@cpp-linter/clang-tidy-darwin-x64 13.0.0 → 16.0.3

package/lib/clang/{13.0.0 → 16}/include/avx512ifmavlintrin.h

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

package/lib/clang/{13.0.0 → 16}/include/avx512vbmi2intrin.h

@@ -129,88 +129,88 @@ _mm512_maskz_expandloadu_epi8(__mmask64 __U, void const *__P)
 }
 
 #define _mm512_shldi_epi64(A, B, I) \
-  (__m512i)__builtin_ia32_vpshldq512((__v8di)(__m512i)(A), \
-                                     (__v8di)(__m512i)(B), (int)(I))
+  ((__m512i)__builtin_ia32_vpshldq512((__v8di)(__m512i)(A), \
+                                      (__v8di)(__m512i)(B), (int)(I)))
 
 #define _mm512_mask_shldi_epi64(S, U, A, B, I) \
-  (__m512i)__builtin_ia32_selectq_512((__mmask8)(U), \
-                                    (__v8di)_mm512_shldi_epi64((A), (B), (I)), \
-                                    (__v8di)(__m512i)(S))
+  ((__m512i)__builtin_ia32_selectq_512((__mmask8)(U), \
+                                     (__v8di)_mm512_shldi_epi64((A), (B), (I)), \
+                                     (__v8di)(__m512i)(S)))
 
 #define _mm512_maskz_shldi_epi64(U, A, B, I) \
-  (__m512i)__builtin_ia32_selectq_512((__mmask8)(U), \
-                                    (__v8di)_mm512_shldi_epi64((A), (B), (I)), \
-                                    (__v8di)_mm512_setzero_si512())
+  ((__m512i)__builtin_ia32_selectq_512((__mmask8)(U), \
+                                     (__v8di)_mm512_shldi_epi64((A), (B), (I)), \
+                                     (__v8di)_mm512_setzero_si512()))
 
 #define _mm512_shldi_epi32(A, B, I) \
-  (__m512i)__builtin_ia32_vpshldd512((__v16si)(__m512i)(A), \
-                                     (__v16si)(__m512i)(B), (int)(I))
+  ((__m512i)__builtin_ia32_vpshldd512((__v16si)(__m512i)(A), \
+                                      (__v16si)(__m512i)(B), (int)(I)))
 
 #define _mm512_mask_shldi_epi32(S, U, A, B, I) \
-  (__m512i)__builtin_ia32_selectd_512((__mmask16)(U), \
-                                   (__v16si)_mm512_shldi_epi32((A), (B), (I)), \
-                                   (__v16si)(__m512i)(S))
+  ((__m512i)__builtin_ia32_selectd_512((__mmask16)(U), \
+                                    (__v16si)_mm512_shldi_epi32((A), (B), (I)), \
+                                    (__v16si)(__m512i)(S)))
 
 #define _mm512_maskz_shldi_epi32(U, A, B, I) \
-  (__m512i)__builtin_ia32_selectd_512((__mmask16)(U), \
-                                   (__v16si)_mm512_shldi_epi32((A), (B), (I)), \
-                                   (__v16si)_mm512_setzero_si512())
+  ((__m512i)__builtin_ia32_selectd_512((__mmask16)(U), \
+                                    (__v16si)_mm512_shldi_epi32((A), (B), (I)), \
+                                    (__v16si)_mm512_setzero_si512()))
 
 #define _mm512_shldi_epi16(A, B, I) \
-  (__m512i)__builtin_ia32_vpshldw512((__v32hi)(__m512i)(A), \
-                                     (__v32hi)(__m512i)(B), (int)(I))
+  ((__m512i)__builtin_ia32_vpshldw512((__v32hi)(__m512i)(A), \
+                                      (__v32hi)(__m512i)(B), (int)(I)))
 
 #define _mm512_mask_shldi_epi16(S, U, A, B, I) \
-  (__m512i)__builtin_ia32_selectw_512((__mmask32)(U), \
-                                   (__v32hi)_mm512_shldi_epi16((A), (B), (I)), \
-                                   (__v32hi)(__m512i)(S))
+  ((__m512i)__builtin_ia32_selectw_512((__mmask32)(U), \
+                                    (__v32hi)_mm512_shldi_epi16((A), (B), (I)), \
+                                    (__v32hi)(__m512i)(S)))
 
