_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svminnmqv_f64)))
-float64x2_t svminnmqv(svbool_t, svfloat64_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svminnmqv_f32)))
-float32x4_t svminnmqv(svbool_t, svfloat32_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svminnmqv_f16)))
-float16x8_t svminnmqv(svbool_t, svfloat16_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svminqv_f64)))
-float64x2_t svminqv(svbool_t, svfloat64_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svminqv_f32)))
-float32x4_t svminqv(svbool_t, svfloat32_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svminqv_f16)))
-float16x8_t svminqv(svbool_t, svfloat16_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svminqv_s8)))
-int8x16_t svminqv(svbool_t, svint8_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svminqv_s32)))
-int32x4_t svminqv(svbool_t, svint32_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svminqv_s64)))
-int64x2_t svminqv(svbool_t, svint64_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svminqv_s16)))
-int16x8_t svminqv(svbool_t, svint16_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svminqv_u8)))
-uint8x16_t svminqv(svbool_t, svuint8_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svminqv_u32)))
-uint32x4_t svminqv(svbool_t, svuint32_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svminqv_u64)))
-uint64x2_t svminqv(svbool_t, svuint64_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svminqv_u16)))
-uint16x8_t svminqv(svbool_t, svuint16_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svorqv_u8)))
-uint8x16_t svorqv(svbool_t, svuint8_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svorqv_u32)))
-uint32x4_t svorqv(svbool_t, svuint32_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svorqv_u64)))
-uint64x2_t svorqv(svbool_t, svuint64_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svorqv_u16)))
-uint16x8_t svorqv(svbool_t, svuint16_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svorqv_s8)))
-int8x16_t svorqv(svbool_t, svint8_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svorqv_s32)))
-int32x4_t svorqv(svbool_t, svint32_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svorqv_s64)))
-int64x2_t svorqv(svbool_t, svint64_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svorqv_s16)))
-int16x8_t svorqv(svbool_t, svint16_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svpmov_u8)))
-svbool_t svpmov(svuint8_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svpmov_s8)))
-svbool_t svpmov(svint8_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svpmov_u64)))
-svbool_t svpmov(svuint64_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svpmov_s64)))
-svbool_t svpmov(svint64_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svpmov_u16)))
-svbool_t svpmov(svuint16_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svpmov_s16)))
-svbool_t svpmov(svint16_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svpmov_u32)))
-svbool_t svpmov(svuint32_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svpmov_s32)))
-svbool_t svpmov(svint32_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svpmov_lane_u8)))
-svbool_t svpmov_lane(svuint8_t, uint64_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svpmov_lane_s8)))
-svbool_t svpmov_lane(svint8_t, uint64_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svpmov_lane_u64)))
-svbool_t svpmov_lane(svuint64_t, uint64_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svpmov_lane_s64)))
-svbool_t svpmov_lane(svint64_t, uint64_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svpmov_lane_u16)))
-svbool_t svpmov_lane(svuint16_t, uint64_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svpmov_lane_s16)))
-svbool_t svpmov_lane(svint16_t, uint64_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svpmov_lane_u32)))
-svbool_t svpmov_lane(svuint32_t, uint64_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svpmov_lane_s32)))
-svbool_t svpmov_lane(svint32_t, uint64_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svpmov_lane_u64_m)))
-svuint64_t svpmov_lane_m(svuint64_t, svbool_t, uint64_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svpmov_lane_s64_m)))
-svint64_t svpmov_lane_m(svint64_t, svbool_t, uint64_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svpmov_lane_u16_m)))
-svuint16_t svpmov_lane_m(svuint16_t, svbool_t, uint64_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svpmov_lane_s16_m)))
-svint16_t svpmov_lane_m(svint16_t, svbool_t, uint64_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svpmov_lane_u32_m)))
-svuint32_t svpmov_lane_m(svuint32_t, svbool_t, uint64_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svpmov_lane_s32_m)))
-svint32_t svpmov_lane_m(svint32_t, svbool_t, uint64_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1dq_u64)))
 void svst1dq(svbool_t, uint64_t *, svuint64_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1dq_f64)))
@@ -10735,6 +11371,8 @@ __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1q_scatter_u64b
 void svst1q_scatter(svbool_t, svuint64_t, svint32_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1q_scatter_u64base_s64)))
 void svst1q_scatter(svbool_t, svuint64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1q_scatter_u64base_mf8)))
+void svst1q_scatter(svbool_t, svuint64_t, svmfloat8_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1q_scatter_u64base_s16)))
 void svst1q_scatter(svbool_t, svuint64_t, svint16_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1q_scatter_u64base_index_u32)))
@@ -10779,6 +11417,8 @@ __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1q_scatter_u64b
 void svst1q_scatter_offset(svbool_t, svuint64_t, int64_t, svint32_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1q_scatter_u64base_offset_s64)))
 void svst1q_scatter_offset(svbool_t, svuint64_t, int64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1q_scatter_u64base_offset_mf8)))
+void svst1q_scatter_offset(svbool_t, svuint64_t, int64_t, svmfloat8_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1q_scatter_u64base_offset_s16)))
 void svst1q_scatter_offset(svbool_t, svuint64_t, int64_t, svint16_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1q_scatter_s64index_u32)))
@@ -10843,6 +11483,8 @@ __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1q_scatter_s64o
 void svst1q_scatter_offset(svbool_t, int32_t *, svint64_t, svint32_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1q_scatter_s64offset_s64)))
 void svst1q_scatter_offset(svbool_t, int64_t *, svint64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1q_scatter_s64offset_mf8)))
+void svst1q_scatter_offset(svbool_t, mfloat8_t *, svint64_t, svmfloat8_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1q_scatter_s64offset_s16)))
 void svst1q_scatter_offset(svbool_t, int16_t *, svint64_t, svint16_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1q_scatter_u64offset_u8)))
@@ -10867,6 +11509,8 @@ __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1q_scatter_u64o
 void svst1q_scatter_offset(svbool_t, int32_t *, svuint64_t, svint32_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1q_scatter_u64offset_s64)))
 void svst1q_scatter_offset(svbool_t, int64_t *, svuint64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1q_scatter_u64offset_mf8)))
+void svst1q_scatter_offset(svbool_t, mfloat8_t *, svuint64_t, svmfloat8_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1q_scatter_u64offset_s16)))
 void svst1q_scatter_offset(svbool_t, int16_t *, svuint64_t, svint16_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1wq_u32)))
@@ -10881,508 +11525,6 @@ __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1wq_vnum_f32)))
 void svst1wq_vnum(svbool_t, float32_t *, int64_t, svfloat32_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1wq_vnum_s32)))
 void svst1wq_vnum(svbool_t, int32_t *, int64_t, svint32_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst2q_u8)))
-void svst2q(svbool_t, uint8_t const *, svuint8x2_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst2q_u32)))
-void svst2q(svbool_t, uint32_t const *, svuint32x2_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst2q_u64)))
-void svst2q(svbool_t, uint64_t const *, svuint64x2_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst2q_u16)))
-void svst2q(svbool_t, uint16_t const *, svuint16x2_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst2q_s8)))
-void svst2q(svbool_t, int8_t const *, svint8x2_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst2q_f64)))
-void svst2q(svbool_t, float64_t const *, svfloat64x2_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst2q_f32)))
-void svst2q(svbool_t, float32_t const *, svfloat32x2_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst2q_f16)))
-void svst2q(svbool_t, float16_t const *, svfloat16x2_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst2q_s32)))
-void svst2q(svbool_t, int32_t const *, svint32x2_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst2q_s64)))
-void svst2q(svbool_t, int64_t const *, svint64x2_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst2q_s16)))
-void svst2q(svbool_t, int16_t const *, svint16x2_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst2q_bf16)))
-void svst2q(svbool_t, bfloat16_t const *, svbfloat16x2_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst2q_vnum_u8)))
-void svst2q_vnum(svbool_t, uint8_t const *, int64_t, svuint8x2_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst2q_vnum_u32)))
-void svst2q_vnum(svbool_t, uint32_t const *, int64_t, svuint32x2_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst2q_vnum_u64)))
-void svst2q_vnum(svbool_t, uint64_t const *, int64_t, svuint64x2_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst2q_vnum_u16)))
-void svst2q_vnum(svbool_t, uint16_t const *, int64_t, svuint16x2_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst2q_vnum_s8)))
-void svst2q_vnum(svbool_t, int8_t const *, int64_t, svint8x2_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst2q_vnum_f64)))
-void svst2q_vnum(svbool_t, float64_t const *, int64_t, svfloat64x2_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst2q_vnum_f32)))
-void svst2q_vnum(svbool_t, float32_t const *, int64_t, svfloat32x2_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst2q_vnum_f16)))
-void svst2q_vnum(svbool_t, float16_t const *, int64_t, svfloat16x2_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst2q_vnum_s32)))
-void svst2q_vnum(svbool_t, int32_t const *, int64_t, svint32x2_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst2q_vnum_s64)))
-void svst2q_vnum(svbool_t, int64_t const *, int64_t, svint64x2_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst2q_vnum_s16)))
-void svst2q_vnum(svbool_t, int16_t const *, int64_t, svint16x2_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst2q_vnum_bf16)))
-void svst2q_vnum(svbool_t, bfloat16_t const *, int64_t, svbfloat16x2_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst3q_u8)))
-void svst3q(svbool_t, uint8_t const *, svuint8x3_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst3q_u32)))
-void svst3q(svbool_t, uint32_t const *, svuint32x3_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst3q_u64)))
-void svst3q(svbool_t, uint64_t const *, svuint64x3_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst3q_u16)))
-void svst3q(svbool_t, uint16_t const *, svuint16x3_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst3q_s8)))
-void svst3q(svbool_t, int8_t const *, svint8x3_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst3q_f64)))
-void svst3q(svbool_t, float64_t const *, svfloat64x3_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst3q_f32)))
-void svst3q(svbool_t, float32_t const *, svfloat32x3_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst3q_f16)))
-void svst3q(svbool_t, float16_t const *, svfloat16x3_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst3q_s32)))
-void svst3q(svbool_t, int32_t const *, svint32x3_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst3q_s64)))
-void svst3q(svbool_t, int64_t const *, svint64x3_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst3q_s16)))
-void svst3q(svbool_t, int16_t const *, svint16x3_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst3q_bf16)))
-void svst3q(svbool_t, bfloat16_t const *, svbfloat16x3_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst3q_vnum_u8)))
-void svst3q_vnum(svbool_t, uint8_t const *, int64_t, svuint8x3_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst3q_vnum_u32)))
-void svst3q_vnum(svbool_t, uint32_t const *, int64_t, svuint32x3_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst3q_vnum_u64)))
-void svst3q_vnum(svbool_t, uint64_t const *, int64_t, svuint64x3_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst3q_vnum_u16)))
-void svst3q_vnum(svbool_t, uint16_t const *, int64_t, svuint16x3_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst3q_vnum_s8)))
-void svst3q_vnum(svbool_t, int8_t const *, int64_t, svint8x3_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst3q_vnum_f64)))
-void svst3q_vnum(svbool_t, float64_t const *, int64_t, svfloat64x3_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst3q_vnum_f32)))
-void svst3q_vnum(svbool_t, float32_t const *, int64_t, svfloat32x3_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst3q_vnum_f16)))
-void svst3q_vnum(svbool_t, float16_t const *, int64_t, svfloat16x3_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst3q_vnum_s32)))
-void svst3q_vnum(svbool_t, int32_t const *, int64_t, svint32x3_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst3q_vnum_s64)))
-void svst3q_vnum(svbool_t, int64_t const *, int64_t, svint64x3_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst3q_vnum_s16)))
-void svst3q_vnum(svbool_t, int16_t const *, int64_t, svint16x3_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst3q_vnum_bf16)))
-void svst3q_vnum(svbool_t, bfloat16_t const *, int64_t, svbfloat16x3_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst4q_u8)))
-void svst4q(svbool_t, uint8_t const *, svuint8x4_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst4q_u32)))
-void svst4q(svbool_t, uint32_t const *, svuint32x4_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst4q_u64)))
-void svst4q(svbool_t, uint64_t const *, svuint64x4_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst4q_u16)))
-void svst4q(svbool_t, uint16_t const *, svuint16x4_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst4q_s8)))
-void svst4q(svbool_t, int8_t const *, svint8x4_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst4q_f64)))
-void svst4q(svbool_t, float64_t const *, svfloat64x4_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst4q_f32)))
-void svst4q(svbool_t, float32_t const *, svfloat32x4_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst4q_f16)))
-void svst4q(svbool_t, float16_t const *, svfloat16x4_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst4q_s32)))
-void svst4q(svbool_t, int32_t const *, svint32x4_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst4q_s64)))
-void svst4q(svbool_t, int64_t const *, svint64x4_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst4q_s16)))
-void svst4q(svbool_t, int16_t const *, svint16x4_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst4q_bf16)))
-void svst4q(svbool_t, bfloat16_t const *, svbfloat16x4_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst4q_vnum_u8)))
-void svst4q_vnum(svbool_t, uint8_t const *, int64_t, svuint8x4_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst4q_vnum_u32)))
-void svst4q_vnum(svbool_t, uint32_t const *, int64_t, svuint32x4_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst4q_vnum_u64)))
-void svst4q_vnum(svbool_t, uint64_t const *, int64_t, svuint64x4_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst4q_vnum_u16)))
-void svst4q_vnum(svbool_t, uint16_t const *, int64_t, svuint16x4_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst4q_vnum_s8)))
-void svst4q_vnum(svbool_t, int8_t const *, int64_t, svint8x4_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst4q_vnum_f64)))
-void svst4q_vnum(svbool_t, float64_t const *, int64_t, svfloat64x4_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst4q_vnum_f32)))
-void svst4q_vnum(svbool_t, float32_t const *, int64_t, svfloat32x4_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst4q_vnum_f16)))
-void svst4q_vnum(svbool_t, float16_t const *, int64_t, svfloat16x4_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst4q_vnum_s32)))
-void svst4q_vnum(svbool_t, int32_t const *, int64_t, svint32x4_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst4q_vnum_s64)))
-void svst4q_vnum(svbool_t, int64_t const *, int64_t, svint64x4_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst4q_vnum_s16)))
-void svst4q_vnum(svbool_t, int16_t const *, int64_t, svint16x4_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst4q_vnum_bf16)))
-void svst4q_vnum(svbool_t, bfloat16_t const *, int64_t, svbfloat16x4_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtblq_u8)))
-svuint8_t svtblq(svuint8_t, svuint8_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtblq_u32)))
-svuint32_t svtblq(svuint32_t, svuint32_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtblq_u64)))
-svuint64_t svtblq(svuint64_t, svuint64_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtblq_u16)))
-svuint16_t svtblq(svuint16_t, svuint16_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtblq_bf16)))
-svbfloat16_t svtblq(svbfloat16_t, svuint16_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtblq_s8)))
-svint8_t svtblq(svint8_t, svuint8_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtblq_f64)))
-svfloat64_t svtblq(svfloat64_t, svuint64_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtblq_f32)))
-svfloat32_t svtblq(svfloat32_t, svuint32_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtblq_f16)))
-svfloat16_t svtblq(svfloat16_t, svuint16_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtblq_s32)))
-svint32_t svtblq(svint32_t, svuint32_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtblq_s64)))
-svint64_t svtblq(svint64_t, svuint64_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtblq_mf8)))
-svmfloat8_t svtblq(svmfloat8_t, svuint8_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtblq_s16)))
-svint16_t svtblq(svint16_t, svuint16_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtbxq_u8)))
-svuint8_t svtbxq(svuint8_t, svuint8_t, svuint8_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtbxq_u32)))
-svuint32_t svtbxq(svuint32_t, svuint32_t, svuint32_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtbxq_u64)))
-svuint64_t svtbxq(svuint64_t, svuint64_t, svuint64_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtbxq_u16)))
-svuint16_t svtbxq(svuint16_t, svuint16_t, svuint16_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtbxq_bf16)))
-svbfloat16_t svtbxq(svbfloat16_t, svbfloat16_t, svuint16_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtbxq_s8)))
-svint8_t svtbxq(svint8_t, svint8_t, svuint8_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtbxq_f64)))
-svfloat64_t svtbxq(svfloat64_t, svfloat64_t, svuint64_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtbxq_f32)))
-svfloat32_t svtbxq(svfloat32_t, svfloat32_t, svuint32_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtbxq_f16)))
-svfloat16_t svtbxq(svfloat16_t, svfloat16_t, svuint16_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtbxq_s32)))
-svint32_t svtbxq(svint32_t, svint32_t, svuint32_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtbxq_s64)))
-svint64_t svtbxq(svint64_t, svint64_t, svuint64_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtbxq_mf8)))
-svmfloat8_t svtbxq(svmfloat8_t, svmfloat8_t, svuint8_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtbxq_s16)))
-svint16_t svtbxq(svint16_t, svint16_t, svuint16_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzpq1_u8)))
-svuint8_t svuzpq1(svuint8_t, svuint8_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzpq1_u32)))
-svuint32_t svuzpq1(svuint32_t, svuint32_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzpq1_u64)))
-svuint64_t svuzpq1(svuint64_t, svuint64_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzpq1_u16)))
-svuint16_t svuzpq1(svuint16_t, svuint16_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzpq1_bf16)))
-svbfloat16_t svuzpq1(svbfloat16_t, svbfloat16_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzpq1_s8)))
-svint8_t svuzpq1(svint8_t, svint8_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzpq1_f64)))
-svfloat64_t svuzpq1(svfloat64_t, svfloat64_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzpq1_f32)))
-svfloat32_t svuzpq1(svfloat32_t, svfloat32_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzpq1_f16)))
-svfloat16_t svuzpq1(svfloat16_t, svfloat16_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzpq1_s32)))
-svint32_t svuzpq1(svint32_t, svint32_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzpq1_s64)))
-svint64_t svuzpq1(svint64_t, svint64_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzpq1_mf8)))
-svmfloat8_t svuzpq1(svmfloat8_t, svmfloat8_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzpq1_s16)))
-svint16_t svuzpq1(svint16_t, svint16_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzpq2_u8)))
-svuint8_t svuzpq2(svuint8_t, svuint8_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzpq2_u32)))
-svuint32_t svuzpq2(svuint32_t, svuint32_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzpq2_u64)))
-svuint64_t svuzpq2(svuint64_t, svuint64_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzpq2_u16)))
-svuint16_t svuzpq2(svuint16_t, svuint16_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzpq2_bf16)))
-svbfloat16_t svuzpq2(svbfloat16_t, svbfloat16_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzpq2_s8)))
-svint8_t svuzpq2(svint8_t, svint8_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzpq2_f64)))
-svfloat64_t svuzpq2(svfloat64_t, svfloat64_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzpq2_f32)))
-svfloat32_t svuzpq2(svfloat32_t, svfloat32_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzpq2_f16)))
-svfloat16_t svuzpq2(svfloat16_t, svfloat16_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzpq2_s32)))
-svint32_t svuzpq2(svint32_t, svint32_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzpq2_s64)))
-svint64_t svuzpq2(svint64_t, svint64_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzpq2_mf8)))
-svmfloat8_t svuzpq2(svmfloat8_t, svmfloat8_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzpq2_s16)))
-svint16_t svuzpq2(svint16_t, svint16_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzipq1_u8)))
-svuint8_t svzipq1(svuint8_t, svuint8_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzipq1_u32)))
-svuint32_t svzipq1(svuint32_t, svuint32_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzipq1_u64)))
-svuint64_t svzipq1(svuint64_t, svuint64_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzipq1_u16)))
-svuint16_t svzipq1(svuint16_t, svuint16_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzipq1_bf16)))
-svbfloat16_t svzipq1(svbfloat16_t, svbfloat16_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzipq1_s8)))
-svint8_t svzipq1(svint8_t, svint8_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzipq1_f64)))
-svfloat64_t svzipq1(svfloat64_t, svfloat64_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzipq1_f32)))
-svfloat32_t svzipq1(svfloat32_t, svfloat32_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzipq1_f16)))
-svfloat16_t svzipq1(svfloat16_t, svfloat16_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzipq1_s32)))
-svint32_t svzipq1(svint32_t, svint32_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzipq1_s64)))
-svint64_t svzipq1(svint64_t, svint64_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzipq1_mf8)))
-svmfloat8_t svzipq1(svmfloat8_t, svmfloat8_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzipq1_s16)))
-svint16_t svzipq1(svint16_t, svint16_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzipq2_u8)))
-svuint8_t svzipq2(svuint8_t, svuint8_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzipq2_u32)))
-svuint32_t svzipq2(svuint32_t, svuint32_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzipq2_u64)))
-svuint64_t svzipq2(svuint64_t, svuint64_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzipq2_u16)))
-svuint16_t svzipq2(svuint16_t, svuint16_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzipq2_bf16)))
-svbfloat16_t svzipq2(svbfloat16_t, svbfloat16_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzipq2_s8)))
-svint8_t svzipq2(svint8_t, svint8_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzipq2_f64)))
-svfloat64_t svzipq2(svfloat64_t, svfloat64_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzipq2_f32)))
-svfloat32_t svzipq2(svfloat32_t, svfloat32_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzipq2_f16)))
-svfloat16_t svzipq2(svfloat16_t, svfloat16_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzipq2_s32)))
-svint32_t svzipq2(svint32_t, svint32_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzipq2_s64)))
-svint64_t svzipq2(svint64_t, svint64_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzipq2_mf8)))
-svmfloat8_t svzipq2(svmfloat8_t, svmfloat8_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzipq2_s16)))
-svint16_t svzipq2(svint16_t, svint16_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdup_laneq_bf16)))
-svbfloat16_t svdup_laneq_bf16(svbfloat16_t, uint64_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdup_laneq_bf16)))
-svbfloat16_t svdup_laneq(svbfloat16_t, uint64_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svclamp_s8)))
-svint8_t svclamp_s8(svint8_t, svint8_t, svint8_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svclamp_s32)))
-svint32_t svclamp_s32(svint32_t, svint32_t, svint32_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svclamp_s64)))
-svint64_t svclamp_s64(svint64_t, svint64_t, svint64_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svclamp_s16)))
-svint16_t svclamp_s16(svint16_t, svint16_t, svint16_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svclamp_u8)))
-svuint8_t svclamp_u8(svuint8_t, svuint8_t, svuint8_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svclamp_u32)))
-svuint32_t svclamp_u32(svuint32_t, svuint32_t, svuint32_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svclamp_u64)))
-svuint64_t svclamp_u64(svuint64_t, svuint64_t, svuint64_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svclamp_u16)))
-svuint16_t svclamp_u16(svuint16_t, svuint16_t, svuint16_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svpsel_lane_b16)))
-svbool_t svpsel_lane_b16(svbool_t, svbool_t, uint32_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svpsel_lane_b32)))
-svbool_t svpsel_lane_b32(svbool_t, svbool_t, uint32_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svpsel_lane_b64)))
-svbool_t svpsel_lane_b64(svbool_t, svbool_t, uint32_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svpsel_lane_b8)))
-svbool_t svpsel_lane_b8(svbool_t, svbool_t, uint32_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrevd_u8_m)))
-svuint8_t svrevd_u8_m(svuint8_t, svbool_t, svuint8_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrevd_u32_m)))
-svuint32_t svrevd_u32_m(svuint32_t, svbool_t, svuint32_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrevd_u64_m)))
-svuint64_t svrevd_u64_m(svuint64_t, svbool_t, svuint64_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrevd_u16_m)))
-svuint16_t svrevd_u16_m(svuint16_t, svbool_t, svuint16_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrevd_bf16_m)))
-svbfloat16_t svrevd_bf16_m(svbfloat16_t, svbool_t, svbfloat16_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrevd_s8_m)))
-svint8_t svrevd_s8_m(svint8_t, svbool_t, svint8_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrevd_f64_m)))
-svfloat64_t svrevd_f64_m(svfloat64_t, svbool_t, svfloat64_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrevd_f32_m)))
-svfloat32_t svrevd_f32_m(svfloat32_t, svbool_t, svfloat32_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrevd_f16_m)))
-svfloat16_t svrevd_f16_m(svfloat16_t, svbool_t, svfloat16_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrevd_s32_m)))
-svint32_t svrevd_s32_m(svint32_t, svbool_t, svint32_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrevd_s64_m)))
-svint64_t svrevd_s64_m(svint64_t, svbool_t, svint64_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrevd_s16_m)))
-svint16_t svrevd_s16_m(svint16_t, svbool_t, svint16_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrevd_u8_x)))
-svuint8_t svrevd_u8_x(svbool_t, svuint8_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrevd_u32_x)))
-svuint32_t svrevd_u32_x(svbool_t, svuint32_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrevd_u64_x)))
-svuint64_t svrevd_u64_x(svbool_t, svuint64_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrevd_u16_x)))
-svuint16_t svrevd_u16_x(svbool_t, svuint16_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrevd_bf16_x)))
-svbfloat16_t svrevd_bf16_x(svbool_t, svbfloat16_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrevd_s8_x)))
-svint8_t svrevd_s8_x(svbool_t, svint8_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrevd_f64_x)))
-svfloat64_t svrevd_f64_x(svbool_t, svfloat64_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrevd_f32_x)))
-svfloat32_t svrevd_f32_x(svbool_t, svfloat32_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrevd_f16_x)))
-svfloat16_t svrevd_f16_x(svbool_t, svfloat16_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrevd_s32_x)))
-svint32_t svrevd_s32_x(svbool_t, svint32_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrevd_s64_x)))
-svint64_t svrevd_s64_x(svbool_t, svint64_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrevd_s16_x)))
-svint16_t svrevd_s16_x(svbool_t, svint16_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrevd_u8_z)))
-svuint8_t svrevd_u8_z(svbool_t, svuint8_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrevd_u32_z)))
-svuint32_t svrevd_u32_z(svbool_t, svuint32_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrevd_u64_z)))
-svuint64_t svrevd_u64_z(svbool_t, svuint64_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrevd_u16_z)))
-svuint16_t svrevd_u16_z(svbool_t, svuint16_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrevd_bf16_z)))
-svbfloat16_t svrevd_bf16_z(svbool_t, svbfloat16_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrevd_s8_z)))
-svint8_t svrevd_s8_z(svbool_t, svint8_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrevd_f64_z)))
-svfloat64_t svrevd_f64_z(svbool_t, svfloat64_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrevd_f32_z)))
-svfloat32_t svrevd_f32_z(svbool_t, svfloat32_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrevd_f16_z)))
-svfloat16_t svrevd_f16_z(svbool_t, svfloat16_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrevd_s32_z)))
-svint32_t svrevd_s32_z(svbool_t, svint32_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrevd_s64_z)))
-svint64_t svrevd_s64_z(svbool_t, svint64_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrevd_s16_z)))
-svint16_t svrevd_s16_z(svbool_t, svint16_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svclamp_s8)))
-svint8_t svclamp(svint8_t, svint8_t, svint8_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svclamp_s32)))
-svint32_t svclamp(svint32_t, svint32_t, svint32_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svclamp_s64)))
-svint64_t svclamp(svint64_t, svint64_t, svint64_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svclamp_s16)))
-svint16_t svclamp(svint16_t, svint16_t, svint16_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svclamp_u8)))
-svuint8_t svclamp(svuint8_t, svuint8_t, svuint8_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svclamp_u32)))
-svuint32_t svclamp(svuint32_t, svuint32_t, svuint32_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svclamp_u64)))
-svuint64_t svclamp(svuint64_t, svuint64_t, svuint64_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svclamp_u16)))
-svuint16_t svclamp(svuint16_t, svuint16_t, svuint16_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrevd_u8_m)))
-svuint8_t svrevd_m(svuint8_t, svbool_t, svuint8_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrevd_u32_m)))
-svuint32_t svrevd_m(svuint32_t, svbool_t, svuint32_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrevd_u64_m)))
-svuint64_t svrevd_m(svuint64_t, svbool_t, svuint64_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrevd_u16_m)))
-svuint16_t svrevd_m(svuint16_t, svbool_t, svuint16_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrevd_bf16_m)))
-svbfloat16_t svrevd_m(svbfloat16_t, svbool_t, svbfloat16_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrevd_s8_m)))
-svint8_t svrevd_m(svint8_t, svbool_t, svint8_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrevd_f64_m)))
-svfloat64_t svrevd_m(svfloat64_t, svbool_t, svfloat64_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrevd_f32_m)))
-svfloat32_t svrevd_m(svfloat32_t, svbool_t, svfloat32_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrevd_f16_m)))
-svfloat16_t svrevd_m(svfloat16_t, svbool_t, svfloat16_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrevd_s32_m)))
-svint32_t svrevd_m(svint32_t, svbool_t, svint32_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrevd_s64_m)))
-svint64_t svrevd_m(svint64_t, svbool_t, svint64_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrevd_s16_m)))
-svint16_t svrevd_m(svint16_t, svbool_t, svint16_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrevd_u8_x)))
-svuint8_t svrevd_x(svbool_t, svuint8_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrevd_u32_x)))
-svuint32_t svrevd_x(svbool_t, svuint32_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrevd_u64_x)))
-svuint64_t svrevd_x(svbool_t, svuint64_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrevd_u16_x)))
-svuint16_t svrevd_x(svbool_t, svuint16_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrevd_bf16_x)))
-svbfloat16_t svrevd_x(svbool_t, svbfloat16_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrevd_s8_x)))
-svint8_t svrevd_x(svbool_t, svint8_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrevd_f64_x)))
-svfloat64_t svrevd_x(svbool_t, svfloat64_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrevd_f32_x)))
-svfloat32_t svrevd_x(svbool_t, svfloat32_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrevd_f16_x)))
-svfloat16_t svrevd_x(svbool_t, svfloat16_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrevd_s32_x)))
-svint32_t svrevd_x(svbool_t, svint32_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrevd_s64_x)))
-svint64_t svrevd_x(svbool_t, svint64_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrevd_s16_x)))
-svint16_t svrevd_x(svbool_t, svint16_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrevd_u8_z)))
-svuint8_t svrevd_z(svbool_t, svuint8_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrevd_u32_z)))
-svuint32_t svrevd_z(svbool_t, svuint32_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrevd_u64_z)))
-svuint64_t svrevd_z(svbool_t, svuint64_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrevd_u16_z)))
-svuint16_t svrevd_z(svbool_t, svuint16_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrevd_bf16_z)))
-svbfloat16_t svrevd_z(svbool_t, svbfloat16_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrevd_s8_z)))
-svint8_t svrevd_z(svbool_t, svint8_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrevd_f64_z)))
-svfloat64_t svrevd_z(svbool_t, svfloat64_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrevd_f32_z)))
-svfloat32_t svrevd_z(svbool_t, svfloat32_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrevd_f16_z)))
-svfloat16_t svrevd_z(svbool_t, svfloat16_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrevd_s32_z)))
-svint32_t svrevd_z(svbool_t, svint32_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrevd_s64_z)))
-svint64_t svrevd_z(svbool_t, svint64_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrevd_s16_z)))
-svint16_t svrevd_z(svbool_t, svint16_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbfmlslb_f32)))
-svfloat32_t svbfmlslb_f32(svfloat32_t, svbfloat16_t, svbfloat16_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbfmlslb_lane_f32)))
-svfloat32_t svbfmlslb_lane_f32(svfloat32_t, svbfloat16_t, svbfloat16_t, uint64_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbfmlslt_f32)))
-svfloat32_t svbfmlslt_f32(svfloat32_t, svbfloat16_t, svbfloat16_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbfmlslt_lane_f32)))
-svfloat32_t svbfmlslt_lane_f32(svfloat32_t, svbfloat16_t, svbfloat16_t, uint64_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svclamp_f64)))
-svfloat64_t svclamp_f64(svfloat64_t, svfloat64_t, svfloat64_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svclamp_f32)))
-svfloat32_t svclamp_f32(svfloat32_t, svfloat32_t, svfloat32_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svclamp_f16)))
-svfloat16_t svclamp_f16(svfloat16_t, svfloat16_t, svfloat16_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcntp_c8)))
 uint64_t svcntp_c8(svcount_t, uint64_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcntp_c32)))
@@ -11391,30 +11533,12 @@ __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcntp_c64)))
 uint64_t svcntp_c64(svcount_t, uint64_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcntp_c16)))
 uint64_t svcntp_c16(svcount_t, uint64_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcreate2_b)))
-svboolx2_t svcreate2_b(svbool_t, svbool_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcreate4_b)))
-svboolx4_t svcreate4_b(svbool_t, svbool_t, svbool_t, svbool_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdot_f32_f16)))
-svfloat32_t svdot_f32_f16(svfloat32_t, svfloat16_t, svfloat16_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdot_s32_s16)))
-svint32_t svdot_s32_s16(svint32_t, svint16_t, svint16_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdot_u32_u16)))
-svuint32_t svdot_u32_u16(svuint32_t, svuint16_t, svuint16_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdot_lane_f32_f16)))
-svfloat32_t svdot_lane_f32_f16(svfloat32_t, svfloat16_t, svfloat16_t, uint64_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdot_lane_s32_s16)))
-svint32_t svdot_lane_s32_s16(svint32_t, svint16_t, svint16_t, uint64_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdot_lane_u32_u16)))
-svuint32_t svdot_lane_u32_u16(svuint32_t, svuint16_t, svuint16_t, uint64_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svget2_b)))
-svbool_t svget2_b(svboolx2_t, uint64_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svget4_b)))
-svbool_t svget4_b(svboolx4_t, uint64_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1_u8_x2)))
 svuint8x2_t svld1_u8_x2(svcount_t, uint8_t const *);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1_s8_x2)))
 svint8x2_t svld1_s8_x2(svcount_t, int8_t const *);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1_mf8_x2)))
+svmfloat8x2_t svld1_mf8_x2(svcount_t, mfloat8_t const *);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1_u64_x2)))
 svuint64x2_t svld1_u64_x2(svcount_t, uint64_t const *);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1_f64_x2)))
@@ -11439,6 +11563,8 @@ __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1_u8_x4)))
 svuint8x4_t svld1_u8_x4(svcount_t, uint8_t const *);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1_s8_x4)))
 svint8x4_t svld1_s8_x4(svcount_t, int8_t const *);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1_mf8_x4)))
+svmfloat8x4_t svld1_mf8_x4(svcount_t, mfloat8_t const *);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1_u64_x4)))
 svuint64x4_t svld1_u64_x4(svcount_t, uint64_t const *);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1_f64_x4)))
@@ -11463,6 +11589,8 @@ __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1_vnum_u8_x2)))
 svuint8x2_t svld1_vnum_u8_x2(svcount_t, uint8_t const *, int64_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1_vnum_s8_x2)))
 svint8x2_t svld1_vnum_s8_x2(svcount_t, int8_t const *, int64_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1_vnum_mf8_x2)))
+svmfloat8x2_t svld1_vnum_mf8_x2(svcount_t, mfloat8_t const *, int64_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1_vnum_u64_x2)))
 svuint64x2_t svld1_vnum_u64_x2(svcount_t, uint64_t const *, int64_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1_vnum_f64_x2)))
@@ -11487,6 +11615,8 @@ __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1_vnum_u8_x4)))
 svuint8x4_t svld1_vnum_u8_x4(svcount_t, uint8_t const *, int64_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1_vnum_s8_x4)))
 svint8x4_t svld1_vnum_s8_x4(svcount_t, int8_t const *, int64_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1_vnum_mf8_x4)))
+svmfloat8x4_t svld1_vnum_mf8_x4(svcount_t, mfloat8_t const *, int64_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1_vnum_u64_x4)))
 svuint64x4_t svld1_vnum_u64_x4(svcount_t, uint64_t const *, int64_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1_vnum_f64_x4)))
@@ -11511,6 +11641,8 @@ __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_u8_x2)))
 svuint8x2_t svldnt1_u8_x2(svcount_t, uint8_t const *);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_s8_x2)))
 svint8x2_t svldnt1_s8_x2(svcount_t, int8_t const *);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_mf8_x2)))
+svmfloat8x2_t svldnt1_mf8_x2(svcount_t, mfloat8_t const *);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_u64_x2)))
 svuint64x2_t svldnt1_u64_x2(svcount_t, uint64_t const *);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_f64_x2)))
@@ -11535,6 +11667,8 @@ __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_u8_x4)))
 svuint8x4_t svldnt1_u8_x4(svcount_t, uint8_t const *);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_s8_x4)))
 svint8x4_t svldnt1_s8_x4(svcount_t, int8_t const *);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_mf8_x4)))
+svmfloat8x4_t svldnt1_mf8_x4(svcount_t, mfloat8_t const *);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_u64_x4)))
 svuint64x4_t svldnt1_u64_x4(svcount_t, uint64_t const *);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_f64_x4)))
@@ -11559,6 +11693,8 @@ __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_vnum_u8_x2))
 svuint8x2_t svldnt1_vnum_u8_x2(svcount_t, uint8_t const *, int64_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_vnum_s8_x2)))
 svint8x2_t svldnt1_vnum_s8_x2(svcount_t, int8_t const *, int64_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_vnum_mf8_x2)))
+svmfloat8x2_t svldnt1_vnum_mf8_x2(svcount_t, mfloat8_t const *, int64_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_vnum_u64_x2)))
 svuint64x2_t svldnt1_vnum_u64_x2(svcount_t, uint64_t const *, int64_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_vnum_f64_x2)))
@@ -11583,6 +11719,8 @@ __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_vnum_u8_x4))
 svuint8x4_t svldnt1_vnum_u8_x4(svcount_t, uint8_t const *, int64_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_vnum_s8_x4)))
 svint8x4_t svldnt1_vnum_s8_x4(svcount_t, int8_t const *, int64_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_vnum_mf8_x4)))
+svmfloat8x4_t svldnt1_vnum_mf8_x4(svcount_t, mfloat8_t const *, int64_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_vnum_u64_x4)))
 svuint64x4_t svldnt1_vnum_u64_x4(svcount_t, uint64_t const *, int64_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_vnum_f64_x4)))
@@ -11637,30 +11775,16 @@ __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svptrue_c64)))
 svcount_t svptrue_c64(void);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svptrue_c16)))
 svcount_t svptrue_c16(void);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqcvtn_s16_s32_x2)))
-svint16_t svqcvtn_s16_s32_x2(svint32x2_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqcvtn_u16_s32_x2)))
-svuint16_t svqcvtn_u16_s32_x2(svint32x2_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqcvtn_u16_u32_x2)))
-svuint16_t svqcvtn_u16_u32_x2(svuint32x2_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshrn_n_s16_s32_x2)))
-svint16_t svqrshrn_n_s16_s32_x2(svint32x2_t, uint64_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshrn_n_u16_u32_x2)))
-svuint16_t svqrshrn_n_u16_u32_x2(svuint32x2_t, uint64_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshrun_n_u16_s32_x2)))
-svuint16_t svqrshrun_n_u16_s32_x2(svint32x2_t, uint64_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svreinterpret_b)))
 svbool_t svreinterpret_b(svcount_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svreinterpret_c)))
 svcount_t svreinterpret_c(svbool_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svset2_b)))
-svboolx2_t svset2_b(svboolx2_t, uint64_t, svbool_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svset4_b)))
-svboolx4_t svset4_b(svboolx4_t, uint64_t, svbool_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1_u8_x2)))
 void svst1_u8_x2(svcount_t, uint8_t *, svuint8x2_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1_s8_x2)))
 void svst1_s8_x2(svcount_t, int8_t *, svint8x2_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1_mf8_x2)))
+void svst1_mf8_x2(svcount_t, mfloat8_t *, svmfloat8x2_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1_u64_x2)))
 void svst1_u64_x2(svcount_t, uint64_t *, svuint64x2_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1_f64_x2)))
@@ -11685,6 +11809,8 @@ __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1_u8_x4)))
 void svst1_u8_x4(svcount_t, uint8_t *, svuint8x4_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1_s8_x4)))
 void svst1_s8_x4(svcount_t, int8_t *, svint8x4_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1_mf8_x4)))
+void svst1_mf8_x4(svcount_t, mfloat8_t *, svmfloat8x4_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1_u64_x4)))
 void svst1_u64_x4(svcount_t, uint64_t *, svuint64x4_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1_f64_x4)))
@@ -11709,6 +11835,8 @@ __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1_vnum_u8_x2)))
 void svst1_vnum_u8_x2(svcount_t, uint8_t *, int64_t, svuint8x2_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1_vnum_s8_x2)))
 void svst1_vnum_s8_x2(svcount_t, int8_t *, int64_t, svint8x2_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1_vnum_mf8_x2)))
+void svst1_vnum_mf8_x2(svcount_t, mfloat8_t *, int64_t, svmfloat8x2_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1_vnum_u64_x2)))
 void svst1_vnum_u64_x2(svcount_t, uint64_t *, int64_t, svuint64x2_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1_vnum_f64_x2)))
@@ -11733,6 +11861,8 @@ __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1_vnum_u8_x4)))
 void svst1_vnum_u8_x4(svcount_t, uint8_t *, int64_t, svuint8x4_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1_vnum_s8_x4)))
 void svst1_vnum_s8_x4(svcount_t, int8_t *, int64_t, svint8x4_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1_vnum_mf8_x4)))
+void svst1_vnum_mf8_x4(svcount_t, mfloat8_t *, int64_t, svmfloat8x4_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1_vnum_u64_x4)))
 void svst1_vnum_u64_x4(svcount_t, uint64_t *, int64_t, svuint64x4_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1_vnum_f64_x4)))
@@ -11757,6 +11887,8 @@ __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_u8_x2)))
 void svstnt1_u8_x2(svcount_t, uint8_t *, svuint8x2_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_s8_x2)))
 void svstnt1_s8_x2(svcount_t, int8_t *, svint8x2_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_mf8_x2)))
+void svstnt1_mf8_x2(svcount_t, mfloat8_t *, svmfloat8x2_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_u64_x2)))
 void svstnt1_u64_x2(svcount_t, uint64_t *, svuint64x2_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_f64_x2)))
@@ -11781,6 +11913,8 @@ __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_u8_x4)))
 void svstnt1_u8_x4(svcount_t, uint8_t *, svuint8x4_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_s8_x4)))
 void svstnt1_s8_x4(svcount_t, int8_t *, svint8x4_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_mf8_x4)))
+void svstnt1_mf8_x4(svcount_t, mfloat8_t *, svmfloat8x4_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_u64_x4)))
 void svstnt1_u64_x4(svcount_t, uint64_t *, svuint64x4_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_f64_x4)))
@@ -11805,6 +11939,8 @@ __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_vnum_u8_x2))
 void svstnt1_vnum_u8_x2(svcount_t, uint8_t *, int64_t, svuint8x2_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_vnum_s8_x2)))
 void svstnt1_vnum_s8_x2(svcount_t, int8_t *, int64_t, svint8x2_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_vnum_mf8_x2)))
+void svstnt1_vnum_mf8_x2(svcount_t, mfloat8_t *, int64_t, svmfloat8x2_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_vnum_u64_x2)))
 void svstnt1_vnum_u64_x2(svcount_t, uint64_t *, int64_t, svuint64x2_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_vnum_f64_x2)))
@@ -11829,6 +11965,8 @@ __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_vnum_u8_x4))
 void svstnt1_vnum_u8_x4(svcount_t, uint8_t *, int64_t, svuint8x4_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_vnum_s8_x4)))
 void svstnt1_vnum_s8_x4(svcount_t, int8_t *, int64_t, svint8x4_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_vnum_mf8_x4)))
+void svstnt1_vnum_mf8_x4(svcount_t, mfloat8_t *, int64_t, svmfloat8x4_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_vnum_u64_x4)))
 void svstnt1_vnum_u64_x4(svcount_t, uint64_t *, int64_t, svuint64x4_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_vnum_f64_x4)))
@@ -11849,10 +11987,6 @@ __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_vnum_f32_x4)
 void svstnt1_vnum_f32_x4(svcount_t, float32_t *, int64_t, svfloat32x4_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_vnum_s32_x4)))
 void svstnt1_vnum_s32_x4(svcount_t, int32_t *, int64_t, svint32x4_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svundef2_b)))
-svboolx2_t svundef2_b();
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svundef4_b)))
-svboolx4_t svundef4_b();
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilege_c8_s64)))
 svcount_t svwhilege_c8_s64(int64_t, int64_t, uint64_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilege_c32_s64)))
@@ -11869,22 +12003,6 @@ __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilege_c64_u64)))
 svcount_t svwhilege_c64_u64(uint64_t, uint64_t, uint64_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilege_c16_u64)))
 svcount_t svwhilege_c16_u64(uint64_t, uint64_t, uint64_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilege_b8_s64_x2)))
-svboolx2_t svwhilege_b8_s64_x2(int64_t, int64_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilege_b32_s64_x2)))
-svboolx2_t svwhilege_b32_s64_x2(int64_t, int64_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilege_b64_s64_x2)))
-svboolx2_t svwhilege_b64_s64_x2(int64_t, int64_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilege_b16_s64_x2)))
-svboolx2_t svwhilege_b16_s64_x2(int64_t, int64_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilege_b8_u64_x2)))
-svboolx2_t svwhilege_b8_u64_x2(uint64_t, uint64_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilege_b32_u64_x2)))
-svboolx2_t svwhilege_b32_u64_x2(uint64_t, uint64_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilege_b64_u64_x2)))
-svboolx2_t svwhilege_b64_u64_x2(uint64_t, uint64_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilege_b16_u64_x2)))
-svboolx2_t svwhilege_b16_u64_x2(uint64_t, uint64_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilegt_c8_s64)))
 svcount_t svwhilegt_c8_s64(int64_t, int64_t, uint64_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilegt_c32_s64)))
@@ -11901,22 +12019,6 @@ __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilegt_c64_u64)))
 svcount_t svwhilegt_c64_u64(uint64_t, uint64_t, uint64_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilegt_c16_u64)))
 svcount_t svwhilegt_c16_u64(uint64_t, uint64_t, uint64_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilegt_b8_s64_x2)))
-svboolx2_t svwhilegt_b8_s64_x2(int64_t, int64_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilegt_b32_s64_x2)))
-svboolx2_t svwhilegt_b32_s64_x2(int64_t, int64_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilegt_b64_s64_x2)))
-svboolx2_t svwhilegt_b64_s64_x2(int64_t, int64_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilegt_b16_s64_x2)))
-svboolx2_t svwhilegt_b16_s64_x2(int64_t, int64_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilegt_b8_u64_x2)))
-svboolx2_t svwhilegt_b8_u64_x2(uint64_t, uint64_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilegt_b32_u64_x2)))
-svboolx2_t svwhilegt_b32_u64_x2(uint64_t, uint64_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilegt_b64_u64_x2)))
-svboolx2_t svwhilegt_b64_u64_x2(uint64_t, uint64_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilegt_b16_u64_x2)))
-svboolx2_t svwhilegt_b16_u64_x2(uint64_t, uint64_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilele_c8_s64)))
 svcount_t svwhilele_c8_s64(int64_t, int64_t, uint64_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilele_c32_s64)))
@@ -11933,22 +12035,6 @@ __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilele_c64_u64)))
 svcount_t svwhilele_c64_u64(uint64_t, uint64_t, uint64_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilele_c16_u64)))
 svcount_t svwhilele_c16_u64(uint64_t, uint64_t, uint64_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilele_b8_s64_x2)))
-svboolx2_t svwhilele_b8_s64_x2(int64_t, int64_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilele_b32_s64_x2)))
-svboolx2_t svwhilele_b32_s64_x2(int64_t, int64_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilele_b64_s64_x2)))
-svboolx2_t svwhilele_b64_s64_x2(int64_t, int64_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilele_b16_s64_x2)))
-svboolx2_t svwhilele_b16_s64_x2(int64_t, int64_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilele_b8_u64_x2)))
-svboolx2_t svwhilele_b8_u64_x2(uint64_t, uint64_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilele_b32_u64_x2)))
-svboolx2_t svwhilele_b32_u64_x2(uint64_t, uint64_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilele_b64_u64_x2)))
-svboolx2_t svwhilele_b64_u64_x2(uint64_t, uint64_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilele_b16_u64_x2)))
-svboolx2_t svwhilele_b16_u64_x2(uint64_t, uint64_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilelt_c8_u64)))
 svcount_t svwhilelt_c8_u64(uint64_t, uint64_t, uint64_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilelt_c32_u64)))
@@ -11965,60 +12051,12 @@ __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilelt_c64_s64)))
 svcount_t svwhilelt_c64_s64(int64_t, int64_t, uint64_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilelt_c16_s64)))
 svcount_t svwhilelt_c16_s64(int64_t, int64_t, uint64_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilelt_b8_u64_x2)))
-svboolx2_t svwhilelt_b8_u64_x2(uint64_t, uint64_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilelt_b32_u64_x2)))
-svboolx2_t svwhilelt_b32_u64_x2(uint64_t, uint64_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilelt_b64_u64_x2)))
-svboolx2_t svwhilelt_b64_u64_x2(uint64_t, uint64_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilelt_b16_u64_x2)))
-svboolx2_t svwhilelt_b16_u64_x2(uint64_t, uint64_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilelt_b8_s64_x2)))
-svboolx2_t svwhilelt_b8_s64_x2(int64_t, int64_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilelt_b32_s64_x2)))
-svboolx2_t svwhilelt_b32_s64_x2(int64_t, int64_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilelt_b64_s64_x2)))
-svboolx2_t svwhilelt_b64_s64_x2(int64_t, int64_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilelt_b16_s64_x2)))
-svboolx2_t svwhilelt_b16_s64_x2(int64_t, int64_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbfmlslb_f32)))
-svfloat32_t svbfmlslb(svfloat32_t, svbfloat16_t, svbfloat16_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbfmlslb_lane_f32)))
-svfloat32_t svbfmlslb_lane(svfloat32_t, svbfloat16_t, svbfloat16_t, uint64_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbfmlslt_f32)))
-svfloat32_t svbfmlslt(svfloat32_t, svbfloat16_t, svbfloat16_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svbfmlslt_lane_f32)))
-svfloat32_t svbfmlslt_lane(svfloat32_t, svbfloat16_t, svbfloat16_t, uint64_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svclamp_f64)))
-svfloat64_t svclamp(svfloat64_t, svfloat64_t, svfloat64_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svclamp_f32)))
-svfloat32_t svclamp(svfloat32_t, svfloat32_t, svfloat32_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svclamp_f16)))
-svfloat16_t svclamp(svfloat16_t, svfloat16_t, svfloat16_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcreate2_b)))
-svboolx2_t svcreate2(svbool_t, svbool_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcreate4_b)))
-svboolx4_t svcreate4(svbool_t, svbool_t, svbool_t, svbool_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdot_f32_f16)))
-svfloat32_t svdot(svfloat32_t, svfloat16_t, svfloat16_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdot_s32_s16)))
-svint32_t svdot(svint32_t, svint16_t, svint16_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdot_u32_u16)))
-svuint32_t svdot(svuint32_t, svuint16_t, svuint16_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdot_lane_f32_f16)))
-svfloat32_t svdot_lane(svfloat32_t, svfloat16_t, svfloat16_t, uint64_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdot_lane_s32_s16)))
-svint32_t svdot_lane(svint32_t, svint16_t, svint16_t, uint64_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdot_lane_u32_u16)))
-svuint32_t svdot_lane(svuint32_t, svuint16_t, svuint16_t, uint64_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svget2_b)))
-svbool_t svget2(svboolx2_t, uint64_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svget4_b)))
-svbool_t svget4(svboolx4_t, uint64_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1_u8_x2)))
 svuint8x2_t svld1_x2(svcount_t, uint8_t const *);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1_s8_x2)))
 svint8x2_t svld1_x2(svcount_t, int8_t const *);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1_mf8_x2)))
+svmfloat8x2_t svld1_x2(svcount_t, mfloat8_t const *);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1_u64_x2)))
 svuint64x2_t svld1_x2(svcount_t, uint64_t const *);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1_f64_x2)))
@@ -12043,6 +12081,8 @@ __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1_u8_x4)))
 svuint8x4_t svld1_x4(svcount_t, uint8_t const *);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1_s8_x4)))
 svint8x4_t svld1_x4(svcount_t, int8_t const *);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1_mf8_x4)))
+svmfloat8x4_t svld1_x4(svcount_t, mfloat8_t const *);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1_u64_x4)))
 svuint64x4_t svld1_x4(svcount_t, uint64_t const *);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1_f64_x4)))
@@ -12067,6 +12107,8 @@ __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1_vnum_u8_x2)))
 svuint8x2_t svld1_vnum_x2(svcount_t, uint8_t const *, int64_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1_vnum_s8_x2)))
 svint8x2_t svld1_vnum_x2(svcount_t, int8_t const *, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1_vnum_mf8_x2)))
+svmfloat8x2_t svld1_vnum_x2(svcount_t, mfloat8_t const *, int64_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1_vnum_u64_x2)))
 svuint64x2_t svld1_vnum_x2(svcount_t, uint64_t const *, int64_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1_vnum_f64_x2)))
@@ -12091,6 +12133,8 @@ __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1_vnum_u8_x4)))
 svuint8x4_t svld1_vnum_x4(svcount_t, uint8_t const *, int64_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1_vnum_s8_x4)))
 svint8x4_t svld1_vnum_x4(svcount_t, int8_t const *, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1_vnum_mf8_x4)))
+svmfloat8x4_t svld1_vnum_x4(svcount_t, mfloat8_t const *, int64_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1_vnum_u64_x4)))
 svuint64x4_t svld1_vnum_x4(svcount_t, uint64_t const *, int64_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1_vnum_f64_x4)))
@@ -12115,6 +12159,8 @@ __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_u8_x2)))
 svuint8x2_t svldnt1_x2(svcount_t, uint8_t const *);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_s8_x2)))
 svint8x2_t svldnt1_x2(svcount_t, int8_t const *);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_mf8_x2)))
+svmfloat8x2_t svldnt1_x2(svcount_t, mfloat8_t const *);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_u64_x2)))
 svuint64x2_t svldnt1_x2(svcount_t, uint64_t const *);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_f64_x2)))
@@ -12139,6 +12185,8 @@ __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_u8_x4)))
 svuint8x4_t svldnt1_x4(svcount_t, uint8_t const *);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_s8_x4)))
 svint8x4_t svldnt1_x4(svcount_t, int8_t const *);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_mf8_x4)))
+svmfloat8x4_t svldnt1_x4(svcount_t, mfloat8_t const *);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_u64_x4)))
 svuint64x4_t svldnt1_x4(svcount_t, uint64_t const *);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_f64_x4)))
@@ -12163,6 +12211,8 @@ __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_vnum_u8_x2)
 svuint8x2_t svldnt1_vnum_x2(svcount_t, uint8_t const *, int64_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_vnum_s8_x2)))
 svint8x2_t svldnt1_vnum_x2(svcount_t, int8_t const *, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_vnum_mf8_x2)))
+svmfloat8x2_t svldnt1_vnum_x2(svcount_t, mfloat8_t const *, int64_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_vnum_u64_x2)))
 svuint64x2_t svldnt1_vnum_x2(svcount_t, uint64_t const *, int64_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_vnum_f64_x2)))
@@ -12187,6 +12237,8 @@ __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_vnum_u8_x4)
 svuint8x4_t svldnt1_vnum_x4(svcount_t, uint8_t const *, int64_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_vnum_s8_x4)))
 svint8x4_t svldnt1_vnum_x4(svcount_t, int8_t const *, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_vnum_mf8_x4)))
+svmfloat8x4_t svldnt1_vnum_x4(svcount_t, mfloat8_t const *, int64_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_vnum_u64_x4)))
 svuint64x4_t svldnt1_vnum_x4(svcount_t, uint64_t const *, int64_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_vnum_f64_x4)))
@@ -12207,30 +12259,16 @@ __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_vnum_f32_x4
 svfloat32x4_t svldnt1_vnum_x4(svcount_t, float32_t const *, int64_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_vnum_s32_x4)))
 svint32x4_t svldnt1_vnum_x4(svcount_t, int32_t const *, int64_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqcvtn_s16_s32_x2)))
-svint16_t svqcvtn_s16(svint32x2_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqcvtn_u16_s32_x2)))
-svuint16_t svqcvtn_u16(svint32x2_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqcvtn_u16_u32_x2)))
-svuint16_t svqcvtn_u16(svuint32x2_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshrn_n_s16_s32_x2)))
-svint16_t svqrshrn_s16(svint32x2_t, uint64_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshrn_n_u16_u32_x2)))
-svuint16_t svqrshrn_u16(svuint32x2_t, uint64_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svqrshrun_n_u16_s32_x2)))
-svuint16_t svqrshrun_u16(svint32x2_t, uint64_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svreinterpret_b)))
 svbool_t svreinterpret(svcount_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svreinterpret_c)))
 svcount_t svreinterpret(svbool_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svset2_b)))
-svboolx2_t svset2(svboolx2_t, uint64_t, svbool_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svset4_b)))
-svboolx4_t svset4(svboolx4_t, uint64_t, svbool_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1_u8_x2)))
 void svst1(svcount_t, uint8_t *, svuint8x2_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1_s8_x2)))
 void svst1(svcount_t, int8_t *, svint8x2_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1_mf8_x2)))
+void svst1(svcount_t, mfloat8_t *, svmfloat8x2_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1_u64_x2)))
 void svst1(svcount_t, uint64_t *, svuint64x2_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1_f64_x2)))
@@ -12255,6 +12293,8 @@ __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1_u8_x4)))
 void svst1(svcount_t, uint8_t *, svuint8x4_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1_s8_x4)))
 void svst1(svcount_t, int8_t *, svint8x4_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1_mf8_x4)))
+void svst1(svcount_t, mfloat8_t *, svmfloat8x4_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1_u64_x4)))
 void svst1(svcount_t, uint64_t *, svuint64x4_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1_f64_x4)))
@@ -12279,6 +12319,8 @@ __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1_vnum_u8_x2)))
 void svst1_vnum(svcount_t, uint8_t *, int64_t, svuint8x2_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1_vnum_s8_x2)))
 void svst1_vnum(svcount_t, int8_t *, int64_t, svint8x2_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1_vnum_mf8_x2)))
+void svst1_vnum(svcount_t, mfloat8_t *, int64_t, svmfloat8x2_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1_vnum_u64_x2)))
 void svst1_vnum(svcount_t, uint64_t *, int64_t, svuint64x2_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1_vnum_f64_x2)))
@@ -12303,6 +12345,8 @@ __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1_vnum_u8_x4)))
 void svst1_vnum(svcount_t, uint8_t *, int64_t, svuint8x4_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1_vnum_s8_x4)))
 void svst1_vnum(svcount_t, int8_t *, int64_t, svint8x4_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1_vnum_mf8_x4)))
+void svst1_vnum(svcount_t, mfloat8_t *, int64_t, svmfloat8x4_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1_vnum_u64_x4)))
 void svst1_vnum(svcount_t, uint64_t *, int64_t, svuint64x4_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1_vnum_f64_x4)))
@@ -12327,6 +12371,8 @@ __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_u8_x2)))
 void svstnt1(svcount_t, uint8_t *, svuint8x2_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_s8_x2)))
 void svstnt1(svcount_t, int8_t *, svint8x2_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_mf8_x2)))
+void svstnt1(svcount_t, mfloat8_t *, svmfloat8x2_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_u64_x2)))
 void svstnt1(svcount_t, uint64_t *, svuint64x2_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_f64_x2)))
@@ -12351,6 +12397,8 @@ __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_u8_x4)))
 void svstnt1(svcount_t, uint8_t *, svuint8x4_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_s8_x4)))
 void svstnt1(svcount_t, int8_t *, svint8x4_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_mf8_x4)))
+void svstnt1(svcount_t, mfloat8_t *, svmfloat8x4_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_u64_x4)))
 void svstnt1(svcount_t, uint64_t *, svuint64x4_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_f64_x4)))
@@ -12375,6 +12423,8 @@ __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_vnum_u8_x2)
 void svstnt1_vnum(svcount_t, uint8_t *, int64_t, svuint8x2_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_vnum_s8_x2)))
 void svstnt1_vnum(svcount_t, int8_t *, int64_t, svint8x2_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_vnum_mf8_x2)))
+void svstnt1_vnum(svcount_t, mfloat8_t *, int64_t, svmfloat8x2_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_vnum_u64_x2)))
 void svstnt1_vnum(svcount_t, uint64_t *, int64_t, svuint64x2_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_vnum_f64_x2)))
@@ -12399,6 +12449,8 @@ __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_vnum_u8_x4)
 void svstnt1_vnum(svcount_t, uint8_t *, int64_t, svuint8x4_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_vnum_s8_x4)))
 void svstnt1_vnum(svcount_t, int8_t *, int64_t, svint8x4_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_vnum_mf8_x4)))
+void svstnt1_vnum(svcount_t, mfloat8_t *, int64_t, svmfloat8x4_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_vnum_u64_x4)))
 void svstnt1_vnum(svcount_t, uint64_t *, int64_t, svuint64x4_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_vnum_f64_x4)))
@@ -12435,22 +12487,6 @@ __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilege_c64_u64))
 svcount_t svwhilege_c64(uint64_t, uint64_t, uint64_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilege_c16_u64)))
 svcount_t svwhilege_c16(uint64_t, uint64_t, uint64_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilege_b8_s64_x2)))
-svboolx2_t svwhilege_b8_x2(int64_t, int64_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilege_b32_s64_x2)))
-svboolx2_t svwhilege_b32_x2(int64_t, int64_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilege_b64_s64_x2)))
-svboolx2_t svwhilege_b64_x2(int64_t, int64_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilege_b16_s64_x2)))
-svboolx2_t svwhilege_b16_x2(int64_t, int64_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilege_b8_u64_x2)))
-svboolx2_t svwhilege_b8_x2(uint64_t, uint64_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilege_b32_u64_x2)))
-svboolx2_t svwhilege_b32_x2(uint64_t, uint64_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilege_b64_u64_x2)))
-svboolx2_t svwhilege_b64_x2(uint64_t, uint64_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilege_b16_u64_x2)))
-svboolx2_t svwhilege_b16_x2(uint64_t, uint64_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilegt_c8_s64)))
 svcount_t svwhilegt_c8(int64_t, int64_t, uint64_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilegt_c32_s64)))
@@ -12467,22 +12503,6 @@ __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilegt_c64_u64))
 svcount_t svwhilegt_c64(uint64_t, uint64_t, uint64_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilegt_c16_u64)))
 svcount_t svwhilegt_c16(uint64_t, uint64_t, uint64_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilegt_b8_s64_x2)))
-svboolx2_t svwhilegt_b8_x2(int64_t, int64_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilegt_b32_s64_x2)))
-svboolx2_t svwhilegt_b32_x2(int64_t, int64_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilegt_b64_s64_x2)))
-svboolx2_t svwhilegt_b64_x2(int64_t, int64_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilegt_b16_s64_x2)))
-svboolx2_t svwhilegt_b16_x2(int64_t, int64_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilegt_b8_u64_x2)))
-svboolx2_t svwhilegt_b8_x2(uint64_t, uint64_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilegt_b32_u64_x2)))
-svboolx2_t svwhilegt_b32_x2(uint64_t, uint64_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilegt_b64_u64_x2)))
-svboolx2_t svwhilegt_b64_x2(uint64_t, uint64_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilegt_b16_u64_x2)))
-svboolx2_t svwhilegt_b16_x2(uint64_t, uint64_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilele_c8_s64)))
 svcount_t svwhilele_c8(int64_t, int64_t, uint64_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilele_c32_s64)))
@@ -12499,22 +12519,6 @@ __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilele_c64_u64))
 svcount_t svwhilele_c64(uint64_t, uint64_t, uint64_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilele_c16_u64)))
 svcount_t svwhilele_c16(uint64_t, uint64_t, uint64_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilele_b8_s64_x2)))
-svboolx2_t svwhilele_b8_x2(int64_t, int64_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilele_b32_s64_x2)))
-svboolx2_t svwhilele_b32_x2(int64_t, int64_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilele_b64_s64_x2)))
-svboolx2_t svwhilele_b64_x2(int64_t, int64_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilele_b16_s64_x2)))
-svboolx2_t svwhilele_b16_x2(int64_t, int64_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilele_b8_u64_x2)))
-svboolx2_t svwhilele_b8_x2(uint64_t, uint64_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilele_b32_u64_x2)))
-svboolx2_t svwhilele_b32_x2(uint64_t, uint64_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilele_b64_u64_x2)))
-svboolx2_t svwhilele_b64_x2(uint64_t, uint64_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilele_b16_u64_x2)))
-svboolx2_t svwhilele_b16_x2(uint64_t, uint64_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilelt_c8_u64)))
 svcount_t svwhilelt_c8(uint64_t, uint64_t, uint64_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilelt_c32_u64)))
@@ -12531,70 +12535,6 @@ __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilelt_c64_s64))
 svcount_t svwhilelt_c64(int64_t, int64_t, uint64_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilelt_c16_s64)))
 svcount_t svwhilelt_c16(int64_t, int64_t, uint64_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilelt_b8_u64_x2)))
-svboolx2_t svwhilelt_b8_x2(uint64_t, uint64_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilelt_b32_u64_x2)))
-svboolx2_t svwhilelt_b32_x2(uint64_t, uint64_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilelt_b64_u64_x2)))
-svboolx2_t svwhilelt_b64_x2(uint64_t, uint64_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilelt_b16_u64_x2)))
-svboolx2_t svwhilelt_b16_x2(uint64_t, uint64_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilelt_b8_s64_x2)))
-svboolx2_t svwhilelt_b8_x2(int64_t, int64_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilelt_b32_s64_x2)))
-svboolx2_t svwhilelt_b32_x2(int64_t, int64_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilelt_b64_s64_x2)))
-svboolx2_t svwhilelt_b64_x2(int64_t, int64_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilelt_b16_s64_x2)))
-svboolx2_t svwhilelt_b16_x2(int64_t, int64_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdup_laneq_u8)))
-svuint8_t svdup_laneq_u8(svuint8_t, uint64_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdup_laneq_s8)))
-svint8_t svdup_laneq_s8(svint8_t, uint64_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdup_laneq_mf8)))
-svmfloat8_t svdup_laneq_mf8(svmfloat8_t, uint64_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdup_laneq_u64)))
-svuint64_t svdup_laneq_u64(svuint64_t, uint64_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdup_laneq_f64)))
-svfloat64_t svdup_laneq_f64(svfloat64_t, uint64_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdup_laneq_s64)))
-svint64_t svdup_laneq_s64(svint64_t, uint64_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdup_laneq_u16)))
-svuint16_t svdup_laneq_u16(svuint16_t, uint64_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdup_laneq_f16)))
-svfloat16_t svdup_laneq_f16(svfloat16_t, uint64_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdup_laneq_s16)))
-svint16_t svdup_laneq_s16(svint16_t, uint64_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdup_laneq_u32)))
-svuint32_t svdup_laneq_u32(svuint32_t, uint64_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdup_laneq_f32)))
-svfloat32_t svdup_laneq_f32(svfloat32_t, uint64_t);
-__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdup_laneq_s32)))
-svint32_t svdup_laneq_s32(svint32_t, uint64_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdup_laneq_u8)))
-svuint8_t svdup_laneq(svuint8_t, uint64_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdup_laneq_s8)))
-svint8_t svdup_laneq(svint8_t, uint64_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdup_laneq_mf8)))
-svmfloat8_t svdup_laneq(svmfloat8_t, uint64_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdup_laneq_u64)))
-svuint64_t svdup_laneq(svuint64_t, uint64_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdup_laneq_f64)))
-svfloat64_t svdup_laneq(svfloat64_t, uint64_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdup_laneq_s64)))
-svint64_t svdup_laneq(svint64_t, uint64_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdup_laneq_u16)))
-svuint16_t svdup_laneq(svuint16_t, uint64_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdup_laneq_f16)))
-svfloat16_t svdup_laneq(svfloat16_t, uint64_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdup_laneq_s16)))
-svint16_t svdup_laneq(svint16_t, uint64_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdup_laneq_u32)))
-svuint32_t svdup_laneq(svuint32_t, uint64_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdup_laneq_f32)))
-svfloat32_t svdup_laneq(svfloat32_t, uint64_t);
-__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdup_laneq_s32)))
-svint32_t svdup_laneq(svint32_t, uint64_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaba_n_s8)))
 svint8_t svaba_n_s8(svint8_t, svint8_t, int8_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svaba_n_s32)))
@@ -15641,6 +15581,8 @@ __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtbl2_u64)))
 svuint64_t svtbl2_u64(svuint64x2_t, svuint64_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtbl2_u16)))
 svuint16_t svtbl2_u16(svuint16x2_t, svuint16_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtbl2_bf16)))
+svbfloat16_t svtbl2_bf16(svbfloat16x2_t, svuint16_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtbl2_s8)))
 svint8_t svtbl2_s8(svint8x2_t, svuint8_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtbl2_f64)))
@@ -15663,6 +15605,8 @@ __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtbx_u64)))
 svuint64_t svtbx_u64(svuint64_t, svuint64_t, svuint64_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtbx_u16)))
 svuint16_t svtbx_u16(svuint16_t, svuint16_t, svuint16_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtbx_bf16)))
+svbfloat16_t svtbx_bf16(svbfloat16_t, svbfloat16_t, svuint16_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtbx_s8)))
 svint8_t svtbx_s8(svint8_t, svint8_t, svuint8_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtbx_f64)))
@@ -15801,6 +15745,8 @@ __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilerw_s64)))
 svbool_t svwhilerw_s64(int64_t const *, int64_t const *);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilerw_u16)))
 svbool_t svwhilerw_u16(uint16_t const *, uint16_t const *);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilerw_bf16)))
+svbool_t svwhilerw_bf16(bfloat16_t const *, bfloat16_t const *);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilerw_f16)))
 svbool_t svwhilerw_f16(float16_t const *, float16_t const *);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilerw_s16)))
@@ -15823,6 +15769,8 @@ __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilewr_s64)))
 svbool_t svwhilewr_s64(int64_t const *, int64_t const *);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilewr_u16)))
 svbool_t svwhilewr_u16(uint16_t const *, uint16_t const *);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilewr_bf16)))
+svbool_t svwhilewr_bf16(bfloat16_t const *, bfloat16_t const *);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilewr_f16)))
 svbool_t svwhilewr_f16(float16_t const *, float16_t const *);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilewr_s16)))
@@ -18895,6 +18843,8 @@ __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtbl2_u64)))
 svuint64_t svtbl2(svuint64x2_t, svuint64_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtbl2_u16)))
 svuint16_t svtbl2(svuint16x2_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtbl2_bf16)))
+svbfloat16_t svtbl2(svbfloat16x2_t, svuint16_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtbl2_s8)))
 svint8_t svtbl2(svint8x2_t, svuint8_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtbl2_f64)))
@@ -18917,6 +18867,8 @@ __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtbx_u64)))
 svuint64_t svtbx(svuint64_t, svuint64_t, svuint64_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtbx_u16)))
 svuint16_t svtbx(svuint16_t, svuint16_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtbx_bf16)))
+svbfloat16_t svtbx(svbfloat16_t, svbfloat16_t, svuint16_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtbx_s8)))
 svint8_t svtbx(svint8_t, svint8_t, svuint8_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtbx_f64)))
@@ -19055,6 +19007,8 @@ __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilerw_s64)))
 svbool_t svwhilerw(int64_t const *, int64_t const *);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilerw_u16)))
 svbool_t svwhilerw(uint16_t const *, uint16_t const *);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilerw_bf16)))
+svbool_t svwhilerw(bfloat16_t const *, bfloat16_t const *);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilerw_f16)))
 svbool_t svwhilerw(float16_t const *, float16_t const *);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilerw_s16)))
@@ -19077,6 +19031,8 @@ __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilewr_s64)))
 svbool_t svwhilewr(int64_t const *, int64_t const *);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilewr_u16)))
 svbool_t svwhilewr(uint16_t const *, uint16_t const *);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilewr_bf16)))
+svbool_t svwhilewr(bfloat16_t const *, bfloat16_t const *);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilewr_f16)))
 svbool_t svwhilewr(float16_t const *, float16_t const *);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svwhilewr_s16)))
@@ -19839,6 +19795,8 @@ __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svclasta_n_u64)))
 uint64_t svclasta_n_u64(svbool_t, uint64_t, svuint64_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svclasta_n_u16)))
 uint16_t svclasta_n_u16(svbool_t, uint16_t, svuint16_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svclasta_n_bf16)))
+bfloat16_t svclasta_n_bf16(svbool_t, bfloat16_t, svbfloat16_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svclasta_n_s8)))
 int8_t svclasta_n_s8(svbool_t, int8_t, svint8_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svclasta_n_f64)))
@@ -19861,6 +19819,8 @@ __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svclasta_u64)))
 svuint64_t svclasta_u64(svbool_t, svuint64_t, svuint64_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svclasta_u16)))
 svuint16_t svclasta_u16(svbool_t, svuint16_t, svuint16_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svclasta_bf16)))
+svbfloat16_t svclasta_bf16(svbool_t, svbfloat16_t, svbfloat16_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svclasta_s8)))
 svint8_t svclasta_s8(svbool_t, svint8_t, svint8_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svclasta_f64)))
@@ -19883,6 +19843,8 @@ __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svclastb_n_u64)))
 uint64_t svclastb_n_u64(svbool_t, uint64_t, svuint64_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svclastb_n_u16)))
 uint16_t svclastb_n_u16(svbool_t, uint16_t, svuint16_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svclastb_n_bf16)))
+bfloat16_t svclastb_n_bf16(svbool_t, bfloat16_t, svbfloat16_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svclastb_n_s8)))
 int8_t svclastb_n_s8(svbool_t, int8_t, svint8_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svclastb_n_f64)))
@@ -19905,6 +19867,8 @@ __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svclastb_u64)))
 svuint64_t svclastb_u64(svbool_t, svuint64_t, svuint64_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svclastb_u16)))
 svuint16_t svclastb_u16(svbool_t, svuint16_t, svuint16_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svclastb_bf16)))
+svbfloat16_t svclastb_bf16(svbool_t, svbfloat16_t, svbfloat16_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svclastb_s8)))
 svint8_t svclastb_s8(svbool_t, svint8_t, svint8_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svclastb_f64)))
@@ -20465,6 +20429,8 @@ __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcnt_u64_m)))
 svuint64_t svcnt_u64_m(svuint64_t, svbool_t, svuint64_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcnt_u16_m)))
 svuint16_t svcnt_u16_m(svuint16_t, svbool_t, svuint16_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcnt_bf16_m)))
+svuint16_t svcnt_bf16_m(svuint16_t, svbool_t, svbfloat16_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcnt_s8_m)))
 svuint8_t svcnt_s8_m(svuint8_t, svbool_t, svint8_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcnt_f64_m)))
@@ -20487,6 +20453,8 @@ __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcnt_u64_x)))
 svuint64_t svcnt_u64_x(svbool_t, svuint64_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcnt_u16_x)))
 svuint16_t svcnt_u16_x(svbool_t, svuint16_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcnt_bf16_x)))
+svuint16_t svcnt_bf16_x(svbool_t, svbfloat16_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcnt_s8_x)))
 svuint8_t svcnt_s8_x(svbool_t, svint8_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcnt_f64_x)))
@@ -20509,6 +20477,8 @@ __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcnt_u64_z)))
 svuint64_t svcnt_u64_z(svbool_t, svuint64_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcnt_u16_z)))
 svuint16_t svcnt_u16_z(svbool_t, svuint16_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcnt_bf16_z)))
+svuint16_t svcnt_bf16_z(svbool_t, svbfloat16_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcnt_s8_z)))
 svuint8_t svcnt_s8_z(svbool_t, svint8_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcnt_f64_z)))
@@ -20555,6 +20525,8 @@ __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcreate2_u64)))
 svuint64x2_t svcreate2_u64(svuint64_t, svuint64_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcreate2_u16)))
 svuint16x2_t svcreate2_u16(svuint16_t, svuint16_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcreate2_bf16)))
+svbfloat16x2_t svcreate2_bf16(svbfloat16_t, svbfloat16_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcreate2_s8)))
 svint8x2_t svcreate2_s8(svint8_t, svint8_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcreate2_f64)))
@@ -20567,6 +20539,8 @@ __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcreate2_s32)))
 svint32x2_t svcreate2_s32(svint32_t, svint32_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcreate2_s64)))
 svint64x2_t svcreate2_s64(svint64_t, svint64_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcreate2_mf8)))
+svmfloat8x2_t svcreate2_mf8(svmfloat8_t, svmfloat8_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcreate2_s16)))
 svint16x2_t svcreate2_s16(svint16_t, svint16_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcreate3_u8)))
@@ -20577,6 +20551,8 @@ __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcreate3_u64)))
 svuint64x3_t svcreate3_u64(svuint64_t, svuint64_t, svuint64_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcreate3_u16)))
 svuint16x3_t svcreate3_u16(svuint16_t, svuint16_t, svuint16_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcreate3_bf16)))
+svbfloat16x3_t svcreate3_bf16(svbfloat16_t, svbfloat16_t, svbfloat16_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcreate3_s8)))
 svint8x3_t svcreate3_s8(svint8_t, svint8_t, svint8_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcreate3_f64)))
@@ -20589,6 +20565,8 @@ __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcreate3_s32)))
 svint32x3_t svcreate3_s32(svint32_t, svint32_t, svint32_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcreate3_s64)))
 svint64x3_t svcreate3_s64(svint64_t, svint64_t, svint64_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcreate3_mf8)))
+svmfloat8x3_t svcreate3_mf8(svmfloat8_t, svmfloat8_t, svmfloat8_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcreate3_s16)))
 svint16x3_t svcreate3_s16(svint16_t, svint16_t, svint16_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcreate4_u8)))
@@ -20599,6 +20577,8 @@ __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcreate4_u64)))
 svuint64x4_t svcreate4_u64(svuint64_t, svuint64_t, svuint64_t, svuint64_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcreate4_u16)))
 svuint16x4_t svcreate4_u16(svuint16_t, svuint16_t, svuint16_t, svuint16_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcreate4_bf16)))
+svbfloat16x4_t svcreate4_bf16(svbfloat16_t, svbfloat16_t, svbfloat16_t, svbfloat16_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcreate4_s8)))
 svint8x4_t svcreate4_s8(svint8_t, svint8_t, svint8_t, svint8_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcreate4_f64)))
@@ -20611,6 +20591,8 @@ __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcreate4_s32)))
 svint32x4_t svcreate4_s32(svint32_t, svint32_t, svint32_t, svint32_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcreate4_s64)))
 svint64x4_t svcreate4_s64(svint64_t, svint64_t, svint64_t, svint64_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcreate4_mf8)))
+svmfloat8x4_t svcreate4_mf8(svmfloat8_t, svmfloat8_t, svmfloat8_t, svmfloat8_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcreate4_s16)))
 svint16x4_t svcreate4_s16(svint16_t, svint16_t, svint16_t, svint16_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcvt_f16_f32_m)))
@@ -21017,6 +20999,8 @@ __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdup_n_u64)))
 svuint64_t svdup_n_u64(uint64_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdup_n_u16)))
 svuint16_t svdup_n_u16(uint16_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdup_n_bf16)))
+svbfloat16_t svdup_n_bf16(bfloat16_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdup_n_s8)))
 svint8_t svdup_n_s8(int8_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdup_n_f64)))
@@ -21039,6 +21023,8 @@ __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdup_n_u64_m)))
 svuint64_t svdup_n_u64_m(svuint64_t, svbool_t, uint64_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdup_n_u16_m)))
 svuint16_t svdup_n_u16_m(svuint16_t, svbool_t, uint16_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdup_n_bf16_m)))
+svbfloat16_t svdup_n_bf16_m(svbfloat16_t, svbool_t, bfloat16_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdup_n_s8_m)))
 svint8_t svdup_n_s8_m(svint8_t, svbool_t, int8_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdup_n_f64_m)))
@@ -21069,6 +21055,8 @@ __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdup_n_u64_x)))
 svuint64_t svdup_n_u64_x(svbool_t, uint64_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdup_n_u16_x)))
 svuint16_t svdup_n_u16_x(svbool_t, uint16_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdup_n_bf16_x)))
+svbfloat16_t svdup_n_bf16_x(svbool_t, bfloat16_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdup_n_s8_x)))
 svint8_t svdup_n_s8_x(svbool_t, int8_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdup_n_f64_x)))
@@ -21091,6 +21079,8 @@ __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdup_n_u64_z)))
 svuint64_t svdup_n_u64_z(svbool_t, uint64_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdup_n_u16_z)))
 svuint16_t svdup_n_u16_z(svbool_t, uint16_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdup_n_bf16_z)))
+svbfloat16_t svdup_n_bf16_z(svbool_t, bfloat16_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdup_n_s8_z)))
 svint8_t svdup_n_s8_z(svbool_t, int8_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdup_n_f64_z)))
@@ -21113,6 +21103,8 @@ __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdup_lane_u64)))
 svuint64_t svdup_lane_u64(svuint64_t, uint64_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdup_lane_u16)))
 svuint16_t svdup_lane_u16(svuint16_t, uint16_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdup_lane_bf16)))
+svbfloat16_t svdup_lane_bf16(svbfloat16_t, uint16_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdup_lane_s8)))
 svint8_t svdup_lane_s8(svint8_t, uint8_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdup_lane_f64)))
@@ -21133,6 +21125,8 @@ __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdupq_n_s8)))
 svint8_t svdupq_n_s8(int8_t, int8_t, int8_t, int8_t, int8_t, int8_t, int8_t, int8_t, int8_t, int8_t, int8_t, int8_t, int8_t, int8_t, int8_t, int8_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdupq_n_u16)))
 svuint16_t svdupq_n_u16(uint16_t, uint16_t, uint16_t, uint16_t, uint16_t, uint16_t, uint16_t, uint16_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdupq_n_bf16)))
+svbfloat16_t svdupq_n_bf16(bfloat16_t, bfloat16_t, bfloat16_t, bfloat16_t, bfloat16_t, bfloat16_t, bfloat16_t, bfloat16_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdupq_n_f16)))
 svfloat16_t svdupq_n_f16(float16_t, float16_t, float16_t, float16_t, float16_t, float16_t, float16_t, float16_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdupq_n_s16)))
@@ -21165,6 +21159,8 @@ __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdupq_lane_u64)))
 svuint64_t svdupq_lane_u64(svuint64_t, uint64_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdupq_lane_u16)))
 svuint16_t svdupq_lane_u16(svuint16_t, uint64_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdupq_lane_bf16)))
+svbfloat16_t svdupq_lane_bf16(svbfloat16_t, uint64_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdupq_lane_s8)))
 svint8_t svdupq_lane_s8(svint8_t, uint64_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdupq_lane_f64)))
@@ -21301,6 +21297,8 @@ __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svext_u64)))
 svuint64_t svext_u64(svuint64_t, svuint64_t, uint64_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svext_u16)))
 svuint16_t svext_u16(svuint16_t, svuint16_t, uint64_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svext_bf16)))
+svbfloat16_t svext_bf16(svbfloat16_t, svbfloat16_t, uint64_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svext_s8)))
 svint8_t svext_s8(svint8_t, svint8_t, uint64_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svext_f64)))
@@ -21395,6 +21393,8 @@ __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svget2_u64)))
 svuint64_t svget2_u64(svuint64x2_t, uint64_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svget2_u16)))
 svuint16_t svget2_u16(svuint16x2_t, uint64_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svget2_bf16)))
+svbfloat16_t svget2_bf16(svbfloat16x2_t, uint64_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svget2_s8)))
 svint8_t svget2_s8(svint8x2_t, uint64_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svget2_f64)))
@@ -21407,6 +21407,8 @@ __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svget2_s32)))
 svint32_t svget2_s32(svint32x2_t, uint64_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svget2_s64)))
 svint64_t svget2_s64(svint64x2_t, uint64_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svget2_mf8)))
+svmfloat8_t svget2_mf8(svmfloat8x2_t, uint64_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svget2_s16)))
 svint16_t svget2_s16(svint16x2_t, uint64_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svget3_u8)))
@@ -21417,6 +21419,8 @@ __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svget3_u64)))
 svuint64_t svget3_u64(svuint64x3_t, uint64_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svget3_u16)))
 svuint16_t svget3_u16(svuint16x3_t, uint64_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svget3_bf16)))
+svbfloat16_t svget3_bf16(svbfloat16x3_t, uint64_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svget3_s8)))
 svint8_t svget3_s8(svint8x3_t, uint64_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svget3_f64)))
@@ -21429,6 +21433,8 @@ __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svget3_s32)))
 svint32_t svget3_s32(svint32x3_t, uint64_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svget3_s64)))
 svint64_t svget3_s64(svint64x3_t, uint64_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svget3_mf8)))
+svmfloat8_t svget3_mf8(svmfloat8x3_t, uint64_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svget3_s16)))
 svint16_t svget3_s16(svint16x3_t, uint64_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svget4_u8)))
@@ -21439,6 +21445,8 @@ __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svget4_u64)))
 svuint64_t svget4_u64(svuint64x4_t, uint64_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svget4_u16)))
 svuint16_t svget4_u16(svuint16x4_t, uint64_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svget4_bf16)))
+svbfloat16_t svget4_bf16(svbfloat16x4_t, uint64_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svget4_s8)))
 svint8_t svget4_s8(svint8x4_t, uint64_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svget4_f64)))
@@ -21451,6 +21459,8 @@ __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svget4_s32)))
 svint32_t svget4_s32(svint32x4_t, uint64_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svget4_s64)))
 svint64_t svget4_s64(svint64x4_t, uint64_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svget4_mf8)))
+svmfloat8_t svget4_mf8(svmfloat8x4_t, uint64_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svget4_s16)))
 svint16_t svget4_s16(svint16x4_t, uint64_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svindex_u8)))
@@ -21477,6 +21487,8 @@ __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svinsr_n_u64)))
 svuint64_t svinsr_n_u64(svuint64_t, uint64_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svinsr_n_u16)))
 svuint16_t svinsr_n_u16(svuint16_t, uint16_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svinsr_n_bf16)))
+svbfloat16_t svinsr_n_bf16(svbfloat16_t, bfloat16_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svinsr_n_s8)))
 svint8_t svinsr_n_s8(svint8_t, int8_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svinsr_n_f64)))
@@ -21499,6 +21511,8 @@ __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlasta_u64)))
 uint64_t svlasta_u64(svbool_t, svuint64_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlasta_u16)))
 uint16_t svlasta_u16(svbool_t, svuint16_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlasta_bf16)))
+bfloat16_t svlasta_bf16(svbool_t, svbfloat16_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlasta_s8)))
 int8_t svlasta_s8(svbool_t, svint8_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlasta_f64)))
@@ -21521,6 +21535,8 @@ __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlastb_u64)))
 uint64_t svlastb_u64(svbool_t, svuint64_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlastb_u16)))
 uint16_t svlastb_u16(svbool_t, svuint16_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlastb_bf16)))
+bfloat16_t svlastb_bf16(svbool_t, svbfloat16_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlastb_s8)))
 int8_t svlastb_s8(svbool_t, svint8_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlastb_f64)))
@@ -21543,6 +21559,8 @@ __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1_u64)))
 svuint64_t svld1_u64(svbool_t, uint64_t const *);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1_u16)))
 svuint16_t svld1_u16(svbool_t, uint16_t const *);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1_bf16)))
+svbfloat16_t svld1_bf16(svbool_t, bfloat16_t const *);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1_s8)))
 svint8_t svld1_s8(svbool_t, int8_t const *);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1_f64)))
@@ -21555,6 +21573,8 @@ __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1_s32)))
 svint32_t svld1_s32(svbool_t, int32_t const *);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1_s64)))
 svint64_t svld1_s64(svbool_t, int64_t const *);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1_mf8)))
+svmfloat8_t svld1_mf8(svbool_t, mfloat8_t const *);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1_s16)))
 svint16_t svld1_s16(svbool_t, int16_t const *);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1_vnum_u8)))
@@ -21565,6 +21585,8 @@ __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1_vnum_u64)))
 svuint64_t svld1_vnum_u64(svbool_t, uint64_t const *, int64_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1_vnum_u16)))
 svuint16_t svld1_vnum_u16(svbool_t, uint16_t const *, int64_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1_vnum_bf16)))
+svbfloat16_t svld1_vnum_bf16(svbool_t, bfloat16_t const *, int64_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1_vnum_s8)))
 svint8_t svld1_vnum_s8(svbool_t, int8_t const *, int64_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1_vnum_f64)))
@@ -21577,6 +21599,8 @@ __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1_vnum_s32)))
 svint32_t svld1_vnum_s32(svbool_t, int32_t const *, int64_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1_vnum_s64)))
 svint64_t svld1_vnum_s64(svbool_t, int64_t const *, int64_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1_vnum_mf8)))
+svmfloat8_t svld1_vnum_mf8(svbool_t, mfloat8_t const *, int64_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1_vnum_s16)))
 svint16_t svld1_vnum_s16(svbool_t, int16_t const *, int64_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1rq_u8)))
@@ -21587,6 +21611,8 @@ __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1rq_u64)))
 svuint64_t svld1rq_u64(svbool_t, uint64_t const *);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1rq_u16)))
 svuint16_t svld1rq_u16(svbool_t, uint16_t const *);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1rq_bf16)))
+svbfloat16_t svld1rq_bf16(svbool_t, bfloat16_t const *);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1rq_s8)))
 svint8_t svld1rq_s8(svbool_t, int8_t const *);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1rq_f64)))
@@ -21599,6 +21625,8 @@ __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1rq_s32)))
 svint32_t svld1rq_s32(svbool_t, int32_t const *);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1rq_s64)))
 svint64_t svld1rq_s64(svbool_t, int64_t const *);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1rq_mf8)))
+svmfloat8_t svld1rq_mf8(svbool_t, mfloat8_t const *);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1rq_s16)))
 svint16_t svld1rq_s16(svbool_t, int16_t const *);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1sb_vnum_u32)))
@@ -21705,6 +21733,8 @@ __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld2_u64)))
 svuint64x2_t svld2_u64(svbool_t, uint64_t const *);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld2_u16)))
 svuint16x2_t svld2_u16(svbool_t, uint16_t const *);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld2_bf16)))
+svbfloat16x2_t svld2_bf16(svbool_t, bfloat16_t const *);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld2_s8)))
 svint8x2_t svld2_s8(svbool_t, int8_t const *);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld2_f64)))
@@ -21717,6 +21747,8 @@ __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld2_s32)))
 svint32x2_t svld2_s32(svbool_t, int32_t const *);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld2_s64)))
 svint64x2_t svld2_s64(svbool_t, int64_t const *);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld2_mf8)))
+svmfloat8x2_t svld2_mf8(svbool_t, mfloat8_t const *);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld2_s16)))
 svint16x2_t svld2_s16(svbool_t, int16_t const *);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld2_vnum_u8)))
@@ -21727,6 +21759,8 @@ __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld2_vnum_u64)))
 svuint64x2_t svld2_vnum_u64(svbool_t, uint64_t const *, int64_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld2_vnum_u16)))
 svuint16x2_t svld2_vnum_u16(svbool_t, uint16_t const *, int64_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld2_vnum_bf16)))
+svbfloat16x2_t svld2_vnum_bf16(svbool_t, bfloat16_t const *, int64_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld2_vnum_s8)))
 svint8x2_t svld2_vnum_s8(svbool_t, int8_t const *, int64_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld2_vnum_f64)))
@@ -21739,6 +21773,8 @@ __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld2_vnum_s32)))
 svint32x2_t svld2_vnum_s32(svbool_t, int32_t const *, int64_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld2_vnum_s64)))
 svint64x2_t svld2_vnum_s64(svbool_t, int64_t const *, int64_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld2_vnum_mf8)))
+svmfloat8x2_t svld2_vnum_mf8(svbool_t, mfloat8_t const *, int64_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld2_vnum_s16)))
 svint16x2_t svld2_vnum_s16(svbool_t, int16_t const *, int64_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld3_u8)))
@@ -21749,6 +21785,8 @@ __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld3_u64)))
 svuint64x3_t svld3_u64(svbool_t, uint64_t const *);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld3_u16)))
 svuint16x3_t svld3_u16(svbool_t, uint16_t const *);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld3_bf16)))
+svbfloat16x3_t svld3_bf16(svbool_t, bfloat16_t const *);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld3_s8)))
 svint8x3_t svld3_s8(svbool_t, int8_t const *);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld3_f64)))
@@ -21761,6 +21799,8 @@ __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld3_s32)))
 svint32x3_t svld3_s32(svbool_t, int32_t const *);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld3_s64)))
 svint64x3_t svld3_s64(svbool_t, int64_t const *);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld3_mf8)))
+svmfloat8x3_t svld3_mf8(svbool_t, mfloat8_t const *);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld3_s16)))
 svint16x3_t svld3_s16(svbool_t, int16_t const *);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld3_vnum_u8)))
@@ -21771,6 +21811,8 @@ __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld3_vnum_u64)))
 svuint64x3_t svld3_vnum_u64(svbool_t, uint64_t const *, int64_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld3_vnum_u16)))
 svuint16x3_t svld3_vnum_u16(svbool_t, uint16_t const *, int64_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld3_vnum_bf16)))
+svbfloat16x3_t svld3_vnum_bf16(svbool_t, bfloat16_t const *, int64_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld3_vnum_s8)))
 svint8x3_t svld3_vnum_s8(svbool_t, int8_t const *, int64_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld3_vnum_f64)))
@@ -21783,6 +21825,8 @@ __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld3_vnum_s32)))
 svint32x3_t svld3_vnum_s32(svbool_t, int32_t const *, int64_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld3_vnum_s64)))
 svint64x3_t svld3_vnum_s64(svbool_t, int64_t const *, int64_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld3_vnum_mf8)))
+svmfloat8x3_t svld3_vnum_mf8(svbool_t, mfloat8_t const *, int64_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld3_vnum_s16)))
 svint16x3_t svld3_vnum_s16(svbool_t, int16_t const *, int64_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld4_u8)))
@@ -21793,6 +21837,8 @@ __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld4_u64)))
 svuint64x4_t svld4_u64(svbool_t, uint64_t const *);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld4_u16)))
 svuint16x4_t svld4_u16(svbool_t, uint16_t const *);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld4_bf16)))
+svbfloat16x4_t svld4_bf16(svbool_t, bfloat16_t const *);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld4_s8)))
 svint8x4_t svld4_s8(svbool_t, int8_t const *);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld4_f64)))
@@ -21805,6 +21851,8 @@ __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld4_s32)))
 svint32x4_t svld4_s32(svbool_t, int32_t const *);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld4_s64)))
 svint64x4_t svld4_s64(svbool_t, int64_t const *);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld4_mf8)))
+svmfloat8x4_t svld4_mf8(svbool_t, mfloat8_t const *);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld4_s16)))
 svint16x4_t svld4_s16(svbool_t, int16_t const *);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld4_vnum_u8)))
@@ -21815,6 +21863,8 @@ __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld4_vnum_u64)))
 svuint64x4_t svld4_vnum_u64(svbool_t, uint64_t const *, int64_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld4_vnum_u16)))
 svuint16x4_t svld4_vnum_u16(svbool_t, uint16_t const *, int64_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld4_vnum_bf16)))
+svbfloat16x4_t svld4_vnum_bf16(svbool_t, bfloat16_t const *, int64_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld4_vnum_s8)))
 svint8x4_t svld4_vnum_s8(svbool_t, int8_t const *, int64_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld4_vnum_f64)))
@@ -21827,6 +21877,8 @@ __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld4_vnum_s32)))
 svint32x4_t svld4_vnum_s32(svbool_t, int32_t const *, int64_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld4_vnum_s64)))
 svint64x4_t svld4_vnum_s64(svbool_t, int64_t const *, int64_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld4_vnum_mf8)))
+svmfloat8x4_t svld4_vnum_mf8(svbool_t, mfloat8_t const *, int64_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld4_vnum_s16)))
 svint16x4_t svld4_vnum_s16(svbool_t, int16_t const *, int64_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_u8)))
@@ -21837,6 +21889,8 @@ __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_u64)))
 svuint64_t svldnt1_u64(svbool_t, uint64_t const *);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_u16)))
 svuint16_t svldnt1_u16(svbool_t, uint16_t const *);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_bf16)))
+svbfloat16_t svldnt1_bf16(svbool_t, bfloat16_t const *);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_s8)))
 svint8_t svldnt1_s8(svbool_t, int8_t const *);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_f64)))
@@ -21849,6 +21903,8 @@ __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_s32)))
 svint32_t svldnt1_s32(svbool_t, int32_t const *);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_s64)))
 svint64_t svldnt1_s64(svbool_t, int64_t const *);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_mf8)))
+svmfloat8_t svldnt1_mf8(svbool_t, mfloat8_t const *);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_s16)))
 svint16_t svldnt1_s16(svbool_t, int16_t const *);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_vnum_u8)))
@@ -21859,6 +21915,8 @@ __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_vnum_u64)))
 svuint64_t svldnt1_vnum_u64(svbool_t, uint64_t const *, int64_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_vnum_u16)))
 svuint16_t svldnt1_vnum_u16(svbool_t, uint16_t const *, int64_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_vnum_bf16)))
+svbfloat16_t svldnt1_vnum_bf16(svbool_t, bfloat16_t const *, int64_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_vnum_s8)))
 svint8_t svldnt1_vnum_s8(svbool_t, int8_t const *, int64_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_vnum_f64)))
@@ -21871,6 +21929,8 @@ __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_vnum_s32)))
 svint32_t svldnt1_vnum_s32(svbool_t, int32_t const *, int64_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_vnum_s64)))
 svint64_t svldnt1_vnum_s64(svbool_t, int64_t const *, int64_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_vnum_mf8)))
+svmfloat8_t svldnt1_vnum_mf8(svbool_t, mfloat8_t const *, int64_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_vnum_s16)))
 svint16_t svldnt1_vnum_s16(svbool_t, int16_t const *, int64_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlen_u8)))
@@ -21881,6 +21941,8 @@ __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlen_u64)))
 uint64_t svlen_u64(svuint64_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlen_u16)))
 uint64_t svlen_u16(svuint16_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlen_bf16)))
+uint64_t svlen_bf16(svbfloat16_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlen_s8)))
 uint64_t svlen_s8(svint8_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlen_f64)))
@@ -24171,6 +24233,8 @@ __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrev_u64)))
 svuint64_t svrev_u64(svuint64_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrev_u16)))
 svuint16_t svrev_u16(svuint16_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrev_bf16)))
+svbfloat16_t svrev_bf16(svbfloat16_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrev_s8)))
 svint8_t svrev_s8(svint8_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrev_f64)))
@@ -24449,6 +24513,8 @@ __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsel_u64)))
 svuint64_t svsel_u64(svbool_t, svuint64_t, svuint64_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsel_u16)))
 svuint16_t svsel_u16(svbool_t, svuint16_t, svuint16_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsel_bf16)))
+svbfloat16_t svsel_bf16(svbool_t, svbfloat16_t, svbfloat16_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsel_s8)))
 svint8_t svsel_s8(svbool_t, svint8_t, svint8_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsel_f64)))
@@ -24471,6 +24537,8 @@ __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svset2_u64)))
 svuint64x2_t svset2_u64(svuint64x2_t, uint64_t, svuint64_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svset2_u16)))
 svuint16x2_t svset2_u16(svuint16x2_t, uint64_t, svuint16_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svset2_bf16)))
+svbfloat16x2_t svset2_bf16(svbfloat16x2_t, uint64_t, svbfloat16_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svset2_s8)))
 svint8x2_t svset2_s8(svint8x2_t, uint64_t, svint8_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svset2_f64)))
@@ -24483,6 +24551,8 @@ __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svset2_s32)))
 svint32x2_t svset2_s32(svint32x2_t, uint64_t, svint32_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svset2_s64)))
 svint64x2_t svset2_s64(svint64x2_t, uint64_t, svint64_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svset2_mf8)))
+svmfloat8x2_t svset2_mf8(svmfloat8x2_t, uint64_t, svmfloat8_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svset2_s16)))
 svint16x2_t svset2_s16(svint16x2_t, uint64_t, svint16_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svset3_u8)))
@@ -24493,6 +24563,8 @@ __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svset3_u64)))
 svuint64x3_t svset3_u64(svuint64x3_t, uint64_t, svuint64_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svset3_u16)))
 svuint16x3_t svset3_u16(svuint16x3_t, uint64_t, svuint16_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svset3_bf16)))
+svbfloat16x3_t svset3_bf16(svbfloat16x3_t, uint64_t, svbfloat16_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svset3_s8)))
 svint8x3_t svset3_s8(svint8x3_t, uint64_t, svint8_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svset3_f64)))
@@ -24505,6 +24577,8 @@ __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svset3_s32)))
 svint32x3_t svset3_s32(svint32x3_t, uint64_t, svint32_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svset3_s64)))
 svint64x3_t svset3_s64(svint64x3_t, uint64_t, svint64_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svset3_mf8)))
+svmfloat8x3_t svset3_mf8(svmfloat8x3_t, uint64_t, svmfloat8_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svset3_s16)))
 svint16x3_t svset3_s16(svint16x3_t, uint64_t, svint16_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svset4_u8)))
@@ -24515,6 +24589,8 @@ __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svset4_u64)))
 svuint64x4_t svset4_u64(svuint64x4_t, uint64_t, svuint64_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svset4_u16)))
 svuint16x4_t svset4_u16(svuint16x4_t, uint64_t, svuint16_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svset4_bf16)))
+svbfloat16x4_t svset4_bf16(svbfloat16x4_t, uint64_t, svbfloat16_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svset4_s8)))
 svint8x4_t svset4_s8(svint8x4_t, uint64_t, svint8_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svset4_f64)))
@@ -24527,6 +24603,8 @@ __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svset4_s32)))
 svint32x4_t svset4_s32(svint32x4_t, uint64_t, svint32_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svset4_s64)))
 svint64x4_t svset4_s64(svint64x4_t, uint64_t, svint64_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svset4_mf8)))
+svmfloat8x4_t svset4_mf8(svmfloat8x4_t, uint64_t, svmfloat8_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svset4_s16)))
 svint16x4_t svset4_s16(svint16x4_t, uint64_t, svint16_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsplice_u8)))
@@ -24537,6 +24615,8 @@ __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsplice_u64)))
 svuint64_t svsplice_u64(svbool_t, svuint64_t, svuint64_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsplice_u16)))
 svuint16_t svsplice_u16(svbool_t, svuint16_t, svuint16_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsplice_bf16)))
+svbfloat16_t svsplice_bf16(svbool_t, svbfloat16_t, svbfloat16_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsplice_s8)))
 svint8_t svsplice_s8(svbool_t, svint8_t, svint8_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsplice_f64)))
@@ -24577,6 +24657,8 @@ __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1_u64)))
 void svst1_u64(svbool_t, uint64_t *, svuint64_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1_u16)))
 void svst1_u16(svbool_t, uint16_t *, svuint16_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1_bf16)))
+void svst1_bf16(svbool_t, bfloat16_t *, svbfloat16_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1_s8)))
 void svst1_s8(svbool_t, int8_t *, svint8_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1_f64)))
@@ -24589,6 +24671,8 @@ __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1_s32)))
 void svst1_s32(svbool_t, int32_t *, svint32_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1_s64)))
 void svst1_s64(svbool_t, int64_t *, svint64_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1_mf8)))
+void svst1_mf8(svbool_t, mfloat8_t *, svmfloat8_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1_s16)))
 void svst1_s16(svbool_t, int16_t *, svint16_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1_vnum_u8)))
@@ -24599,6 +24683,8 @@ __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1_vnum_u64)))
 void svst1_vnum_u64(svbool_t, uint64_t *, int64_t, svuint64_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1_vnum_u16)))
 void svst1_vnum_u16(svbool_t, uint16_t *, int64_t, svuint16_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1_vnum_bf16)))
+void svst1_vnum_bf16(svbool_t, bfloat16_t *, int64_t, svbfloat16_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1_vnum_s8)))
 void svst1_vnum_s8(svbool_t, int8_t *, int64_t, svint8_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1_vnum_f64)))
@@ -24611,6 +24697,8 @@ __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1_vnum_s32)))
 void svst1_vnum_s32(svbool_t, int32_t *, int64_t, svint32_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1_vnum_s64)))
 void svst1_vnum_s64(svbool_t, int64_t *, int64_t, svint64_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1_vnum_mf8)))
+void svst1_vnum_mf8(svbool_t, mfloat8_t *, int64_t, svmfloat8_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1_vnum_s16)))
 void svst1_vnum_s16(svbool_t, int16_t *, int64_t, svint16_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1b_s32)))
@@ -24669,6 +24757,8 @@ __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst2_u64)))
 void svst2_u64(svbool_t, uint64_t *, svuint64x2_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst2_u16)))
 void svst2_u16(svbool_t, uint16_t *, svuint16x2_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst2_bf16)))
+void svst2_bf16(svbool_t, bfloat16_t *, svbfloat16x2_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst2_s8)))
 void svst2_s8(svbool_t, int8_t *, svint8x2_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst2_f64)))
@@ -24681,6 +24771,8 @@ __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst2_s32)))
 void svst2_s32(svbool_t, int32_t *, svint32x2_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst2_s64)))
 void svst2_s64(svbool_t, int64_t *, svint64x2_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst2_mf8)))
+void svst2_mf8(svbool_t, mfloat8_t *, svmfloat8x2_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst2_s16)))
 void svst2_s16(svbool_t, int16_t *, svint16x2_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst2_vnum_u8)))
@@ -24691,6 +24783,8 @@ __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst2_vnum_u64)))
 void svst2_vnum_u64(svbool_t, uint64_t *, int64_t, svuint64x2_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst2_vnum_u16)))
 void svst2_vnum_u16(svbool_t, uint16_t *, int64_t, svuint16x2_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst2_vnum_bf16)))
+void svst2_vnum_bf16(svbool_t, bfloat16_t *, int64_t, svbfloat16x2_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst2_vnum_s8)))
 void svst2_vnum_s8(svbool_t, int8_t *, int64_t, svint8x2_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst2_vnum_f64)))
@@ -24703,6 +24797,8 @@ __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst2_vnum_s32)))
 void svst2_vnum_s32(svbool_t, int32_t *, int64_t, svint32x2_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst2_vnum_s64)))
 void svst2_vnum_s64(svbool_t, int64_t *, int64_t, svint64x2_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst2_vnum_mf8)))
+void svst2_vnum_mf8(svbool_t, mfloat8_t *, int64_t, svmfloat8x2_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst2_vnum_s16)))
 void svst2_vnum_s16(svbool_t, int16_t *, int64_t, svint16x2_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst3_u8)))
@@ -24713,6 +24809,8 @@ __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst3_u64)))
 void svst3_u64(svbool_t, uint64_t *, svuint64x3_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst3_u16)))
 void svst3_u16(svbool_t, uint16_t *, svuint16x3_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst3_bf16)))
+void svst3_bf16(svbool_t, bfloat16_t *, svbfloat16x3_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst3_s8)))
 void svst3_s8(svbool_t, int8_t *, svint8x3_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst3_f64)))
@@ -24725,6 +24823,8 @@ __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst3_s32)))
 void svst3_s32(svbool_t, int32_t *, svint32x3_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst3_s64)))
 void svst3_s64(svbool_t, int64_t *, svint64x3_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst3_mf8)))
+void svst3_mf8(svbool_t, mfloat8_t *, svmfloat8x3_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst3_s16)))
 void svst3_s16(svbool_t, int16_t *, svint16x3_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst3_vnum_u8)))
@@ -24735,6 +24835,8 @@ __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst3_vnum_u64)))
 void svst3_vnum_u64(svbool_t, uint64_t *, int64_t, svuint64x3_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst3_vnum_u16)))
 void svst3_vnum_u16(svbool_t, uint16_t *, int64_t, svuint16x3_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst3_vnum_bf16)))
+void svst3_vnum_bf16(svbool_t, bfloat16_t *, int64_t, svbfloat16x3_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst3_vnum_s8)))
 void svst3_vnum_s8(svbool_t, int8_t *, int64_t, svint8x3_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst3_vnum_f64)))
@@ -24747,6 +24849,8 @@ __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst3_vnum_s32)))
 void svst3_vnum_s32(svbool_t, int32_t *, int64_t, svint32x3_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst3_vnum_s64)))
 void svst3_vnum_s64(svbool_t, int64_t *, int64_t, svint64x3_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst3_vnum_mf8)))
+void svst3_vnum_mf8(svbool_t, mfloat8_t *, int64_t, svmfloat8x3_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst3_vnum_s16)))
 void svst3_vnum_s16(svbool_t, int16_t *, int64_t, svint16x3_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst4_u8)))
@@ -24757,6 +24861,8 @@ __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst4_u64)))
 void svst4_u64(svbool_t, uint64_t *, svuint64x4_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst4_u16)))
 void svst4_u16(svbool_t, uint16_t *, svuint16x4_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst4_bf16)))
+void svst4_bf16(svbool_t, bfloat16_t *, svbfloat16x4_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst4_s8)))
 void svst4_s8(svbool_t, int8_t *, svint8x4_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst4_f64)))
@@ -24769,6 +24875,8 @@ __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst4_s32)))
 void svst4_s32(svbool_t, int32_t *, svint32x4_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst4_s64)))
 void svst4_s64(svbool_t, int64_t *, svint64x4_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst4_mf8)))
+void svst4_mf8(svbool_t, mfloat8_t *, svmfloat8x4_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst4_s16)))
 void svst4_s16(svbool_t, int16_t *, svint16x4_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst4_vnum_u8)))
@@ -24779,6 +24887,8 @@ __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst4_vnum_u64)))
 void svst4_vnum_u64(svbool_t, uint64_t *, int64_t, svuint64x4_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst4_vnum_u16)))
 void svst4_vnum_u16(svbool_t, uint16_t *, int64_t, svuint16x4_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst4_vnum_bf16)))
+void svst4_vnum_bf16(svbool_t, bfloat16_t *, int64_t, svbfloat16x4_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst4_vnum_s8)))
 void svst4_vnum_s8(svbool_t, int8_t *, int64_t, svint8x4_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst4_vnum_f64)))
@@ -24791,6 +24901,8 @@ __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst4_vnum_s32)))
 void svst4_vnum_s32(svbool_t, int32_t *, int64_t, svint32x4_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst4_vnum_s64)))
 void svst4_vnum_s64(svbool_t, int64_t *, int64_t, svint64x4_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst4_vnum_mf8)))
+void svst4_vnum_mf8(svbool_t, mfloat8_t *, int64_t, svmfloat8x4_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst4_vnum_s16)))
 void svst4_vnum_s16(svbool_t, int16_t *, int64_t, svint16x4_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_u8)))
@@ -24801,6 +24913,8 @@ __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_u64)))
 void svstnt1_u64(svbool_t, uint64_t *, svuint64_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_u16)))
 void svstnt1_u16(svbool_t, uint16_t *, svuint16_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_bf16)))
+void svstnt1_bf16(svbool_t, bfloat16_t *, svbfloat16_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_s8)))
 void svstnt1_s8(svbool_t, int8_t *, svint8_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_f64)))
@@ -24813,6 +24927,8 @@ __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_s32)))
 void svstnt1_s32(svbool_t, int32_t *, svint32_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_s64)))
 void svstnt1_s64(svbool_t, int64_t *, svint64_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_mf8)))
+void svstnt1_mf8(svbool_t, mfloat8_t *, svmfloat8_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_s16)))
 void svstnt1_s16(svbool_t, int16_t *, svint16_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_vnum_u8)))
@@ -24823,6 +24939,8 @@ __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_vnum_u64)))
 void svstnt1_vnum_u64(svbool_t, uint64_t *, int64_t, svuint64_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_vnum_u16)))
 void svstnt1_vnum_u16(svbool_t, uint16_t *, int64_t, svuint16_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_vnum_bf16)))
+void svstnt1_vnum_bf16(svbool_t, bfloat16_t *, int64_t, svbfloat16_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_vnum_s8)))
 void svstnt1_vnum_s8(svbool_t, int8_t *, int64_t, svint8_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_vnum_f64)))
@@ -24835,6 +24953,8 @@ __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_vnum_s32)))
 void svstnt1_vnum_s32(svbool_t, int32_t *, int64_t, svint32_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_vnum_s64)))
 void svstnt1_vnum_s64(svbool_t, int64_t *, int64_t, svint64_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_vnum_mf8)))
+void svstnt1_vnum_mf8(svbool_t, mfloat8_t *, int64_t, svmfloat8_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_vnum_s16)))
 void svstnt1_vnum_s16(svbool_t, int16_t *, int64_t, svint16_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsub_n_f64_m)))
@@ -25109,6 +25229,8 @@ __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtbl_u64)))
 svuint64_t svtbl_u64(svuint64_t, svuint64_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtbl_u16)))
 svuint16_t svtbl_u16(svuint16_t, svuint16_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtbl_bf16)))
+svbfloat16_t svtbl_bf16(svbfloat16_t, svuint16_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtbl_s8)))
 svint8_t svtbl_s8(svint8_t, svuint8_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtbl_f64)))
@@ -25131,6 +25253,8 @@ __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtrn1_u64)))
 svuint64_t svtrn1_u64(svuint64_t, svuint64_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtrn1_u16)))
 svuint16_t svtrn1_u16(svuint16_t, svuint16_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtrn1_bf16)))
+svbfloat16_t svtrn1_bf16(svbfloat16_t, svbfloat16_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtrn1_s8)))
 svint8_t svtrn1_s8(svint8_t, svint8_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtrn1_f64)))
@@ -25161,6 +25285,8 @@ __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtrn2_u64)))
 svuint64_t svtrn2_u64(svuint64_t, svuint64_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtrn2_u16)))
 svuint16_t svtrn2_u16(svuint16_t, svuint16_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtrn2_bf16)))
+svbfloat16_t svtrn2_bf16(svbfloat16_t, svbfloat16_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtrn2_s8)))
 svint8_t svtrn2_s8(svint8_t, svint8_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtrn2_f64)))
@@ -25191,6 +25317,8 @@ __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svundef2_u64)))
 svuint64x2_t svundef2_u64(void);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svundef2_u16)))
 svuint16x2_t svundef2_u16(void);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svundef2_bf16)))
+svbfloat16x2_t svundef2_bf16(void);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svundef2_s8)))
 svint8x2_t svundef2_s8(void);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svundef2_f64)))
@@ -25203,6 +25331,8 @@ __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svundef2_s32)))
 svint32x2_t svundef2_s32(void);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svundef2_s64)))
 svint64x2_t svundef2_s64(void);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svundef2_mf8)))
+svmfloat8x2_t svundef2_mf8(void);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svundef2_s16)))
 svint16x2_t svundef2_s16(void);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svundef3_u8)))
@@ -25213,6 +25343,8 @@ __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svundef3_u64)))
 svuint64x3_t svundef3_u64(void);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svundef3_u16)))
 svuint16x3_t svundef3_u16(void);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svundef3_bf16)))
+svbfloat16x3_t svundef3_bf16(void);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svundef3_s8)))
 svint8x3_t svundef3_s8(void);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svundef3_f64)))
@@ -25225,6 +25357,8 @@ __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svundef3_s32)))
 svint32x3_t svundef3_s32(void);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svundef3_s64)))
 svint64x3_t svundef3_s64(void);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svundef3_mf8)))
+svmfloat8x3_t svundef3_mf8(void);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svundef3_s16)))
 svint16x3_t svundef3_s16(void);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svundef4_u8)))
@@ -25235,6 +25369,8 @@ __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svundef4_u64)))
 svuint64x4_t svundef4_u64(void);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svundef4_u16)))
 svuint16x4_t svundef4_u16(void);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svundef4_bf16)))
+svbfloat16x4_t svundef4_bf16(void);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svundef4_s8)))
 svint8x4_t svundef4_s8(void);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svundef4_f64)))
@@ -25247,6 +25383,8 @@ __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svundef4_s32)))
 svint32x4_t svundef4_s32(void);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svundef4_s64)))
 svint64x4_t svundef4_s64(void);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svundef4_mf8)))
+svmfloat8x4_t svundef4_mf8(void);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svundef4_s16)))
 svint16x4_t svundef4_s16(void);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svundef_u8)))
@@ -25257,6 +25395,8 @@ __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svundef_u64)))
 svuint64_t svundef_u64(void);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svundef_u16)))
 svuint16_t svundef_u16(void);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svundef_bf16)))
+svbfloat16_t svundef_bf16(void);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svundef_s8)))
 svint8_t svundef_s8(void);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svundef_f64)))
@@ -25269,6 +25409,8 @@ __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svundef_s32)))
 svint32_t svundef_s32(void);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svundef_s64)))
 svint64_t svundef_s64(void);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svundef_mf8)))
+svmfloat8_t svundef_mf8(void);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svundef_s16)))
 svint16_t svundef_s16(void);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svunpkhi_b)))
@@ -25307,6 +25449,8 @@ __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzp1_u64)))
 svuint64_t svuzp1_u64(svuint64_t, svuint64_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzp1_u16)))
 svuint16_t svuzp1_u16(svuint16_t, svuint16_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzp1_bf16)))
+svbfloat16_t svuzp1_bf16(svbfloat16_t, svbfloat16_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzp1_s8)))
 svint8_t svuzp1_s8(svint8_t, svint8_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzp1_f64)))
@@ -25337,6 +25481,8 @@ __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzp2_u64)))
 svuint64_t svuzp2_u64(svuint64_t, svuint64_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzp2_u16)))
 svuint16_t svuzp2_u16(svuint16_t, svuint16_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzp2_bf16)))
+svbfloat16_t svuzp2_bf16(svbfloat16_t, svbfloat16_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzp2_s8)))
 svint8_t svuzp2_s8(svint8_t, svint8_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzp2_f64)))
@@ -25431,6 +25577,8 @@ __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzip1_u64)))
 svuint64_t svzip1_u64(svuint64_t, svuint64_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzip1_u16)))
 svuint16_t svzip1_u16(svuint16_t, svuint16_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzip1_bf16)))
+svbfloat16_t svzip1_bf16(svbfloat16_t, svbfloat16_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzip1_s8)))
 svint8_t svzip1_s8(svint8_t, svint8_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzip1_f64)))
@@ -25461,6 +25609,8 @@ __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzip2_u64)))
 svuint64_t svzip2_u64(svuint64_t, svuint64_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzip2_u16)))
 svuint16_t svzip2_u16(svuint16_t, svuint16_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzip2_bf16)))
+svbfloat16_t svzip2_bf16(svbfloat16_t, svbfloat16_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzip2_s8)))
 svint8_t svzip2_s8(svint8_t, svint8_t);
 __ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzip2_f64)))
@@ -26219,6 +26369,8 @@ __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svclasta_n_u64)))
 uint64_t svclasta(svbool_t, uint64_t, svuint64_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svclasta_n_u16)))
 uint16_t svclasta(svbool_t, uint16_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svclasta_n_bf16)))
+bfloat16_t svclasta(svbool_t, bfloat16_t, svbfloat16_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svclasta_n_s8)))
 int8_t svclasta(svbool_t, int8_t, svint8_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svclasta_n_f64)))
@@ -26241,6 +26393,8 @@ __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svclasta_u64)))
 svuint64_t svclasta(svbool_t, svuint64_t, svuint64_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svclasta_u16)))
 svuint16_t svclasta(svbool_t, svuint16_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svclasta_bf16)))
+svbfloat16_t svclasta(svbool_t, svbfloat16_t, svbfloat16_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svclasta_s8)))
 svint8_t svclasta(svbool_t, svint8_t, svint8_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svclasta_f64)))
@@ -26263,6 +26417,8 @@ __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svclastb_n_u64)))
 uint64_t svclastb(svbool_t, uint64_t, svuint64_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svclastb_n_u16)))
 uint16_t svclastb(svbool_t, uint16_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svclastb_n_bf16)))
+bfloat16_t svclastb(svbool_t, bfloat16_t, svbfloat16_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svclastb_n_s8)))
 int8_t svclastb(svbool_t, int8_t, svint8_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svclastb_n_f64)))
@@ -26285,6 +26441,8 @@ __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svclastb_u64)))
 svuint64_t svclastb(svbool_t, svuint64_t, svuint64_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svclastb_u16)))
 svuint16_t svclastb(svbool_t, svuint16_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svclastb_bf16)))
+svbfloat16_t svclastb(svbool_t, svbfloat16_t, svbfloat16_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svclastb_s8)))
 svint8_t svclastb(svbool_t, svint8_t, svint8_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svclastb_f64)))
@@ -26845,6 +27003,8 @@ __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcnt_u64_m)))
 svuint64_t svcnt_m(svuint64_t, svbool_t, svuint64_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcnt_u16_m)))
 svuint16_t svcnt_m(svuint16_t, svbool_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcnt_bf16_m)))
+svuint16_t svcnt_m(svuint16_t, svbool_t, svbfloat16_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcnt_s8_m)))
 svuint8_t svcnt_m(svuint8_t, svbool_t, svint8_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcnt_f64_m)))
@@ -26867,6 +27027,8 @@ __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcnt_u64_x)))
 svuint64_t svcnt_x(svbool_t, svuint64_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcnt_u16_x)))
 svuint16_t svcnt_x(svbool_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcnt_bf16_x)))
+svuint16_t svcnt_x(svbool_t, svbfloat16_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcnt_s8_x)))
 svuint8_t svcnt_x(svbool_t, svint8_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcnt_f64_x)))
@@ -26889,6 +27051,8 @@ __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcnt_u64_z)))
 svuint64_t svcnt_z(svbool_t, svuint64_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcnt_u16_z)))
 svuint16_t svcnt_z(svbool_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcnt_bf16_z)))
+svuint16_t svcnt_z(svbool_t, svbfloat16_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcnt_s8_z)))
 svuint8_t svcnt_z(svbool_t, svint8_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcnt_f64_z)))
@@ -26911,6 +27075,8 @@ __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcreate2_u64)))
 svuint64x2_t svcreate2(svuint64_t, svuint64_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcreate2_u16)))
 svuint16x2_t svcreate2(svuint16_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcreate2_bf16)))
+svbfloat16x2_t svcreate2(svbfloat16_t, svbfloat16_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcreate2_s8)))
 svint8x2_t svcreate2(svint8_t, svint8_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcreate2_f64)))
@@ -26923,6 +27089,8 @@ __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcreate2_s32)))
 svint32x2_t svcreate2(svint32_t, svint32_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcreate2_s64)))
 svint64x2_t svcreate2(svint64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcreate2_mf8)))
+svmfloat8x2_t svcreate2(svmfloat8_t, svmfloat8_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcreate2_s16)))
 svint16x2_t svcreate2(svint16_t, svint16_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcreate3_u8)))
@@ -26933,6 +27101,8 @@ __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcreate3_u64)))
 svuint64x3_t svcreate3(svuint64_t, svuint64_t, svuint64_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcreate3_u16)))
 svuint16x3_t svcreate3(svuint16_t, svuint16_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcreate3_bf16)))
+svbfloat16x3_t svcreate3(svbfloat16_t, svbfloat16_t, svbfloat16_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcreate3_s8)))
 svint8x3_t svcreate3(svint8_t, svint8_t, svint8_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcreate3_f64)))
@@ -26945,6 +27115,8 @@ __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcreate3_s32)))
 svint32x3_t svcreate3(svint32_t, svint32_t, svint32_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcreate3_s64)))
 svint64x3_t svcreate3(svint64_t, svint64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcreate3_mf8)))
+svmfloat8x3_t svcreate3(svmfloat8_t, svmfloat8_t, svmfloat8_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcreate3_s16)))
 svint16x3_t svcreate3(svint16_t, svint16_t, svint16_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcreate4_u8)))
@@ -26955,6 +27127,8 @@ __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcreate4_u64)))
 svuint64x4_t svcreate4(svuint64_t, svuint64_t, svuint64_t, svuint64_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcreate4_u16)))
 svuint16x4_t svcreate4(svuint16_t, svuint16_t, svuint16_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcreate4_bf16)))
+svbfloat16x4_t svcreate4(svbfloat16_t, svbfloat16_t, svbfloat16_t, svbfloat16_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcreate4_s8)))
 svint8x4_t svcreate4(svint8_t, svint8_t, svint8_t, svint8_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcreate4_f64)))
@@ -26967,6 +27141,8 @@ __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcreate4_s32)))
 svint32x4_t svcreate4(svint32_t, svint32_t, svint32_t, svint32_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcreate4_s64)))
 svint64x4_t svcreate4(svint64_t, svint64_t, svint64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcreate4_mf8)))
+svmfloat8x4_t svcreate4(svmfloat8_t, svmfloat8_t, svmfloat8_t, svmfloat8_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcreate4_s16)))
 svint16x4_t svcreate4(svint16_t, svint16_t, svint16_t, svint16_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svcvt_f16_f32_m)))
@@ -27373,6 +27549,8 @@ __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdup_n_u64)))
 svuint64_t svdup_u64(uint64_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdup_n_u16)))
 svuint16_t svdup_u16(uint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdup_n_bf16)))
+svbfloat16_t svdup_bf16(bfloat16_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdup_n_s8)))
 svint8_t svdup_s8(int8_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdup_n_f64)))
@@ -27395,6 +27573,8 @@ __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdup_n_u64_m)))
 svuint64_t svdup_u64_m(svuint64_t, svbool_t, uint64_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdup_n_u16_m)))
 svuint16_t svdup_u16_m(svuint16_t, svbool_t, uint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdup_n_bf16_m)))
+svbfloat16_t svdup_bf16_m(svbfloat16_t, svbool_t, bfloat16_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdup_n_s8_m)))
 svint8_t svdup_s8_m(svint8_t, svbool_t, int8_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdup_n_f64_m)))
@@ -27425,6 +27605,8 @@ __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdup_n_u64_x)))
 svuint64_t svdup_u64_x(svbool_t, uint64_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdup_n_u16_x)))
 svuint16_t svdup_u16_x(svbool_t, uint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdup_n_bf16_x)))
+svbfloat16_t svdup_bf16_x(svbool_t, bfloat16_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdup_n_s8_x)))
 svint8_t svdup_s8_x(svbool_t, int8_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdup_n_f64_x)))
@@ -27447,6 +27629,8 @@ __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdup_n_u64_z)))
 svuint64_t svdup_u64_z(svbool_t, uint64_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdup_n_u16_z)))
 svuint16_t svdup_u16_z(svbool_t, uint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdup_n_bf16_z)))
+svbfloat16_t svdup_bf16_z(svbool_t, bfloat16_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdup_n_s8_z)))
 svint8_t svdup_s8_z(svbool_t, int8_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdup_n_f64_z)))
@@ -27469,6 +27653,8 @@ __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdup_lane_u64)))
 svuint64_t svdup_lane(svuint64_t, uint64_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdup_lane_u16)))
 svuint16_t svdup_lane(svuint16_t, uint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdup_lane_bf16)))
+svbfloat16_t svdup_lane(svbfloat16_t, uint16_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdup_lane_s8)))
 svint8_t svdup_lane(svint8_t, uint8_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdup_lane_f64)))
@@ -27489,6 +27675,8 @@ __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdupq_n_s8)))
 svint8_t svdupq_s8(int8_t, int8_t, int8_t, int8_t, int8_t, int8_t, int8_t, int8_t, int8_t, int8_t, int8_t, int8_t, int8_t, int8_t, int8_t, int8_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdupq_n_u16)))
 svuint16_t svdupq_u16(uint16_t, uint16_t, uint16_t, uint16_t, uint16_t, uint16_t, uint16_t, uint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdupq_n_bf16)))
+svbfloat16_t svdupq_bf16(bfloat16_t, bfloat16_t, bfloat16_t, bfloat16_t, bfloat16_t, bfloat16_t, bfloat16_t, bfloat16_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdupq_n_f16)))
 svfloat16_t svdupq_f16(float16_t, float16_t, float16_t, float16_t, float16_t, float16_t, float16_t, float16_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdupq_n_s16)))
@@ -27521,6 +27709,8 @@ __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdupq_lane_u64)))
 svuint64_t svdupq_lane(svuint64_t, uint64_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdupq_lane_u16)))
 svuint16_t svdupq_lane(svuint16_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdupq_lane_bf16)))
+svbfloat16_t svdupq_lane(svbfloat16_t, uint64_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdupq_lane_s8)))
 svint8_t svdupq_lane(svint8_t, uint64_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svdupq_lane_f64)))
@@ -27657,6 +27847,8 @@ __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svext_u64)))
 svuint64_t svext(svuint64_t, svuint64_t, uint64_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svext_u16)))
 svuint16_t svext(svuint16_t, svuint16_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svext_bf16)))
+svbfloat16_t svext(svbfloat16_t, svbfloat16_t, uint64_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svext_s8)))
 svint8_t svext(svint8_t, svint8_t, uint64_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svext_f64)))
@@ -27751,6 +27943,8 @@ __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svget2_u64)))
 svuint64_t svget2(svuint64x2_t, uint64_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svget2_u16)))
 svuint16_t svget2(svuint16x2_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svget2_bf16)))
+svbfloat16_t svget2(svbfloat16x2_t, uint64_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svget2_s8)))
 svint8_t svget2(svint8x2_t, uint64_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svget2_f64)))
@@ -27763,6 +27957,8 @@ __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svget2_s32)))
 svint32_t svget2(svint32x2_t, uint64_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svget2_s64)))
 svint64_t svget2(svint64x2_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svget2_mf8)))
+svmfloat8_t svget2(svmfloat8x2_t, uint64_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svget2_s16)))
 svint16_t svget2(svint16x2_t, uint64_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svget3_u8)))
@@ -27773,6 +27969,8 @@ __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svget3_u64)))
 svuint64_t svget3(svuint64x3_t, uint64_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svget3_u16)))
 svuint16_t svget3(svuint16x3_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svget3_bf16)))
+svbfloat16_t svget3(svbfloat16x3_t, uint64_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svget3_s8)))
 svint8_t svget3(svint8x3_t, uint64_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svget3_f64)))
@@ -27785,6 +27983,8 @@ __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svget3_s32)))
 svint32_t svget3(svint32x3_t, uint64_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svget3_s64)))
 svint64_t svget3(svint64x3_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svget3_mf8)))
+svmfloat8_t svget3(svmfloat8x3_t, uint64_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svget3_s16)))
 svint16_t svget3(svint16x3_t, uint64_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svget4_u8)))
@@ -27795,6 +27995,8 @@ __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svget4_u64)))
 svuint64_t svget4(svuint64x4_t, uint64_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svget4_u16)))
 svuint16_t svget4(svuint16x4_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svget4_bf16)))
+svbfloat16_t svget4(svbfloat16x4_t, uint64_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svget4_s8)))
 svint8_t svget4(svint8x4_t, uint64_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svget4_f64)))
@@ -27807,6 +28009,8 @@ __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svget4_s32)))
 svint32_t svget4(svint32x4_t, uint64_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svget4_s64)))
 svint64_t svget4(svint64x4_t, uint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svget4_mf8)))
+svmfloat8_t svget4(svmfloat8x4_t, uint64_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svget4_s16)))
 svint16_t svget4(svint16x4_t, uint64_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svinsr_n_u8)))
@@ -27817,6 +28021,8 @@ __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svinsr_n_u64)))
 svuint64_t svinsr(svuint64_t, uint64_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svinsr_n_u16)))
 svuint16_t svinsr(svuint16_t, uint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svinsr_n_bf16)))
+svbfloat16_t svinsr(svbfloat16_t, bfloat16_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svinsr_n_s8)))
 svint8_t svinsr(svint8_t, int8_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svinsr_n_f64)))
@@ -27839,6 +28045,8 @@ __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlasta_u64)))
 uint64_t svlasta(svbool_t, svuint64_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlasta_u16)))
 uint16_t svlasta(svbool_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlasta_bf16)))
+bfloat16_t svlasta(svbool_t, svbfloat16_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlasta_s8)))
 int8_t svlasta(svbool_t, svint8_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlasta_f64)))
@@ -27861,6 +28069,8 @@ __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlastb_u64)))
 uint64_t svlastb(svbool_t, svuint64_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlastb_u16)))
 uint16_t svlastb(svbool_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlastb_bf16)))
+bfloat16_t svlastb(svbool_t, svbfloat16_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlastb_s8)))
 int8_t svlastb(svbool_t, svint8_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlastb_f64)))
@@ -27883,6 +28093,8 @@ __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1_u64)))
 svuint64_t svld1(svbool_t, uint64_t const *);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1_u16)))
 svuint16_t svld1(svbool_t, uint16_t const *);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1_bf16)))
+svbfloat16_t svld1(svbool_t, bfloat16_t const *);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1_s8)))
 svint8_t svld1(svbool_t, int8_t const *);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1_f64)))
@@ -27895,6 +28107,8 @@ __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1_s32)))
 svint32_t svld1(svbool_t, int32_t const *);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1_s64)))
 svint64_t svld1(svbool_t, int64_t const *);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1_mf8)))
+svmfloat8_t svld1(svbool_t, mfloat8_t const *);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1_s16)))
 svint16_t svld1(svbool_t, int16_t const *);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1_vnum_u8)))
@@ -27905,6 +28119,8 @@ __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1_vnum_u64)))
 svuint64_t svld1_vnum(svbool_t, uint64_t const *, int64_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1_vnum_u16)))
 svuint16_t svld1_vnum(svbool_t, uint16_t const *, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1_vnum_bf16)))
+svbfloat16_t svld1_vnum(svbool_t, bfloat16_t const *, int64_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1_vnum_s8)))
 svint8_t svld1_vnum(svbool_t, int8_t const *, int64_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1_vnum_f64)))
@@ -27917,6 +28133,8 @@ __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1_vnum_s32)))
 svint32_t svld1_vnum(svbool_t, int32_t const *, int64_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1_vnum_s64)))
 svint64_t svld1_vnum(svbool_t, int64_t const *, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1_vnum_mf8)))
+svmfloat8_t svld1_vnum(svbool_t, mfloat8_t const *, int64_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1_vnum_s16)))
 svint16_t svld1_vnum(svbool_t, int16_t const *, int64_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1rq_u8)))
@@ -27927,6 +28145,8 @@ __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1rq_u64)))
 svuint64_t svld1rq(svbool_t, uint64_t const *);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1rq_u16)))
 svuint16_t svld1rq(svbool_t, uint16_t const *);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1rq_bf16)))
+svbfloat16_t svld1rq(svbool_t, bfloat16_t const *);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1rq_s8)))
 svint8_t svld1rq(svbool_t, int8_t const *);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1rq_f64)))
@@ -27939,6 +28159,8 @@ __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1rq_s32)))
 svint32_t svld1rq(svbool_t, int32_t const *);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1rq_s64)))
 svint64_t svld1rq(svbool_t, int64_t const *);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1rq_mf8)))
+svmfloat8_t svld1rq(svbool_t, mfloat8_t const *);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld1rq_s16)))
 svint16_t svld1rq(svbool_t, int16_t const *);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld2_u8)))
@@ -27949,6 +28171,8 @@ __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld2_u64)))
 svuint64x2_t svld2(svbool_t, uint64_t const *);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld2_u16)))
 svuint16x2_t svld2(svbool_t, uint16_t const *);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld2_bf16)))
+svbfloat16x2_t svld2(svbool_t, bfloat16_t const *);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld2_s8)))
 svint8x2_t svld2(svbool_t, int8_t const *);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld2_f64)))
@@ -27961,6 +28185,8 @@ __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld2_s32)))
 svint32x2_t svld2(svbool_t, int32_t const *);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld2_s64)))
 svint64x2_t svld2(svbool_t, int64_t const *);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld2_mf8)))
+svmfloat8x2_t svld2(svbool_t, mfloat8_t const *);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld2_s16)))
 svint16x2_t svld2(svbool_t, int16_t const *);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld2_vnum_u8)))
@@ -27971,6 +28197,8 @@ __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld2_vnum_u64)))
 svuint64x2_t svld2_vnum(svbool_t, uint64_t const *, int64_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld2_vnum_u16)))
 svuint16x2_t svld2_vnum(svbool_t, uint16_t const *, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld2_vnum_bf16)))
+svbfloat16x2_t svld2_vnum(svbool_t, bfloat16_t const *, int64_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld2_vnum_s8)))
 svint8x2_t svld2_vnum(svbool_t, int8_t const *, int64_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld2_vnum_f64)))
@@ -27983,6 +28211,8 @@ __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld2_vnum_s32)))
 svint32x2_t svld2_vnum(svbool_t, int32_t const *, int64_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld2_vnum_s64)))
 svint64x2_t svld2_vnum(svbool_t, int64_t const *, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld2_vnum_mf8)))
+svmfloat8x2_t svld2_vnum(svbool_t, mfloat8_t const *, int64_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld2_vnum_s16)))
 svint16x2_t svld2_vnum(svbool_t, int16_t const *, int64_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld3_u8)))
@@ -27993,6 +28223,8 @@ __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld3_u64)))
 svuint64x3_t svld3(svbool_t, uint64_t const *);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld3_u16)))
 svuint16x3_t svld3(svbool_t, uint16_t const *);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld3_bf16)))
+svbfloat16x3_t svld3(svbool_t, bfloat16_t const *);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld3_s8)))
 svint8x3_t svld3(svbool_t, int8_t const *);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld3_f64)))
@@ -28005,6 +28237,8 @@ __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld3_s32)))
 svint32x3_t svld3(svbool_t, int32_t const *);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld3_s64)))
 svint64x3_t svld3(svbool_t, int64_t const *);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld3_mf8)))
+svmfloat8x3_t svld3(svbool_t, mfloat8_t const *);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld3_s16)))
 svint16x3_t svld3(svbool_t, int16_t const *);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld3_vnum_u8)))
@@ -28015,6 +28249,8 @@ __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld3_vnum_u64)))
 svuint64x3_t svld3_vnum(svbool_t, uint64_t const *, int64_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld3_vnum_u16)))
 svuint16x3_t svld3_vnum(svbool_t, uint16_t const *, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld3_vnum_bf16)))
+svbfloat16x3_t svld3_vnum(svbool_t, bfloat16_t const *, int64_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld3_vnum_s8)))
 svint8x3_t svld3_vnum(svbool_t, int8_t const *, int64_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld3_vnum_f64)))
@@ -28027,6 +28263,8 @@ __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld3_vnum_s32)))
 svint32x3_t svld3_vnum(svbool_t, int32_t const *, int64_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld3_vnum_s64)))
 svint64x3_t svld3_vnum(svbool_t, int64_t const *, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld3_vnum_mf8)))
+svmfloat8x3_t svld3_vnum(svbool_t, mfloat8_t const *, int64_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld3_vnum_s16)))
 svint16x3_t svld3_vnum(svbool_t, int16_t const *, int64_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld4_u8)))
@@ -28037,6 +28275,8 @@ __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld4_u64)))
 svuint64x4_t svld4(svbool_t, uint64_t const *);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld4_u16)))
 svuint16x4_t svld4(svbool_t, uint16_t const *);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld4_bf16)))
+svbfloat16x4_t svld4(svbool_t, bfloat16_t const *);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld4_s8)))
 svint8x4_t svld4(svbool_t, int8_t const *);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld4_f64)))
@@ -28049,6 +28289,8 @@ __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld4_s32)))
 svint32x4_t svld4(svbool_t, int32_t const *);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld4_s64)))
 svint64x4_t svld4(svbool_t, int64_t const *);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld4_mf8)))
+svmfloat8x4_t svld4(svbool_t, mfloat8_t const *);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld4_s16)))
 svint16x4_t svld4(svbool_t, int16_t const *);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld4_vnum_u8)))
@@ -28059,6 +28301,8 @@ __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld4_vnum_u64)))
 svuint64x4_t svld4_vnum(svbool_t, uint64_t const *, int64_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld4_vnum_u16)))
 svuint16x4_t svld4_vnum(svbool_t, uint16_t const *, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld4_vnum_bf16)))
+svbfloat16x4_t svld4_vnum(svbool_t, bfloat16_t const *, int64_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld4_vnum_s8)))
 svint8x4_t svld4_vnum(svbool_t, int8_t const *, int64_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld4_vnum_f64)))
@@ -28071,6 +28315,8 @@ __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld4_vnum_s32)))
 svint32x4_t svld4_vnum(svbool_t, int32_t const *, int64_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld4_vnum_s64)))
 svint64x4_t svld4_vnum(svbool_t, int64_t const *, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld4_vnum_mf8)))
+svmfloat8x4_t svld4_vnum(svbool_t, mfloat8_t const *, int64_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svld4_vnum_s16)))
 svint16x4_t svld4_vnum(svbool_t, int16_t const *, int64_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_u8)))
@@ -28081,6 +28327,8 @@ __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_u64)))
 svuint64_t svldnt1(svbool_t, uint64_t const *);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_u16)))
 svuint16_t svldnt1(svbool_t, uint16_t const *);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_bf16)))
+svbfloat16_t svldnt1(svbool_t, bfloat16_t const *);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_s8)))
 svint8_t svldnt1(svbool_t, int8_t const *);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_f64)))
@@ -28093,6 +28341,8 @@ __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_s32)))
 svint32_t svldnt1(svbool_t, int32_t const *);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_s64)))
 svint64_t svldnt1(svbool_t, int64_t const *);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_mf8)))
+svmfloat8_t svldnt1(svbool_t, mfloat8_t const *);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_s16)))
 svint16_t svldnt1(svbool_t, int16_t const *);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_vnum_u8)))
@@ -28103,6 +28353,8 @@ __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_vnum_u64)))
 svuint64_t svldnt1_vnum(svbool_t, uint64_t const *, int64_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_vnum_u16)))
 svuint16_t svldnt1_vnum(svbool_t, uint16_t const *, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_vnum_bf16)))
+svbfloat16_t svldnt1_vnum(svbool_t, bfloat16_t const *, int64_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_vnum_s8)))
 svint8_t svldnt1_vnum(svbool_t, int8_t const *, int64_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_vnum_f64)))
@@ -28115,6 +28367,8 @@ __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_vnum_s32)))
 svint32_t svldnt1_vnum(svbool_t, int32_t const *, int64_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_vnum_s64)))
 svint64_t svldnt1_vnum(svbool_t, int64_t const *, int64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_vnum_mf8)))
+svmfloat8_t svldnt1_vnum(svbool_t, mfloat8_t const *, int64_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svldnt1_vnum_s16)))
 svint16_t svldnt1_vnum(svbool_t, int16_t const *, int64_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlen_u8)))
@@ -28125,6 +28379,8 @@ __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlen_u64)))
 uint64_t svlen(svuint64_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlen_u16)))
 uint64_t svlen(svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlen_bf16)))
+uint64_t svlen(svbfloat16_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlen_s8)))
 uint64_t svlen(svint8_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svlen_f64)))
@@ -30369,6 +30625,8 @@ __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrev_u64)))
 svuint64_t svrev(svuint64_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrev_u16)))
 svuint16_t svrev(svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrev_bf16)))
+svbfloat16_t svrev(svbfloat16_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrev_s8)))
 svint8_t svrev(svint8_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svrev_f64)))
@@ -30639,6 +30897,8 @@ __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsel_u64)))
 svuint64_t svsel(svbool_t, svuint64_t, svuint64_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsel_u16)))
 svuint16_t svsel(svbool_t, svuint16_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsel_bf16)))
+svbfloat16_t svsel(svbool_t, svbfloat16_t, svbfloat16_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsel_s8)))
 svint8_t svsel(svbool_t, svint8_t, svint8_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsel_f64)))
@@ -30661,6 +30921,8 @@ __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svset2_u64)))
 svuint64x2_t svset2(svuint64x2_t, uint64_t, svuint64_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svset2_u16)))
 svuint16x2_t svset2(svuint16x2_t, uint64_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svset2_bf16)))
+svbfloat16x2_t svset2(svbfloat16x2_t, uint64_t, svbfloat16_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svset2_s8)))
 svint8x2_t svset2(svint8x2_t, uint64_t, svint8_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svset2_f64)))
@@ -30673,6 +30935,8 @@ __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svset2_s32)))
 svint32x2_t svset2(svint32x2_t, uint64_t, svint32_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svset2_s64)))
 svint64x2_t svset2(svint64x2_t, uint64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svset2_mf8)))
+svmfloat8x2_t svset2(svmfloat8x2_t, uint64_t, svmfloat8_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svset2_s16)))
 svint16x2_t svset2(svint16x2_t, uint64_t, svint16_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svset3_u8)))
@@ -30683,6 +30947,8 @@ __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svset3_u64)))
 svuint64x3_t svset3(svuint64x3_t, uint64_t, svuint64_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svset3_u16)))
 svuint16x3_t svset3(svuint16x3_t, uint64_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svset3_bf16)))
+svbfloat16x3_t svset3(svbfloat16x3_t, uint64_t, svbfloat16_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svset3_s8)))
 svint8x3_t svset3(svint8x3_t, uint64_t, svint8_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svset3_f64)))
@@ -30695,6 +30961,8 @@ __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svset3_s32)))
 svint32x3_t svset3(svint32x3_t, uint64_t, svint32_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svset3_s64)))
 svint64x3_t svset3(svint64x3_t, uint64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svset3_mf8)))
+svmfloat8x3_t svset3(svmfloat8x3_t, uint64_t, svmfloat8_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svset3_s16)))
 svint16x3_t svset3(svint16x3_t, uint64_t, svint16_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svset4_u8)))
@@ -30705,6 +30973,8 @@ __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svset4_u64)))
 svuint64x4_t svset4(svuint64x4_t, uint64_t, svuint64_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svset4_u16)))
 svuint16x4_t svset4(svuint16x4_t, uint64_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svset4_bf16)))
+svbfloat16x4_t svset4(svbfloat16x4_t, uint64_t, svbfloat16_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svset4_s8)))
 svint8x4_t svset4(svint8x4_t, uint64_t, svint8_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svset4_f64)))
@@ -30717,6 +30987,8 @@ __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svset4_s32)))
 svint32x4_t svset4(svint32x4_t, uint64_t, svint32_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svset4_s64)))
 svint64x4_t svset4(svint64x4_t, uint64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svset4_mf8)))
+svmfloat8x4_t svset4(svmfloat8x4_t, uint64_t, svmfloat8_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svset4_s16)))
 svint16x4_t svset4(svint16x4_t, uint64_t, svint16_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsplice_u8)))
@@ -30727,6 +30999,8 @@ __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsplice_u64)))
 svuint64_t svsplice(svbool_t, svuint64_t, svuint64_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsplice_u16)))
 svuint16_t svsplice(svbool_t, svuint16_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsplice_bf16)))
+svbfloat16_t svsplice(svbool_t, svbfloat16_t, svbfloat16_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsplice_s8)))
 svint8_t svsplice(svbool_t, svint8_t, svint8_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsplice_f64)))
@@ -30767,6 +31041,8 @@ __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1_u64)))
 void svst1(svbool_t, uint64_t *, svuint64_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1_u16)))
 void svst1(svbool_t, uint16_t *, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1_bf16)))
+void svst1(svbool_t, bfloat16_t *, svbfloat16_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1_s8)))
 void svst1(svbool_t, int8_t *, svint8_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1_f64)))
@@ -30779,6 +31055,8 @@ __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1_s32)))
 void svst1(svbool_t, int32_t *, svint32_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1_s64)))
 void svst1(svbool_t, int64_t *, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1_mf8)))
+void svst1(svbool_t, mfloat8_t *, svmfloat8_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1_s16)))
 void svst1(svbool_t, int16_t *, svint16_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1_vnum_u8)))
@@ -30789,6 +31067,8 @@ __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1_vnum_u64)))
 void svst1_vnum(svbool_t, uint64_t *, int64_t, svuint64_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1_vnum_u16)))
 void svst1_vnum(svbool_t, uint16_t *, int64_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1_vnum_bf16)))
+void svst1_vnum(svbool_t, bfloat16_t *, int64_t, svbfloat16_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1_vnum_s8)))
 void svst1_vnum(svbool_t, int8_t *, int64_t, svint8_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1_vnum_f64)))
@@ -30801,6 +31081,8 @@ __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1_vnum_s32)))
 void svst1_vnum(svbool_t, int32_t *, int64_t, svint32_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1_vnum_s64)))
 void svst1_vnum(svbool_t, int64_t *, int64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1_vnum_mf8)))
+void svst1_vnum(svbool_t, mfloat8_t *, int64_t, svmfloat8_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1_vnum_s16)))
 void svst1_vnum(svbool_t, int16_t *, int64_t, svint16_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst1b_s32)))
@@ -30859,6 +31141,8 @@ __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst2_u64)))
 void svst2(svbool_t, uint64_t *, svuint64x2_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst2_u16)))
 void svst2(svbool_t, uint16_t *, svuint16x2_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst2_bf16)))
+void svst2(svbool_t, bfloat16_t *, svbfloat16x2_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst2_s8)))
 void svst2(svbool_t, int8_t *, svint8x2_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst2_f64)))
@@ -30871,6 +31155,8 @@ __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst2_s32)))
 void svst2(svbool_t, int32_t *, svint32x2_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst2_s64)))
 void svst2(svbool_t, int64_t *, svint64x2_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst2_mf8)))
+void svst2(svbool_t, mfloat8_t *, svmfloat8x2_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst2_s16)))
 void svst2(svbool_t, int16_t *, svint16x2_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst2_vnum_u8)))
@@ -30881,6 +31167,8 @@ __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst2_vnum_u64)))
 void svst2_vnum(svbool_t, uint64_t *, int64_t, svuint64x2_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst2_vnum_u16)))
 void svst2_vnum(svbool_t, uint16_t *, int64_t, svuint16x2_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst2_vnum_bf16)))
+void svst2_vnum(svbool_t, bfloat16_t *, int64_t, svbfloat16x2_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst2_vnum_s8)))
 void svst2_vnum(svbool_t, int8_t *, int64_t, svint8x2_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst2_vnum_f64)))
@@ -30893,6 +31181,8 @@ __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst2_vnum_s32)))
 void svst2_vnum(svbool_t, int32_t *, int64_t, svint32x2_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst2_vnum_s64)))
 void svst2_vnum(svbool_t, int64_t *, int64_t, svint64x2_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst2_vnum_mf8)))
+void svst2_vnum(svbool_t, mfloat8_t *, int64_t, svmfloat8x2_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst2_vnum_s16)))
 void svst2_vnum(svbool_t, int16_t *, int64_t, svint16x2_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst3_u8)))
@@ -30903,6 +31193,8 @@ __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst3_u64)))
 void svst3(svbool_t, uint64_t *, svuint64x3_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst3_u16)))
 void svst3(svbool_t, uint16_t *, svuint16x3_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst3_bf16)))
+void svst3(svbool_t, bfloat16_t *, svbfloat16x3_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst3_s8)))
 void svst3(svbool_t, int8_t *, svint8x3_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst3_f64)))
@@ -30915,6 +31207,8 @@ __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst3_s32)))
 void svst3(svbool_t, int32_t *, svint32x3_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst3_s64)))
 void svst3(svbool_t, int64_t *, svint64x3_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst3_mf8)))
+void svst3(svbool_t, mfloat8_t *, svmfloat8x3_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst3_s16)))
 void svst3(svbool_t, int16_t *, svint16x3_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst3_vnum_u8)))
@@ -30925,6 +31219,8 @@ __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst3_vnum_u64)))
 void svst3_vnum(svbool_t, uint64_t *, int64_t, svuint64x3_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst3_vnum_u16)))
 void svst3_vnum(svbool_t, uint16_t *, int64_t, svuint16x3_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst3_vnum_bf16)))
+void svst3_vnum(svbool_t, bfloat16_t *, int64_t, svbfloat16x3_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst3_vnum_s8)))
 void svst3_vnum(svbool_t, int8_t *, int64_t, svint8x3_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst3_vnum_f64)))
@@ -30937,6 +31233,8 @@ __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst3_vnum_s32)))
 void svst3_vnum(svbool_t, int32_t *, int64_t, svint32x3_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst3_vnum_s64)))
 void svst3_vnum(svbool_t, int64_t *, int64_t, svint64x3_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst3_vnum_mf8)))
+void svst3_vnum(svbool_t, mfloat8_t *, int64_t, svmfloat8x3_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst3_vnum_s16)))
 void svst3_vnum(svbool_t, int16_t *, int64_t, svint16x3_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst4_u8)))
@@ -30947,6 +31245,8 @@ __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst4_u64)))
 void svst4(svbool_t, uint64_t *, svuint64x4_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst4_u16)))
 void svst4(svbool_t, uint16_t *, svuint16x4_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst4_bf16)))
+void svst4(svbool_t, bfloat16_t *, svbfloat16x4_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst4_s8)))
 void svst4(svbool_t, int8_t *, svint8x4_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst4_f64)))
@@ -30959,6 +31259,8 @@ __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst4_s32)))
 void svst4(svbool_t, int32_t *, svint32x4_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst4_s64)))
 void svst4(svbool_t, int64_t *, svint64x4_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst4_mf8)))
+void svst4(svbool_t, mfloat8_t *, svmfloat8x4_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst4_s16)))
 void svst4(svbool_t, int16_t *, svint16x4_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst4_vnum_u8)))
@@ -30969,6 +31271,8 @@ __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst4_vnum_u64)))
 void svst4_vnum(svbool_t, uint64_t *, int64_t, svuint64x4_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst4_vnum_u16)))
 void svst4_vnum(svbool_t, uint16_t *, int64_t, svuint16x4_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst4_vnum_bf16)))
+void svst4_vnum(svbool_t, bfloat16_t *, int64_t, svbfloat16x4_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst4_vnum_s8)))
 void svst4_vnum(svbool_t, int8_t *, int64_t, svint8x4_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst4_vnum_f64)))
@@ -30981,6 +31285,8 @@ __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst4_vnum_s32)))
 void svst4_vnum(svbool_t, int32_t *, int64_t, svint32x4_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst4_vnum_s64)))
 void svst4_vnum(svbool_t, int64_t *, int64_t, svint64x4_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst4_vnum_mf8)))
+void svst4_vnum(svbool_t, mfloat8_t *, int64_t, svmfloat8x4_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svst4_vnum_s16)))
 void svst4_vnum(svbool_t, int16_t *, int64_t, svint16x4_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_u8)))
@@ -30991,6 +31297,8 @@ __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_u64)))
 void svstnt1(svbool_t, uint64_t *, svuint64_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_u16)))
 void svstnt1(svbool_t, uint16_t *, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_bf16)))
+void svstnt1(svbool_t, bfloat16_t *, svbfloat16_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_s8)))
 void svstnt1(svbool_t, int8_t *, svint8_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_f64)))
@@ -31003,6 +31311,8 @@ __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_s32)))
 void svstnt1(svbool_t, int32_t *, svint32_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_s64)))
 void svstnt1(svbool_t, int64_t *, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_mf8)))
+void svstnt1(svbool_t, mfloat8_t *, svmfloat8_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_s16)))
 void svstnt1(svbool_t, int16_t *, svint16_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_vnum_u8)))
@@ -31013,6 +31323,8 @@ __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_vnum_u64)))
 void svstnt1_vnum(svbool_t, uint64_t *, int64_t, svuint64_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_vnum_u16)))
 void svstnt1_vnum(svbool_t, uint16_t *, int64_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_vnum_bf16)))
+void svstnt1_vnum(svbool_t, bfloat16_t *, int64_t, svbfloat16_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_vnum_s8)))
 void svstnt1_vnum(svbool_t, int8_t *, int64_t, svint8_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_vnum_f64)))
@@ -31025,6 +31337,8 @@ __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_vnum_s32)))
 void svstnt1_vnum(svbool_t, int32_t *, int64_t, svint32_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_vnum_s64)))
 void svstnt1_vnum(svbool_t, int64_t *, int64_t, svint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_vnum_mf8)))
+void svstnt1_vnum(svbool_t, mfloat8_t *, int64_t, svmfloat8_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svstnt1_vnum_s16)))
 void svstnt1_vnum(svbool_t, int16_t *, int64_t, svint16_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svsub_n_f64_m)))
@@ -31299,6 +31613,8 @@ __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtbl_u64)))
 svuint64_t svtbl(svuint64_t, svuint64_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtbl_u16)))
 svuint16_t svtbl(svuint16_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtbl_bf16)))
+svbfloat16_t svtbl(svbfloat16_t, svuint16_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtbl_s8)))
 svint8_t svtbl(svint8_t, svuint8_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtbl_f64)))
@@ -31321,6 +31637,8 @@ __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtrn1_u64)))
 svuint64_t svtrn1(svuint64_t, svuint64_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtrn1_u16)))
 svuint16_t svtrn1(svuint16_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtrn1_bf16)))
+svbfloat16_t svtrn1(svbfloat16_t, svbfloat16_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtrn1_s8)))
 svint8_t svtrn1(svint8_t, svint8_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtrn1_f64)))
@@ -31343,6 +31661,8 @@ __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtrn2_u64)))
 svuint64_t svtrn2(svuint64_t, svuint64_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtrn2_u16)))
 svuint16_t svtrn2(svuint16_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtrn2_bf16)))
+svbfloat16_t svtrn2(svbfloat16_t, svbfloat16_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtrn2_s8)))
 svint8_t svtrn2(svint8_t, svint8_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svtrn2_f64)))
@@ -31393,6 +31713,8 @@ __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzp1_u64)))
 svuint64_t svuzp1(svuint64_t, svuint64_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzp1_u16)))
 svuint16_t svuzp1(svuint16_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzp1_bf16)))
+svbfloat16_t svuzp1(svbfloat16_t, svbfloat16_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzp1_s8)))
 svint8_t svuzp1(svint8_t, svint8_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzp1_f64)))
@@ -31415,6 +31737,8 @@ __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzp2_u64)))
 svuint64_t svuzp2(svuint64_t, svuint64_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzp2_u16)))
 svuint16_t svuzp2(svuint16_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzp2_bf16)))
+svbfloat16_t svuzp2(svbfloat16_t, svbfloat16_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzp2_s8)))
 svint8_t svuzp2(svint8_t, svint8_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svuzp2_f64)))
@@ -31501,6 +31825,8 @@ __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzip1_u64)))
 svuint64_t svzip1(svuint64_t, svuint64_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzip1_u16)))
 svuint16_t svzip1(svuint16_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzip1_bf16)))
+svbfloat16_t svzip1(svbfloat16_t, svbfloat16_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzip1_s8)))
 svint8_t svzip1(svint8_t, svint8_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzip1_f64)))
@@ -31523,6 +31849,8 @@ __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzip2_u64)))
 svuint64_t svzip2(svuint64_t, svuint64_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzip2_u16)))
 svuint16_t svzip2(svuint16_t, svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzip2_bf16)))
+svbfloat16_t svzip2(svbfloat16_t, svbfloat16_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzip2_s8)))
 svint8_t svzip2(svint8_t, svint8_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzip2_f64)))
@@ -31537,6 +31865,18 @@ __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzip2_s64)))
 svint64_t svzip2(svint64_t, svint64_t);
 __aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svzip2_s16)))
 svint16_t svzip2(svint16_t, svint16_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svexpa_f64)))
+svfloat64_t svexpa_f64(svuint64_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svexpa_f32)))
+svfloat32_t svexpa_f32(svuint32_t);
+__ai __attribute__((__clang_arm_builtin_alias(__builtin_sve_svexpa_f16)))
+svfloat16_t svexpa_f16(svuint16_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svexpa_f64)))
+svfloat64_t svexpa(svuint64_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svexpa_f32)))
+svfloat32_t svexpa(svuint32_t);
+__aio __attribute__((__clang_arm_builtin_alias(__builtin_sve_svexpa_f16)))
+svfloat16_t svexpa(svuint16_t);
 #define svcvtnt_bf16_x      svcvtnt_bf16_m
 #define svcvtnt_bf16_f32_x  svcvtnt_bf16_f32_m
 #define svcvtnt_f16_x      svcvtnt_f16_m

