@cpp-linter/clang-tidy-win32-x64 20.1.0 → 22.1.8

package/lib/clang/{20 → 22}/include/f16cintrin.h

@@ -15,10 +15,21 @@
 #define __F16CINTRIN_H
 
 /* Define the default attributes for the functions in this file. */
-#define __DEFAULT_FN_ATTRS128 \
-  __attribute__((__always_inline__, __nodebug__, __target__("f16c"), __min_vector_width__(128)))
-#define __DEFAULT_FN_ATTRS256 \
-  __attribute__((__always_inline__, __nodebug__, __target__("f16c"), __min_vector_width__(256)))
+#if defined(__cplusplus) && (__cplusplus >= 201103L)
+#define __DEFAULT_FN_ATTRS128                                                  \
+  __attribute__((__always_inline__, __nodebug__, __target__("f16c"),           \
+                 __min_vector_width__(128))) constexpr
+#define __DEFAULT_FN_ATTRS256                                                  \
+  __attribute__((__always_inline__, __nodebug__, __target__("f16c"),           \
+                 __min_vector_width__(256))) constexpr
+#else
+#define __DEFAULT_FN_ATTRS128                                                  \
+  __attribute__((__always_inline__, __nodebug__, __target__("f16c"),           \
+                 __min_vector_width__(128)))
+#define __DEFAULT_FN_ATTRS256                                                  \
+  __attribute__((__always_inline__, __nodebug__, __target__("f16c"),           \
+                 __min_vector_width__(256)))
+#endif
 
 /* NOTE: Intel documents the 128-bit versions of these as being in emmintrin.h,
  * but that's because icc can emulate these without f16c using a library call.
@@ -38,9 +49,7 @@
 static __inline float __DEFAULT_FN_ATTRS128
 _cvtsh_ss(unsigned short __a)
 {
-  __v8hi __v = {(short)__a, 0, 0, 0, 0, 0, 0, 0};
-  __v4sf __r = __builtin_ia32_vcvtph2ps(__v);
-  return __r[0];
+  return (float)__builtin_bit_cast(__fp16, __a);
 }
 
 /// Converts a 32-bit single-precision float value to a 16-bit
@@ -109,7 +118,10 @@ _cvtsh_ss(unsigned short __a)
 static __inline __m128 __DEFAULT_FN_ATTRS128
 _mm_cvtph_ps(__m128i __a)
 {
-  return (__m128)__builtin_ia32_vcvtph2ps((__v8hi)__a);
+  typedef __fp16 __v4fp16 __attribute__((__vector_size__(8)));
+
+  __v4hi __v = __builtin_shufflevector((__v8hi)__a, (__v8hi)__a, 0, 1, 2, 3);
+  return (__m128) __builtin_convertvector((__v4fp16)__v, __v4sf);
 }
 
 /// Converts a 256-bit vector of [8 x float] into a 128-bit vector
@@ -153,7 +165,9 @@ _mm_cvtph_ps(__m128i __a)
 static __inline __m256 __DEFAULT_FN_ATTRS256
 _mm256_cvtph_ps(__m128i __a)
 {
-  return (__m256)__builtin_ia32_vcvtph2ps256((__v8hi)__a);
+  typedef __fp16 __v8fp16 __attribute__((__vector_size__(16), __aligned__(16)));
+
+  return (__m256) __builtin_convertvector((__v8fp16)__a, __v8sf);
 }
 
 #undef __DEFAULT_FN_ATTRS128

package/lib/clang/22/include/float.h

@@ -0,0 +1,48 @@
+/*===---- float.h - Characteristics of floating point types ----------------===
+ *
+ * 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
+ *
+ *===-----------------------------------------------------------------------===
+ */
+
+#if defined(__MVS__) && __has_include_next(<float.h>)
+#include <__float_header_macro.h>
+#include_next <float.h>
+#else
+
+#if !defined(__need_infinity_nan)
+#define __need_float_float
+#if (defined(__STDC_VERSION__) && __STDC_VERSION__ >= 202311L) ||              \
+    !defined(__STRICT_ANSI__)
+#define __need_infinity_nan
+#endif
+#include <__float_header_macro.h>
+#endif
+
+#ifdef __need_float_float
+/* If we're on MinGW, fall back to the system's float.h, which might have
+ * additional definitions provided for Windows.
+ * For more details see http://msdn.microsoft.com/en-us/library/y0ybw9fy.aspx
+ *
+ * Also fall back on AIX to allow additional definitions and
+ * implementation-defined values.
+ */
+#if (defined(__MINGW32__) || defined(_MSC_VER) || defined(_AIX)) &&            \
+    __STDC_HOSTED__ && __has_include_next(<float.h>)
+
+#  include_next <float.h>
+
+#endif
+
+#include <__float_float.h>
+#undef __need_float_float
+#endif
+
+#ifdef __need_infinity_nan
+#include <__float_infinity_nan.h>
+#undef __need_infinity_nan
+#endif
+
+#endif /* __MVS__ */

