llvm-runtime-resources-darwin 0.0.0 → 22.1.8

package/include/avx10_2_512convertintrin.h

@@ -0,0 +1,322 @@
+/*===--------- avx10_2_512convertintrin.h - AVX10_2_512CONVERT -------------===
+ *
+ * Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+ * See https://llvm.org/LICENSE.txt for license information.
+ * SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+ *
+ *===-----------------------------------------------------------------------===
+ */
+#ifndef __IMMINTRIN_H
+#error                                                                         \
+    "Never use <avx10_2_512convertintrin.h> directly; include <immintrin.h> instead."
+#endif // __IMMINTRIN_H
+
+#ifdef __SSE2__
+
+#ifndef __AVX10_2_512CONVERTINTRIN_H
+#define __AVX10_2_512CONVERTINTRIN_H
+
+/* Define the default attributes for the functions in this file. */
+#define __DEFAULT_FN_ATTRS512                                                  \
+  __attribute__((__always_inline__, __nodebug__, __target__("avx10.2"),        \
+                 __min_vector_width__(512)))
+
+static __inline__ __m512h __DEFAULT_FN_ATTRS512 _mm512_cvtx2ps_ph(__m512 __A,
+                                                                  __m512 __B) {
+  return (__m512h)__builtin_ia32_vcvt2ps2phx512_mask(
+      (__v16sf)__A, (__v16sf)__B, (__v32hf)_mm512_setzero_ph(), (__mmask32)(-1),
+      _MM_FROUND_CUR_DIRECTION);
+}
+
+static __inline__ __m512h __DEFAULT_FN_ATTRS512
+_mm512_mask_cvtx2ps_ph(__m512h __W, __mmask32 __U, __m512 __A, __m512 __B) {
+  return (__m512h)__builtin_ia32_vcvt2ps2phx512_mask(
+      (__v16sf)__A, (__v16sf)__B, (__v32hf)__W, (__mmask32)__U,
+      _MM_FROUND_CUR_DIRECTION);
+}
+
+static __inline__ __m512h __DEFAULT_FN_ATTRS512
+_mm512_maskz_cvtx2ps_ph(__mmask32 __U, __m512 __A, __m512 __B) {
+  return (__m512h)__builtin_ia32_vcvt2ps2phx512_mask(
+      (__v16sf)__A, (__v16sf)__B, (__v32hf)_mm512_setzero_ph(), (__mmask32)__U,
+      _MM_FROUND_CUR_DIRECTION);
+}
+
+#define _mm512_cvtx_round2ps_ph(A, B, R)                                       \
+  ((__m512h)__builtin_ia32_vcvt2ps2phx512_mask(                                \
+      (__v16sf)(A), (__v16sf)(B), (__v32hf)_mm512_undefined_ph(),              \
+      (__mmask32)(-1), (const int)(R)))
+
+#define _mm512_mask_cvtx_round2ps_ph(W, U, A, B, R)                            \
+  ((__m512h)__builtin_ia32_vcvt2ps2phx512_mask((__v16sf)(A), (__v16sf)(B),     \
+                                               (__v32hf)(W), (__mmask32)(U),   \
+                                               (const int)(R)))
+
+#define _mm512_maskz_cvtx_round2ps_ph(U, A, B, R)                              \
+  ((__m512h)__builtin_ia32_vcvt2ps2phx512_mask(                                \
+      (__v16sf)(A), (__v16sf)(B), (__v32hf)_mm512_setzero_ph(),                \
+      (__mmask32)(U), (const int)(R)))
+
+static __inline__ __m256i __DEFAULT_FN_ATTRS512
+_mm512_cvtbiasph_bf8(__m512i __A, __m512h __B) {
+  return (__m256i)__builtin_ia32_vcvtbiasph2bf8_512_mask(
+      (__v64qi)__A, (__v32hf)__B, (__v32qi)_mm256_undefined_si256(),
+      (__mmask32)-1);
+}
+
+static __inline__ __m256i __DEFAULT_FN_ATTRS512 _mm512_mask_cvtbiasph_bf8(
+    __m256i __W, __mmask32 __U, __m512i __A, __m512h __B) {
+  return (__m256i)__builtin_ia32_vcvtbiasph2bf8_512_mask(
+      (__v64qi)__A, (__v32hf)__B, (__v32qi)(__m256i)__W, (__mmask32)__U);
+}
+
+static __inline__ __m256i __DEFAULT_FN_ATTRS512
+_mm512_maskz_cvtbiasph_bf8(__mmask32 __U, __m512i __A, __m512h __B) {
+  return (__m256i)__builtin_ia32_vcvtbiasph2bf8_512_mask(
+      (__v64qi)__A, (__v32hf)__B, (__v32qi)(__m256i)_mm256_setzero_si256(),
+      (__mmask32)__U);