lib/include/avx10_2_512bf16intrin.h

@@ -441,8 +441,8 @@ _mm512_maskz_sqrt_pbh(__mmask32 __U, __m512bh __A) {
 
 static __inline__ __m512bh __DEFAULT_FN_ATTRS512
 _mm512_fmadd_pbh(__m512bh __A, __m512bh __B, __m512bh __C) {
-  return (__m512bh)__builtin_ia32_vfmaddnepbh512((__v32bf)__A, (__v32bf)__B,
-                                                 (__v32bf)__C);
+  return (__m512bh)__builtin_ia32_vfmaddbf16512((__v32bf)__A, (__v32bf)__B,
+                                                (__v32bf)__C);
 }
 
 static __inline__ __m512bh __DEFAULT_FN_ATTRS512
@@ -469,8 +469,8 @@ static __inline__ __m512bh __DEFAULT_FN_ATTRS512 _mm512_maskz_fmadd_pbh(
 
 static __inline__ __m512bh __DEFAULT_FN_ATTRS512
 _mm512_fmsub_pbh(__m512bh __A, __m512bh __B, __m512bh __C) {
-  return (__m512bh)__builtin_ia32_vfmaddnepbh512((__v32bf)__A, (__v32bf)__B,
-                                                 -(__v32bf)__C);
+  return (__m512bh)__builtin_ia32_vfmaddbf16512((__v32bf)__A, (__v32bf)__B,
+                                                -(__v32bf)__C);
 }
 
 static __inline__ __m512bh __DEFAULT_FN_ATTRS512
@@ -497,8 +497,8 @@ static __inline__ __m512bh __DEFAULT_FN_ATTRS512 _mm512_maskz_fmsub_pbh(
 
 static __inline__ __m512bh __DEFAULT_FN_ATTRS512
 _mm512_fnmadd_pbh(__m512bh __A, __m512bh __B, __m512bh __C) {
-  return (__m512bh)__builtin_ia32_vfmaddnepbh512((__v32bf)__A, -(__v32bf)__B,
-                                                 (__v32bf)__C);
+  return (__m512bh)__builtin_ia32_vfmaddbf16512((__v32bf)__A, -(__v32bf)__B,
+                                                (__v32bf)__C);
 }
 
 static __inline__ __m512bh __DEFAULT_FN_ATTRS512 _mm512_mask_fnmadd_pbh(
@@ -527,8 +527,8 @@ static __inline__ __m512bh __DEFAULT_FN_ATTRS512 _mm512_maskz_fnmadd_pbh(
 
 static __inline__ __m512bh __DEFAULT_FN_ATTRS512
 _mm512_fnmsub_pbh(__m512bh __A, __m512bh __B, __m512bh __C) {
-  return (__m512bh)__builtin_ia32_vfmaddnepbh512((__v32bf)__A, -(__v32bf)__B,
-                                                 -(__v32bf)__C);
+  return (__m512bh)__builtin_ia32_vfmaddbf16512((__v32bf)__A, -(__v32bf)__B,
+                                                -(__v32bf)__C);
 }
 
 static __inline__ __m512bh __DEFAULT_FN_ATTRS512 _mm512_mask_fnmsub_pbh(

lib/include/avx10_2_512convertintrin.h

@@ -78,20 +78,20 @@ _mm512_maskz_cvtbiasph_bf8(__mmask32 __U, __m512i __A, __m512h __B) {
 }
 
 static __inline__ __m256i __DEFAULT_FN_ATTRS512
-_mm512_cvtbiassph_bf8(__m512i __A, __m512h __B) {
+_mm512_cvts_biasph_bf8(__m512i __A, __m512h __B) {
   return (__m256i)__builtin_ia32_vcvtbiasph2bf8s_512_mask(
       (__v64qi)__A, (__v32hf)__B, (__v32qi)_mm256_undefined_si256(),
       (__mmask32)-1);
 }
 
