@cpp-linter/clang-tidy-win32-x64 19.1.0 → 23.1.3

package/lib/clang/{19 → 23}/include/avx512vlvnniintrin.h

@@ -15,14 +15,25 @@
 #define __AVX512VLVNNIINTRIN_H
 
 /* Define the default attributes for the functions in this file. */
+#if defined(__cplusplus) && (__cplusplus >= 201103L)
 #define __DEFAULT_FN_ATTRS128                                                  \
   __attribute__((__always_inline__, __nodebug__,                               \
-                 __target__("avx512vl,avx512vnni,no-evex512"),                 \
+                 __target__("avx512vl,avx512vnni"),                            \
+                 __min_vector_width__(128))) constexpr
+#define __DEFAULT_FN_ATTRS256                                                  \
+  __attribute__((__always_inline__, __nodebug__,                               \
+                 __target__("avx512vl,avx512vnni"),                            \
+                 __min_vector_width__(256))) constexpr
+#else
+#define __DEFAULT_FN_ATTRS128                                                  \
+  __attribute__((__always_inline__, __nodebug__,                               \
+                 __target__("avx512vl,avx512vnni"),                            \
                  __min_vector_width__(128)))
 #define __DEFAULT_FN_ATTRS256                                                  \
   __attribute__((__always_inline__, __nodebug__,                               \
-                 __target__("avx512vl,avx512vnni,no-evex512"),                 \
+                 __target__("avx512vl,avx512vnni"),                            \
                  __min_vector_width__(256)))
+#endif
 
 /// Multiply groups of 4 adjacent pairs of unsigned 8-bit integers in \a A with
 /// corresponding signed 8-bit integers in \a B, producing 4 intermediate signed
@@ -41,8 +52,8 @@
 ///    ENDFOR
 ///    DST[MAX:256] := 0
 /// \endcode
-#define _mm256_dpbusd_epi32(S, A, B) \
-  ((__m256i)__builtin_ia32_vpdpbusd256((__v8si)(S), (__v8si)(A), (__v8si)(B)))
+#define _mm256_dpbusd_epi32(S, A, B)                                           \
+  ((__m256i)__builtin_ia32_vpdpbusd256((__v8si)(S), (__v32qu)(A), (__v32qi)(B)))
 
 /// Multiply groups of 4 adjacent pairs of unsigned 8-bit integers in \a A with
 /// corresponding signed 8-bit integers in \a B, producing 4 intermediate signed
@@ -61,8 +72,9 @@
 ///    ENDFOR
 ///    DST[MAX:256] := 0
 /// \endcode
-#define _mm256_dpbusds_epi32(S, A, B) \
-  ((__m256i)__builtin_ia32_vpdpbusds256((__v8si)(S), (__v8si)(A), (__v8si)(B)))
+#define _mm256_dpbusds_epi32(S, A, B)                                          \
+  ((__m256i)__builtin_ia32_vpdpbusds256((__v8si)(S), (__v32qu)(A),             \
+                                        (__v32qi)(B)))
 
 /// Multiply groups of 2 adjacent pairs of signed 16-bit integers in \a A with
 /// corresponding 16-bit integers in \a B, producing 2 intermediate signed 32-bit
@@ -79,8 +91,8 @@
 ///    ENDFOR
 ///    DST[MAX:256] := 0
 /// \endcode
-#define _mm256_dpwssd_epi32(S, A, B) \
-  ((__m256i)__builtin_ia32_vpdpwssd256((__v8si)(S), (__v8si)(A), (__v8si)(B)))
+#define _mm256_dpwssd_epi32(S, A, B)                                           \
+  ((__m256i)__builtin_ia32_vpdpwssd256((__v8si)(S), (__v16hi)(A), (__v16hi)(B)))
 
 /// Multiply groups of 2 adjacent pairs of signed 16-bit integers in \a A with
 /// corresponding 16-bit integers in \a B, producing 2 intermediate signed 32-bit
@@ -97,8 +109,9 @@
 ///    ENDFOR
 ///    DST[MAX:256] := 0
 /// \endcode
-#define _mm256_dpwssds_epi32(S, A, B) \
-  ((__m256i)__builtin_ia32_vpdpwssds256((__v8si)(S), (__v8si)(A), (__v8si)(B)))
+#define _mm256_dpwssds_epi32(S, A, B)                                          \
+  ((__m256i)__builtin_ia32_vpdpwssds256((__v8si)(S), (__v16hi)(A),             \
+                                        (__v16hi)(B)))
 