 #define _mm512_maskz_shldi_epi16(U, A, B, I) \
-  (__m512i)__builtin_ia32_selectw_512((__mmask32)(U), \
-                                   (__v32hi)_mm512_shldi_epi16((A), (B), (I)), \
-                                   (__v32hi)_mm512_setzero_si512())
+  ((__m512i)__builtin_ia32_selectw_512((__mmask32)(U), \
+                                    (__v32hi)_mm512_shldi_epi16((A), (B), (I)), \
+                                    (__v32hi)_mm512_setzero_si512()))
 
 #define _mm512_shrdi_epi64(A, B, I) \
-  (__m512i)__builtin_ia32_vpshrdq512((__v8di)(__m512i)(A), \
-                                     (__v8di)(__m512i)(B), (int)(I))
+  ((__m512i)__builtin_ia32_vpshrdq512((__v8di)(__m512i)(A), \
+                                      (__v8di)(__m512i)(B), (int)(I)))
 
 #define _mm512_mask_shrdi_epi64(S, U, A, B, I) \
-  (__m512i)__builtin_ia32_selectq_512((__mmask8)(U), \
-                                    (__v8di)_mm512_shrdi_epi64((A), (B), (I)), \
-                                    (__v8di)(__m512i)(S))
+  ((__m512i)__builtin_ia32_selectq_512((__mmask8)(U), \
+                                     (__v8di)_mm512_shrdi_epi64((A), (B), (I)), \
+                                     (__v8di)(__m512i)(S)))
 
 #define _mm512_maskz_shrdi_epi64(U, A, B, I) \
-  (__m512i)__builtin_ia32_selectq_512((__mmask8)(U), \
-                                    (__v8di)_mm512_shrdi_epi64((A), (B), (I)), \
-                                    (__v8di)_mm512_setzero_si512())
+  ((__m512i)__builtin_ia32_selectq_512((__mmask8)(U), \
+                                     (__v8di)_mm512_shrdi_epi64((A), (B), (I)), \
+                                     (__v8di)_mm512_setzero_si512()))
 
 #define _mm512_shrdi_epi32(A, B, I) \
-  (__m512i)__builtin_ia32_vpshrdd512((__v16si)(__m512i)(A), \
-                                     (__v16si)(__m512i)(B), (int)(I))
+  ((__m512i)__builtin_ia32_vpshrdd512((__v16si)(__m512i)(A), \
+                                      (__v16si)(__m512i)(B), (int)(I)))
 
 #define _mm512_mask_shrdi_epi32(S, U, A, B, I) \
-  (__m512i)__builtin_ia32_selectd_512((__mmask16)(U), \
-                                   (__v16si)_mm512_shrdi_epi32((A), (B), (I)), \
-                                   (__v16si)(__m512i)(S))
+  ((__m512i)__builtin_ia32_selectd_512((__mmask16)(U), \
+                                    (__v16si)_mm512_shrdi_epi32((A), (B), (I)), \
+                                    (__v16si)(__m512i)(S)))
 
 #define _mm512_maskz_shrdi_epi32(U, A, B, I) \
-  (__m512i)__builtin_ia32_selectd_512((__mmask16)(U), \
-                                   (__v16si)_mm512_shrdi_epi32((A), (B), (I)), \
-                                   (__v16si)_mm512_setzero_si512())
+  ((__m512i)__builtin_ia32_selectd_512((__mmask16)(U), \
+                                    (__v16si)_mm512_shrdi_epi32((A), (B), (I)), \
+                                    (__v16si)_mm512_setzero_si512()))
 
 #define _mm512_shrdi_epi16(A, B, I) \
-  (__m512i)__builtin_ia32_vpshrdw512((__v32hi)(__m512i)(A), \
-                                     (__v32hi)(__m512i)(B), (int)(I))
+  ((__m512i)__builtin_ia32_vpshrdw512((__v32hi)(__m512i)(A), \
+                                      (__v32hi)(__m512i)(B), (int)(I)))
 
 #define _mm512_mask_shrdi_epi16(S, U, A, B, I) \
-  (__m512i)__builtin_ia32_selectw_512((__mmask32)(U), \
-                                   (__v32hi)_mm512_shrdi_epi16((A), (B), (I)), \
-                                   (__v32hi)(__m512i)(S))
+  ((__m512i)__builtin_ia32_selectw_512((__mmask32)(U), \
+                                    (__v32hi)_mm512_shrdi_epi16((A), (B), (I)), \
+                                    (__v32hi)(__m512i)(S)))
 