-static __inline__ __m256i __DEFAULT_FN_ATTRS512 _mm512_mask_cvtbiassph_bf8(
+static __inline__ __m256i __DEFAULT_FN_ATTRS512 _mm512_mask_cvts_biasph_bf8(
     __m256i __W, __mmask32 __U, __m512i __A, __m512h __B) {
   return (__m256i)__builtin_ia32_vcvtbiasph2bf8s_512_mask(
       (__v64qi)__A, (__v32hf)__B, (__v32qi)(__m256i)__W, (__mmask32)__U);
 }
 
 static __inline__ __m256i __DEFAULT_FN_ATTRS512
-_mm512_maskz_cvtbiassph_bf8(__mmask32 __U, __m512i __A, __m512h __B) {
+_mm512_maskz_cvts_biasph_bf8(__mmask32 __U, __m512i __A, __m512h __B) {
   return (__m256i)__builtin_ia32_vcvtbiasph2bf8s_512_mask(
       (__v64qi)__A, (__v32hf)__B, (__v32qi)(__m256i)_mm256_setzero_si256(),
       (__mmask32)__U);
@@ -118,20 +118,20 @@ _mm512_maskz_cvtbiasph_hf8(__mmask32 __U, __m512i __A, __m512h __B) {
 }
 
 static __inline__ __m256i __DEFAULT_FN_ATTRS512
-_mm512_cvtbiassph_hf8(__m512i __A, __m512h __B) {
+_mm512_cvts_biasph_hf8(__m512i __A, __m512h __B) {
   return (__m256i)__builtin_ia32_vcvtbiasph2hf8s_512_mask(
       (__v64qi)__A, (__v32hf)__B, (__v32qi)_mm256_undefined_si256(),
       (__mmask32)-1);
 }
 
-static __inline__ __m256i __DEFAULT_FN_ATTRS512 _mm512_mask_cvtbiassph_hf8(
+static __inline__ __m256i __DEFAULT_FN_ATTRS512 _mm512_mask_cvts_biasph_hf8(
     __m256i __W, __mmask32 __U, __m512i __A, __m512h __B) {
   return (__m256i)__builtin_ia32_vcvtbiasph2hf8s_512_mask(
       (__v64qi)__A, (__v32hf)__B, (__v32qi)(__m256i)__W, (__mmask32)__U);
 }
 
 static __inline__ __m256i __DEFAULT_FN_ATTRS512
-_mm512_maskz_cvtbiassph_hf8(__mmask32 __U, __m512i __A, __m512h __B) {
+_mm512_maskz_cvts_biasph_hf8(__mmask32 __U, __m512i __A, __m512h __B) {
   return (__m256i)__builtin_ia32_vcvtbiasph2hf8s_512_mask(
       (__v64qi)__A, (__v32hf)__B, (__v32qi)(__m256i)_mm256_setzero_si256(),
       (__mmask32)__U);
@@ -157,21 +157,21 @@ _mm512_maskz_cvt2ph_bf8(__mmask64 __U, __m512h __A, __m512h __B) {
 }
 
 static __inline__ __m512i __DEFAULT_FN_ATTRS512
-_mm512_cvts2ph_bf8(__m512h __A, __m512h __B) {
+_mm512_cvts_2ph_bf8(__m512h __A, __m512h __B) {
   return (__m512i)__builtin_ia32_vcvt2ph2bf8s_512((__v32hf)(__A),
                                                   (__v32hf)(__B));
 }
 
 static __inline__ __m512i __DEFAULT_FN_ATTRS512
-_mm512_mask_cvts2ph_bf8(__m512i __W, __mmask64 __U, __m512h __A, __m512h __B) {
+_mm512_mask_cvts_2ph_bf8(__m512i __W, __mmask64 __U, __m512h __A, __m512h __B) {
   return (__m512i)__builtin_ia32_selectb_512(
-      (__mmask64)__U, (__v64qi)_mm512_cvts2ph_bf8(__A, __B), (__v64qi)__W);
+      (__mmask64)__U, (__v64qi)_mm512_cvts_2ph_bf8(__A, __B), (__v64qi)__W);
 }
 
 static __inline__ __m512i __DEFAULT_FN_ATTRS512
-_mm512_maskz_cvts2ph_bf8(__mmask64 __U, __m512h __A, __m512h __B) {
+_mm512_maskz_cvts_2ph_bf8(__mmask64 __U, __m512h __A, __m512h __B) {
   return (__m512i)__builtin_ia32_selectb_512(
-      (__mmask64)__U, (__v64qi)_mm512_cvts2ph_bf8(__A, __B),
+      (__mmask64)__U, (__v64qi)_mm512_cvts_2ph_bf8(__A, __B),
       (__v64qi)(__m512i)_mm512_setzero_si512());
 }
 
@@ -195,37 +195,37 @@ _mm512_maskz_cvt2ph_hf8(__mmask64 __U, __m512h __A, __m512h __B) {
 }
 
 static __inline__ __m512i __DEFAULT_FN_ATTRS512
-_mm512_cvts2ph_hf8(__m512h __A, __m512h __B) {
+_mm512_cvts_2ph_hf8(__m512h __A, __m512h __B) {
   return (__m512i)__builtin_ia32_vcvt2ph2hf8s_512((__v32hf)(__A),
                                                   (__v32hf)(__B));
 }
 
 static __inline__ __m512i __DEFAULT_FN_ATTRS512
-_mm512_mask_cvts2ph_hf8(__m512i __W, __mmask64 __U, __m512h __A, __m512h __B) {
+_mm512_mask_cvts_2ph_hf8(__m512i __W, __mmask64 __U, __m512h __A, __m512h __B) {
   return (__m512i)__builtin_ia32_selectb_512(
-      (__mmask64)__U, (__v64qi)_mm512_cvts2ph_hf8(__A, __B), (__v64qi)__W);
+      (__mmask64)__U, (__v64qi)_mm512_cvts_2ph_hf8(__A, __B), (__v64qi)__W);
 }
 
 static __inline__ __m512i __DEFAULT_FN_ATTRS512
-_mm512_maskz_cvts2ph_hf8(__mmask64 __U, __m512h __A, __m512h __B) {
+_mm512_maskz_cvts_2ph_hf8(__mmask64 __U, __m512h __A, __m512h __B) {
   return (__m512i)__builtin_ia32_selectb_512(
-      (__mmask64)__U, (__v64qi)_mm512_cvts2ph_hf8(__A, __B),
+      (__mmask64)__U, (__v64qi)_mm512_cvts_2ph_hf8(__A, __B),
       (__v64qi)(__m512i)_mm512_setzero_si512());
 }
 
-static __inline__ __m512h __DEFAULT_FN_ATTRS512 _mm512_cvthf8(__m256i __A) {
+static __inline__ __m512h __DEFAULT_FN_ATTRS512 _mm512_cvthf8_ph(__m256i __A) {
   return (__m512h)__builtin_ia32_vcvthf8_2ph512_mask(
       (__v32qi)__A, (__v32hf)(__m512h)_mm512_undefined_ph(), (__mmask32)-1);
 }
 
 static __inline__ __m512h __DEFAULT_FN_ATTRS512
-_mm512_mask_cvthf8(__m512h __W, __mmask32 __U, __m256i __A) {
+_mm512_mask_cvthf8_ph(__m512h __W, __mmask32 __U, __m256i __A) {
   return (__m512h)__builtin_ia32_vcvthf8_2ph512_mask(
       (__v32qi)__A, (__v32hf)(__m512h)__W, (__mmask32)__U);
 }
 
 static __inline__ __m512h __DEFAULT_FN_ATTRS512
-_mm512_maskz_cvthf8(__mmask32 __U, __m256i __A) {
+_mm512_maskz_cvthf8_ph(__mmask32 __U, __m256i __A) {
   return (__m512h)__builtin_ia32_vcvthf8_2ph512_mask(
       (__v32qi)__A, (__v32hf)(__m512h)_mm512_setzero_ph(), (__mmask32)__U);
 }
@@ -247,19 +247,20 @@ _mm512_maskz_cvtph_bf8(__mmask32 __U, __m512h __A) {
       (__v32hf)__A, (__v32qi)(__m256i)_mm256_setzero_si256(), (__mmask32)__U);
 }
 
-static __inline__ __m256i __DEFAULT_FN_ATTRS512 _mm512_cvtsph_bf8(__m512h __A) {
+static __inline__ __m256i __DEFAULT_FN_ATTRS512
+_mm512_cvts_ph_bf8(__m512h __A) {
   return (__m256i)__builtin_ia32_vcvtph2bf8s_512_mask(
       (__v32hf)__A, (__v32qi)(__m256i)_mm256_undefined_si256(), (__mmask32)-1);
 }
 
 static __inline__ __m256i __DEFAULT_FN_ATTRS512
-_mm512_mask_cvtsph_bf8(__m256i __W, __mmask32 __U, __m512h __A) {
+_mm512_mask_cvts_ph_bf8(__m256i __W, __mmask32 __U, __m512h __A) {
   return (__m256i)__builtin_ia32_vcvtph2bf8s_512_mask(
       (__v32hf)__A, (__v32qi)(__m256i)__W, (__mmask32)__U);
 }
 
 static __inline__ __m256i __DEFAULT_FN_ATTRS512
-_mm512_maskz_cvtsph_bf8(__mmask32 __U, __m512h __A) {
+_mm512_maskz_cvts_ph_bf8(__mmask32 __U, __m512h __A) {
   return (__m256i)__builtin_ia32_vcvtph2bf8s_512_mask(
       (__v32hf)__A, (__v32qi)(__m256i)_mm256_setzero_si256(), (__mmask32)__U);
 }
@@ -281,19 +282,20 @@ _mm512_maskz_cvtph_hf8(__mmask32 __U, __m512h __A) {
       (__v32hf)__A, (__v32qi)(__m256i)_mm256_setzero_si256(), (__mmask32)__U);
 }
 
-static __inline__ __m256i __DEFAULT_FN_ATTRS512 _mm512_cvtsph_hf8(__m512h __A) {
+static __inline__ __m256i __DEFAULT_FN_ATTRS512
+_mm512_cvts_ph_hf8(__m512h __A) {
   return (__m256i)__builtin_ia32_vcvtph2hf8s_512_mask(
       (__v32hf)__A, (__v32qi)(__m256i)_mm256_undefined_si256(), (__mmask32)-1);
 }
 
 static __inline__ __m256i __DEFAULT_FN_ATTRS512
-_mm512_mask_cvtsph_hf8(__m256i __W, __mmask32 __U, __m512h __A) {
+_mm512_mask_cvts_ph_hf8(__m256i __W, __mmask32 __U, __m512h __A) {
   return (__m256i)__builtin_ia32_vcvtph2hf8s_512_mask(
       (__v32hf)__A, (__v32qi)(__m256i)__W, (__mmask32)__U);
 }
 
 static __inline__ __m256i __DEFAULT_FN_ATTRS512
-_mm512_maskz_cvtsph_hf8(__mmask32 __U, __m512h __A) {
+_mm512_maskz_cvts_ph_hf8(__mmask32 __U, __m512h __A) {
   return (__m256i)__builtin_ia32_vcvtph2hf8s_512_mask(
       (__v32hf)__A, (__v32qi)(__m256i)_mm256_setzero_si256(), (__mmask32)__U);
 }

lib/include/avx10_2_512satcvtdsintrin.h

@@ -20,20 +20,21 @@
                  __min_vector_width__(512)))
 
 // 512 bit : Double -> Int
-static __inline__ __m256i __DEFAULT_FN_ATTRS _mm512_cvttspd_epi32(__m512d __A) {
+static __inline__ __m256i __DEFAULT_FN_ATTRS
+_mm512_cvtts_pd_epi32(__m512d __A) {
   return ((__m256i)__builtin_ia32_vcvttpd2dqs512_round_mask(
       (__v8df)__A, (__v8si)_mm256_undefined_si256(), (__mmask8)-1,
       _MM_FROUND_CUR_DIRECTION));
 }
 
 static __inline__ __m256i __DEFAULT_FN_ATTRS
-_mm512_mask_cvttspd_epi32(__m256i __W, __mmask8 __U, __m512d __A) {
+_mm512_mask_cvtts_pd_epi32(__m256i __W, __mmask8 __U, __m512d __A) {
   return ((__m256i)__builtin_ia32_vcvttpd2dqs512_round_mask(
       (__v8df)__A, (__v8si)__W, __U, _MM_FROUND_CUR_DIRECTION));
 }
 
 static __inline__ __m256i __DEFAULT_FN_ATTRS
-_mm512_maskz_cvttspd_epi32(__mmask8 __U, __m512d __A) {
+_mm512_maskz_cvtts_pd_epi32(__mmask8 __U, __m512d __A) {
   return ((__m256i)__builtin_ia32_vcvttpd2dqs512_round_mask(
       (__v8df)__A, (__v8si)_mm256_setzero_si256(), __U,
       _MM_FROUND_CUR_DIRECTION));
@@ -55,20 +56,21 @@ _mm512_maskz_cvttspd_epi32(__mmask8 __U, __m512d __A) {
       (const int)(__R)))
 
 // 512 bit : Double -> uInt
-static __inline__ __m256i __DEFAULT_FN_ATTRS _mm512_cvttspd_epu32(__m512d __A) {
+static __inline__ __m256i __DEFAULT_FN_ATTRS
+_mm512_cvtts_pd_epu32(__m512d __A) {
   return ((__m256i)__builtin_ia32_vcvttpd2udqs512_round_mask(
       (__v8df)__A, (__v8si)_mm256_undefined_si256(), (__mmask8)-1,
       _MM_FROUND_CUR_DIRECTION));
 }
 
 static __inline__ __m256i __DEFAULT_FN_ATTRS
-_mm512_mask_cvttspd_epu32(__m256i __W, __mmask8 __U, __m512d __A) {
+_mm512_mask_cvtts_pd_epu32(__m256i __W, __mmask8 __U, __m512d __A) {
   return ((__m256i)__builtin_ia32_vcvttpd2udqs512_round_mask(
       (__v8df)__A, (__v8si)__W, __U, _MM_FROUND_CUR_DIRECTION));
 }
 
 static __inline__ __m256i __DEFAULT_FN_ATTRS
-_mm512_maskz_cvttspd_epu32(__mmask8 __U, __m512d __A) {
+_mm512_maskz_cvtts_pd_epu32(__mmask8 __U, __m512d __A) {
   return ((__m256i)__builtin_ia32_vcvttpd2udqs512_round_mask(
       (__v8df)__A, (__v8si)_mm256_setzero_si256(), __U,
       _MM_FROUND_CUR_DIRECTION));
@@ -91,18 +93,19 @@ _mm512_maskz_cvttspd_epu32(__mmask8 __U, __m512d __A) {
 
 //  512 bit : Double -> Long
 
-static __inline__ __m512i __DEFAULT_FN_ATTRS _mm512_cvttspd_epi64(__m512d __A) {
+static __inline__ __m512i __DEFAULT_FN_ATTRS
+_mm512_cvtts_pd_epi64(__m512d __A) {
   return ((__m512i)__builtin_ia32_vcvttpd2qqs512_round_mask(
       (__v8df)__A, (__v8di)_mm512_undefined_epi32(), (__mmask8)-1,
       _MM_FROUND_CUR_DIRECTION));
 }
 static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_mask_cvttspd_epi64(__m512i __W, __mmask8 __U, __m512d __A) {
+_mm512_mask_cvtts_pd_epi64(__m512i __W, __mmask8 __U, __m512d __A) {
   return ((__m512i)__builtin_ia32_vcvttpd2qqs512_round_mask(
       (__v8df)__A, (__v8di)__W, __U, _MM_FROUND_CUR_DIRECTION));
 }
 static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_maskz_cvttspd_epi64(__mmask8 __U, __m512d __A) {
+_mm512_maskz_cvtts_pd_epi64(__mmask8 __U, __m512d __A) {
   return ((__m512i)__builtin_ia32_vcvttpd2qqs512_round_mask(
       (__v8df)__A, (__v8di)_mm512_setzero_si512(), __U,
       _MM_FROUND_CUR_DIRECTION));
@@ -125,20 +128,21 @@ _mm512_maskz_cvttspd_epi64(__mmask8 __U, __m512d __A) {
 
 // 512 bit : Double -> ULong
 
-static __inline__ __m512i __DEFAULT_FN_ATTRS _mm512_cvttspd_epu64(__m512d __A) {
+static __inline__ __m512i __DEFAULT_FN_ATTRS
+_mm512_cvtts_pd_epu64(__m512d __A) {
   return ((__m512i)__builtin_ia32_vcvttpd2uqqs512_round_mask(
       (__v8df)__A, (__v8di)_mm512_undefined_epi32(), (__mmask8)-1,
       _MM_FROUND_CUR_DIRECTION));
 }
 
 static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_mask_cvttspd_epu64(__m512i __W, __mmask8 __U, __m512d __A) {
+_mm512_mask_cvtts_pd_epu64(__m512i __W, __mmask8 __U, __m512d __A) {
   return ((__m512i)__builtin_ia32_vcvttpd2uqqs512_round_mask(
       (__v8df)__A, (__v8di)__W, __U, _MM_FROUND_CUR_DIRECTION));
 }
 
 static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_maskz_cvttspd_epu64(__mmask8 __U, __m512d __A) {
+_mm512_maskz_cvtts_pd_epu64(__mmask8 __U, __m512d __A) {
   return ((__m512i)__builtin_ia32_vcvttpd2uqqs512_round_mask(
       (__v8df)__A, (__v8di)_mm512_setzero_si512(), __U,
       _MM_FROUND_CUR_DIRECTION));
@@ -160,20 +164,20 @@ _mm512_maskz_cvttspd_epu64(__mmask8 __U, __m512d __A) {
       (const int)(__R)))
 
 // 512 bit: Float -> int
-static __inline__ __m512i __DEFAULT_FN_ATTRS _mm512_cvttsps_epi32(__m512 __A) {
+static __inline__ __m512i __DEFAULT_FN_ATTRS _mm512_cvtts_ps_epi32(__m512 __A) {
   return ((__m512i)__builtin_ia32_vcvttps2dqs512_round_mask(
       (__v16sf)(__A), (__v16si)_mm512_undefined_epi32(), (__mmask16)-1,
       _MM_FROUND_CUR_DIRECTION));
 }
 
 static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_mask_cvttsps_epi32(__m512i __W, __mmask16 __U, __m512 __A) {
+_mm512_mask_cvtts_ps_epi32(__m512i __W, __mmask16 __U, __m512 __A) {
   return ((__m512i)__builtin_ia32_vcvttps2dqs512_round_mask(
       (__v16sf)(__A), (__v16si)(__W), __U, _MM_FROUND_CUR_DIRECTION));
 }
 
 static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_maskz_cvttsps_epi32(__mmask16 __U, __m512 __A) {
+_mm512_maskz_cvtts_ps_epi32(__mmask16 __U, __m512 __A) {
   return ((__m512i)__builtin_ia32_vcvttps2dqs512_round_mask(
       (__v16sf)(__A), (__v16si)_mm512_setzero_si512(), __U,
       _MM_FROUND_CUR_DIRECTION));
@@ -195,20 +199,20 @@ _mm512_maskz_cvttsps_epi32(__mmask16 __U, __m512 __A) {
       (__mmask16)(__U), (const int)(__R)))
 
 // 512 bit: Float -> uint
-static __inline__ __m512i __DEFAULT_FN_ATTRS _mm512_cvttsps_epu32(__m512 __A) {
+static __inline__ __m512i __DEFAULT_FN_ATTRS _mm512_cvtts_ps_epu32(__m512 __A) {
   return ((__m512i)__builtin_ia32_vcvttps2udqs512_round_mask(
       (__v16sf)(__A), (__v16si)_mm512_undefined_epi32(), (__mmask16)-1,
       _MM_FROUND_CUR_DIRECTION));
 }
 
 static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_mask_cvttsps_epu32(__m512i __W, __mmask16 __U, __m512 __A) {
+_mm512_mask_cvtts_ps_epu32(__m512i __W, __mmask16 __U, __m512 __A) {
   return ((__m512i)__builtin_ia32_vcvttps2udqs512_round_mask(
       (__v16sf)(__A), (__v16si)(__W), __U, _MM_FROUND_CUR_DIRECTION));
 }
 
 static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_maskz_cvttsps_epu32(__mmask16 __U, __m512 __A) {
+_mm512_maskz_cvtts_ps_epu32(__mmask16 __U, __m512 __A) {
   return ((__m512i)__builtin_ia32_vcvttps2udqs512_round_mask(
       (__v16sf)(__A), (__v16si)_mm512_setzero_si512(), __U,
       _MM_FROUND_CUR_DIRECTION));
@@ -230,20 +234,20 @@ _mm512_maskz_cvttsps_epu32(__mmask16 __U, __m512 __A) {
       (__mmask16)(__U), (const int)(__R)))
 
 // 512 bit : float -> long
-static __inline__ __m512i __DEFAULT_FN_ATTRS _mm512_cvttsps_epi64(__m256 __A) {
+static __inline__ __m512i __DEFAULT_FN_ATTRS _mm512_cvtts_ps_epi64(__m256 __A) {
   return ((__m512i)__builtin_ia32_vcvttps2qqs512_round_mask(
       (__v8sf)__A, (__v8di)_mm512_undefined_epi32(), (__mmask8)-1,
       _MM_FROUND_CUR_DIRECTION));
 }
 
 static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_mask_cvttsps_epi64(__m512i __W, __mmask8 __U, __m256 __A) {
+_mm512_mask_cvtts_ps_epi64(__m512i __W, __mmask8 __U, __m256 __A) {
   return ((__m512i)__builtin_ia32_vcvttps2qqs512_round_mask(
       (__v8sf)__A, (__v8di)__W, __U, _MM_FROUND_CUR_DIRECTION));
 }
 
 static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_maskz_cvttsps_epi64(__mmask8 __U, __m256 __A) {
+_mm512_maskz_cvtts_ps_epi64(__mmask8 __U, __m256 __A) {
   return ((__m512i)__builtin_ia32_vcvttps2qqs512_round_mask(
       (__v8sf)__A, (__v8di)_mm512_setzero_si512(), __U,
       _MM_FROUND_CUR_DIRECTION));
@@ -265,20 +269,20 @@ _mm512_maskz_cvttsps_epi64(__mmask8 __U, __m256 __A) {
       (const int)(__R)))
 
 // 512 bit : float -> ulong
-static __inline__ __m512i __DEFAULT_FN_ATTRS _mm512_cvttsps_epu64(__m256 __A) {
+static __inline__ __m512i __DEFAULT_FN_ATTRS _mm512_cvtts_ps_epu64(__m256 __A) {
   return ((__m512i)__builtin_ia32_vcvttps2uqqs512_round_mask(
       (__v8sf)__A, (__v8di)_mm512_undefined_epi32(), (__mmask8)-1,
       _MM_FROUND_CUR_DIRECTION));
 }
 
 static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_mask_cvttsps_epu64(__m512i __W, __mmask8 __U, __m256 __A) {
+_mm512_mask_cvtts_ps_epu64(__m512i __W, __mmask8 __U, __m256 __A) {
   return ((__m512i)__builtin_ia32_vcvttps2uqqs512_round_mask(
       (__v8sf)__A, (__v8di)__W, __U, _MM_FROUND_CUR_DIRECTION));
 }
 
 static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_maskz_cvttsps_epu64(__mmask8 __U, __m256 __A) {
+_mm512_maskz_cvtts_ps_epu64(__mmask8 __U, __m256 __A) {
   return ((__m512i)__builtin_ia32_vcvttps2uqqs512_round_mask(
       (__v8sf)__A, (__v8di)_mm512_setzero_si512(), __U,
       _MM_FROUND_CUR_DIRECTION));

lib/include/avx10_2_512satcvtintrin.h

@@ -14,286 +14,286 @@
 #ifndef __AVX10_2_512SATCVTINTRIN_H
 #define __AVX10_2_512SATCVTINTRIN_H
 
-#define _mm512_ipcvtbf16_epi8(A)                                               \
+#define _mm512_ipcvts_bf16_epi8(A)                                             \
   ((__m512i)__builtin_ia32_vcvtbf162ibs512((__v32bf)(__m512bh)(A)))
 
-#define _mm512_mask_ipcvtbf16_epi8(W, U, A)                                    \
+#define _mm512_mask_ipcvts_bf16_epi8(W, U, A)                                  \
   ((__m512i)__builtin_ia32_selectw_512((__mmask32)(U),                         \
-                                       (__v32hi)_mm512_ipcvtbf16_epi8(A),      \
+                                       (__v32hi)_mm512_ipcvts_bf16_epi8(A),    \
                                        (__v32hi)(__m512i)(W)))
 
-#define _mm512_maskz_ipcvtbf16_epi8(U, A)                                      \
+#define _mm512_maskz_ipcvts_bf16_epi8(U, A)                                    \
   ((__m512i)__builtin_ia32_selectw_512((__mmask32)(U),                         \
-                                       (__v32hi)_mm512_ipcvtbf16_epi8(A),      \
+                                       (__v32hi)_mm512_ipcvts_bf16_epi8(A),    \
                                        (__v32hi)_mm512_setzero_si512()))
 
-#define _mm512_ipcvtbf16_epu8(A)                                               \
+#define _mm512_ipcvts_bf16_epu8(A)                                             \
   ((__m512i)__builtin_ia32_vcvtbf162iubs512((__v32bf)(__m512bh)(A)))
 
-#define _mm512_mask_ipcvtbf16_epu8(W, U, A)                                    \
+#define _mm512_mask_ipcvts_bf16_epu8(W, U, A)                                  \
   ((__m512i)__builtin_ia32_selectw_512((__mmask32)(U),                         \
-                                       (__v32hi)_mm512_ipcvtbf16_epu8(A),      \
+                                       (__v32hi)_mm512_ipcvts_bf16_epu8(A),    \
                                        (__v32hi)(__m512i)(W)))
 
-#define _mm512_maskz_ipcvtbf16_epu8(U, A)                                      \
+#define _mm512_maskz_ipcvts_bf16_epu8(U, A)                                    \
   ((__m512i)__builtin_ia32_selectw_512((__mmask32)(U),                         \
-                                       (__v32hi)_mm512_ipcvtbf16_epu8(A),      \
+                                       (__v32hi)_mm512_ipcvts_bf16_epu8(A),    \
                                        (__v32hi)_mm512_setzero_si512()))
 
-#define _mm512_ipcvttbf16_epi8(A)                                              \
+#define _mm512_ipcvtts_bf16_epi8(A)                                            \
   ((__m512i)__builtin_ia32_vcvttbf162ibs512((__v32bf)(__m512bh)(A)))
 
-#define _mm512_mask_ipcvttbf16_epi8(W, U, A)                                   \
+#define _mm512_mask_ipcvtts_bf16_epi8(W, U, A)                                 \
   ((__m512i)__builtin_ia32_selectw_512((__mmask32)(U),                         \
-                                       (__v32hi)_mm512_ipcvttbf16_epi8(A),     \
+                                       (__v32hi)_mm512_ipcvtts_bf16_epi8(A),   \
                                        (__v32hi)(__m512i)(W)))
 
-#define _mm512_maskz_ipcvttbf16_epi8(U, A)                                     \
+#define _mm512_maskz_ipcvtts_bf16_epi8(U, A)                                   \
   ((__m512i)__builtin_ia32_selectw_512((__mmask32)(U),                         \
-                                       (__v32hi)_mm512_ipcvttbf16_epi8(A),     \
+                                       (__v32hi)_mm512_ipcvtts_bf16_epi8(A),   \
                                        (__v32hi)_mm512_setzero_si512()))
 
-#define _mm512_ipcvttbf16_epu8(A)                                              \
+#define _mm512_ipcvtts_bf16_epu8(A)                                            \
   ((__m512i)__builtin_ia32_vcvttbf162iubs512((__v32bf)(__m512bh)(A)))
 
-#define _mm512_mask_ipcvttbf16_epu8(W, U, A)                                   \
+#define _mm512_mask_ipcvtts_bf16_epu8(W, U, A)                                 \
   ((__m512i)__builtin_ia32_selectw_512((__mmask32)(U),                         \
-                                       (__v32hi)_mm512_ipcvttbf16_epu8(A),     \
+                                       (__v32hi)_mm512_ipcvtts_bf16_epu8(A),   \
                                        (__v32hi)(__m512i)(W)))
 
-#define _mm512_maskz_ipcvttbf16_epu8(U, A)                                     \
+#define _mm512_maskz_ipcvtts_bf16_epu8(U, A)                                   \
   ((__m512i)__builtin_ia32_selectw_512((__mmask32)(U),                         \
-                                       (__v32hi)_mm512_ipcvttbf16_epu8(A),     \
+                                       (__v32hi)_mm512_ipcvtts_bf16_epu8(A),   \
                                        (__v32hi)_mm512_setzero_si512()))
 
-#define _mm512_ipcvtph_epi8(A)                                                 \
+#define _mm512_ipcvts_ph_epi8(A)                                               \
   ((__m512i)__builtin_ia32_vcvtph2ibs512_mask(                                 \
-      (__v32hf)(__m512h)(A), (__v32hu)_mm512_setzero_si512(), (__mmask32)-1,   \
+      (__v32hf)(__m512h)(A), (__v32hu)_mm512_setzero_si512(), (__mmask32) - 1, \
       _MM_FROUND_CUR_DIRECTION))
 
-#define _mm512_mask_ipcvtph_epi8(W, U, A)                                      \
+#define _mm512_mask_ipcvts_ph_epi8(W, U, A)                                    \
   ((__m512i)__builtin_ia32_vcvtph2ibs512_mask((__v32hf)(__m512h)(A),           \
                                               (__v32hu)(W), (__mmask32)(U),    \
                                               _MM_FROUND_CUR_DIRECTION))
 
-#define _mm512_maskz_ipcvtph_epi8(U, A)                                        \
+#define _mm512_maskz_ipcvts_ph_epi8(U, A)                                      \
   ((__m512i)__builtin_ia32_vcvtph2ibs512_mask(                                 \
       (__v32hf)(__m512h)(A), (__v32hu)_mm512_setzero_si512(), (__mmask32)(U),  \
       _MM_FROUND_CUR_DIRECTION))
 
-#define _mm512_ipcvt_roundph_epi8(A, R)                                        \
+#define _mm512_ipcvts_roundph_epi8(A, R)                                       \
   ((__m512i)__builtin_ia32_vcvtph2ibs512_mask((__v32hf)(__m512h)(A),           \
                                               (__v32hu)_mm512_setzero_si512(), \
-                                              (__mmask32)-1, (const int)R))
+                                              (__mmask32) - 1, (const int)R))
 
-#define _mm512_mask_ipcvt_roundph_epi8(W, U, A, R)                             \
+#define _mm512_mask_ipcvts_roundph_epi8(W, U, A, R)                            \
   ((__m512i)__builtin_ia32_vcvtph2ibs512_mask(                                 \
       (__v32hf)(__m512h)(A), (__v32hu)(W), (__mmask32)(U), (const int)R))
 
-#define _mm512_maskz_ipcvt_roundph_epi8(U, A, R)                               \
+#define _mm512_maskz_ipcvts_roundph_epi8(U, A, R)                              \
   ((__m512i)__builtin_ia32_vcvtph2ibs512_mask((__v32hf)(__m512h)(A),           \
                                               (__v32hu)_mm512_setzero_si512(), \
                                               (__mmask32)(U), (const int)R))
 
-#define _mm512_ipcvtph_epu8(A)                                                 \
+#define _mm512_ipcvts_ph_epu8(A)                                               \
   ((__m512i)__builtin_ia32_vcvtph2iubs512_mask(                                \
-      (__v32hf)(__m512h)(A), (__v32hu)_mm512_setzero_si512(), (__mmask32)-1,   \
+      (__v32hf)(__m512h)(A), (__v32hu)_mm512_setzero_si512(), (__mmask32) - 1, \
       _MM_FROUND_CUR_DIRECTION))
 
-#define _mm512_mask_ipcvtph_epu8(W, U, A)                                      \
+#define _mm512_mask_ipcvts_ph_epu8(W, U, A)                                    \
   ((__m512i)__builtin_ia32_vcvtph2iubs512_mask((__v32hf)(__m512h)(A),          \
                                                (__v32hu)(W), (__mmask32)(U),   \
                                                _MM_FROUND_CUR_DIRECTION))
 
-#define _mm512_maskz_ipcvtph_epu8(U, A)                                        \
+#define _mm512_maskz_ipcvts_ph_epu8(U, A)                                      \
   ((__m512i)__builtin_ia32_vcvtph2iubs512_mask(                                \
       (__v32hf)(__m512h)(A), (__v32hu)_mm512_setzero_si512(), (__mmask32)(U),  \
       _MM_FROUND_CUR_DIRECTION))
 
-#define _mm512_ipcvt_roundph_epu8(A, R)                                        \
+#define _mm512_ipcvts_roundph_epu8(A, R)                                       \
   ((__m512i)__builtin_ia32_vcvtph2iubs512_mask(                                \
-      (__v32hf)(__m512h)(A), (__v32hu)_mm512_setzero_si512(), (__mmask32)-1,   \
+      (__v32hf)(__m512h)(A), (__v32hu)_mm512_setzero_si512(), (__mmask32) - 1, \
       (const int)R))
 
-#define _mm512_mask_ipcvt_roundph_epu8(W, U, A, R)                             \
+#define _mm512_mask_ipcvts_roundph_epu8(W, U, A, R)                            \
   ((__m512i)__builtin_ia32_vcvtph2iubs512_mask(                                \
       (__v32hf)(__m512h)(A), (__v32hu)(W), (__mmask32)(U), (const int)R))
 
-#define _mm512_maskz_ipcvt_roundph_epu8(U, A, R)                               \
+#define _mm512_maskz_ipcvts_roundph_epu8(U, A, R)                              \
   ((__m512i)__builtin_ia32_vcvtph2iubs512_mask(                                \
       (__v32hf)(__m512h)(A), (__v32hu)_mm512_setzero_si512(), (__mmask32)(U),  \
       (const int)R))
 
-#define _mm512_ipcvtps_epi8(A)                                                 \
+#define _mm512_ipcvts_ps_epi8(A)                                               \
   ((__m512i)__builtin_ia32_vcvtps2ibs512_mask(                                 \
-      (__v16sf)(__m512)(A), (__v16su)_mm512_setzero_si512(), (__mmask16)-1,    \
+      (__v16sf)(__m512)(A), (__v16su)_mm512_setzero_si512(), (__mmask16) - 1,  \
       _MM_FROUND_CUR_DIRECTION))
 
-#define _mm512_mask_ipcvtps_epi8(W, U, A)                                      \
+#define _mm512_mask_ipcvts_ps_epi8(W, U, A)                                    \
   ((__m512i)__builtin_ia32_vcvtps2ibs512_mask((__v16sf)(__m512)(A),            \
                                               (__v16su)(W), (__mmask16)(U),    \
                                               _MM_FROUND_CUR_DIRECTION))
 
-#define _mm512_maskz_ipcvtps_epi8(U, A)                                        \
+#define _mm512_maskz_ipcvts_ps_epi8(U, A)                                      \
   ((__m512i)__builtin_ia32_vcvtps2ibs512_mask(                                 \
       (__v16sf)(__m512)(A), (__v16su)_mm512_setzero_si512(), (__mmask16)(U),   \
       _MM_FROUND_CUR_DIRECTION))
 
-#define _mm512_ipcvt_roundps_epi8(A, R)                                        \
+#define _mm512_ipcvts_roundps_epi8(A, R)                                       \
   ((__m512i)__builtin_ia32_vcvtps2ibs512_mask((__v16sf)(__m512)(A),            \
                                               (__v16su)_mm512_setzero_si512(), \
-                                              (__mmask16)-1, (const int)R))
+                                              (__mmask16) - 1, (const int)R))
 
-#define _mm512_mask_ipcvt_roundps_epi8(W, U, A, R)                             \
+#define _mm512_mask_ipcvts_roundps_epi8(W, U, A, R)                            \
   ((__m512i)__builtin_ia32_vcvtps2ibs512_mask(                                 \
       (__v16sf)(__m512)(A), (__v16su)(W), (__mmask16)(U), (const int)R))
 
-#define _mm512_maskz_ipcvt_roundps_epi8(U, A, R)                               \
+#define _mm512_maskz_ipcvts_roundps_epi8(U, A, R)                              \
   ((__m512i)__builtin_ia32_vcvtps2ibs512_mask((__v16sf)(__m512)(A),            \
                                               (__v16su)_mm512_setzero_si512(), \
                                               (__mmask16)(U), (const int)R))
 
-#define _mm512_ipcvtps_epu8(A)                                                 \
+#define _mm512_ipcvts_ps_epu8(A)                                               \
   ((__m512i)__builtin_ia32_vcvtps2iubs512_mask(                                \
-      (__v16sf)(__m512)(A), (__v16su)_mm512_setzero_si512(), (__mmask16)-1,    \
+      (__v16sf)(__m512)(A), (__v16su)_mm512_setzero_si512(), (__mmask16) - 1,  \
       _MM_FROUND_CUR_DIRECTION))
 
-#define _mm512_mask_ipcvtps_epu8(W, U, A)                                      \
+#define _mm512_mask_ipcvts_ps_epu8(W, U, A)                                    \
   ((__m512i)__builtin_ia32_vcvtps2iubs512_mask((__v16sf)(__m512)(A),           \
                                                (__v16su)(W), (__mmask16)(U),   \
                                                _MM_FROUND_CUR_DIRECTION))
 
-#define _mm512_maskz_ipcvtps_epu8(U, A)                                        \
+#define _mm512_maskz_ipcvts_ps_epu8(U, A)                                      \
   ((__m512i)__builtin_ia32_vcvtps2iubs512_mask(                                \
       (__v16sf)(__m512)(A), (__v16su)_mm512_setzero_si512(), (__mmask16)(U),   \
       _MM_FROUND_CUR_DIRECTION))
 
-#define _mm512_ipcvt_roundps_epu8(A, R)                                        \
+#define _mm512_ipcvts_roundps_epu8(A, R)                                       \
   ((__m512i)__builtin_ia32_vcvtps2iubs512_mask(                                \
-      (__v16sf)(__m512)(A), (__v16su)_mm512_setzero_si512(), (__mmask16)-1,    \
+      (__v16sf)(__m512)(A), (__v16su)_mm512_setzero_si512(), (__mmask16) - 1,  \
       (const int)R))
 
-#define _mm512_mask_ipcvt_roundps_epu8(W, U, A, R)                             \
+#define _mm512_mask_ipcvts_roundps_epu8(W, U, A, R)                            \
   ((__m512i)__builtin_ia32_vcvtps2iubs512_mask(                                \
       (__v16sf)(__m512)(A), (__v16su)(W), (__mmask16)(U), (const int)R))
 
-#define _mm512_maskz_ipcvt_roundps_epu8(U, A, R)                               \
+#define _mm512_maskz_ipcvts_roundps_epu8(U, A, R)                              \
   ((__m512i)__builtin_ia32_vcvtps2iubs512_mask(                                \
       (__v16sf)(__m512)(A), (__v16su)_mm512_setzero_si512(), (__mmask16)(U),   \
       (const int)R))
 
-#define _mm512_ipcvttph_epi8(A)                                                \
+#define _mm512_ipcvtts_ph_epi8(A)                                              \
   ((__m512i)__builtin_ia32_vcvttph2ibs512_mask(                                \
-      (__v32hf)(__m512h)(A), (__v32hu)_mm512_setzero_si512(), (__mmask32)-1,   \
+      (__v32hf)(__m512h)(A), (__v32hu)_mm512_setzero_si512(), (__mmask32) - 1, \
       _MM_FROUND_CUR_DIRECTION))
 
-#define _mm512_mask_ipcvttph_epi8(W, U, A)                                     \
+#define _mm512_mask_ipcvtts_ph_epi8(W, U, A)                                   \
   ((__m512i)__builtin_ia32_vcvttph2ibs512_mask((__v32hf)(__m512h)(A),          \
                                                (__v32hu)(W), (__mmask32)(U),   \
                                                _MM_FROUND_CUR_DIRECTION))
 
-#define _mm512_maskz_ipcvttph_epi8(U, A)                                       \
+#define _mm512_maskz_ipcvtts_ph_epi8(U, A)                                     \
   ((__m512i)__builtin_ia32_vcvttph2ibs512_mask(                                \
       (__v32hf)(__m512h)(A), (__v32hu)_mm512_setzero_si512(), (__mmask32)(U),  \
       _MM_FROUND_CUR_DIRECTION))
 
-#define _mm512_ipcvtt_roundph_epi8(A, S)                                       \
+#define _mm512_ipcvtts_roundph_epi8(A, S)                                      \
   ((__m512i)__builtin_ia32_vcvttph2ibs512_mask(                                \
-      (__v32hf)(__m512h)(A), (__v32hu)_mm512_setzero_si512(), (__mmask32)-1,   \
+      (__v32hf)(__m512h)(A), (__v32hu)_mm512_setzero_si512(), (__mmask32) - 1, \
       S))
 
-#define _mm512_mask_ipcvtt_roundph_epi8(W, U, A, S)                            \
+#define _mm512_mask_ipcvtts_roundph_epi8(W, U, A, S)                           \
   ((__m512i)__builtin_ia32_vcvttph2ibs512_mask(                                \
       (__v32hf)(__m512h)(A), (__v32hu)(W), (__mmask32)(U), S))
 
-#define _mm512_maskz_ipcvtt_roundph_epi8(U, A, S)                              \
+#define _mm512_maskz_ipcvtts_roundph_epi8(U, A, S)                             \
   ((__m512i)__builtin_ia32_vcvttph2ibs512_mask(                                \
       (__v32hf)(__m512h)(A), (__v32hu)_mm512_setzero_si512(), (__mmask32)(U),  \
       S))
 
-#define _mm512_ipcvttph_epu8(A)                                                \
+#define _mm512_ipcvtts_ph_epu8(A)                                              \
   ((__m512i)__builtin_ia32_vcvttph2iubs512_mask(                               \
-      (__v32hf)(__m512h)(A), (__v32hu)_mm512_setzero_si512(), (__mmask32)-1,   \
+      (__v32hf)(__m512h)(A), (__v32hu)_mm512_setzero_si512(), (__mmask32) - 1, \
       _MM_FROUND_CUR_DIRECTION))
 
-#define _mm512_mask_ipcvttph_epu8(W, U, A)                                     \
+#define _mm512_mask_ipcvtts_ph_epu8(W, U, A)                                   \
   ((__m512i)__builtin_ia32_vcvttph2iubs512_mask((__v32hf)(__m512h)(A),         \
                                                 (__v32hu)(W), (__mmask32)(U),  \
                                                 _MM_FROUND_CUR_DIRECTION))
 
-#define _mm512_maskz_ipcvttph_epu8(U, A)                                       \
+#define _mm512_maskz_ipcvtts_ph_epu8(U, A)                                     \
   ((__m512i)__builtin_ia32_vcvttph2iubs512_mask(                               \
       (__v32hf)(__m512h)(A), (__v32hu)_mm512_setzero_si512(), (__mmask32)(U),  \
       _MM_FROUND_CUR_DIRECTION))
 
-#define _mm512_ipcvtt_roundph_epu8(A, S)                                       \
+#define _mm512_ipcvtts_roundph_epu8(A, S)                                      \
   ((__m512i)__builtin_ia32_vcvttph2iubs512_mask(                               \
-      (__v32hf)(__m512h)(A), (__v32hu)_mm512_setzero_si512(), (__mmask32)-1,   \
+      (__v32hf)(__m512h)(A), (__v32hu)_mm512_setzero_si512(), (__mmask32) - 1, \
       S))
 
-#define _mm512_mask_ipcvtt_roundph_epu8(W, U, A, S)                            \
+#define _mm512_mask_ipcvtts_roundph_epu8(W, U, A, S)                           \
   ((__m512i)__builtin_ia32_vcvttph2iubs512_mask(                               \
       (__v32hf)(__m512h)(A), (__v32hu)(W), (__mmask32)(U), S))
 
-#define _mm512_maskz_ipcvtt_roundph_epu8(U, A, S)                              \
+#define _mm512_maskz_ipcvtts_roundph_epu8(U, A, S)                             \
   ((__m512i)__builtin_ia32_vcvttph2iubs512_mask(                               \
       (__v32hf)(__m512h)(A), (__v32hu)_mm512_setzero_si512(), (__mmask32)(U),  \
       S))
 
-#define _mm512_ipcvttps_epi8(A)                                                \
+#define _mm512_ipcvtts_ps_epi8(A)                                              \
   ((__m512i)__builtin_ia32_vcvttps2ibs512_mask(                                \
-      (__v16sf)(__m512h)(A), (__v16su)_mm512_setzero_si512(), (__mmask16)-1,   \
+      (__v16sf)(__m512h)(A), (__v16su)_mm512_setzero_si512(), (__mmask16) - 1, \
       _MM_FROUND_CUR_DIRECTION))
 
-#define _mm512_mask_ipcvttps_epi8(W, U, A)                                     \
+#define _mm512_mask_ipcvtts_ps_epi8(W, U, A)                                   \
   ((__m512i)__builtin_ia32_vcvttps2ibs512_mask((__v16sf)(__m512h)(A),          \
                                                (__v16su)(W), (__mmask16)(U),   \
                                                _MM_FROUND_CUR_DIRECTION))
 
-#define _mm512_maskz_ipcvttps_epi8(U, A)                                       \
+#define _mm512_maskz_ipcvtts_ps_epi8(U, A)                                     \
   ((__m512i)__builtin_ia32_vcvttps2ibs512_mask(                                \
       (__v16sf)(__m512h)(A), (__v16su)_mm512_setzero_si512(), (__mmask16)(U),  \
       _MM_FROUND_CUR_DIRECTION))
 
-#define _mm512_ipcvtt_roundps_epi8(A, S)                                       \
+#define _mm512_ipcvtts_roundps_epi8(A, S)                                      \
   ((__m512i)__builtin_ia32_vcvttps2ibs512_mask(                                \
-      (__v16sf)(__m512h)(A), (__v16su)_mm512_setzero_si512(), (__mmask16)-1,   \
+      (__v16sf)(__m512h)(A), (__v16su)_mm512_setzero_si512(), (__mmask16) - 1, \
       S))
 
-#define _mm512_mask_ipcvtt_roundps_epi8(W, U, A, S)                            \
+#define _mm512_mask_ipcvtts_roundps_epi8(W, U, A, S)                           \
   ((__m512i)__builtin_ia32_vcvttps2ibs512_mask(                                \
       (__v16sf)(__m512h)(A), (__v16su)(W), (__mmask16)(U), S))
 
-#define _mm512_maskz_ipcvtt_roundps_epi8(U, A, S)                              \
+#define _mm512_maskz_ipcvtts_roundps_epi8(U, A, S)                             \
   ((__m512i)__builtin_ia32_vcvttps2ibs512_mask(                                \
       (__v16sf)(__m512h)(A), (__v16su)_mm512_setzero_si512(), (__mmask16)(U),  \
       S))
 
-#define _mm512_ipcvttps_epu8(A)                                                \
+#define _mm512_ipcvtts_ps_epu8(A)                                              \
   ((__m512i)__builtin_ia32_vcvttps2iubs512_mask(                               \
-      (__v16sf)(__m512h)(A), (__v16su)_mm512_setzero_si512(), (__mmask16)-1,   \
+      (__v16sf)(__m512h)(A), (__v16su)_mm512_setzero_si512(), (__mmask16) - 1, \
       _MM_FROUND_CUR_DIRECTION))
 
-#define _mm512_mask_ipcvttps_epu8(W, U, A)                                     \
+#define _mm512_mask_ipcvtts_ps_epu8(W, U, A)                                   \
   ((__m512i)__builtin_ia32_vcvttps2iubs512_mask((__v16sf)(__m512h)(A),         \
                                                 (__v16su)(W), (__mmask16)(U),  \
                                                 _MM_FROUND_CUR_DIRECTION))
 
-#define _mm512_maskz_ipcvttps_epu8(U, A)                                       \
+#define _mm512_maskz_ipcvtts_ps_epu8(U, A)                                     \
   ((__m512i)__builtin_ia32_vcvttps2iubs512_mask(                               \
       (__v16sf)(__m512h)(A), (__v16su)_mm512_setzero_si512(), (__mmask16)(U),  \
       _MM_FROUND_CUR_DIRECTION))
 
-#define _mm512_ipcvtt_roundps_epu8(A, S)                                       \
+#define _mm512_ipcvtts_roundps_epu8(A, S)                                      \
   ((__m512i)__builtin_ia32_vcvttps2iubs512_mask(                               \
-      (__v16sf)(__m512h)(A), (__v16su)_mm512_setzero_si512(), (__mmask16)-1,   \
+      (__v16sf)(__m512h)(A), (__v16su)_mm512_setzero_si512(), (__mmask16) - 1, \
       S))
 
-#define _mm512_mask_ipcvtt_roundps_epu8(W, U, A, S)                            \
+#define _mm512_mask_ipcvtts_roundps_epu8(W, U, A, S)                           \
   ((__m512i)__builtin_ia32_vcvttps2iubs512_mask(                               \
       (__v16sf)(__m512h)(A), (__v16su)(W), (__mmask16)(U), S))
 
-#define _mm512_maskz_ipcvtt_roundps_epu8(U, A, S)                              \
+#define _mm512_maskz_ipcvtts_roundps_epu8(U, A, S)                             \
   ((__m512i)__builtin_ia32_vcvttps2iubs512_mask(                               \
       (__v16sf)(__m512h)(A), (__v16su)_mm512_setzero_si512(), (__mmask16)(U),  \
       S))

lib/include/avx10_2bf16intrin.h

@@ -852,8 +852,8 @@ _mm_maskz_sqrt_pbh(__mmask8 __U, __m128bh __A) {
 
 static __inline__ __m256bh __DEFAULT_FN_ATTRS256
 _mm256_fmadd_pbh(__m256bh __A, __m256bh __B, __m256bh __C) {
-  return (__m256bh)__builtin_ia32_vfmaddnepbh256((__v16bf)__A, (__v16bf)__B,
-                                                 (__v16bf)__C);
+  return (__m256bh)__builtin_ia32_vfmaddbf16256((__v16bf)__A, (__v16bf)__B,
+                                                (__v16bf)__C);
 }
 
 static __inline__ __m256bh __DEFAULT_FN_ATTRS256
@@ -880,8 +880,8 @@ static __inline__ __m256bh __DEFAULT_FN_ATTRS256 _mm256_maskz_fmadd_pbh(
 
 static __inline__ __m256bh __DEFAULT_FN_ATTRS256
 _mm256_fmsub_pbh(__m256bh __A, __m256bh __B, __m256bh __C) {
-  return (__m256bh)__builtin_ia32_vfmaddnepbh256((__v16bf)__A, (__v16bf)__B,
-                                                 -(__v16bf)__C);
+  return (__m256bh)__builtin_ia32_vfmaddbf16256((__v16bf)__A, (__v16bf)__B,
+                                                -(__v16bf)__C);
 }
 
 static __inline__ __m256bh __DEFAULT_FN_ATTRS256
@@ -908,8 +908,8 @@ static __inline__ __m256bh __DEFAULT_FN_ATTRS256 _mm256_maskz_fmsub_pbh(
 
 static __inline__ __m256bh __DEFAULT_FN_ATTRS256
 _mm256_fnmadd_pbh(__m256bh __A, __m256bh __B, __m256bh __C) {
-  return (__m256bh)__builtin_ia32_vfmaddnepbh256((__v16bf)__A, -(__v16bf)__B,
-                                                 (__v16bf)__C);
+  return (__m256bh)__builtin_ia32_vfmaddbf16256((__v16bf)__A, -(__v16bf)__B,
+                                                (__v16bf)__C);
 }
 
 static __inline__ __m256bh __DEFAULT_FN_ATTRS256 _mm256_mask_fnmadd_pbh(
@@ -938,8 +938,8 @@ static __inline__ __m256bh __DEFAULT_FN_ATTRS256 _mm256_maskz_fnmadd_pbh(
 
 static __inline__ __m256bh __DEFAULT_FN_ATTRS256
 _mm256_fnmsub_pbh(__m256bh __A, __m256bh __B, __m256bh __C) {
-  return (__m256bh)__builtin_ia32_vfmaddnepbh256((__v16bf)__A, -(__v16bf)__B,
-                                                 -(__v16bf)__C);
+  return (__m256bh)__builtin_ia32_vfmaddbf16256((__v16bf)__A, -(__v16bf)__B,
+                                                -(__v16bf)__C);
 }
 
 static __inline__ __m256bh __DEFAULT_FN_ATTRS256 _mm256_mask_fnmsub_pbh(
@@ -969,8 +969,8 @@ static __inline__ __m256bh __DEFAULT_FN_ATTRS256 _mm256_maskz_fnmsub_pbh(
 static __inline__ __m128bh __DEFAULT_FN_ATTRS128 _mm_fmadd_pbh(__m128bh __A,
                                                                __m128bh __B,
                                                                __m128bh __C) {
-  return (__m128bh)__builtin_ia32_vfmaddnepbh128((__v8bf)__A, (__v8bf)__B,
-                                                 (__v8bf)__C);
+  return (__m128bh)__builtin_ia32_vfmaddbf16128((__v8bf)__A, (__v8bf)__B,
+                                                (__v8bf)__C);
 }
 
 static __inline__ __m128bh __DEFAULT_FN_ATTRS128
@@ -997,8 +997,8 @@ _mm_maskz_fmadd_pbh(__mmask8 __U, __m128bh __A, __m128bh __B, __m128bh __C) {
 static __inline__ __m128bh __DEFAULT_FN_ATTRS128 _mm_fmsub_pbh(__m128bh __A,
                                                                __m128bh __B,
                                                                __m128bh __C) {
-  return (__m128bh)__builtin_ia32_vfmaddnepbh128((__v8bf)__A, (__v8bf)__B,
-                                                 -(__v8bf)__C);
+  return (__m128bh)__builtin_ia32_vfmaddbf16128((__v8bf)__A, (__v8bf)__B,
+                                                -(__v8bf)__C);
 }
 
 static __inline__ __m128bh __DEFAULT_FN_ATTRS128
@@ -1025,8 +1025,8 @@ _mm_maskz_fmsub_pbh(__mmask8 __U, __m128bh __A, __m128bh __B, __m128bh __C) {
 static __inline__ __m128bh __DEFAULT_FN_ATTRS128 _mm_fnmadd_pbh(__m128bh __A,
                                                                 __m128bh __B,
                                                                 __m128bh __C) {
-  return (__m128bh)__builtin_ia32_vfmaddnepbh128((__v8bf)__A, -(__v8bf)__B,
-                                                 (__v8bf)__C);
+  return (__m128bh)__builtin_ia32_vfmaddbf16128((__v8bf)__A, -(__v8bf)__B,
+                                                (__v8bf)__C);
 }
 
 static __inline__ __m128bh __DEFAULT_FN_ATTRS128
@@ -1053,8 +1053,8 @@ _mm_maskz_fnmadd_pbh(__mmask8 __U, __m128bh __A, __m128bh __B, __m128bh __C) {
 static __inline__ __m128bh __DEFAULT_FN_ATTRS128 _mm_fnmsub_pbh(__m128bh __A,
                                                                 __m128bh __B,
                                                                 __m128bh __C) {
-  return (__m128bh)__builtin_ia32_vfmaddnepbh128((__v8bf)__A, -(__v8bf)__B,
-                                                 -(__v8bf)__C);
+  return (__m128bh)__builtin_ia32_vfmaddbf16128((__v8bf)__A, -(__v8bf)__B,
+                                                -(__v8bf)__C);
 }
 
 static __inline__ __m128bh __DEFAULT_FN_ATTRS128

lib/include/avx10_2convertintrin.h

@@ -24,71 +24,351 @@
   __attribute__((__always_inline__, __nodebug__, __target__("avx10.2-256"),    \
                  __min_vector_width__(256)))
 
+// clang-format off
+
+/// Convert two 128-bit vectors, \a __A and \a __B, containing packed
+///    single-precision (32-bit) floating-point elements to a 128-bit vector
+///    containing FP16 elements.
+///
+/// \code{.operation}
+/// FOR i := 0 to 7
+/// 	IF i < 4
+/// 		dst.fp16[i] := convert_fp32_to_fp16(__B.fp32[i])
+/// 	ELSE
+/// 		dst.fp16[i] := convert_fp32_to_fp16(__A.fp32[i - 4])
+/// 	FI
+///
+/// ENDFOR
+///
+/// dst[MAX:128] := 0
+/// \endcode
+///
+/// \headerfile <immintrin.h>
+///
+/// This intrinsic corresponds to the \c VCVT2PS2PHX instruction.
+///
+/// \param __A
+///    A 128-bit vector of [4 x float].
+/// \param __B
+///    A 128-bit vector of [4 x float].
+/// \returns
+///    A 128-bit vector of [8 x fp16]. Lower 4 elements correspond to the
+///    (converted) elements from \a __B; higher order elements correspond to the
+///    (converted) elements from \a __A.
 static __inline__ __m128h __DEFAULT_FN_ATTRS128 _mm_cvtx2ps_ph(__m128 __A,
                                                                __m128 __B) {
   return (__m128h)__builtin_ia32_vcvt2ps2phx128_mask(
       (__v4sf)__A, (__v4sf)__B, (__v8hf)_mm_setzero_ph(), (__mmask8)(-1));
 }
 
+/// Convert two 128-bit vectors, \a __A and \a __B, containing packed
+///    single-precision (32-bit) floating-point elements to a 128-bit vector
+///    containing FP16 elements. Merging mask \a __U is used to determine if given
+///    element should be taken from \a __W instead.
+///
+/// \code{.operation}
+/// FOR i := 0 to 7
+/// 	IF __U[i]
+/// 		IF i < 4
+/// 			dst.fp16[i] := convert_fp32_to_fp16(__B.fp32[i])
+/// 		ELSE
+/// 			dst.fp16[i] := convert_fp32_to_fp16(__A.fp32[i - 4])
+/// 		FI
+/// 	ELSE
+/// 		dst.fp16[i] := __W.fp16[i]
+/// 	FI
+/// ENDFOR
+///
+/// dst[MAX:128] := 0
+/// \endcode
+///
+/// \headerfile <immintrin.h>
+///
+/// This intrinsic corresponds to the \c VCVT2PS2PHX instruction.
+///
+/// \param __W
+///    A 128-bit vector of [8 x fp16].
+/// \param __U
+///    A 8-bit merging mask.
+/// \param __A
+///    A 128-bit vector of [4 x float].
+/// \param __B
+///    A 128-bit vector of [4 x float].
+/// \returns
+///    A 128-bit vector of [8 x fp16]. Lower elements correspond to the
+///    (converted) elements from \a __B; higher order elements correspond to the
+///    (converted) elements from \a __A. If corresponding mask bit is not set, then
+///    element from \a __W is taken instead.
 static __inline__ __m128h __DEFAULT_FN_ATTRS128
 _mm_mask_cvtx2ps_ph(__m128h __W, __mmask8 __U, __m128 __A, __m128 __B) {
   return (__m128h)__builtin_ia32_vcvt2ps2phx128_mask(
       (__v4sf)__A, (__v4sf)__B, (__v8hf)__W, (__mmask8)__U);
 }
 
+/// Convert two 128-bit vectors, \a __A and \a __B, containing packed
+///    single-precision (32-bit) floating-point elements to a 128-bit vector
+///    containing FP16 elements. Zeroing mask \a __U is used to determine if given
+///    element should be zeroed instead.
+///
+/// \code{.operation}
+/// FOR i := 0 to 7
+/// 	IF __U[i]
+/// 		IF i < 4
+/// 			dst.fp16[i] := convert_fp32_to_fp16(__B.fp32[i])
+/// 		ELSE
+/// 			dst.fp16[i] := convert_fp32_to_fp16(__A.fp32[i - 4])
+/// 		FI
+/// 	ELSE
+/// 		dst.fp16[i] := 0
+/// 	FI
+/// ENDFOR
+///
+/// dst[MAX:128] := 0
+/// \endcode
+///
+/// \headerfile <immintrin.h>
+///
+/// This intrinsic corresponds to the \c VCVT2PS2PHX instruction.
+///
+/// \param __U
+///    A 8-bit zeroing mask.
+/// \param __A
+///    A 128-bit vector of [4 x float].
+/// \param __B
+///    A 128-bit vector of [4 x float].
+/// \returns
+///    A 128-bit vector of [8 x fp16]. Lower elements correspond to the
+///    (converted) elements from \a __B; higher order elements correspond to the
+///    (converted) elements from \a __A. If corresponding mask bit is not set,
+///    then zero is taken instead.
 static __inline__ __m128h __DEFAULT_FN_ATTRS128
 _mm_maskz_cvtx2ps_ph(__mmask8 __U, __m128 __A, __m128 __B) {
   return (__m128h)__builtin_ia32_vcvt2ps2phx128_mask(
       (__v4sf)__A, (__v4sf)__B, (__v8hf)_mm_setzero_ph(), (__mmask8)__U);
 }
 
+/// Convert two 256-bit vectors, \a __A and \a __B, containing packed
+///    single-precision (32-bit) floating-point elements to a 256-bit vector
+///    containing FP16 elements.
+///   
+/// \code{.operation}
+/// FOR i := 0 to 15 
+/// 	IF i < 8
+/// 		dst.fp16[i] := convert_fp32_to_fp16(__B.fp32[i])
+/// 	ELSE
+/// 		dst.fp16[i] := convert_fp32_to_fp16(__A.fp32[i - 8])
+/// 	FI
+/// ENDFOR
+///
+/// dst[MAX:256] := 0
+/// \endcode
+///
+/// \headerfile <immintrin.h>
+///
+/// This intrinsic corresponds to the \c VCVT2PS2PHX instruction.
+///
+/// \param __A
+///    A 256-bit vector of [8 x float].
+/// \param __B
+///    A 256-bit vector of [8 x float].
+/// \returns
+///    A 256-bit vector of [16 x fp16]. Lower elements correspond to the
+///    (converted) elements from \a __B; higher order elements correspond to the
+///    (converted) elements from \a __A.
 static __inline__ __m256h __DEFAULT_FN_ATTRS256 _mm256_cvtx2ps_ph(__m256 __A,
                                                                   __m256 __B) {
   return (__m256h)__builtin_ia32_vcvt2ps2phx256_mask(
-      (__v8sf)__A, (__v8sf)__B, (__v16hf)_mm256_setzero_ph(), (__mmask16)(-1),
-      _MM_FROUND_CUR_DIRECTION);
+      (__v8sf)__A, (__v8sf)__B, (__v16hf)_mm256_setzero_ph(), (__mmask16)(-1));
 }
 
+/// Convert two 256-bit vectors, \a __A and \a __B, containing packed
+///    single-precision (32-bit) floating-point elements to a 256-bit vector
+///    containing FP16 elements. Merging mask \a __U is used to determine if given
+///    element should be taken from \a __W instead.
+///
+/// \code{.operation}
+/// FOR i := 0 to 15
+/// 	IF __U[i]
+/// 		IF i < 8
+/// 			dst.fp16[i] := convert_fp32_to_fp16(__B.fp32[i])
+/// 		ELSE
+/// 			dst.fp16[i] := convert_fp32_to_fp16(__A.fp32[i - 8])
+/// 		FI
+/// 	ELSE
+/// 		dst.fp16[i] := __W.fp16[i]
+/// 	FI
+/// ENDFOR
+///
+/// dst[MAX:256] := 0
+/// \endcode
+///
+/// \headerfile <immintrin.h>
+///
+/// This intrinsic corresponds to the \c VCVT2PS2PHX instruction.
+///
+/// \param __W
+///    A 256-bit vector of [16 x fp16].
+/// \param __U
+///    A 16-bit merging mask.
+/// \param __A
+///    A 256-bit vector of [8 x float].
+/// \param __B
+///    A 256-bit vector of [8 x float].
+/// \returns
+///    A 256-bit vector of [16 x fp16]. Lower elements correspond to the
+///    (converted) elements from \a __B; higher order elements correspond to the
+///    (converted) elements from \a __A. If corresponding mask bit is not set, then
+///    element from \a __W is taken instead.
 static __inline__ __m256h __DEFAULT_FN_ATTRS256
 _mm256_mask_cvtx2ps_ph(__m256h __W, __mmask16 __U, __m256 __A, __m256 __B) {
   return (__m256h)__builtin_ia32_vcvt2ps2phx256_mask(
-      (__v8sf)__A, (__v8sf)__B, (__v16hf)__W, (__mmask16)__U,
-      _MM_FROUND_CUR_DIRECTION);
+      (__v8sf)__A, (__v8sf)__B, (__v16hf)__W, (__mmask16)__U);
 }
 
+/// Convert two 256-bit vectors, \a __A and \a __B, containing packed
+///    single-precision (32-bit) floating-point elements to a 256-bit vector
+///    containing FP16 elements. Zeroing mask \a __U is used to determine if given
+///    element should be zeroed instead.
+///
+/// \code{.operation}
+/// FOR i := 0 to 15 
+/// 	IF __U[i]
+/// 		IF i < 8
+/// 			dst.fp16[i] := convert_fp32_to_fp16(__B.fp32[i])
+/// 		ELSE
+/// 			dst.fp16[i] := convert_fp32_to_fp16(__A.fp32[i - 8])
+/// 		FI
+/// 	ELSE
+/// 		dst.fp16[i] := 0
+/// 	FI
+/// ENDFOR
+///
+/// dst[MAX:256] := 0
+/// \endcode
+///
+/// \headerfile <immintrin.h>
+///
+/// This intrinsic corresponds to the \c VCVT2PS2PHX instruction.
+///
+/// \param __U
+///    A 16-bit zeroing mask.
+/// \param __A
+///    A 256-bit vector of [8 x float].
+/// \param __B
+///    A 256-bit vector of [8 x float].
+/// \returns
+///    A 256-bit vector of [16 x fp16]. Lower elements correspond to the
+///    (converted) elements from \a __B; higher order elements correspond to the
+///    (converted) elements from \a __A. If corresponding mask bit is not set,
+///    then zero is taken instead.
 static __inline__ __m256h __DEFAULT_FN_ATTRS256
 _mm256_maskz_cvtx2ps_ph(__mmask16 __U, __m256 __A, __m256 __B) {
   return (__m256h)__builtin_ia32_vcvt2ps2phx256_mask(
-      (__v8sf)__A, (__v8sf)__B, (__v16hf)_mm256_setzero_ph(), (__mmask16)__U,
-      _MM_FROUND_CUR_DIRECTION);
+      (__v8sf)__A, (__v8sf)__B, (__v16hf)_mm256_setzero_ph(), (__mmask16)__U);
 }
 
-#define _mm256_cvtx_round2ps_ph(A, B, R)                                       \
-  ((__m256h)__builtin_ia32_vcvt2ps2phx256_mask(                                \
-      (__v8sf)(A), (__v8sf)(B), (__v16hf)_mm256_undefined_ph(),                \
-      (__mmask16)(-1), (const int)(R)))
-
-#define _mm256_mask_cvtx_round2ps_ph(W, U, A, B, R)                            \
-  ((__m256h)__builtin_ia32_vcvt2ps2phx256_mask(                                \
-      (__v8sf)(A), (__v8sf)(B), (__v16hf)(W), (__mmask16)(U), (const int)(R)))
-
-#define _mm256_maskz_cvtx_round2ps_ph(U, A, B, R)                              \
-  ((__m256h)__builtin_ia32_vcvt2ps2phx256_mask(                                \
-      (__v8sf)(A), (__v8sf)(B), (__v16hf)(_mm256_setzero_ph()),                \
-      (__mmask16)(U), (const int)(R)))
-
-static __inline__ __m128i __DEFAULT_FN_ATTRS128 _mm_cvtbiasph_bf8(__m128i __A,
-                                                                  __m128h __B) {
+/// Convert 128-bit vector \a __B containing packed FP16 floating-point elements
+///    to FP8 E5M2 numbers, using conversion biases stored in lower 8 bits of each
+///    16-bit integer stored in \a __B.
+///
+/// \code{.operation}
+/// FOR i := 0 to 7
+/// 	dst.bf8[i] := convert_fp16_to_bf8_with_bias(__A.int8[2 * i], __B.fp16[i])
+/// ENDFOR
+///
+/// dst[MAX:64] := 0
+/// \endcode
+///
+/// \headerfile <immintrin.h>
+///
+/// This intrinsic corresponds to the \c VCVTBIASPH2BF8 instruction.
+///
+/// \param __A
+///    A 128-bit vector of [8 x int16].
+/// \param __B
+///    A 128-bit vector of [8 x fp16].
+/// \returns
+///    A 128-bit vector of [16 x bf8]. Lower elements correspond to the
+///    converted elements from \a __B using biases from \a __A; higher order
+///    elements are zeroed.
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
+_mm_cvtbiasph_bf8(__m128i __A, __m128h __B) {
   return (__m128i)__builtin_ia32_vcvtbiasph2bf8_128_mask(
       (__v16qi)__A, (__v8hf)__B, (__v16qi)_mm_undefined_si128(), (__mmask8)-1);
 }
 