 /// Multiply groups of 4 adjacent pairs of unsigned 8-bit integers in \a A with
 /// corresponding signed 8-bit integers in \a B, producing 4 intermediate signed
@@ -117,8 +130,8 @@
 ///    ENDFOR
 ///    DST[MAX:128] := 0
 /// \endcode
-#define _mm_dpbusd_epi32(S, A, B) \
-  ((__m128i)__builtin_ia32_vpdpbusd128((__v4si)(S), (__v4si)(A), (__v4si)(B)))
+#define _mm_dpbusd_epi32(S, A, B)                                              \
+  ((__m128i)__builtin_ia32_vpdpbusd128((__v4si)(S), (__v16qu)(A), (__v16qi)(B)))
 
 /// Multiply groups of 4 adjacent pairs of unsigned 8-bit integers in \a A with
 /// corresponding signed 8-bit integers in \a B, producing 4 intermediate signed
@@ -137,8 +150,9 @@
 ///    ENDFOR
 ///    DST[MAX:128] := 0
 /// \endcode
-#define _mm_dpbusds_epi32(S, A, B) \
-  ((__m128i)__builtin_ia32_vpdpbusds128((__v4si)(S), (__v4si)(A), (__v4si)(B)))
+#define _mm_dpbusds_epi32(S, A, B)                                             \
+  ((__m128i)__builtin_ia32_vpdpbusds128((__v4si)(S), (__v16qu)(A),             \
+                                        (__v16qi)(B)))
 
 /// Multiply groups of 2 adjacent pairs of signed 16-bit integers in \a A with
 /// corresponding 16-bit integers in \a B, producing 2 intermediate signed 32-bit
@@ -155,8 +169,8 @@
 ///    ENDFOR
 ///    DST[MAX:128] := 0
 /// \endcode
-#define _mm_dpwssd_epi32(S, A, B) \
-  ((__m128i)__builtin_ia32_vpdpwssd128((__v4si)(S), (__v4si)(A), (__v4si)(B)))
+#define _mm_dpwssd_epi32(S, A, B)                                              \
+  ((__m128i)__builtin_ia32_vpdpwssd128((__v4si)(S), (__v8hi)(A), (__v8hi)(B)))
 
 /// Multiply groups of 2 adjacent pairs of signed 16-bit integers in \a A with
 /// corresponding 16-bit integers in \a B, producing 2 intermediate signed 32-bit
@@ -173,132 +187,116 @@
 ///    ENDFOR
 ///    DST[MAX:128] := 0
 /// \endcode
-#define _mm_dpwssds_epi32(S, A, B) \
-  ((__m128i)__builtin_ia32_vpdpwssds128((__v4si)(S), (__v4si)(A), (__v4si)(B)))
+#define _mm_dpwssds_epi32(S, A, B)                                             \
+  ((__m128i)__builtin_ia32_vpdpwssds128((__v4si)(S), (__v8hi)(A), (__v8hi)(B)))
 
 static __inline__ __m256i __DEFAULT_FN_ATTRS256
-_mm256_mask_dpbusd_epi32(__m256i __S, __mmask8 __U, __m256i __A, __m256i __B)
-{
+_mm256_mask_dpbusd_epi32(__m256i __S, __mmask8 __U, __m256i __A, __m256i __B) {
   return (__m256i)__builtin_ia32_selectd_256(__U,
                                      (__v8si)_mm256_dpbusd_epi32(__S, __A, __B),
                                      (__v8si)__S);
 }
 
 static __inline__ __m256i __DEFAULT_FN_ATTRS256
-_mm256_maskz_dpbusd_epi32(__mmask8 __U, __m256i __S, __m256i __A, __m256i __B)
-{
+_mm256_maskz_dpbusd_epi32(__mmask8 __U, __m256i __S, __m256i __A, __m256i __B) {
   return (__m256i)__builtin_ia32_selectd_256(__U,
                                      (__v8si)_mm256_dpbusd_epi32(__S, __A, __B),
                                      (__v8si)_mm256_setzero_si256());
 }
 
 static __inline__ __m256i __DEFAULT_FN_ATTRS256
-_mm256_mask_dpbusds_epi32(__m256i __S, __mmask8 __U, __m256i __A, __m256i __B)
-{
+_mm256_mask_dpbusds_epi32(__m256i __S, __mmask8 __U, __m256i __A, __m256i __B) {
   return (__m256i)__builtin_ia32_selectd_256(__U,
                                     (__v8si)_mm256_dpbusds_epi32(__S, __A, __B),
                                     (__v8si)__S);
 }
 