 #define _mm512_maskz_shrdi_epi16(U, A, B, I) \
-  (__m512i)__builtin_ia32_selectw_512((__mmask32)(U), \
-                                   (__v32hi)_mm512_shrdi_epi16((A), (B), (I)), \
-                                   (__v32hi)_mm512_setzero_si512())
+  ((__m512i)__builtin_ia32_selectw_512((__mmask32)(U), \
+                                    (__v32hi)_mm512_shrdi_epi16((A), (B), (I)), \
+                                    (__v32hi)_mm512_setzero_si512()))
 
 static __inline__ __m512i __DEFAULT_FN_ATTRS
 _mm512_shldv_epi64(__m512i __A, __m512i __B, __m512i __C)

package/lib/clang/{13.0.0 → 16}/include/avx512vlbf16intrin.h

@@ -10,11 +10,11 @@
 #error "Never use <avx512vlbf16intrin.h> directly; include <immintrin.h> instead."
 #endif
 
+#ifdef __SSE2__
+
 #ifndef __AVX512VLBF16INTRIN_H
 #define __AVX512VLBF16INTRIN_H
 
-typedef short __m128bh __attribute__((__vector_size__(16), __aligned__(16)));
-
 #define __DEFAULT_FN_ATTRS128 \
   __attribute__((__always_inline__, __nodebug__, \
                  __target__("avx512vl, avx512bf16"), __min_vector_width__(128)))
@@ -59,9 +59,9 @@ _mm_cvtne2ps_pbh(__m128 __A, __m128 __B) {
 ///    conversion of __B, and higher 64 bits come from conversion of __A.
 static __inline__ __m128bh __DEFAULT_FN_ATTRS128
 _mm_mask_cvtne2ps_pbh(__m128bh __W, __mmask8 __U, __m128 __A, __m128 __B) {
-  return (__m128bh)__builtin_ia32_selectw_128((__mmask8)__U,
-                                             (__v8hi)_mm_cvtne2ps_pbh(__A, __B),
-                                             (__v8hi)__W);
+  return (__m128bh)__builtin_ia32_selectpbf_128((__mmask8)__U,
+                                             (__v8bf)_mm_cvtne2ps_pbh(__A, __B),
+                                             (__v8bf)__W);
 }
 
 /// Convert Two Packed Single Data to One Packed BF16 Data.
@@ -81,9 +81,9 @@ _mm_mask_cvtne2ps_pbh(__m128bh __W, __mmask8 __U, __m128 __A, __m128 __B) {
 ///    conversion of __B, and higher 64 bits come from conversion of __A.
 static __inline__ __m128bh __DEFAULT_FN_ATTRS128
 _mm_maskz_cvtne2ps_pbh(__mmask8 __U, __m128 __A, __m128 __B) {
-  return (__m128bh)__builtin_ia32_selectw_128((__mmask8)__U,
-                                             (__v8hi)_mm_cvtne2ps_pbh(__A, __B),
-                                             (__v8hi)_mm_setzero_si128());
+  return (__m128bh)__builtin_ia32_selectpbf_128((__mmask8)__U,
+                                             (__v8bf)_mm_cvtne2ps_pbh(__A, __B),
+                                             (__v8bf)_mm_setzero_si128());
 }
 
 /// Convert Two Packed Single Data to One Packed BF16 Data.
@@ -123,9 +123,9 @@ _mm256_cvtne2ps_pbh(__m256 __A, __m256 __B) {
 ///    conversion of __B, and higher 128 bits come from conversion of __A.
 static __inline__ __m256bh __DEFAULT_FN_ATTRS256
 _mm256_mask_cvtne2ps_pbh(__m256bh __W, __mmask16 __U, __m256 __A, __m256 __B) {
-  return (__m256bh)__builtin_ia32_selectw_256((__mmask16)__U,
-                                         (__v16hi)_mm256_cvtne2ps_pbh(__A, __B),
-                                         (__v16hi)__W);
+  return (__m256bh)__builtin_ia32_selectpbf_256((__mmask16)__U,
+                                         (__v16bf)_mm256_cvtne2ps_pbh(__A, __B),
+                                         (__v16bf)__W);
 }
 