+/// Convert 128-bit vector \a __B containing packed FP16 floating-point elements
+///    to FP8 E5M2 numbers, using conversion biases stored in lower 8 bits of each
+///    16-bit integer stored in \a __B. Merging mask \a __U is used to determine if
+///    given element should be taken from \a __W instead.
+///
+/// \code{.operation}
+/// FOR i := 0 to 7
+/// 	IF __U[i]
+/// 		dst.bf8[i] := convert_fp16_to_bf8_with_bias(__A.int8[2 * i], __B.fp16[i])
+/// 	ELSE
+/// 		dst.bf8[i] := __W.bf8[i]
+/// 	FI
+/// ENDFOR
+///
+/// dst[MAX:64] := 0
+/// \endcode
+///
+/// \headerfile <immintrin.h>
+///
+/// This intrinsic corresponds to the \c VCVTBIASPH2BF8 instruction.
+///
+/// \param __W
+///    A 128-bit vector of [16 x bf8].
+/// \param __U
+///    A 8-bit merging mask.
+/// \param __A
+///    A 128-bit vector of [8 x int16].
+/// \param __B
+///    A 128-bit vector of [8 x fp16].
+/// \returns
+///    A 128-bit vector of [16 x bf8]. Lower elements correspond to the
+///    converted elements from \a __B, using biases from \a __A; higher order
+///    elements are zeroed. If corresponding mask bit is not set, then element
+///    from \a __W is taken instead.
 static __inline__ __m128i __DEFAULT_FN_ATTRS128
 _mm_mask_cvtbiasph_bf8(__m128i __W, __mmask8 __U, __m128i __A, __m128h __B) {
   return (__m128i)__builtin_ia32_vcvtbiasph2bf8_128_mask(
       (__v16qi)__A, (__v8hf)__B, (__v16qi)(__m128i)__W, (__mmask8)__U);
 }
 
+/// Convert 128-bit vector \a __B containing packed FP16 floating-point elements
+///    to FP8 E5M2 numbers, using conversion biases stored in lower 8 bits of each
+///    16-bit integer stored in \a __B. Zeroing mask \a __U is used to determine if
+///    given element should be zeroed instead.
+///
+/// \code{.operation}
+/// FOR i := 0 to 7
+/// 	IF __U[i]
+///	 	dst.bf8[i] := convert_fp16_to_bf8_with_bias(__A.int8[2 * i], __B.fp16[i])
+///	 ELSE
+///	 	dst.bf8[i] := 0
+///	 FI
+/// ENDFOR
+///
+/// dst[MAX:64] := 0
+/// \endcode
+///
+/// \headerfile <immintrin.h>
+///
+/// This intrinsic corresponds to the \c VCVTBIASPH2BF8 instruction.
+///
+/// \param __U
+///    A 8-bit zeroing mask.
+/// \param __A
+///    A 128-bit vector of [8 x int16].
+/// \param __B
+///    A 128-bit vector of [8 x fp16].
+/// \returns
+///    A 128-bit vector of [16 x bf8]. Lower elements correspond to the
+///    converted elements from \a __B, using biases from \a __A; higher order
+///    elements are zeroed. If corresponding mask bit is not set, then element
+///    is zeroed.
 static __inline__ __m128i __DEFAULT_FN_ATTRS128
 _mm_maskz_cvtbiasph_bf8(__mmask8 __U, __m128i __A, __m128h __B) {
   return (__m128i)__builtin_ia32_vcvtbiasph2bf8_128_mask(
@@ -96,6 +376,29 @@ _mm_maskz_cvtbiasph_bf8(__mmask8 __U, __m128i __A, __m128h __B) {
       (__mmask8)__U);
 }
 
+/// Convert 256-bit vector \a __B containing packed FP16 floating-point elements
+///    to FP8 E5M2 numbers, using conversion biases stored in lower 8 bits of each
+///    16-bit integer stored in \a __B.
+///
+/// \code{.operation}
+/// FOR i := 0 to 15
+/// 	dst.bf8[i] := convert_fp16_to_bf8_with_bias(__A.int8[2 * i], __B.fp16[i])
+/// ENDFOR
+///
+/// dst[MAX:128] := 0
+/// \endcode
+///
+/// \headerfile <immintrin.h>
+///
+/// This intrinsic corresponds to the \c VCVTBIASPH2BF8 instruction.
+///
+/// \param __A
+///    A 256-bit vector of [16 x int16].
+/// \param __B
+///    A 256-bit vector of [16 x fp16].
+/// \returns
+///    A 128-bit vector of [16 x bf8]. Elements correspond to the
+///    converted elements from \a __B using biases from \a __A.
 static __inline__ __m128i __DEFAULT_FN_ATTRS256
 _mm256_cvtbiasph_bf8(__m256i __A, __m256h __B) {
   return (__m128i)__builtin_ia32_vcvtbiasph2bf8_256_mask(
@@ -103,12 +406,76 @@ _mm256_cvtbiasph_bf8(__m256i __A, __m256h __B) {
       (__mmask16)-1);
 }
 
+/// Convert 256-bit vector \a __B containing packed FP16 floating-point elements
+///    to FP8 E5M2 numbers, using conversion biases stored in lower 8 bits of each
+///    16-bit integer stored in \a __B. Merging mask \a __U is used to determine if
+///    given element should be taken from \a __W instead.
+///
+/// \code{.operation}
+/// FOR i := 0 to 15
+/// 	IF __U[i]
+/// 		dst.bf8[i] := convert_fp16_to_bf8_with_bias(__A.int8[2 * i], __B.fp16[i])
+/// 	ELSE
+/// 		dst.bf8[i] := __W.bf8[i]
+/// 	FI
+/// ENDFOR
+///
+/// dst[MAX:128] := 0
+/// \endcode
+///
+/// \headerfile <immintrin.h>
+///
+/// This intrinsic corresponds to the \c VCVTBIASPH2BF8 instruction.
+///
+/// \param __W
+///    A 128-bit vector of [16 x bf8].
+/// \param __U
+///    A 16-bit merging mask.
+/// \param __A
+///    A 256-bit vector of [16 x int16].
+/// \param __B
+///    A 256-bit vector of [16 x fp16].
+/// \returns
+///    A 128-bit vector of [16 x bf8]. Elements correspond to the converted
+///    elements from \a __B, using biases from \a __A. If corresponding mask bit
+///    is not set, then element from \a __W is taken instead.
 static __inline__ __m128i __DEFAULT_FN_ATTRS256 _mm256_mask_cvtbiasph_bf8(
     __m128i __W, __mmask16 __U, __m256i __A, __m256h __B) {
   return (__m128i)__builtin_ia32_vcvtbiasph2bf8_256_mask(
       (__v32qi)__A, (__v16hf)__B, (__v16qi)(__m128i)__W, (__mmask16)__U);
 }
 
+/// Convert 256-bit vector \a __B containing packed FP16 floating-point elements
+///    to FP8 E5M2 numbers, using conversion biases stored in lower 8 bits of each
+///    16-bit integer stored in \a __B. Zeroing mask \a __U is used to determine if
+///    given element should be zeroed instead.
+///
+/// \code{.operation}
+/// FOR i := 0 to 15
+/// 	IF __U[i]
+///	 	dst.bf8[i] := convert_fp16_to_bf8_with_bias(__A.int8[2 * i], __B.fp16[i])
+///	 ELSE
+///	 	dst.bf8[i] := 0
+///	 FI
+/// ENDFOR
+///
+/// dst[MAX:128] := 0
+/// \endcode
+///
+/// \headerfile <immintrin.h>
+///
+/// This intrinsic corresponds to the \c VCVTBIASPH2BF8 instruction.
+///
+/// \param __U
+///    A 16-bit zeroing mask.
+/// \param __A
+///    A 256-bit vector of [16 x int16].
+/// \param __B
+///    A 256-bit vector of [16 x fp16].
+/// \returns
+///    A 128-bit vector of [16 x bf8]. Elements correspond to the converted
+///    elements from \a __B, using biases from \a __A. If corresponding mask bit
+///    is not set, then element is zeroed.
 static __inline__ __m128i __DEFAULT_FN_ATTRS256
 _mm256_maskz_cvtbiasph_bf8(__mmask16 __U, __m256i __A, __m256h __B) {
   return (__m128i)__builtin_ia32_vcvtbiasph2bf8_256_mask(
@@ -116,57 +483,324 @@ _mm256_maskz_cvtbiasph_bf8(__mmask16 __U, __m256i __A, __m256h __B) {
       (__mmask16)__U);
 }
 
+/// Convert 128-bit vector \a __B containing packed FP16 floating-point elements
+///    to FP8 E5M2 numbers, using conversion biases stored in lower 8 bits of each
+///    16-bit integer stored in \a __B. Results are saturated.
+///
+/// \code{.operation}
+/// FOR i := 0 to 7
+/// 	dst.bf8[i] := convert_fp16_to_bf8_with_bias_saturate(__A.int8[2 * i], __B.fp16[i])
+/// ENDFOR
+///
+/// dst[MAX:64] := 0
+/// \endcode
+///
+/// \headerfile <immintrin.h>
+///
+/// This intrinsic corresponds to the \c VCVTBIASPH2BF8 instruction.
+///
+/// \param __A
+///    A 128-bit vector of [8 x int16].
+/// \param __B
+///    A 128-bit vector of [8 x fp16].
+/// \returns
+///    A 128-bit vector of [16 x bf8]. Lower elements correspond to the
+///    converted elements from \a __B using biases from \a __A; higher order
+///    elements are zeroed.
 static __inline__ __m128i __DEFAULT_FN_ATTRS128
-_mm_cvtbiassph_bf8(__m128i __A, __m128h __B) {
+_mm_cvts_biasph_bf8(__m128i __A, __m128h __B) {
   return (__m128i)__builtin_ia32_vcvtbiasph2bf8s_128_mask(
       (__v16qi)__A, (__v8hf)__B, (__v16qi)_mm_undefined_si128(), (__mmask8)-1);
 }
 
-static __inline__ __m128i __DEFAULT_FN_ATTRS128
-_mm_mask_cvtbiassph_bf8(__m128i __W, __mmask8 __U, __m128i __A, __m128h __B) {
-  return (__m128i)__builtin_ia32_vcvtbiasph2bf8s_128_mask(
-      (__v16qi)__A, (__v8hf)__B, (__v16qi)(__m128i)__W, (__mmask8)__U);
-}
+/// Convert 128-bit vector \a __B containing packed FP16 floating-point elements
+///    to FP8 E5M2 numbers, using conversion biases stored in lower 8 bits of each
+///    16-bit integer stored in \a __B. Results are saturated. Merging mask \a __U
+///    is used to determine if given element should be taken from \a __W instead.
+///
+/// \code{.operation}
+/// FOR i := 0 to 7
+/// 	IF __U[i]
+/// 		dst.bf8[i] := convert_fp16_to_bf8_with_bias_saturate(__A.int8[2 * i], __B.fp16[i])
+/// 	ELSE
+/// 		dst.bf8[i] := __W.bf8[i]
+/// 	FI
+/// ENDFOR
+///
+/// dst[MAX:64] := 0
+/// \endcode
+///
+/// \headerfile <immintrin.h>
+///
+/// This intrinsic corresponds to the \c VCVTBIASPH2BF8S instruction.
+///
+/// \param __W
+///    A 128-bit vector of [16 x bf8].
+/// \param __U
+///    A 8-bit merging mask.
+/// \param __A
+///    A 128-bit vector of [8 x int16].
+/// \param __B
+///    A 128-bit vector of [8 x fp16].
+/// \returns
+///    A 128-bit vector of [16 x bf8]. Lower elements correspond to the
+///    converted elements from \a __B, using biases from \a __A; higher order
+///    elements are zeroed. If corresponding mask bit is not set, then element
+///    from \a __W is taken instead.
+static __inline__ __m128i __DEFAULT_FN_ATTRS128 _mm_mask_cvts_biasph_bf8(__m128i
+		__W, __mmask8 __U, __m128i __A, __m128h __B) { return
+	(__m128i)__builtin_ia32_vcvtbiasph2bf8s_128_mask( (__v16qi)__A,
+			(__v8hf)__B, (__v16qi)(__m128i)__W, (__mmask8)__U); }
 
+/// Convert 128-bit vector \a __B containing packed FP16 floating-point elements
+///    to FP8 E5M2 numbers, using conversion biases stored in lower 8 bits of each
+///    16-bit integer stored in \a __B. Results are saturated. Zeroing mask \a __U
+///    is used to determine if given element should be zeroed instead.
+///
+/// \code{.operation}
+/// FOR i := 0 to 7
+/// 	IF __U[i]
+///	 	dst.bf8[i] := convert_fp16_to_bf8_with_bias_saturate(__A.int8[2 * i], __B.fp16[i])
+///	 ELSE
+///	 	dst.bf8[i] := 0
+///	 FI
+/// ENDFOR
+///
+/// dst[MAX:64] := 0
+/// \endcode
+///
+/// \headerfile <immintrin.h>
+///
+/// This intrinsic corresponds to the \c VCVTBIASPH2BF8S instruction.
+///
+/// \param __U
+///    A 8-bit zeroing mask.
+/// \param __A
+///    A 128-bit vector of [8 x int16].
+/// \param __B
+///    A 128-bit vector of [8 x fp16].
+/// \returns
+///    A 128-bit vector of [16 x bf8]. Lower elements correspond to the
+///    converted elements from \a __B, using biases from \a __A; higher order
+///    elements are zeroed. If corresponding mask bit is not set, then element
+///    is zeroed.
 static __inline__ __m128i __DEFAULT_FN_ATTRS128
-_mm_maskz_cvtbiassph_bf8(__mmask8 __U, __m128i __A, __m128h __B) {
+_mm_maskz_cvts_biasph_bf8(__mmask8 __U, __m128i __A, __m128h __B) {
   return (__m128i)__builtin_ia32_vcvtbiasph2bf8s_128_mask(
       (__v16qi)__A, (__v8hf)__B, (__v16qi)(__m128i)_mm_setzero_si128(),
       (__mmask8)__U);
 }
 
+
+/// Convert 256-bit vector \a __B containing packed FP16 floating-point elements
+///    to FP8 E5M2 numbers, using conversion biases stored in lower 8 bits of each
+///    16-bit integer stored in \a __B. Results are saturated.
+///
+/// \code{.operation}
+/// FOR i := 0 to 15
+/// 	dst.bf8[i] := convert_fp16_to_bf8_with_bias_saturate(__A.int8[2 * i], __B.fp16[i])
+/// ENDFOR
+///
+/// dst[MAX:128] := 0
+/// \endcode
+///
+/// \headerfile <immintrin.h>
+///
+/// This intrinsic corresponds to the \c VCVTBIASPH2BF8S instruction.
+///
+/// \param __A
+///    A 256-bit vector of [16 x int16].
+/// \param __B
+///    A 256-bit vector of [16 x fp16].
+/// \returns
+///    A 128-bit vector of [16 x bf8]. Elements correspond to the
+///    converted elements from \a __B using biases from \a __A.
 static __inline__ __m128i __DEFAULT_FN_ATTRS256
-_mm256_cvtbiassph_bf8(__m256i __A, __m256h __B) {
+_mm256_cvts_biasph_bf8(__m256i __A, __m256h __B) {
   return (__m128i)__builtin_ia32_vcvtbiasph2bf8s_256_mask(
       (__v32qi)__A, (__v16hf)__B, (__v16qi)(__m128i)_mm_undefined_si128(),
       (__mmask16)-1);
 }
 
-static __inline__ __m128i __DEFAULT_FN_ATTRS256 _mm256_mask_cvtbiassph_bf8(
+/// Convert 256-bit vector \a __B containing packed FP16 floating-point elements
+///    to FP8 E5M2 numbers, using conversion biases stored in lower 8 bits of each
+///    16-bit integer stored in \a __B. Results are saturated. Merging mask \a __U
+///    is used to determine if given element should be taken from \a __W instead.
+///
+/// \code{.operation}
+/// FOR i := 0 to 15
+/// 	IF __U[i]
+/// 		dst.bf8[i] := convert_fp16_to_bf8_with_bias_saturate(__A.int8[2 * i], __B.fp16[i])
+/// 	ELSE
+/// 		dst.bf8[i] := __W.bf8[i]
+/// 	FI
+/// ENDFOR
+///
+/// dst[MAX:128] := 0
+/// \endcode
+///
+/// \headerfile <immintrin.h>
+///
+/// This intrinsic corresponds to the \c VCVTBIASPH2BF8S instruction.
+///
+/// \param __W
+///    A 128-bit vector of [16 x bf8].
+/// \param __U
+///    A 16-bit merging mask.
+/// \param __A
+///    A 256-bit vector of [16 x int16].
+/// \param __B
+///    A 256-bit vector of [16 x fp16].
+/// \returns
+///    A 128-bit vector of [16 x bf8]. Elements correspond to the converted
+///    elements from \a __B, using biases from \a __A. If corresponding mask bit
+///    is not set, then element from \a __W is taken instead.
+static __inline__ __m128i __DEFAULT_FN_ATTRS256 _mm256_mask_cvts_biasph_bf8(
     __m128i __W, __mmask16 __U, __m256i __A, __m256h __B) {
   return (__m128i)__builtin_ia32_vcvtbiasph2bf8s_256_mask(
       (__v32qi)__A, (__v16hf)__B, (__v16qi)(__m128i)__W, (__mmask16)__U);
 }
 
+/// Convert 256-bit vector \a __B containing packed FP16 floating-point elements
+///    to FP8 E5M2 numbers, using conversion biases stored in lower 8 bits of each
+///    16-bit integer stored in \a __B. Results are saturated. Zeroing mask \a __U
+///    is used to determine if given element should be zeroed instead.
+///
+/// \code{.operation}
+/// FOR i := 0 to 15
+/// 	IF __U[i]
+///	 	dst.bf8[i] := convert_fp16_to_bf8_with_bias_saturate(__A.int8[2 * i], __B.fp16[i])
+///	 ELSE
+///	 	dst.bf8[i] := 0
+///	 FI
+/// ENDFOR
+///
+/// dst[MAX:128] := 0
+/// \endcode
+///
+/// \headerfile <immintrin.h>
+///
+/// This intrinsic corresponds to the \c VCVTBIASPH2BF8S instruction.
+///
+/// \param __U
+///    A 16-bit zeroing mask.
+/// \param __A
+///    A 256-bit vector of [16 x int16].
+/// \param __B
+///    A 256-bit vector of [16 x fp16].
+/// \returns
+///    A 128-bit vector of [16 x bf8]. Elements correspond to the converted
+///    elements from \a __B, using biases from \a __A. If corresponding mask bit
+///    is not set, then element is zeroed.
 static __inline__ __m128i __DEFAULT_FN_ATTRS256
-_mm256_maskz_cvtbiassph_bf8(__mmask16 __U, __m256i __A, __m256h __B) {
+_mm256_maskz_cvts_biasph_bf8(__mmask16 __U, __m256i __A, __m256h __B) {
   return (__m128i)__builtin_ia32_vcvtbiasph2bf8s_256_mask(
       (__v32qi)__A, (__v16hf)__B, (__v16qi)(__m128i)_mm_setzero_si128(),
       (__mmask16)__U);
 }
 
-static __inline__ __m128i __DEFAULT_FN_ATTRS128 _mm_cvtbiasph_hf8(__m128i __A,
-                                                                  __m128h __B) {
+/// Convert 128-bit vector \a __B containing packed FP16 floating-point elements
+///    to FP8 E4M3 numbers, using conversion biases stored in lower 8 bits of each
+///    16-bit integer stored in \a __B.
+///
+/// \code{.operation}
+/// FOR i := 0 to 7
+/// 	dst.hf8[i] := convert_fp16_to_hf8_with_bias(__A.int8[2 * i], __B.fp16[i])
+/// ENDFOR
+///
+/// dst[MAX:64] := 0
+/// \endcode
+///
+/// \headerfile <immintrin.h>
+///
+/// This intrinsic corresponds to the \c VCVTBIASPH2HF8 instruction.
+///
+/// \param __A
+///    A 128-bit vector of [8 x int16].
+/// \param __B
+///    A 128-bit vector of [8 x fp16].
+/// \returns
+///    A 128-bit vector of [16 x hf8]. Lower elements correspond to the
+///    converted elements from \a __B using biases from \a __A; higher order
+///    elements are zeroed.
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
+_mm_cvtbiasph_hf8(__m128i __A, __m128h __B) {
   return (__m128i)__builtin_ia32_vcvtbiasph2hf8_128_mask(
       (__v16qi)__A, (__v8hf)__B, (__v16qi)_mm_undefined_si128(), (__mmask8)-1);
 }
 
+/// Convert 128-bit vector \a __B containing packed FP16 floating-point elements
+///    to FP8 E4M3 numbers, using conversion biases stored in lower 8 bits of each
+///    16-bit integer stored in \a __B. Merging mask \a __U is used to determine if
+///    given element should be taken from \a __W instead.
+///
+/// \code{.operation}
+/// FOR i := 0 to 7
+/// 	IF __U[i]
+/// 		dst.hf8[i] := convert_fp16_to_hf8_with_bias(__A.int8[2 * i], __B.fp16[i])
+/// 	ELSE
+/// 		dst.hf8[i] := __W.hf8[i]
+/// 	FI
+/// ENDFOR
+///
+/// dst[MAX:64] := 0
+/// \endcode
+///
+/// \headerfile <immintrin.h>
+///
+/// This intrinsic corresponds to the \c VCVTBIASPH2HF8 instruction.
+///
+/// \param __W
+///    A 128-bit vector of [16 x hf8].
+/// \param __U
+///    A 8-bit merging mask.
+/// \param __A
+///    A 128-bit vector of [8 x int16].
+/// \param __B
+///    A 128-bit vector of [8 x fp16].
+/// \returns
+///    A 128-bit vector of [16 x hf8]. Lower elements correspond to the
+///    converted elements from \a __B, using biases from \a __A; higher order
+///    elements are zeroed. If corresponding mask bit is not set, then element
+///    from \a __W is taken instead.
 static __inline__ __m128i __DEFAULT_FN_ATTRS128
 _mm_mask_cvtbiasph_hf8(__m128i __W, __mmask8 __U, __m128i __A, __m128h __B) {
   return (__m128i)__builtin_ia32_vcvtbiasph2hf8_128_mask(
       (__v16qi)__A, (__v8hf)__B, (__v16qi)(__m128i)__W, (__mmask8)__U);
 }
 
+/// Convert 128-bit vector \a __B containing packed FP16 floating-point elements
+///    to FP8 E4M3 numbers, using conversion biases stored in lower 8 bits of each
+///    16-bit integer stored in \a __B. Zeroing mask \a __U is used to determine if
+///    given element should be zeroed instead.
+///
+/// \code{.operation}
+/// FOR i := 0 to 7
+/// 	IF __U[i]
+///	 	dst.hf8[i] := convert_fp16_to_hf8_with_bias(__A.int8[2 * i], __B.fp16[i])
+///	 ELSE
+///	 	dst.hf8[i] := 0
+///	 FI
+/// ENDFOR
+///
+/// dst[MAX:64] := 0
+/// \endcode
+///
+/// \headerfile <immintrin.h>
+///
+/// This intrinsic corresponds to the \c VCVTBIASPH2HF8 instruction.
+///
+/// \param __U
+///    A 8-bit zeroing mask.
+/// \param __A
+///    A 128-bit vector of [8 x int16].
+/// \param __B
+///    A 128-bit vector of [8 x fp16].
+/// \returns
+///    A 128-bit vector of [16 x hf8]. Lower elements correspond to the
+///    converted elements from \a __B, using biases from \a __A; higher order
+///    elements are zeroed. If corresponding mask bit is not set, then element
+///    is zeroed.
 static __inline__ __m128i __DEFAULT_FN_ATTRS128
 _mm_maskz_cvtbiasph_hf8(__mmask8 __U, __m128i __A, __m128h __B) {
   return (__m128i)__builtin_ia32_vcvtbiasph2hf8_128_mask(
@@ -174,6 +808,29 @@ _mm_maskz_cvtbiasph_hf8(__mmask8 __U, __m128i __A, __m128h __B) {
       (__mmask8)__U);
 }
 
+/// Convert 256-bit vector \a __B containing packed FP16 floating-point elements
+///    to FP8 E4M3 numbers, using conversion biases stored in lower 8 bits of each
+///    16-bit integer stored in \a __B.
+///
+/// \code{.operation}
+/// FOR i := 0 to 15
+/// 	dst.hf8[i] := convert_fp16_to_hf8_with_bias(__A.int8[2 * i], __B.fp16[i])
+/// ENDFOR
+///
+/// dst[MAX:128] := 0
+/// \endcode
+///
+/// \headerfile <immintrin.h>
+///
+/// This intrinsic corresponds to the \c VCVTBIASPH2HF8 instruction.
+///
+/// \param __A
+///    A 256-bit vector of [16 x half].
+/// \param __B
+///    A 256-bit vector of [16 x i16].
+/// \returns
+///    A 128-bit vector of [16 x hf8]. Elements correspond to the
+///    converted elements from \a __B using biases from \a __A.
 static __inline__ __m128i __DEFAULT_FN_ATTRS256
 _mm256_cvtbiasph_hf8(__m256i __A, __m256h __B) {
   return (__m128i)__builtin_ia32_vcvtbiasph2hf8_256_mask(
@@ -181,12 +838,76 @@ _mm256_cvtbiasph_hf8(__m256i __A, __m256h __B) {
       (__mmask16)-1);
 }
 
+/// Convert 256-bit vector \a __B containing packed FP16 floating-point elements
+///    to FP8 E4M3 numbers, using conversion biases stored in lower 8 bits of each
+///    16-bit integer stored in \a __B. Merging mask \a __U is used to determine if
+///    given element should be taken from \a __W instead.
+///
+/// \code{.operation}
+/// FOR i := 0 to 15
+/// 	IF __U[i]
+/// 		dst.hf8[i] := convert_fp16_to_hf8_with_bias(__A.int8[2 * i], __B.fp16[i])
+/// 	ELSE
+/// 		dst.hf8[i] := __W.hf8[i]
+/// 	FI
+/// ENDFOR
+///
+/// dst[MAX:128] := 0
+/// \endcode
+///
+/// \headerfile <immintrin.h>
+///
+/// This intrinsic corresponds to the \c VCVTBIASPH2HF8 instruction.
+///
+/// \param __W
+///    A 128-bit vector of [16 x hf8].
+/// \param __U
+///    A 16-bit merging mask.
+/// \param __A
+///    A 256-bit vector of [16 x int16].
+/// \param __B
+///    A 256-bit vector of [16 x fp16].
+/// \returns
+///    A 128-bit vector of [16 x hf8]. Elements correspond to the converted
+///    elements from \a __B, using biases from \a __A. If corresponding mask bit
+///    is not set, then element from \a __W is taken instead.
 static __inline__ __m128i __DEFAULT_FN_ATTRS256 _mm256_mask_cvtbiasph_hf8(
     __m128i __W, __mmask16 __U, __m256i __A, __m256h __B) {
   return (__m128i)__builtin_ia32_vcvtbiasph2hf8_256_mask(
       (__v32qi)__A, (__v16hf)__B, (__v16qi)(__m128i)__W, (__mmask16)__U);
 }
 
+/// Convert 256-bit vector \a __B containing packed FP16 floating-point elements
+///    to FP8 E4M3 numbers, using conversion biases stored in lower 8 bits of each
+///    16-bit integer stored in \a __B. Zeroing mask \a __U is used to determine if
+///    given element should be taken zeroed instead.
+///
+/// \code{.operation}
+/// FOR i := 0 to 15
+/// 	IF __U[i]
+///	 	dst.hf8[i] := convert_fp16_to_hf8_with_bias(__A.int8[2 * i], __B.fp16[i])
+///	 ELSE
+///	 	dst.hf8[i] := 0
+///	 FI
+/// ENDFOR
+///
+/// dst[MAX:128] := 0
+/// \endcode
+///
+/// \headerfile <immintrin.h>
+///
+/// This intrinsic corresponds to the \c VCVTBIASPH2HF8 instruction.
+///
+/// \param __U
+///    A 16-bit zeroing mask.
+/// \param __A
+///    A 256-bit vector of [16 x half].
+/// \param __B
+///    A 256-bit vector of [16 x i16].
+/// \returns
+///    A 128-bit vector of [16 x hf8]. Elements correspond to the converted
+///    elements from \a __B, using biases from \a __A. If corresponding mask bit
+///    is not set, then element is zeroed.
 static __inline__ __m128i __DEFAULT_FN_ATTRS256
 _mm256_maskz_cvtbiasph_hf8(__mmask16 __U, __m256i __A, __m256h __B) {
   return (__m128i)__builtin_ia32_vcvtbiasph2hf8_256_mask(
@@ -194,56 +915,335 @@ _mm256_maskz_cvtbiasph_hf8(__mmask16 __U, __m256i __A, __m256h __B) {
       (__mmask16)__U);
 }
 
+/// Convert 128-bit vector \a __B containing packed FP16 floating-point elements
+///    to FP8 E4M3 numbers, using conversion biases stored in lower 8 bits of each
+///    16-bit integer stored in \a __B. Results are saturated.
+///
+/// \code{.operation}
+/// FOR i := 0 to 7
+/// 	dst.hf8[i] := convert_fp16_to_hf8_with_bias_saturate(__A.int8[2 * i], __B.fp16[i])
+/// ENDFOR
+///
+/// dst[MAX:64] := 0
+/// \endcode
+///
+/// \headerfile <immintrin.h>
+///
+/// This intrinsic corresponds to the \c VCVTBIASPH2HF8S`instruction.
+///
+/// \param __A
+///    A 128-bit vector of [8 x int16].
+/// \param __B
+///    A 128-bit vector of [8 x fp16].
+/// \returns
+///    A 128-bit vector of [16 x hf8]. Lower elements correspond to the
+///    converted elements from \a __B using biases from \a __A; higher order
+///    elements are zeroed.
 static __inline__ __m128i __DEFAULT_FN_ATTRS128
-_mm_cvtbiassph_hf8(__m128i __A, __m128h __B) {
+_mm_cvts_biasph_hf8(__m128i __A, __m128h __B) {
   return (__m128i)__builtin_ia32_vcvtbiasph2hf8s_128_mask(
       (__v16qi)__A, (__v8hf)__B, (__v16qi)_mm_undefined_si128(), (__mmask8)-1);
 }
 
+/// Convert 128-bit vector \a __B containing packed FP16 floating-point elements
+///    to FP8 E4M3 numbers, using conversion biases stored in lower 8 bits of each
+///    16-bit integer stored in \a __B. Results are saturated. Merging mask \a __U
+///    is used to determine if given element should be taken from \a __W instead.
+///
+/// \code{.operation}
+/// FOR i := 0 to 7
+/// 	IF __U[i]
+/// 		dst.hf8[i] := convert_fp16_to_hf8_with_bias_saturate(__A.int8[2 * i], __B.fp16[i])
+/// 	ELSE
+/// 		dst.hf8[i] := __W.hf8[i]
+/// 	FI
+/// ENDFOR
+///
+/// dst[MAX:64] := 0
+/// \endcode
+///
+/// \headerfile <immintrin.h>
+///
+/// This intrinsic corresponds to the \c VCVTBIASPH2HF8S instruction.
+///
+/// \param __W
+///    A 128-bit vector of [16 x hf8].
+/// \param __U
+///    A 8-bit merging mask.
+/// \param __A
+///    A 128-bit vector of [8 x int16].
+/// \param __B
+///    A 128-bit vector of [8 x fp16].
+/// \returns
+///    A 128-bit vector of [16 x hf8]. Lower elements correspond to the
+///    converted elements from \a __B, using biases from \a __A; higher order
+///    elements are zeroed. If corresponding mask bit is not set, then element
+///    from \a __W is taken instead.
 static __inline__ __m128i __DEFAULT_FN_ATTRS128
-_mm_mask_cvtbiassph_hf8(__m128i __W, __mmask8 __U, __m128i __A, __m128h __B) {
+_mm_mask_cvts_biasph_hf8(__m128i __W, __mmask8 __U, __m128i __A, __m128h __B) {
   return (__m128i)__builtin_ia32_vcvtbiasph2hf8s_128_mask(
       (__v16qi)__A, (__v8hf)__B, (__v16qi)(__m128i)__W, (__mmask8)__U);
 }
 
+/// Convert 128-bit vector \a __B containing packed FP16 floating-point elements
+///    to FP8 E4M3 numbers, using conversion biases stored in lower 8 bits of each
+///    16-bit integer stored in \a __B. Results are saturated. Zeroing mask \a __U
+///    is used to determine if given element should be zeroed instead.
+///
+/// \code{.operation}
+/// FOR i := 0 to 7
+/// 	IF __U[i]
+///	 	dst.hf8[i] := convert_fp16_to_hf8_with_bias_saturate(__A.int8[2 * i], __B.fp16[i])
+///	 ELSE
+///	 	dst.hf8[i] := 0
+///	 FI
+/// ENDFOR
+///
+/// dst[MAX:64] := 0
+/// \endcode
+///
+/// \headerfile <immintrin.h>
+///
+/// This intrinsic corresponds to the \c VCVTBIASPH2HF8S instruction.
+///
+/// \param __U
+///    A 8-bit zeroing mask.
+/// \param __A
+///    A 128-bit vector of [8 x int16].
+/// \param __B
+///    A 128-bit vector of [8 x fp16].
+/// \returns
+///    A 128-bit vector of [16 x hf8]. Lower elements correspond to the
+///    converted elements from \a __B, using biases from \a __A; higher order
+///    elements are zeroed. If corresponding mask bit is not set, then element
+///    is zeroed.
 static __inline__ __m128i __DEFAULT_FN_ATTRS128
-_mm_maskz_cvtbiassph_hf8(__mmask8 __U, __m128i __A, __m128h __B) {
+_mm_maskz_cvts_biasph_hf8(__mmask8 __U, __m128i __A, __m128h __B) {
   return (__m128i)__builtin_ia32_vcvtbiasph2hf8s_128_mask(
       (__v16qi)__A, (__v8hf)__B, (__v16qi)(__m128i)_mm_setzero_si128(),
       (__mmask8)__U);
 }
 
+/// Convert 256-bit vector \a __B containing packed FP16 floating-point elements
+///    to FP8 E4M3 numbers, using conversion biases stored in lower 8 bits of each
+///    16-bit integer stored in \a __B. Results are saturated.
+///
+/// \code{.operation}
+/// FOR i := 0 to 15
+/// 	dst.hf8[i] := convert_fp16_to_hf8_with_bias_saturate(__A.int8[2 * i], __B.fp16[i])
+/// ENDFOR
+///
+/// dst[MAX:128] := 0
+/// \endcode
+///
+/// \headerfile <immintrin.h>
+///
+/// This intrinsic corresponds to the \c VCVTBIASPH2HF8S instruction.
+///
+/// \param __A
+///    A 256-bit vector of [16 x int16].
+/// \param __B
+///    A 256-bit vector of [16 x fp16].
+/// \returns
+///    A 128-bit vector of [16 x hf8]. Elements correspond to the
+///    converted elements from \a __B using biases from \a __A.
 static __inline__ __m128i __DEFAULT_FN_ATTRS256
-_mm256_cvtbiassph_hf8(__m256i __A, __m256h __B) {
+_mm256_cvts_biasph_hf8(__m256i __A, __m256h __B) {
   return (__m128i)__builtin_ia32_vcvtbiasph2hf8s_256_mask(
       (__v32qi)__A, (__v16hf)__B, (__v16qi)(__m128i)_mm_undefined_si128(),
       (__mmask16)-1);
 }
 
-static __inline__ __m128i __DEFAULT_FN_ATTRS256 _mm256_mask_cvtbiassph_hf8(
+/// Convert 256-bit vector \a __B containing packed FP16 floating-point elements
+///    to FP8 E4M3 numbers, using conversion biases stored in lower 8 bits of each
+///    16-bit integer stored in \a __B. Results are saturated. Merging mask \a __U
+///    is used to determine if given element should be taken from \a __W instead.
+///
+/// \code{.operation}
+/// FOR i := 0 to 15
+/// 	IF __U[i]
+/// 		dst.hf8[i] := convert_fp16_to_hf8_with_bias_saturate(__A.int8[2 * i], __B.fp16[i])
+/// 	ELSE
+/// 		dst.hf8[i] := __W.hf8[i]
+/// 	FI
+/// ENDFOR
+///
+/// dst[MAX:128] := 0
+/// \endcode
+///
+/// \headerfile <immintrin.h>
+///
+/// This intrinsic corresponds to the \c VCVTBIASPH2HF8S instruction.
+///
+/// \param __W
+///    A 128-bit vector of [16 x hf8].
+/// \param __U
+///    A 16-bit merging mask.
+/// \param __A
+///    A 256-bit vector of [16 x int16].
+/// \param __B
+///    A 256-bit vector of [16 x fp16].
+/// \returns
+///    A 128-bit vector of [16 x hf8]. Elements correspond to the converted
+///    elements from \a __B, using biases from \a __A. If corresponding mask bit
+///    is not set, then element from \a __W is taken instead.
+static __inline__ __m128i __DEFAULT_FN_ATTRS256 _mm256_mask_cvts_biasph_hf8(
     __m128i __W, __mmask16 __U, __m256i __A, __m256h __B) {
   return (__m128i)__builtin_ia32_vcvtbiasph2hf8s_256_mask(
       (__v32qi)__A, (__v16hf)__B, (__v16qi)(__m128i)__W, (__mmask16)__U);
 }
 
+/// Convert 256-bit vector \a __B containing packed FP16 floating-point elements
+///    to FP8 E4M3 numbers, using conversion biases stored in lower 8 bits of each
+///    16-bit integer stored in \a __B. Results are saturated. Zeroing mask \a __U
+///    is used to determine if given element should be zeroed instead.
+///
+/// \code{.operation}
+/// FOR i := 0 to 15
+/// 	IF __U[i]
+///	 	dst.hf8[i] := convert_fp16_to_hf8_with_bias_saturate(__A.int8[2 * i], __B.fp16[i])
+///	 ELSE
+///	 	dst.hf8[i] := 0
+///	 FI
+/// ENDFOR
+///
+/// dst[MAX:128] := 0
+/// \endcode
+///
+/// \headerfile <immintrin.h>
+///
+/// This intrinsic corresponds to the \c VCVTBIASPH2HF8S instruction.
+///
+/// \param __U
+///    A 16-bit zeroing mask.
+/// \param __A
+///    A 256-bit vector of [16 x int16].
+/// \param __B
+///    A 256-bit vector of [16 x fp16].
+/// \returns
+///    A 128-bit vector of [16 x hf8]. Elements correspond to the converted
+///    elements from \a __B, using biases from \a __A. If corresponding mask bit
+///    is not set, then element is zeroed.
 static __inline__ __m128i __DEFAULT_FN_ATTRS256
-_mm256_maskz_cvtbiassph_hf8(__mmask16 __U, __m256i __A, __m256h __B) {
+_mm256_maskz_cvts_biasph_hf8(__mmask16 __U, __m256i __A, __m256h __B) {
   return (__m128i)__builtin_ia32_vcvtbiasph2hf8s_256_mask(
       (__v32qi)__A, (__v16hf)__B, (__v16qi)(__m128i)_mm_setzero_si128(),
       (__mmask16)__U);
 }
 
+/// Convert two 128-bit vectors, \a __A and \a __B, containing packed FP16
+///    floating-point elements to a 128-bit vector containing E5M2 FP8 elements.
+///
+/// \code{.operation}
+/// FOR i := 0 to 15 
+/// 	IF i < 8
+/// 		dst.bf8[i] := convert_fp16_to_bf8(__B.fp16[i])
+/// 	ELSE
+/// 		dst.bf8[i] := convert_fp16_to_bf8(__A.fp16[i - 8])
+/// 	FI
+/// ENDFOR
+///
+/// dst[MAX:128] := 0
+/// \endcode
+///
+/// \headerfile <immintrin.h>
+///
+/// This intrinsic corresponds to the \c VCVT2PH2BF8 instruction.
+///
+/// \param __A
+///    A 128-bit vector of [8 x fp16].
+/// \param __B
+///    A 128-bit vector of [8 x fp16].
+/// \returns
+///    A 128-bit vector of [16 x bf8]. Lower 8 elements correspond to the
+///    (converted) elements from \a __B; higher order elements correspond to the
+///    (converted) elements from \a __A.
 static __inline__ __m128i __DEFAULT_FN_ATTRS128 _mm_cvt2ph_bf8(__m128h __A,
-                                                               __m128h __B) {
-  return (__m128i)__builtin_ia32_vcvt2ph2bf8_128((__v8hf)(__A), (__v8hf)(__B));
+                                                                  __m128h __B) {
+  return (__m128i)__builtin_ia32_vcvt2ph2bf8_128((__v8hf)(__A),
+                                                   (__v8hf)(__B));
 }
 
+/// Convert two 128-bit vectors, \a __A and \a __B, containing packed FP16
+///    floating-point elements to a 128-bit vector containing E5M2 FP8 elements.
+///    Merging mask \a __U is used to determine if given element should be taken
+///    from \a __W instead.
+///
+/// \code{.operation}
+/// FOR i := 0 to 15
+/// 	IF __U[i]
+/// 		IF i < 8
+/// 			dst.bf8[i] := convert_fp16_to_bf8(__B.fp16[i])
+/// 		ELSE
+/// 			dst.bf8[i] := convert_fp16_to_bf8(__A.fp16[i - 8])
+/// 		FI
+/// 	ELSE
+/// 		dst.bf8[i] := __W.bf8[i]
+/// 	FI
+/// ENDFOR
+///
+/// dst[MAX:128] := 0
+/// \endcode
+///
+/// \headerfile <immintrin.h>
+///
+/// This intrinsic corresponds to the \c VCVT2PH2BF8 instruction.
+///
+/// \param __W
+///    A 128-bit vector of [16 x bf8].
+/// \param __U
+///    A 16-bit merging mask.
+/// \param __A
+///    A 128-bit vector of [8 x fp16].
+/// \param __B
+///    A 128-bit vector of [8 x fp16].
+/// \returns
+///    A 128-bit vector of [16 x bf8]. Lower 8 elements correspond to the
+///    (converted) elements from \a __B; higher order elements correspond to the
+///    (converted) elements from \a __A. If corresponding mask bit is not set, then
+///    element from \a __W is taken instead.
 static __inline__ __m128i __DEFAULT_FN_ATTRS128
 _mm_mask_cvt2ph_bf8(__m128i __W, __mmask16 __U, __m128h __A, __m128h __B) {
   return (__m128i)__builtin_ia32_selectb_128(
       (__mmask16)__U, (__v16qi)_mm_cvt2ph_bf8(__A, __B), (__v16qi)__W);
 }
 
+/// Convert two 128-bit vectors, \a __A and \a __B, containing packed FP16
+///    floating-point elements to a 128-bit vector containing E5M2 FP8 elements.
+///    Zeroing mask \a __U is used to determine if given element should be zeroed
+///    instead.
+///
+/// \code{.operation}
+/// FOR i := 0 to 15 
+/// 	IF __U[i]
+/// 		IF i < 8
+/// 			dst.bf8[i] := convert_fp16_to_bf8(__B.fp16[i])
+/// 		ELSE
+/// 			dst.bf8[i] := convert_fp16_to_bf8(__A.fp16[i - 8])
+/// 		FI
+/// 	ELSE
+/// 		dst.bf8[i] := 0
+/// 	FI
+/// ENDFOR
+///
+/// dst[MAX:128] := 0
+/// \endcode
+///
+/// \headerfile <immintrin.h>
+///
+/// This intrinsic corresponds to the \c VCVT2PH2BF8 instruction.
+///
+/// \param __U
+///    A 16-bit zeroing mask.
+/// \param __A
+///    A 128-bit vector of [8 x fp16].
+/// \param __B
+///    A 128-bit vector of [8 x fp16].
+/// \returns
+///    A 128-bit vector of [16 x bf8]. Lower 8 elements correspond to the
+///    (converted) elements from \a __B; higher order elements correspond to the
+///    (converted) elements from \a __A. If corresponding mask bit is not set, then
+///    zero is taken instead.
 static __inline__ __m128i __DEFAULT_FN_ATTRS128
 _mm_maskz_cvt2ph_bf8(__mmask16 __U, __m128h __A, __m128h __B) {
   return (__m128i)__builtin_ia32_selectb_128(
@@ -251,73 +1251,481 @@ _mm_maskz_cvt2ph_bf8(__mmask16 __U, __m128h __A, __m128h __B) {
       (__v16qi)(__m128i)_mm_setzero_si128());
 }
 
-static __inline__ __m256i __DEFAULT_FN_ATTRS256 _mm256_cvt2ph_bf8(__m256h __A,
-                                                                  __m256h __B) {
-  return (__m256i)__builtin_ia32_vcvt2ph2bf8_256((__v16hf)(__A),
-                                                 (__v16hf)(__B));
-}
-
+/// Convert two 256-bit vectors, \a __A and \a __B, containing packed FP16
+///    floating-point elements to a 256-bit vector containing E5M2 FP8 elements.
+///
+/// \code{.operation}
+/// FOR i := 0 to 31 
+/// 	IF i < 16 
+/// 		dst.bf8[i] := convert_fp16_to_bf8(__B.fp16[i])
+/// 	ELSE
+/// 		dst.bf8[i] := convert_fp16_to_bf8(__A.fp16[i - 16])
+/// 	FI
+/// ENDFOR
+///
+/// dst[MAX:256] := 0
+/// \endcode
+///
+/// \headerfile <immintrin.h>
+///
+/// This intrinsic corresponds to the \c VCVT2PH2BF8 instruction.
+///
+/// \param __A
+///    A 256-bit vector of [16 x fp16].
+/// \param __B
+///    A 256-bit vector of [16 x fp16].
+/// \returns
+///    A 256-bit vector of [32 x bf8]. Lower 16 elements correspond to the
+///    (converted) elements from \a __B; higher order elements correspond to the
+///    (converted) elements from \a __A.
 static __inline__ __m256i __DEFAULT_FN_ATTRS256
-_mm256_mask_cvt2ph_bf8(__m256i __W, __mmask32 __U, __m256h __A, __m256h __B) {
-  return (__m256i)__builtin_ia32_selectb_256(
-      (__mmask16)__U, (__v32qi)_mm256_cvt2ph_bf8(__A, __B), (__v32qi)__W);
+_mm256_cvt2ph_bf8(__m256h __A, __m256h __B) {
+  return (__m256i)__builtin_ia32_vcvt2ph2bf8_256((__v16hf)(__A),
+                                                   (__v16hf)(__B));
 }
 
+/// Convert two 256-bit vectors, \a __A and \a __B, containing packed FP16
+///    floating-point elements to a 256-bit vector containing E5M2 FP8 elements.
+///    Merging mask \a __U is used to determine if given element should be taken
+///    from \a __W instead.
+///
+/// \code{.operation}
+/// FOR i := 0 to 31 
+/// 	IF __U[i]
+/// 		IF i < 16 
+/// 			dst.bf8[i] := convert_fp16_to_bf8(__B.fp16[i])
+/// 		ELSE
+/// 			dst.bf8[i] := convert_fp16_to_bf8(__A.fp16[i - 16])
+/// 		FI
+/// 	ELSE
+/// 		dst.bf8[i] := __W.bf8[i]
+/// 	FI
+/// ENDFOR
+///
+/// dst[MAX:256] := 0
+/// \endcode
+///
+/// \headerfile <immintrin.h>
+///
+/// This intrinsic corresponds to the \c VCVT2PH2BF8 instruction.
+///
+/// \param __W
+///    A 256-bit vector of [32 x bf8].
+/// \param __U
+///    A 32-bit merging mask.
+/// \param __A
+///    A 256-bit vector of [16 x fp16].
+/// \param __B
+///    A 256-bit vector of [16 x fp16].
+/// \returns
+///    A 256-bit vector of [32 x bf8]. Lower 16 elements correspond to the
+///    (converted) elements from \a __B; higher order elements correspond to the
+///    (converted) elements from \a __A. If corresponding mask bit is not set, then
+///    element from \a __W is taken instead.
+static __inline__ __m256i __DEFAULT_FN_ATTRS256 _mm256_mask_cvt2ph_bf8(
+    __m256i __W, __mmask32 __U, __m256h __A, __m256h __B) {
+  return (__m256i)__builtin_ia32_selectb_256(
+      (__mmask32)__U, (__v32qi)_mm256_cvt2ph_bf8(__A, __B), (__v32qi)__W);
+}
+
+/// Convert two 256-bit vectors, \a __A and \a __B, containing packed FP16
+///    floating-point elements to a 256-bit vector containing E5M2 FP8 elements.
+///    Zeroing mask \a __U is used to determine if given element should be zeroed
+///    instead.
+///
+/// \code{.operation}
+/// FOR i := 0 to 31 
+/// 	IF __U[i]
+/// 		IF i < 16 
+/// 			dst.bf8[i] := convert_fp16_to_bf8(__B.fp16[i])
+/// 		ELSE
+/// 			dst.bf8[i] := convert_fp16_to_bf8(__A.fp16[i - 16])
+/// 		FI
+/// 	ELSE
+/// 		dst.bf8[i] := 0
+/// 	FI
+/// ENDFOR
+///
+/// dst[MAX:256] := 0
+/// \endcode
+///
+/// \headerfile <immintrin.h>
+///
+/// This intrinsic corresponds to the \c VCVT2PH2BF8 instruction.
+///
+/// \param __U
+///    A 32-bit zeroing mask.
+/// \param __A
+///    A 256-bit vector of [16 x fp16].
+/// \param __B
+///    A 256-bit vector of [16 x fp16].
+/// \returns
+///    A 256-bit vector of [32 x bf8]. Lower 16 elements correspond to the
+///    (converted) elements from \a __B; higher order elements correspond to the
+///    (converted) elements from \a __A. If corresponding mask bit is not set,
+///    zero is taken instead.
 static __inline__ __m256i __DEFAULT_FN_ATTRS256
 _mm256_maskz_cvt2ph_bf8(__mmask32 __U, __m256h __A, __m256h __B) {
   return (__m256i)__builtin_ia32_selectb_256(
-      (__mmask16)__U, (__v32qi)_mm256_cvt2ph_bf8(__A, __B),
+      (__mmask32)__U, (__v32qi)_mm256_cvt2ph_bf8(__A, __B),
       (__v32qi)(__m256i)_mm256_setzero_si256());
 }
 
-static __inline__ __m128i __DEFAULT_FN_ATTRS128 _mm_cvts2ph_bf8(__m128h __A,
-                                                                __m128h __B) {
-  return (__m128i)__builtin_ia32_vcvt2ph2bf8s_128((__v8hf)(__A), (__v8hf)(__B));
+/// Convert two 128-bit vectors, \a __A and \a __B, containing packed FP16
+///    floating-point elements to a 128-bit vector containing E5M2 FP8 elements.
+///    Resulting elements are saturated in case of overflow.
+///
+/// \code{.operation}
+/// FOR i := 0 to 15 
+/// 	IF i < 8
+/// 		dst.bf8[i] := convert_fp16_to_bf8_saturate(__B.fp16[i])
+/// 	ELSE
+/// 		dst.bf8[i] := convert_fp16_to_bf8_saturate(__A.fp16[i - 8])
+/// 	FI
+/// ENDFOR
+///
+/// dst[MAX:128] := 0
+/// \endcode
+///
+/// \headerfile <immintrin.h>
+///
+/// This intrinsic corresponds to the \c VCVT2PH2BF8S instruction.
+///
+/// \param __A
+///    A 128-bit vector of [8 x fp16].
+/// \param __B
+///    A 128-bit vector of [8 x fp16].
+/// \returns
+///    A 128-bit vector of [16 x bf8]. Lower 8 elements correspond to the
+///    (converted) elements from \a __B; higher order elements correspond to the
+///    (converted) elements from \a __A.
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
+_mm_cvts_2ph_bf8(__m128h __A, __m128h __B) {
+  return (__m128i)__builtin_ia32_vcvt2ph2bf8s_128((__v8hf)(__A),
+                                                    (__v8hf)(__B));
 }
 
+/// Convert two 128-bit vectors, \a __A and \a __B, containing packed FP16
+///    floating-point elements to a 128-bit vector containing E5M2 FP8 elements.
+///    Merging mask \a __U is used to determine if given element should be taken
+///    from \a __W instead. Resulting elements are saturated in case of overflow.
+///
+/// \code{.operation}
+/// FOR i := 0 to 15 
+/// 	IF __U[i]
+/// 		IF i < 8
+/// 			dst.bf8[i] := convert_fp16_to_bf8_saturate(__B.fp16[i])
+/// 		ELSE
+/// 			dst.bf8[i] := convert_fp16_to_bf8_saturate(__A.fp16[i - 8])
+/// 		FI
+/// 	ELSE
+/// 		dst.bf8[i] := __W.bf8[i]
+/// 	FI
+/// ENDFOR
+///
+/// dst[MAX:128] := 0
+/// \endcode
+///
+/// \headerfile <immintrin.h>
+///
+/// This intrinsic corresponds to the \c VCVT2PH2BF8S instruction.
+///
+/// \param __W
+///    A 128-bit vector of [16 x bf8].
+/// \param __U
+///    A 16-bit merging mask.
+/// \param __A
+///    A 128-bit vector of [8 x fp16].
+/// \param __B
+///    A 128-bit vector of [8 x fp16].
+/// \returns
+///    A 128-bit vector of [16 x bf8]. Lower 8 elements correspond to the
+///    (converted) elements from \a __B; higher order elements correspond to the
+///    (converted) elements from \a __A. If corresponding mask bit is not set, then
+///    element from \a __W is taken instead.
 static __inline__ __m128i __DEFAULT_FN_ATTRS128
-_mm_mask_cvts2ph_bf8(__m128i __W, __mmask16 __U, __m128h __A, __m128h __B) {
+_mm_mask_cvts_2ph_bf8(__m128i __W, __mmask16 __U, __m128h __A, __m128h __B) {
   return (__m128i)__builtin_ia32_selectb_128(
-      (__mmask16)__U, (__v16qi)_mm_cvts2ph_bf8(__A, __B), (__v16qi)__W);
+      (__mmask16)__U, (__v16qi)_mm_cvts_2ph_bf8(__A, __B), (__v16qi)__W);
 }
 
+/// Convert two 128-bit vectors, \a __A and \a __B, containing packed FP16
+///    floating-point elements to a 128-bit vector containing E5M2 FP8 elements.
+///    Zeroing mask \a __U is used to determine if given element should be zeroed
+///    instead. Resulting elements are saturated in case of overflow.
+///
+/// \code{.operation}
+/// FOR i := 0 to 15 
+/// 	IF __U[i]
+/// 		IF i < 8
+/// 			dst.bf8[i] := convert_fp16_to_bf8_saturate(__B.fp16[i])
+/// 		ELSE
+/// 			dst.bf8[i] := convert_fp16_to_bf8_saturate(__A.fp16[i - 8])
+/// 		FI
+/// 	ELSE
+/// 		dst.bf8[i] := 0
+/// 	FI
+/// ENDFOR
+///
+/// dst[MAX:128] := 0
+/// \endcode
+///
+/// \headerfile <immintrin.h>
+///
+/// This intrinsic corresponds to the \c VCVT2PH2BF8S instruction.
+///
+/// \param __U
+///    A 16-bit zeroing mask.
+/// \param __A
+///    A 128-bit vector of [8 x fp16].
+/// \param __B
+///    A 128-bit vector of [8 x fp16].
+/// \returns
+///    A 128-bit vector of [16 x bf8]. Lower 8 elements correspond to the
+///    (converted) elements from \a __B; higher order elements correspond to the
+///    (converted) elements from \a __A. If corresponding mask bit is not set, then
+///    zero is taken instead.
 static __inline__ __m128i __DEFAULT_FN_ATTRS128
-_mm_maskz_cvts2ph_bf8(__mmask16 __U, __m128h __A, __m128h __B) {
+_mm_maskz_cvts_2ph_bf8(__mmask16 __U, __m128h __A, __m128h __B) {
   return (__m128i)__builtin_ia32_selectb_128(
-      (__mmask16)__U, (__v16qi)_mm_cvts2ph_bf8(__A, __B),
+      (__mmask16)__U, (__v16qi)_mm_cvts_2ph_bf8(__A, __B),
       (__v16qi)(__m128i)_mm_setzero_si128());
 }
 
+/// Convert two 256-bit vectors, \a __A and \a __B, containing packed FP16
+///    floating-point elements to a 256-bit vector containing E5M2 FP8 elements.
+///    Resulting elements are saturated in case of overflow.
+///
+/// \code{.operation}
+/// FOR i := 0 to 31 
+/// 	IF i < 16 
+/// 		dst.bf8[i] := convert_fp16_to_bf8_saturate(__B.fp16[i])
+/// 	ELSE
+/// 		dst.bf8[i] := convert_fp16_to_bf8_saturate(__A.fp16[i - 16])
+/// 	FI
+/// ENDFOR
+///
+/// dst[MAX:256] := 0
+/// \endcode
+///
+/// \headerfile <immintrin.h>
+///
+/// This intrinsic corresponds to the \c VCVT2PH2BF8S instruction.
+///
+/// \param __A
+///    A 256-bit vector of [16 x fp16].
+/// \param __B
+///    A 256-bit vector of [16 x fp16].
+/// \returns
+///    A 256-bit vector of [32 x bf8]. Lower 16 elements correspond to the
+///    (converted) elements from \a __B; higher order elements correspond to the
+///    (converted) elements from \a __A.
 static __inline__ __m256i __DEFAULT_FN_ATTRS256
-_mm256_cvts2ph_bf8(__m256h __A, __m256h __B) {
+_mm256_cvts_2ph_bf8(__m256h __A, __m256h __B) {
   return (__m256i)__builtin_ia32_vcvt2ph2bf8s_256((__v16hf)(__A),
-                                                  (__v16hf)(__B));
+                                                    (__v16hf)(__B));
 }
 
-static __inline__ __m256i __DEFAULT_FN_ATTRS256
-_mm256_mask_cvts2ph_bf8(__m256i __W, __mmask32 __U, __m256h __A, __m256h __B) {
+/// Convert two 256-bit vectors, \a __A and \a __B, containing packed FP16
+///    floating-point elements to a 256-bit vector containing E5M2 FP8 elements.
+///    Merging mask \a __U is used to determine if given element should be taken
+///    from \a __W instead. Resulting elements are saturated in case of overflow.
+///
+/// \code{.operation}
+/// FOR i := 0 to 31 
+/// 	IF __U[i]
+/// 		IF i < 16 
+/// 			dst.bf8[i] := convert_fp16_to_bf8_saturate(__B.fp16[i])
+/// 		ELSE
+/// 			dst.bf8[i] := convert_fp16_to_bf8_saturate(__A.fp16[i - 16])
+/// 		FI
+/// 	ELSE
+/// 		dst.bf8[i] := __W.bf8[i]
+/// 	FI
+/// ENDFOR
+///
+/// dst[MAX:256] := 0
+/// \endcode
+///
+/// \headerfile <immintrin.h>
+///
+/// This intrinsic corresponds to the \c VCVT2PH2BF8S instruction.
+///
+/// \param __W
+///    A 256-bit vector of [32 x bf8].
+/// \param __U
+///    A 32-bit merging mask.
+/// \param __A
+///    A 256-bit vector of [16 x fp16].
+/// \param __B
+///    A 256-bit vector of [16 x fp16].
+/// \returns
+///    A 256-bit vector of [32 x bf8]. Lower 16 elements correspond to the
+///    (converted) elements from \a __B; higher order elements correspond to the
+///    (converted) elements from \a __A. If corresponding mask bit is not set, then
+///    element from \a __W is taken instead.
+static __inline__ __m256i __DEFAULT_FN_ATTRS256 _mm256_mask_cvts_2ph_bf8(
+    __m256i __W, __mmask32 __U, __m256h __A, __m256h __B) {
   return (__m256i)__builtin_ia32_selectb_256(
-      (__mmask16)__U, (__v32qi)_mm256_cvts2ph_bf8(__A, __B), (__v32qi)__W);
+      (__mmask32)__U, (__v32qi)_mm256_cvts_2ph_bf8(__A, __B), (__v32qi)__W);
 }
 
+/// Convert two 256-bit vectors, \a __A and \a __B, containing packed FP16
+///    floating-point elements to a 256-bit vector containing E5M2 FP8 elements.
+///    Zeroing mask \a __U is used to determine if given element should be zeroed
+///    instead. Resulting elements are saturated in case of overflow.
+///
+/// \code{.operation}
+/// FOR i := 0 to 31 
+/// 	IF __U[i]
+/// 		IF i < 16 
+/// 			dst.bf8[i] := convert_fp16_to_bf8_saturate(__B.fp16[i])
+/// 		ELSE
+/// 			dst.bf8[i] := convert_fp16_to_bf8_saturate(__A.fp16[i - 16])
+/// 		FI
+/// 	ELSE
+/// 		dst.bf8[i] := 0
+/// 	FI
+/// ENDFOR
+///
+/// dst[MAX:256] := 0
+/// \endcode
+///
+/// \headerfile <immintrin.h>
+///
+/// This intrinsic corresponds to the \c VCVT2PH2BF8S instruction.
+///
+/// \param __U
+///    A 32-bit zeroing mask.
+/// \param __A
+///    A 256-bit vector of [16 x fp16].
+/// \param __B
+///    A 256-bit vector of [16 x fp16].
+/// \returns
+///    A 256-bit vector of [32 x bf8]. Lower 16 elements correspond to the
+///    (converted) elements from \a __B; higher order elements correspond to the
+///    (converted) elements from \a __A. If corresponding mask bit is not set,
+///    zero is taken instead.
 static __inline__ __m256i __DEFAULT_FN_ATTRS256
-_mm256_maskz_cvts2ph_bf8(__mmask32 __U, __m256h __A, __m256h __B) {
+_mm256_maskz_cvts_2ph_bf8(__mmask32 __U, __m256h __A, __m256h __B) {
   return (__m256i)__builtin_ia32_selectb_256(
-      (__mmask16)__U, (__v32qi)_mm256_cvts2ph_bf8(__A, __B),
+      (__mmask32)__U, (__v32qi)_mm256_cvts_2ph_bf8(__A, __B),
       (__v32qi)(__m256i)_mm256_setzero_si256());
 }
 
+/// Convert two 128-bit vectors, \a __A and \a __B, containing packed FP16
+///    floating-point elements to a 128-bit vector containing E4M3 FP8 elements.
+///
+/// \code{.operation}
+/// FOR i := 0 to 15 
+/// 	IF i < 8
+/// 		dst.hf8[i] := convert_fp16_to_hf8(__B.fp16[i])
+/// 	ELSE
+/// 		dst.hf8[i] := convert_fp16_to_hf8(__A.fp16[i - 8])
+/// 	FI
+/// ENDFOR
+///
+/// dst[MAX:128] := 0
+/// \endcode
+///
+/// \headerfile <immintrin.h>
+///
+/// This intrinsic corresponds to the \c VCVT2PH2HF8 instruction.
+///
+/// \param __A
+///    A 128-bit vector of [8 x fp16].
+/// \param __B
+///    A 128-bit vector of [8 x fp16].
+/// \returns
+///    A 128-bit vector of [16 x hf8]. Lower 8 elements correspond to the
+///    (converted) elements from \a __B; higher order elements correspond to the
+///    (converted) elements from \a __A.
 static __inline__ __m128i __DEFAULT_FN_ATTRS128 _mm_cvt2ph_hf8(__m128h __A,
-                                                               __m128h __B) {
-  return (__m128i)__builtin_ia32_vcvt2ph2hf8_128((__v8hf)(__A), (__v8hf)(__B));
+                                                                  __m128h __B) {
+  return (__m128i)__builtin_ia32_vcvt2ph2hf8_128((__v8hf)(__A),
+                                                   (__v8hf)(__B));
 }
 
+/// Convert two 128-bit vectors, \a __A and \a __B, containing packed FP16
+///    floating-point elements to a 128-bit vector containing E4M3 FP8 elements.
+///    Merging mask \a __U is used to determine if given element should be taken
+///    from \a __W instead.
+///
+/// \code{.operation}
+/// FOR i := 0 to 15 
+/// 	IF __U[i]
+/// 		IF i < 8
+/// 			dst.hf8[i] := convert_fp16_to_hf8(__B.fp16[i])
+/// 		ELSE
+/// 			dst.hf8[i] := convert_fp16_to_hf8(__A.fp16[i - 8])
+/// 		FI
+/// 	ELSE
+/// 		dst.hf8[i] := __W.hf8[i]
+/// 	FI
+/// ENDFOR
+///
+/// dst[MAX:128] := 0
+/// \endcode
+///
+/// \headerfile <immintrin.h>
+///
+/// This intrinsic corresponds to the \c VCVT2PH2HF8 instruction.
+///
+/// \param __W
+///    A 128-bit vector of [16 x hf8].
+/// \param __U
+///    A 16-bit merging mask.
+/// \param __A
+///    A 128-bit vector of [8 x fp16].
+/// \param __B
+///    A 128-bit vector of [8 x fp16].
+/// \returns
+///    A 128-bit vector of [16 x hf8]. Lower 8 elements correspond to the
+///    (converted) elements from \a __B; higher order elements correspond to the
+///    (converted) elements from \a __A. If corresponding mask bit is not set, then
+///    element from \a __W is taken instead.
 static __inline__ __m128i __DEFAULT_FN_ATTRS128
 _mm_mask_cvt2ph_hf8(__m128i __W, __mmask16 __U, __m128h __A, __m128h __B) {
   return (__m128i)__builtin_ia32_selectb_128(
       (__mmask16)__U, (__v16qi)_mm_cvt2ph_hf8(__A, __B), (__v16qi)__W);
 }
 
+/// Convert two 128-bit vectors, \a __A and \a __B, containing packed FP16
+///    floating-point elements to a 128-bit vector containing E4M3 FP8 elements.
+///    Zeroing mask \a __U is used to determine if given element should be zeroed
+///    instead.
+///
+/// \code{.operation}
+/// FOR i := 0 to 15 
+/// 	IF __U[i]
+/// 		IF i < 8
+/// 			dst.hf8[i] := convert_fp16_to_hf8(__B.fp16[i])
+/// 		ELSE
+/// 			dst.hf8[i] := convert_fp16_to_hf8(__A.fp16[i - 8])
+/// 		FI
+/// 	ELSE
+/// 		dst.hf8[i] := 0
+/// 	FI
+/// ENDFOR
+///
+/// dst[MAX:128] := 0
+/// \endcode
+///
+/// \headerfile <immintrin.h>
+///
+/// This intrinsic corresponds to the \c VCVT2PH2HF8 instruction.
+///
+/// \param __U
+///    A 16-bit zeroing mask.
+/// \param __A
+///    A 128-bit vector of [8 x fp16].
+/// \param __B
+///    A 128-bit vector of [8 x fp16].
+/// \returns
+///    A 128-bit vector of [16 x hf8]. Lower 8 elements correspond to the
+///    (converted) elements from \a __B; higher order elements correspond to the
+///    (converted) elements from \a __A. If corresponding mask bit is not set, then
+///    zero is taken instead.
 static __inline__ __m128i __DEFAULT_FN_ATTRS128
 _mm_maskz_cvt2ph_hf8(__mmask16 __U, __m128h __A, __m128h __B) {
   return (__m128i)__builtin_ia32_selectb_128(
@@ -325,264 +1733,1523 @@ _mm_maskz_cvt2ph_hf8(__mmask16 __U, __m128h __A, __m128h __B) {
       (__v16qi)(__m128i)_mm_setzero_si128());
 }
 
-static __inline__ __m256i __DEFAULT_FN_ATTRS256 _mm256_cvt2ph_hf8(__m256h __A,
-                                                                  __m256h __B) {
-  return (__m256i)__builtin_ia32_vcvt2ph2hf8_256((__v16hf)(__A),
-                                                 (__v16hf)(__B));
-}
-
+/// Convert two 256-bit vectors, \a __A and \a __B, containing packed FP16
+///    floating-point elements to a 256-bit vector containing E4M3 FP8 elements.
+///
+/// \code{.operation}
+/// FOR i := 0 to 31 
+/// 	IF i < 16 
+/// 		dst.hf8[i] := convert_fp16_to_hf8(__B.fp16[i])
+/// 	ELSE
+/// 		dst.hf8[i] := convert_fp16_to_hf8(__A.fp16[i - 16])
+/// 	FI
+/// ENDFOR
+///
+/// dst[MAX:256] := 0
+/// \endcode
+///
+/// \headerfile <immintrin.h>
+///
+/// This intrinsic corresponds to the \c VCVT2PH2HF8 instruction.
+///
+/// \param __A
+///    A 256-bit vector of [16 x fp16].
+/// \param __B
+///    A 256-bit vector of [16 x fp16].
+/// \returns
+///    A 256-bit vector of [32 x hf8]. Lower 16 elements correspond to the
+///    (converted) elements from \a __B; higher order elements correspond to the
+///    (converted) elements from \a __A.
 static __inline__ __m256i __DEFAULT_FN_ATTRS256
-_mm256_mask_cvt2ph_hf8(__m256i __W, __mmask32 __U, __m256h __A, __m256h __B) {
-  return (__m256i)__builtin_ia32_selectb_256(
-      (__mmask16)__U, (__v32qi)_mm256_cvt2ph_hf8(__A, __B), (__v32qi)__W);
+_mm256_cvt2ph_hf8(__m256h __A, __m256h __B) {
+  return (__m256i)__builtin_ia32_vcvt2ph2hf8_256((__v16hf)(__A),
+                                                   (__v16hf)(__B));
 }
 
+/// Convert two 256-bit vectors, \a __A and \a __B, containing packed FP16
+///    floating-point elements to a 256-bit vector containing E4M3 FP8 elements.
+///    Merging mask \a __U is used to determine if given element should be taken
+///    from \a __W instead.
+///
+/// \code{.operation}
+/// FOR i := 0 to 31 
+/// 	IF __U[i]
+/// 		IF i < 16 
+/// 			dst.hf8[i] := convert_fp16_to_hf8(__B.fp16[i])
+/// 		ELSE
+/// 			dst.hf8[i] := convert_fp16_to_hf8(__A.fp16[i - 16])
+/// 		FI
+/// 	ELSE
+/// 		dst.hf8[i] := __W.hf8[i]
+/// 	FI
+/// ENDFOR
+///
+/// dst[MAX:256] := 0
+/// \endcode
+///
+/// \headerfile <immintrin.h>
+///
+/// This intrinsic corresponds to the \c VCVT2PH2HF8 instruction.
+///
+/// \param __W
+///    A 256-bit vector of [32 x hf8].
+/// \param __U
+///    A 32-bit merging mask.
+/// \param __A
+///    A 256-bit vector of [16 x fp16].
+/// \param __B
+///    A 256-bit vector of [16 x fp16].
+/// \returns
+///    A 256-bit vector of [32 x hf8]. Lower 16 elements correspond to the
+///    (converted) elements from \a __B; higher order elements correspond to the
+///    (converted) elements from \a __A. If corresponding mask bit is not set, then
+///    element from \a __W is taken instead.
+static __inline__ __m256i __DEFAULT_FN_ATTRS256 _mm256_mask_cvt2ph_hf8(
+    __m256i __W, __mmask32 __U, __m256h __A, __m256h __B) {
+  return (__m256i)__builtin_ia32_selectb_256(
+      (__mmask32)__U, (__v32qi)_mm256_cvt2ph_hf8(__A, __B), (__v32qi)__W);
+}
+
+/// Convert two 256-bit vectors, \a __A and \a __B, containing packed FP16
+///    floating-point elements to a 256-bit vector containing E4M3 FP8 elements.
+///    Zeroing mask \a __U is used to determine if given element should be zeroed
+///    instead.
+///
+/// \code{.operation}
+/// FOR i := 0 to 31 
+/// 	IF __U[i]
+/// 		IF i < 16 
+/// 			dst.hf8[i] := convert_fp16_to_hf8(__B.fp16[i])
+/// 		ELSE
+/// 			dst.hf8[i] := convert_fp16_to_hf8(__A.fp16[i - 16])
+/// 		FI
+/// 	ELSE
+/// 		dst.hf8[i] := 0
+/// 	FI
+/// ENDFOR
+///
+/// dst[MAX:256] := 0
+/// \endcode
+///
+/// \headerfile <immintrin.h>
+///
+/// This intrinsic corresponds to the \c VCVT2PH2HF8 instruction.
+///
+/// \param __U
+///    A 32-bit zeroing mask.
+/// \param __A
+///    A 256-bit vector of [16 x fp16].
+/// \param __B
+///    A 256-bit vector of [16 x fp16].
+/// \returns
+///    A 256-bit vector of [32 x hf8]. Lower 16 elements correspond to the
+///    (converted) elements from \a __B; higher order elements correspond to the
+///    (converted) elements from \a __A. If corresponding mask bit is not set,
+///    zero is taken instead.
 static __inline__ __m256i __DEFAULT_FN_ATTRS256
 _mm256_maskz_cvt2ph_hf8(__mmask32 __U, __m256h __A, __m256h __B) {
   return (__m256i)__builtin_ia32_selectb_256(
-      (__mmask16)__U, (__v32qi)_mm256_cvt2ph_hf8(__A, __B),
+      (__mmask32)__U, (__v32qi)_mm256_cvt2ph_hf8(__A, __B),
       (__v32qi)(__m256i)_mm256_setzero_si256());
 }
 
-static __inline__ __m128i __DEFAULT_FN_ATTRS128 _mm_cvts2ph_hf8(__m128h __A,
-                                                                __m128h __B) {
-  return (__m128i)__builtin_ia32_vcvt2ph2hf8s_128((__v8hf)(__A), (__v8hf)(__B));
+/// Convert two 128-bit vectors, \a __A and \a __B, containing packed FP16
+///    floating-point elements to a 128-bit vector containing E4M3 FP8 elements.
+///    Resulting elements are saturated in case of overflow.
+///
+/// \code{.operation}
+/// FOR i := 0 to 15 
+/// 	IF i < 8
+/// 		dst.hf8[i] := convert_fp16_to_hf8_saturate(__B.fp16[i])
+/// 	ELSE
+/// 		dst.hf8[i] := convert_fp16_to_hf8_saturate(__A.fp16[i - 8])
+/// 	FI
+/// ENDFOR
+///
+/// dst[MAX:128] := 0
+/// \endcode
+///
+/// \headerfile <immintrin.h>
+///
+/// This intrinsic corresponds to the \c VCVT2PH2HF8S instruction.
+///
+/// \param __A
+///    A 128-bit vector of [8 x fp16].
+/// \param __B
+///    A 128-bit vector of [8 x fp16].
+/// \returns
+///    A 128-bit vector of [16 x hf8]. Lower 8 elements correspond to the
+///    (converted) elements from \a __B; higher order elements correspond to the
+///    (converted) elements from \a __A.
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
+_mm_cvts_2ph_hf8(__m128h __A, __m128h __B) {
+  return (__m128i)__builtin_ia32_vcvt2ph2hf8s_128((__v8hf)(__A),
+                                                    (__v8hf)(__B));
 }
 