package/lib/clang/22/include/fma4intrin.h

@@ -0,0 +1,220 @@
+/*===---- fma4intrin.h - FMA4 intrinsics -----------------------------------===
+ *
+ * Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+ * See https://llvm.org/LICENSE.txt for license information.
+ * SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+ *
+ *===-----------------------------------------------------------------------===
+ */
+
+#ifndef __X86INTRIN_H
+#error "Never use <fma4intrin.h> directly; include <x86intrin.h> instead."
+#endif
+
+#ifndef __FMA4INTRIN_H
+#define __FMA4INTRIN_H
+
+#include <pmmintrin.h>
+
+/* Define the default attributes for the functions in this file. */
+#define __DEFAULT_FN_ATTRS128 __attribute__((__always_inline__, __nodebug__, __target__("fma4"), __min_vector_width__(128)))
+#define __DEFAULT_FN_ATTRS256 __attribute__((__always_inline__, __nodebug__, __target__("fma4"), __min_vector_width__(256)))
+
+#if defined(__cplusplus) && (__cplusplus >= 201103L)
+#define __DEFAULT_FN_ATTRS128_CONSTEXPR __DEFAULT_FN_ATTRS128 constexpr
+#define __DEFAULT_FN_ATTRS256_CONSTEXPR __DEFAULT_FN_ATTRS256 constexpr
+#else
+#define __DEFAULT_FN_ATTRS128_CONSTEXPR __DEFAULT_FN_ATTRS128
+#define __DEFAULT_FN_ATTRS256_CONSTEXPR __DEFAULT_FN_ATTRS256
+#endif
+
+static __inline__ __m128 __DEFAULT_FN_ATTRS128_CONSTEXPR
+_mm_macc_ps(__m128 __A, __m128 __B, __m128 __C) {
+  return (__m128)__builtin_elementwise_fma((__v4sf)__A, (__v4sf)__B,
+                                           (__v4sf)__C);
+}
+
+static __inline__ __m128d __DEFAULT_FN_ATTRS128_CONSTEXPR
+_mm_macc_pd(__m128d __A, __m128d __B, __m128d __C) {
+  return (__m128d)__builtin_elementwise_fma((__v2df)__A, (__v2df)__B,
+                                            (__v2df)__C);
+}
+
+static __inline__ __m128 __DEFAULT_FN_ATTRS128_CONSTEXPR
+_mm_macc_ss(__m128 __A, __m128 __B, __m128 __C) {
+  return _mm_set_ss(__builtin_elementwise_fma(__A[0], __B[0], __C[0]));
+}
+
+static __inline__ __m128d __DEFAULT_FN_ATTRS128_CONSTEXPR
+_mm_macc_sd(__m128d __A, __m128d __B, __m128d __C) {
+  return _mm_set_sd(__builtin_elementwise_fma(__A[0], __B[0], __C[0]));
+}
+
+static __inline__ __m128 __DEFAULT_FN_ATTRS128_CONSTEXPR
+_mm_msub_ps(__m128 __A, __m128 __B, __m128 __C) {
+  return (__m128)__builtin_elementwise_fma((__v4sf)__A, (__v4sf)__B,
+                                           -(__v4sf)__C);
+}
+
+static __inline__ __m128d __DEFAULT_FN_ATTRS128_CONSTEXPR
+_mm_msub_pd(__m128d __A, __m128d __B, __m128d __C) {
+  return (__m128d)__builtin_elementwise_fma((__v2df)__A, (__v2df)__B,
+                                            -(__v2df)__C);
+}
+
+static __inline__ __m128 __DEFAULT_FN_ATTRS128_CONSTEXPR
+_mm_msub_ss(__m128 __A, __m128 __B, __m128 __C) {
+  return _mm_set_ss(__builtin_elementwise_fma(__A[0], __B[0], -__C[0]));
+}
+