 /// Convert Two Packed Single Data to One Packed BF16 Data.
@@ -145,9 +145,9 @@ _mm256_mask_cvtne2ps_pbh(__m256bh __W, __mmask16 __U, __m256 __A, __m256 __B) {
 ///    conversion of __B, and higher 128 bits come from conversion of __A.
 static __inline__ __m256bh __DEFAULT_FN_ATTRS256
 _mm256_maskz_cvtne2ps_pbh(__mmask16 __U, __m256 __A, __m256 __B) {
-  return (__m256bh)__builtin_ia32_selectw_256((__mmask16)__U,
-                                         (__v16hi)_mm256_cvtne2ps_pbh(__A, __B),
-                                         (__v16hi)_mm256_setzero_si256());
+  return (__m256bh)__builtin_ia32_selectpbf_256((__mmask16)__U,
+                                         (__v16bf)_mm256_cvtne2ps_pbh(__A, __B),
+                                         (__v16bf)_mm256_setzero_si256());
 }
 
 /// Convert Packed Single Data to Packed BF16 Data.
@@ -160,12 +160,8 @@ _mm256_maskz_cvtne2ps_pbh(__mmask16 __U, __m256 __A, __m256 __B) {
 ///    A 128-bit vector of [4 x float].
 /// \returns A 128-bit vector of [8 x bfloat] whose lower 64 bits come from
 ///    conversion of __A, and higher 64 bits are 0.
-static __inline__ __m128bh __DEFAULT_FN_ATTRS128
-_mm_cvtneps_pbh(__m128 __A) {
-  return (__m128bh)__builtin_ia32_cvtneps2bf16_128_mask((__v4sf) __A,
-                                                  (__v8hi)_mm_undefined_si128(),
-                                                  (__mmask8)-1);
-}
+#define _mm_cvtneps_pbh(A)                                                     \
+  ((__m128bh)__builtin_ia32_vcvtneps2bf16128((__v4sf)(A)))
 
 /// Convert Packed Single Data to Packed BF16 Data.
 ///
@@ -185,7 +181,7 @@ _mm_cvtneps_pbh(__m128 __A) {
 static __inline__ __m128bh __DEFAULT_FN_ATTRS128
 _mm_mask_cvtneps_pbh(__m128bh __W, __mmask8 __U, __m128 __A) {
   return (__m128bh)__builtin_ia32_cvtneps2bf16_128_mask((__v4sf) __A,
-                                                        (__v8hi)__W,
+                                                        (__v8bf)__W,
                                                         (__mmask8)__U);
 }
 
@@ -205,7 +201,7 @@ _mm_mask_cvtneps_pbh(__m128bh __W, __mmask8 __U, __m128 __A) {
 static __inline__ __m128bh __DEFAULT_FN_ATTRS128
 _mm_maskz_cvtneps_pbh(__mmask8 __U, __m128 __A) {
   return (__m128bh)__builtin_ia32_cvtneps2bf16_128_mask((__v4sf) __A,
-                                                    (__v8hi)_mm_setzero_si128(),
+                                                    (__v8bf)_mm_setzero_si128(),
                                                     (__mmask8)__U);
 }
 
@@ -218,12 +214,8 @@ _mm_maskz_cvtneps_pbh(__mmask8 __U, __m128 __A) {
 /// \param __A
 ///    A 256-bit vector of [8 x float].
 /// \returns A 128-bit vector of [8 x bfloat] comes from conversion of __A.
-static __inline__ __m128bh __DEFAULT_FN_ATTRS256
-_mm256_cvtneps_pbh(__m256 __A) {
-  return (__m128bh)__builtin_ia32_cvtneps2bf16_256_mask((__v8sf)__A,
-                                                  (__v8hi)_mm_undefined_si128(),
-                                                  (__mmask8)-1);
-}
+#define _mm256_cvtneps_pbh(A)                                                  \
+  ((__m128bh)__builtin_ia32_vcvtneps2bf16256((__v8sf)(A)))
 