+/// Convert two 128-bit vectors, \a __A and \a __B, containing packed FP16
+///    floating-point elements to a 128-bit vector containing E4M3 FP8 elements.
+///    Merging mask \a __U is used to determine if given element should be taken
+///    from \a __W instead. Resulting elements are saturated in case of overflow.
+///
+/// \code{.operation}
+/// FOR i := 0 to 15 
+/// 	IF __U[i]
+/// 		IF i < 8
+/// 			dst.hf8[i] := convert_fp16_to_hf8_saturate(__B.fp16[i])
+/// 		ELSE
+/// 			dst.hf8[i] := convert_fp16_to_hf8_saturate(__A.fp16[i - 8])
+/// 		FI
+/// 	ELSE
+/// 		dst.hf8[i] := __W.hf8[i]
+/// 	FI
+/// ENDFOR
+///
+/// dst[MAX:128] := 0
+/// \endcode
+///
+/// \headerfile <immintrin.h>
+///
+/// This intrinsic corresponds to the \c VCVT2PH2HF8S instruction.
+///
+/// \param __W
+///    A 128-bit vector of [16 x hf8].
+/// \param __U
+///    A 16-bit merging mask.
+/// \param __A
+///    A 128-bit vector of [8 x fp16].
+/// \param __B
+///    A 128-bit vector of [8 x fp16].
+/// \returns
+///    A 128-bit vector of [16 x hf8]. Lower 8 elements correspond to the
+///    (converted) elements from \a __B; higher order elements correspond to the
+///    (converted) elements from \a __A. If corresponding mask bit is not set, then
+///    element from \a __W is taken instead.
 static __inline__ __m128i __DEFAULT_FN_ATTRS128
-_mm_mask_cvts2ph_hf8(__m128i __W, __mmask16 __U, __m128h __A, __m128h __B) {
+_mm_mask_cvts_2ph_hf8(__m128i __W, __mmask16 __U, __m128h __A, __m128h __B) {
   return (__m128i)__builtin_ia32_selectb_128(
-      (__mmask16)__U, (__v16qi)_mm_cvts2ph_hf8(__A, __B), (__v16qi)__W);
+      (__mmask16)__U, (__v16qi)_mm_cvts_2ph_hf8(__A, __B), (__v16qi)__W);
 }
 
+/// Convert two 128-bit vectors, \a __A and \a __B, containing packed FP16
+///    floating-point elements to a 128-bit vector containing E4M3 FP8 elements.
+///    Zeroing mask \a __U is used to determine if given element should be zeroed
+///    instead. Resulting elements are saturated in case of overflow.
+///
+/// \code{.operation}
+/// FOR i := 0 to 15 
+/// 	IF __U[i]
+/// 		IF i < 8
+/// 			dst.hf8[i] := convert_fp16_to_hf8_saturate(__B.fp16[i])
+/// 		ELSE
+/// 			dst.hf8[i] := convert_fp16_to_hf8_saturate(__A.fp16[i - 8])
+/// 		FI
+/// 	ELSE
+/// 		dst.hf8[i] := 0
+/// 	FI
+/// ENDFOR
+///
+/// dst[MAX:128] := 0
+/// \endcode
+///
+/// \headerfile <immintrin.h>
+///
+/// This intrinsic corresponds to the \c VCVT2PH2HF8S instruction.
+///
+/// \param __U
+///    A 16-bit zeroing mask.
+/// \param __A
+///    A 128-bit vector of [8 x fp16].
+/// \param __B
+///    A 128-bit vector of [8 x fp16].
+/// \returns
+///    A 128-bit vector of [16 x hf8]. Lower 8 elements correspond to the
+///    (converted) elements from \a __B; higher order elements correspond to the
+///    (converted) elements from \a __A. If corresponding mask bit is not set, then
+///    zero is taken instead.
 static __inline__ __m128i __DEFAULT_FN_ATTRS128
-_mm_maskz_cvts2ph_hf8(__mmask16 __U, __m128h __A, __m128h __B) {
+_mm_maskz_cvts_2ph_hf8(__mmask16 __U, __m128h __A, __m128h __B) {
   return (__m128i)__builtin_ia32_selectb_128(
-      (__mmask16)__U, (__v16qi)_mm_cvts2ph_hf8(__A, __B),
+      (__mmask16)__U, (__v16qi)_mm_cvts_2ph_hf8(__A, __B),
       (__v16qi)(__m128i)_mm_setzero_si128());
 }
 
+/// Convert two 256-bit vectors, \a __A and \a __B, containing packed FP16
+///    floating-point elements to a 256-bit vector containing E4M3 FP8 elements.
+///    Resulting elements are saturated in case of overflow.
+///
+/// \code{.operation}
+/// FOR i := 0 to 31 
+/// 	IF i < 16 
+/// 		dst.hf8[i] := convert_fp16_to_hf8_saturate(__B.fp16[i])
+/// 	ELSE
+/// 		dst.hf8[i] := convert_fp16_to_hf8_saturate(__A.fp16[i - 16])
+/// 	FI
+/// ENDFOR
+///
+/// dst[MAX:256] := 0
+/// \endcode
+///
+/// \headerfile <immintrin.h>
+///
+/// This intrinsic corresponds to the \c VCVT2PH2HF8S instruction.
+///
+/// \param __A
+///    A 256-bit vector of [16 x fp16].
+/// \param __B
+///    A 256-bit vector of [16 x fp16].
+/// \returns
+///    A 256-bit vector of [32 x hf8]. Lower 16 elements correspond to the
+///    (converted) elements from \a __B; higher order elements correspond to the
+///    (converted) elements from \a __A.
 static __inline__ __m256i __DEFAULT_FN_ATTRS256
-_mm256_cvts2ph_hf8(__m256h __A, __m256h __B) {
+_mm256_cvts_2ph_hf8(__m256h __A, __m256h __B) {
   return (__m256i)__builtin_ia32_vcvt2ph2hf8s_256((__v16hf)(__A),
-                                                  (__v16hf)(__B));
+                                                    (__v16hf)(__B));
 }
 
-static __inline__ __m256i __DEFAULT_FN_ATTRS256
-_mm256_mask_cvts2ph_hf8(__m256i __W, __mmask32 __U, __m256h __A, __m256h __B) {
+/// Convert two 256-bit vectors, \a __A and \a __B, containing packed FP16
+///    floating-point elements to a 256-bit vector containing E4M3 FP8 elements.
+///    Merging mask \a __U is used to determine if given element should be taken
+///    from \a __W instead. Resulting elements are saturated in case of overflow.
+///
+/// \code{.operation}
+/// FOR i := 0 to 31 
+/// 	IF __U[i]
+/// 		IF i < 16 
+/// 			dst.hf8[i] := convert_fp16_to_hf8_saturate(__B.fp16[i])
+/// 		ELSE
+/// 			dst.hf8[i] := convert_fp16_to_hf8_saturate(__A.fp16[i - 16])
+/// 		FI
+/// 	ELSE
+/// 		dst.hf8[i] := __W.hf8[i]
+/// 	FI
+/// ENDFOR
+///
+/// dst[MAX:256] := 0
+/// \endcode
+///
+/// \headerfile <immintrin.h>
+///
+/// This intrinsic corresponds to the \c VCVT2PH2HF8S instruction.
+///
+/// \param __W
+///    A 256-bit vector of [32 x hf8].
+/// \param __U
+///    A 32-bit merging mask.
+/// \param __A
+///    A 256-bit vector of [16 x fp16].
+/// \param __B
+///    A 256-bit vector of [16 x fp16].
+/// \returns
+///    A 256-bit vector of [32 x hf8]. Lower 16 elements correspond to the
+///    (converted) elements from \a __B; higher order elements correspond to the
+///    (converted) elements from \a __A. If corresponding mask bit is not set, then
+///    element from \a __W is taken instead.
+static __inline__ __m256i __DEFAULT_FN_ATTRS256 _mm256_mask_cvts_2ph_hf8(
+    __m256i __W, __mmask32 __U, __m256h __A, __m256h __B) {
   return (__m256i)__builtin_ia32_selectb_256(
-      (__mmask16)__U, (__v32qi)_mm256_cvts2ph_hf8(__A, __B), (__v32qi)__W);
+      (__mmask32)__U, (__v32qi)_mm256_cvts_2ph_hf8(__A, __B), (__v32qi)__W);
 }
 
+/// Convert two 256-bit vectors, \a __A and \a __B, containing packed FP16
+///    floating-point elements to a 256-bit vector containing E4M3 FP8 elements.
+///    Zeroing mask \a __U is used to determine if given element should be zeroed
+///    instead. Resulting elements are saturated in case of overflow.
+///
+/// \code{.operation}
+/// FOR i := 0 to 31 
+/// 	IF __U[i]
+/// 		IF i < 16 
+/// 			dst.hf8[i] := convert_fp16_to_hf8_saturate(__B.fp16[i])
+/// 		ELSE
+/// 			dst.hf8[i] := convert_fp16_to_hf8_saturate(__A.fp16[i - 16])
+/// 		FI
+/// 	ELSE
+/// 		dst.hf8[i] := 0
+/// 	FI
+/// ENDFOR
+///
+/// dst[MAX:256] := 0
+/// \endcode
+///
+/// \headerfile <immintrin.h>
+///
+/// This intrinsic corresponds to the \c VCVT2PH2HF8S instruction.
+///
+/// \param __U
+///    A 32-bit zeroing mask.
+/// \param __A
+///    A 256-bit vector of [16 x fp16].
+/// \param __B
+///    A 256-bit vector of [16 x fp16].
+/// \returns
+///    A 256-bit vector of [32 x hf8]. Lower 16 elements correspond to the
+///    (converted) elements from \a __B; higher order elements correspond to the
+///    (converted) elements from \a __A. If corresponding mask bit is not set,
+///    zero is taken instead.
 static __inline__ __m256i __DEFAULT_FN_ATTRS256
-_mm256_maskz_cvts2ph_hf8(__mmask32 __U, __m256h __A, __m256h __B) {
+_mm256_maskz_cvts_2ph_hf8(__mmask32 __U, __m256h __A, __m256h __B) {
   return (__m256i)__builtin_ia32_selectb_256(
-      (__mmask16)__U, (__v32qi)_mm256_cvts2ph_hf8(__A, __B),
+      (__mmask32)__U, (__v32qi)_mm256_cvts_2ph_hf8(__A, __B),
       (__v32qi)(__m256i)_mm256_setzero_si256());
 }
 
-static __inline__ __m128h __DEFAULT_FN_ATTRS128 _mm_cvthf8(__m128i __A) {
+/// Convert 128-bit vector \a __A, containing packed FP8 E4M3 floating-point
+///    elements to a 128-bit vector containing FP16 elements. The conversion is exact.
+///
+/// \code{.operation}
+/// FOR i := 0 to 7
+/// 	dst.fp16[i] := convert_hf8_to_fp16(__A.hf8[i])
+/// ENDFOR
+///
+/// dst[MAX:128] := 0
+/// \endcode
+///
+/// \headerfile <immintrin.h>
+///
+/// This intrinsic corresponds to the \c VCVTHF82PH instruction.
+///
+/// \param __A
+///    A 128-bit vector of [16 x hf8].
+/// \returns
+///    A 128-bit vector of [8 x fp16]. Resulting elements correspond to the
+///    (converted) elements from \a __A.
+static __inline__ __m128h __DEFAULT_FN_ATTRS128 _mm_cvthf8_ph(__m128i __A) {
   return (__m128h)__builtin_ia32_vcvthf8_2ph128_mask(
       (__v16qi)__A, (__v8hf)(__m128h)_mm_undefined_ph(), (__mmask8)-1);
 }
 
-static __inline__ __m128h __DEFAULT_FN_ATTRS128 _mm_mask_cvthf8(__m128h __W,
-                                                                __mmask8 __U,
-                                                                __m128i __A) {
+/// Convert 128-bit vector \a __A, containing packed FP8 E4M3 floating-point
+///    elements to a 128-bit vector containing FP16 elements. The conversion is
+///    exact. Merging mask \a __U is used to determine if given element should be
+///    taken from \a __W instead.
+///
+/// \code{.operation}
+/// FOR i := 0 to 7
+/// 	IF __U[i]
+/// 		dst.fp16[i] := convert_hf8_to_fp16(__A.hf8[i])
+/// 	ELSE
+/// 		dst.fp16[i] := __W.fp16[i]
+/// 	FI
+/// ENDFOR
+///
+/// dst[MAX:128] := 0
+/// \endcode
+///
+/// \headerfile <immintrin.h>
+///
+/// This intrinsic corresponds to the \c VCVTHF82PH instruction.
+///
+/// \param __W
+///    A 128-bit vector of [8 x fp16].
+/// \param __U
+///    A 8-bit merging mask.
+/// \param __A
+///    A 128-bit vector of [16 x hf8].
+/// \returns
+///    A 128-bit vector of [8 x fp16]. Resulting elements correspond to the
+///    (converted) elements from \a __A. If corresponding mask bit is not set, then
+///    element from \a __W is taken instead.
+static __inline__ __m128h __DEFAULT_FN_ATTRS128
+_mm_mask_cvthf8_ph(__m128h __W, __mmask8 __U, __m128i __A) {
   return (__m128h)__builtin_ia32_vcvthf8_2ph128_mask(
       (__v16qi)__A, (__v8hf)(__m128h)__W, (__mmask8)__U);
 }
 
-static __inline__ __m128h __DEFAULT_FN_ATTRS128 _mm_maskz_cvthf8(__mmask8 __U,
-                                                                 __m128i __A) {
+/// Convert 128-bit vector \a __A, containing packed FP8 E4M3 floating-point
+///    elements to a 128-bit vector containing FP16 elements. The conversion is
+///    exact. Zeroing mask \a __U is used to determine if given element should be
+///    zeroed instead.
+///
+/// \code{.operation}
+/// FOR i := 0 to 7
+/// 	IF __U[i]
+/// 		dst.fp16[i] := convert_hf8_to_fp16(__A.hf8[i])
+/// 	ELSE
+/// 		dst.fp16[i] := 0
+/// 	FI
+/// ENDFOR
+///
+/// dst[MAX:128] := 0
+/// \endcode
+///
+/// \headerfile <immintrin.h>
+///
+/// This intrinsic corresponds to the \c VCVTHF82PH instruction.
+///
+/// \param __U
+///    A 8-bit zeroing mask.
+/// \param __A
+///    A 128-bit vector of [16 x hf8].
+/// \returns
+///    A 128-bit vector of [8 x fp16]. Resulting elements correspond to the
+///    (converted) elements from \a __A. If corresponding mask bit is not set, then
+///    zero is taken instead.
+static __inline__ __m128h __DEFAULT_FN_ATTRS128
+_mm_maskz_cvthf8_ph(__mmask8 __U, __m128i __A) {
   return (__m128h)__builtin_ia32_vcvthf8_2ph128_mask(
       (__v16qi)__A, (__v8hf)(__m128h)_mm_setzero_ph(), (__mmask8)__U);
 }
 
-static __inline__ __m256h __DEFAULT_FN_ATTRS256 _mm256_cvthf8(__m128i __A) {
+/// Convert 256-bit vector \a __A, containing packed FP8 E4M3 floating-point
+///    elements to a 256-bit vector containing FP16 elements. The conversion is exact.
+///
+/// \code{.operation}
+/// FOR i := 0 to 15
+/// 	dst.fp16[i] := convert_hf8_to_fp16(__A.hf8[i])
+/// ENDFOR
+///
+/// dst[MAX:256] := 0
+/// \endcode
+///
+/// \headerfile <immintrin.h>
+///
+/// This intrinsic corresponds to the \c VCVTHF82PH instruction.
+///
+/// \param __A
+///    A 256-bit vector of [32 x hf8].
+/// \returns
+///    A 256-bit vector of [16 x fp16]. Resulting elements correspond to the
+///    (converted) elements from \a __A.
+static __inline__ __m256h __DEFAULT_FN_ATTRS256 _mm256_cvthf8_ph(__m128i __A) {
   return (__m256h)__builtin_ia32_vcvthf8_2ph256_mask(
       (__v16qi)__A, (__v16hf)(__m256h)_mm256_undefined_ph(), (__mmask16)-1);
 }
 
+/// Convert 256-bit vector \a __A, containing packed FP8 E4M3 floating-point
+///    elements to a 256-bit vector containing FP16 elements. The conversion is
+///    exact. Merging mask \a __U is used to determine if given element should be
+///    taken from \a __W instead.
+///
+/// \code{.operation}
+/// FOR i := 0 to 15 
+/// 	IF __U[i]
+/// 		dst.fp16[i] := convert_hf8_to_fp16(__A.hf8[i])
+/// 	ELSE
+/// 		dst.fp16[i] := __W.fp16[i]
+/// 	FI
+/// ENDFOR
+///
+/// dst[MAX:256] := 0
+/// \endcode
+///
+/// \headerfile <immintrin.h>
+///
+/// This intrinsic corresponds to the \c VCVTHF82PH instruction.
+///
+/// \param __W
+///    A 256-bit vector of [16 x fp16].
+/// \param __U
+///    A 16-bit merging mask.
+/// \param __A
+///    A 256-bit vector of [32 x hf8].
+/// \returns
+///    A 256-bit vector of [16 x fp16]. Resulting elements correspond to the
+///    (converted) elements from \a __A. If corresponding mask bit is not set, then
+///    element from \a __W is taken instead.
 static __inline__ __m256h __DEFAULT_FN_ATTRS256
-_mm256_mask_cvthf8(__m256h __W, __mmask16 __U, __m128i __A) {
+_mm256_mask_cvthf8_ph(__m256h __W, __mmask16 __U, __m128i __A) {
   return (__m256h)__builtin_ia32_vcvthf8_2ph256_mask(
       (__v16qi)__A, (__v16hf)(__m256h)__W, (__mmask16)__U);
 }
 
+/// Convert 256-bit vector \a __A, containing packed FP8 E4M3 floating-point
+///    elements to a 256-bit vector containing FP16 elements. The conversion is
+///    exact. Zeroing mask \a __U is used to determine if given element should be
+///    zeroed instead.
+///
+/// \code{.operation}
+/// FOR i := 0 to 15 
+/// 	IF __U[i]
+/// 		dst.fp16[i] := convert_hf8_to_fp16(__A.hf8[i])
+/// 	ELSE
+/// 		dst.fp16[i] := 0
+/// 	FI
+/// ENDFOR
+///
+/// dst[MAX:256] := 0
+/// \endcode
+///
+/// \headerfile <immintrin.h>
+///
+/// This intrinsic corresponds to the \c VCVTHF82PH instruction.
+///
+/// \param __U
+///    A 16-bit zeroing mask.
+/// \param __A
+///    A 256-bit vector of [32 x hf8].
+/// \returns
+///    A 256-bit vector of [16 x fp16]. Resulting elements correspond to the
+///    (converted) elements from \a __A. If corresponding mask bit is not set, then
+///    zero is taken instead.
 static __inline__ __m256h __DEFAULT_FN_ATTRS256
-_mm256_maskz_cvthf8(__mmask16 __U, __m128i __A) {
+_mm256_maskz_cvthf8_ph(__mmask16 __U, __m128i __A) {
   return (__m256h)__builtin_ia32_vcvthf8_2ph256_mask(
       (__v16qi)__A, (__v16hf)(__m256h)_mm256_setzero_ph(), (__mmask16)__U);
 }
 
+/// Convert 128-bit vector \a __A containing packed FP16 floating-point elements
+///    to a 128-bit vector containing E5M2 FP8 elements. Upper elements of
+///    resulting vector are zeroed.
+///
+/// \code{.operation}
+/// FOR i := 0 to 7
+/// 	dst.bf8[i] := convert_fp16_to_bf8(__A.fp16[i])
+/// ENDFOR
+///
+/// dst[MAX:64] := 0
+/// \endcode
+///
+/// \headerfile <immintrin.h>
+///
+/// This intrinsic corresponds to the \c VCVTPH2BF8 instruction.
+///
+/// \param __A
+///    A 128-bit vector of [8 x fp16].
+/// \returns
+///    A 128-bit vector of [16 x bf8]. Lower elements correspond to the (converted)
+///    elements from \a __A; upper elements are zeroed. 
 static __inline__ __m128i __DEFAULT_FN_ATTRS128 _mm_cvtph_bf8(__m128h __A) {
   return (__m128i)__builtin_ia32_vcvtph2bf8_128_mask(
       (__v8hf)__A, (__v16qi)(__m128i)_mm_undefined_si128(), (__mmask8)-1);
 }
 
+/// Convert 128-bit vector \a __A containing packed FP16 floating-point elements
+///    to a 128-bit vector containing E5M2 FP8 elements. Upper elements of
+///    resulting vector are zeroed. Merging mask \a __U is used to determine if
+///    given element should be taken from \a __W instead.
+///
+/// \code{.operation}
+/// FOR i := 0 to 7
+/// 	IF __U[i]
+/// 		dst.bf8[i] := convert_fp16_to_bf8(__A.fp16[i])
+/// 	ELSE
+/// 		dst.bf8[i] := __W.bf8[i]
+/// 	FI
+/// ENDFOR
+///
+/// dst[MAX:64] := 0
+/// \endcode
+///
+/// \headerfile <immintrin.h>
+///
+/// This intrinsic corresponds to the \c VCVTPH2BF8 instruction.
+///
+/// \param __W
+///    A 128-bit vector of [16 x bf8].
+/// \param __U
+///    A 8-bit merging mask.
+/// \param __A
+///    A 128-bit vector of [8 x fp16].
+/// \returns
+///    A 128-bit vector of [16 x bf8]. Lower elements correspond to the
+///    (converted) elements from \a __A; upper elements are zeroed. If
+///    corresponding mask bit is not set, then element from \a __W is taken instead.
 static __inline__ __m128i __DEFAULT_FN_ATTRS128
 _mm_mask_cvtph_bf8(__m128i __W, __mmask8 __U, __m128h __A) {
   return (__m128i)__builtin_ia32_vcvtph2bf8_128_mask(
       (__v8hf)__A, (__v16qi)(__m128i)__W, (__mmask8)__U);
 }
 
+/// Convert 128-bit vector \a __A containing packed FP16 floating-point elements
+///    to a 128-bit vector containing E5M2 FP8 elements. Upper elements of
+///    resulting vector are zeroed. Zeroing mask \a __U is used to determine if
+///    given element should be zeroed instead.
+///
+/// \code{.operation}
+/// FOR i := 0 to 7
+/// 	IF __U[i]
+/// 		dst.bf8[i] := convert_fp16_to_bf8(__A.fp16[i])
+/// 	ELSE
+/// 		dst.bf8[i] := 0
+/// 	FI
+/// ENDFOR
+///
+/// dst[MAX:64] := 0
+/// \endcode
+///
+/// \headerfile <immintrin.h>
+///
+/// This intrinsic corresponds to the \c VCVTPH2BF8 instruction.
+///
+/// \param __U
+///    A 8-bit zeroing mask.
+/// \param __A
+///    A 128-bit vector of [8 x fp16].
+/// \returns
+///    A 128-bit vector of [16 x bf8]. Lower elements correspond to the
+///    (converted) elements from \a __A; upper elements are zeroed. If
+///    corresponding mask bit is not set, then element is zeroed.
 static __inline__ __m128i __DEFAULT_FN_ATTRS128
 _mm_maskz_cvtph_bf8(__mmask8 __U, __m128h __A) {
   return (__m128i)__builtin_ia32_vcvtph2bf8_128_mask(
       (__v8hf)__A, (__v16qi)(__m128i)_mm_setzero_si128(), (__mmask8)__U);
 }
 
-static __inline__ __m128i __DEFAULT_FN_ATTRS256 _mm256_cvtph_bf8(__m256h __A) {
+/// Convert 256-bit vector \a __A containing packed FP16 floating-point elements
+///    to a 128-bit vector containing E5M2 FP8 elements.
+///
+/// \code{.operation}
+/// FOR i := 0 to 15
+/// 	dst.bf8[i] := convert_fp16_to_bf8(__A.fp16[i])
+/// ENDFOR
+///
+/// dst[MAX:128] := 0
+/// \endcode
+///
+/// \headerfile <immintrin.h>
+///
+/// This intrinsic corresponds to the \c VCVTPH2BF8 instruction.
+///
+/// \param __A
+///    A 256-bit vector of [16 x fp16].
+/// \returns
+///    A 128-bit vector of [16 x bf8]. Resulting elements correspond to the (converted)
+///    elements from \a __A.
+static __inline__ __m128i __DEFAULT_FN_ATTRS256
+_mm256_cvtph_bf8(__m256h __A) {
   return (__m128i)__builtin_ia32_vcvtph2bf8_256_mask(
       (__v16hf)__A, (__v16qi)(__m128i)_mm_undefined_si128(), (__mmask16)-1);
 }
 
+/// Convert 256-bit vector \a __A containing packed FP16 floating-point elements
+///    to a 128-bit vector containing E5M2 FP8 elements. Merging mask \a __U is
+///    used to determine if given element should be taken from \a __W instead.
+///   
+/// \code{.operation}
+/// FOR i := 0 to 15
+/// 	IF __U[i]
+/// 		dst.bf8[i] := convert_fp16_to_bf8(__A.fp16[i])
+/// 	ELSE
+/// 		dst.bf8[i] := __W.bf8[i]
+/// 	FI
+/// ENDFOR
+///
+/// dst[MAX:128] := 0
+/// \endcode
+///
+/// \headerfile <immintrin.h>
+///
+/// This intrinsic corresponds to the \c VCVTPH2BF8 instruction.
+///
+/// \param __W
+///    A 128-bit vector of [16 x bf8].
+/// \param __U
+///    A 16-bit merging mask.
+/// \param __A
+///    A 256-bit vector of [8 x fp16].
+/// \returns
+///    A 128-bit vector of [16 x bf8]. Resulting elements correspond to the
+///    (converted) elements from \a __A. If
+///    corresponding mask bit is not set, then element from \a __W is taken instead.
 static __inline__ __m128i __DEFAULT_FN_ATTRS256
 _mm256_mask_cvtph_bf8(__m128i __W, __mmask16 __U, __m256h __A) {
   return (__m128i)__builtin_ia32_vcvtph2bf8_256_mask(
       (__v16hf)__A, (__v16qi)(__m128i)__W, (__mmask16)__U);
 }
 
+/// Convert 256-bit vector \a __A containing packed FP16 floating-point elements
+///    to a 128-bit vector containing E5M2 FP8 elements. Zeroing mask \a __U is
+///    used to determine if given element should be zeroed instead.
+///   
+/// \code{.operation}
+/// FOR i := 0 to 15
+/// 	IF __U[i]
+/// 		dst.bf8[i] := convert_fp16_to_bf8(__A.fp16[i])
+/// 	ELSE
+/// 		dst.bf8[i] := 0
+/// 	FI
+/// ENDFOR
+///
+/// dst[MAX:128] := 0
+/// \endcode
+///
+/// \headerfile <immintrin.h>
+///
+/// This intrinsic corresponds to the \c VCVTPH2BF8 instruction.
+///
+/// \param __U
+///    A 16-bit zeroing mask.
+/// \param __A
+///    A 256-bit vector of [16 x fp16].
+/// \returns
+///    A 128-bit vector of [16 x bf8]. Resulting elements correspond to the
+///    (converted) elements from \a __A. If corresponding mask bit is not set,
+///    then element is zeroed instead.
 static __inline__ __m128i __DEFAULT_FN_ATTRS256
 _mm256_maskz_cvtph_bf8(__mmask16 __U, __m256h __A) {
   return (__m128i)__builtin_ia32_vcvtph2bf8_256_mask(
       (__v16hf)__A, (__v16qi)(__m128i)_mm_setzero_si128(), (__mmask16)__U);
 }
 
-static __inline__ __m128i __DEFAULT_FN_ATTRS128 _mm_cvtsph_bf8(__m128h __A) {
+/// Convert 128-bit vector \a __A containing packed FP16 floating-point elements
+///    to a 128-bit vector containing E5M2 FP8 elements. Upper elements of
+///    resulting vector are zeroed. Results are saturated.
+///   
+/// \code{.operation}
+/// FOR i := 0 to 7
+/// 	dst.bf8[i] := convert_fp16_to_bf8_saturate(__A.fp16[i])
+/// ENDFOR
+///
+/// dst[MAX:64] := 0
+/// \endcode
+///
+/// \headerfile <immintrin.h>
+///
+/// This intrinsic corresponds to the \c VCVTPH2BF8S instruction.
+///
+/// \param __A
+///    A 128-bit vector of [8 x fp16].
+/// \returns
+///    A 128-bit vector of [16 x bf8]. Lower elements correspond to the (converted)
+///    elements from \a __A; upper elements are zeroed. 
+static __inline__ __m128i __DEFAULT_FN_ATTRS128 _mm_cvts_ph_bf8(__m128h __A) {
   return (__m128i)__builtin_ia32_vcvtph2bf8s_128_mask(
       (__v8hf)__A, (__v16qi)(__m128i)_mm_undefined_si128(), (__mmask8)-1);
 }
 
+/// Convert 128-bit vector \a __A containing packed FP16 floating-point elements
+///    to a 128-bit vector containing E5M2 FP8 elements. Upper elements of
+///    resulting vector are zeroed. Results are saturated. Merging mask \a __U is
+///    used to determine if given element should be taken from \a __W instead.
+///
+/// \code{.operation}
+/// FOR i := 0 to 7
+/// 	IF __U[i]
+/// 		dst.bf8[i] := convert_fp16_to_bf8_saturate(__A.fp16[i])
+/// 	ELSE
+/// 		dst.bf8[i] := __W.bf8[i]
+/// 	FI
+/// ENDFOR
+///
+/// dst[MAX:64] := 0
+/// \endcode
+///
+/// \headerfile <immintrin.h>
+///
+/// This intrinsic corresponds to the \c VCVTPH2BF8S instruction.
+///
+/// \param __W
+///    A 128-bit vector of [16 x bf8].
+/// \param __U
+///    A 8-bit merging mask.
+/// \param __A
+///    A 128-bit vector of [8 x fp16].
+/// \returns
+///    A 128-bit vector of [16 x bf8]. Lower elements correspond to the
+///    (converted) elements from \a __A; upper elements are zeroed. If
+///    corresponding mask bit is not set, then element from \a __W is taken instead.
 static __inline__ __m128i __DEFAULT_FN_ATTRS128
-_mm_mask_cvtsph_bf8(__m128i __W, __mmask8 __U, __m128h __A) {
+_mm_mask_cvts_ph_bf8(__m128i __W, __mmask8 __U, __m128h __A) {
   return (__m128i)__builtin_ia32_vcvtph2bf8s_128_mask(
       (__v8hf)__A, (__v16qi)(__m128i)__W, (__mmask8)__U);
 }
 
+/// Convert 128-bit vector \a __A containing packed FP16 floating-point elements
+///    to a 128-bit vector containing E5M2 FP8 elements. Upper elements of
+///    resulting vector are zeroed. Results are saturated. Zeroing mask \a __U is
+///    used to determine if given element should be zeroed instead.
+///
+/// \code{.operation}
+/// FOR i := 0 to 7
+/// 	IF __U[i]
+/// 		dst.bf8[i] := convert_fp16_to_bf8_saturate(__A.fp16[i])
+/// 	ELSE
+/// 		dst.bf8[i] := 0
+/// 	FI
+/// ENDFOR
+///
+/// dst[MAX:64] := 0
+/// \endcode
+///
+/// \headerfile <immintrin.h>
+///
+/// This intrinsic corresponds to the \c VCVTPH2BF8S instruction.
+///
+/// \param __U
+///    A 8-bit zeroing mask.
+/// \param __A
+///    A 128-bit vector of [8 x fp16].
+/// \returns
+///    A 128-bit vector of [16 x bf8]. Lower elements correspond to the
+///    (converted) elements from \a __A; upper elements are zeroed. If
+///    corresponding mask bit is not set, then element is zeroed.
 static __inline__ __m128i __DEFAULT_FN_ATTRS128
-_mm_maskz_cvtsph_bf8(__mmask8 __U, __m128h __A) {
+_mm_maskz_cvts_ph_bf8(__mmask8 __U, __m128h __A) {
   return (__m128i)__builtin_ia32_vcvtph2bf8s_128_mask(
       (__v8hf)__A, (__v16qi)(__m128i)_mm_setzero_si128(), (__mmask8)__U);
 }
 
-static __inline__ __m128i __DEFAULT_FN_ATTRS256 _mm256_cvtsph_bf8(__m256h __A) {
+/// Convert 256-bit vector \a __A containing packed FP16 floating-point elements
+///    to a 128-bit vector containing E5M2 FP8 elements. Results are saturated.
+///
+/// \code{.operation}
+/// FOR i := 0 to 15
+/// 	dst.bf8[i] := convert_fp16_to_bf8_saturate(__A.fp16[i])
+/// ENDFOR
+///
+/// dst[MAX:128] := 0
+/// \endcode
+///
+/// \headerfile <immintrin.h>
+///
+/// This intrinsic corresponds to the \c VCVTPH2BF8S instruction.
+///
+/// \param __A
+///    A 256-bit vector of [16 x fp16].
+/// \returns
+///    A 128-bit vector of [16 x bf8]. Resulting elements correspond to the (converted)
+///    elements from \a __A.
+static __inline__ __m128i __DEFAULT_FN_ATTRS256
+_mm256_cvts_ph_bf8(__m256h __A) {
   return (__m128i)__builtin_ia32_vcvtph2bf8s_256_mask(
       (__v16hf)__A, (__v16qi)(__m128i)_mm_undefined_si128(), (__mmask16)-1);
 }
 
+/// Convert 256-bit vector \a __A containing packed FP16 floating-point elements
+///    to a 128-bit vector containing E5M2 FP8 elements. Results are saturated.
+///    Merging mask \a __U is used to determine if given element should be taken
+///    from \a __W instead.
+///
+/// \code{.operation}
+/// FOR i := 0 to 15
+/// 	IF __U[i]
+/// 		dst.bf8[i] := convert_fp16_to_bf8_saturate(__A.fp16[i])
+/// 	ELSE
+/// 		dst.bf8[i] := __W.bf8[i]
+/// 	FI
+/// ENDFOR
+///
+/// dst[MAX:128] := 0
+/// \endcode
+///
+/// \headerfile <immintrin.h>
+///
+/// This intrinsic corresponds to the \c VCVTPH2BF8S instruction.
+///
+/// \param __W
+///    A 128-bit vector of [16 x bf8].
+/// \param __U
+///    A 16-bit merging mask.
+/// \param __A
+///    A 256-bit vector of [8 x fp16].
+/// \returns
+///    A 128-bit vector of [16 x bf8]. Resulting elements correspond to the
+///    (converted) elements from \a __A. If
+///    corresponding mask bit is not set, then element from \a __W is taken instead.
 static __inline__ __m128i __DEFAULT_FN_ATTRS256
-_mm256_mask_cvtsph_bf8(__m128i __W, __mmask16 __U, __m256h __A) {
+_mm256_mask_cvts_ph_bf8(__m128i __W, __mmask16 __U, __m256h __A) {
   return (__m128i)__builtin_ia32_vcvtph2bf8s_256_mask(
       (__v16hf)__A, (__v16qi)(__m128i)__W, (__mmask16)__U);
 }
 
+/// Convert 256-bit vector \a __A containing packed FP16 floating-point elements
+///    to a 128-bit vector containing E5M2 FP8 elements. Results are saturated.
+///    Zeroing mask \a __U is used to determine if given element should be zeroed
+///    instead.
+///
+/// \code{.operation}
+/// FOR i := 0 to 15 
+/// 	IF __U[i]
+/// 		dst.bf8[i] := convert_fp16_to_bf8_saturate(__A.fp16[i])
+/// 	ELSE
+/// 		dst.bf8[i] := 0
+/// 	FI
+/// ENDFOR
+///
+/// dst[MAX:128] := 0
+/// \endcode
+///
+/// \headerfile <immintrin.h>
+///
+/// This intrinsic corresponds to the \c VCVTPH2BF8S instruction.
+///
+/// \param __U
+///    A 16-bit zeroing mask.
+/// \param __A
+///    A 256-bit vector of [16 x fp16].
+/// \returns
+///    A 128-bit vector of [16 x bf8]. Resulting elements correspond to the
+///    (converted) elements from \a __A. If corresponding mask bit is not set,
+///    then element is zeroed instead.
 static __inline__ __m128i __DEFAULT_FN_ATTRS256
-_mm256_maskz_cvtsph_bf8(__mmask16 __U, __m256h __A) {
+_mm256_maskz_cvts_ph_bf8(__mmask16 __U, __m256h __A) {
   return (__m128i)__builtin_ia32_vcvtph2bf8s_256_mask(
       (__v16hf)__A, (__v16qi)(__m128i)_mm_setzero_si128(), (__mmask16)__U);
 }
 
+/// Convert 128-bit vector \a __A containing packed FP16 floating-point elements
+///    to a 128-bit vector containing E5M2 FP8 elements. Upper elements of
+///    resulting vector are zeroed.
+///
+/// \code{.operation}
+/// FOR i := 0 to 7
+/// 	dst.hf8[i] := convert_fp16_to_hf8(__A.fp16[i])
+/// ENDFOR
+///
+/// dst[MAX:64] := 0
+/// \endcode
+///
+/// \headerfile <immintrin.h>
+///
+/// This intrinsic corresponds to the \c VCVTPH2HF8 instruction.
+///
+/// \param __A
+///    A 128-bit vector of [8 x fp16].
+/// \returns
+///    A 128-bit vector of [16 x hf8]. Lower elements correspond to the (converted)
+///    elements from \a __A; upper elements are zeroed. 
 static __inline__ __m128i __DEFAULT_FN_ATTRS128 _mm_cvtph_hf8(__m128h __A) {
   return (__m128i)__builtin_ia32_vcvtph2hf8_128_mask(
       (__v8hf)__A, (__v16qi)(__m128i)_mm_undefined_si128(), (__mmask8)-1);
 }
 
+/// Convert 128-bit vector \a __A containing packed FP16 floating-point elements
+///    to a 128-bit vector containing E4M3 FP8 elements. Upper elements of
+///    resulting vector are zeroed. Merging mask \a __U is used to determine if
+///    given element should be taken from \a __W instead.
+///
+/// \code{.operation}
+/// FOR i := 0 to 7
+/// 	IF __U[i]
+/// 		dst.hf8[i] := convert_fp16_to_hf8(__A.fp16[i])
+/// 	ELSE
+/// 		dst.hf8[i] := __W.hf8[i]
+/// 	FI
+/// ENDFOR
+///
+/// dst[MAX:64] := 0
+/// \endcode
+///
+/// \headerfile <immintrin.h>
+///
+/// This intrinsic corresponds to the \c VCVTPH2HF8 instruction.
+///
+/// \param __W
+///    A 128-bit vector of [16 x hf8].
+/// \param __U
+///    A 8-bit merging mask.
+/// \param __A
+///    A 128-bit vector of [8 x fp16].
+/// \returns
+///    A 128-bit vector of [16 x hf8]. Lower elements correspond to the
+///    (converted) elements from \a __A; upper elements are zeroed. If
+///    corresponding mask bit is not set, then element from \a __W is taken instead.
 static __inline__ __m128i __DEFAULT_FN_ATTRS128
 _mm_mask_cvtph_hf8(__m128i __W, __mmask8 __U, __m128h __A) {
   return (__m128i)__builtin_ia32_vcvtph2hf8_128_mask(
       (__v8hf)__A, (__v16qi)(__m128i)__W, (__mmask8)__U);
 }
 
+/// Convert 128-bit vector \a __A containing packed FP16 floating-point elements
+///    to a 128-bit vector containing E4M3 FP8 elements. Upper elements of
+///    resulting vector are zeroed. Zeroing mask \a __U is used to determine if
+///    given element should be zeroed instead.
+///
+/// \code{.operation}
+/// FOR i := 0 to 7
+/// 	IF __U[i]
+/// 		dst.hf8[i] := convert_fp16_to_hf8(__A.fp16[i])
+/// 	ELSE
+/// 		dst.hf8[i] := 0
+/// 	FI
+/// ENDFOR
+///
+/// dst[MAX:64] := 0
+/// \endcode
+///
+/// \headerfile <immintrin.h>
+///
+/// This intrinsic corresponds to the \c VCVTPH2HF8 instruction.
+///
+/// \param __U
+///    A 8-bit zeroing mask.
+/// \param __A
+///    A 128-bit vector of [8 x fp16].
+/// \returns
+///    A 128-bit vector of [16 x hf8]. Lower elements correspond to the
+///    (converted) elements from \a __A; upper elements are zeroed. If
+///    corresponding mask bit is not set, then element is zeroed.
 static __inline__ __m128i __DEFAULT_FN_ATTRS128
 _mm_maskz_cvtph_hf8(__mmask8 __U, __m128h __A) {
   return (__m128i)__builtin_ia32_vcvtph2hf8_128_mask(
       (__v8hf)__A, (__v16qi)(__m128i)_mm_setzero_si128(), (__mmask8)__U);
 }
 
-static __inline__ __m128i __DEFAULT_FN_ATTRS256 _mm256_cvtph_hf8(__m256h __A) {
+/// Convert 256-bit vector \a __A containing packed FP16 floating-point elements
+///    to a 128-bit vector containing E4M3 FP8 elements.
+///
+/// \code{.operation}
+/// FOR i := 0 to 15
+/// 	dst.hf8[i] := convert_fp16_to_hf8(__A.fp16[i])
+/// ENDFOR
+///
+/// dst[MAX:128] := 0
+/// \endcode
+///
+/// \headerfile <immintrin.h>
+///
+/// This intrinsic corresponds to the \c VCVTPH2HF8 instruction.
+///
+/// \param __A
+///    A 256-bit vector of [16 x fp16].
+/// \returns
+///    A 128-bit vector of [16 x hf8]. Resulting elements correspond to the (converted)
+///    elements from \a __A.
+static __inline__ __m128i __DEFAULT_FN_ATTRS256
+_mm256_cvtph_hf8(__m256h __A) {
   return (__m128i)__builtin_ia32_vcvtph2hf8_256_mask(
       (__v16hf)__A, (__v16qi)(__m128i)_mm_undefined_si128(), (__mmask16)-1);
 }
 
+/// Convert 256-bit vector \a __A containing packed FP16 floating-point elements
+///    to a 128-bit vector containing E4M3 FP8 elements. Merging mask \a __U is
+///    used to determine if given element should be taken from \a __W instead.
+///   
+/// \code{.operation}
+/// FOR i := 0 to 15
+/// 	IF __U[i]
+/// 		dst.hf8[i] := convert_fp16_to_hf8(__A.fp16[i])
+/// 	ELSE
+/// 		dst.hf8[i] := __W.hf8[i]
+/// 	FI
+/// ENDFOR
+///
+/// dst[MAX:128] := 0
+/// \endcode
+///
+/// \headerfile <immintrin.h>
+///
+/// This intrinsic corresponds to the \c VCVTPH2HF8 instruction.
+///
+/// \param __W
+///    A 128-bit vector of [16 x hf8].
+/// \param __U
+///    A 16-bit merging mask.
+/// \param __A
+///    A 256-bit vector of [8 x fp16].
+/// \returns
+///    A 128-bit vector of [16 x hf8]. Resulting elements correspond to the
+///    (converted) elements from \a __A. If
+///    corresponding mask bit is not set, then element from \a __W is taken instead.
 static __inline__ __m128i __DEFAULT_FN_ATTRS256
 _mm256_mask_cvtph_hf8(__m128i __W, __mmask16 __U, __m256h __A) {
   return (__m128i)__builtin_ia32_vcvtph2hf8_256_mask(
       (__v16hf)__A, (__v16qi)(__m128i)__W, (__mmask16)__U);
 }
 
+/// Convert 256-bit vector \a __A containing packed FP16 floating-point elements
+///    to a 128-bit vector containing E4M3 FP8 elements. Zeroing mask \a __U is
+///    used to determine if given element should be zeroed instead.
+///   
+/// \code{.operation}
+/// FOR i := 0 to 15
+/// 	IF __U[i]
+/// 		dst.hf8[i] := convert_fp16_to_hf8(__A.fp16[i])
+/// 	ELSE
+/// 		dst.hf8[i] := 0
+/// 	FI
+/// ENDFOR
+///
+/// dst[MAX:128] := 0
+/// \endcode
+///
+/// \headerfile <immintrin.h>
+///
+/// This intrinsic corresponds to the \c VCVTPH2HF8 instruction.
+///
+/// \param __U
+///    A 16-bit zeroing mask.
+/// \param __A
+///    A 256-bit vector of [16 x fp16].
+/// \returns
+///    A 128-bit vector of [16 x hf8]. Resulting elements correspond to the
+///    (converted) elements from \a __A. If corresponding mask bit is not set,
+///    then element is zeroed instead.
 static __inline__ __m128i __DEFAULT_FN_ATTRS256
 _mm256_maskz_cvtph_hf8(__mmask16 __U, __m256h __A) {
   return (__m128i)__builtin_ia32_vcvtph2hf8_256_mask(
       (__v16hf)__A, (__v16qi)(__m128i)_mm_setzero_si128(), (__mmask16)__U);
 }
 
-static __inline__ __m128i __DEFAULT_FN_ATTRS128 _mm_cvtsph_hf8(__m128h __A) {
+/// Convert 128-bit vector \a __A containing packed FP16 floating-point elements
+///    to a 128-bit vector containing E4M3 FP8 elements. Upper elements of
+///    resulting vector are zeroed. Results are saturated.
+///   
+/// \code{.operation}
+/// FOR i := 0 to 7
+/// 	dst.hf8[i] := convert_fp16_to_hf8_saturate(__A.fp16[i])
+/// ENDFOR
+///
+/// dst[MAX:64] := 0
+/// \endcode
+///
+/// \headerfile <immintrin.h>
+///
+/// This intrinsic corresponds to the \c VCVTPH2HF8S instruction.
+///
+/// \param __A
+///    A 128-bit vector of [8 x fp16].
+/// \returns
+///    A 128-bit vector of [16 x hf8]. Lower elements correspond to the (converted)
+///    elements from \a __A; upper elements are zeroed. 
+static __inline__ __m128i __DEFAULT_FN_ATTRS128 _mm_cvts_ph_hf8(__m128h __A) {
   return (__m128i)__builtin_ia32_vcvtph2hf8s_128_mask(
       (__v8hf)__A, (__v16qi)(__m128i)_mm_undefined_si128(), (__mmask8)-1);
 }
 
+/// Convert 128-bit vector \a __A containing packed FP16 floating-point elements
+///    to a 128-bit vector containing E4M3 FP8 elements. Upper elements of
+///    resulting vector are zeroed. Results are saturated. Merging mask \a __U is
+///    used to determine if given element should be taken from \a __W instead.
+///
+/// \code{.operation}
+/// FOR i := 0 to 7
+/// 	IF __U[i]
+/// 		dst.hf8[i] := convert_fp16_to_hf8_saturate(__A.fp16[i])
+/// 	ELSE
+/// 		dst.hf8[i] := __W.hf8[i]
+/// 	FI
+/// ENDFOR
+///
+/// dst[MAX:64] := 0
+/// \endcode
+///
+/// \headerfile <immintrin.h>
+///
+/// This intrinsic corresponds to the \c VCVTPH2HF8S instruction.
+///
+/// \param __W
+///    A 128-bit vector of [16 x hf8].
+/// \param __U
+///    A 8-bit merging mask.
+/// \param __A
+///    A 128-bit vector of [8 x fp16].
+/// \returns
+///    A 128-bit vector of [16 x hf8]. Lower elements correspond to the
+///    (converted) elements from \a __A; upper elements are zeroed. If
+///    corresponding mask bit is not set, then element from \a __W is taken instead.
 static __inline__ __m128i __DEFAULT_FN_ATTRS128
-_mm_mask_cvtsph_hf8(__m128i __W, __mmask8 __U, __m128h __A) {
+_mm_mask_cvts_ph_hf8(__m128i __W, __mmask8 __U, __m128h __A) {
   return (__m128i)__builtin_ia32_vcvtph2hf8s_128_mask(
       (__v8hf)__A, (__v16qi)(__m128i)__W, (__mmask8)__U);
 }
 
+/// Convert 128-bit vector \a __A containing packed FP16 floating-point elements
+///    to a 128-bit vector containing E4M3 FP8 elements. Upper elements of
+///    resulting vector are zeroed. Results are saturated. Zeroing mask \a __U is
+///    used to determine if given element should be zeroed instead.
+///
+/// \code{.operation}
+/// FOR i := 0 to 7
+/// 	IF __U[i]
+/// 		dst.hf8[i] := convert_fp16_to_hf8_saturate(__A.fp16[i])
+/// 	ELSE
+/// 		dst.hf8[i] := 0
+/// 	FI
+/// ENDFOR
+///
+/// dst[MAX:64] := 0
+/// \endcode
+///
+/// \headerfile <immintrin.h>
+///
+/// This intrinsic corresponds to the \c VCVTPH2HF8S instruction.
+///
+/// \param __U
+///    A 8-bit zeroing mask.
+/// \param __A
+///    A 128-bit vector of [8 x fp16].
+/// \returns
+///    A 128-bit vector of [16 x hf8]. Lower elements correspond to the
+///    (converted) elements from \a __A; upper elements are zeroed. If
+///    corresponding mask bit is not set, then element is zeroed.
 static __inline__ __m128i __DEFAULT_FN_ATTRS128
-_mm_maskz_cvtsph_hf8(__mmask8 __U, __m128h __A) {
+_mm_maskz_cvts_ph_hf8(__mmask8 __U, __m128h __A) {
   return (__m128i)__builtin_ia32_vcvtph2hf8s_128_mask(
       (__v8hf)__A, (__v16qi)(__m128i)_mm_setzero_si128(), (__mmask8)__U);
 }
 
-static __inline__ __m128i __DEFAULT_FN_ATTRS256 _mm256_cvtsph_hf8(__m256h __A) {
+/// Convert 256-bit vector \a __A containing packed FP16 floating-point elements
+///    to a 128-bit vector containing E4M3 FP8 elements. Results are saturated.
+///
+/// \code{.operation}
+/// FOR i := 0 to 15
+/// 	dst.hf8[i] := convert_fp16_to_hf8_saturate(__A.fp16[i])
+/// ENDFOR
+///
+/// dst[MAX:128] := 0
+/// \endcode
+///
+/// \headerfile <immintrin.h>
+///
+/// This intrinsic corresponds to the \c VCVTPH2HF8S instruction.
+///
+/// \param __A
+///    A 256-bit vector of [16 x fp16].
+/// \returns
+///    A 128-bit vector of [16 x hf8]. Resulting elements correspond to the (converted)
+///    elements from \a __A.
+static __inline__ __m128i __DEFAULT_FN_ATTRS256
+_mm256_cvts_ph_hf8(__m256h __A) {
   return (__m128i)__builtin_ia32_vcvtph2hf8s_256_mask(
       (__v16hf)__A, (__v16qi)(__m128i)_mm_undefined_si128(), (__mmask16)-1);
 }
 
+/// Convert 256-bit vector \a __A containing packed FP16 floating-point elements
+///    to a 128-bit vector containing E4M3 FP8 elements. Results are saturated.
+///    Merging mask \a __U is used to determine if given element should be taken
+///    from \a __W instead.
+///
+/// \code{.operation}
+/// FOR i := 0 to 15
+/// 	IF __U[i]
+/// 		dst.hf8[i] := convert_fp16_to_hf8_saturate(__A.fp16[i])
+/// 	ELSE
+/// 		dst.hf8[i] := __W.hf8[i]
+/// 	FI
+/// ENDFOR
+///
+/// dst[MAX:128] := 0
+/// \endcode
+///
+/// \headerfile <immintrin.h>
+///
+/// This intrinsic corresponds to the \c VCVTPH2HF8S instruction.
+///
+/// \param __W
+///    A 128-bit vector of [16 x hf8].
+/// \param __U
+///    A 16-bit merging mask.
+/// \param __A
+///    A 256-bit vector of [8 x fp16].
+/// \returns
+///    A 128-bit vector of [16 x hf8]. Resulting elements correspond to the
+///    (converted) elements from \a __A. If
+///    corresponding mask bit is not set, then element from \a __W is taken instead.
 static __inline__ __m128i __DEFAULT_FN_ATTRS256
-_mm256_mask_cvtsph_hf8(__m128i __W, __mmask16 __U, __m256h __A) {
+_mm256_mask_cvts_ph_hf8(__m128i __W, __mmask16 __U, __m256h __A) {
   return (__m128i)__builtin_ia32_vcvtph2hf8s_256_mask(
       (__v16hf)__A, (__v16qi)(__m128i)__W, (__mmask16)__U);
 }
 
+/// Convert 256-bit vector \a __A containing packed FP16 floating-point elements
+///    to a 128-bit vector containing E4M3 FP8 elements. Results are saturated.
+///    Zeroing mask \a __U is used to determine if given element should be zeroed
+///    instead.
+///
+/// \code{.operation}
+/// FOR i := 0 to 15 
+/// 	IF __U[i]
+/// 		dst.hf8[i] := convert_fp16_to_hf8_saturate(__A.fp16[i])
+/// 	ELSE
+/// 		dst.hf8[i] := 0
+/// 	FI
+/// ENDFOR
+///
+/// dst[MAX:128] := 0
+/// \endcode
+///
+/// \headerfile <immintrin.h>
+///
+/// This intrinsic corresponds to the \c VCVTPH2HF8S instruction.
+///
+/// \param __U
+///    A 16-bit zeroing mask.
+/// \param __A
+///    A 256-bit vector of [16 x fp16].
+/// \returns
+///    A 128-bit vector of [16 x hf8]. Resulting elements correspond to the
+///    (converted) elements from \a __A. If corresponding mask bit is not set,
+///    then element is zeroed instead.
 static __inline__ __m128i __DEFAULT_FN_ATTRS256
-_mm256_maskz_cvtsph_hf8(__mmask16 __U, __m256h __A) {
+_mm256_maskz_cvts_ph_hf8(__mmask16 __U, __m256h __A) {
   return (__m128i)__builtin_ia32_vcvtph2hf8s_256_mask(
       (__v16hf)__A, (__v16qi)(__m128i)_mm_setzero_si128(), (__mmask16)__U);
 }
 
+/// Convert 128-bit vector \a __A, containing packed FP8 E5M2 floating-point
+///    elements to a 128-bit vector containing FP16 elements. The conversion is exact.
+///
+/// \code{.operation}
+/// FOR i := 0 to 7
+/// 	dst.fp16[i] := convert_bf8_to_fp16(__A.bf8[i])
+/// ENDFOR
+/// \endcode
+///
+/// \headerfile <immintrin.h>
+///
+/// This intrinsic does not correspond to a single instruction.
+///
+/// \param __A
+///    A 128-bit vector of [16 x bf8].
+/// \returns
+///    A 128-bit vector of [8 x fp16]. Resulting elements correspond to the
+///    (converted) elements from \a __A.
 static __inline__ __m128h __DEFAULT_FN_ATTRS128 _mm_cvtbf8_ph(__m128i __A) {
   return _mm_castsi128_ph(_mm_slli_epi16(_mm_cvtepi8_epi16(__A), 8));
 }
 
+/// Convert 128-bit vector \a __A, containing packed FP8 E5M2 floating-point
+///    elements to a 128-bit vector containing FP16 elements. The conversion is
+///    exact. Merging mask \a __U is used to determine if given element should be
+///    taken from \a __W instead.
+///
+/// \code{.operation}
+/// FOR i := 0 to 7
+/// 	IF __U[i]
+/// 		dst.fp16[i] := convert_bf8_to_fp16(__A.bf8[i])
+/// 	ELSE
+/// 		dst.fp16[i] := __W.fp16[i]
+/// 	FI
+/// ENDFOR
+/// \endcode
+///
+/// \headerfile <immintrin.h>
+///
+/// This intrinsic does not correspond to a single instruction.
+///
+/// \param __W
+///    A 128-bit vector of [8 x fp16].
+/// \param __U
+///    A 8-bit merging mask.
+/// \param __A
+///    A 128-bit vector of [16 x bf8].
+/// \returns
+///    A 128-bit vector of [8 x fp16]. Resulting elements correspond to the
+///    (converted) elements from \a __A. If corresponding mask bit is not set, then
+///    element from \a __W is taken instead.
 static __inline__ __m128h __DEFAULT_FN_ATTRS128
-_mm_mask_cvtbf8_ph(__m128h __S, __mmask8 __U, __m128i __A) {
+_mm_mask_cvtbf8_ph(__m128h __W, __mmask8 __U, __m128i __A) {
   return _mm_castsi128_ph(
-      _mm_mask_slli_epi16((__m128i)__S, __U, _mm_cvtepi8_epi16(__A), 8));
+      _mm_mask_slli_epi16((__m128i)__W, __U, _mm_cvtepi8_epi16(__A), 8));
 }
 
+/// Convert 128-bit vector \a __A, containing packed FP8 E5M2 floating-point
+///    elements to a 128-bit vector containing FP16 elements. The conversion is
+///    exact. Zeroing mask \a __U is used to determine if given element should be
+///    zeroed instead.
+///
+/// \code{.operation}
+/// FOR i := 0 to 7
+/// 	IF __U[i]
+/// 		dst.fp16[i] := convert_bf8_to_fp16(__A.bf8[i])
+/// 	ELSE
+/// 		dst.fp16[i] := 0
+/// 	FI
+/// ENDFOR
+/// \endcode
+///
+/// \headerfile <immintrin.h>
+///
+/// This intrinsic does not correspond to a single instruction.
+///
+/// \param __U
+///    A 8-bit zeroing mask.
+/// \param __A
+///    A 128-bit vector of [16 x bf8].
+/// \returns
+///    A 128-bit vector of [8 x fp16]. Resulting elements correspond to the
+///    (converted) elements from \a __A. If corresponding mask bit is not set, then
+///    zero is taken instead.
 static __inline__ __m128h __DEFAULT_FN_ATTRS128
 _mm_maskz_cvtbf8_ph(__mmask8 __U, __m128i __A) {
   return _mm_castsi128_ph(_mm_slli_epi16(_mm_maskz_cvtepi8_epi16(__U, __A), 8));
 }
 
+/// Convert 256-bit vector \a __A, containing packed FP8 E4M3 floating-point
+///    elements to a 256-bit vector containing FP16 elements. The conversion is exact.
+///
+/// \code{.operation}
+/// FOR i := 0 to 15
+/// 	dst.fp16[i] := convert_bf8_to_fp16(__A.bf8[i])
+/// ENDFOR
+/// \endcode
+///
+/// \headerfile <immintrin.h>
+///
+/// This intrinsic does not correspond to a single instruction.
+///
+/// \param __A
+///    A 256-bit vector of [32 x bf8].
+/// \returns
+///    A 256-bit vector of [16 x fp16]. Resulting elements correspond to the
+///    (converted) elements from \a __A.
 static __inline__ __m256h __DEFAULT_FN_ATTRS256 _mm256_cvtbf8_ph(__m128i __A) {
   return _mm256_castsi256_ph(_mm256_slli_epi16(_mm256_cvtepi8_epi16(__A), 8));
 }
 
+/// Convert 256-bit vector \a __A, containing packed FP8 E5M2 floating-point
+///    elements to a 256-bit vector containing FP16 elements. The conversion is
+///    exact. Merging mask \a __U is used to determine if given element should be
+///    taken from \a __W instead.
+///
+/// \code{.operation}
+/// FOR i := 0 to 15 
+/// 	IF __U[i]
+/// 		dst.fp16[i] := convert_bf8_to_fp16(__A.bf8[i])
+/// 	ELSE
+/// 		dst.fp16[i] := __W.fp16[i]
+/// 	FI
+/// ENDFOR
+/// \endcode
+///
+/// \headerfile <immintrin.h>
+///
+/// This intrinsic does not correspond to a single instruction.
+///
+/// \param __W
+///    A 256-bit vector of [16 x fp16].
+/// \param __U
+///    A 16-bit merging mask.
+/// \param __A
+///    A 256-bit vector of [32 x bf8].
+/// \returns
+///    A 256-bit vector of [16 x fp16]. Resulting elements correspond to the
+///    (converted) elements from \a __A. If corresponding mask bit is not set, then
+///    element from \a __W is taken instead.
 static __inline__ __m256h __DEFAULT_FN_ATTRS256
-_mm256_mask_cvtbf8_ph(__m256h __S, __mmask16 __U, __m128i __A) {
+_mm256_mask_cvtbf8_ph(__m256h __W, __mmask16 __U, __m128i __A) {
   return _mm256_castsi256_ph(
-      _mm256_mask_slli_epi16((__m256i)__S, __U, _mm256_cvtepi8_epi16(__A), 8));
+      _mm256_mask_slli_epi16((__m256i)__W, __U, _mm256_cvtepi8_epi16(__A), 8));
 }
 
+/// Convert 256-bit vector \a __A, containing packed FP8 E5M2 floating-point
+///    elements to a 256-bit vector containing FP16 elements. The conversion is
+///    exact. Zeroing mask \a __U is used to determine if given element should be
+///    zeroed instead.
+///
+/// \code{.operation}
+/// FOR i := 0 to 15 
+/// 	IF __U[i]
+/// 		dst.fp16[i] := convert_bf8_to_fp16(__A.bf8[i])
+/// 	ELSE
+/// 		dst.fp16[i] := 0
+/// 	FI
+/// ENDFOR
+/// \endcode
+///
+/// \headerfile <immintrin.h>
+///
+/// This intrinsic does not correspond to a single instruction.
+///
+/// \param __U
+///    A 16-bit zeroing mask.
+/// \param __A
+///    A 256-bit vector of [32 x bf8].
+/// \returns
+///    A 256-bit vector of [16 x fp16]. Resulting elements correspond to the
+///    (converted) elements from \a __A. If corresponding mask bit is not set, then
+///    zero is taken instead.
 static __inline__ __m256h __DEFAULT_FN_ATTRS256
 _mm256_maskz_cvtbf8_ph(__mmask16 __U, __m128i __A) {
   return _mm256_castsi256_ph(
       _mm256_slli_epi16(_mm256_maskz_cvtepi8_epi16(__U, __A), 8));
 }
 
+// clang-format on
+
 #undef __DEFAULT_FN_ATTRS128
 #undef __DEFAULT_FN_ATTRS256
 

lib/include/avx10_2minmaxintrin.h

@@ -66,34 +66,19 @@
       (__v2df)_mm_setzero_pd(), (__mmask8)(U)))
 
 #define _mm256_minmax_pd(A, B, C)                                              \
-  ((__m256d)__builtin_ia32_vminmaxpd256_round_mask(                            \
+  ((__m256d)__builtin_ia32_vminmaxpd256_mask(                                  \
       (__v4df)(__m256d)(A), (__v4df)(__m256d)(B), (int)(C),                    \
-      (__v4df)_mm256_setzero_pd(), (__mmask8)-1, _MM_FROUND_NO_EXC))
+      (__v4df)_mm256_setzero_pd(), (__mmask8)-1))
 
 #define _mm256_mask_minmax_pd(W, U, A, B, C)                                   \
-  ((__m256d)__builtin_ia32_vminmaxpd256_round_mask(                            \
+  ((__m256d)__builtin_ia32_vminmaxpd256_mask(                                  \
       (__v4df)(__m256d)(A), (__v4df)(__m256d)(B), (int)(C),                    \
-      (__v4df)(__m256d)(W), (__mmask8)(U), _MM_FROUND_NO_EXC))
+      (__v4df)(__m256d)(W), (__mmask8)(U)))
 
 #define _mm256_maskz_minmax_pd(U, A, B, C)                                     \
-  ((__m256d)__builtin_ia32_vminmaxpd256_round_mask(                            \
+  ((__m256d)__builtin_ia32_vminmaxpd256_mask(                                  \
       (__v4df)(__m256d)(A), (__v4df)(__m256d)(B), (int)(C),                    \
-      (__v4df)_mm256_setzero_pd(), (__mmask8)(U), _MM_FROUND_NO_EXC))
-
-#define _mm256_minmax_round_pd(A, B, C, R)                                     \
-  ((__m256d)__builtin_ia32_vminmaxpd256_round_mask(                            \
-      (__v4df)(__m256d)(A), (__v4df)(__m256d)(B), (int)(C),                    \
-      (__v4df)_mm256_undefined_pd(), (__mmask8)-1, (int)(R)))
-
-#define _mm256_mask_minmax_round_pd(W, U, A, B, C, R)                          \
-  ((__m256d)__builtin_ia32_vminmaxpd256_round_mask(                            \
-      (__v4df)(__m256d)(A), (__v4df)(__m256d)(B), (int)(C),                    \
-      (__v4df)(__m256d)(W), (__mmask8)(U), (int)(R)))
-
-#define _mm256_maskz_minmax_round_pd(U, A, B, C, R)                            \
-  ((__m256d)__builtin_ia32_vminmaxpd256_round_mask(                            \
-      (__v4df)(__m256d)(A), (__v4df)(__m256d)(B), (int)(C),                    \
-      (__v4df)_mm256_setzero_pd(), (__mmask8)(U), (int)(R)))
+      (__v4df)_mm256_setzero_pd(), (__mmask8)(U)))
 
 #define _mm_minmax_ph(A, B, C)                                                 \
   ((__m128h)__builtin_ia32_vminmaxph128_mask(                                  \
@@ -111,34 +96,19 @@
       (__v8hf)_mm_setzero_ph(), (__mmask8)(U)))
 
 #define _mm256_minmax_ph(A, B, C)                                              \
-  ((__m256h)__builtin_ia32_vminmaxph256_round_mask(                            \
+  ((__m256h)__builtin_ia32_vminmaxph256_mask(                                  \
       (__v16hf)(__m256h)(A), (__v16hf)(__m256h)(B), (int)(C),                  \
-      (__v16hf)_mm256_setzero_ph(), (__mmask16)-1, _MM_FROUND_NO_EXC))
+      (__v16hf)_mm256_setzero_ph(), (__mmask16)-1))
 
 #define _mm256_mask_minmax_ph(W, U, A, B, C)                                   \
-  ((__m256h)__builtin_ia32_vminmaxph256_round_mask(                            \
+  ((__m256h)__builtin_ia32_vminmaxph256_mask(                                  \
       (__v16hf)(__m256h)(A), (__v16hf)(__m256h)(B), (int)(C),                  \
-      (__v16hf)(__m256h)(W), (__mmask16)(U), _MM_FROUND_NO_EXC))
+      (__v16hf)(__m256h)(W), (__mmask16)(U)))
 
 #define _mm256_maskz_minmax_ph(U, A, B, C)                                     \
-  ((__m256h)__builtin_ia32_vminmaxph256_round_mask(                            \
+  ((__m256h)__builtin_ia32_vminmaxph256_mask(                                  \
       (__v16hf)(__m256h)(A), (__v16hf)(__m256h)(B), (int)(C),                  \
-      (__v16hf)_mm256_setzero_ph(), (__mmask16)(U), _MM_FROUND_NO_EXC))
-
-#define _mm256_minmax_round_ph(A, B, C, R)                                     \
-  ((__m256h)__builtin_ia32_vminmaxph256_round_mask(                            \
-      (__v16hf)(__m256h)(A), (__v16hf)(__m256h)(B), (int)(C),                  \
-      (__v16hf)_mm256_undefined_ph(), (__mmask16)-1, (int)(R)))
-
-#define _mm256_mask_minmax_round_ph(W, U, A, B, C, R)                          \
-  ((__m256h)__builtin_ia32_vminmaxph256_round_mask(                            \
-      (__v16hf)(__m256h)(A), (__v16hf)(__m256h)(B), (C),                       \
-      (__v16hf)(__m256h)(W), (__mmask16)(U), (int)(R)))
-
-#define _mm256_maskz_minmax_round_ph(U, A, B, C, R)                            \
-  ((__m256h)__builtin_ia32_vminmaxph256_round_mask(                            \
-      (__v16hf)(__m256h)(A), (__v16hf)(__m256h)(B), (int)(C),                  \
-      (__v16hf)_mm256_setzero_ph(), (__mmask16)(U), (int)(R)))
+      (__v16hf)_mm256_setzero_ph(), (__mmask16)(U)))
 
 #define _mm_minmax_ps(A, B, C)                                                 \
   ((__m128)__builtin_ia32_vminmaxps128_mask(                                   \
@@ -156,34 +126,19 @@
       (__v4sf)_mm_setzero_ps(), (__mmask8)(U)))
 
 #define _mm256_minmax_ps(A, B, C)                                              \
-  ((__m256)__builtin_ia32_vminmaxps256_round_mask(                             \
+  ((__m256)__builtin_ia32_vminmaxps256_mask(                                   \
       (__v8sf)(__m256)(A), (__v8sf)(__m256)(B), (int)(C),                      \
-      (__v8sf)_mm256_setzero_ps(), (__mmask8)-1, _MM_FROUND_NO_EXC))
+      (__v8sf)_mm256_setzero_ps(), (__mmask8)-1))
 
 #define _mm256_mask_minmax_ps(W, U, A, B, C)                                   \
-  ((__m256)__builtin_ia32_vminmaxps256_round_mask(                             \
+  ((__m256)__builtin_ia32_vminmaxps256_mask(                                   \
       (__v8sf)(__m256)(A), (__v8sf)(__m256)(B), (int)(C), (__v8sf)(__m256)(W), \
-      (__mmask8)(U), _MM_FROUND_NO_EXC))
+      (__mmask8)(U)))
 
 #define _mm256_maskz_minmax_ps(U, A, B, C)                                     \
-  ((__m256)__builtin_ia32_vminmaxps256_round_mask(                             \
+  ((__m256)__builtin_ia32_vminmaxps256_mask(                                   \
       (__v8sf)(__m256)(A), (__v8sf)(__m256)(B), (int)(C),                      \
-      (__v8sf)_mm256_setzero_ps(), (__mmask8)(U), _MM_FROUND_NO_EXC))
-
-#define _mm256_minmax_round_ps(A, B, C, R)                                     \
-  ((__m256)__builtin_ia32_vminmaxps256_round_mask(                             \
-      (__v8sf)(__m256)(A), (__v8sf)(__m256)(B), (int)(C),                      \
-      (__v8sf)_mm256_undefined_ps(), (__mmask8)-1, (int)(R)))
-
-#define _mm256_mask_minmax_round_ps(W, U, A, B, C, R)                          \
-  ((__m256)__builtin_ia32_vminmaxps256_round_mask(                             \
-      (__v8sf)(__m256)(A), (__v8sf)(__m256)(B), (int)(C), (__v8sf)(__m256)(W), \
-      (__mmask8)(U), (int)(R)))
-
-#define _mm256_maskz_minmax_round_ps(U, A, B, C, R)                            \
-  ((__m256)__builtin_ia32_vminmaxps256_round_mask(                             \
-      (__v8sf)(__m256)(A), (__v8sf)(__m256)(B), (int)(C),                      \
-      (__v8sf)_mm256_setzero_ps(), (__mmask8)(U), (int)(R)))
+      (__v8sf)_mm256_setzero_ps(), (__mmask8)(U)))
 
 #define _mm_minmax_sd(A, B, C)                                                 \
   ((__m128d)__builtin_ia32_vminmaxsd_round_mask(                               \

lib/include/avx10_2niintrin.h

@@ -402,1672 +402,6 @@ static __inline__ __m256i __DEFAULT_FN_ATTRS256 _mm256_maskz_dpwuuds_epi32(
       (__v8si)_mm256_setzero_si256());
 }
 