+}
+
+static __inline__ __m256i __DEFAULT_FN_ATTRS512
+_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_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_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);
+}
+
+static __inline__ __m256i __DEFAULT_FN_ATTRS512
+_mm512_cvtbiasph_hf8(__m512i __A, __m512h __B) {
+  return (__m256i)__builtin_ia32_vcvtbiasph2hf8_512_mask(
+      (__v64qi)__A, (__v32hf)__B, (__v32qi)_mm256_undefined_si256(),
+      (__mmask32)-1);
+}
+
+static __inline__ __m256i __DEFAULT_FN_ATTRS512 _mm512_mask_cvtbiasph_hf8(
+    __m256i __W, __mmask32 __U, __m512i __A, __m512h __B) {
+  return (__m256i)__builtin_ia32_vcvtbiasph2hf8_512_mask(
+      (__v64qi)__A, (__v32hf)__B, (__v32qi)(__m256i)__W, (__mmask32)__U);
+}
+
+static __inline__ __m256i __DEFAULT_FN_ATTRS512
+_mm512_maskz_cvtbiasph_hf8(__mmask32 __U, __m512i __A, __m512h __B) {
+  return (__m256i)__builtin_ia32_vcvtbiasph2hf8_512_mask(
+      (__v64qi)__A, (__v32hf)__B, (__v32qi)(__m256i)_mm256_setzero_si256(),
+      (__mmask32)__U);
+}
+
+static __inline__ __m256i __DEFAULT_FN_ATTRS512
+_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_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_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);
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_cvt2ph_bf8(__m512h __A,
+                                                                  __m512h __B) {
+  return (__m512i)__builtin_ia32_vcvt2ph2bf8_512((__v32hf)(__A),
+                                                 (__v32hf)(__B));
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
+_mm512_mask_cvt2ph_bf8(__m512i __W, __mmask64 __U, __m512h __A, __m512h __B) {
+  return (__m512i)__builtin_ia32_selectb_512(
+      (__mmask64)__U, (__v64qi)_mm512_cvt2ph_bf8(__A, __B), (__v64qi)__W);
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
+_mm512_maskz_cvt2ph_bf8(__mmask64 __U, __m512h __A, __m512h __B) {
+  return (__m512i)__builtin_ia32_selectb_512(
+      (__mmask64)__U, (__v64qi)_mm512_cvt2ph_bf8(__A, __B),
+      (__v64qi)(__m512i)_mm512_setzero_si512());
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
+_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_cvts_2ph_bf8(__m512i __W, __mmask64 __U, __m512h __A, __m512h __B) {
+  return (__m512i)__builtin_ia32_selectb_512(
+      (__mmask64)__U, (__v64qi)_mm512_cvts_2ph_bf8(__A, __B), (__v64qi)__W);
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
+_mm512_maskz_cvts_2ph_bf8(__mmask64 __U, __m512h __A, __m512h __B) {
+  return (__m512i)__builtin_ia32_selectb_512(
+      (__mmask64)__U, (__v64qi)_mm512_cvts_2ph_bf8(__A, __B),
+      (__v64qi)(__m512i)_mm512_setzero_si512());
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_cvt2ph_hf8(__m512h __A,
+                                                                  __m512h __B) {
+  return (__m512i)__builtin_ia32_vcvt2ph2hf8_512((__v32hf)(__A),
+                                                 (__v32hf)(__B));
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
+_mm512_mask_cvt2ph_hf8(__m512i __W, __mmask64 __U, __m512h __A, __m512h __B) {
+  return (__m512i)__builtin_ia32_selectb_512(
+      (__mmask64)__U, (__v64qi)_mm512_cvt2ph_hf8(__A, __B), (__v64qi)__W);
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
+_mm512_maskz_cvt2ph_hf8(__mmask64 __U, __m512h __A, __m512h __B) {
+  return (__m512i)__builtin_ia32_selectb_512(
+      (__mmask64)__U, (__v64qi)_mm512_cvt2ph_hf8(__A, __B),
+      (__v64qi)(__m512i)_mm512_setzero_si512());
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
+_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_cvts_2ph_hf8(__m512i __W, __mmask64 __U, __m512h __A, __m512h __B) {