 /// Convert Packed Single Data to Packed BF16 Data.
 ///
@@ -242,7 +234,7 @@ _mm256_cvtneps_pbh(__m256 __A) {
 static __inline__ __m128bh __DEFAULT_FN_ATTRS256
 _mm256_mask_cvtneps_pbh(__m128bh __W, __mmask8 __U, __m256 __A) {
   return (__m128bh)__builtin_ia32_cvtneps2bf16_256_mask((__v8sf)__A,
-                                                        (__v8hi)__W,
+                                                        (__v8bf)__W,
                                                         (__mmask8)__U);
 }
 
@@ -261,7 +253,7 @@ _mm256_mask_cvtneps_pbh(__m128bh __W, __mmask8 __U, __m256 __A) {
 static __inline__ __m128bh __DEFAULT_FN_ATTRS256
 _mm256_maskz_cvtneps_pbh(__mmask8 __U, __m256 __A) {
   return (__m128bh)__builtin_ia32_cvtneps2bf16_256_mask((__v8sf)__A,
-                                                    (__v8hi)_mm_setzero_si128(),
+                                                    (__v8bf)_mm_setzero_si128(),
                                                     (__mmask8)__U);
 }
 
@@ -282,8 +274,8 @@ _mm256_maskz_cvtneps_pbh(__mmask8 __U, __m256 __A) {
 static __inline__ __m128 __DEFAULT_FN_ATTRS128
 _mm_dpbf16_ps(__m128 __D, __m128bh __A, __m128bh __B) {
   return (__m128)__builtin_ia32_dpbf16ps_128((__v4sf)__D,
-                                             (__v4si)__A,
-                                             (__v4si)__B);
+                                             (__v8bf)__A,
+                                             (__v8bf)__B);
 }
 
 /// Dot Product of BF16 Pairs Accumulated into Packed Single Precision.
@@ -351,8 +343,8 @@ _mm_maskz_dpbf16_ps(__mmask8 __U, __m128 __D, __m128bh __A, __m128bh __B) {
 static __inline__ __m256 __DEFAULT_FN_ATTRS256
 _mm256_dpbf16_ps(__m256 __D, __m256bh __A, __m256bh __B) {
   return (__m256)__builtin_ia32_dpbf16ps_256((__v8sf)__D,
-                                             (__v8si)__A,
-                                             (__v8si)__B);
+                                             (__v16bf)__A,
+                                             (__v16bf)__B);
 }
 
 /// Dot Product of BF16 Pairs Accumulated into Packed Single Precision.
@@ -413,11 +405,23 @@ _mm256_maskz_dpbf16_ps(__mmask8 __U, __m256 __D, __m256bh __A, __m256bh __B) {
 ///    A float data.
 /// \returns A bf16 data whose sign field and exponent field keep unchanged,
 ///    and fraction field is truncated to 7 bits.
-static __inline__ __bfloat16 __DEFAULT_FN_ATTRS128 _mm_cvtness_sbh(float __A) {
+static __inline__ __bf16 __DEFAULT_FN_ATTRS128 _mm_cvtness_sbh(float __A) {
   __v4sf __V = {__A, 0, 0, 0};
-  __v8hi __R = __builtin_ia32_cvtneps2bf16_128_mask(
-      (__v4sf)__V, (__v8hi)_mm_undefined_si128(), (__mmask8)-1);
-  return __R[0];
+  __v8bf __R = __builtin_ia32_cvtneps2bf16_128_mask(
+      (__v4sf)__V, (__v8bf)_mm_undefined_si128(), (__mmask8)-1);
+  return (__bf16)__R[0];
+}
+
+/// Convert Packed BF16 Data to Packed float Data.
+///
+/// \headerfile <x86intrin.h>
+///
+/// \param __A
+///    A 128-bit vector of [4 x bfloat].
+/// \returns A 128-bit vector of [4 x float] come from conversion of __A
+static __inline__ __m128 __DEFAULT_FN_ATTRS128 _mm_cvtpbh_ps(__m128bh __A) {
+  return _mm_castsi128_ps(
+      (__m128i)_mm_slli_epi32((__m128i)_mm_cvtepi16_epi32((__m128i)__A), 16));
 }
 