-/* YMM Rounding */
-#define _mm256_add_round_pd(A, B, R)                                           \
-  ((__m256d)__builtin_ia32_vaddpd256_round((__v4df)(__m256d)(A),               \
-                                           (__v4df)(__m256d)(B), (int)(R)))
-
-#define _mm256_mask_add_round_pd(W, U, A, B, R)                                \
-  ((__m256d)__builtin_ia32_selectpd_256(                                       \
-      (__mmask8)(U), (__v4df)_mm256_add_round_pd((A), (B), (R)),               \
-      (__v4df)(__m256d)(W)))
-
-#define _mm256_maskz_add_round_pd(U, A, B, R)                                  \
-  ((__m256d)__builtin_ia32_selectpd_256(                                       \
-      (__mmask8)(U), (__v4df)_mm256_add_round_pd((A), (B), (R)),               \
-      (__v4df)_mm256_setzero_pd()))
-
-#define _mm256_add_round_ph(A, B, R)                                           \
-  ((__m256h)__builtin_ia32_vaddph256_round((__v16hf)(__m256h)(A),              \
-                                           (__v16hf)(__m256h)(B), (int)(R)))
-
-#define _mm256_mask_add_round_ph(W, U, A, B, R)                                \
-  ((__m256h)__builtin_ia32_selectph_256(                                       \
-      (__mmask16)(U), (__v16hf)_mm256_add_round_ph((A), (B), (R)),             \
-      (__v16hf)(__m256h)(W)))
-
-#define _mm256_maskz_add_round_ph(U, A, B, R)                                  \
-  ((__m256h)__builtin_ia32_selectph_256(                                       \
-      (__mmask16)(U), (__v16hf)_mm256_add_round_ph((A), (B), (R)),             \
-      (__v16hf)_mm256_setzero_ph()))
-
-#define _mm256_add_round_ps(A, B, R)                                           \
-  ((__m256)__builtin_ia32_vaddps256_round((__v8sf)(__m256)(A),                 \
-                                          (__v8sf)(__m256)(B), (int)(R)))
-
-#define _mm256_mask_add_round_ps(W, U, A, B, R)                                \
-  ((__m256)__builtin_ia32_selectps_256(                                        \
-      (__mmask8)(U), (__v8sf)_mm256_add_round_ps((A), (B), (R)),               \
-      (__v8sf)(__m256)(W)))
-
-#define _mm256_maskz_add_round_ps(U, A, B, R)                                  \
-  ((__m256)__builtin_ia32_selectps_256(                                        \
-      (__mmask8)(U), (__v8sf)_mm256_add_round_ps((A), (B), (R)),               \
-      (__v8sf)_mm256_setzero_ps()))
-
-#define _mm256_cmp_round_pd_mask(A, B, P, R)                                   \
-  ((__mmask8)__builtin_ia32_vcmppd256_round_mask(                              \
-      (__v4df)(__m256d)(A), (__v4df)(__m256d)(B), (int)(P), (__mmask8)-1,      \
-      (int)(R)))
-
-#define _mm256_mask_cmp_round_pd_mask(U, A, B, P, R)                           \
-  ((__mmask8)__builtin_ia32_vcmppd256_round_mask(                              \
-      (__v4df)(__m256d)(A), (__v4df)(__m256d)(B), (int)(P), (__mmask8)(U),     \
-      (int)(R)))
-
-#define _mm256_cmp_round_ph_mask(A, B, P, R)                                   \
-  ((__mmask16)__builtin_ia32_vcmpph256_round_mask(                             \
-      (__v16hf)(__m256h)(A), (__v16hf)(__m256h)(B), (int)(P), (__mmask16)-1,   \
-      (int)(R)))
-
-#define _mm256_mask_cmp_round_ph_mask(U, A, B, P, R)                           \
-  ((__mmask16)__builtin_ia32_vcmpph256_round_mask(                             \
-      (__v16hf)(__m256h)(A), (__v16hf)(__m256h)(B), (int)(P), (__mmask16)(U),  \
-      (int)(R)))
-
-#define _mm256_cmp_round_ps_mask(A, B, P, R)                                   \
-  ((__mmask8)__builtin_ia32_vcmpps256_round_mask(                              \
-      (__v8sf)(__m256)(A), (__v8sf)(__m256)(B), (int)(P), (__mmask8)-1,        \
-      (int)(R)))
-
-#define _mm256_mask_cmp_round_ps_mask(U, A, B, P, R)                           \
-  ((__mmask8)__builtin_ia32_vcmpps256_round_mask(                              \
-      (__v8sf)(__m256)(A), (__v8sf)(__m256)(B), (int)(P), (__mmask8)(U),       \
-      (int)(R)))
-
-#define _mm256_cvt_roundepi32_ph(A, R)                                         \
-  ((__m128h)__builtin_ia32_vcvtdq2ph256_round_mask(                            \
-      (__v8si)(A), (__v8hf)_mm_undefined_ph(), (__mmask8)(-1), (int)(R)))
-
-#define _mm256_mask_cvt_roundepi32_ph(W, U, A, R)                              \
-  ((__m128h)__builtin_ia32_vcvtdq2ph256_round_mask((__v8si)(A), (__v8hf)(W),   \
-                                                   (__mmask8)(U), (int)(R)))
-
-#define _mm256_maskz_cvt_roundepi32_ph(U, A, R)                                \
-  ((__m128h)__builtin_ia32_vcvtdq2ph256_round_mask(                            \
-      (__v8si)(A), (__v8hf)_mm_setzero_ph(), (__mmask8)(U), (int)(R)))
-
-#define _mm256_cvt_roundepi32_ps(A, R)                                         \
-  ((__m256)__builtin_ia32_vcvtdq2ps256_round_mask((__v8si)(__m256i)(A),        \
-                                                  (__v8sf)_mm256_setzero_ps(), \
-                                                  (__mmask8)-1, (int)(R)))
-
-#define _mm256_mask_cvt_roundepi32_ps(W, U, A, R)                              \
-  ((__m256)__builtin_ia32_vcvtdq2ps256_round_mask(                             \
-      (__v8si)(__m256i)(A), (__v8sf)(__m256)(W), (__mmask8)(U), (int)(R)))
-
-#define _mm256_maskz_cvt_roundepi32_ps(U, A, R)                                \
-  ((__m256)__builtin_ia32_vcvtdq2ps256_round_mask((__v8si)(__m256i)(A),        \
-                                                  (__v8sf)_mm256_setzero_ps(), \
-                                                  (__mmask8)(U), (int)(R)))
-
-#define _mm256_cvt_roundpd_epi32(A, R)                                         \
-  ((__m128i)__builtin_ia32_vcvtpd2dq256_round_mask(                            \
-      (__v4df)(__m256d)(A), (__v4si)_mm_setzero_si128(), (__mmask8)-1,         \
-      (int)(R)))
-
-#define _mm256_mask_cvt_roundpd_epi32(W, U, A, R)                              \
-  ((__m128i)__builtin_ia32_vcvtpd2dq256_round_mask(                            \
-      (__v4df)(__m256d)(A), (__v4si)(__m128i)(W), (__mmask8)(U), (int)(R)))
-
-#define _mm256_maskz_cvt_roundpd_epi32(U, A, R)                                \
-  ((__m128i)__builtin_ia32_vcvtpd2dq256_round_mask(                            \
-      (__v4df)(__m256d)(A), (__v4si)_mm_setzero_si128(), (__mmask8)(U),        \
-      (int)(R)))
-
-#define _mm256_cvt_roundpd_ph(A, R)                                            \
-  ((__m128h)__builtin_ia32_vcvtpd2ph256_round_mask(                            \
-      (__v4df)(A), (__v8hf)_mm_undefined_ph(), (__mmask8)(-1), (int)(R)))
-
-#define _mm256_mask_cvt_roundpd_ph(W, U, A, R)                                 \
-  ((__m128h)__builtin_ia32_vcvtpd2ph256_round_mask((__v4df)(A), (__v8hf)(W),   \
-                                                   (__mmask8)(U), (int)(R)))
-
-#define _mm256_maskz_cvt_roundpd_ph(U, A, R)                                   \
-  ((__m128h)__builtin_ia32_vcvtpd2ph256_round_mask(                            \
-      (__v4df)(A), (__v8hf)_mm_setzero_ph(), (__mmask8)(U), (int)(R)))
-
-#define _mm256_cvt_roundpd_ps(A, R)                                            \
-  ((__m128)__builtin_ia32_vcvtpd2ps256_round_mask(                             \
-      (__v4df)(__m256d)(A), (__v4sf)_mm_setzero_ps(), (__mmask8)-1, (int)(R)))
-
-#define _mm256_mask_cvt_roundpd_ps(W, U, A, R)                                 \
-  ((__m128)__builtin_ia32_vcvtpd2ps256_round_mask(                             \
-      (__v4df)(__m256d)(A), (__v4sf)(__m128)(W), (__mmask8)(U), (int)(R)))
-
-#define _mm256_maskz_cvt_roundpd_ps(U, A, R)                                   \
-  ((__m128)__builtin_ia32_vcvtpd2ps256_round_mask((__v4df)(__m256d)(A),        \
-                                                  (__v4sf)_mm_setzero_ps(),    \
-                                                  (__mmask8)(U), (int)(R)))
-
-#define _mm256_cvt_roundpd_epi64(A, R)                                         \
-  ((__m256i)__builtin_ia32_vcvtpd2qq256_round_mask(                            \
-      (__v4df)(__m256d)(A), (__v4di)_mm256_setzero_si256(), (__mmask8)-1,      \
-      (int)(R)))
-
-#define _mm256_mask_cvt_roundpd_epi64(W, U, A, R)                              \
-  ((__m256i)__builtin_ia32_vcvtpd2qq256_round_mask(                            \
-      (__v4df)(__m256d)(A), (__v4di)(__m256i)(W), (__mmask8)(U), (int)(R)))
-
-#define _mm256_maskz_cvt_roundpd_epi64(U, A, R)                                \
-  ((__m256i)__builtin_ia32_vcvtpd2qq256_round_mask(                            \
-      (__v4df)(__m256d)(A), (__v4di)_mm256_setzero_si256(), (__mmask8)(U),     \
-      (int)(R)))
-
-#define _mm256_cvt_roundpd_epu32(A, R)                                         \
-  ((__m128i)__builtin_ia32_vcvtpd2udq256_round_mask(                           \
-      (__v4df)(__m256d)(A), (__v4su)_mm_setzero_si128(), (__mmask8)-1,         \
-      (int)(R)))
-
-#define _mm256_mask_cvt_roundpd_epu32(W, U, A, R)                              \
-  ((__m128i)__builtin_ia32_vcvtpd2udq256_round_mask(                           \
-      (__v4df)(__m256d)(A), (__v4su)(__m128i)(W), (__mmask8)(U), (int)(R)))
-
-#define _mm256_maskz_cvt_roundpd_epu32(U, A, R)                                \
-  ((__m128i)__builtin_ia32_vcvtpd2udq256_round_mask(                           \
-      (__v4df)(__m256d)(A), (__v4su)_mm_setzero_si128(), (__mmask8)(U),        \
-      (int)(R)))
-
-#define _mm256_cvt_roundpd_epu64(A, R)                                         \
-  ((__m256i)__builtin_ia32_vcvtpd2uqq256_round_mask(                           \
-      (__v4df)(__m256d)(A), (__v4du)_mm256_setzero_si256(), (__mmask8)-1,      \
-      (int)(R)))
-
-#define _mm256_mask_cvt_roundpd_epu64(W, U, A, R)                              \
-  ((__m256i)__builtin_ia32_vcvtpd2uqq256_round_mask(                           \
-      (__v4df)(__m256d)(A), (__v4du)(__m256i)(W), (__mmask8)(U), (int)(R)))
-
-#define _mm256_maskz_cvt_roundpd_epu64(U, A, R)                                \
-  ((__m256i)__builtin_ia32_vcvtpd2uqq256_round_mask(                           \
-      (__v4df)(__m256d)(A), (__v4du)_mm256_setzero_si256(), (__mmask8)(U),     \
-      (int)(R)))
-
-#define _mm256_cvt_roundph_epi32(A, R)                                         \
-  ((__m256i)__builtin_ia32_vcvtph2dq256_round_mask(                            \
-      (__v8hf)(A), (__v8si)_mm256_undefined_si256(), (__mmask8)(-1),           \
-      (int)(R)))
-
-#define _mm256_mask_cvt_roundph_epi32(W, U, A, R)                              \
-  ((__m256i)__builtin_ia32_vcvtph2dq256_round_mask((__v8hf)(A), (__v8si)(W),   \
-                                                   (__mmask8)(U), (int)(R)))
-
-#define _mm256_maskz_cvt_roundph_epi32(U, A, R)                                \
-  ((__m256i)__builtin_ia32_vcvtph2dq256_round_mask(                            \
-      (__v8hf)(A), (__v8si)_mm256_setzero_si256(), (__mmask8)(U), (int)(R)))
-
-#define _mm256_cvt_roundph_pd(A, R)                                            \
-  ((__m256d)__builtin_ia32_vcvtph2pd256_round_mask(                            \
-      (__v8hf)(A), (__v4df)_mm256_undefined_pd(), (__mmask8)(-1), (int)(R)))
-
-#define _mm256_mask_cvt_roundph_pd(W, U, A, R)                                 \
-  ((__m256d)__builtin_ia32_vcvtph2pd256_round_mask((__v8hf)(A), (__v4df)(W),   \
-                                                   (__mmask8)(U), (int)(R)))
-
-#define _mm256_maskz_cvt_roundph_pd(U, A, R)                                   \
-  ((__m256d)__builtin_ia32_vcvtph2pd256_round_mask(                            \
-      (__v8hf)(A), (__v4df)_mm256_setzero_pd(), (__mmask8)(U), (int)(R)))
-
-#define _mm256_cvtx_roundph_ps(A, R)                                           \
-  ((__m256)__builtin_ia32_vcvtph2psx256_round_mask(                            \
-      (__v8hf)(A), (__v8sf)_mm256_undefined_ps(), (__mmask8)(-1), (int)(R)))
-
-#define _mm256_mask_cvtx_roundph_ps(W, U, A, R)                                \
-  ((__m256)__builtin_ia32_vcvtph2psx256_round_mask((__v8hf)(A), (__v8sf)(W),   \
-                                                   (__mmask8)(U), (int)(R)))
-
-#define _mm256_maskz_cvtx_roundph_ps(U, A, R)                                  \
-  ((__m256)__builtin_ia32_vcvtph2psx256_round_mask(                            \
-      (__v8hf)(A), (__v8sf)_mm256_setzero_ps(), (__mmask8)(U), (int)(R)))
-
-#define _mm256_cvt_roundph_epi64(A, R)                                         \
-  ((__m256i)__builtin_ia32_vcvtph2qq256_round_mask(                            \
-      (__v8hf)(A), (__v4di)_mm256_undefined_si256(), (__mmask8)(-1),           \
-      (int)(R)))
-
-#define _mm256_mask_cvt_roundph_epi64(W, U, A, R)                              \
-  ((__m256i)__builtin_ia32_vcvtph2qq256_round_mask((__v8hf)(A), (__v4di)(W),   \
-                                                   (__mmask8)(U), (int)(R)))
-
-#define _mm256_maskz_cvt_roundph_epi64(U, A, R)                                \
-  ((__m256i)__builtin_ia32_vcvtph2qq256_round_mask(                            \
-      (__v8hf)(A), (__v4di)_mm256_setzero_si256(), (__mmask8)(U), (int)(R)))
-
-#define _mm256_cvt_roundph_epu32(A, R)                                         \
-  ((__m256i)__builtin_ia32_vcvtph2udq256_round_mask(                           \
-      (__v8hf)(A), (__v8su)_mm256_undefined_si256(), (__mmask8)(-1),           \
-      (int)(R)))
-
-#define _mm256_mask_cvt_roundph_epu32(W, U, A, R)                              \
-  ((__m256i)__builtin_ia32_vcvtph2udq256_round_mask((__v8hf)(A), (__v8su)(W),  \
-                                                    (__mmask8)(U), (int)(R)))
-
-#define _mm256_maskz_cvt_roundph_epu32(U, A, R)                                \
-  ((__m256i)__builtin_ia32_vcvtph2udq256_round_mask(                           \
-      (__v8hf)(A), (__v8su)_mm256_setzero_si256(), (__mmask8)(U), (int)(R)))
-
-#define _mm256_cvt_roundph_epu64(A, R)                                         \
-  ((__m256i)__builtin_ia32_vcvtph2uqq256_round_mask(                           \
-      (__v8hf)(A), (__v4du)_mm256_undefined_si256(), (__mmask8)(-1),           \
-      (int)(R)))
-
-#define _mm256_mask_cvt_roundph_epu64(W, U, A, R)                              \
-  ((__m256i)__builtin_ia32_vcvtph2uqq256_round_mask((__v8hf)(A), (__v4du)(W),  \
-                                                    (__mmask8)(U), (int)(R)))
-
-#define _mm256_maskz_cvt_roundph_epu64(U, A, R)                                \
-  ((__m256i)__builtin_ia32_vcvtph2uqq256_round_mask(                           \
-      (__v8hf)(A), (__v4du)_mm256_setzero_si256(), (__mmask8)(U), (int)(R)))
-
-#define _mm256_cvt_roundph_epu16(A, R)                                         \
-  ((__m256i)__builtin_ia32_vcvtph2uw256_round_mask(                            \
-      (__v16hf)(A), (__v16hu)_mm256_undefined_si256(), (__mmask16)(-1),        \
-      (int)(R)))
-
-#define _mm256_mask_cvt_roundph_epu16(W, U, A, R)                              \
-  ((__m256i)__builtin_ia32_vcvtph2uw256_round_mask((__v16hf)(A), (__v16hu)(W), \
-                                                   (__mmask16)(U), (int)(R)))
-
-#define _mm256_maskz_cvt_roundph_epu16(U, A, R)                                \
-  ((__m256i)__builtin_ia32_vcvtph2uw256_round_mask(                            \
-      (__v16hf)(A), (__v16hu)_mm256_setzero_si256(), (__mmask16)(U),           \
-      (int)(R)))
-
-#define _mm256_cvt_roundph_epi16(A, R)                                         \
-  ((__m256i)__builtin_ia32_vcvtph2w256_round_mask(                             \
-      (__v16hf)(A), (__v16hi)_mm256_undefined_si256(), (__mmask16)(-1),        \
-      (int)(R)))
-
-#define _mm256_mask_cvt_roundph_epi16(W, U, A, R)                              \
-  ((__m256i)__builtin_ia32_vcvtph2w256_round_mask((__v16hf)(A), (__v16hi)(W),  \
-                                                  (__mmask16)(U), (int)(R)))
-
-#define _mm256_maskz_cvt_roundph_epi16(U, A, R)                                \
-  ((__m256i)__builtin_ia32_vcvtph2w256_round_mask(                             \
-      (__v16hf)(A), (__v16hi)_mm256_setzero_si256(), (__mmask16)(U),           \
-      (int)(R)))
-
-#define _mm256_cvt_roundps_epi32(A, R)                                         \
-  ((__m256i)__builtin_ia32_vcvtps2dq256_round_mask(                            \
-      (__v8sf)(__m256)(A), (__v8si)_mm256_setzero_si256(), (__mmask8)-1,       \
-      (int)(R)))
-
-#define _mm256_mask_cvt_roundps_epi32(W, U, A, R)                              \
-  ((__m256i)__builtin_ia32_vcvtps2dq256_round_mask(                            \
-      (__v8sf)(__m256)(A), (__v8si)(__m256i)(W), (__mmask8)(U), (int)(R)))
-
-#define _mm256_maskz_cvt_roundps_epi32(U, A, R)                                \
-  ((__m256i)__builtin_ia32_vcvtps2dq256_round_mask(                            \
-      (__v8sf)(__m256)(A), (__v8si)_mm256_setzero_si256(), (__mmask8)(U),      \
-      (int)(R)))
-
-#define _mm256_cvt_roundps_pd(A, R)                                            \
-  ((__m256d)__builtin_ia32_vcvtps2pd256_round_mask(                            \
-      (__v4sf)(__m128)(A), (__v4df)_mm256_undefined_pd(), (__mmask8)-1,        \
-      (int)(R)))
-
-#define _mm256_mask_cvt_roundps_pd(W, U, A, R)                                 \
-  ((__m256d)__builtin_ia32_vcvtps2pd256_round_mask(                            \
-      (__v4sf)(__m128)(A), (__v4df)(__m256d)(W), (__mmask8)(U), (int)(R)))
-
-#define _mm256_maskz_cvt_roundps_pd(U, A, R)                                   \
-  ((__m256d)__builtin_ia32_vcvtps2pd256_round_mask(                            \
-      (__v4sf)(__m128)(A), (__v4df)_mm256_setzero_pd(), (__mmask8)(U),         \
-      (int)(R)))
-
-#define _mm256_cvt_roundps_ph(A, I)                                            \
-  ((__m128i)__builtin_ia32_vcvtps2ph256_mask((__v8sf)(__m256)(A), (int)(I),    \
-                                             (__v8hi)_mm_undefined_si128(),    \
-                                             (__mmask8)-1))
-
-/* FIXME: We may use these way in future.
-#define _mm256_cvt_roundps_ph(A, I)                                            \
-  ((__m128i)__builtin_ia32_vcvtps2ph256_round_mask(                            \
-      (__v8sf)(__m256)(A), (int)(I), (__v8hi)_mm_undefined_si128(),            \
-      (__mmask8)-1))
-#define _mm256_mask_cvt_roundps_ph(U, W, A, I)                                 \
-  ((__m128i)__builtin_ia32_vcvtps2ph256_round_mask(                            \
-      (__v8sf)(__m256)(A), (int)(I), (__v8hi)(__m128i)(U), (__mmask8)(W)))
-#define _mm256_maskz_cvt_roundps_ph(W, A, I)                                   \
-  ((__m128i)__builtin_ia32_vcvtps2ph256_round_mask(                            \
-      (__v8sf)(__m256)(A), (int)(I), (__v8hi)_mm_setzero_si128(),              \
-      (__mmask8)(W))) */
-
-#define _mm256_cvtx_roundps_ph(A, R)                                           \
-  ((__m128h)__builtin_ia32_vcvtps2phx256_round_mask(                           \
-      (__v8sf)(A), (__v8hf)_mm_undefined_ph(), (__mmask8)(-1), (int)(R)))
-
-#define _mm256_mask_cvtx_roundps_ph(W, U, A, R)                                \
-  ((__m128h)__builtin_ia32_vcvtps2phx256_round_mask((__v8sf)(A), (__v8hf)(W),  \
-                                                    (__mmask8)(U), (int)(R)))
-
-#define _mm256_maskz_cvtx_roundps_ph(U, A, R)                                  \
-  ((__m128h)__builtin_ia32_vcvtps2phx256_round_mask(                           \
-      (__v8sf)(A), (__v8hf)_mm_setzero_ph(), (__mmask8)(U), (int)(R)))
-
-#define _mm256_cvt_roundps_epi64(A, R)                                         \
-  ((__m256i)__builtin_ia32_vcvtps2qq256_round_mask(                            \
-      (__v4sf)(__m128)(A), (__v4di)_mm256_setzero_si256(), (__mmask8)-1,       \
-      (int)(R)))
-
-#define _mm256_mask_cvt_roundps_epi64(W, U, A, R)                              \
-  ((__m256i)__builtin_ia32_vcvtps2qq256_round_mask(                            \
-      (__v4sf)(__m128)(A), (__v4di)(__m256i)(W), (__mmask8)(U), (int)(R)))
-
-#define _mm256_maskz_cvt_roundps_epi64(U, A, R)                                \
-  ((__m256i)__builtin_ia32_vcvtps2qq256_round_mask(                            \
-      (__v4sf)(__m128)(A), (__v4di)_mm256_setzero_si256(), (__mmask8)(U),      \
-      (int)(R)))
-
-#define _mm256_cvt_roundps_epu32(A, R)                                         \
-  ((__m256i)__builtin_ia32_vcvtps2udq256_round_mask(                           \
-      (__v8sf)(__m256)(A), (__v8su)_mm256_setzero_si256(), (__mmask8)-1,       \
-      (int)(R)))
-
-#define _mm256_mask_cvt_roundps_epu32(W, U, A, R)                              \
-  ((__m256i)__builtin_ia32_vcvtps2udq256_round_mask(                           \
-      (__v8sf)(__m256)(A), (__v8su)(__m256i)(W), (__mmask8)(U), (int)(R)))
-
-#define _mm256_maskz_cvt_roundps_epu32(U, A, R)                                \
-  ((__m256i)__builtin_ia32_vcvtps2udq256_round_mask(                           \
-      (__v8sf)(__m256)(A), (__v8su)_mm256_setzero_si256(), (__mmask8)(U),      \
-      (int)(R)))
-
-#define _mm256_cvt_roundps_epu64(A, R)                                         \
-  ((__m256i)__builtin_ia32_vcvtps2uqq256_round_mask(                           \
-      (__v4sf)(__m128)(A), (__v4du)_mm256_setzero_si256(), (__mmask8)-1,       \
-      (int)(R)))
-
-#define _mm256_mask_cvt_roundps_epu64(W, U, A, R)                              \
-  ((__m256i)__builtin_ia32_vcvtps2uqq256_round_mask(                           \
-      (__v4sf)(__m128)(A), (__v4du)(__m256i)(W), (__mmask8)(U), (int)(R)))
-
-#define _mm256_maskz_cvt_roundps_epu64(U, A, R)                                \
-  ((__m256i)__builtin_ia32_vcvtps2uqq256_round_mask(                           \
-      (__v4sf)(__m128)(A), (__v4du)_mm256_setzero_si256(), (__mmask8)(U),      \
-      (int)(R)))
-
-#define _mm256_cvt_roundepi64_pd(A, R)                                         \
-  ((__m256d)__builtin_ia32_vcvtqq2pd256_round_mask(                            \
-      (__v4di)(__m256i)(A), (__v4df)_mm256_setzero_pd(), (__mmask8)-1,         \
-      (int)(R)))
-
-#define _mm256_mask_cvt_roundepi64_pd(W, U, A, R)                              \
-  ((__m256d)__builtin_ia32_vcvtqq2pd256_round_mask(                            \
-      (__v4di)(__m256i)(A), (__v4df)(__m256d)(W), (__mmask8)(U), (int)(R)))
-
-#define _mm256_maskz_cvt_roundepi64_pd(U, A, R)                                \
-  ((__m256d)__builtin_ia32_vcvtqq2pd256_round_mask(                            \
-      (__v4di)(__m256i)(A), (__v4df)_mm256_setzero_pd(), (__mmask8)(U),        \
-      (int)(R)))
-
-#define _mm256_cvt_roundepi64_ph(A, R)                                         \
-  ((__m128h)__builtin_ia32_vcvtqq2ph256_round_mask(                            \
-      (__v4di)(A), (__v8hf)_mm_undefined_ph(), (__mmask8)(-1), (int)(R)))
-
-#define _mm256_mask_cvt_roundepi64_ph(W, U, A, R)                              \
-  ((__m128h)__builtin_ia32_vcvtqq2ph256_round_mask((__v4di)(A), (__v8hf)(W),   \
-                                                   (__mmask8)(U), (int)(R)))
-
-#define _mm256_maskz_cvt_roundepi64_ph(U, A, R)                                \
-  ((__m128h)__builtin_ia32_vcvtqq2ph256_round_mask(                            \
-      (__v4di)(A), (__v8hf)_mm_setzero_ph(), (__mmask8)(U), (int)(R)))
-
-#define _mm256_cvt_roundepi64_ps(A, R)                                         \
-  ((__m128)__builtin_ia32_vcvtqq2ps256_round_mask(                             \
-      (__v4di)(__m256i)(A), (__v4sf)_mm_setzero_ps(), (__mmask8)-1, (int)(R)))
-
-#define _mm256_mask_cvt_roundepi64_ps(W, U, A, R)                              \
-  ((__m128)__builtin_ia32_vcvtqq2ps256_round_mask(                             \
-      (__v4di)(__m256i)(A), (__v4sf)(__m128)(W), (__mmask8)(U), (int)(R)))
-
-#define _mm256_maskz_cvt_roundepi64_ps(U, A, R)                                \
-  ((__m128)__builtin_ia32_vcvtqq2ps256_round_mask((__v4di)(__m256i)(A),        \
-                                                  (__v4sf)_mm_setzero_ps(),    \
-                                                  (__mmask8)(U), (int)(R)))
-
-#define _mm256_cvtt_roundpd_epi32(A, R)                                        \
-  ((__m128i)__builtin_ia32_vcvttpd2dq256_round_mask(                           \
-      (__v4df)(__m256d)(A), (__v4si)_mm_setzero_si128(), (__mmask8)-1,         \
-      (int)(R)))
-
-#define _mm256_mask_cvtt_roundpd_epi32(W, U, A, R)                             \
-  ((__m128i)__builtin_ia32_vcvttpd2dq256_round_mask(                           \
-      (__v4df)(__m256d)(A), (__v4si)(__m128i)(W), (__mmask8)(U), (int)(R)))
-
-#define _mm256_maskz_cvtt_roundpd_epi32(U, A, R)                               \
-  ((__m128i)__builtin_ia32_vcvttpd2dq256_round_mask(                           \
-      (__v4df)(__m256d)(A), (__v4si)_mm_setzero_si128(), (__mmask8)(U),        \
-      (int)(R)))
-
-#define _mm256_cvtt_roundpd_epi64(A, R)                                        \
-  ((__m256i)__builtin_ia32_vcvttpd2qq256_round_mask(                           \
-      (__v4df)(__m256d)(A), (__v4di)_mm256_setzero_si256(), (__mmask8)-1,      \
-      (int)(R)))
-
-#define _mm256_mask_cvtt_roundpd_epi64(W, U, A, R)                             \
-  ((__m256i)__builtin_ia32_vcvttpd2qq256_round_mask(                           \
-      (__v4df)(__m256d)(A), (__v4di)(__m256i)(W), (__mmask8)(U), (int)(R)))
-
-#define _mm256_maskz_cvtt_roundpd_epi64(U, A, R)                               \
-  ((__m256i)__builtin_ia32_vcvttpd2qq256_round_mask(                           \
-      (__v4df)(__m256d)(A), (__v4di)_mm256_setzero_si256(), (__mmask8)(U),     \
-      (int)(R)))
-
-#define _mm256_cvtt_roundpd_epu32(A, R)                                        \
-  ((__m128i)__builtin_ia32_vcvttpd2udq256_round_mask(                          \
-      (__v4df)(__m256d)(A), (__v4su)_mm_setzero_si128(), (__mmask8)-1,         \
-      (int)(R)))
-
-#define _mm256_mask_cvtt_roundpd_epu32(W, U, A, R)                             \
-  ((__m128i)__builtin_ia32_vcvttpd2udq256_round_mask(                          \
-      (__v4df)(__m256d)(A), (__v4su)(__m128i)(W), (__mmask8)(U), (int)(R)))
-
-#define _mm256_maskz_cvtt_roundpd_epu32(U, A, R)                               \
-  ((__m128i)__builtin_ia32_vcvttpd2udq256_round_mask(                          \
-      (__v4df)(__m256d)(A), (__v4su)_mm_setzero_si128(), (__mmask8)(U),        \
-      (int)(R)))
-
-#define _mm256_cvtt_roundpd_epu64(A, R)                                        \
-  ((__m256i)__builtin_ia32_vcvttpd2uqq256_round_mask(                          \
-      (__v4df)(__m256d)(A), (__v4du)_mm256_setzero_si256(), (__mmask8)-1,      \
-      (int)(R)))
-
-#define _mm256_mask_cvtt_roundpd_epu64(W, U, A, R)                             \
-  ((__m256i)__builtin_ia32_vcvttpd2uqq256_round_mask(                          \
-      (__v4df)(__m256d)(A), (__v4du)(__m256i)(W), (__mmask8)(U), (int)(R)))
-
-#define _mm256_maskz_cvtt_roundpd_epu64(U, A, R)                               \
-  ((__m256i)__builtin_ia32_vcvttpd2uqq256_round_mask(                          \
-      (__v4df)(__m256d)(A), (__v4du)_mm256_setzero_si256(), (__mmask8)(U),     \
-      (int)(R)))
-
-#define _mm256_cvtt_roundph_epi32(A, R)                                        \
-  ((__m256i)__builtin_ia32_vcvttph2dq256_round_mask(                           \
-      (__v8hf)(A), (__v8si)_mm256_undefined_si256(), (__mmask8)(-1),           \
-      (int)(R)))
-
-#define _mm256_mask_cvtt_roundph_epi32(W, U, A, R)                             \
-  ((__m256i)__builtin_ia32_vcvttph2dq256_round_mask((__v8hf)(A), (__v8si)(W),  \
-                                                    (__mmask8)(U), (int)(R)))
-
-#define _mm256_maskz_cvtt_roundph_epi32(U, A, R)                               \
-  ((__m256i)__builtin_ia32_vcvttph2dq256_round_mask(                           \
-      (__v8hf)(A), (__v8si)_mm256_setzero_si256(), (__mmask8)(U), (int)(R)))
-
-#define _mm256_cvtt_roundph_epi64(A, R)                                        \
-  ((__m256i)__builtin_ia32_vcvttph2qq256_round_mask(                           \
-      (__v8hf)(A), (__v4di)_mm256_undefined_si256(), (__mmask8)(-1),           \
-      (int)(R)))
-
-#define _mm256_mask_cvtt_roundph_epi64(W, U, A, R)                             \
-  ((__m256i)__builtin_ia32_vcvttph2qq256_round_mask((__v8hf)(A), (__v4di)(W),  \
-                                                    (__mmask8)(U), (int)(R)))
-
-#define _mm256_maskz_cvtt_roundph_epi64(U, A, R)                               \
-  ((__m256i)__builtin_ia32_vcvttph2qq256_round_mask(                           \
-      (__v8hf)(A), (__v4di)_mm256_setzero_si256(), (__mmask8)(U), (int)(R)))
-
-#define _mm256_cvtt_roundph_epu32(A, R)                                        \
-  ((__m256i)__builtin_ia32_vcvttph2udq256_round_mask(                          \
-      (__v8hf)(A), (__v8su)_mm256_undefined_si256(), (__mmask8)(-1),           \
-      (int)(R)))
-
-#define _mm256_mask_cvtt_roundph_epu32(W, U, A, R)                             \
-  ((__m256i)__builtin_ia32_vcvttph2udq256_round_mask((__v8hf)(A), (__v8su)(W), \
-                                                     (__mmask8)(U), (int)(R)))
-
-#define _mm256_maskz_cvtt_roundph_epu32(U, A, R)                               \
-  ((__m256i)__builtin_ia32_vcvttph2udq256_round_mask(                          \
-      (__v8hf)(A), (__v8su)_mm256_setzero_si256(), (__mmask8)(U), (int)(R)))
-
-#define _mm256_cvtt_roundph_epu64(A, R)                                        \
-  ((__m256i)__builtin_ia32_vcvttph2uqq256_round_mask(                          \
-      (__v8hf)(A), (__v4du)_mm256_undefined_si256(), (__mmask8)(-1),           \
-      (int)(R)))
-
-#define _mm256_mask_cvtt_roundph_epu64(W, U, A, R)                             \
-  ((__m256i)__builtin_ia32_vcvttph2uqq256_round_mask((__v8hf)(A), (__v4du)(W), \
-                                                     (__mmask8)(U), (int)(R)))
-
-#define _mm256_maskz_cvtt_roundph_epu64(U, A, R)                               \
-  ((__m256i)__builtin_ia32_vcvttph2uqq256_round_mask(                          \
-      (__v8hf)(A), (__v4du)_mm256_setzero_si256(), (__mmask8)(U), (int)(R)))
-
-#define _mm256_cvtt_roundph_epu16(A, R)                                        \
-  ((__m256i)__builtin_ia32_vcvttph2uw256_round_mask(                           \
-      (__v16hf)(A), (__v16hu)_mm256_undefined_si256(), (__mmask16)(-1),        \
-      (int)(R)))
-
-#define _mm256_mask_cvtt_roundph_epu16(W, U, A, R)                             \
-  ((__m256i)__builtin_ia32_vcvttph2uw256_round_mask(                           \
-      (__v16hf)(A), (__v16hu)(W), (__mmask16)(U), (int)(R)))
-
-#define _mm256_maskz_cvtt_roundph_epu16(U, A, R)                               \
-  ((__m256i)__builtin_ia32_vcvttph2uw256_round_mask(                           \
-      (__v16hf)(A), (__v16hu)_mm256_setzero_si256(), (__mmask16)(U),           \
-      (int)(R)))
-
-#define _mm256_cvtt_roundph_epi16(A, R)                                        \
-  ((__m256i)__builtin_ia32_vcvttph2w256_round_mask(                            \
-      (__v16hf)(A), (__v16hi)_mm256_undefined_si256(), (__mmask16)(-1),        \
-      (int)(R)))
-
-#define _mm256_mask_cvtt_roundph_epi16(W, U, A, R)                             \
-  ((__m256i)__builtin_ia32_vcvttph2w256_round_mask((__v16hf)(A), (__v16hi)(W), \
-                                                   (__mmask16)(U), (int)(R)))
-
-#define _mm256_maskz_cvtt_roundph_epi16(U, A, R)                               \
-  ((__m256i)__builtin_ia32_vcvttph2w256_round_mask(                            \
-      (__v16hf)(A), (__v16hi)_mm256_setzero_si256(), (__mmask16)(U),           \
-      (int)(R)))
-
-#define _mm256_cvtt_roundps_epi32(A, R)                                        \
-  ((__m256i)__builtin_ia32_vcvttps2dq256_round_mask(                           \
-      (__v8sf)(__m256)(A), (__v8si)_mm256_setzero_si256(), (__mmask8)-1,       \
-      (int)(R)))
-
-#define _mm256_mask_cvtt_roundps_epi32(W, U, A, R)                             \
-  ((__m256i)__builtin_ia32_vcvttps2dq256_round_mask(                           \
-      (__v8sf)(__m256)(A), (__v8si)(__m256i)(W), (__mmask8)(U), (int)(R)))
-
-#define _mm256_maskz_cvtt_roundps_epi32(U, A, R)                               \
-  ((__m256i)__builtin_ia32_vcvttps2dq256_round_mask(                           \
-      (__v8sf)(__m256)(A), (__v8si)_mm256_setzero_si256(), (__mmask8)(U),      \
-      (int)(R)))
-
-#define _mm256_cvtt_roundps_epi64(A, R)                                        \
-  ((__m256i)__builtin_ia32_vcvttps2qq256_round_mask(                           \
-      (__v4sf)(__m128)(A), (__v4di)_mm256_setzero_si256(), (__mmask8)-1,       \
-      (int)(R)))
-
-#define _mm256_mask_cvtt_roundps_epi64(W, U, A, R)                             \
-  ((__m256i)__builtin_ia32_vcvttps2qq256_round_mask(                           \
-      (__v4sf)(__m128)(A), (__v4di)(__m256i)(W), (__mmask8)(U), (int)(R)))
-
-#define _mm256_maskz_cvtt_roundps_epi64(U, A, R)                               \
-  ((__m256i)__builtin_ia32_vcvttps2qq256_round_mask(                           \
-      (__v4sf)(__m128)(A), (__v4di)_mm256_setzero_si256(), (__mmask8)(U),      \
-      (int)(R)))
-
-#define _mm256_cvtt_roundps_epu32(A, R)                                        \
-  ((__m256i)__builtin_ia32_vcvttps2udq256_round_mask(                          \
-      (__v8sf)(__m256)(A), (__v8su)_mm256_setzero_si256(), (__mmask8)-1,       \
-      (int)(R)))
-
-#define _mm256_mask_cvtt_roundps_epu32(W, U, A, R)                             \
-  ((__m256i)__builtin_ia32_vcvttps2udq256_round_mask(                          \
-      (__v8sf)(__m256)(A), (__v8su)(__m256i)(W), (__mmask8)(U), (int)(R)))
-
-#define _mm256_maskz_cvtt_roundps_epu32(U, A, R)                               \
-  ((__m256i)__builtin_ia32_vcvttps2udq256_round_mask(                          \
-      (__v8sf)(__m256)(A), (__v8su)_mm256_setzero_si256(), (__mmask8)(U),      \
-      (int)(R)))
-
-#define _mm256_cvtt_roundps_epu64(A, R)                                        \
-  ((__m256i)__builtin_ia32_vcvttps2uqq256_round_mask(                          \
-      (__v4sf)(__m128)(A), (__v4du)_mm256_setzero_si256(), (__mmask8)-1,       \
-      (int)(R)))
-
-#define _mm256_mask_cvtt_roundps_epu64(W, U, A, R)                             \
-  ((__m256i)__builtin_ia32_vcvttps2uqq256_round_mask(                          \
-      (__v4sf)(__m128)(A), (__v4du)(__m256i)(W), (__mmask8)(U), (int)(R)))
-
-#define _mm256_maskz_cvtt_roundps_epu64(U, A, R)                               \
-  ((__m256i)__builtin_ia32_vcvttps2uqq256_round_mask(                          \
-      (__v4sf)(__m128)(A), (__v4du)_mm256_setzero_si256(), (__mmask8)(U),      \
-      (int)(R)))
-
-#define _mm256_cvt_roundepu32_ph(A, R)                                         \
-  ((__m128h)__builtin_ia32_vcvtudq2ph256_round_mask(                           \
-      (__v8su)(A), (__v8hf)_mm_undefined_ph(), (__mmask8)(-1), (int)(R)))
-
-#define _mm256_mask_cvt_roundepu32_ph(W, U, A, R)                              \
-  ((__m128h)__builtin_ia32_vcvtudq2ph256_round_mask((__v8su)(A), (__v8hf)(W),  \
-                                                    (__mmask8)(U), (int)(R)))
-
-#define _mm256_maskz_cvt_roundepu32_ph(U, A, R)                                \
-  ((__m128h)__builtin_ia32_vcvtudq2ph256_round_mask(                           \
-      (__v8su)(A), (__v8hf)_mm_setzero_ph(), (__mmask8)(U), (int)(R)))
-
-#define _mm256_cvt_roundepu32_ps(A, R)                                         \
-  ((__m256)__builtin_ia32_vcvtudq2ps256_round_mask(                            \
-      (__v8su)(__m256i)(A), (__v8sf)_mm256_setzero_ps(), (__mmask8)-1,         \
-      (int)(R)))
-
-#define _mm256_mask_cvt_roundepu32_ps(W, U, A, R)                              \
-  ((__m256)__builtin_ia32_vcvtudq2ps256_round_mask(                            \
-      (__v8su)(__m256i)(A), (__v8sf)(__m256)(W), (__mmask8)(U), (int)(R)))
-
-#define _mm256_maskz_cvt_roundepu32_ps(U, A, R)                                \
-  ((__m256)__builtin_ia32_vcvtudq2ps256_round_mask(                            \
-      (__v8su)(__m256i)(A), (__v8sf)_mm256_setzero_ps(), (__mmask8)(U),        \
-      (int)(R)))
-
-#define _mm256_cvt_roundepu64_pd(A, R)                                         \
-  ((__m256d)__builtin_ia32_vcvtuqq2pd256_round_mask(                           \
-      (__v4du)(__m256i)(A), (__v4df)_mm256_setzero_pd(), (__mmask8)-1,         \
-      (int)(R)))
-
-#define _mm256_mask_cvt_roundepu64_pd(W, U, A, R)                              \
-  ((__m256d)__builtin_ia32_vcvtuqq2pd256_round_mask(                           \
-      (__v4du)(__m256i)(A), (__v4df)(__m256d)(W), (__mmask8)(U), (int)(R)))
-
-#define _mm256_maskz_cvt_roundepu64_pd(U, A, R)                                \
-  ((__m256d)__builtin_ia32_vcvtuqq2pd256_round_mask(                           \
-      (__v4du)(__m256i)(A), (__v4df)_mm256_setzero_pd(), (__mmask8)(U),        \
-      (int)(R)))
-
-#define _mm256_cvt_roundepu64_ph(A, R)                                         \
-  ((__m128h)__builtin_ia32_vcvtuqq2ph256_round_mask(                           \
-      (__v4du)(A), (__v8hf)_mm_undefined_ph(), (__mmask8)(-1), (int)(R)))
-
-#define _mm256_mask_cvt_roundepu64_ph(W, U, A, R)                              \
-  ((__m128h)__builtin_ia32_vcvtuqq2ph256_round_mask((__v4du)(A), (__v8hf)(W),  \
-                                                    (__mmask8)(U), (int)(R)))
-
-#define _mm256_maskz_cvt_roundepu64_ph(U, A, R)                                \
-  ((__m128h)__builtin_ia32_vcvtuqq2ph256_round_mask(                           \
-      (__v4du)(A), (__v8hf)_mm_setzero_ph(), (__mmask8)(U), (int)(R)))
-
-#define _mm256_cvt_roundepu64_ps(A, R)                                         \
-  ((__m128)__builtin_ia32_vcvtuqq2ps256_round_mask(                            \
-      (__v4du)(__m256i)(A), (__v4sf)_mm_setzero_ps(), (__mmask8)-1, (int)(R)))
-
-#define _mm256_mask_cvt_roundepu64_ps(W, U, A, R)                              \
-  ((__m128)__builtin_ia32_vcvtuqq2ps256_round_mask(                            \
-      (__v4du)(__m256i)(A), (__v4sf)(__m128)(W), (__mmask8)(U), (int)(R)))
-
-#define _mm256_maskz_cvt_roundepu64_ps(U, A, R)                                \
-  ((__m128)__builtin_ia32_vcvtuqq2ps256_round_mask((__v4du)(__m256i)(A),       \
-                                                   (__v4sf)_mm_setzero_ps(),   \
-                                                   (__mmask8)(U), (int)(R)))
-
-#define _mm256_cvt_roundepu16_ph(A, R)                                         \
-  ((__m256h)__builtin_ia32_vcvtuw2ph256_round_mask(                            \
-      (__v16hu)(A), (__v16hf)_mm256_undefined_ph(), (__mmask16)(-1),           \
-      (int)(R)))
-
-#define _mm256_mask_cvt_roundepu16_ph(W, U, A, R)                              \
-  ((__m256h)__builtin_ia32_vcvtuw2ph256_round_mask((__v16hu)(A), (__v16hf)(W), \
-                                                   (__mmask16)(U), (int)(R)))
-
-#define _mm256_maskz_cvt_roundepu16_ph(U, A, R)                                \
-  ((__m256h)__builtin_ia32_vcvtuw2ph256_round_mask(                            \
-      (__v16hu)(A), (__v16hf)_mm256_setzero_ph(), (__mmask16)(U), (int)(R)))
-
-#define _mm256_cvt_roundepi16_ph(A, R)                                         \
-  ((__m256h)__builtin_ia32_vcvtw2ph256_round_mask(                             \
-      (__v16hi)(A), (__v16hf)_mm256_undefined_ph(), (__mmask16)(-1),           \
-      (int)(R)))
-
-#define _mm256_mask_cvt_roundepi16_ph(W, U, A, R)                              \
-  ((__m256h)__builtin_ia32_vcvtw2ph256_round_mask((__v16hi)(A), (__v16hf)(W),  \
-                                                  (__mmask16)(U), (int)(R)))
-
-#define _mm256_maskz_cvt_roundepi16_ph(U, A, R)                                \
-  ((__m256h)__builtin_ia32_vcvtw2ph256_round_mask(                             \
-      (__v16hi)(A), (__v16hf)_mm256_setzero_ph(), (__mmask16)(U), (int)(R)))
-
-#define _mm256_div_round_pd(A, B, R)                                           \
-  ((__m256d)__builtin_ia32_vdivpd256_round((__v4df)(__m256d)(A),               \
-                                           (__v4df)(__m256d)(B), (int)(R)))
-
-#define _mm256_mask_div_round_pd(W, U, A, B, R)                                \
-  ((__m256d)__builtin_ia32_selectpd_256(                                       \
-      (__mmask8)(U), (__v4df)_mm256_div_round_pd((A), (B), (R)),               \
-      (__v4df)(__m256d)(W)))
-
-#define _mm256_maskz_div_round_pd(U, A, B, R)                                  \
-  ((__m256d)__builtin_ia32_selectpd_256(                                       \
-      (__mmask8)(U), (__v4df)_mm256_div_round_pd((A), (B), (R)),               \
-      (__v4df)_mm256_setzero_pd()))
-
-#define _mm256_div_round_ph(A, B, R)                                           \
-  ((__m256h)__builtin_ia32_vdivph256_round((__v16hf)(__m256h)(A),              \
-                                           (__v16hf)(__m256h)(B), (int)(R)))
-
-#define _mm256_mask_div_round_ph(W, U, A, B, R)                                \
-  ((__m256h)__builtin_ia32_selectph_256(                                       \
-      (__mmask16)(U), (__v16hf)_mm256_div_round_ph((A), (B), (R)),             \
-      (__v16hf)(__m256h)(W)))
-
-#define _mm256_maskz_div_round_ph(U, A, B, R)                                  \
-  ((__m256h)__builtin_ia32_selectph_256(                                       \
-      (__mmask16)(U), (__v16hf)_mm256_div_round_ph((A), (B), (R)),             \
-      (__v16hf)_mm256_setzero_ph()))
-
-#define _mm256_div_round_ps(A, B, R)                                           \
-  ((__m256)__builtin_ia32_vdivps256_round((__v8sf)(__m256)(A),                 \
-                                          (__v8sf)(__m256)(B), (int)(R)))
-
-#define _mm256_mask_div_round_ps(W, U, A, B, R)                                \
-  ((__m256)__builtin_ia32_selectps_256(                                        \
-      (__mmask8)(U), (__v8sf)_mm256_div_round_ps((A), (B), (R)),               \
-      (__v8sf)(__m256)(W)))
-
-#define _mm256_maskz_div_round_ps(U, A, B, R)                                  \
-  ((__m256)__builtin_ia32_selectps_256(                                        \
-      (__mmask8)(U), (__v8sf)_mm256_div_round_ps((A), (B), (R)),               \
-      (__v8sf)_mm256_setzero_ps()))
-
-#define _mm256_fcmadd_round_pch(A, B, C, R)                                    \
-  ((__m256h)__builtin_ia32_vfcmaddcph256_round_mask3(                          \
-      (__v8sf)(__m256h)(A), (__v8sf)(__m256h)(B), (__v8sf)(__m256h)(C),        \
-      (__mmask8)-1, (int)(R)))
-
-#define _mm256_mask_fcmadd_round_pch(A, U, B, C, R)                            \
-  ((__m256h)__builtin_ia32_vfcmaddcph256_round_mask(                           \
-      (__v8sf)(__m256h)(A), (__v8sf)(__m256h)(B), (__v8sf)(__m256h)(C),        \
-      (__mmask8)(U), (int)(R)))
-
-#define _mm256_mask3_fcmadd_round_pch(A, B, C, U, R)                           \
-  ((__m256h)__builtin_ia32_vfcmaddcph256_round_mask3(                          \
-      (__v8sf)(__m256h)(A), (__v8sf)(__m256h)(B), (__v8sf)(__m256h)(C),        \
-      (__mmask8)(U), (int)(R)))
-
-#define _mm256_maskz_fcmadd_round_pch(U, A, B, C, R)                           \
-  ((__m256h)__builtin_ia32_vfcmaddcph256_round_maskz(                          \
-      (__v8sf)(__m256h)(A), (__v8sf)(__m256h)(B), (__v8sf)(__m256h)(C),        \
-      (__mmask8)(U), (int)(R)))
-
-#define _mm256_cmul_round_pch(A, B, R)                                         \
-  ((__m256h)__builtin_ia32_vfcmulcph256_round_mask(                            \
-      (__v8sf)(__m256h)(A), (__v8sf)(__m256h)(B),                              \
-      (__v8sf)(__m256h)_mm256_undefined_ph(), (__mmask8)-1, (int)(R)))
-
-#define _mm256_mask_cmul_round_pch(W, U, A, B, R)                              \
-  ((__m256h)__builtin_ia32_vfcmulcph256_round_mask(                            \
-      (__v8sf)(__m256h)(A), (__v8sf)(__m256h)(B), (__v8sf)(__m256h)(W),        \
-      (__mmask8)(U), (int)(R)))
-
-#define _mm256_maskz_cmul_round_pch(U, A, B, R)                                \
-  ((__m256h)__builtin_ia32_vfcmulcph256_round_mask(                            \
-      (__v8sf)(__m256h)(A), (__v8sf)(__m256h)(B),                              \
-      (__v8sf)(__m256h)_mm256_setzero_ph(), (__mmask8)(U), (int)(R)))
-
-#define _mm256_fixupimm_round_pd(A, B, C, imm, R)                              \
-  ((__m256d)__builtin_ia32_vfixupimmpd256_round_mask(                          \
-      (__v4df)(__m256d)(A), (__v4df)(__m256d)(B), (__v4di)(__m256i)(C),        \
-      (int)(imm), (__mmask8)-1, (int)(R)))
-
-#define _mm256_mask_fixupimm_round_pd(A, U, B, C, imm, R)                      \
-  ((__m256d)__builtin_ia32_vfixupimmpd256_round_mask(                          \
-      (__v4df)(__m256d)(A), (__v4df)(__m256d)(B), (__v4di)(__m256i)(C),        \
-      (int)(imm), (__mmask8)(U), (int)(R)))
-
-#define _mm256_maskz_fixupimm_round_pd(U, A, B, C, imm, R)                     \
-  ((__m256d)__builtin_ia32_vfixupimmpd256_round_maskz(                         \
-      (__v4df)(__m256d)(A), (__v4df)(__m256d)(B), (__v4di)(__m256i)(C),        \
-      (int)(imm), (__mmask8)(U), (int)(R)))
-
-#define _mm256_fixupimm_round_ps(A, B, C, imm, R)                              \
-  ((__m256)__builtin_ia32_vfixupimmps256_round_mask(                           \
-      (__v8sf)(__m256)(A), (__v8sf)(__m256)(B), (__v8si)(__m256i)(C),          \
-      (int)(imm), (__mmask8)-1, (int)(R)))
-
-#define _mm256_mask_fixupimm_round_ps(A, U, B, C, imm, R)                      \
-  ((__m256)__builtin_ia32_vfixupimmps256_round_mask(                           \
-      (__v8sf)(__m256)(A), (__v8sf)(__m256)(B), (__v8si)(__m256i)(C),          \
-      (int)(imm), (__mmask8)(U), (int)(R)))
-
-#define _mm256_maskz_fixupimm_round_ps(U, A, B, C, imm, R)                     \
-  ((__m256)__builtin_ia32_vfixupimmps256_round_maskz(                          \
-      (__v8sf)(__m256)(A), (__v8sf)(__m256)(B), (__v8si)(__m256i)(C),          \
-      (int)(imm), (__mmask8)(U), (int)(R)))
-
-#define _mm256_fmadd_round_pd(A, B, C, R)                                      \
-  ((__m256d)__builtin_ia32_vfmaddpd256_round_mask(                             \
-      (__v4df)(__m256d)(A), (__v4df)(__m256d)(B), (__v4df)(__m256d)(C),        \
-      (__mmask8)-1, (int)(R)))
-
-#define _mm256_mask_fmadd_round_pd(A, U, B, C, R)                              \
-  ((__m256d)__builtin_ia32_vfmaddpd256_round_mask(                             \
-      (__v4df)(__m256d)(A), (__v4df)(__m256d)(B), (__v4df)(__m256d)(C),        \
-      (__mmask8)(U), (int)(R)))
-
-#define _mm256_mask3_fmadd_round_pd(A, B, C, U, R)                             \
-  ((__m256d)__builtin_ia32_vfmaddpd256_round_mask3(                            \
-      (__v4df)(__m256d)(A), (__v4df)(__m256d)(B), (__v4df)(__m256d)(C),        \
-      (__mmask8)(U), (int)(R)))
-
-#define _mm256_maskz_fmadd_round_pd(U, A, B, C, R)                             \
-  ((__m256d)__builtin_ia32_vfmaddpd256_round_maskz(                            \
-      (__v4df)(__m256d)(A), (__v4df)(__m256d)(B), (__v4df)(__m256d)(C),        \
-      (__mmask8)(U), (int)(R)))
-
-#define _mm256_fmsub_round_pd(A, B, C, R)                                      \
-  ((__m256d)__builtin_ia32_vfmaddpd256_round_mask(                             \
-      (__v4df)(__m256d)(A), (__v4df)(__m256d)(B), -(__v4df)(__m256d)(C),       \
-      (__mmask8)-1, (int)(R)))
-
-#define _mm256_mask_fmsub_round_pd(A, U, B, C, R)                              \
-  ((__m256d)__builtin_ia32_vfmaddpd256_round_mask(                             \
-      (__v4df)(__m256d)(A), (__v4df)(__m256d)(B), -(__v4df)(__m256d)(C),       \
-      (__mmask8)(U), (int)(R)))
-
-#define _mm256_maskz_fmsub_round_pd(U, A, B, C, R)                             \
-  ((__m256d)__builtin_ia32_vfmaddpd256_round_maskz(                            \
-      (__v4df)(__m256d)(A), (__v4df)(__m256d)(B), -(__v4df)(__m256d)(C),       \
-      (__mmask8)(U), (int)(R)))
-
-#define _mm256_fnmadd_round_pd(A, B, C, R)                                     \
-  ((__m256d)__builtin_ia32_vfmaddpd256_round_mask(                             \
-      -(__v4df)(__m256d)(A), (__v4df)(__m256d)(B), (__v4df)(__m256d)(C),       \
-      (__mmask8)-1, (int)(R)))
-
-#define _mm256_mask3_fnmadd_round_pd(A, B, C, U, R)                            \
-  ((__m256d)__builtin_ia32_vfmaddpd256_round_mask3(                            \
-      -(__v4df)(__m256d)(A), (__v4df)(__m256d)(B), (__v4df)(__m256d)(C),       \
-      (__mmask8)(U), (int)(R)))
-
-#define _mm256_maskz_fnmadd_round_pd(U, A, B, C, R)                            \
-  ((__m256d)__builtin_ia32_vfmaddpd256_round_maskz(                            \
-      -(__v4df)(__m256d)(A), (__v4df)(__m256d)(B), (__v4df)(__m256d)(C),       \
-      (__mmask8)(U), (int)(R)))
-
-#define _mm256_fnmsub_round_pd(A, B, C, R)                                     \
-  ((__m256d)__builtin_ia32_vfmaddpd256_round_mask(                             \
-      -(__v4df)(__m256d)(A), (__v4df)(__m256d)(B), -(__v4df)(__m256d)(C),      \
-      (__mmask8)-1, (int)(R)))
-
-#define _mm256_maskz_fnmsub_round_pd(U, A, B, C, R)                            \
-  ((__m256d)__builtin_ia32_vfmaddpd256_round_maskz(                            \
-      -(__v4df)(__m256d)(A), (__v4df)(__m256d)(B), -(__v4df)(__m256d)(C),      \
-      (__mmask8)(U), (int)(R)))
-
-#define _mm256_fmadd_round_ph(A, B, C, R)                                      \
-  ((__m256h)__builtin_ia32_vfmaddph256_round_mask(                             \
-      (__v16hf)(__m256h)(A), (__v16hf)(__m256h)(B), (__v16hf)(__m256h)(C),     \
-      (__mmask16)-1, (int)(R)))
-
-#define _mm256_mask_fmadd_round_ph(A, U, B, C, R)                              \
-  ((__m256h)__builtin_ia32_vfmaddph256_round_mask(                             \
-      (__v16hf)(__m256h)(A), (__v16hf)(__m256h)(B), (__v16hf)(__m256h)(C),     \
-      (__mmask16)(U), (int)(R)))
-
-#define _mm256_mask3_fmadd_round_ph(A, B, C, U, R)                             \
-  ((__m256h)__builtin_ia32_vfmaddph256_round_mask3(                            \
-      (__v16hf)(__m256h)(A), (__v16hf)(__m256h)(B), (__v16hf)(__m256h)(C),     \
-      (__mmask16)(U), (int)(R)))
-
-#define _mm256_maskz_fmadd_round_ph(U, A, B, C, R)                             \
-  ((__m256h)__builtin_ia32_vfmaddph256_round_maskz(                            \
-      (__v16hf)(__m256h)(A), (__v16hf)(__m256h)(B), (__v16hf)(__m256h)(C),     \
-      (__mmask16)(U), (int)(R)))
-
-#define _mm256_fmsub_round_ph(A, B, C, R)                                      \
-  ((__m256h)__builtin_ia32_vfmaddph256_round_mask(                             \
-      (__v16hf)(__m256h)(A), (__v16hf)(__m256h)(B), -(__v16hf)(__m256h)(C),    \
-      (__mmask16)-1, (int)(R)))
-
-#define _mm256_mask_fmsub_round_ph(A, U, B, C, R)                              \
-  ((__m256h)__builtin_ia32_vfmaddph256_round_mask(                             \
-      (__v16hf)(__m256h)(A), (__v16hf)(__m256h)(B), -(__v16hf)(__m256h)(C),    \
-      (__mmask16)(U), (int)(R)))
-
-#define _mm256_maskz_fmsub_round_ph(U, A, B, C, R)                             \
-  ((__m256h)__builtin_ia32_vfmaddph256_round_maskz(                            \
-      (__v16hf)(__m256h)(A), (__v16hf)(__m256h)(B), -(__v16hf)(__m256h)(C),    \
-      (__mmask16)(U), (int)(R)))
-
-#define _mm256_fnmadd_round_ph(A, B, C, R)                                     \
-  ((__m256h)__builtin_ia32_vfmaddph256_round_mask(                             \
-      (__v16hf)(__m256h)(A), -(__v16hf)(__m256h)(B), (__v16hf)(__m256h)(C),    \
-      (__mmask16)-1, (int)(R)))
-
-#define _mm256_mask3_fnmadd_round_ph(A, B, C, U, R)                            \
-  ((__m256h)__builtin_ia32_vfmaddph256_round_mask3(                            \
-      -(__v16hf)(__m256h)(A), (__v16hf)(__m256h)(B), (__v16hf)(__m256h)(C),    \
-      (__mmask16)(U), (int)(R)))
-
-#define _mm256_maskz_fnmadd_round_ph(U, A, B, C, R)                            \
-  ((__m256h)__builtin_ia32_vfmaddph256_round_maskz(                            \
-      -(__v16hf)(__m256h)(A), (__v16hf)(__m256h)(B), (__v16hf)(__m256h)(C),    \
-      (__mmask16)(U), (int)(R)))
-
-#define _mm256_fnmsub_round_ph(A, B, C, R)                                     \
-  ((__m256h)__builtin_ia32_vfmaddph256_round_mask(                             \
-      (__v16hf)(__m256h)(A), -(__v16hf)(__m256h)(B), -(__v16hf)(__m256h)(C),   \
-      (__mmask16)-1, (int)(R)))
-
-#define _mm256_maskz_fnmsub_round_ph(U, A, B, C, R)                            \
-  ((__m256h)__builtin_ia32_vfmaddph256_round_maskz(                            \
-      -(__v16hf)(__m256h)(A), (__v16hf)(__m256h)(B), -(__v16hf)(__m256h)(C),   \
-      (__mmask16)(U), (int)(R)))
-
-#define _mm256_fmadd_round_ps(A, B, C, R)                                      \
-  ((__m256)__builtin_ia32_vfmaddps256_round_mask(                              \
-      (__v8sf)(__m256)(A), (__v8sf)(__m256)(B), (__v8sf)(__m256)(C),           \
-      (__mmask8)-1, (int)(R)))
-
-#define _mm256_mask_fmadd_round_ps(A, U, B, C, R)                              \
-  ((__m256)__builtin_ia32_vfmaddps256_round_mask(                              \
-      (__v8sf)(__m256)(A), (__v8sf)(__m256)(B), (__v8sf)(__m256)(C),           \
-      (__mmask8)(U), (int)(R)))
-
-#define _mm256_mask3_fmadd_round_ps(A, B, C, U, R)                             \
-  ((__m256)__builtin_ia32_vfmaddps256_round_mask3(                             \
-      (__v8sf)(__m256)(A), (__v8sf)(__m256)(B), (__v8sf)(__m256)(C),           \
-      (__mmask8)(U), (int)(R)))
-
-#define _mm256_maskz_fmadd_round_ps(U, A, B, C, R)                             \
-  ((__m256)__builtin_ia32_vfmaddps256_round_maskz(                             \
-      (__v8sf)(__m256)(A), (__v8sf)(__m256)(B), (__v8sf)(__m256)(C),           \
-      (__mmask8)(U), (int)(R)))
-
-#define _mm256_fmsub_round_ps(A, B, C, R)                                      \
-  ((__m256)__builtin_ia32_vfmaddps256_round_mask(                              \
-      (__v8sf)(__m256)(A), (__v8sf)(__m256)(B), -(__v8sf)(__m256)(C),          \
-      (__mmask8)-1, (int)(R)))
-
-#define _mm256_mask_fmsub_round_ps(A, U, B, C, R)                              \
-  ((__m256)__builtin_ia32_vfmaddps256_round_mask(                              \
-      (__v8sf)(__m256)(A), (__v8sf)(__m256)(B), -(__v8sf)(__m256)(C),          \
-      (__mmask8)(U), (int)(R)))
-
-#define _mm256_maskz_fmsub_round_ps(U, A, B, C, R)                             \
-  ((__m256)__builtin_ia32_vfmaddps256_round_maskz(                             \
-      (__v8sf)(__m256)(A), (__v8sf)(__m256)(B), -(__v8sf)(__m256)(C),          \
-      (__mmask8)(U), (int)(R)))
-
-#define _mm256_fnmadd_round_ps(A, B, C, R)                                     \
-  ((__m256)__builtin_ia32_vfmaddps256_round_mask(                              \
-      (__v8sf)(__m256)(A), -(__v8sf)(__m256)(B), (__v8sf)(__m256)(C),          \
-      (__mmask8)-1, (int)(R)))
-
-#define _mm256_mask3_fnmadd_round_ps(A, B, C, U, R)                            \
-  ((__m256)__builtin_ia32_vfmaddps256_round_mask3(                             \
-      -(__v8sf)(__m256)(A), (__v8sf)(__m256)(B), (__v8sf)(__m256)(C),          \
-      (__mmask8)(U), (int)(R)))
-
-#define _mm256_maskz_fnmadd_round_ps(U, A, B, C, R)                            \
-  ((__m256)__builtin_ia32_vfmaddps256_round_maskz(                             \
-      -(__v8sf)(__m256)(A), (__v8sf)(__m256)(B), (__v8sf)(__m256)(C),          \
-      (__mmask8)(U), (int)(R)))
-
-#define _mm256_fnmsub_round_ps(A, B, C, R)                                     \
-  ((__m256)__builtin_ia32_vfmaddps256_round_mask(                              \
-      (__v8sf)(__m256)(A), -(__v8sf)(__m256)(B), -(__v8sf)(__m256)(C),         \
-      (__mmask8)-1, (int)(R)))
-
-#define _mm256_maskz_fnmsub_round_ps(U, A, B, C, R)                            \
-  ((__m256)__builtin_ia32_vfmaddps256_round_maskz(                             \
-      -(__v8sf)(__m256)(A), (__v8sf)(__m256)(B), -(__v8sf)(__m256)(C),         \
-      (__mmask8)(U), (int)(R)))
-
-#define _mm256_fmadd_round_pch(A, B, C, R)                                     \
-  ((__m256h)__builtin_ia32_vfmaddcph256_round_mask3(                           \
-      (__v8sf)(__m256h)(A), (__v8sf)(__m256h)(B), (__v8sf)(__m256h)(C),        \
-      (__mmask8)-1, (int)(R)))
-
-#define _mm256_mask_fmadd_round_pch(A, U, B, C, R)                             \
-  ((__m256h)__builtin_ia32_vfmaddcph256_round_mask(                            \
-      (__v8sf)(__m256h)(A), (__v8sf)(__m256h)(B), (__v8sf)(__m256h)(C),        \
-      (__mmask8)(U), (int)(R)))
-
-#define _mm256_mask3_fmadd_round_pch(A, B, C, U, R)                            \
-  ((__m256h)__builtin_ia32_vfmaddcph256_round_mask3(                           \
-      (__v8sf)(__m256h)(A), (__v8sf)(__m256h)(B), (__v8sf)(__m256h)(C),        \
-      (__mmask8)(U), (int)(R)))
-
-#define _mm256_maskz_fmadd_round_pch(U, A, B, C, R)                            \
-  ((__m256h)__builtin_ia32_vfmaddcph256_round_maskz(                           \
-      (__v8sf)(__m256h)(A), (__v8sf)(__m256h)(B), (__v8sf)(__m256h)(C),        \
-      (__mmask8)(U), (int)(R)))
-
-#define _mm256_fmaddsub_round_pd(A, B, C, R)                                   \
-  ((__m256d)__builtin_ia32_vfmaddsubpd256_round_mask(                          \
-      (__v4df)(__m256d)(A), (__v4df)(__m256d)(B), (__v4df)(__m256d)(C),        \
-      (__mmask8)-1, (int)(R)))
-
-#define _mm256_mask_fmaddsub_round_pd(A, U, B, C, R)                           \
-  ((__m256d)__builtin_ia32_vfmaddsubpd256_round_mask(                          \
-      (__v4df)(__m256d)(A), (__v4df)(__m256d)(B), (__v4df)(__m256d)(C),        \
-      (__mmask8)(U), (int)(R)))
-
-#define _mm256_mask3_fmaddsub_round_pd(A, B, C, U, R)                          \
-  ((__m256d)__builtin_ia32_vfmaddsubpd256_round_mask3(                         \
-      (__v4df)(__m256d)(A), (__v4df)(__m256d)(B), (__v4df)(__m256d)(C),        \
-      (__mmask8)(U), (int)(R)))
-
-#define _mm256_maskz_fmaddsub_round_pd(U, A, B, C, R)                          \
-  ((__m256d)__builtin_ia32_vfmaddsubpd256_round_maskz(                         \
-      (__v4df)(__m256d)(A), (__v4df)(__m256d)(B), (__v4df)(__m256d)(C),        \
-      (__mmask8)(U), (int)(R)))
-
-#define _mm256_fmsubadd_round_pd(A, B, C, R)                                   \
-  ((__m256d)__builtin_ia32_vfmaddsubpd256_round_mask(                          \
-      (__v4df)(__m256d)(A), (__v4df)(__m256d)(B), -(__v4df)(__m256d)(C),       \
-      (__mmask8)-1, (int)(R)))
-
-#define _mm256_mask_fmsubadd_round_pd(A, U, B, C, R)                           \
-  ((__m256d)__builtin_ia32_vfmaddsubpd256_round_mask(                          \
-      (__v4df)(__m256d)(A), (__v4df)(__m256d)(B), -(__v4df)(__m256d)(C),       \
-      (__mmask8)(U), (int)(R)))
-
-#define _mm256_maskz_fmsubadd_round_pd(U, A, B, C, R)                          \
-  ((__m256d)__builtin_ia32_vfmaddsubpd256_round_maskz(                         \
-      (__v4df)(__m256d)(A), (__v4df)(__m256d)(B), -(__v4df)(__m256d)(C),       \
-      (__mmask8)(U), (int)(R)))
-
-#define _mm256_fmaddsub_round_ph(A, B, C, R)                                   \
-  ((__m256h)__builtin_ia32_vfmaddsubph256_round_mask(                          \
-      (__v16hf)(__m256h)(A), (__v16hf)(__m256h)(B), (__v16hf)(__m256h)(C),     \
-      (__mmask16)-1, (int)(R)))
-
-#define _mm256_mask_fmaddsub_round_ph(A, U, B, C, R)                           \
-  ((__m256h)__builtin_ia32_vfmaddsubph256_round_mask(                          \
-      (__v16hf)(__m256h)(A), (__v16hf)(__m256h)(B), (__v16hf)(__m256h)(C),     \
-      (__mmask16)(U), (int)(R)))
-
-#define _mm256_mask3_fmaddsub_round_ph(A, B, C, U, R)                          \
-  ((__m256h)__builtin_ia32_vfmaddsubph256_round_mask3(                         \
-      (__v16hf)(__m256h)(A), (__v16hf)(__m256h)(B), (__v16hf)(__m256h)(C),     \
-      (__mmask16)(U), (int)(R)))
-
-#define _mm256_maskz_fmaddsub_round_ph(U, A, B, C, R)                          \
-  ((__m256h)__builtin_ia32_vfmaddsubph256_round_maskz(                         \
-      (__v16hf)(__m256h)(A), (__v16hf)(__m256h)(B), (__v16hf)(__m256h)(C),     \
-      (__mmask16)(U), (int)(R)))
-
-#define _mm256_fmsubadd_round_ph(A, B, C, R)                                   \
-  ((__m256h)__builtin_ia32_vfmaddsubph256_round_mask(                          \
-      (__v16hf)(__m256h)(A), (__v16hf)(__m256h)(B), -(__v16hf)(__m256h)(C),    \
-      (__mmask16)-1, (int)(R)))
-
-#define _mm256_mask_fmsubadd_round_ph(A, U, B, C, R)                           \
-  ((__m256h)__builtin_ia32_vfmaddsubph256_round_mask(                          \
-      (__v16hf)(__m256h)(A), (__v16hf)(__m256h)(B), -(__v16hf)(__m256h)(C),    \
-      (__mmask16)(U), (int)(R)))
-
-#define _mm256_maskz_fmsubadd_round_ph(U, A, B, C, R)                          \
-  ((__m256h)__builtin_ia32_vfmaddsubph256_round_maskz(                         \
-      (__v16hf)(__m256h)(A), (__v16hf)(__m256h)(B), -(__v16hf)(__m256h)(C),    \
-      (__mmask16)(U), (int)(R)))
-
-#define _mm256_fmaddsub_round_ps(A, B, C, R)                                   \
-  ((__m256)__builtin_ia32_vfmaddsubps256_round_mask(                           \
-      (__v8sf)(__m256)(A), (__v8sf)(__m256)(B), (__v8sf)(__m256)(C),           \
-      (__mmask8)-1, (int)(R)))
-
-#define _mm256_mask_fmaddsub_round_ps(A, U, B, C, R)                           \
-  ((__m256)__builtin_ia32_vfmaddsubps256_round_mask(                           \
-      (__v8sf)(__m256)(A), (__v8sf)(__m256)(B), (__v8sf)(__m256)(C),           \
-      (__mmask8)(U), (int)(R)))
-
-#define _mm256_mask3_fmaddsub_round_ps(A, B, C, U, R)                          \
-  ((__m256)__builtin_ia32_vfmaddsubps256_round_mask3(                          \
-      (__v8sf)(__m256)(A), (__v8sf)(__m256)(B), (__v8sf)(__m256)(C),           \
-      (__mmask8)(U), (int)(R)))
-
-#define _mm256_maskz_fmaddsub_round_ps(U, A, B, C, R)                          \
-  ((__m256)__builtin_ia32_vfmaddsubps256_round_maskz(                          \
-      (__v8sf)(__m256)(A), (__v8sf)(__m256)(B), (__v8sf)(__m256)(C),           \
-      (__mmask8)(U), (int)(R)))
-
-#define _mm256_fmsubadd_round_ps(A, B, C, R)                                   \
-  ((__m256)__builtin_ia32_vfmaddsubps256_round_mask(                           \
-      (__v8sf)(__m256)(A), (__v8sf)(__m256)(B), -(__v8sf)(__m256)(C),          \
-      (__mmask8)-1, (int)(R)))
-
-#define _mm256_mask_fmsubadd_round_ps(A, U, B, C, R)                           \
-  ((__m256)__builtin_ia32_vfmaddsubps256_round_mask(                           \
-      (__v8sf)(__m256)(A), (__v8sf)(__m256)(B), -(__v8sf)(__m256)(C),          \
-      (__mmask8)(U), (int)(R)))
-
-#define _mm256_maskz_fmsubadd_round_ps(U, A, B, C, R)                          \
-  ((__m256)__builtin_ia32_vfmaddsubps256_round_maskz(                          \
-      (__v8sf)(__m256)(A), (__v8sf)(__m256)(B), -(__v8sf)(__m256)(C),          \
-      (__mmask8)(U), (int)(R)))
-#define _mm256_mask3_fmsub_round_pd(A, B, C, U, R)                             \
-  ((__m256d)__builtin_ia32_vfmsubpd256_round_mask3(                            \
-      (__v4df)(__m256d)(A), (__v4df)(__m256d)(B), (__v4df)(__m256d)(C),        \
-      (__mmask8)(U), (int)(R)))
-
-#define _mm256_mask3_fmsubadd_round_pd(A, B, C, U, R)                          \
-  ((__m256d)__builtin_ia32_vfmsubaddpd256_round_mask3(                         \
-      (__v4df)(__m256d)(A), (__v4df)(__m256d)(B), (__v4df)(__m256d)(C),        \
-      (__mmask8)(U), (int)(R)))
-
-#define _mm256_mask_fnmadd_round_pd(A, U, B, C, R)                             \
-  ((__m256d)__builtin_ia32_vfmaddpd256_round_mask(                             \
-      (__v4df)(__m256d)(A), -(__v4df)(__m256d)(B), (__v4df)(__m256d)(C),       \
-      (__mmask8)(U), (int)(R)))
-
-#define _mm256_mask_fnmsub_round_pd(A, U, B, C, R)                             \
-  ((__m256d)__builtin_ia32_vfmaddpd256_round_mask(                             \
-      (__v4df)(__m256d)(A), -(__v4df)(__m256d)(B), -(__v4df)(__m256d)(C),      \
-      (__mmask8)(U), (int)(R)))
-
-#define _mm256_mask3_fnmsub_round_pd(A, B, C, U, R)                            \
-  ((__m256d)__builtin_ia32_vfmsubpd256_round_mask3(                            \
-      -(__v4df)(__m256d)(A), (__v4df)(__m256d)(B), (__v4df)(__m256d)(C),       \
-      (__mmask8)(U), (int)(R)))
-
-#define _mm256_mask3_fmsub_round_ph(A, B, C, U, R)                             \
-  ((__m256h)__builtin_ia32_vfmsubph256_round_mask3(                            \
-      (__v16hf)(__m256h)(A), (__v16hf)(__m256h)(B), (__v16hf)(__m256h)(C),     \
-      (__mmask16)(U), (int)(R)))
-
-#define _mm256_mask3_fmsubadd_round_ph(A, B, C, U, R)                          \
-  ((__m256h)__builtin_ia32_vfmsubaddph256_round_mask3(                         \
-      (__v16hf)(__m256h)(A), (__v16hf)(__m256h)(B), (__v16hf)(__m256h)(C),     \
-      (__mmask16)(U), (int)(R)))
-
-#define _mm256_mask_fnmadd_round_ph(A, U, B, C, R)                             \
-  ((__m256h)__builtin_ia32_vfmaddph256_round_mask(                             \
-      (__v16hf)(__m256h)(A), -(__v16hf)(__m256h)(B), (__v16hf)(__m256h)(C),    \
-      (__mmask16)(U), (int)(R)))
-
-#define _mm256_mask_fnmsub_round_ph(A, U, B, C, R)                             \
-  ((__m256h)__builtin_ia32_vfmaddph256_round_mask(                             \
-      (__v16hf)(__m256h)(A), -(__v16hf)(__m256h)(B), -(__v16hf)(__m256h)(C),   \
-      (__mmask16)(U), (int)(R)))
-
-#define _mm256_mask3_fnmsub_round_ph(A, B, C, U, R)                            \
-  ((__m256h)__builtin_ia32_vfmsubph256_round_mask3(                            \
-      -(__v16hf)(__m256h)(A), (__v16hf)(__m256h)(B), (__v16hf)(__m256h)(C),    \
-      (__mmask16)(U), (int)(R)))
-
-#define _mm256_mask3_fmsub_round_ps(A, B, C, U, R)                             \
-  ((__m256)__builtin_ia32_vfmsubps256_round_mask3(                             \
-      (__v8sf)(__m256)(A), (__v8sf)(__m256)(B), (__v8sf)(__m256)(C),           \
-      (__mmask8)(U), (int)(R)))
-
-#define _mm256_mask3_fmsubadd_round_ps(A, B, C, U, R)                          \
-  ((__m256)__builtin_ia32_vfmsubaddps256_round_mask3(                          \
-      (__v8sf)(__m256)(A), (__v8sf)(__m256)(B), (__v8sf)(__m256)(C),           \
-      (__mmask8)(U), (int)(R)))
-
-#define _mm256_mask_fnmadd_round_ps(A, U, B, C, R)                             \
-  ((__m256)__builtin_ia32_vfmaddps256_round_mask(                              \
-      (__v8sf)(__m256)(A), -(__v8sf)(__m256)(B), (__v8sf)(__m256)(C),          \
-      (__mmask8)(U), (int)(R)))
-
-#define _mm256_mask_fnmsub_round_ps(A, U, B, C, R)                             \
-  ((__m256)__builtin_ia32_vfmaddps256_round_mask(                              \
-      (__v8sf)(__m256)(A), -(__v8sf)(__m256)(B), -(__v8sf)(__m256)(C),         \
-      (__mmask8)(U), (int)(R)))
-
-#define _mm256_mask3_fnmsub_round_ps(A, B, C, U, R)                            \
-  ((__m256)__builtin_ia32_vfmsubps256_round_mask3(                             \
-      -(__v8sf)(__m256)(A), (__v8sf)(__m256)(B), (__v8sf)(__m256)(C),          \
-      (__mmask8)(U), (int)(R)))
-
-#define _mm256_mul_round_pch(A, B, R)                                          \
-  ((__m256h)__builtin_ia32_vfmulcph256_round_mask(                             \
-      (__v8sf)(__m256h)(A), (__v8sf)(__m256h)(B),                              \
-      (__v8sf)(__m256h)_mm256_undefined_ph(), (__mmask8)-1, (int)(R)))
-
-#define _mm256_mask_mul_round_pch(W, U, A, B, R)                               \
-  ((__m256h)__builtin_ia32_vfmulcph256_round_mask(                             \
-      (__v8sf)(__m256h)(A), (__v8sf)(__m256h)(B), (__v8sf)(__m256h)(W),        \
-      (__mmask8)(U), (int)(R)))
-
-#define _mm256_maskz_mul_round_pch(U, A, B, R)                                 \
-  ((__m256h)__builtin_ia32_vfmulcph256_round_mask(                             \
-      (__v8sf)(__m256h)(A), (__v8sf)(__m256h)(B),                              \
-      (__v8sf)(__m256h)_mm256_setzero_ph(), (__mmask8)(U), (int)(R)))
-
-#define _mm256_getexp_round_pd(A, R)                                           \
-  ((__m256d)__builtin_ia32_vgetexppd256_round_mask(                            \
-      (__v4df)(__m256d)(A), (__v4df)_mm256_undefined_pd(), (__mmask8)-1,       \
-      (int)(R)))
-
-#define _mm256_mask_getexp_round_pd(W, U, A, R)                                \
-  ((__m256d)__builtin_ia32_vgetexppd256_round_mask(                            \
-      (__v4df)(__m256d)(A), (__v4df)(__m256d)(W), (__mmask8)(U), (int)(R)))
-
-#define _mm256_maskz_getexp_round_pd(U, A, R)                                  \
-  ((__m256d)__builtin_ia32_vgetexppd256_round_mask(                            \
-      (__v4df)(__m256d)(A), (__v4df)_mm256_setzero_pd(), (__mmask8)(U),        \
-      (int)(R)))
-
-#define _mm256_getexp_round_ph(A, R)                                           \
-  ((__m256h)__builtin_ia32_vgetexpph256_round_mask(                            \
-      (__v16hf)(__m256h)(A), (__v16hf)_mm256_undefined_ph(), (__mmask16)-1,    \
-      (int)(R)))
-
-#define _mm256_mask_getexp_round_ph(W, U, A, R)                                \
-  ((__m256h)__builtin_ia32_vgetexpph256_round_mask(                            \
-      (__v16hf)(__m256h)(A), (__v16hf)(__m256h)(W), (__mmask16)(U), (int)(R)))
-
-#define _mm256_maskz_getexp_round_ph(U, A, R)                                  \
-  ((__m256h)__builtin_ia32_vgetexpph256_round_mask(                            \
-      (__v16hf)(__m256h)(A), (__v16hf)_mm256_setzero_ph(), (__mmask16)(U),     \
-      (int)(R)))
-
-#define _mm256_getexp_round_ps(A, R)                                           \
-  ((__m256)__builtin_ia32_vgetexpps256_round_mask(                             \
-      (__v8sf)(__m256)(A), (__v8sf)_mm256_undefined_ps(), (__mmask8)-1,        \
-      (int)(R)))
-
-#define _mm256_mask_getexp_round_ps(W, U, A, R)                                \
-  ((__m256)__builtin_ia32_vgetexpps256_round_mask(                             \
-      (__v8sf)(__m256)(A), (__v8sf)(__m256)(W), (__mmask8)(U), (int)(R)))
-
-#define _mm256_maskz_getexp_round_ps(U, A, R)                                  \
-  ((__m256)__builtin_ia32_vgetexpps256_round_mask((__v8sf)(__m256)(A),         \
-                                                  (__v8sf)_mm256_setzero_ps(), \
-                                                  (__mmask8)(U), (int)(R)))
-
-#define _mm256_getmant_round_pd(A, B, C, R)                                    \
-  ((__m256d)__builtin_ia32_vgetmantpd256_round_mask(                           \
-      (__v4df)(__m256d)(A), (int)(((C) << 2) | (B)),                           \
-      (__v4df)_mm256_undefined_pd(), (__mmask8)-1, (int)(R)))
-
-#define _mm256_mask_getmant_round_pd(W, U, A, B, C, R)                         \
-  ((__m256d)__builtin_ia32_vgetmantpd256_round_mask(                           \
-      (__v4df)(__m256d)(A), (int)(((C) << 2) | (B)), (__v4df)(__m256d)(W),     \
-      (__mmask8)(U), (int)(R)))
-
-#define _mm256_maskz_getmant_round_pd(U, A, B, C, R)                           \
-  ((__m256d)__builtin_ia32_vgetmantpd256_round_mask(                           \
-      (__v4df)(__m256d)(A), (int)(((C) << 2) | (B)),                           \
-      (__v4df)_mm256_setzero_pd(), (__mmask8)(U), (int)(R)))
-
-#define _mm256_getmant_round_ph(A, B, C, R)                                    \
-  ((__m256h)__builtin_ia32_vgetmantph256_round_mask(                           \
-      (__v16hf)(__m256h)(A), (int)(((C) << 2) | (B)),                          \
-      (__v16hf)_mm256_undefined_ph(), (__mmask16)-1, (int)(R)))
-
-#define _mm256_mask_getmant_round_ph(W, U, A, B, C, R)                         \
-  ((__m256h)__builtin_ia32_vgetmantph256_round_mask(                           \
-      (__v16hf)(__m256h)(A), (int)(((C) << 2) | (B)), (__v16hf)(__m256h)(W),   \
-      (__mmask16)(U), (int)(R)))
-
-#define _mm256_maskz_getmant_round_ph(U, A, B, C, R)                           \
-  ((__m256h)__builtin_ia32_vgetmantph256_round_mask(                           \
-      (__v16hf)(__m256h)(A), (int)(((C) << 2) | (B)),                          \
-      (__v16hf)_mm256_setzero_ph(), (__mmask16)(U), (int)(R)))
-
-#define _mm256_getmant_round_ps(A, B, C, R)                                    \
-  ((__m256)__builtin_ia32_vgetmantps256_round_mask(                            \
-      (__v8sf)(__m256)(A), (int)(((C) << 2) | (B)),                            \
-      (__v8sf)_mm256_undefined_ps(), (__mmask8)-1, (int)(R)))
-
-#define _mm256_mask_getmant_round_ps(W, U, A, B, C, R)                         \
-  ((__m256)__builtin_ia32_vgetmantps256_round_mask(                            \
-      (__v8sf)(__m256)(A), (int)(((C) << 2) | (B)), (__v8sf)(__m256)(W),       \
-      (__mmask8)(U), (int)(R)))
-
-#define _mm256_maskz_getmant_round_ps(U, A, B, C, R)                           \
-  ((__m256)__builtin_ia32_vgetmantps256_round_mask(                            \
-      (__v8sf)(__m256)(A), (int)(((C) << 2) | (B)),                            \
-      (__v8sf)_mm256_setzero_ps(), (__mmask8)(U), (int)(R)))
-
-#define _mm256_max_round_pd(A, B, R)                                           \
-  ((__m256d)__builtin_ia32_vmaxpd256_round((__v4df)(__m256d)(A),               \
-                                           (__v4df)(__m256d)(B), (int)(R)))
-
-#define _mm256_mask_max_round_pd(W, U, A, B, R)                                \
-  ((__m256d)__builtin_ia32_selectpd_256(                                       \
-      (__mmask8)(U), (__v4df)_mm256_max_round_pd((A), (B), (R)),               \
-      (__v4df)(__m256d)(W)))
-
-#define _mm256_maskz_max_round_pd(U, A, B, R)                                  \
-  ((__m256d)__builtin_ia32_selectpd_256(                                       \
-      (__mmask8)(U), (__v4df)_mm256_max_round_pd((A), (B), (R)),               \
-      (__v4df)_mm256_setzero_pd()))
-
-#define _mm256_max_round_ph(A, B, R)                                           \
-  ((__m256h)__builtin_ia32_vmaxph256_round((__v16hf)(__m256h)(A),              \
-                                           (__v16hf)(__m256h)(B), (int)(R)))
-
-#define _mm256_mask_max_round_ph(W, U, A, B, R)                                \
-  ((__m256h)__builtin_ia32_selectph_256(                                       \
-      (__mmask16)(U), (__v16hf)_mm256_max_round_ph((A), (B), (R)),             \
-      (__v16hf)(__m256h)(W)))
-
-#define _mm256_maskz_max_round_ph(U, A, B, R)                                  \
-  ((__m256h)__builtin_ia32_selectph_256(                                       \
-      (__mmask16)(U), (__v16hf)_mm256_max_round_ph((A), (B), (R)),             \
-      (__v16hf)_mm256_setzero_ph()))
-
-#define _mm256_max_round_ps(A, B, R)                                           \
-  ((__m256)__builtin_ia32_vmaxps256_round((__v8sf)(__m256)(A),                 \
-                                          (__v8sf)(__m256)(B), (int)(R)))
-
-#define _mm256_mask_max_round_ps(W, U, A, B, R)                                \
-  ((__m256)__builtin_ia32_selectps_256(                                        \
-      (__mmask8)(U), (__v8sf)_mm256_max_round_ps((A), (B), (R)),               \
-      (__v8sf)(__m256)(W)))
-
-#define _mm256_maskz_max_round_ps(U, A, B, R)                                  \
-  ((__m256)__builtin_ia32_selectps_256(                                        \
-      (__mmask8)(U), (__v8sf)_mm256_max_round_ps((A), (B), (R)),               \
-      (__v8sf)_mm256_setzero_ps()))
-
-#define _mm256_min_round_pd(A, B, R)                                           \
-  ((__m256d)__builtin_ia32_vminpd256_round((__v4df)(__m256d)(A),               \
-                                           (__v4df)(__m256d)(B), (int)(R)))
-
-#define _mm256_mask_min_round_pd(W, U, A, B, R)                                \
-  ((__m256d)__builtin_ia32_selectpd_256(                                       \
-      (__mmask8)(U), (__v4df)_mm256_min_round_pd((A), (B), (R)),               \
-      (__v4df)(__m256d)(W)))
-
-#define _mm256_maskz_min_round_pd(U, A, B, R)                                  \
-  ((__m256d)__builtin_ia32_selectpd_256(                                       \
-      (__mmask8)(U), (__v4df)_mm256_min_round_pd((A), (B), (R)),               \
-      (__v4df)_mm256_setzero_pd()))
-
-#define _mm256_min_round_ph(A, B, R)                                           \
-  ((__m256h)__builtin_ia32_vminph256_round((__v16hf)(__m256h)(A),              \
-                                           (__v16hf)(__m256h)(B), (int)(R)))
-
-#define _mm256_mask_min_round_ph(W, U, A, B, R)                                \
-  ((__m256h)__builtin_ia32_selectph_256(                                       \
-      (__mmask16)(U), (__v16hf)_mm256_min_round_ph((A), (B), (R)),             \
-      (__v16hf)(__m256h)(W)))
-
-#define _mm256_maskz_min_round_ph(U, A, B, R)                                  \
-  ((__m256h)__builtin_ia32_selectph_256(                                       \
-      (__mmask16)(U), (__v16hf)_mm256_min_round_ph((A), (B), (R)),             \
-      (__v16hf)_mm256_setzero_ph()))
-
-#define _mm256_min_round_ps(A, B, R)                                           \
-  ((__m256)__builtin_ia32_vminps256_round((__v8sf)(__m256)(A),                 \
-                                          (__v8sf)(__m256)(B), (int)(R)))
-
-#define _mm256_mask_min_round_ps(W, U, A, B, R)                                \
-  ((__m256)__builtin_ia32_selectps_256(                                        \
-      (__mmask8)(U), (__v8sf)_mm256_min_round_ps((A), (B), (R)),               \
-      (__v8sf)(__m256)(W)))
-
-#define _mm256_maskz_min_round_ps(U, A, B, R)                                  \
-  ((__m256)__builtin_ia32_selectps_256(                                        \
-      (__mmask8)(U), (__v8sf)_mm256_min_round_ps((A), (B), (R)),               \
-      (__v8sf)_mm256_setzero_ps()))
-
-#define _mm256_mul_round_pd(A, B, R)                                           \
-  ((__m256d)__builtin_ia32_vmulpd256_round((__v4df)(__m256d)(A),               \
-                                           (__v4df)(__m256d)(B), (int)(R)))
-
-#define _mm256_mask_mul_round_pd(W, U, A, B, R)                                \
-  ((__m256d)__builtin_ia32_selectpd_256(                                       \
-      (__mmask8)(U), (__v4df)_mm256_mul_round_pd((A), (B), (R)),               \
-      (__v4df)(__m256d)(W)))
-
-#define _mm256_maskz_mul_round_pd(U, A, B, R)                                  \
-  ((__m256d)__builtin_ia32_selectpd_256(                                       \
-      (__mmask8)(U), (__v4df)_mm256_mul_round_pd((A), (B), (R)),               \
-      (__v4df)_mm256_setzero_pd()))
-
-#define _mm256_mul_round_ph(A, B, R)                                           \
-  ((__m256h)__builtin_ia32_vmulph256_round((__v16hf)(__m256h)(A),              \
-                                           (__v16hf)(__m256h)(B), (int)(R)))
-
-#define _mm256_mask_mul_round_ph(W, U, A, B, R)                                \
-  ((__m256h)__builtin_ia32_selectph_256(                                       \
-      (__mmask16)(U), (__v16hf)_mm256_mul_round_ph((A), (B), (R)),             \
-      (__v16hf)(__m256h)(W)))
-
-#define _mm256_maskz_mul_round_ph(U, A, B, R)                                  \
-  ((__m256h)__builtin_ia32_selectph_256(                                       \
-      (__mmask16)(U), (__v16hf)_mm256_mul_round_ph((A), (B), (R)),             \
-      (__v16hf)_mm256_setzero_ph()))
-
-#define _mm256_mul_round_ps(A, B, R)                                           \
-  ((__m256)__builtin_ia32_vmulps256_round((__v8sf)(__m256)(A),                 \
-                                          (__v8sf)(__m256)(B), (int)(R)))
-
-#define _mm256_mask_mul_round_ps(W, U, A, B, R)                                \
-  ((__m256)__builtin_ia32_selectps_256(                                        \
-      (__mmask8)(U), (__v8sf)_mm256_mul_round_ps((A), (B), (R)),               \
-      (__v8sf)(__m256)(W)))
-
-#define _mm256_maskz_mul_round_ps(U, A, B, R)                                  \
-  ((__m256)__builtin_ia32_selectps_256(                                        \
-      (__mmask8)(U), (__v8sf)_mm256_mul_round_ps((A), (B), (R)),               \
-      (__v8sf)_mm256_setzero_ps()))
-
-#define _mm256_range_round_pd(A, B, C, R)                                      \
-  ((__m256d)__builtin_ia32_vrangepd256_round_mask(                             \
-      (__v4df)(__m256d)(A), (__v4df)(__m256d)(B), (int)(C),                    \
-      (__v4df)_mm256_setzero_pd(), (__mmask8)-1, (int)(R)))
-
-#define _mm256_mask_range_round_pd(W, U, A, B, C, R)                           \
-  ((__m256d)__builtin_ia32_vrangepd256_round_mask(                             \
-      (__v4df)(__m256d)(A), (__v4df)(__m256d)(B), (int)(C),                    \
-      (__v4df)(__m256d)(W), (__mmask8)(U), (int)(R)))
-
-#define _mm256_maskz_range_round_pd(U, A, B, C, R)                             \
-  ((__m256d)__builtin_ia32_vrangepd256_round_mask(                             \
-      (__v4df)(__m256d)(A), (__v4df)(__m256d)(B), (int)(C),                    \
-      (__v4df)_mm256_setzero_pd(), (__mmask8)(U), (int)(R)))
-
-#define _mm256_range_round_ps(A, B, C, R)                                      \
-  ((__m256)__builtin_ia32_vrangeps256_round_mask(                              \
-      (__v8sf)(__m256)(A), (__v8sf)(__m256)(B), (int)(C),                      \
-      (__v8sf)_mm256_setzero_ps(), (__mmask8)-1, (int)(R)))
-
-#define _mm256_mask_range_round_ps(W, U, A, B, C, R)                           \
-  ((__m256)__builtin_ia32_vrangeps256_round_mask(                              \
-      (__v8sf)(__m256)(A), (__v8sf)(__m256)(B), (int)(C), (__v8sf)(__m256)(W), \
-      (__mmask8)(U), (int)(R)))
-
-#define _mm256_maskz_range_round_ps(U, A, B, C, R)                             \
-  ((__m256)__builtin_ia32_vrangeps256_round_mask(                              \
-      (__v8sf)(__m256)(A), (__v8sf)(__m256)(B), (int)(C),                      \
-      (__v8sf)_mm256_setzero_ps(), (__mmask8)(U), (int)(R)))
-
-#define _mm256_reduce_round_pd(A, B, R)                                        \
-  ((__m256d)__builtin_ia32_vreducepd256_round_mask(                            \
-      (__v4df)(__m256d)(A), (int)(B), (__v4df)_mm256_setzero_pd(),             \
-      (__mmask8)-1, (int)(R)))
-
-#define _mm256_mask_reduce_round_pd(W, U, A, B, R)                             \
-  ((__m256d)__builtin_ia32_vreducepd256_round_mask(                            \
-      (__v4df)(__m256d)(A), (int)(B), (__v4df)(__m256d)(W), (__mmask8)(U),     \
-      (int)(R)))
-
-#define _mm256_maskz_reduce_round_pd(U, A, B, R)                               \
-  ((__m256d)__builtin_ia32_vreducepd256_round_mask(                            \
-      (__v4df)(__m256d)(A), (int)(B), (__v4df)_mm256_setzero_pd(),             \
-      (__mmask8)(U), (int)(R)))
-
-#define _mm256_mask_reduce_round_ph(W, U, A, imm, R)                           \
-  ((__m256h)__builtin_ia32_vreduceph256_round_mask(                            \
-      (__v16hf)(__m256h)(A), (int)(imm), (__v16hf)(__m256h)(W),                \
-      (__mmask16)(U), (int)(R)))
-
-#define _mm256_maskz_reduce_round_ph(U, A, imm, R)                             \
-  ((__m256h)__builtin_ia32_vreduceph256_round_mask(                            \
-      (__v16hf)(__m256h)(A), (int)(imm), (__v16hf)_mm256_setzero_ph(),         \
-      (__mmask16)(U), (int)(R)))
-
-#define _mm256_reduce_round_ph(A, imm, R)                                      \
-  ((__m256h)__builtin_ia32_vreduceph256_round_mask(                            \
-      (__v16hf)(__m256h)(A), (int)(imm), (__v16hf)_mm256_undefined_ph(),       \
-      (__mmask16)-1, (int)(R)))
-
-#define _mm256_reduce_round_ps(A, B, R)                                        \
-  ((__m256)__builtin_ia32_vreduceps256_round_mask(                             \
-      (__v8sf)(__m256)(A), (int)(B), (__v8sf)_mm256_setzero_ps(),              \
-      (__mmask8)-1, (int)(R)))
-
-#define _mm256_mask_reduce_round_ps(W, U, A, B, R)                             \
-  ((__m256)__builtin_ia32_vreduceps256_round_mask(                             \
-      (__v8sf)(__m256)(A), (int)(B), (__v8sf)(__m256)(W), (__mmask8)(U),       \
-      (int)(R)))
-
-#define _mm256_maskz_reduce_round_ps(U, A, B, R)                               \
-  ((__m256)__builtin_ia32_vreduceps256_round_mask(                             \
-      (__v8sf)(__m256)(A), (int)(B), (__v8sf)_mm256_setzero_ps(),              \
-      (__mmask8)(U), (int)(R)))
-
-#define _mm256_roundscale_round_pd(A, imm, R)                                  \
-  ((__m256d)__builtin_ia32_vrndscalepd256_round_mask(                          \
-      (__v4df)(__m256d)(A), (int)(imm), (__v4df)_mm256_undefined_pd(),         \
-      (__mmask8)-1, (int)(R)))
-
-#define _mm256_mask_roundscale_round_pd(A, B, C, imm, R)                       \
-  ((__m256d)__builtin_ia32_vrndscalepd256_round_mask(                          \
-      (__v4df)(__m256d)(C), (int)(imm), (__v4df)(__m256d)(A), (__mmask8)(B),   \
-      (int)(R)))
-
-#define _mm256_maskz_roundscale_round_pd(A, B, imm, R)                         \
-  ((__m256d)__builtin_ia32_vrndscalepd256_round_mask(                          \
-      (__v4df)(__m256d)(B), (int)(imm), (__v4df)_mm256_setzero_pd(),           \
-      (__mmask8)(A), (int)(R)))
-
-#define _mm256_roundscale_round_ph(A, imm, R)                                  \
-  ((__m256h)__builtin_ia32_vrndscaleph256_round_mask(                          \
-      (__v16hf)(__m256h)(A), (int)(imm), (__v16hf)_mm256_undefined_ph(),       \
-      (__mmask16)-1, (int)(R)))
-
-#define _mm256_mask_roundscale_round_ph(A, B, C, imm, R)                       \
-  ((__m256h)__builtin_ia32_vrndscaleph256_round_mask(                          \
-      (__v16hf)(__m256h)(C), (int)(imm), (__v16hf)(__m256h)(A),                \
-      (__mmask16)(B), (int)(R)))
-
-#define _mm256_maskz_roundscale_round_ph(A, B, imm, R)                         \
-  ((__m256h)__builtin_ia32_vrndscaleph256_round_mask(                          \
-      (__v16hf)(__m256h)(B), (int)(imm), (__v16hf)_mm256_setzero_ph(),         \
-      (__mmask16)(A), (int)(R)))
-
-#define _mm256_roundscale_round_ps(A, imm, R)                                  \
-  ((__m256)__builtin_ia32_vrndscaleps256_round_mask(                           \
-      (__v8sf)(__m256)(A), (int)(imm), (__v8sf)_mm256_undefined_ps(),          \
-      (__mmask8)-1, (int)(R)))
-
-#define _mm256_mask_roundscale_round_ps(A, B, C, imm, R)                       \
-  ((__m256)__builtin_ia32_vrndscaleps256_round_mask(                           \
-      (__v8sf)(__m256)(C), (int)(imm), (__v8sf)(__m256)(A), (__mmask8)(B),     \
-      (int)(R)))
-
-#define _mm256_maskz_roundscale_round_ps(A, B, imm, R)                         \
-  ((__m256)__builtin_ia32_vrndscaleps256_round_mask(                           \
-      (__v8sf)(__m256)(B), (int)(imm), (__v8sf)_mm256_setzero_ps(),            \
-      (__mmask8)(A), (int)(R)))
-
-#define _mm256_scalef_round_pd(A, B, R)                                        \
-  ((__m256d)__builtin_ia32_vscalefpd256_round_mask(                            \
-      (__v4df)(__m256d)(A), (__v4df)(__m256d)(B),                              \
-      (__v4df)_mm256_undefined_pd(), (__mmask8)-1, (int)(R)))
-
-#define _mm256_mask_scalef_round_pd(W, U, A, B, R)                             \
-  ((__m256d)__builtin_ia32_vscalefpd256_round_mask(                            \
-      (__v4df)(__m256d)(A), (__v4df)(__m256d)(B), (__v4df)(__m256d)(W),        \
-      (__mmask8)(U), (int)(R)))
-
-#define _mm256_maskz_scalef_round_pd(U, A, B, R)                               \
-  ((__m256d)__builtin_ia32_vscalefpd256_round_mask(                            \
-      (__v4df)(__m256d)(A), (__v4df)(__m256d)(B), (__v4df)_mm256_setzero_pd(), \
-      (__mmask8)(U), (int)(R)))
-
-#define _mm256_scalef_round_ph(A, B, R)                                        \
-  ((__m256h)__builtin_ia32_vscalefph256_round_mask(                            \
-      (__v16hf)(__m256h)(A), (__v16hf)(__m256h)(B),                            \
-      (__v16hf)_mm256_undefined_ph(), (__mmask16)-1, (int)(R)))
-
-#define _mm256_mask_scalef_round_ph(W, U, A, B, R)                             \
-  ((__m256h)__builtin_ia32_vscalefph256_round_mask(                            \
-      (__v16hf)(__m256h)(A), (__v16hf)(__m256h)(B), (__v16hf)(__m256h)(W),     \
-      (__mmask16)(U), (int)(R)))
-
-#define _mm256_maskz_scalef_round_ph(U, A, B, R)                               \
-  ((__m256h)__builtin_ia32_vscalefph256_round_mask(                            \
-      (__v16hf)(__m256h)(A), (__v16hf)(__m256h)(B),                            \
-      (__v16hf)_mm256_setzero_ph(), (__mmask16)(U), (int)(R)))
-
-#define _mm256_scalef_round_ps(A, B, R)                                        \
-  ((__m256)__builtin_ia32_vscalefps256_round_mask(                             \
-      (__v8sf)(__m256)(A), (__v8sf)(__m256)(B), (__v8sf)_mm256_undefined_ps(), \
-      (__mmask8)-1, (int)(R)))
-
-#define _mm256_mask_scalef_round_ps(W, U, A, B, R)                             \
-  ((__m256)__builtin_ia32_vscalefps256_round_mask(                             \
-      (__v8sf)(__m256)(A), (__v8sf)(__m256)(B), (__v8sf)(__m256)(W),           \
-      (__mmask8)(U), (int)(R)))
-
-#define _mm256_maskz_scalef_round_ps(U, A, B, R)                               \
-  ((__m256)__builtin_ia32_vscalefps256_round_mask(                             \
-      (__v8sf)(__m256)(A), (__v8sf)(__m256)(B), (__v8sf)_mm256_setzero_ps(),   \
-      (__mmask8)(U), (int)(R)))
-
-#define _mm256_sqrt_round_pd(A, R)                                             \
-  ((__m256d)__builtin_ia32_vsqrtpd256_round((__v4df)(__m256d)(A), (int)(R)))
-
-#define _mm256_mask_sqrt_round_pd(W, U, A, R)                                  \
-  ((__m256d)__builtin_ia32_selectpd_256(                                       \
-      (__mmask8)(U), (__v4df)_mm256_sqrt_round_pd((A), (R)),                   \
-      (__v4df)(__m256d)(W)))
-
-#define _mm256_maskz_sqrt_round_pd(U, A, R)                                    \
-  ((__m256d)__builtin_ia32_selectpd_256(                                       \
-      (__mmask8)(U), (__v4df)_mm256_sqrt_round_pd((A), (R)),                   \
-      (__v4df)_mm256_setzero_pd()))
-
-#define _mm256_sqrt_round_ph(A, R)                                             \
-  ((__m256h)__builtin_ia32_vsqrtph256_round((__v16hf)(__m256h)(A), (int)(R)))
-
-#define _mm256_mask_sqrt_round_ph(W, U, A, R)                                  \
-  ((__m256h)__builtin_ia32_selectph_256(                                       \
-      (__mmask16)(U), (__v16hf)_mm256_sqrt_round_ph((A), (R)),                 \
-      (__v16hf)(__m256h)(W)))
-
-#define _mm256_maskz_sqrt_round_ph(U, A, R)                                    \
-  ((__m256h)__builtin_ia32_selectph_256(                                       \
-      (__mmask16)(U), (__v16hf)_mm256_sqrt_round_ph((A), (R)),                 \
-      (__v16hf)_mm256_setzero_ph()))
-
-#define _mm256_sqrt_round_ps(A, R)                                             \
-  ((__m256)__builtin_ia32_vsqrtps256_round((__v8sf)(__m256)(A), (int)(R)))
-
-#define _mm256_mask_sqrt_round_ps(W, U, A, R)                                  \
-  ((__m256)__builtin_ia32_selectps_256((__mmask8)(U),                          \
-                                       (__v8sf)_mm256_sqrt_round_ps((A), (R)), \
-                                       (__v8sf)(__m256)(W)))
-
-#define _mm256_maskz_sqrt_round_ps(U, A, R)                                    \
-  ((__m256)__builtin_ia32_selectps_256((__mmask8)(U),                          \
-                                       (__v8sf)_mm256_sqrt_round_ps((A), (R)), \
-                                       (__v8sf)_mm256_setzero_ps()))
-
-#define _mm256_sub_round_pd(A, B, R)                                           \
-  ((__m256d)__builtin_ia32_vsubpd256_round((__v4df)(__m256d)(A),               \
-                                           (__v4df)(__m256d)(B), (int)(R)))
-
-#define _mm256_mask_sub_round_pd(W, U, A, B, R)                                \
-  ((__m256d)__builtin_ia32_selectpd_256(                                       \
-      (__mmask8)(U), (__v4df)_mm256_sub_round_pd((A), (B), (R)),               \
-      (__v4df)(__m256d)(W)))
-
-#define _mm256_maskz_sub_round_pd(U, A, B, R)                                  \
-  ((__m256d)__builtin_ia32_selectpd_256(                                       \
-      (__mmask8)(U), (__v4df)_mm256_sub_round_pd((A), (B), (R)),               \
-      (__v4df)_mm256_setzero_pd()))
-
-#define _mm256_sub_round_ph(A, B, R)                                           \
-  ((__m256h)__builtin_ia32_vsubph256_round((__v16hf)(__m256h)(A),              \
-                                           (__v16hf)(__m256h)(B), (int)(R)))
-
-#define _mm256_mask_sub_round_ph(W, U, A, B, R)                                \
-  ((__m256h)__builtin_ia32_selectph_256(                                       \
-      (__mmask16)(U), (__v16hf)_mm256_sub_round_ph((A), (B), (R)),             \
-      (__v16hf)(__m256h)(W)))
-
-#define _mm256_maskz_sub_round_ph(U, A, B, R)                                  \
-  ((__m256h)__builtin_ia32_selectph_256(                                       \
-      (__mmask16)(U), (__v16hf)_mm256_sub_round_ph((A), (B), (R)),             \
-      (__v16hf)_mm256_setzero_ph()))
-
-#define _mm256_sub_round_ps(A, B, R)                                           \
-  ((__m256)__builtin_ia32_vsubps256_round((__v8sf)(__m256)(A),                 \
-                                          (__v8sf)(__m256)(B), (int)(R)))
-
-#define _mm256_mask_sub_round_ps(W, U, A, B, R)                                \
-  ((__m256)__builtin_ia32_selectps_256(                                        \
-      (__mmask8)(U), (__v8sf)_mm256_sub_round_ps((A), (B), (R)),               \
-      (__v8sf)(__m256)(W)))
-
-#define _mm256_maskz_sub_round_ps(U, A, B, R)                                  \
-  ((__m256)__builtin_ia32_selectps_256(                                        \
-      (__mmask8)(U), (__v8sf)_mm256_sub_round_ps((A), (B), (R)),               \
-      (__v8sf)_mm256_setzero_ps()))
-
 #undef __DEFAULT_FN_ATTRS256
 #undef __DEFAULT_FN_ATTRS128
 