-static __inline__ __m256i __DEFAULT_FN_ATTRS256
-_mm256_maskz_dpbusds_epi32(__mmask8 __U, __m256i __S, __m256i __A, __m256i __B)
-{
+static __inline__ __m256i __DEFAULT_FN_ATTRS256 _mm256_maskz_dpbusds_epi32(
+    __mmask8 __U, __m256i __S, __m256i __A, __m256i __B) {
   return (__m256i)__builtin_ia32_selectd_256(__U,
                                      (__v8si)_mm256_dpbusds_epi32(__S, __A, __B),
                                      (__v8si)_mm256_setzero_si256());
 }
 
 static __inline__ __m256i __DEFAULT_FN_ATTRS256
-_mm256_mask_dpwssd_epi32(__m256i __S, __mmask8 __U, __m256i __A, __m256i __B)
-{
+_mm256_mask_dpwssd_epi32(__m256i __S, __mmask8 __U, __m256i __A, __m256i __B) {
   return (__m256i)__builtin_ia32_selectd_256(__U,
                                      (__v8si)_mm256_dpwssd_epi32(__S, __A, __B),
                                      (__v8si)__S);
 }
 
 static __inline__ __m256i __DEFAULT_FN_ATTRS256
-_mm256_maskz_dpwssd_epi32(__mmask8 __U, __m256i __S, __m256i __A, __m256i __B)
-{
+_mm256_maskz_dpwssd_epi32(__mmask8 __U, __m256i __S, __m256i __A, __m256i __B) {
   return (__m256i)__builtin_ia32_selectd_256(__U,
                                      (__v8si)_mm256_dpwssd_epi32(__S, __A, __B),
                                      (__v8si)_mm256_setzero_si256());
 }
 
 static __inline__ __m256i __DEFAULT_FN_ATTRS256
-_mm256_mask_dpwssds_epi32(__m256i __S, __mmask8 __U, __m256i __A, __m256i __B)
-{
+_mm256_mask_dpwssds_epi32(__m256i __S, __mmask8 __U, __m256i __A, __m256i __B) {
   return (__m256i)__builtin_ia32_selectd_256(__U,
                                     (__v8si)_mm256_dpwssds_epi32(__S, __A, __B),
                                     (__v8si)__S);
 }
 
-static __inline__ __m256i __DEFAULT_FN_ATTRS256
-_mm256_maskz_dpwssds_epi32(__mmask8 __U, __m256i __S, __m256i __A, __m256i __B)
-{
+static __inline__ __m256i __DEFAULT_FN_ATTRS256 _mm256_maskz_dpwssds_epi32(
+    __mmask8 __U, __m256i __S, __m256i __A, __m256i __B) {
   return (__m256i)__builtin_ia32_selectd_256(__U,
                                     (__v8si)_mm256_dpwssds_epi32(__S, __A, __B),
                                     (__v8si)_mm256_setzero_si256());
 }
 
 static __inline__ __m128i __DEFAULT_FN_ATTRS128
-_mm_mask_dpbusd_epi32(__m128i __S, __mmask8 __U, __m128i __A, __m128i __B)
-{
+_mm_mask_dpbusd_epi32(__m128i __S, __mmask8 __U, __m128i __A, __m128i __B) {
   return (__m128i)__builtin_ia32_selectd_128(__U,
                                         (__v4si)_mm_dpbusd_epi32(__S, __A, __B),
                                         (__v4si)__S);
 }
 
 static __inline__ __m128i __DEFAULT_FN_ATTRS128
-_mm_maskz_dpbusd_epi32(__mmask8 __U, __m128i __S, __m128i __A, __m128i __B)
-{
+_mm_maskz_dpbusd_epi32(__mmask8 __U, __m128i __S, __m128i __A, __m128i __B) {
   return (__m128i)__builtin_ia32_selectd_128(__U,
                                         (__v4si)_mm_dpbusd_epi32(__S, __A, __B),
                                         (__v4si)_mm_setzero_si128());
 }
 