 /// Convert Packed BF16 Data to Packed float Data.
@@ -426,12 +430,28 @@ static __inline__ __bfloat16 __DEFAULT_FN_ATTRS128 _mm_cvtness_sbh(float __A) {
 ///
 /// \param __A
 ///    A 128-bit vector of [8 x bfloat].
-/// \returns A 256-bit vector of [8 x float] come from convertion of __A
+/// \returns A 256-bit vector of [8 x float] come from conversion of __A
 static __inline__ __m256 __DEFAULT_FN_ATTRS256 _mm256_cvtpbh_ps(__m128bh __A) {
   return _mm256_castsi256_ps((__m256i)_mm256_slli_epi32(
       (__m256i)_mm256_cvtepi16_epi32((__m128i)__A), 16));
 }
 
+/// Convert Packed BF16 Data to Packed float Data using zeroing mask.
+///
+/// \headerfile <x86intrin.h>
+///
+/// \param __U
+///    A 4-bit mask. Elements are zeroed out when the corresponding mask
+///    bit is not set.
+/// \param __A
+///    A 128-bit vector of [4 x bfloat].
+/// \returns A 128-bit vector of [4 x float] come from conversion of __A
+static __inline__ __m128 __DEFAULT_FN_ATTRS128
+_mm_maskz_cvtpbh_ps(__mmask8 __U, __m128bh __A) {
+  return _mm_castsi128_ps((__m128i)_mm_slli_epi32(
+      (__m128i)_mm_maskz_cvtepi16_epi32((__mmask8)__U, (__m128i)__A), 16));
+}
+
 /// Convert Packed BF16 Data to Packed float Data using zeroing mask.
 ///
 /// \headerfile <x86intrin.h>
@@ -441,13 +461,33 @@ static __inline__ __m256 __DEFAULT_FN_ATTRS256 _mm256_cvtpbh_ps(__m128bh __A) {
 ///    bit is not set.
 /// \param __A
 ///    A 128-bit vector of [8 x bfloat].
-/// \returns A 256-bit vector of [8 x float] come from convertion of __A
+/// \returns A 256-bit vector of [8 x float] come from conversion of __A
 static __inline__ __m256 __DEFAULT_FN_ATTRS256
 _mm256_maskz_cvtpbh_ps(__mmask8 __U, __m128bh __A) {
   return _mm256_castsi256_ps((__m256i)_mm256_slli_epi32(
       (__m256i)_mm256_maskz_cvtepi16_epi32((__mmask8)__U, (__m128i)__A), 16));
 }
 
+/// Convert Packed BF16 Data to Packed float Data using merging mask.
+///
+/// \headerfile <x86intrin.h>
+///
+/// \param __S
+///    A 128-bit vector of [4 x float]. Elements are copied from __S when
+///     the corresponding mask bit is not set.
+/// \param __U
+///    A 4-bit mask. Elements are zeroed out when the corresponding mask
+///    bit is not set.
+/// \param __A
+///    A 128-bit vector of [4 x bfloat].
+/// \returns A 128-bit vector of [4 x float] come from conversion of __A
+static __inline__ __m128 __DEFAULT_FN_ATTRS128
+_mm_mask_cvtpbh_ps(__m128 __S, __mmask8 __U, __m128bh __A) {
+  return _mm_castsi128_ps((__m128i)_mm_mask_slli_epi32(
+      (__m128i)__S, (__mmask8)__U, (__m128i)_mm_cvtepi16_epi32((__m128i)__A),
+      16));
+}
+
 /// Convert Packed BF16 Data to Packed float Data using merging mask.
 ///
 /// \headerfile <x86intrin.h>
@@ -460,7 +500,7 @@ _mm256_maskz_cvtpbh_ps(__mmask8 __U, __m128bh __A) {
 ///    bit is not set.
 /// \param __A
 ///    A 128-bit vector of [8 x bfloat].
-/// \returns A 256-bit vector of [8 x float] come from convertion of __A
+/// \returns A 256-bit vector of [8 x float] come from conversion of __A
 static __inline__ __m256 __DEFAULT_FN_ATTRS256
 _mm256_mask_cvtpbh_ps(__m256 __S, __mmask8 __U, __m128bh __A) {
   return _mm256_castsi256_ps((__m256i)_mm256_mask_slli_epi32(
@@ -472,3 +512,4 @@ _mm256_mask_cvtpbh_ps(__m256 __S, __mmask8 __U, __m128bh __A) {
 #undef __DEFAULT_FN_ATTRS256
 
 #endif
+#endif