lib/include/avx10_2satcvtdsintrin.h

@@ -71,175 +71,134 @@
 #endif /* __x86_64__ */
 
 // 128 Bit : Double -> int
-static __inline__ __m128i __DEFAULT_FN_ATTRS128 _mm_cvttspd_epi32(__m128d __A) {
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
+_mm_cvtts_pd_epi32(__m128d __A) {
   return ((__m128i)__builtin_ia32_vcvttpd2dqs128_mask(
       (__v2df)__A, (__v4si)(__m128i)_mm_undefined_si128(), (__mmask8)(-1)));
 }
 
 static __inline__ __m128i __DEFAULT_FN_ATTRS128
-_mm_mask_cvttspd_epi32(__m128i __W, __mmask8 __U, __m128d __A) {
+_mm_mask_cvtts_pd_epi32(__m128i __W, __mmask8 __U, __m128d __A) {
   return ((__m128i)__builtin_ia32_vcvttpd2dqs128_mask((__v2df)__A, (__v4si)__W,
                                                       __U));
 }
 
 static __inline__ __m128i __DEFAULT_FN_ATTRS128
-_mm_maskz_cvttspd_epi32(__mmask16 __U, __m128d __A) {
+_mm_maskz_cvtts_pd_epi32(__mmask16 __U, __m128d __A) {
   return ((__m128i)__builtin_ia32_vcvttpd2dqs128_mask(
       (__v2df)__A, (__v4si)(__m128i)_mm_setzero_si128(), __U));
 }
 
 // 256 Bit : Double -> int
 static __inline__ __m128i __DEFAULT_FN_ATTRS256
-_mm256_cvttspd_epi32(__m256d __A) {
-  return ((__m128i)__builtin_ia32_vcvttpd2dqs256_round_mask(
-      (__v4df)__A, (__v4si)_mm_undefined_si128(), (__mmask8)-1,
-      _MM_FROUND_CUR_DIRECTION));
+_mm256_cvtts_pd_epi32(__m256d __A) {
+  return ((__m128i)__builtin_ia32_vcvttpd2dqs256_mask(
+      (__v4df)__A, (__v4si)_mm_undefined_si128(), (__mmask8)-1));
 }
 
 static __inline__ __m128i __DEFAULT_FN_ATTRS256
-_mm256_mask_cvttspd_epi32(__m128i __W, __mmask8 __U, __m256d __A) {
-  return ((__m128i)__builtin_ia32_vcvttpd2dqs256_round_mask(
-      (__v4df)__A, (__v4si)__W, __U, _MM_FROUND_CUR_DIRECTION));
+_mm256_mask_cvtts_pd_epi32(__m128i __W, __mmask8 __U, __m256d __A) {
+  return ((__m128i)__builtin_ia32_vcvttpd2dqs256_mask((__v4df)__A, (__v4si)__W,
+                                                      __U));
 }
 
 static __inline__ __m128i __DEFAULT_FN_ATTRS256
-_mm256_maskz_cvttspd_epi32(__mmask8 __U, __m256d __A) {
-  return ((__m128i)__builtin_ia32_vcvttpd2dqs256_round_mask(
-      (__v4df)__A, (__v4si)_mm_setzero_si128(), __U, _MM_FROUND_CUR_DIRECTION));
+_mm256_maskz_cvtts_pd_epi32(__mmask8 __U, __m256d __A) {
+  return ((__m128i)__builtin_ia32_vcvttpd2dqs256_mask(
+      (__v4df)__A, (__v4si)_mm_setzero_si128(), __U));
 }
 
-#define _mm256_cvtts_roundpd_epi32(__A, __R)                                   \
-  ((__m128i)__builtin_ia32_vcvttpd2dqs256_round_mask(                          \
-      (__v4df)(__m256d)__A, (__v4si)(__m128i)_mm_undefined_si128(),            \
-      (__mmask8) - 1, (int)(__R)))
-
-#define _mm256_mask_cvtts_roundpd_epi32(__W, __U, __A, __R)                    \
-  ((__m128i)__builtin_ia32_vcvttpd2dqs256_round_mask(                          \
-      (__v4df)(__m256d)__A, (__v4si)(__m128i)__W, (__mmask8)__U, (int)(__R)))
-
-#define _mm256_maskz_cvtts_roundpd_epi32(__U, __A, __R)                        \
-  ((__m128i)__builtin_ia32_vcvttpd2dqs256_round_mask(                          \
-      (__v4df)(__m256d)__A, (__v4si)(__m128i)_mm_setzero_si128(),              \
-      (__mmask8)__U, (int)(__R)))
-
 // 128 Bit : Double -> uint
-static __inline__ __m128i __DEFAULT_FN_ATTRS128 _mm_cvttspd_epu32(__m128d __A) {
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
+_mm_cvtts_pd_epu32(__m128d __A) {
   return ((__m128i)__builtin_ia32_vcvttpd2udqs128_mask(
       (__v2df)__A, (__v4si)(__m128i)_mm_undefined_si128(), (__mmask8)(-1)));
 }
 
 static __inline__ __m128i __DEFAULT_FN_ATTRS128
-_mm_mask_cvttspd_epu32(__m128i __W, __mmask8 __U, __m128d __A) {
+_mm_mask_cvtts_pd_epu32(__m128i __W, __mmask8 __U, __m128d __A) {
   return ((__m128i)__builtin_ia32_vcvttpd2udqs128_mask(
       (__v2df)__A, (__v4si)(__m128i)__W, (__mmask8)__U));
 }
 
 static __inline__ __m128i __DEFAULT_FN_ATTRS128
-_mm_maskz_cvttspd_epu32(__mmask8 __U, __m128d __A) {
+_mm_maskz_cvtts_pd_epu32(__mmask8 __U, __m128d __A) {
   return ((__m128i)__builtin_ia32_vcvttpd2udqs128_mask(
       (__v2df)__A, (__v4si)(__m128i)_mm_setzero_si128(), __U));
 }
 
 // 256 Bit : Double -> uint
 static __inline__ __m128i __DEFAULT_FN_ATTRS256
-_mm256_cvttspd_epu32(__m256d __A) {
-  return ((__m128i)__builtin_ia32_vcvttpd2udqs256_round_mask(
-      (__v4df)__A, (__v4si)_mm_undefined_si128(), (__mmask8)-1,
-      _MM_FROUND_CUR_DIRECTION));
+_mm256_cvtts_pd_epu32(__m256d __A) {
+  return ((__m128i)__builtin_ia32_vcvttpd2udqs256_mask(
+      (__v4df)__A, (__v4si)_mm_undefined_si128(), (__mmask8)-1));
 }
 
 static __inline__ __m128i __DEFAULT_FN_ATTRS256
-_mm256_mask_cvttspd_epu32(__m128i __W, __mmask8 __U, __m256d __A) {
-  return ((__m128i)__builtin_ia32_vcvttpd2udqs256_round_mask(
-      (__v4df)__A, (__v4si)__W, __U, _MM_FROUND_CUR_DIRECTION));
+_mm256_mask_cvtts_pd_epu32(__m128i __W, __mmask8 __U, __m256d __A) {
+  return ((__m128i)__builtin_ia32_vcvttpd2udqs256_mask((__v4df)__A, (__v4si)__W,
+                                                       __U));
 }
 
 static __inline__ __m128i __DEFAULT_FN_ATTRS256
-_mm256_maskz_cvttspd_epu32(__mmask8 __U, __m256d __A) {
-  return ((__m128i)__builtin_ia32_vcvttpd2udqs256_round_mask(
-      (__v4df)__A, (__v4si)_mm_setzero_si128(), __U, _MM_FROUND_CUR_DIRECTION));
+_mm256_maskz_cvtts_pd_epu32(__mmask8 __U, __m256d __A) {
+  return ((__m128i)__builtin_ia32_vcvttpd2udqs256_mask(
+      (__v4df)__A, (__v4si)_mm_setzero_si128(), __U));
 }
 
-#define _mm256_cvtts_roundpd_epu32(__A, __R)                                   \
-  ((__m128i)__builtin_ia32_vcvttpd2udqs256_round_mask(                         \
-      (__v4df)(__m256d)__A, (__v4si)(__m128i)_mm_undefined_si128(),            \
-      (__mmask8) - 1, (int)(__R)))
-
-#define _mm256_mask_cvtts_roundpd_epu32(__W, __U, __A, __R)                    \
-  ((__m128i)__builtin_ia32_vcvttpd2udqs256_round_mask(                         \
-      (__v4df)(__m256d)__A, (__v4si)(__m128i)__W, (__mmask8)__U, (int)(__R)))
-
-#define _mm256_maskz_cvtts_roundpd_epu32(__U, __A, __R)                        \
-  ((__m128i)__builtin_ia32_vcvttpd2udqs256_round_mask(                         \
-      (__v4df)(__m256d)__A, (__v4si)(__m128i)_mm_setzero_si128(),              \
-      (__mmask8)__U, (int)(__R)))
-
 // 128 Bit : Double -> long
-static __inline__ __m128i __DEFAULT_FN_ATTRS128 _mm_cvttspd_epi64(__m128d __A) {
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
+_mm_cvtts_pd_epi64(__m128d __A) {
   return ((__m128i)__builtin_ia32_vcvttpd2qqs128_mask(
       (__v2df)__A, (__v2di)_mm_undefined_si128(), (__mmask8)-1));
 }
 
 static __inline__ __m128i __DEFAULT_FN_ATTRS128
-_mm_mask_cvttspd_epi64(__m128i __W, __mmask8 __U, __m128d __A) {
+_mm_mask_cvtts_pd_epi64(__m128i __W, __mmask8 __U, __m128d __A) {
   return ((__m128i)__builtin_ia32_vcvttpd2qqs128_mask((__v2df)__A, (__v2di)__W,
                                                       (__mmask8)__U));
 }
 
 static __inline__ __m128i __DEFAULT_FN_ATTRS128
-_mm_maskz_cvttspd_epi64(__mmask8 __U, __m128d __A) {
+_mm_maskz_cvtts_pd_epi64(__mmask8 __U, __m128d __A) {
   return ((__m128i)__builtin_ia32_vcvttpd2qqs128_mask(
       (__v2df)__A, (__v2di)_mm_setzero_si128(), (__mmask8)__U));
 }
 
 // 256 Bit : Double -> long
 static __inline__ __m256i __DEFAULT_FN_ATTRS256
-_mm256_cvttspd_epi64(__m256d __A) {
-  return ((__m256i)__builtin_ia32_vcvttpd2qqs256_round_mask(
-      (__v4df)__A, (__v4di)_mm256_undefined_si256(), (__mmask8)-1,
-      _MM_FROUND_CUR_DIRECTION));
+_mm256_cvtts_pd_epi64(__m256d __A) {
+  return ((__m256i)__builtin_ia32_vcvttpd2qqs256_mask(
+      (__v4df)__A, (__v4di)_mm256_undefined_si256(), (__mmask8)-1));
 }
 
 static __inline__ __m256i __DEFAULT_FN_ATTRS256
-_mm256_mask_cvttspd_epi64(__m256i __W, __mmask8 __U, __m256d __A) {
-  return ((__m256i)__builtin_ia32_vcvttpd2qqs256_round_mask(
-      (__v4df)__A, (__v4di)__W, __U, _MM_FROUND_CUR_DIRECTION));
+_mm256_mask_cvtts_pd_epi64(__m256i __W, __mmask8 __U, __m256d __A) {
+  return ((__m256i)__builtin_ia32_vcvttpd2qqs256_mask((__v4df)__A, (__v4di)__W,
+                                                      __U));
 }
 
 static __inline__ __m256i __DEFAULT_FN_ATTRS256
-_mm256_maskz_cvttspd_epi64(__mmask8 __U, __m256d __A) {
-  return ((__m256i)__builtin_ia32_vcvttpd2qqs256_round_mask(
-      (__v4df)__A, (__v4di)_mm256_setzero_si256(), __U,
-      _MM_FROUND_CUR_DIRECTION));
+_mm256_maskz_cvtts_pd_epi64(__mmask8 __U, __m256d __A) {
+  return ((__m256i)__builtin_ia32_vcvttpd2qqs256_mask(
+      (__v4df)__A, (__v4di)_mm256_setzero_si256(), __U));
 }
 
-#define _mm256_cvtts_roundpd_epi64(__A, __R)                                   \
-  ((__m256i)__builtin_ia32_vcvttpd2qqs256_round_mask(                          \
-      (__v4df)__A, (__v4di)_mm256_undefined_si256(), (__mmask8) - 1,           \
-      (int)__R))
-
-#define _mm256_mask_cvtts_roundpd_epi64(__W, __U, __A, __R)                    \
-  ((__m256i)__builtin_ia32_vcvttpd2qqs256_round_mask((__v4df)__A, (__v4di)__W, \
-                                                     (__mmask8)__U, (int)__R))
-
-#define _mm256_maskz_cvtts_roundpd_epi64(__U, __A, __R)                        \
-  ((__m256i)__builtin_ia32_vcvttpd2qqs256_round_mask(                          \
-      (__v4df)__A, (__v4di)_mm256_setzero_si256(), (__mmask8)__U, (int)__R))
-
 // 128 Bit : Double -> ulong
-static __inline__ __m128i __DEFAULT_FN_ATTRS128 _mm_cvttspd_epu64(__m128d __A) {
+static __inline__ __m128i __DEFAULT_FN_ATTRS128
+_mm_cvtts_pd_epu64(__m128d __A) {
   return ((__m128i)__builtin_ia32_vcvttpd2uqqs128_mask(
       (__v2df)__A, (__v2di)_mm_undefined_si128(), (__mmask8)-1));
 }
 
 static __inline__ __m128i __DEFAULT_FN_ATTRS128
-_mm_mask_cvttspd_epu64(__m128i __W, __mmask8 __U, __m128d __A) {
+_mm_mask_cvtts_pd_epu64(__m128i __W, __mmask8 __U, __m128d __A) {
   return ((__m128i)__builtin_ia32_vcvttpd2uqqs128_mask((__v2df)__A, (__v2di)__W,
                                                        (__mmask8)__U));
 }
 
 static __inline__ __m128i __DEFAULT_FN_ATTRS128
-_mm_maskz_cvttspd_epu64(__mmask8 __U, __m128d __A) {
+_mm_maskz_cvtts_pd_epu64(__mmask8 __U, __m128d __A) {
   return ((__m128i)__builtin_ia32_vcvttpd2uqqs128_mask(
       (__v2df)__A, (__v2di)_mm_setzero_si128(), (__mmask8)__U));
 }
@@ -247,105 +206,74 @@ _mm_maskz_cvttspd_epu64(__mmask8 __U, __m128d __A) {
 // 256 Bit : Double -> ulong
 
 static __inline__ __m256i __DEFAULT_FN_ATTRS256
-_mm256_cvttspd_epu64(__m256d __A) {
-  return ((__m256i)__builtin_ia32_vcvttpd2uqqs256_round_mask(
-      (__v4df)__A, (__v4di)_mm256_undefined_si256(), (__mmask8)-1,
-      _MM_FROUND_CUR_DIRECTION));
+_mm256_cvtts_pd_epu64(__m256d __A) {
+  return ((__m256i)__builtin_ia32_vcvttpd2uqqs256_mask(
+      (__v4df)__A, (__v4di)_mm256_undefined_si256(), (__mmask8)-1));
 }
 
 static __inline__ __m256i __DEFAULT_FN_ATTRS256
-_mm256_mask_cvttspd_epu64(__m256i __W, __mmask8 __U, __m256d __A) {
-  return ((__m256i)__builtin_ia32_vcvttpd2uqqs256_round_mask(
-      (__v4df)__A, (__v4di)__W, __U, _MM_FROUND_CUR_DIRECTION));
+_mm256_mask_cvtts_pd_epu64(__m256i __W, __mmask8 __U, __m256d __A) {
+  return ((__m256i)__builtin_ia32_vcvttpd2uqqs256_mask((__v4df)__A, (__v4di)__W,
+                                                       __U));
 }
 
 static __inline__ __m256i __DEFAULT_FN_ATTRS256
-_mm256_maskz_cvttspd_epu64(__mmask8 __U, __m256d __A) {
-  return ((__m256i)__builtin_ia32_vcvttpd2uqqs256_round_mask(
-      (__v4df)__A, (__v4di)_mm256_setzero_si256(), __U,
-      _MM_FROUND_CUR_DIRECTION));
+_mm256_maskz_cvtts_pd_epu64(__mmask8 __U, __m256d __A) {
+  return ((__m256i)__builtin_ia32_vcvttpd2uqqs256_mask(
+      (__v4df)__A, (__v4di)_mm256_setzero_si256(), __U));
 }
 
-#define _mm256_cvtts_roundpd_epu64(__A, __R)                                   \
-  ((__m256i)__builtin_ia32_vcvttpd2uqqs256_round_mask(                         \
-      (__v4df)__A, (__v4di)_mm256_undefined_si256(), (__mmask8) - 1,           \
-      (int)__R))
-
-#define _mm256_mask_cvtts_roundpd_epu64(__W, __U, __A, __R)                    \
-  ((__m256i)__builtin_ia32_vcvttpd2uqqs256_round_mask(                         \
-      (__v4df)__A, (__v4di)__W, (__mmask8)__U, (int)__R))
-
-#define _mm256_maskz_cvtts_roundpd_epu64(__U, __A, __R)                        \
-  ((__m256i)__builtin_ia32_vcvttpd2uqqs256_round_mask(                         \
-      (__v4df)__A, (__v4di)_mm256_setzero_si256(), (__mmask8)__U, (int)__R))
-
 // 128 Bit : float -> int
-static __inline__ __m128i __DEFAULT_FN_ATTRS128 _mm_cvttsps_epi32(__m128 __A) {
+static __inline__ __m128i __DEFAULT_FN_ATTRS128 _mm_cvtts_ps_epi32(__m128 __A) {
   return ((__m128i)__builtin_ia32_vcvttps2dqs128_mask(
       (__v4sf)__A, (__v4si)(__m128i)_mm_undefined_si128(), (__mmask8)(-1)));
 }
 
 static __inline__ __m128i __DEFAULT_FN_ATTRS128
-_mm_mask_cvttsps_epi32(__m128i __W, __mmask8 __U, __m128 __A) {
+_mm_mask_cvtts_ps_epi32(__m128i __W, __mmask8 __U, __m128 __A) {
   return ((__m128i)__builtin_ia32_vcvttps2dqs128_mask((__v4sf)__A, (__v4si)__W,
                                                       (__mmask8)__U));
 }
 
 static __inline__ __m128i __DEFAULT_FN_ATTRS128
-_mm_maskz_cvttsps_epi32(__mmask8 __U, __m128 __A) {
+_mm_maskz_cvtts_ps_epi32(__mmask8 __U, __m128 __A) {
   return ((__m128i)__builtin_ia32_vcvttps2dqs128_mask(
       (__v4sf)__A, (__v4si)(__m128i)_mm_setzero_si128(), (__mmask8)__U));
 }
 
 // 256 Bit : float -> int
 static __inline__ __m256i __DEFAULT_FN_ATTRS256
-_mm256_cvttsps_epi32(__m256 __A) {
-  return ((__m256i)__builtin_ia32_vcvttps2dqs256_round_mask(
-      (__v8sf)__A, (__v8si)_mm256_undefined_si256(), (__mmask8)-1,
-      _MM_FROUND_CUR_DIRECTION));
+_mm256_cvtts_ps_epi32(__m256 __A) {
+  return ((__m256i)__builtin_ia32_vcvttps2dqs256_mask(
+      (__v8sf)__A, (__v8si)_mm256_undefined_si256(), (__mmask8)-1));
 }
 
 static __inline__ __m256i __DEFAULT_FN_ATTRS256
-_mm256_mask_cvttsps_epi32(__m256i __W, __mmask8 __U, __m256 __A) {
-  return ((__m256i)__builtin_ia32_vcvttps2dqs256_round_mask(
-      (__v8sf)__A, (__v8si)__W, __U, _MM_FROUND_CUR_DIRECTION));
+_mm256_mask_cvtts_ps_epi32(__m256i __W, __mmask8 __U, __m256 __A) {
+  return ((__m256i)__builtin_ia32_vcvttps2dqs256_mask((__v8sf)__A, (__v8si)__W,
+                                                      __U));
 }
 
 static __inline__ __m256i __DEFAULT_FN_ATTRS256
-_mm256_maskz_cvttsps_epi32(__mmask8 __U, __m256 __A) {
-  return ((__m256i)__builtin_ia32_vcvttps2dqs256_round_mask(
-      (__v8sf)__A, (__v8si)_mm256_setzero_si256(), __U,
-      _MM_FROUND_CUR_DIRECTION));
+_mm256_maskz_cvtts_ps_epi32(__mmask8 __U, __m256 __A) {
+  return ((__m256i)__builtin_ia32_vcvttps2dqs256_mask(
+      (__v8sf)__A, (__v8si)_mm256_setzero_si256(), __U));
 }
 
-#define _mm256_cvtts_roundps_epi32(__A, __R)                                   \
-  ((__m256i)__builtin_ia32_vcvttps2dqs256_round_mask(                          \
-      (__v8sf)(__m256)__A, (__v8si)(__m256i)_mm256_undefined_si256(),          \
-      (__mmask8) - 1, (int)(__R)))
-
-#define _mm256_mask_cvtts_roundps_epi32(__W, __U, __A, __R)                    \
-  ((__m256i)__builtin_ia32_vcvttps2dqs256_round_mask(                          \
-      (__v8sf)(__m256)__A, (__v8si)(__m256i)__W, (__mmask8)__U, (int)(__R)))
-
-#define _mm256_maskz_cvtts_roundps_epi32(__U, __A, __R)                        \
-  ((__m256i)__builtin_ia32_vcvttps2dqs256_round_mask(                          \
-      (__v8sf)(__m256)__A, (__v8si)(__m256i)_mm256_setzero_si256(),            \
-      (__mmask8)__U, (int)(__R)))
-
 // 128 Bit : float -> uint
-static __inline__ __m128i __DEFAULT_FN_ATTRS128 _mm_cvttsps_epu32(__m128 __A) {
+static __inline__ __m128i __DEFAULT_FN_ATTRS128 _mm_cvtts_ps_epu32(__m128 __A) {
   return ((__m128i)__builtin_ia32_vcvttps2udqs128_mask(
       (__v4sf)__A, (__v4si)(__m128i)_mm_undefined_si128(), (__mmask8)(-1)));
 }
 
 static __inline__ __m128i __DEFAULT_FN_ATTRS128
-_mm_mask_cvttsps_epu32(__m128i __W, __mmask8 __U, __m128 __A) {
+_mm_mask_cvtts_ps_epu32(__m128i __W, __mmask8 __U, __m128 __A) {
   return ((__m128i)__builtin_ia32_vcvttps2udqs128_mask((__v4sf)__A, (__v4si)__W,
                                                        (__mmask8)__U));
 }
 
 static __inline__ __m128i __DEFAULT_FN_ATTRS128
-_mm_maskz_cvttsps_epu32(__mmask8 __U, __m128 __A) {
+_mm_maskz_cvtts_ps_epu32(__mmask8 __U, __m128 __A) {
   return ((__m128i)__builtin_ia32_vcvttps2udqs128_mask(
       (__v4sf)__A, (__v4si)_mm_setzero_si128(), (__mmask8)__U));
 }
@@ -353,144 +281,96 @@ _mm_maskz_cvttsps_epu32(__mmask8 __U, __m128 __A) {
 // 256 Bit : float -> uint
 
 static __inline__ __m256i __DEFAULT_FN_ATTRS256
-_mm256_cvttsps_epu32(__m256 __A) {
-  return ((__m256i)__builtin_ia32_vcvttps2udqs256_round_mask(
-      (__v8sf)__A, (__v8si)_mm256_undefined_si256(), (__mmask8)-1,
-      _MM_FROUND_CUR_DIRECTION));
+_mm256_cvtts_ps_epu32(__m256 __A) {
+  return ((__m256i)__builtin_ia32_vcvttps2udqs256_mask(
+      (__v8sf)__A, (__v8si)_mm256_undefined_si256(), (__mmask8)-1));
 }
 
 static __inline__ __m256i __DEFAULT_FN_ATTRS256
-_mm256_mask_cvttsps_epu32(__m256i __W, __mmask8 __U, __m256 __A) {
-  return ((__m256i)__builtin_ia32_vcvttps2udqs256_round_mask(
-      (__v8sf)__A, (__v8si)__W, __U, _MM_FROUND_CUR_DIRECTION));
+_mm256_mask_cvtts_ps_epu32(__m256i __W, __mmask8 __U, __m256 __A) {
+  return ((__m256i)__builtin_ia32_vcvttps2udqs256_mask((__v8sf)__A, (__v8si)__W,
+                                                       __U));
 }
 
 static __inline__ __m256i __DEFAULT_FN_ATTRS256
-_mm256_maskz_cvttsps_epu32(__mmask8 __U, __m256 __A) {
-  return ((__m256i)__builtin_ia32_vcvttps2udqs256_round_mask(
-      (__v8sf)__A, (__v8si)_mm256_setzero_si256(), __U,
-      _MM_FROUND_CUR_DIRECTION));
+_mm256_maskz_cvtts_ps_epu32(__mmask8 __U, __m256 __A) {
+  return ((__m256i)__builtin_ia32_vcvttps2udqs256_mask(
+      (__v8sf)__A, (__v8si)_mm256_setzero_si256(), __U));
 }
 
-#define _mm256_cvtts_roundps_epu32(__A, __R)                                   \
-  ((__m256i)__builtin_ia32_vcvttps2udqs256_round_mask(                         \
-      (__v8sf)(__m256)__A, (__v8si)(__m256i)_mm256_undefined_si256(),          \
-      (__mmask8) - 1, (int)(__R)))
-
-#define _mm256_mask_cvtts_roundps_epu32(__W, __U, __A, __R)                    \
-  ((__m256i)__builtin_ia32_vcvttps2udqs256_round_mask(                         \
-      (__v8sf)(__m256)__A, (__v8si)(__m256i)__W, (__mmask8)__U, (int)(__R)))
-
-#define _mm256_maskz_cvtts_roundps_epu32(__U, __A, __R)                        \
-  ((__m256i)__builtin_ia32_vcvttps2udqs256_round_mask(                         \
-      (__v8sf)(__m256)__A, (__v8si)(__m256i)_mm256_setzero_si256(),            \
-      (__mmask8)__U, (int)(__R)))
-
 // 128 bit : float -> long
-static __inline__ __m128i __DEFAULT_FN_ATTRS128 _mm_cvttsps_epi64(__m128 __A) {
+static __inline__ __m128i __DEFAULT_FN_ATTRS128 _mm_cvtts_ps_epi64(__m128 __A) {
   return ((__m128i)__builtin_ia32_vcvttps2qqs128_mask(
       (__v4sf)__A, (__v2di)_mm_undefined_si128(), (__mmask8)-1));
 }
 
 static __inline__ __m128i __DEFAULT_FN_ATTRS128
-_mm_mask_cvttsps_epi64(__m128i __W, __mmask8 __U, __m128 __A) {
+_mm_mask_cvtts_ps_epi64(__m128i __W, __mmask8 __U, __m128 __A) {
   return ((__m128i)__builtin_ia32_vcvttps2qqs128_mask(
       (__v4sf)__A, (__v2di)(__m128i)__W, (__mmask8)__U));
 }
 
 static __inline__ __m128i __DEFAULT_FN_ATTRS128
-_mm_maskz_cvttsps_epi64(__mmask8 __U, __m128 __A) {
+_mm_maskz_cvtts_ps_epi64(__mmask8 __U, __m128 __A) {
   return ((__m128i)__builtin_ia32_vcvttps2qqs128_mask(
       (__v4sf)__A, (__v2di)_mm_setzero_si128(), (__mmask8)__U));
 }
 // 256 bit : float -> long
 
 static __inline__ __m256i __DEFAULT_FN_ATTRS256
-_mm256_cvttsps_epi64(__m128 __A) {
-  return ((__m256i)__builtin_ia32_vcvttps2qqs256_round_mask(
-      (__v4sf)__A, (__v4di)_mm256_undefined_si256(), (__mmask8)-1,
-      _MM_FROUND_CUR_DIRECTION));
+_mm256_cvtts_ps_epi64(__m128 __A) {
+  return ((__m256i)__builtin_ia32_vcvttps2qqs256_mask(
+      (__v4sf)__A, (__v4di)_mm256_undefined_si256(), (__mmask8)-1));
 }
 static __inline__ __m256i __DEFAULT_FN_ATTRS256
-_mm256_mask_cvttsps_epi64(__m256i __W, __mmask8 __U, __m128 __A) {
-  return ((__m256i)__builtin_ia32_vcvttps2qqs256_round_mask(
-      (__v4sf)__A, (__v4di)__W, __U, _MM_FROUND_CUR_DIRECTION));
+_mm256_mask_cvtts_ps_epi64(__m256i __W, __mmask8 __U, __m128 __A) {
+  return ((__m256i)__builtin_ia32_vcvttps2qqs256_mask((__v4sf)__A, (__v4di)__W,
+                                                      __U));
 }
 
 static __inline__ __m256i __DEFAULT_FN_ATTRS256
-_mm256_maskz_cvttsps_epi64(__mmask8 __U, __m128 __A) {
-  return ((__m256i)__builtin_ia32_vcvttps2qqs256_round_mask(
-      (__v4sf)__A, (__v4di)_mm256_setzero_si256(), __U,
-      _MM_FROUND_CUR_DIRECTION));
+_mm256_maskz_cvtts_ps_epi64(__mmask8 __U, __m128 __A) {
+  return ((__m256i)__builtin_ia32_vcvttps2qqs256_mask(
+      (__v4sf)__A, (__v4di)_mm256_setzero_si256(), __U));
 }
 
-#define _mm256_cvtts_roundps_epi64(__A, __R)                                   \
-  ((__m256i)__builtin_ia32_vcvttps2qqs256_round_mask(                          \
-      (__v4sf)(__m128)__A, (__v4di)_mm256_undefined_si256(), (__mmask8) - 1,   \
-      (int)__R))
-
-#define _mm256_mask_cvtts_roundps_epi64(__W, __U, __A, __R)                    \
-  ((__m256i)__builtin_ia32_vcvttps2qqs256_round_mask(                          \
-      (__v4sf)(__m128)__A, (__v4di)__W, (__mmask8)__U, (int)__R))
-
-#define _mm256_maskz_cvtts_roundps_epi64(__U, __A, __R)                        \
-  ((__m256i)__builtin_ia32_vcvttps2qqs256_round_mask(                          \
-      (__v4sf)(__m128)__A, (__v4di)_mm256_setzero_si256(), (__mmask8)__U,      \
-      (int)__R))
-
 // 128 bit : float -> ulong
-static __inline__ __m128i __DEFAULT_FN_ATTRS128 _mm_cvttsps_epu64(__m128 __A) {
+static __inline__ __m128i __DEFAULT_FN_ATTRS128 _mm_cvtts_ps_epu64(__m128 __A) {
   return ((__m128i)__builtin_ia32_vcvttps2uqqs128_mask(
       (__v4sf)__A, (__v2di)_mm_undefined_si128(), (__mmask8)-1));
 }
 
 static __inline__ __m128i __DEFAULT_FN_ATTRS128
-_mm_mask_cvttsps_epu64(__m128i __W, __mmask8 __U, __m128 __A) {
+_mm_mask_cvtts_ps_epu64(__m128i __W, __mmask8 __U, __m128 __A) {
   return ((__m128i)__builtin_ia32_vcvttps2uqqs128_mask(
       (__v4sf)__A, (__v2di)(__m128i)__W, (__mmask8)__U));
 }
 
 static __inline__ __m128i __DEFAULT_FN_ATTRS128
-_mm_maskz_cvttsps_epu64(__mmask8 __U, __m128 __A) {
+_mm_maskz_cvtts_ps_epu64(__mmask8 __U, __m128 __A) {
   return ((__m128i)__builtin_ia32_vcvttps2uqqs128_mask(
       (__v4sf)__A, (__v2di)_mm_setzero_si128(), (__mmask8)__U));
 }
 // 256 bit : float -> ulong
 
 static __inline__ __m256i __DEFAULT_FN_ATTRS256
-_mm256_cvttsps_epu64(__m128 __A) {
-  return ((__m256i)__builtin_ia32_vcvttps2uqqs256_round_mask(
-      (__v4sf)__A, (__v4di)_mm256_undefined_si256(), (__mmask8)-1,
-      _MM_FROUND_CUR_DIRECTION));
+_mm256_cvtts_ps_epu64(__m128 __A) {
+  return ((__m256i)__builtin_ia32_vcvttps2uqqs256_mask(
+      (__v4sf)__A, (__v4di)_mm256_undefined_si256(), (__mmask8)-1));
 }
 
 static __inline__ __m256i __DEFAULT_FN_ATTRS256
-_mm256_mask_cvttsps_epu64(__m256i __W, __mmask8 __U, __m128 __A) {
-  return ((__m256i)__builtin_ia32_vcvttps2uqqs256_round_mask(
-      (__v4sf)__A, (__v4di)__W, __U, _MM_FROUND_CUR_DIRECTION));
+_mm256_mask_cvtts_ps_epu64(__m256i __W, __mmask8 __U, __m128 __A) {
+  return ((__m256i)__builtin_ia32_vcvttps2uqqs256_mask((__v4sf)__A, (__v4di)__W,
+                                                       __U));
 }
 
 static __inline__ __m256i __DEFAULT_FN_ATTRS256
-_mm256_maskz_cvttsps_epu64(__mmask8 __U, __m128 __A) {
-  return ((__m256i)__builtin_ia32_vcvttps2uqqs256_round_mask(
-      (__v4sf)__A, (__v4di)_mm256_setzero_si256(), __U,
-      _MM_FROUND_CUR_DIRECTION));
+_mm256_maskz_cvtts_ps_epu64(__mmask8 __U, __m128 __A) {
+  return ((__m256i)__builtin_ia32_vcvttps2uqqs256_mask(
+      (__v4sf)__A, (__v4di)_mm256_setzero_si256(), __U));
 }
 
-#define _mm256_cvtts_roundps_epu64(__A, __R)                                   \
-  ((__m256i)__builtin_ia32_vcvttps2uqqs256_round_mask(                         \
-      (__v4sf)(__m128)__A, (__v4di)_mm256_undefined_si256(), (__mmask8) - 1,   \
-      (int)__R))
-
-#define _mm256_mask_cvtts_roundps_epu64(__W, __U, __A, __R)                    \
-  ((__m256i)__builtin_ia32_vcvttps2uqqs256_round_mask(                         \
-      (__v4sf)(__m128)__A, (__v4di)__W, (__mmask8)__U, (int)__R))
-
-#define _mm256_maskz_cvtts_roundps_epu64(__U, __A, __R)                        \
-  ((__m256i)__builtin_ia32_vcvttps2uqqs256_round_mask(                         \
-      (__v4sf)(__m128)__A, (__v4di)_mm256_setzero_si256(), (__mmask8)__U,      \
-      (int)__R))
-
 #undef __DEFAULT_FN_ATTRS128
 #undef __DEFAULT_FN_ATTRS256
 #endif // __AVX10_2SATCVTDSINTRIN_H

lib/include/avx10_2satcvtintrin.h

@@ -14,431 +14,299 @@
 #ifndef __AVX10_2SATCVTINTRIN_H
 #define __AVX10_2SATCVTINTRIN_H
 
-#define _mm_ipcvtbf16_epi8(A)                                                  \
+#define _mm_ipcvts_bf16_epi8(A)                                                \
   ((__m128i)__builtin_ia32_vcvtbf162ibs128((__v8bf)(__m128bh)(A)))
 
-#define _mm_mask_ipcvtbf16_epi8(W, U, A)                                       \
+#define _mm_mask_ipcvts_bf16_epi8(W, U, A)                                     \
   ((__m128i)__builtin_ia32_selectw_128(                                        \
-      (__mmask8)(U), (__v8hi)_mm_ipcvtbf16_epi8(A), (__v8hi)(__m128i)(W)))
+      (__mmask8)(U), (__v8hi)_mm_ipcvts_bf16_epi8(A), (__v8hi)(__m128i)(W)))
 
-#define _mm_maskz_ipcvtbf16_epi8(U, A)                                         \
+#define _mm_maskz_ipcvts_bf16_epi8(U, A)                                       \
   ((__m128i)__builtin_ia32_selectw_128((__mmask8)(U),                          \
-                                       (__v8hi)_mm_ipcvtbf16_epi8(A),          \
+                                       (__v8hi)_mm_ipcvts_bf16_epi8(A),        \
                                        (__v8hi)_mm_setzero_si128()))
 
-#define _mm256_ipcvtbf16_epi8(A)                                               \
+#define _mm256_ipcvts_bf16_epi8(A)                                             \
   ((__m256i)__builtin_ia32_vcvtbf162ibs256((__v16bf)(__m256bh)(A)))
 
-#define _mm256_mask_ipcvtbf16_epi8(W, U, A)                                    \
+#define _mm256_mask_ipcvts_bf16_epi8(W, U, A)                                  \
   ((__m256i)__builtin_ia32_selectw_256((__mmask16)(U),                         \
-                                       (__v16hi)_mm256_ipcvtbf16_epi8(A),      \
+                                       (__v16hi)_mm256_ipcvts_bf16_epi8(A),    \
                                        (__v16hi)(__m256i)(W)))
 
-#define _mm256_maskz_ipcvtbf16_epi8(U, A)                                      \
+#define _mm256_maskz_ipcvts_bf16_epi8(U, A)                                    \
   ((__m256i)__builtin_ia32_selectw_256((__mmask16)(U),                         \
-                                       (__v16hi)_mm256_ipcvtbf16_epi8(A),      \
+                                       (__v16hi)_mm256_ipcvts_bf16_epi8(A),    \
                                        (__v16hi)_mm256_setzero_si256()))
 
-#define _mm_ipcvtbf16_epu8(A)                                                  \
+#define _mm_ipcvts_bf16_epu8(A)                                                \
   ((__m128i)__builtin_ia32_vcvtbf162iubs128((__v8bf)(__m128bh)(A)))
 
-#define _mm_mask_ipcvtbf16_epu8(W, U, A)                                       \
+#define _mm_mask_ipcvts_bf16_epu8(W, U, A)                                     \
   ((__m128i)__builtin_ia32_selectw_128(                                        \
-      (__mmask8)(U), (__v8hi)_mm_ipcvtbf16_epu8(A), (__v8hi)(__m128i)(W)))
+      (__mmask8)(U), (__v8hi)_mm_ipcvts_bf16_epu8(A), (__v8hi)(__m128i)(W)))
 
-#define _mm_maskz_ipcvtbf16_epu8(U, A)                                         \
+#define _mm_maskz_ipcvts_bf16_epu8(U, A)                                       \
   ((__m128i)__builtin_ia32_selectw_128((__mmask8)(U),                          \
-                                       (__v8hi)_mm_ipcvtbf16_epu8(A),          \
+                                       (__v8hi)_mm_ipcvts_bf16_epu8(A),        \
                                        (__v8hi)_mm_setzero_si128()))
 
-#define _mm256_ipcvtbf16_epu8(A)                                               \
+#define _mm256_ipcvts_bf16_epu8(A)                                             \
   ((__m256i)__builtin_ia32_vcvtbf162iubs256((__v16bf)(__m256bh)(A)))
 
-#define _mm256_mask_ipcvtbf16_epu8(W, U, A)                                    \
+#define _mm256_mask_ipcvts_bf16_epu8(W, U, A)                                  \
   ((__m256i)__builtin_ia32_selectw_256((__mmask16)(U),                         \
-                                       (__v16hi)_mm256_ipcvtbf16_epu8(A),      \
+                                       (__v16hi)_mm256_ipcvts_bf16_epu8(A),    \
                                        (__v16hi)(__m256i)(W)))
 
-#define _mm256_maskz_ipcvtbf16_epu8(U, A)                                      \
+#define _mm256_maskz_ipcvts_bf16_epu8(U, A)                                    \
   ((__m256i)__builtin_ia32_selectw_256((__mmask16)(U),                         \
-                                       (__v16hi)_mm256_ipcvtbf16_epu8(A),      \
+                                       (__v16hi)_mm256_ipcvts_bf16_epu8(A),    \
                                        (__v16hi)_mm256_setzero_si256()))
 
-#define _mm_ipcvtph_epi8(A)                                                    \
+#define _mm_ipcvts_ph_epi8(A)                                                  \
   ((__m128i)__builtin_ia32_vcvtph2ibs128_mask(                                 \
       (__v8hf)(__m128h)(A), (__v8hu)_mm_setzero_si128(), (__mmask8)-1))
 
-#define _mm_mask_ipcvtph_epi8(W, U, A)                                         \
+#define _mm_mask_ipcvts_ph_epi8(W, U, A)                                       \
   ((__m128i)__builtin_ia32_vcvtph2ibs128_mask((__v8hf)(__m128h)(A),            \
                                               (__v8hu)(W), (__mmask8)(U)))
 
-#define _mm_maskz_ipcvtph_epi8(U, A)                                           \
+#define _mm_maskz_ipcvts_ph_epi8(U, A)                                         \
   ((__m128i)__builtin_ia32_vcvtph2ibs128_mask(                                 \
       (__v8hf)(__m128h)(A), (__v8hu)(_mm_setzero_si128()), (__mmask8)(U)))
 
-#define _mm256_ipcvtph_epi8(A)                                                 \
+#define _mm256_ipcvts_ph_epi8(A)                                               \
   ((__m256i)__builtin_ia32_vcvtph2ibs256_mask(                                 \
-      (__v16hf)(__m256h)(A), (__v16hu)_mm256_setzero_si256(), (__mmask16)-1,   \
-      _MM_FROUND_CUR_DIRECTION))
+      (__v16hf)(__m256h)(A), (__v16hu)_mm256_setzero_si256(), (__mmask16)-1))
 
-#define _mm256_mask_ipcvtph_epi8(W, U, A)                                      \
+#define _mm256_mask_ipcvts_ph_epi8(W, U, A)                                    \
   ((__m256i)__builtin_ia32_vcvtph2ibs256_mask((__v16hf)(__m256h)(A),           \
-                                              (__v16hu)(W), (__mmask16)(U),    \
-                                              _MM_FROUND_CUR_DIRECTION))
+                                              (__v16hu)(W), (__mmask16)(U)))
 
-#define _mm256_maskz_ipcvtph_epi8(U, A)                                        \
+#define _mm256_maskz_ipcvts_ph_epi8(U, A)                                      \
   ((__m256i)__builtin_ia32_vcvtph2ibs256_mask(                                 \
       (__v16hf)(__m256h)(A), (__v16hu)(_mm256_setzero_si256()),                \
-      (__mmask16)(U), _MM_FROUND_CUR_DIRECTION))
+      (__mmask16)(U)))
 
-#define _mm256_ipcvt_roundph_epi8(A, R)                                        \
-  ((__m256i)__builtin_ia32_vcvtph2ibs256_mask((__v16hf)(__m256h)(A),           \
-                                              (__v16hu)_mm256_setzero_si256(), \
-                                              (__mmask16)-1, (const int)R))
-
-#define _mm256_mask_ipcvt_roundph_epi8(W, U, A, R)                             \
-  ((__m256i)__builtin_ia32_vcvtph2ibs256_mask(                                 \
-      (__v16hf)(__m256h)(A), (__v16hu)(W), (__mmask16)(U), (const int)R))
-
-#define _mm256_maskz_ipcvt_roundph_epi8(U, A, R)                               \
-  ((__m256i)__builtin_ia32_vcvtph2ibs256_mask((__v16hf)(__m256h)(A),           \
-                                              (__v16hu)_mm256_setzero_si256(), \
-                                              (__mmask16)(U), (const int)R))
-
-#define _mm_ipcvtph_epu8(A)                                                    \
+#define _mm_ipcvts_ph_epu8(A)                                                  \
   ((__m128i)__builtin_ia32_vcvtph2iubs128_mask(                                \
       (__v8hf)(__m128h)(A), (__v8hu)_mm_setzero_si128(), (__mmask8)-1))
 
-#define _mm_mask_ipcvtph_epu8(W, U, A)                                         \
+#define _mm_mask_ipcvts_ph_epu8(W, U, A)                                       \
   ((__m128i)__builtin_ia32_vcvtph2iubs128_mask((__v8hf)(__m128h)(A),           \
                                                (__v8hu)(W), (__mmask8)(U)))
 
-#define _mm_maskz_ipcvtph_epu8(U, A)                                           \
+#define _mm_maskz_ipcvts_ph_epu8(U, A)                                         \
   ((__m128i)__builtin_ia32_vcvtph2iubs128_mask(                                \
       (__v8hf)(__m128h)(A), (__v8hu)(_mm_setzero_si128()), (__mmask8)(U)))
 
-#define _mm256_ipcvtph_epu8(A)                                                 \
+#define _mm256_ipcvts_ph_epu8(A)                                               \
   ((__m256i)__builtin_ia32_vcvtph2iubs256_mask(                                \
-      (__v16hf)(__m256h)(A), (__v16hu)_mm256_setzero_si256(), (__mmask16)-1,   \
-      _MM_FROUND_CUR_DIRECTION))
+      (__v16hf)(__m256h)(A), (__v16hu)_mm256_setzero_si256(), (__mmask16)-1))
 
-#define _mm256_mask_ipcvtph_epu8(W, U, A)                                      \
+#define _mm256_mask_ipcvts_ph_epu8(W, U, A)                                    \
   ((__m256i)__builtin_ia32_vcvtph2iubs256_mask((__v16hf)(__m256h)(A),          \
-                                               (__v16hu)(W), (__mmask16)(U),   \
-                                               _MM_FROUND_CUR_DIRECTION))
+                                               (__v16hu)(W), (__mmask16)(U)))
 
-#define _mm256_maskz_ipcvtph_epu8(U, A)                                        \
+#define _mm256_maskz_ipcvts_ph_epu8(U, A)                                      \
   ((__m256i)__builtin_ia32_vcvtph2iubs256_mask(                                \
       (__v16hf)(__m256h)(A), (__v16hu)(_mm256_setzero_si256()),                \
-      (__mmask16)(U), _MM_FROUND_CUR_DIRECTION))
+      (__mmask16)(U)))
 
-#define _mm256_ipcvt_roundph_epu8(A, R)                                        \
-  ((__m256i)__builtin_ia32_vcvtph2iubs256_mask(                                \
-      (__v16hf)(__m256h)(A), (__v16hu)_mm256_setzero_si256(), (__mmask16)-1,   \
-      (const int)R))
-
-#define _mm256_mask_ipcvt_roundph_epu8(W, U, A, R)                             \
-  ((__m256i)__builtin_ia32_vcvtph2iubs256_mask(                                \
-      (__v16hf)(__m256h)(A), (__v16hu)(W), (__mmask16)(U), (const int)R))
-
-#define _mm256_maskz_ipcvt_roundph_epu8(U, A, R)                               \
-  ((__m256i)__builtin_ia32_vcvtph2iubs256_mask(                                \
-      (__v16hf)(__m256h)(A), (__v16hu)_mm256_setzero_si256(), (__mmask16)(U),  \
-      (const int)R))
-
-#define _mm_ipcvtps_epi8(A)                                                    \
+#define _mm_ipcvts_ps_epi8(A)                                                  \
   ((__m128i)__builtin_ia32_vcvtps2ibs128_mask(                                 \
       (__v4sf)(__m128)(A), (__v4su)_mm_setzero_si128(), (__mmask8)-1))
 
-#define _mm_mask_ipcvtps_epi8(W, U, A)                                         \
+#define _mm_mask_ipcvts_ps_epi8(W, U, A)                                       \
   ((__m128i)__builtin_ia32_vcvtps2ibs128_mask((__v4sf)(__m128)(A),             \
                                               (__v4su)(W), (__mmask8)(U)))
 
-#define _mm_maskz_ipcvtps_epi8(U, A)                                           \
+#define _mm_maskz_ipcvts_ps_epi8(U, A)                                         \
   ((__m128i)__builtin_ia32_vcvtps2ibs128_mask(                                 \
       (__v4sf)(__m128)(A), (__v4su)(_mm_setzero_si128()), (__mmask8)(U)))
 
-#define _mm256_ipcvtps_epi8(A)                                                 \
+#define _mm256_ipcvts_ps_epi8(A)                                               \
   ((__m256i)__builtin_ia32_vcvtps2ibs256_mask(                                 \
-      (__v8sf)(__m256)(A), (__v8su)_mm256_setzero_si256(), (__mmask8)-1,       \
-      _MM_FROUND_CUR_DIRECTION))
+      (__v8sf)(__m256)(A), (__v8su)_mm256_setzero_si256(), (__mmask8)-1))
 