 static __inline__ __m128i __DEFAULT_FN_ATTRS128
-_mm_mask_dpbusds_epi32(__m128i __S, __mmask8 __U, __m128i __A, __m128i __B)
-{
+_mm_mask_dpbusds_epi32(__m128i __S, __mmask8 __U, __m128i __A, __m128i __B) {
   return (__m128i)__builtin_ia32_selectd_128(__U,
                                        (__v4si)_mm_dpbusds_epi32(__S, __A, __B),
                                        (__v4si)__S);
 }
 
 static __inline__ __m128i __DEFAULT_FN_ATTRS128
-_mm_maskz_dpbusds_epi32(__mmask8 __U, __m128i __S, __m128i __A, __m128i __B)
-{
+_mm_maskz_dpbusds_epi32(__mmask8 __U, __m128i __S, __m128i __A, __m128i __B) {
   return (__m128i)__builtin_ia32_selectd_128(__U,
                                        (__v4si)_mm_dpbusds_epi32(__S, __A, __B),
                                        (__v4si)_mm_setzero_si128());
 }
 
 static __inline__ __m128i __DEFAULT_FN_ATTRS128
-_mm_mask_dpwssd_epi32(__m128i __S, __mmask8 __U, __m128i __A, __m128i __B)
-{
+_mm_mask_dpwssd_epi32(__m128i __S, __mmask8 __U, __m128i __A, __m128i __B) {
   return (__m128i)__builtin_ia32_selectd_128(__U,
                                         (__v4si)_mm_dpwssd_epi32(__S, __A, __B),
                                         (__v4si)__S);
 }
 
 static __inline__ __m128i __DEFAULT_FN_ATTRS128
-_mm_maskz_dpwssd_epi32(__mmask8 __U, __m128i __S, __m128i __A, __m128i __B)
-{
+_mm_maskz_dpwssd_epi32(__mmask8 __U, __m128i __S, __m128i __A, __m128i __B) {
   return (__m128i)__builtin_ia32_selectd_128(__U,
                                         (__v4si)_mm_dpwssd_epi32(__S, __A, __B),
                                         (__v4si)_mm_setzero_si128());
 }
 
 static __inline__ __m128i __DEFAULT_FN_ATTRS128
-_mm_mask_dpwssds_epi32(__m128i __S, __mmask8 __U, __m128i __A, __m128i __B)
-{
+_mm_mask_dpwssds_epi32(__m128i __S, __mmask8 __U, __m128i __A, __m128i __B) {
   return (__m128i)__builtin_ia32_selectd_128(__U,
                                        (__v4si)_mm_dpwssds_epi32(__S, __A, __B),
                                        (__v4si)__S);
 }
 
 static __inline__ __m128i __DEFAULT_FN_ATTRS128
-_mm_maskz_dpwssds_epi32(__mmask8 __U, __m128i __S, __m128i __A, __m128i __B)
-{
+_mm_maskz_dpwssds_epi32(__mmask8 __U, __m128i __S, __m128i __A, __m128i __B) {
   return (__m128i)__builtin_ia32_selectd_128(__U,
                                        (__v4si)_mm_dpwssds_epi32(__S, __A, __B),
                                        (__v4si)_mm_setzero_si128());

package/lib/clang/{19 → 23}/include/avx512vlvp2intersectintrin.h

@@ -30,12 +30,12 @@
 
 #define __DEFAULT_FN_ATTRS128                                                  \
   __attribute__((__always_inline__, __nodebug__,                               \
-                 __target__("avx512vl,avx512vp2intersect,no-evex512"),         \
+                 __target__("avx512vl,avx512vp2intersect"),                    \
                  __min_vector_width__(128)))
 
 #define __DEFAULT_FN_ATTRS256                                                  \
   __attribute__((__always_inline__, __nodebug__,                               \
-                 __target__("avx512vl,avx512vp2intersect,no-evex512"),         \
+                 __target__("avx512vl,avx512vp2intersect"),                    \
                  __min_vector_width__(256)))
 /// Store, in an even/odd pair of mask registers, the indicators of the
 /// locations of value matches between dwords in operands __a and __b.