+static __inline__ __m128d __DEFAULT_FN_ATTRS128_CONSTEXPR
+_mm_msub_sd(__m128d __A, __m128d __B, __m128d __C) {
+  return _mm_set_sd(__builtin_elementwise_fma(__A[0], __B[0], -__C[0]));
+}
+
+static __inline__ __m128 __DEFAULT_FN_ATTRS128_CONSTEXPR
+_mm_nmacc_ps(__m128 __A, __m128 __B, __m128 __C) {
+  return (__m128)__builtin_elementwise_fma(-(__v4sf)__A, (__v4sf)__B,
+                                           (__v4sf)__C);
+}
+
+static __inline__ __m128d __DEFAULT_FN_ATTRS128_CONSTEXPR
+_mm_nmacc_pd(__m128d __A, __m128d __B, __m128d __C) {
+  return (__m128d)__builtin_elementwise_fma(-(__v2df)__A, (__v2df)__B,
+                                            (__v2df)__C);
+}
+
+static __inline__ __m128 __DEFAULT_FN_ATTRS128_CONSTEXPR
+_mm_nmacc_ss(__m128 __A, __m128 __B, __m128 __C) {
+  return _mm_set_ss(__builtin_elementwise_fma(-__A[0], __B[0], __C[0]));
+}
+
+static __inline__ __m128d __DEFAULT_FN_ATTRS128_CONSTEXPR
+_mm_nmacc_sd(__m128d __A, __m128d __B, __m128d __C) {
+  return _mm_set_sd(__builtin_elementwise_fma(-__A[0], __B[0], __C[0]));
+}
+
+static __inline__ __m128 __DEFAULT_FN_ATTRS128_CONSTEXPR
+_mm_nmsub_ps(__m128 __A, __m128 __B, __m128 __C) {
+  return (__m128)__builtin_elementwise_fma(-(__v4sf)__A, (__v4sf)__B,
+                                           -(__v4sf)__C);
+}
+
+static __inline__ __m128d __DEFAULT_FN_ATTRS128_CONSTEXPR
+_mm_nmsub_pd(__m128d __A, __m128d __B, __m128d __C) {
+  return (__m128d)__builtin_elementwise_fma(-(__v2df)__A, (__v2df)__B,
+                                            -(__v2df)__C);
+}
+
+static __inline__ __m128 __DEFAULT_FN_ATTRS128_CONSTEXPR
+_mm_nmsub_ss(__m128 __A, __m128 __B, __m128 __C) {
+  return _mm_set_ss(__builtin_elementwise_fma(-__A[0], __B[0], -__C[0]));
+}
+
+static __inline__ __m128d __DEFAULT_FN_ATTRS128_CONSTEXPR
+_mm_nmsub_sd(__m128d __A, __m128d __B, __m128d __C) {
+  return _mm_set_sd(__builtin_elementwise_fma(-__A[0], __B[0], -__C[0]));
+}
+
+static __inline__ __m128 __DEFAULT_FN_ATTRS128
+_mm_maddsub_ps(__m128 __A, __m128 __B, __m128 __C)
+{
+  return (__m128)__builtin_ia32_vfmaddsubps((__v4sf)__A, (__v4sf)__B, (__v4sf)__C);
+}
+
+static __inline__ __m128d __DEFAULT_FN_ATTRS128
+_mm_maddsub_pd(__m128d __A, __m128d __B, __m128d __C)
+{
+  return (__m128d)__builtin_ia32_vfmaddsubpd((__v2df)__A, (__v2df)__B, (__v2df)__C);
+}
+
+static __inline__ __m128 __DEFAULT_FN_ATTRS128
+_mm_msubadd_ps(__m128 __A, __m128 __B, __m128 __C)
+{
+  return (__m128)__builtin_ia32_vfmaddsubps((__v4sf)__A, (__v4sf)__B, -(__v4sf)__C);
+}
+
+static __inline__ __m128d __DEFAULT_FN_ATTRS128
+_mm_msubadd_pd(__m128d __A, __m128d __B, __m128d __C)
+{
+  return (__m128d)__builtin_ia32_vfmaddsubpd((__v2df)__A, (__v2df)__B, -(__v2df)__C);
+}
+
+static __inline__ __m256 __DEFAULT_FN_ATTRS256_CONSTEXPR
+_mm256_macc_ps(__m256 __A, __m256 __B, __m256 __C) {