+  return (__m512i)__builtin_ia32_selectb_512(
+      (__mmask64)__U, (__v64qi)_mm512_cvts_2ph_hf8(__A, __B), (__v64qi)__W);
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS512
+_mm512_maskz_cvts_2ph_hf8(__mmask64 __U, __m512h __A, __m512h __B) {
+  return (__m512i)__builtin_ia32_selectb_512(
+      (__mmask64)__U, (__v64qi)_mm512_cvts_2ph_hf8(__A, __B),
+      (__v64qi)(__m512i)_mm512_setzero_si512());
+}
+
+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_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_ph(__mmask32 __U, __m256i __A) {
+  return (__m512h)__builtin_ia32_vcvthf8_2ph512_mask(
+      (__v32qi)__A, (__v32hf)(__m512h)_mm512_setzero_ph(), (__mmask32)__U);
+}
+
+static __inline__ __m256i __DEFAULT_FN_ATTRS512 _mm512_cvtph_bf8(__m512h __A) {
+  return (__m256i)__builtin_ia32_vcvtph2bf8_512_mask(
+      (__v32hf)__A, (__v32qi)(__m256i)_mm256_undefined_si256(), (__mmask32)-1);
+}
+
+static __inline__ __m256i __DEFAULT_FN_ATTRS512
+_mm512_mask_cvtph_bf8(__m256i __W, __mmask32 __U, __m512h __A) {
+  return (__m256i)__builtin_ia32_vcvtph2bf8_512_mask(
+      (__v32hf)__A, (__v32qi)(__m256i)__W, (__mmask32)__U);
+}
+
+static __inline__ __m256i __DEFAULT_FN_ATTRS512
+_mm512_maskz_cvtph_bf8(__mmask32 __U, __m512h __A) {
+  return (__m256i)__builtin_ia32_vcvtph2bf8_512_mask(
+      (__v32hf)__A, (__v32qi)(__m256i)_mm256_setzero_si256(), (__mmask32)__U);
+}
+
+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_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_cvts_ph_bf8(__mmask32 __U, __m512h __A) {
+  return (__m256i)__builtin_ia32_vcvtph2bf8s_512_mask(
+      (__v32hf)__A, (__v32qi)(__m256i)_mm256_setzero_si256(), (__mmask32)__U);
+}
+
+static __inline__ __m256i __DEFAULT_FN_ATTRS512 _mm512_cvtph_hf8(__m512h __A) {
+  return (__m256i)__builtin_ia32_vcvtph2hf8_512_mask(
+      (__v32hf)__A, (__v32qi)(__m256i)_mm256_undefined_si256(), (__mmask32)-1);
+}
+
+static __inline__ __m256i __DEFAULT_FN_ATTRS512
+_mm512_mask_cvtph_hf8(__m256i __W, __mmask32 __U, __m512h __A) {
+  return (__m256i)__builtin_ia32_vcvtph2hf8_512_mask(
+      (__v32hf)__A, (__v32qi)(__m256i)__W, (__mmask32)__U);
+}
+
+static __inline__ __m256i __DEFAULT_FN_ATTRS512
+_mm512_maskz_cvtph_hf8(__mmask32 __U, __m512h __A) {
+  return (__m256i)__builtin_ia32_vcvtph2hf8_512_mask(
+      (__v32hf)__A, (__v32qi)(__m256i)_mm256_setzero_si256(), (__mmask32)__U);
+}
+
+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_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_cvts_ph_hf8(__mmask32 __U, __m512h __A) {
+  return (__m256i)__builtin_ia32_vcvtph2hf8s_512_mask(
+      (__v32hf)__A, (__v32qi)(__m256i)_mm256_setzero_si256(), (__mmask32)__U);
+}
+
+static __inline __m512h __DEFAULT_FN_ATTRS512 _mm512_cvtbf8_ph(__m256i __A) {
+  return _mm512_castsi512_ph(_mm512_slli_epi16(_mm512_cvtepi8_epi16(__A), 8));
+}
+
+static __inline __m512h __DEFAULT_FN_ATTRS512
+_mm512_mask_cvtbf8_ph(__m512h __S, __mmask32 __U, __m256i __A) {
+  return _mm512_castsi512_ph(
+      _mm512_mask_slli_epi16((__m512i)__S, __U, _mm512_cvtepi8_epi16(__A), 8));
+}
+
+static __inline __m512h __DEFAULT_FN_ATTRS512
+_mm512_maskz_cvtbf8_ph(__mmask32 __U, __m256i __A) {
+  return _mm512_castsi512_ph(
+      _mm512_slli_epi16(_mm512_maskz_cvtepi8_epi16(__U, __A), 8));
+}
+
+#undef __DEFAULT_FN_ATTRS512
+
+#endif // __AVX10_2_512CONVERTINTRIN_H
+#endif // __SSE2__

package/include/avx10_2_512minmaxintrin.h

@@ -0,0 +1,127 @@
+/*===---- avx10_2_512minmaxintrin.h - AVX10_2_512MINMAX intrinsics ---------===
+ *
+ * Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+ * See https://llvm.org/LICENSE.txt for license information.
+ * SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+ *
+ *===-----------------------------------------------------------------------===
+ */
+#ifndef __IMMINTRIN_H
+#error                                                                         \
+    "Never use <avx10_2_512minmaxintrin.h> directly; include <immintrin.h> instead."
+#endif // __IMMINTRIN_H
+
+#ifndef __AVX10_2_512MINMAXINTRIN_H
+#define __AVX10_2_512MINMAXINTRIN_H
+
+#define _mm512_minmax_pbh(A, B, C)                                             \
+  ((__m512bh)__builtin_ia32_vminmaxbf16512((__v32bf)(__m512bh)(A),             \
+                                           (__v32bf)(__m512bh)(A), (int)(C)))
+
+#define _mm512_mask_minmax_pbh(W, U, A, B, C)                                  \
+  ((__m512bh)__builtin_ia32_selectpbf_512(                                     \
+      (__mmask32)(U),                                                          \
+      (__v32bf)_mm512_minmax_pbh((__v32bf)(__m512bh)(A),                       \
+                                 (__v32bf)(__m512bh)(B), (int)(C)),            \
+      (__v32bf)(__m512bh)(W)))
+
+#define _mm512_maskz_minmax_pbh(U, A, B, C)                                    \
+  ((__m512bh)__builtin_ia32_selectpbf_512(                                     \
+      (__mmask32)(U),                                                          \
+      (__v32bf)_mm512_minmax_pbh((__v32bf)(__m512bh)(A),                       \
+                                 (__v32bf)(__m512bh)(B), (int)(C)),            \
+      (__v32bf) __builtin_bit_cast(__m512bh, _mm512_setzero_ps())))
+
+#define _mm512_minmax_pd(A, B, C)                                              \
+  ((__m512d)__builtin_ia32_vminmaxpd512_round_mask(                            \
+      (__v8df)(__m512d)(A), (__v8df)(__m512d)(B), (int)(C),                    \
+      (__v8df)_mm512_undefined_pd(), (__mmask8)-1,                             \
+      _MM_FROUND_CUR_DIRECTION))
+
+#define _mm512_mask_minmax_pd(W, U, A, B, C)                                   \
+  ((__m512d)__builtin_ia32_vminmaxpd512_round_mask(                            \
+      (__v8df)(__m512d)(A), (__v8df)(__m512d)(B), (int)(C),                    \
+      (__v8df)(__m512d)(W), (__mmask8)(U), _MM_FROUND_CUR_DIRECTION))
+
+#define _mm512_maskz_minmax_pd(U, A, B, C)                                     \
+  ((__m512d)__builtin_ia32_vminmaxpd512_round_mask(                            \
+      (__v8df)(__m512d)(A), (__v8df)(__m512d)(B), (int)(C),                    \
+      (__v8df)_mm512_setzero_pd(), (__mmask8)(U), _MM_FROUND_CUR_DIRECTION))
+
+#define _mm512_minmax_round_pd(A, B, C, R)                                     \
+  ((__m512d)__builtin_ia32_vminmaxpd512_round_mask(                            \
+      (__v8df)(__m512d)(A), (__v8df)(__m512d)(B), (int)(C),                    \
+      (__v8df)_mm512_undefined_pd(), (__mmask8)-1, (int)(R)))
+
+#define _mm512_mask_minmax_round_pd(W, U, A, B, C, R)                          \
+  ((__m512d)__builtin_ia32_vminmaxpd512_round_mask(                            \
+      (__v8df)(__m512d)(A), (__v8df)(__m512d)(B), (int)(C),                    \
+      (__v8df)(__m512d)(W), (__mmask8)(U), (int)(R)))
+
+#define _mm512_maskz_minmax_round_pd(U, A, B, C, R)                            \
+  ((__m512d)__builtin_ia32_vminmaxpd512_round_mask(                            \
+      (__v8df)(__m512d)(A), (__v8df)(__m512d)(B), (int)(C),                    \
+      (__v8df)_mm512_setzero_pd(), (__mmask8)(U), (int)(R)))
+
+#define _mm512_minmax_ph(A, B, C)                                              \
+  ((__m512h)__builtin_ia32_vminmaxph512_round_mask(                            \