package/lib/clang/{19 → 23}/include/avx512vnniintrin.h

@@ -15,102 +15,99 @@
 #define __AVX512VNNIINTRIN_H
 
 /* Define the default attributes for the functions in this file. */
+#if defined(__cplusplus) && (__cplusplus >= 201103L)
 #define __DEFAULT_FN_ATTRS                                                     \
-  __attribute__((__always_inline__, __nodebug__,                               \
-                 __target__("avx512vnni,evex512"), __min_vector_width__(512)))
+  __attribute__((__always_inline__, __nodebug__, __target__("avx512vnni"),     \
+                 __min_vector_width__(512))) constexpr
+#else
+#define __DEFAULT_FN_ATTRS                                                     \
+  __attribute__((__always_inline__, __nodebug__, __target__("avx512vnni"),     \
+                 __min_vector_width__(512)))
+#endif
 
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_dpbusd_epi32(__m512i __S, __m512i __A, __m512i __B)
-{
-  return (__m512i)__builtin_ia32_vpdpbusd512((__v16si)__S, (__v16si)__A,
-                                             (__v16si)__B);
+static __inline__ __m512i __DEFAULT_FN_ATTRS _mm512_dpbusd_epi32(__m512i __S,
+                                                                 __m512i __A,
+                                                                 __m512i __B) {
+  return (__m512i)__builtin_ia32_vpdpbusd512((__v16si)__S, (__v64qu)__A,
+                                             (__v64qi)__B);
 }
 
 static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_mask_dpbusd_epi32(__m512i __S, __mmask16 __U, __m512i __A, __m512i __B)
-{
+_mm512_mask_dpbusd_epi32(__m512i __S, __mmask16 __U, __m512i __A, __m512i __B) {
   return (__m512i)__builtin_ia32_selectd_512(__U,
                                     (__v16si)_mm512_dpbusd_epi32(__S, __A, __B),
                                     (__v16si)__S);
 }
 
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_maskz_dpbusd_epi32(__mmask16 __U, __m512i __S, __m512i __A, __m512i __B)
-{
+static __inline__ __m512i __DEFAULT_FN_ATTRS _mm512_maskz_dpbusd_epi32(
+    __mmask16 __U, __m512i __S, __m512i __A, __m512i __B) {
   return (__m512i)__builtin_ia32_selectd_512(__U,
                                     (__v16si)_mm512_dpbusd_epi32(__S, __A, __B),
                                     (__v16si)_mm512_setzero_si512());
 }
 
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_dpbusds_epi32(__m512i __S, __m512i __A, __m512i __B)
-{
-  return (__m512i)__builtin_ia32_vpdpbusds512((__v16si)__S, (__v16si)__A,
-                                              (__v16si)__B);
+static __inline__ __m512i __DEFAULT_FN_ATTRS _mm512_dpbusds_epi32(__m512i __S,
+                                                                  __m512i __A,
+                                                                  __m512i __B) {
+  return (__m512i)__builtin_ia32_vpdpbusds512((__v16si)__S, (__v64qu)__A,
+                                              (__v64qi)__B);
 }
 
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_mask_dpbusds_epi32(__m512i __S, __mmask16 __U, __m512i __A, __m512i __B)
-{
+static __inline__ __m512i __DEFAULT_FN_ATTRS _mm512_mask_dpbusds_epi32(
+    __m512i __S, __mmask16 __U, __m512i __A, __m512i __B) {
   return (__m512i)__builtin_ia32_selectd_512(__U,
                                    (__v16si)_mm512_dpbusds_epi32(__S, __A, __B),
                                    (__v16si)__S);
 }
 
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_maskz_dpbusds_epi32(__mmask16 __U, __m512i __S, __m512i __A, __m512i __B)
-{
+static __inline__ __m512i __DEFAULT_FN_ATTRS _mm512_maskz_dpbusds_epi32(
+    __mmask16 __U, __m512i __S, __m512i __A, __m512i __B) {
   return (__m512i)__builtin_ia32_selectd_512(__U,
                                    (__v16si)_mm512_dpbusds_epi32(__S, __A, __B),
                                    (__v16si)_mm512_setzero_si512());
 }
 