+  return (__m256)__builtin_elementwise_fma((__v8sf)__A, (__v8sf)__B,
+                                           (__v8sf)__C);
+}
+
+static __inline__ __m256d __DEFAULT_FN_ATTRS256_CONSTEXPR
+_mm256_macc_pd(__m256d __A, __m256d __B, __m256d __C) {
+  return (__m256d)__builtin_elementwise_fma((__v4df)__A, (__v4df)__B,
+                                            (__v4df)__C);
+}
+
+static __inline__ __m256 __DEFAULT_FN_ATTRS256_CONSTEXPR
+_mm256_msub_ps(__m256 __A, __m256 __B, __m256 __C) {
+  return (__m256)__builtin_elementwise_fma((__v8sf)__A, (__v8sf)__B,
+                                           -(__v8sf)__C);
+}
+
+static __inline__ __m256d __DEFAULT_FN_ATTRS256_CONSTEXPR
+_mm256_msub_pd(__m256d __A, __m256d __B, __m256d __C) {
+  return (__m256d)__builtin_elementwise_fma((__v4df)__A, (__v4df)__B,
+                                            -(__v4df)__C);
+}
+
+static __inline__ __m256 __DEFAULT_FN_ATTRS256_CONSTEXPR
+_mm256_nmacc_ps(__m256 __A, __m256 __B, __m256 __C) {
+  return (__m256)__builtin_elementwise_fma(-(__v8sf)__A, (__v8sf)__B,
+                                           (__v8sf)__C);
+}
+
+static __inline__ __m256d __DEFAULT_FN_ATTRS256_CONSTEXPR
+_mm256_nmacc_pd(__m256d __A, __m256d __B, __m256d __C) {
+  return (__m256d)__builtin_elementwise_fma(-(__v4df)__A, (__v4df)__B,
+                                            (__v4df)__C);
+}
+
+static __inline__ __m256 __DEFAULT_FN_ATTRS256_CONSTEXPR
+_mm256_nmsub_ps(__m256 __A, __m256 __B, __m256 __C) {
+  return (__m256)__builtin_elementwise_fma(-(__v8sf)__A, (__v8sf)__B,
+                                           -(__v8sf)__C);
+}
+
+static __inline__ __m256d __DEFAULT_FN_ATTRS256_CONSTEXPR
+_mm256_nmsub_pd(__m256d __A, __m256d __B, __m256d __C) {
+  return (__m256d)__builtin_elementwise_fma(-(__v4df)__A, (__v4df)__B,
+                                            -(__v4df)__C);
+}
+
+static __inline__ __m256 __DEFAULT_FN_ATTRS256
+_mm256_maddsub_ps(__m256 __A, __m256 __B, __m256 __C)
+{
+  return (__m256)__builtin_ia32_vfmaddsubps256((__v8sf)__A, (__v8sf)__B, (__v8sf)__C);
+}
+
+static __inline__ __m256d __DEFAULT_FN_ATTRS256
+_mm256_maddsub_pd(__m256d __A, __m256d __B, __m256d __C)
+{
+  return (__m256d)__builtin_ia32_vfmaddsubpd256((__v4df)__A, (__v4df)__B, (__v4df)__C);
+}
+
+static __inline__ __m256 __DEFAULT_FN_ATTRS256
+_mm256_msubadd_ps(__m256 __A, __m256 __B, __m256 __C)
+{
+  return (__m256)__builtin_ia32_vfmaddsubps256((__v8sf)__A, (__v8sf)__B, -(__v8sf)__C);
+}
+
+static __inline__ __m256d __DEFAULT_FN_ATTRS256
+_mm256_msubadd_pd(__m256d __A, __m256d __B, __m256d __C)
+{
+  return (__m256d)__builtin_ia32_vfmaddsubpd256((__v4df)__A, (__v4df)__B, -(__v4df)__C);
+}
+
+#undef __DEFAULT_FN_ATTRS128
+#undef __DEFAULT_FN_ATTRS256
+#undef __DEFAULT_FN_ATTRS128_CONSTEXPR
+#undef __DEFAULT_FN_ATTRS256_CONSTEXPR
+
+#endif /* __FMA4INTRIN_H */