-#define _mm256_mask_ipcvtps_epi8(W, U, A)                                      \
+#define _mm256_mask_ipcvts_ps_epi8(W, U, A)                                    \
   ((__m256i)__builtin_ia32_vcvtps2ibs256_mask((__v8sf)(__m256)(A),             \
-                                              (__v8su)(W), (__mmask8)(U),      \
-                                              _MM_FROUND_CUR_DIRECTION))
+                                              (__v8su)(W), (__mmask8)(U)))
 
-#define _mm256_maskz_ipcvtps_epi8(U, A)                                        \
+#define _mm256_maskz_ipcvts_ps_epi8(U, A)                                      \
   ((__m256i)__builtin_ia32_vcvtps2ibs256_mask(                                 \
-      (__v8sf)(__m256)(A), (__v8su)(_mm256_setzero_si256()), (__mmask8)(U),    \
-      _MM_FROUND_CUR_DIRECTION))
+      (__v8sf)(__m256)(A), (__v8su)(_mm256_setzero_si256()), (__mmask8)(U)))
 
-#define _mm256_ipcvt_roundps_epi8(A, R)                                        \
-  ((__m256i)__builtin_ia32_vcvtps2ibs256_mask((__v8sf)(__m256)(A),             \
-                                              (__v8su)_mm256_setzero_si256(),  \
-                                              (__mmask8)-1, (const int)R))
-
-#define _mm256_mask_ipcvt_roundps_epi8(W, U, A, R)                             \
-  ((__m256i)__builtin_ia32_vcvtps2ibs256_mask(                                 \
-      (__v8sf)(__m256)(A), (__v8su)(W), (__mmask8)(U), (const int)R))
-
-#define _mm256_maskz_ipcvt_roundps_epi8(U, A, R)                               \
-  ((__m256i)__builtin_ia32_vcvtps2ibs256_mask((__v8sf)(__m256)(A),             \
-                                              (__v8su)_mm256_setzero_si256(),  \
-                                              (__mmask8)(U), (const int)R))
-
-#define _mm_ipcvtps_epu8(A)                                                    \
+#define _mm_ipcvts_ps_epu8(A)                                                  \
   ((__m128i)__builtin_ia32_vcvtps2iubs128_mask(                                \
       (__v4sf)(__m128)(A), (__v4su)_mm_setzero_si128(), (__mmask8)-1))
 
-#define _mm_mask_ipcvtps_epu8(W, U, A)                                         \
+#define _mm_mask_ipcvts_ps_epu8(W, U, A)                                       \
   ((__m128i)__builtin_ia32_vcvtps2iubs128_mask((__v4sf)(__m128)(A),            \
                                                (__v4su)(W), (__mmask8)(U)))
 
-#define _mm_maskz_ipcvtps_epu8(U, A)                                           \
+#define _mm_maskz_ipcvts_ps_epu8(U, A)                                         \
   ((__m128i)__builtin_ia32_vcvtps2iubs128_mask(                                \
       (__v4sf)(__m128)(A), (__v4su)(_mm_setzero_si128()), (__mmask8)(U)))
 
-#define _mm256_ipcvtps_epu8(A)                                                 \
+#define _mm256_ipcvts_ps_epu8(A)                                               \
   ((__m256i)__builtin_ia32_vcvtps2iubs256_mask(                                \
-      (__v8sf)(__m256)(A), (__v8su)_mm256_setzero_si256(), (__mmask8)-1,       \
-      _MM_FROUND_CUR_DIRECTION))
+      (__v8sf)(__m256)(A), (__v8su)_mm256_setzero_si256(), (__mmask8)-1))
 
-#define _mm256_mask_ipcvtps_epu8(W, U, A)                                      \
+#define _mm256_mask_ipcvts_ps_epu8(W, U, A)                                    \
   ((__m256i)__builtin_ia32_vcvtps2iubs256_mask((__v8sf)(__m256)(A),            \
-                                               (__v8su)(W), (__mmask8)(U),     \
-                                               _MM_FROUND_CUR_DIRECTION))
+                                               (__v8su)(W), (__mmask8)(U)))
 
-#define _mm256_maskz_ipcvtps_epu8(U, A)                                        \
+#define _mm256_maskz_ipcvts_ps_epu8(U, A)                                      \
   ((__m256i)__builtin_ia32_vcvtps2iubs256_mask(                                \
-      (__v8sf)(__m256)(A), (__v8su)(_mm256_setzero_si256()), (__mmask8)(U),    \
-      _MM_FROUND_CUR_DIRECTION))
+      (__v8sf)(__m256)(A), (__v8su)(_mm256_setzero_si256()), (__mmask8)(U)))
 
-#define _mm256_ipcvt_roundps_epu8(A, R)                                        \
-  ((__m256i)__builtin_ia32_vcvtps2iubs256_mask((__v8sf)(__m256)(A),            \
-                                               (__v8su)_mm256_setzero_si256(), \
-                                               (__mmask8)-1, (const int)R))
-
-#define _mm256_mask_ipcvt_roundps_epu8(W, U, A, R)                             \
-  ((__m256i)__builtin_ia32_vcvtps2iubs256_mask(                                \
-      (__v8sf)(__m256)(A), (__v8su)(W), (__mmask8)(U), (const int)R))
-
-#define _mm256_maskz_ipcvt_roundps_epu8(U, A, R)                               \
-  ((__m256i)__builtin_ia32_vcvtps2iubs256_mask((__v8sf)(__m256)(A),            \
-                                               (__v8su)_mm256_setzero_si256(), \
-                                               (__mmask8)(U), (const int)R))
-
-#define _mm_ipcvttbf16_epi8(A)                                                 \
+#define _mm_ipcvtts_bf16_epi8(A)                                               \
   ((__m128i)__builtin_ia32_vcvttbf162ibs128((__v8bf)(__m128bh)(A)))
 
-#define _mm_mask_ipcvttbf16_epi8(W, U, A)                                      \
+#define _mm_mask_ipcvtts_bf16_epi8(W, U, A)                                    \
   ((__m128i)__builtin_ia32_selectw_128(                                        \
-      (__mmask8)(U), (__v8hi)_mm_ipcvttbf16_epi8(A), (__v8hi)(__m128i)(W)))
+      (__mmask8)(U), (__v8hi)_mm_ipcvtts_bf16_epi8(A), (__v8hi)(__m128i)(W)))
 
-#define _mm_maskz_ipcvttbf16_epi8(U, A)                                        \
+#define _mm_maskz_ipcvtts_bf16_epi8(U, A)                                      \
   ((__m128i)__builtin_ia32_selectw_128((__mmask8)(U),                          \
-                                       (__v8hi)_mm_ipcvttbf16_epi8(A),         \
+                                       (__v8hi)_mm_ipcvtts_bf16_epi8(A),       \
                                        (__v8hi)_mm_setzero_si128()))
 
-#define _mm256_ipcvttbf16_epi8(A)                                              \
+#define _mm256_ipcvtts_bf16_epi8(A)                                            \
   ((__m256i)__builtin_ia32_vcvttbf162ibs256((__v16bf)(__m256bh)(A)))
 
-#define _mm256_mask_ipcvttbf16_epi8(W, U, A)                                   \
+#define _mm256_mask_ipcvtts_bf16_epi8(W, U, A)                                 \
   ((__m256i)__builtin_ia32_selectw_256((__mmask16)(U),                         \
-                                       (__v16hi)_mm256_ipcvttbf16_epi8(A),     \
+                                       (__v16hi)_mm256_ipcvtts_bf16_epi8(A),   \
                                        (__v16hi)(__m256i)(W)))
 
-#define _mm256_maskz_ipcvttbf16_epi8(U, A)                                     \
+#define _mm256_maskz_ipcvtts_bf16_epi8(U, A)                                   \
   ((__m256i)__builtin_ia32_selectw_256((__mmask16)(U),                         \
-                                       (__v16hi)_mm256_ipcvttbf16_epi8(A),     \
+                                       (__v16hi)_mm256_ipcvtts_bf16_epi8(A),   \
                                        (__v16hi)_mm256_setzero_si256()))
 
-#define _mm_ipcvttbf16_epu8(A)                                                 \
+#define _mm_ipcvtts_bf16_epu8(A)                                               \
   ((__m128i)__builtin_ia32_vcvttbf162iubs128((__v8bf)(__m128bh)(A)))
 
-#define _mm_mask_ipcvttbf16_epu8(W, U, A)                                      \
+#define _mm_mask_ipcvtts_bf16_epu8(W, U, A)                                    \
   ((__m128i)__builtin_ia32_selectw_128(                                        \
-      (__mmask8)(U), (__v8hi)_mm_ipcvttbf16_epu8(A), (__v8hi)(__m128i)(W)))
+      (__mmask8)(U), (__v8hi)_mm_ipcvtts_bf16_epu8(A), (__v8hi)(__m128i)(W)))
 
-#define _mm_maskz_ipcvttbf16_epu8(U, A)                                        \
+#define _mm_maskz_ipcvtts_bf16_epu8(U, A)                                      \
   ((__m128i)__builtin_ia32_selectw_128((__mmask8)(U),                          \
-                                       (__v8hi)_mm_ipcvttbf16_epu8(A),         \
+                                       (__v8hi)_mm_ipcvtts_bf16_epu8(A),       \
                                        (__v8hi)_mm_setzero_si128()))
 
-#define _mm256_ipcvttbf16_epu8(A)                                              \
+#define _mm256_ipcvtts_bf16_epu8(A)                                            \
   ((__m256i)__builtin_ia32_vcvttbf162iubs256((__v16bf)(__m256bh)(A)))
 
-#define _mm256_mask_ipcvttbf16_epu8(W, U, A)                                   \
+#define _mm256_mask_ipcvtts_bf16_epu8(W, U, A)                                 \
   ((__m256i)__builtin_ia32_selectw_256((__mmask16)(U),                         \
-                                       (__v16hi)_mm256_ipcvttbf16_epu8(A),     \
+                                       (__v16hi)_mm256_ipcvtts_bf16_epu8(A),   \
                                        (__v16hi)(__m256i)(W)))
 
-#define _mm256_maskz_ipcvttbf16_epu8(U, A)                                     \
+#define _mm256_maskz_ipcvtts_bf16_epu8(U, A)                                   \
   ((__m256i)__builtin_ia32_selectw_256((__mmask16)(U),                         \
-                                       (__v16hi)_mm256_ipcvttbf16_epu8(A),     \
+                                       (__v16hi)_mm256_ipcvtts_bf16_epu8(A),   \
                                        (__v16hi)_mm256_setzero_si256()))
 
-#define _mm_ipcvttph_epi8(A)                                                   \
+#define _mm_ipcvtts_ph_epi8(A)                                                 \
   ((__m128i)__builtin_ia32_vcvttph2ibs128_mask(                                \
       (__v8hf)(__m128h)(A), (__v8hu)_mm_setzero_si128(), (__mmask8)-1))
 
-#define _mm_mask_ipcvttph_epi8(W, U, A)                                        \
+#define _mm_mask_ipcvtts_ph_epi8(W, U, A)                                      \
   ((__m128i)__builtin_ia32_vcvttph2ibs128_mask((__v8hf)(__m128h)(A),           \
                                                (__v8hu)(W), (__mmask8)(U)))
 
-#define _mm_maskz_ipcvttph_epi8(U, A)                                          \
+#define _mm_maskz_ipcvtts_ph_epi8(U, A)                                        \
   ((__m128i)__builtin_ia32_vcvttph2ibs128_mask(                                \
       (__v8hf)(__m128h)(A), (__v8hu)(_mm_setzero_si128()), (__mmask8)(U)))
 
-#define _mm256_ipcvttph_epi8(A)                                                \
+#define _mm256_ipcvtts_ph_epi8(A)                                              \
   ((__m256i)__builtin_ia32_vcvttph2ibs256_mask(                                \
-      (__v16hf)(__m256h)(A), (__v16hu)_mm256_setzero_si256(), (__mmask16)-1,   \
-      _MM_FROUND_CUR_DIRECTION))
+      (__v16hf)(__m256h)(A), (__v16hu)_mm256_setzero_si256(), (__mmask16)-1))
 
-#define _mm256_mask_ipcvttph_epi8(W, U, A)                                     \
+#define _mm256_mask_ipcvtts_ph_epi8(W, U, A)                                   \
   ((__m256i)__builtin_ia32_vcvttph2ibs256_mask((__v16hf)(__m256h)(A),          \
-                                               (__v16hu)(W), (__mmask16)(U),   \
-                                               _MM_FROUND_CUR_DIRECTION))
+                                               (__v16hu)(W), (__mmask16)(U)))
 
-#define _mm256_maskz_ipcvttph_epi8(U, A)                                       \
+#define _mm256_maskz_ipcvtts_ph_epi8(U, A)                                     \
   ((__m256i)__builtin_ia32_vcvttph2ibs256_mask(                                \
       (__v16hf)(__m256h)(A), (__v16hu)(_mm256_setzero_si256()),                \
-      (__mmask16)(U), _MM_FROUND_CUR_DIRECTION))
+      (__mmask16)(U)))
 
-#define _mm256_ipcvtt_roundph_epi8(A, R)                                       \
-  ((__m256i)__builtin_ia32_vcvttph2ibs256_mask(                                \
-      (__v16hf)(__m256h)(A), (__v16hu)_mm256_setzero_si256(), (__mmask16)-1,   \
-      (const int)R))
-
-#define _mm256_mask_ipcvtt_roundph_epi8(W, U, A, R)                            \
-  ((__m256i)__builtin_ia32_vcvttph2ibs256_mask(                                \
-      (__v16hf)(__m256h)(A), (__v16hu)(W), (__mmask16)(U), (const int)R))
-
-#define _mm256_maskz_ipcvtt_roundph_epi8(U, A, R)                              \
-  ((__m256i)__builtin_ia32_vcvttph2ibs256_mask(                                \
-      (__v16hf)(__m256h)(A), (__v16hu)_mm256_setzero_si256(), (__mmask16)(U),  \
-      (const int)R))
-
-#define _mm_ipcvttph_epu8(A)                                                   \
+#define _mm_ipcvtts_ph_epu8(A)                                                 \
   ((__m128i)__builtin_ia32_vcvttph2iubs128_mask(                               \
       (__v8hf)(__m128h)(A), (__v8hu)_mm_setzero_si128(), (__mmask8)-1))
 
-#define _mm_mask_ipcvttph_epu8(W, U, A)                                        \
+#define _mm_mask_ipcvtts_ph_epu8(W, U, A)                                      \
   ((__m128i)__builtin_ia32_vcvttph2iubs128_mask((__v8hf)(__m128h)(A),          \
                                                 (__v8hu)(W), (__mmask8)(U)))
 
-#define _mm_maskz_ipcvttph_epu8(U, A)                                          \
+#define _mm_maskz_ipcvtts_ph_epu8(U, A)                                        \
   ((__m128i)__builtin_ia32_vcvttph2iubs128_mask(                               \
       (__v8hf)(__m128h)(A), (__v8hu)(_mm_setzero_si128()), (__mmask8)(U)))
 
-#define _mm256_ipcvttph_epu8(A)                                                \
+#define _mm256_ipcvtts_ph_epu8(A)                                              \
   ((__m256i)__builtin_ia32_vcvttph2iubs256_mask(                               \
-      (__v16hf)(__m256h)(A), (__v16hu)_mm256_setzero_si256(), (__mmask16)-1,   \
-      _MM_FROUND_CUR_DIRECTION))
+      (__v16hf)(__m256h)(A), (__v16hu)_mm256_setzero_si256(), (__mmask16)-1))
 
-#define _mm256_mask_ipcvttph_epu8(W, U, A)                                     \
+#define _mm256_mask_ipcvtts_ph_epu8(W, U, A)                                   \
   ((__m256i)__builtin_ia32_vcvttph2iubs256_mask((__v16hf)(__m256h)(A),         \
-                                                (__v16hu)(W), (__mmask16)(U),  \
-                                                _MM_FROUND_CUR_DIRECTION))
+                                                (__v16hu)(W), (__mmask16)(U)))
 
-#define _mm256_maskz_ipcvttph_epu8(U, A)                                       \
+#define _mm256_maskz_ipcvtts_ph_epu8(U, A)                                     \
   ((__m256i)__builtin_ia32_vcvttph2iubs256_mask(                               \
       (__v16hf)(__m256h)(A), (__v16hu)(_mm256_setzero_si256()),                \
-      (__mmask16)(U), _MM_FROUND_CUR_DIRECTION))
+      (__mmask16)(U)))
 
-#define _mm256_ipcvtt_roundph_epu8(A, R)                                       \
-  ((__m256i)__builtin_ia32_vcvttph2iubs256_mask(                               \
-      (__v16hf)(__m256h)(A), (__v16hu)_mm256_setzero_si256(), (__mmask16)-1,   \
-      (const int)R))
-
-#define _mm256_mask_ipcvtt_roundph_epu8(W, U, A, R)                            \
-  ((__m256i)__builtin_ia32_vcvttph2iubs256_mask(                               \
-      (__v16hf)(__m256h)(A), (__v16hu)(W), (__mmask16)(U), (const int)R))
-
-#define _mm256_maskz_ipcvtt_roundph_epu8(U, A, R)                              \
-  ((__m256i)__builtin_ia32_vcvttph2iubs256_mask(                               \
-      (__v16hf)(__m256h)(A), (__v16hu)_mm256_setzero_si256(), (__mmask16)(U),  \
-      (const int)R))
-
-#define _mm_ipcvttps_epi8(A)                                                   \
+#define _mm_ipcvtts_ps_epi8(A)                                                 \
   ((__m128i)__builtin_ia32_vcvttps2ibs128_mask(                                \
       (__v4sf)(__m128)(A), (__v4su)_mm_setzero_si128(), (__mmask8)-1))
 
-#define _mm_mask_ipcvttps_epi8(W, U, A)                                        \
+#define _mm_mask_ipcvtts_ps_epi8(W, U, A)                                      \
   ((__m128i)__builtin_ia32_vcvttps2ibs128_mask((__v4sf)(__m128)(A),            \
                                                (__v4su)(W), (__mmask8)(U)))
 
-#define _mm_maskz_ipcvttps_epi8(U, A)                                          \
+#define _mm_maskz_ipcvtts_ps_epi8(U, A)                                        \
   ((__m128i)__builtin_ia32_vcvttps2ibs128_mask(                                \
       (__v4sf)(__m128)(A), (__v4su)(_mm_setzero_si128()), (__mmask8)(U)))
 
-#define _mm256_ipcvttps_epi8(A)                                                \
+#define _mm256_ipcvtts_ps_epi8(A)                                              \
   ((__m256i)__builtin_ia32_vcvttps2ibs256_mask(                                \
-      (__v8sf)(__m256)(A), (__v8su)_mm256_setzero_si256(), (__mmask8)-1,       \
-      _MM_FROUND_CUR_DIRECTION))
+      (__v8sf)(__m256)(A), (__v8su)_mm256_setzero_si256(), (__mmask8)-1))
 
-#define _mm256_mask_ipcvttps_epi8(W, U, A)                                     \
+#define _mm256_mask_ipcvtts_ps_epi8(W, U, A)                                   \
   ((__m256i)__builtin_ia32_vcvttps2ibs256_mask((__v8sf)(__m256)(A),            \
-                                               (__v8su)(W), (__mmask8)(U),     \
-                                               _MM_FROUND_CUR_DIRECTION))
+                                               (__v8su)(W), (__mmask8)(U)))
 
-#define _mm256_maskz_ipcvttps_epi8(U, A)                                       \
+#define _mm256_maskz_ipcvtts_ps_epi8(U, A)                                     \
   ((__m256i)__builtin_ia32_vcvttps2ibs256_mask(                                \
-      (__v8sf)(__m256)(A), (__v8su)(_mm256_setzero_si256()), (__mmask8)(U),    \
-      _MM_FROUND_CUR_DIRECTION))
+      (__v8sf)(__m256)(A), (__v8su)(_mm256_setzero_si256()), (__mmask8)(U)))
 
-#define _mm256_ipcvtt_roundps_epi8(A, R)                                       \
-  ((__m256i)__builtin_ia32_vcvttps2ibs256_mask((__v8sf)(__m256)(A),            \
-                                               (__v8su)_mm256_setzero_si256(), \
-                                               (__mmask8)-1, (const int)R))
-
-#define _mm256_mask_ipcvtt_roundps_epi8(W, U, A, R)                            \
-  ((__m256i)__builtin_ia32_vcvttps2ibs256_mask(                                \
-      (__v8sf)(__m256)(A), (__v8su)(W), (__mmask8)(U), (const int)R))
-
-#define _mm256_maskz_ipcvtt_roundps_epi8(U, A, R)                              \
-  ((__m256i)__builtin_ia32_vcvttps2ibs256_mask((__v8sf)(__m256)(A),            \
-                                               (__v8su)_mm256_setzero_si256(), \
-                                               (__mmask8)(U), (const int)R))
-
-#define _mm_ipcvttps_epu8(A)                                                   \
+#define _mm_ipcvtts_ps_epu8(A)                                                 \
   ((__m128i)__builtin_ia32_vcvttps2iubs128_mask(                               \
       (__v4sf)(__m128)(A), (__v4su)_mm_setzero_si128(), (__mmask8)-1))
 
-#define _mm_mask_ipcvttps_epu8(W, U, A)                                        \
+#define _mm_mask_ipcvtts_ps_epu8(W, U, A)                                      \
   ((__m128i)__builtin_ia32_vcvttps2iubs128_mask((__v4sf)(__m128)(A),           \
                                                 (__v4su)(W), (__mmask8)(U)))
 
-#define _mm_maskz_ipcvttps_epu8(U, A)                                          \
+#define _mm_maskz_ipcvtts_ps_epu8(U, A)                                        \
   ((__m128i)__builtin_ia32_vcvttps2iubs128_mask(                               \
       (__v4sf)(__m128)(A), (__v4su)(_mm_setzero_si128()), (__mmask8)(U)))
 
-#define _mm256_ipcvttps_epu8(A)                                                \
+#define _mm256_ipcvtts_ps_epu8(A)                                              \
   ((__m256i)__builtin_ia32_vcvttps2iubs256_mask(                               \
-      (__v8sf)(__m256)(A), (__v8su)_mm256_setzero_si256(), (__mmask8)-1,       \
-      _MM_FROUND_CUR_DIRECTION))
+      (__v8sf)(__m256)(A), (__v8su)_mm256_setzero_si256(), (__mmask8)-1))
 
-#define _mm256_mask_ipcvttps_epu8(W, U, A)                                     \
+#define _mm256_mask_ipcvtts_ps_epu8(W, U, A)                                   \
   ((__m256i)__builtin_ia32_vcvttps2iubs256_mask((__v8sf)(__m256)(A),           \
-                                                (__v8su)(W), (__mmask8)(U),    \
-                                                _MM_FROUND_CUR_DIRECTION))
+                                                (__v8su)(W), (__mmask8)(U)))
 
-#define _mm256_maskz_ipcvttps_epu8(U, A)                                       \
+#define _mm256_maskz_ipcvtts_ps_epu8(U, A)                                     \
   ((__m256i)__builtin_ia32_vcvttps2iubs256_mask(                               \
-      (__v8sf)(__m256)(A), (__v8su)(_mm256_setzero_si256()), (__mmask8)(U),    \
-      _MM_FROUND_CUR_DIRECTION))
-
-#define _mm256_ipcvtt_roundps_epu8(A, R)                                       \
-  ((__m256i)__builtin_ia32_vcvttps2iubs256_mask(                               \
-      (__v8sf)(__m256)(A), (__v8su)_mm256_setzero_si256(), (__mmask8)-1,       \
-      (const int)R))
-
-#define _mm256_mask_ipcvtt_roundps_epu8(W, U, A, R)                            \
-  ((__m256i)__builtin_ia32_vcvttps2iubs256_mask(                               \
-      (__v8sf)(__m256)(A), (__v8su)(W), (__mmask8)(U), (const int)R))
-
-#define _mm256_maskz_ipcvtt_roundps_epu8(U, A, R)                              \
-  ((__m256i)__builtin_ia32_vcvttps2iubs256_mask(                               \
-      (__v8sf)(__m256)(A), (__v8su)_mm256_setzero_si256(), (__mmask8)(U),      \
-      (const int)R))
+      (__v8sf)(__m256)(A), (__v8su)(_mm256_setzero_si256()), (__mmask8)(U)))
 #endif // __AVX10_2SATCVTINTRIN_H

lib/include/avx512fp16intrin.h

@@ -553,7 +553,8 @@ static __inline__ __m512h __DEFAULT_FN_ATTRS512 _mm512_abs_ph(__m512h __A) {
 }
 
 static __inline__ __m512h __DEFAULT_FN_ATTRS512 _mm512_conj_pch(__m512h __A) {
-  return (__m512h)_mm512_xor_ps((__m512)__A, _mm512_set1_ps(-0.0f));
+  return (__m512h)_mm512_xor_epi32((__m512i)__A,
+                                   _mm512_set1_epi32(-2147483648));
 }
 
 static __inline__ __m512h __DEFAULT_FN_ATTRS512

lib/include/bmiintrin.h

@@ -161,8 +161,6 @@ _mm_tzcnt_64(unsigned long long __X) {
 
 #undef __RELAXED_FN_ATTRS
 
-#if !defined(__SCE__) || __has_feature(modules) || defined(__BMI__)
-
 /* Define the default attributes for the functions in this file. */
 #if defined(__cplusplus) && (__cplusplus >= 201103L)
 #define __DEFAULT_FN_ATTRS                                                     \
@@ -603,6 +601,4 @@ __blsr_u64(unsigned long long __X) {
 
 #undef __DEFAULT_FN_ATTRS
 
-#endif /* !defined(__SCE__) || __has_feature(modules) || defined(__BMI__) */
-
 #endif /* __BMIINTRIN_H */

lib/include/cpuid.h

@@ -267,18 +267,18 @@
                   : "0"(__leaf), "2"(__count))
 #else
 /* x86-64 uses %rbx as the base register, so preserve it. */
-#define __cpuid(__leaf, __eax, __ebx, __ecx, __edx) \
-    __asm("  xchgq  %%rbx,%q1\n" \
-          "  cpuid\n" \
-          "  xchgq  %%rbx,%q1" \
-        : "=a"(__eax), "=r" (__ebx), "=c"(__ecx), "=d"(__edx) \
+#define __cpuid(__leaf, __eax, __ebx, __ecx, __edx)                            \
+  __asm("  xchg{q|}  {%%|}rbx,%q1\n"                                           \
+        "  cpuid\n"                                                            \
+        "  xchg{q|}  {%%|}rbx,%q1"                                             \
+        : "=a"(__eax), "=r"(__ebx), "=c"(__ecx), "=d"(__edx)                   \
         : "0"(__leaf))
 
-#define __cpuid_count(__leaf, __count, __eax, __ebx, __ecx, __edx) \
-    __asm("  xchgq  %%rbx,%q1\n" \
-          "  cpuid\n" \
-          "  xchgq  %%rbx,%q1" \
-        : "=a"(__eax), "=r" (__ebx), "=c"(__ecx), "=d"(__edx) \
+#define __cpuid_count(__leaf, __count, __eax, __ebx, __ecx, __edx)             \
+  __asm("  xchg{q|}  {%%|}rbx,%q1\n"                                           \
+        "  cpuid\n"                                                            \
+        "  xchg{q|}  {%%|}rbx,%q1"                                             \
+        : "=a"(__eax), "=r"(__ebx), "=c"(__ecx), "=d"(__edx)                   \
         : "0"(__leaf), "2"(__count))
 #endif
 
@@ -289,20 +289,22 @@ static __inline unsigned int __get_cpuid_max (unsigned int __leaf,
 #ifdef __i386__
     int __cpuid_supported;
 
-    __asm("  pushfl\n"
-          "  popl   %%eax\n"
-          "  movl   %%eax,%%ecx\n"
-          "  xorl   $0x00200000,%%eax\n"
-          "  pushl  %%eax\n"
-          "  popfl\n"
-          "  pushfl\n"
-          "  popl   %%eax\n"
-          "  movl   $0,%0\n"
-          "  cmpl   %%eax,%%ecx\n"
+    __asm("  pushf{l|d}\n"
+          "  pop{l|}   {%%|}eax\n"
+          "  mov{l|}   {%%eax,%%ecx|ecx,eax}\n"
+          "  xor{l|}   {$0x00200000,%%eax|eax,0x00200000}\n"
+          "  push{l|}  {%%|}eax\n"
+          "  popf{l|d}\n"
+          "  pushf{l|d}\n"
+          "  pop{l|}   {%%|}eax\n"
+          "  mov{l|}   {$0,%0|%0,0}\n"
+          "  cmp{l|}   {%%eax,%%ecx|ecx,eax}\n"
           "  je     1f\n"
-          "  movl   $1,%0\n"
+          "  mov{l|}   {$1,%0|%0,1}\n"
           "1:"
-        : "=r" (__cpuid_supported) : : "eax", "ecx");
+          : "=r"(__cpuid_supported)
+          :
+          : "eax", "ecx");
     if (!__cpuid_supported)
         return 0;
 #endif

lib/include/float.h

@@ -18,21 +18,12 @@
  * additional definitions provided for Windows.
  * For more details see http://msdn.microsoft.com/en-us/library/y0ybw9fy.aspx
  *
- * Also fall back on Darwin and AIX to allow additional definitions and
+ * Also fall back on AIX to allow additional definitions and
  * implementation-defined values.
  */
-#if (defined(__APPLE__) || defined(__MINGW32__) || defined(_MSC_VER) ||        \
-     defined(_AIX)) &&                                                         \
+#if (defined(__MINGW32__) || defined(_MSC_VER) || defined(_AIX)) &&            \
     __STDC_HOSTED__ && __has_include_next(<float.h>)
 
-/* Prior to Apple's 10.7 SDK, float.h SDK header used to apply an extra level
- * of #include_next<float.h> to keep Metrowerks compilers happy. Avoid this
- * extra indirection.
- */
-#ifdef __APPLE__
-#define _FLOAT_H_
-#endif
-
 #  include_next <float.h>
 
 /* Undefine anything that we'll be redefining below. */

lib/include/immintrin.h

@@ -16,231 +16,112 @@
 
 #include <x86gprintrin.h>
 
-#if !defined(__SCE__) || __has_feature(modules) || defined(__MMX__)
 #include <mmintrin.h>
-#endif
 
-#if !defined(__SCE__) || __has_feature(modules) || defined(__SSE__)
 #include <xmmintrin.h>
-#endif
 
-#if !defined(__SCE__) || __has_feature(modules) || defined(__SSE2__)
 #include <emmintrin.h>
-#endif
 
-#if !defined(__SCE__) || __has_feature(modules) || defined(__SSE3__)
 #include <pmmintrin.h>
-#endif
 
-#if !defined(__SCE__) || __has_feature(modules) || defined(__SSSE3__)
 #include <tmmintrin.h>
-#endif
 
-#if !defined(__SCE__) || __has_feature(modules) ||                             \
-    (defined(__SSE4_2__) || defined(__SSE4_1__))
 #include <smmintrin.h>
-#endif
 
-#if !defined(__SCE__) || __has_feature(modules) ||                             \
-    (defined(__AES__) || defined(__PCLMUL__))
 #include <wmmintrin.h>
-#endif
 
-#if !defined(__SCE__) || __has_feature(modules) || defined(__CLFLUSHOPT__)
 #include <clflushoptintrin.h>
-#endif
 
-#if !defined(__SCE__) || __has_feature(modules) || defined(__CLWB__)
 #include <clwbintrin.h>
-#endif
 
-#if !defined(__SCE__) || __has_feature(modules) || defined(__AVX__)
 #include <avxintrin.h>
-#endif
 
-#if !defined(__SCE__) || __has_feature(modules) || defined(__AVX2__)
 #include <avx2intrin.h>
-#endif
 
-#if !defined(__SCE__) || __has_feature(modules) || defined(__F16C__)
 #include <f16cintrin.h>
-#endif
 
-/* No feature check desired due to internal checks */
 #include <bmiintrin.h>
 
-#if !defined(__SCE__) || __has_feature(modules) || defined(__BMI2__)
 #include <bmi2intrin.h>
-#endif
 
-#if !defined(__SCE__) || __has_feature(modules) || defined(__LZCNT__)
 #include <lzcntintrin.h>
-#endif
 
-#if !defined(__SCE__) || __has_feature(modules) || defined(__POPCNT__)
 #include <popcntintrin.h>
-#endif
 
-#if !defined(__SCE__) || __has_feature(modules) || defined(__FMA__)
 #include <fmaintrin.h>
-#endif
 
-#if !defined(__SCE__) || __has_feature(modules) || defined(__AVX512F__)
 #include <avx512fintrin.h>
-#endif
 
-#if !defined(__SCE__) || __has_feature(modules) || defined(__AVX512VL__)
 #include <avx512vlintrin.h>
-#endif
 
-#if !defined(__SCE__) || __has_feature(modules) || defined(__AVX512BW__)
 #include <avx512bwintrin.h>
-#endif
 
-#if !defined(__SCE__) || __has_feature(modules) || defined(__AVX512BITALG__)
 #include <avx512bitalgintrin.h>
-#endif
 
-#if !defined(__SCE__) || __has_feature(modules) || defined(__AVX512CD__)
 #include <avx512cdintrin.h>
-#endif
 
-#if !defined(__SCE__) || __has_feature(modules) || defined(__AVX512VPOPCNTDQ__)
 #include <avx512vpopcntdqintrin.h>
-#endif
 
-#if !defined(__SCE__) || __has_feature(modules) ||                             \
-    (defined(__AVX512VL__) && defined(__AVX512VPOPCNTDQ__))
 #include <avx512vpopcntdqvlintrin.h>
-#endif
 
-#if !defined(__SCE__) || __has_feature(modules) || defined(__AVX512VNNI__)
 #include <avx512vnniintrin.h>
-#endif
 
-#if !defined(__SCE__) || __has_feature(modules) ||                             \
-    (defined(__AVX512VL__) && defined(__AVX512VNNI__))
 #include <avx512vlvnniintrin.h>
-#endif
 
-#if !defined(__SCE__) || __has_feature(modules) || defined(__AVXVNNI__)
 #include <avxvnniintrin.h>
-#endif
 
-#if !defined(__SCE__) || __has_feature(modules) || defined(__AVX512DQ__)
 #include <avx512dqintrin.h>
-#endif
 
-#if !defined(__SCE__) || __has_feature(modules) ||                             \
-    (defined(__AVX512VL__) && defined(__AVX512BITALG__))
 #include <avx512vlbitalgintrin.h>
-#endif
 
-#if !defined(__SCE__) || __has_feature(modules) ||                             \
-    (defined(__AVX512VL__) && defined(__AVX512BW__))
 #include <avx512vlbwintrin.h>
-#endif
 
-#if !defined(__SCE__) || __has_feature(modules) ||                             \
-    (defined(__AVX512VL__) && defined(__AVX512CD__))
 #include <avx512vlcdintrin.h>
-#endif
 
-#if !defined(__SCE__) || __has_feature(modules) ||                             \
-    (defined(__AVX512VL__) && defined(__AVX512DQ__))
 #include <avx512vldqintrin.h>
-#endif
 
-#if !defined(__SCE__) || __has_feature(modules) || defined(__AVX512IFMA__)
 #include <avx512ifmaintrin.h>
-#endif
 
-#if !defined(__SCE__) || __has_feature(modules) ||                             \
-    (defined(__AVX512IFMA__) && defined(__AVX512VL__))
 #include <avx512ifmavlintrin.h>
-#endif
 
-#if !defined(__SCE__) || __has_feature(modules) || defined(__AVXIFMA__)
 #include <avxifmaintrin.h>
-#endif
 
-#if !defined(__SCE__) || __has_feature(modules) || defined(__AVX512VBMI__)
 #include <avx512vbmiintrin.h>
-#endif
 
-#if !defined(__SCE__) || __has_feature(modules) ||                             \
-    (defined(__AVX512VBMI__) && defined(__AVX512VL__))
 #include <avx512vbmivlintrin.h>
-#endif
 
-#if !defined(__SCE__) || __has_feature(modules) || defined(__AVX512VBMI2__)
 #include <avx512vbmi2intrin.h>
-#endif
 
-#if !defined(__SCE__) || __has_feature(modules) ||                             \
-    (defined(__AVX512VBMI2__) && defined(__AVX512VL__))
 #include <avx512vlvbmi2intrin.h>
-#endif
 
-#if !defined(__SCE__) || __has_feature(modules) || defined(__AVX512FP16__)
 #include <avx512fp16intrin.h>
-#endif
 
-#if !defined(__SCE__) || __has_feature(modules) ||                             \
-    (defined(__AVX512VL__) && defined(__AVX512FP16__))
 #include <avx512vlfp16intrin.h>
-#endif
 
-#if !defined(__SCE__) || __has_feature(modules) || defined(__AVX512BF16__)
 #include <avx512bf16intrin.h>
-#endif
 
-#if !defined(__SCE__) || __has_feature(modules) ||                             \
-    (defined(__AVX512VL__) && defined(__AVX512BF16__))
 #include <avx512vlbf16intrin.h>
-#endif
 
-#if !defined(__SCE__) || __has_feature(modules) || defined(__PKU__)
 #include <pkuintrin.h>
-#endif
 
-#if !defined(__SCE__) || __has_feature(modules) || defined(__VPCLMULQDQ__)
 #include <vpclmulqdqintrin.h>
-#endif
 
-#if !defined(__SCE__) || __has_feature(modules) || defined(__VAES__)
 #include <vaesintrin.h>
-#endif
 
-#if !defined(__SCE__) || __has_feature(modules) || defined(__GFNI__)
 #include <gfniintrin.h>
-#endif
 
-#if !defined(__SCE__) || __has_feature(modules) || defined(__AVXVNNIINT8__)
 #include <avxvnniint8intrin.h>
-#endif
 
-#if !defined(__SCE__) || __has_feature(modules) || defined(__AVXNECONVERT__)
 #include <avxneconvertintrin.h>
-#endif
 
-#if !defined(__SCE__) || __has_feature(modules) || defined(__SHA512__)
 #include <sha512intrin.h>
-#endif
 
-#if !defined(__SCE__) || __has_feature(modules) || defined(__SM3__)
 #include <sm3intrin.h>
-#endif
 
-#if !defined(__SCE__) || __has_feature(modules) || defined(__SM4__)
 #include <sm4intrin.h>
-#endif
 
-#if !defined(__SCE__) || __has_feature(modules) || defined(__AVXVNNIINT16__)
 #include <avxvnniint16intrin.h>
-#endif
 
-#if !defined(__SCE__) || __has_feature(modules) || defined(__RDPID__)
 /// Reads the value of the IA32_TSC_AUX MSR (0xc0000103).
 ///
 /// \headerfile <immintrin.h>
@@ -252,9 +133,7 @@ static __inline__ unsigned int __attribute__((__always_inline__, __nodebug__, __
 _rdpid_u32(void) {
   return __builtin_ia32_rdpid();
 }
-#endif // __RDPID__
 
-#if !defined(__SCE__) || __has_feature(modules) || defined(__RDRND__)
 /// Returns a 16-bit hardware-generated random value.
 ///
 /// \headerfile <immintrin.h>
@@ -314,9 +193,7 @@ _rdrand64_step(unsigned long long *__p)
   }
 #endif
 }
-#endif /* __RDRND__ */
 
-#if !defined(__SCE__) || __has_feature(modules) || defined(__FSGSBASE__)
 #ifdef __x86_64__
 /// Reads the FS base register.
 ///
@@ -427,9 +304,6 @@ _writegsbase_u64(unsigned long long __V)
 }
 
 #endif
-#endif /* __FSGSBASE__ */
-
-#if !defined(__SCE__) || __has_feature(modules) || defined(__MOVBE__)
 
 /* The structs used below are to force the load/store to be unaligned. This
  * is accomplished with the __packed__ attribute. The __may_alias__ prevents
@@ -543,172 +417,86 @@ _storebe_i64(void * __P, long long __D) {
   ((struct __storeu_i64*)__P)->__v = __builtin_bswap64((unsigned long long)__D);
 }
 #endif
-#endif /* __MOVBE */
 
-#if !defined(__SCE__) || __has_feature(modules) || defined(__RTM__)
 #include <rtmintrin.h>
 #include <xtestintrin.h>
-#endif
 
-#if !defined(__SCE__) || __has_feature(modules) || defined(__SHA__)
 #include <shaintrin.h>
-#endif
 
-#if !defined(__SCE__) || __has_feature(modules) || defined(__FXSR__)
 #include <fxsrintrin.h>
-#endif
 
 /* No feature check desired due to internal MSC_VER checks */
 #include <xsaveintrin.h>
 
-#if !defined(__SCE__) || __has_feature(modules) || defined(__XSAVEOPT__)
 #include <xsaveoptintrin.h>
-#endif
 
-#if !defined(__SCE__) || __has_feature(modules) || defined(__XSAVEC__)
 #include <xsavecintrin.h>
-#endif
 
-#if !defined(__SCE__) || __has_feature(modules) || defined(__XSAVES__)
 #include <xsavesintrin.h>
-#endif
 
-#if !defined(__SCE__) || __has_feature(modules) || defined(__SHSTK__)
 #include <cetintrin.h>
-#endif
 
 /* Intrinsics inside adcintrin.h are available at all times. */
 #include <adcintrin.h>
 
-#if !defined(__SCE__) || __has_feature(modules) || defined(__ADX__)
 #include <adxintrin.h>
-#endif
 
-#if !defined(__SCE__) || __has_feature(modules) || defined(__RDSEED__)
 #include <rdseedintrin.h>
-#endif
 
-#if !defined(__SCE__) || __has_feature(modules) || defined(__WBNOINVD__)
 #include <wbnoinvdintrin.h>
-#endif
 
-#if !defined(__SCE__) || __has_feature(modules) || defined(__CLDEMOTE__)
 #include <cldemoteintrin.h>
-#endif
 
-#if !defined(__SCE__) || __has_feature(modules) || defined(__WAITPKG__)
 #include <waitpkgintrin.h>
-#endif
 
-#if !defined(__SCE__) || __has_feature(modules) || defined(__MOVDIRI__) ||     \
-    defined(__MOVDIR64B__)
 #include <movdirintrin.h>
-#endif
 
-#if !defined(__SCE__) || __has_feature(modules) || defined(__MOVRS__)
 #include <movrsintrin.h>
-#endif
 
-#if !defined(__SCE__) || __has_feature(modules) ||                             \
-    (defined(__AVX10_2__) && defined(__MOVRS__))
 #include <movrs_avx10_2intrin.h>
-#endif
 
-#if !defined(__SCE__) || __has_feature(modules) ||                             \
-    (defined(__AVX10_2_512__) && defined(__MOVRS__))
 #include <movrs_avx10_2_512intrin.h>
-#endif
 
-#if !defined(__SCE__) || __has_feature(modules) || defined(__PCONFIG__)
 #include <pconfigintrin.h>
-#endif
 
-#if !defined(__SCE__) || __has_feature(modules) || defined(__SGX__)
 #include <sgxintrin.h>
-#endif
 
-#if !defined(__SCE__) || __has_feature(modules) || defined(__PTWRITE__)
 #include <ptwriteintrin.h>
-#endif
 
-#if !defined(__SCE__) || __has_feature(modules) || defined(__INVPCID__)
 #include <invpcidintrin.h>
-#endif
 
-#if !defined(__SCE__) || __has_feature(modules) || defined(__KL__) ||          \
-    defined(__WIDEKL__)
 #include <keylockerintrin.h>
-#endif
 
-#if !defined(__SCE__) || __has_feature(modules) || defined(__AMX_TILE__) ||    \
-    defined(__AMX_INT8__) || defined(__AMX_BF16__)
 #include <amxintrin.h>
-#endif
 
-#if !defined(__SCE__) || __has_feature(modules) || defined(__AMX_FP16__)
 #include <amxfp16intrin.h>
-#endif
 
-#if !defined(__SCE__) || __has_feature(modules) || defined(__AMX_COMPLEX__)
 #include <amxcomplexintrin.h>
-#endif
 
-#if !defined(__SCE__) || __has_feature(modules) || defined(__AMX_FP8__)
 #include <amxfp8intrin.h>
-#endif
 
-#if !defined(__SCE__) || __has_feature(modules) || defined(__AMX_TRANSPOSE__)
 #include <amxtransposeintrin.h>
-#endif
 
-#if !defined(__SCE__) || __has_feature(modules) || defined(__AMX_MOVRS__)
 #include <amxmovrsintrin.h>
-#endif
 
-#if !defined(__SCE__) || __has_feature(modules) ||                             \
-    (defined(__AMX_MOVRS__) && defined(__AMX_TRANSPOSE__))
 #include <amxmovrstransposeintrin.h>
-#endif
 
-#if !defined(__SCE__) || __has_feature(modules) || defined(__AMX_AVX512__)
 #include <amxavx512intrin.h>
-#endif
 
-#if !defined(__SCE__) || __has_feature(modules) || defined(__AMX_TF32__)
 #include <amxtf32intrin.h>
-#endif
 
-#if !defined(__SCE__) || __has_feature(modules) ||                             \
-    (defined(__AMX_TF32__) && defined(__AMX_TRANSPOSE__))
 #include <amxtf32transposeintrin.h>
-#endif
 
-#if !defined(__SCE__) || __has_feature(modules) ||                             \
-    (defined(__AMX_BF16__) && defined(__AMX_TRANSPOSE__))
 #include <amxbf16transposeintrin.h>
-#endif
 
-#if !defined(__SCE__) || __has_feature(modules) ||                             \
-    (defined(__AMX_FP16__) && defined(__AMX_TRANSPOSE__))
 #include <amxfp16transposeintrin.h>
-#endif
 
-#if !defined(__SCE__) || __has_feature(modules) ||                             \
-    (defined(__AMX_COMPLEX__) && defined(__AMX_TRANSPOSE__))
 #include <amxcomplextransposeintrin.h>
-#endif
 
-#if !defined(__SCE__) || __has_feature(modules) ||                             \
-    defined(__AVX512VP2INTERSECT__)
 #include <avx512vp2intersectintrin.h>
-#endif
 
-#if !defined(__SCE__) || __has_feature(modules) ||                             \
-    (defined(__AVX512VL__) && defined(__AVX512VP2INTERSECT__))
 #include <avx512vlvp2intersectintrin.h>
-#endif
 
-#if !defined(__SCE__) || __has_feature(modules) || defined(__AVX10_2__)
 #include <avx10_2bf16intrin.h>
 #include <avx10_2convertintrin.h>
 #include <avx10_2copyintrin.h>
@@ -716,33 +504,21 @@ _storebe_i64(void * __P, long long __D) {
 #include <avx10_2niintrin.h>
 #include <avx10_2satcvtdsintrin.h>
 #include <avx10_2satcvtintrin.h>
-#endif
 
-#if !defined(__SCE__) || __has_feature(modules) || defined(__AVX10_2_512__)
 #include <avx10_2_512bf16intrin.h>
 #include <avx10_2_512convertintrin.h>
 #include <avx10_2_512minmaxintrin.h>
 #include <avx10_2_512niintrin.h>
 #include <avx10_2_512satcvtdsintrin.h>
 #include <avx10_2_512satcvtintrin.h>
-#endif
 
-#if !defined(__SCE__) || __has_feature(modules) ||                             \
-    (defined(__AVX10_2_512__) && defined(__SM4__))
 #include <sm4evexintrin.h>
-#endif
 
-#if !defined(__SCE__) || __has_feature(modules) || defined(__ENQCMD__)
 #include <enqcmdintrin.h>
-#endif
 
-#if !defined(__SCE__) || __has_feature(modules) || defined(__SERIALIZE__)
 #include <serializeintrin.h>
-#endif
 
-#if !defined(__SCE__) || __has_feature(modules) || defined(__TSXLDTRK__)
 #include <tsxldtrkintrin.h>
-#endif
 
 #if defined(_MSC_VER) && __has_extension(gnu_asm)
 /* Define the default attributes for these intrinsics */

lib/include/intrin.h

@@ -162,8 +162,6 @@ void _Store_HLERelease(long volatile *, long);
 void _Store64_HLERelease(__int64 volatile *, __int64);
 void _StorePointer_HLERelease(void *volatile *, void *);
 void _WriteBarrier(void);
-unsigned __int32 xbegin(void);
-void _xend(void);
 
 /* These additional intrinsics are turned on in x64/amd64/x86_64 mode. */
 #if defined(__x86_64__) && !defined(__arm64ec__)
@@ -372,10 +370,29 @@ static __inline__ void __DEFAULT_FN_ATTRS __nop(void) {
 \*----------------------------------------------------------------------------*/
 #if defined(__aarch64__) || defined(__arm64ec__)
 unsigned __int64 __getReg(int);
-long _InterlockedAdd(long volatile *Addend, long Value);
-__int64 _InterlockedAdd64(__int64 volatile *Addend, __int64 Value);
+unsigned char _interlockedbittestandreset_acq(long volatile *, long);
+unsigned char _interlockedbittestandreset_nf(long volatile *, long);
+unsigned char _interlockedbittestandreset_rel(long volatile *, long);
+unsigned char _interlockedbittestandreset64_acq(__int64 volatile *, __int64);
+unsigned char _interlockedbittestandreset64_nf(__int64 volatile *, __int64);
+unsigned char _interlockedbittestandreset64_rel(__int64 volatile *, __int64);
+unsigned char _interlockedbittestandset_acq(long volatile *, long);
+unsigned char _interlockedbittestandset_nf(long volatile *, long);
+unsigned char _interlockedbittestandset_rel(long volatile *, long);
+unsigned char _interlockedbittestandset64_acq(__int64 volatile *, __int64);
+unsigned char _interlockedbittestandset64_nf(__int64 volatile *, __int64);
+unsigned char _interlockedbittestandset64_rel(__int64 volatile *, __int64);
+long _InterlockedAdd(long volatile *, long);
+long _InterlockedAdd_acq(long volatile *, long);
+long _InterlockedAdd_nf(long volatile *, long);
+long _InterlockedAdd_rel(long volatile *, long);
+__int64 _InterlockedAdd64(__int64 volatile *, __int64);
+__int64 _InterlockedAdd64_acq(__int64 volatile *, __int64);
+__int64 _InterlockedAdd64_nf(__int64 volatile *, __int64);
+__int64 _InterlockedAdd64_rel(__int64 volatile *, __int64);
 __int64 _ReadStatusReg(int);
 void _WriteStatusReg(int, __int64);
+unsigned int __sys(int, __int64);
 
 unsigned short __cdecl _byteswap_ushort(unsigned short val);
 unsigned long __cdecl _byteswap_ulong (unsigned long val);

lib/include/keylockerintrin.h

@@ -28,8 +28,6 @@
 #ifndef _KEYLOCKERINTRIN_H
 #define _KEYLOCKERINTRIN_H
 
-#if !defined(__SCE__) || __has_feature(modules) || defined(__KL__)
-
 /* Define the default attributes for the functions in this file. */
 #define __DEFAULT_FN_ATTRS \
   __attribute__((__always_inline__, __nodebug__, __target__("kl"),\
@@ -326,10 +324,6 @@ _mm_aesdec256kl_u8(__m128i* __odata, __m128i __idata, const void *__h) {
 
 #undef __DEFAULT_FN_ATTRS
 
-#endif /* !defined(__SCE__ || __has_feature(modules) || defined(__KL__) */
-
-#if !defined(__SCE__) || __has_feature(modules) || defined(__WIDEKL__)
-
 /* Define the default attributes for the functions in this file. */
 #define __DEFAULT_FN_ATTRS \
   __attribute__((__always_inline__, __nodebug__, __target__("kl,widekl"),\
@@ -521,7 +515,4 @@ _mm_aesdecwide256kl_u8(__m128i __odata[8], const __m128i __idata[8], const void*
 
 #undef __DEFAULT_FN_ATTRS
 
-#endif /* !defined(__SCE__) || __has_feature(modules) || defined(__WIDEKL__)   \
-        */
-
 #endif /* _KEYLOCKERINTRIN_H */

lib/include/llvm_libc_wrappers/stdlib.h

@@ -25,7 +25,7 @@
 
 // The LLVM C library uses these named types so we forward declare them.
 typedef void (*__atexithandler_t)(void);
-typedef int (*__bsearchcompare_t)(const void *, const void *);
+typedef int (*__search_compare_t)(const void *, const void *);
 typedef int (*__qsortcompare_t)(const void *, const void *);
 typedef int (*__qsortrcompare_t)(const void *, const void *, void *);
 

lib/include/lzcntintrin.h

@@ -14,13 +14,15 @@
 #ifndef __LZCNTINTRIN_H
 #define __LZCNTINTRIN_H
 
-/* Define the default attributes for the functions in this file. */
+/* Define the default attributes for the functions in this file.
+   Allow using the lzcnt intrinsics even for non-LZCNT targets. Since the LZCNT
+   intrinsics are mapped to llvm.ctlz.*, false, which can be lowered to BSR on
+   non-LZCNT targets with zero-value input handled correctly. */
 #if defined(__cplusplus) && (__cplusplus >= 201103L)
 #define __DEFAULT_FN_ATTRS                                                     \
-  __attribute__((__always_inline__, __nodebug__, __target__("lzcnt"))) constexpr
+  __attribute__((__always_inline__, __nodebug__)) constexpr
 #else
-#define __DEFAULT_FN_ATTRS                                                     \
-  __attribute__((__always_inline__, __nodebug__, __target__("lzcnt")))
+#define __DEFAULT_FN_ATTRS __attribute__((__always_inline__, __nodebug__))
 #endif
 
 #ifndef _MSC_VER

lib/include/module.modulemap

@@ -35,6 +35,14 @@ module _Builtin_intrinsics [system] [extern_c] {
     }
   }
 
+  explicit module arm64 {
+    requires arm64
+    requires windows
+
+    header "arm64intr.h"
+    export *
+  }
+
   explicit module intel {
     requires x86
     export *
@@ -231,6 +239,11 @@ module _Builtin_stdbool [system] {
   export *
 }
 
+module _Builtin_stdcountof [system] {
+  header "stdcountof.h"
+  export *
+}
+
 module _Builtin_stddef [system] {
   textual header "stddef.h"
 

lib/include/movrsintrin.h

@@ -56,4 +56,4 @@ _m_prefetchrs(volatile const void *__P) {
 }
 
 #undef __DEFAULT_FN_ATTRS
-#endif // __MOVRSINTRIN_H
+#endif // __MOVRSINTRIN_H

lib/include/prfchwintrin.h

@@ -14,6 +14,10 @@
 #ifndef __PRFCHWINTRIN_H
 #define __PRFCHWINTRIN_H
 
+#if defined(__cplusplus)
+extern "C" {
+#endif
+
 /// Loads a memory sequence containing the specified memory address into
 ///    all data cache levels.
 ///
@@ -26,11 +30,7 @@
 ///
 /// \param __P
 ///    A pointer specifying the memory address to be prefetched.
-static __inline__ void __attribute__((__always_inline__, __nodebug__))
-_m_prefetch(void *__P)
-{
-  __builtin_prefetch (__P, 0, 3 /* _MM_HINT_T0 */);
-}
+void _m_prefetch(void *__P);
 
 /// Loads a memory sequence containing the specified memory address into
 ///    the L1 data cache and sets the cache-coherency state to modified.
@@ -48,13 +48,10 @@ _m_prefetch(void *__P)
 ///
 /// \param __P
 ///    A pointer specifying the memory address to be prefetched.
-static __inline__ void __attribute__((__always_inline__, __nodebug__))
-_m_prefetchw(volatile const void *__P)
-{
-#pragma clang diagnostic push
-#pragma clang diagnostic ignored "-Wcast-qual"
-  __builtin_prefetch ((const void*)__P, 1, 3 /* _MM_HINT_T0 */);
-#pragma clang diagnostic pop
-}
+void _m_prefetchw(volatile const void *__P);
+
+#if defined(__cplusplus)
+} // extern "C"
+#endif
 
 #endif /* __PRFCHWINTRIN_H */

lib/include/ptrauth.h

@@ -42,6 +42,19 @@ typedef enum {
      The extra data is always 0. */
   ptrauth_key_cxx_vtable_pointer = ptrauth_key_process_independent_data,
 
+  /* The key used to sign metadata pointers to Objective-C method-lists. */
+  ptrauth_key_method_list_pointer = ptrauth_key_asda,
+
+  /* The key used to sign Objective-C isa and super pointers. */
+  ptrauth_key_objc_isa_pointer = ptrauth_key_process_independent_data,
+  ptrauth_key_objc_super_pointer = ptrauth_key_process_independent_data,
+
+  /* The key used to sign selector pointers */
+  ptrauth_key_objc_sel_pointer = ptrauth_key_process_dependent_data,
+
+  /* The key used to sign Objective-C class_ro_t pointers. */
+  ptrauth_key_objc_class_ro_pointer = ptrauth_key_process_independent_data,
+
   /* The key used to sign pointers in ELF .init_array/.fini_array. */
   ptrauth_key_init_fini_pointer = ptrauth_key_process_independent_code,
 
@@ -259,6 +272,46 @@ typedef __UINTPTR_TYPE__ ptrauth_generic_signature_t;
 /* The value is ptrauth_string_discriminator("init_fini") */
 #define __ptrauth_init_fini_discriminator 0xd9d4
 
+/* Objective-C pointer auth ABI qualifiers */
+#define __ptrauth_objc_method_list_imp                                         \
+  __ptrauth(ptrauth_key_function_pointer, 1, 0)
+
+#if __has_feature(ptrauth_objc_method_list_pointer)
+#define __ptrauth_objc_method_list_pointer                                     \
+  __ptrauth(ptrauth_key_method_list_pointer, 1, 0xC310)
+#else
+#define __ptrauth_objc_method_list_pointer
+#endif
+
+#define __ptrauth_isa_discriminator 0x6AE1
+#define __ptrauth_super_discriminator 0xB5AB
+#define __ptrauth_objc_isa_pointer                                             \
+  __ptrauth(ptrauth_key_objc_isa_pointer, 1, __ptrauth_isa_discriminator)
+#if __has_feature(ptrauth_restricted_intptr_qualifier)
+#define __ptrauth_objc_isa_uintptr                                             \
+  __ptrauth_restricted_intptr(ptrauth_key_objc_isa_pointer, 1,                 \
+                              __ptrauth_isa_discriminator)
+#else
+#define __ptrauth_objc_isa_uintptr                                             \
+  __ptrauth(ptrauth_key_objc_isa_pointer, 1, __ptrauth_isa_discriminator)
+#endif
+
+#define __ptrauth_objc_super_pointer                                           \
+  __ptrauth(ptrauth_key_objc_super_pointer, 1, __ptrauth_super_discriminator)
+
+#define __ptrauth_objc_sel_discriminator 0x57c2
+#if __has_feature(ptrauth_objc_interface_sel)
+#define __ptrauth_objc_sel                                                     \
+  __ptrauth(ptrauth_key_objc_sel_pointer, 1, __ptrauth_objc_sel_discriminator)
+#else
+#define __ptrauth_objc_sel
+#endif
+
+#define __ptrauth_objc_class_ro_discriminator 0x61f8
+#define __ptrauth_objc_class_ro                                                \
+  __ptrauth(ptrauth_key_objc_class_ro_pointer, 1,                              \
+            __ptrauth_objc_class_ro_discriminator)
+
 #else
 
 #define ptrauth_strip(__value, __key)                                          \
@@ -331,6 +384,10 @@ typedef __UINTPTR_TYPE__ ptrauth_generic_signature_t;
 #define ptrauth_cxx_vtable_pointer(key, address_discrimination,                \
                                    extra_discrimination...)
 
+#define __ptrauth_objc_isa_pointer
+#define __ptrauth_objc_isa_uintptr
+#define __ptrauth_objc_super_pointer
+
 #endif /* __has_feature(ptrauth_intrinsics) */
 
 #endif /* __PTRAUTH_H */

lib/include/riscv_corev_alu.h

@@ -24,13 +24,13 @@ static __inline__ long __DEFAULT_FN_ATTRS __riscv_cv_abs(long a) {
   return __builtin_abs(a);
 }
 
-static __inline__ long __DEFAULT_FN_ATTRS __riscv_cv_alu_slet(long a, long b) {
-  return __builtin_riscv_cv_alu_slet(a, b);
+static __inline__ long __DEFAULT_FN_ATTRS __riscv_cv_alu_sle(long a, long b) {
+  return __builtin_riscv_cv_alu_sle(a, b);
 }
 
 static __inline__ long __DEFAULT_FN_ATTRS
-__riscv_cv_alu_sletu(unsigned long a, unsigned long b) {
-  return __builtin_riscv_cv_alu_sletu(a, b);
+__riscv_cv_alu_sleu(unsigned long a, unsigned long b) {
+  return __builtin_riscv_cv_alu_sleu(a, b);
 }
 
 static __inline__ long __DEFAULT_FN_ATTRS __riscv_cv_alu_min(long a, long b) {

lib/include/riscv_vector.h

@@ -49,7 +49,6 @@ enum __RISCV_FRM {
 #define __riscv_vsetvl_e32m4(avl) __builtin_rvv_vsetvli((size_t)(avl), 2, 2)
 #define __riscv_vsetvl_e32m8(avl) __builtin_rvv_vsetvli((size_t)(avl), 2, 3)
 
-#if __riscv_v_elen >= 64
 #define __riscv_vsetvl_e8mf8(avl) __builtin_rvv_vsetvli((size_t)(avl), 0, 5)
 #define __riscv_vsetvl_e16mf4(avl) __builtin_rvv_vsetvli((size_t)(avl), 1, 6)
 #define __riscv_vsetvl_e32mf2(avl) __builtin_rvv_vsetvli((size_t)(avl), 2, 7)
@@ -58,7 +57,6 @@ enum __RISCV_FRM {
 #define __riscv_vsetvl_e64m2(avl) __builtin_rvv_vsetvli((size_t)(avl), 3, 1)
 #define __riscv_vsetvl_e64m4(avl) __builtin_rvv_vsetvli((size_t)(avl), 3, 2)
 #define __riscv_vsetvl_e64m8(avl) __builtin_rvv_vsetvli((size_t)(avl), 3, 3)
-#endif
 
 #define __riscv_vsetvlmax_e8mf4() __builtin_rvv_vsetvlimax(0, 6)
 #define __riscv_vsetvlmax_e8mf2() __builtin_rvv_vsetvlimax(0, 7)
@@ -78,7 +76,6 @@ enum __RISCV_FRM {
 #define __riscv_vsetvlmax_e32m4() __builtin_rvv_vsetvlimax(2, 2)
 #define __riscv_vsetvlmax_e32m8() __builtin_rvv_vsetvlimax(2, 3)
 
-#if __riscv_v_elen >= 64
 #define __riscv_vsetvlmax_e8mf8() __builtin_rvv_vsetvlimax(0, 5)
 #define __riscv_vsetvlmax_e16mf4() __builtin_rvv_vsetvlimax(1, 6)
 #define __riscv_vsetvlmax_e32mf2() __builtin_rvv_vsetvlimax(2, 7)
@@ -87,7 +84,6 @@ enum __RISCV_FRM {
 #define __riscv_vsetvlmax_e64m2() __builtin_rvv_vsetvlimax(3, 1)
 #define __riscv_vsetvlmax_e64m4() __builtin_rvv_vsetvlimax(3, 2)
 #define __riscv_vsetvlmax_e64m8() __builtin_rvv_vsetvlimax(3, 3)
-#endif
 
 
 enum __RISCV_VXRM {

lib/include/shaintrin.h

@@ -47,8 +47,9 @@
 ///    An immediate value where bits [1:0] select among four possible
 ///    combining functions and rounding constants (not specified here).
 /// \returns A 128-bit vector of [4 x i32] containing the updated SHA-1 state.
-#define _mm_sha1rnds4_epu32(V1, V2, M) \
-  __builtin_ia32_sha1rnds4((__v4si)(__m128i)(V1), (__v4si)(__m128i)(V2), (M))
+#define _mm_sha1rnds4_epu32(V1, V2, M)                                         \
+  ((__m128i)__builtin_ia32_sha1rnds4((__v4si)(__m128i)(V1),                    \
+                                     (__v4si)(__m128i)(V2), (M)))
 
 /// Calculates the SHA-1 state variable E from the SHA-1 state variables in
 ///    the 128-bit vector of [4 x i32] in \a __X, adds that to the next set of

lib/include/stdcountof.h

@@ -0,0 +1,15 @@
+/*===---- stdcountof.h - Standard header for countof -----------------------===
+ *
+ * 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 __STDCOUNTOF_H
+#define __STDCOUNTOF_H
+
+#define countof _Countof
+
+#endif /* __STDCOUNTOF_H */

lib/include/stdint.h

@@ -317,166 +317,55 @@ typedef __UINTMAX_TYPE__ uintmax_t;
  * integer width that the target implements, so corresponding macros are
  * defined below, too.
  *
- * These macros are defined using the same successive-shrinking approach as
- * the type definitions above. It is likewise important that macros are defined
- * in order of decending width.
- *
  * Note that C++ should not check __STDC_CONSTANT_MACROS here, contrary to the
  * claims of the C standard (see C++ 18.3.1p2, [cstdint.syn]).
  */
 
-#define __int_c_join(a, b) a ## b
-#define __int_c(v, suffix) __int_c_join(v, suffix)
-#define __uint_c(v, suffix) __int_c_join(v##U, suffix)
-
-
-#ifdef __INT64_TYPE__
-# undef __int64_c_suffix
-# undef __int32_c_suffix
-# undef __int16_c_suffix
-# undef  __int8_c_suffix
-# ifdef __INT64_C_SUFFIX__
-#  define __int64_c_suffix __INT64_C_SUFFIX__
-#  define __int32_c_suffix __INT64_C_SUFFIX__
-#  define __int16_c_suffix __INT64_C_SUFFIX__
-#  define  __int8_c_suffix __INT64_C_SUFFIX__
-# endif /* __INT64_C_SUFFIX__ */
-#endif /* __INT64_TYPE__ */
-
 #ifdef __int_least64_t
-# ifdef __int64_c_suffix
-#  define INT64_C(v) __int_c(v, __int64_c_suffix)
-#  define UINT64_C(v) __uint_c(v, __int64_c_suffix)
-# else
-#  define INT64_C(v) v
-#  define UINT64_C(v) v ## U
-# endif /* __int64_c_suffix */
+#define INT64_C(v) __INT64_C(v)
+#define UINT64_C(v) __UINT64_C(v)
 #endif /* __int_least64_t */
 
 
 #ifdef __INT56_TYPE__
-# undef __int32_c_suffix
-# undef __int16_c_suffix
-# undef  __int8_c_suffix
-# ifdef __INT56_C_SUFFIX__
-#  define INT56_C(v) __int_c(v, __INT56_C_SUFFIX__)
-#  define UINT56_C(v) __uint_c(v, __INT56_C_SUFFIX__)
-#  define __int32_c_suffix __INT56_C_SUFFIX__
-#  define __int16_c_suffix __INT56_C_SUFFIX__
-#  define __int8_c_suffix  __INT56_C_SUFFIX__
-# else
-#  define INT56_C(v) v
-#  define UINT56_C(v) v ## U
-# endif /* __INT56_C_SUFFIX__ */
+#define INT56_C(v) __INT56_C(v)
+#define UINT56_C(v) __UINT56_C(v)
 #endif /* __INT56_TYPE__ */
 
 
 #ifdef __INT48_TYPE__
-# undef __int32_c_suffix
-# undef __int16_c_suffix
-# undef  __int8_c_suffix
-# ifdef __INT48_C_SUFFIX__
-#  define INT48_C(v) __int_c(v, __INT48_C_SUFFIX__)
-#  define UINT48_C(v) __uint_c(v, __INT48_C_SUFFIX__)
-#  define __int32_c_suffix __INT48_C_SUFFIX__
-#  define __int16_c_suffix __INT48_C_SUFFIX__
-#  define __int8_c_suffix  __INT48_C_SUFFIX__
-# else
-#  define INT48_C(v) v
-#  define UINT48_C(v) v ## U
-# endif /* __INT48_C_SUFFIX__ */
+#define INT48_C(v) __INT48_C(v)
+#define UINT48_C(v) __UINT48_C(v)
 #endif /* __INT48_TYPE__ */
 
 
 #ifdef __INT40_TYPE__
-# undef __int32_c_suffix
-# undef __int16_c_suffix
-# undef  __int8_c_suffix
-# ifdef __INT40_C_SUFFIX__
-#  define INT40_C(v) __int_c(v, __INT40_C_SUFFIX__)
-#  define UINT40_C(v) __uint_c(v, __INT40_C_SUFFIX__)
-#  define __int32_c_suffix __INT40_C_SUFFIX__
-#  define __int16_c_suffix __INT40_C_SUFFIX__
-#  define __int8_c_suffix  __INT40_C_SUFFIX__
-# else
-#  define INT40_C(v) v
-#  define UINT40_C(v) v ## U
-# endif /* __INT40_C_SUFFIX__ */
+#define INT40_C(v) __INT40_C(v)
+#define UINT40_C(v) __UINT40_C(v)
 #endif /* __INT40_TYPE__ */
 
 
-#ifdef __INT32_TYPE__
-# undef __int32_c_suffix
-# undef __int16_c_suffix
-# undef  __int8_c_suffix
-# ifdef __INT32_C_SUFFIX__
-#  define __int32_c_suffix __INT32_C_SUFFIX__
-#  define __int16_c_suffix __INT32_C_SUFFIX__
-#  define __int8_c_suffix  __INT32_C_SUFFIX__
-# endif /* __INT32_C_SUFFIX__ */
-#endif /* __INT32_TYPE__ */
-
 #ifdef __int_least32_t
-# ifdef __int32_c_suffix
-#  define INT32_C(v) __int_c(v, __int32_c_suffix)
-#  define UINT32_C(v) __uint_c(v, __int32_c_suffix)
-# else
-#  define INT32_C(v) v
-#  define UINT32_C(v) v ## U
-# endif /* __int32_c_suffix */
+#define INT32_C(v) __INT32_C(v)
+#define UINT32_C(v) __UINT32_C(v)
 #endif /* __int_least32_t */
 
 
 #ifdef __INT24_TYPE__
-# undef __int16_c_suffix
-# undef  __int8_c_suffix
-# ifdef __INT24_C_SUFFIX__
-#  define INT24_C(v) __int_c(v, __INT24_C_SUFFIX__)
-#  define UINT24_C(v) __uint_c(v, __INT24_C_SUFFIX__)
-#  define __int16_c_suffix __INT24_C_SUFFIX__
-#  define __int8_c_suffix  __INT24_C_SUFFIX__
-# else
-#  define INT24_C(v) v
-#  define UINT24_C(v) v ## U
-# endif /* __INT24_C_SUFFIX__ */
+#define INT24_C(v) __INT24_C(v)
+#define UINT24_C(v) __UINT24_C(v)
 #endif /* __INT24_TYPE__ */
 
 
-#ifdef __INT16_TYPE__
-# undef __int16_c_suffix
-# undef  __int8_c_suffix
-# ifdef __INT16_C_SUFFIX__
-#  define __int16_c_suffix __INT16_C_SUFFIX__
-#  define __int8_c_suffix  __INT16_C_SUFFIX__
-# endif /* __INT16_C_SUFFIX__ */
-#endif /* __INT16_TYPE__ */
-
 #ifdef __int_least16_t
-# ifdef __int16_c_suffix
-#  define INT16_C(v) __int_c(v, __int16_c_suffix)
-#  define UINT16_C(v) __uint_c(v, __int16_c_suffix)
-# else
-#  define INT16_C(v) v
-#  define UINT16_C(v) v ## U
-# endif /* __int16_c_suffix */
+#define INT16_C(v) __INT16_C(v)
+#define UINT16_C(v) __UINT16_C(v)
 #endif /* __int_least16_t */
 
 
-#ifdef __INT8_TYPE__
-# undef  __int8_c_suffix
-# ifdef __INT8_C_SUFFIX__
-#  define __int8_c_suffix __INT8_C_SUFFIX__
-# endif /* __INT8_C_SUFFIX__ */
-#endif /* __INT8_TYPE__ */
-
 #ifdef __int_least8_t
-# ifdef __int8_c_suffix
-#  define INT8_C(v) __int_c(v, __int8_c_suffix)
-#  define UINT8_C(v) __uint_c(v, __int8_c_suffix)
-# else
-#  define INT8_C(v) v
-#  define UINT8_C(v) v ## U
-# endif /* __int8_c_suffix */
+#define INT8_C(v) __INT8_C(v)
+#define UINT8_C(v) __UINT8_C(v)
 #endif /* __int_least8_t */
 
 
@@ -938,8 +827,8 @@ typedef __UINTMAX_TYPE__ uintmax_t;
 #endif
 
 /* 7.18.4.2 Macros for greatest-width integer constants. */
-#define  INTMAX_C(v) __int_c(v,  __INTMAX_C_SUFFIX__)
-#define UINTMAX_C(v) __int_c(v, __UINTMAX_C_SUFFIX__)
+#define  INTMAX_C(v) __INTMAX_C(v)
+#define UINTMAX_C(v) __UINTMAX_C(v)
 
 /* C23 7.22.3.x Width of other integer types. */
 #if defined(__STDC_VERSION__) && __STDC_VERSION__ >= 202311L

lib/include/vecintrin.h

@@ -7,6 +7,9 @@
  *===-----------------------------------------------------------------------===
  */
 
+#ifndef _VECINTRIN_H
+#define _VECINTRIN_H
+
 #if defined(__s390x__) && defined(__VEC__)
 
 #define __ATTRS_ai __attribute__((__always_inline__))
@@ -12861,3 +12864,5 @@ vec_search_string_until_zero_cc(__vector unsigned int __a,
 #error "Use -fzvector to enable vector extensions"
 
 #endif
+
+#endif /* _VECINTRIN_H */

lib/include/x86gprintrin.h

@@ -10,33 +10,19 @@
 #ifndef __X86GPRINTRIN_H
 #define __X86GPRINTRIN_H
 
-#if !defined(__SCE__) || __has_feature(modules) || defined(__HRESET__)
 #include <hresetintrin.h>
-#endif
 
-#if !defined(__SCE__) || __has_feature(modules) || defined(__UINTR__)
 #include <uintrintrin.h>
-#endif
 
-#if !defined(__SCE__) || __has_feature(modules) || defined(__USERMSR__)
 #include <usermsrintrin.h>
-#endif
 
-#if !defined(__SCE__) || __has_feature(modules) || defined(__CRC32__)
 #include <crc32intrin.h>
-#endif
 
-#if !defined(__SCE__) || __has_feature(modules) || defined(__PRFCHI__)
 #include <prfchiintrin.h>
-#endif
 
-#if !defined(__SCE__) || __has_feature(modules) || defined(__RAOINT__)
 #include <raointintrin.h>
-#endif
 
-#if !defined(__SCE__) || __has_feature(modules) || defined(__CMPCCXADD__)
 #include <cmpccxaddintrin.h>
-#endif
 
 #if defined(__i386__)
 #define __SAVE_GPRBX "mov {%%ebx, %%eax |eax, ebx};"

lib/include/x86intrin.h

@@ -14,40 +14,22 @@
 
 #include <immintrin.h>
 
-#if !defined(__SCE__) || __has_feature(modules) || defined(__PRFCHW__)
 #include <prfchwintrin.h>
-#endif
 
-#if !defined(__SCE__) || __has_feature(modules) || defined(__SSE4A__)
 #include <ammintrin.h>
-#endif
 
-#if !defined(__SCE__) || __has_feature(modules) || defined(__FMA4__)
 #include <fma4intrin.h>
-#endif
 
-#if !defined(__SCE__) || __has_feature(modules) || defined(__XOP__)
 #include <xopintrin.h>
-#endif
 
-#if !defined(__SCE__) || __has_feature(modules) || defined(__TBM__)
 #include <tbmintrin.h>
-#endif
 
-#if !defined(__SCE__) || __has_feature(modules) || defined(__LWP__)
 #include <lwpintrin.h>
-#endif
 
-#if !defined(__SCE__) || __has_feature(modules) || defined(__MWAITX__)
 #include <mwaitxintrin.h>
-#endif
 
-#if !defined(__SCE__) || __has_feature(modules) || defined(__CLZERO__)
 #include <clzerointrin.h>
-#endif
 
-#if !defined(__SCE__) || __has_feature(modules) || defined(__RDPRU__)
 #include <rdpruintrin.h>
-#endif
 
 #endif /* __X86INTRIN_H */

lib/include/xmmintrin.h

@@ -2198,8 +2198,9 @@ _mm_storer_ps(float *__p, __m128 __a)
 #define _MM_HINT_NTA 0
 
 #ifndef _MSC_VER
-/* FIXME: We have to #define this because "sel" must be a constant integer, and
-   Sema doesn't do any form of constant propagation yet. */
+// If _MSC_VER is defined, we use the builtin variant of _mm_prefetch.
+// Otherwise, we provide this macro, which includes a cast, allowing the user
+// to pass a pointer of any time. The _mm_prefetch accepts char to match MSVC.
 
 /// Loads one cache line of data from the specified address to a location
 ///    closer to the processor.