+      (__v32hf)(__m512h)(A), (__v32hf)(__m512h)(B), (int)(C),                  \
+      (__v32hf)_mm512_undefined_ph(), (__mmask32)-1,                           \
+      _MM_FROUND_CUR_DIRECTION))
+
+#define _mm512_mask_minmax_ph(W, U, A, B, C)                                   \
+  ((__m512h)__builtin_ia32_vminmaxph512_round_mask(                            \
+      (__v32hf)(__m512h)(A), (__v32hf)(__m512h)(B), (int)(C),                  \
+      (__v32hf)(__m512h)(W), (__mmask32)(U), _MM_FROUND_CUR_DIRECTION))
+
+#define _mm512_maskz_minmax_ph(U, A, B, C)                                     \
+  ((__m512h)__builtin_ia32_vminmaxph512_round_mask(                            \
+      (__v32hf)(__m512h)(A), (__v32hf)(__m512h)(B), (int)(C),                  \
+      (__v32hf)_mm512_setzero_ph(), (__mmask32)(U), _MM_FROUND_CUR_DIRECTION))
+
+#define _mm512_minmax_round_ph(A, B, C, R)                                     \
+  ((__m512h)__builtin_ia32_vminmaxph512_round_mask(                            \
+      (__v32hf)(__m512h)(A), (__v32hf)(__m512h)(B), (int)(C),                  \
+      (__v32hf)_mm512_undefined_ph(), (__mmask32)-1, (int)(R)))
+
+#define _mm512_mask_minmax_round_ph(W, U, A, B, C, R)                          \
+  ((__m512h)__builtin_ia32_vminmaxph512_round_mask(                            \
+      (__v32hf)(__m512h)(A), (__v32hf)(__m512h)(B), (int)(C),                  \
+      (__v32hf)(__m512h)(W), (__mmask32)(U), (int)(R)))
+
+#define _mm512_maskz_minmax_round_ph(U, A, B, C, R)                            \
+  ((__m512h)__builtin_ia32_vminmaxph512_round_mask(                            \
+      (__v32hf)(__m512h)(A), (__v32hf)(__m512h)(B), (int)(C),                  \
+      (__v32hf)_mm512_setzero_ph(), (__mmask32)(U), (int)(R)))
+
+#define _mm512_minmax_ps(A, B, C)                                              \
+  ((__m512)__builtin_ia32_vminmaxps512_round_mask(                             \
+      (__v16sf)(__m512)(A), (__v16sf)(__m512)(B), (int)(C),                    \
+      (__v16sf)_mm512_undefined_ps(), (__mmask16)-1,                           \
+      _MM_FROUND_CUR_DIRECTION))
+
+#define _mm512_mask_minmax_ps(W, U, A, B, C)                                   \
+  ((__m512)__builtin_ia32_vminmaxps512_round_mask(                             \
+      (__v16sf)(__m512)(A), (__v16sf)(__m512)(B), (int)(C), (__v16sf)(W),      \
+      (__mmask16)(U), _MM_FROUND_CUR_DIRECTION))
+
+#define _mm512_maskz_minmax_ps(U, A, B, C)                                     \
+  ((__m512)__builtin_ia32_vminmaxps512_round_mask(                             \
+      (__v16sf)(__m512)(A), (__v16sf)(__m512)(B), (int)(C),                    \
+      (__v16sf)_mm512_setzero_ps(), (__mmask16)(U), _MM_FROUND_CUR_DIRECTION))
+
+#define _mm512_minmax_round_ps(A, B, C, R)                                     \
+  ((__m512)__builtin_ia32_vminmaxps512_round_mask(                             \
+      (__v16sf)(__m512)(A), (__v16sf)(__m512)(B), (int)(C),                    \
+      (__v16sf)_mm512_undefined_ps(), (__mmask16)-1, (int)(R)))
+
+#define _mm512_mask_minmax_round_ps(W, U, A, B, C, R)                          \
+  ((__m512)__builtin_ia32_vminmaxps512_round_mask(                             \
+      (__v16sf)(__m512)(A), (__v16sf)(__m512)(B), (int)(C), (__v16sf)(W),      \
+      (__mmask16)(U), (int)(R)))
+
+#define _mm512_maskz_minmax_round_ps(U, A, B, C, R)                            \
+  ((__m512)__builtin_ia32_vminmaxps512_round_mask(                             \
+      (__v16sf)(__m512)(A), (__v16sf)(__m512)(B), (int)(C),                    \
+      (__v16sf)_mm512_setzero_ps(), (__mmask16)(U), (int)(R)))
+#endif // __AVX10_2_512MINMAXINTRIN_H