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_dpwssd_epi32(__m512i __S, __m512i __A, __m512i __B)
-{
-  return (__m512i)__builtin_ia32_vpdpwssd512((__v16si)__S, (__v16si)__A,
-                                             (__v16si)__B);
+static __inline__ __m512i __DEFAULT_FN_ATTRS _mm512_dpwssd_epi32(__m512i __S,
+                                                                 __m512i __A,
+                                                                 __m512i __B) {
+  return (__m512i)__builtin_ia32_vpdpwssd512((__v16si)__S, (__v32hi)__A,
+                                             (__v32hi)__B);
 }
 
 static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_mask_dpwssd_epi32(__m512i __S, __mmask16 __U, __m512i __A, __m512i __B)
-{
+_mm512_mask_dpwssd_epi32(__m512i __S, __mmask16 __U, __m512i __A, __m512i __B) {
   return (__m512i)__builtin_ia32_selectd_512(__U,
                                     (__v16si)_mm512_dpwssd_epi32(__S, __A, __B),
                                     (__v16si)__S);
 }
 
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_maskz_dpwssd_epi32(__mmask16 __U, __m512i __S, __m512i __A, __m512i __B)
-{
+static __inline__ __m512i __DEFAULT_FN_ATTRS _mm512_maskz_dpwssd_epi32(
+    __mmask16 __U, __m512i __S, __m512i __A, __m512i __B) {
   return (__m512i)__builtin_ia32_selectd_512(__U,
                                     (__v16si)_mm512_dpwssd_epi32(__S, __A, __B),
                                     (__v16si)_mm512_setzero_si512());
 }
 
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_dpwssds_epi32(__m512i __S, __m512i __A, __m512i __B)
-{
-  return (__m512i)__builtin_ia32_vpdpwssds512((__v16si)__S, (__v16si)__A,
-                                              (__v16si)__B);
+static __inline__ __m512i __DEFAULT_FN_ATTRS _mm512_dpwssds_epi32(__m512i __S,
+                                                                  __m512i __A,
+                                                                  __m512i __B) {
+  return (__m512i)__builtin_ia32_vpdpwssds512((__v16si)__S, (__v32hi)__A,
+                                              (__v32hi)__B);
 }
 
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_mask_dpwssds_epi32(__m512i __S, __mmask16 __U, __m512i __A, __m512i __B)
-{
+static __inline__ __m512i __DEFAULT_FN_ATTRS _mm512_mask_dpwssds_epi32(
+    __m512i __S, __mmask16 __U, __m512i __A, __m512i __B) {
   return (__m512i)__builtin_ia32_selectd_512(__U,
                                    (__v16si)_mm512_dpwssds_epi32(__S, __A, __B),
                                    (__v16si)__S);
 }
 
-static __inline__ __m512i __DEFAULT_FN_ATTRS
-_mm512_maskz_dpwssds_epi32(__mmask16 __U, __m512i __S, __m512i __A, __m512i __B)
-{
+static __inline__ __m512i __DEFAULT_FN_ATTRS _mm512_maskz_dpwssds_epi32(
+    __mmask16 __U, __m512i __S, __m512i __A, __m512i __B) {
   return (__m512i)__builtin_ia32_selectd_512(__U,
                                    (__v16si)_mm512_dpwssds_epi32(__S, __A, __B),
                                    (__v16si)_mm512_setzero_si512());
 }
 
 #undef __DEFAULT_FN_ATTRS
-
 #endif

package/lib/clang/{19 → 23}/include/avx512vp2intersectintrin.h

@@ -22,7 +22,7 @@
  *===-----------------------------------------------------------------------===
  */
 #ifndef __IMMINTRIN_H
-#error "Never use <avx512vp2intersect.h> directly; include <immintrin.h> instead."
+#error "Never use <avx512vp2intersectintrin.h> directly; include <immintrin.h> instead."
 #endif
 
 #ifndef _AVX512VP2INTERSECT_H
@@ -30,8 +30,7 @@
 
 #define __DEFAULT_FN_ATTRS                                                     \
   __attribute__((__always_inline__, __nodebug__,                               \
-                 __target__("avx512vp2intersect,evex512"),                     \
-                 __min_vector_width__(512)))
+                 __target__("avx512vp2intersect"), __min_vector_width__(512)))
 
 /// Store, in an even/odd pair of mask registers, the indicators of the
 /// locations of value matches between dwords in operands __a and __b.

package/lib/clang/{19 → 23}/include/avx512vpopcntdqintrin.h

@@ -16,13 +16,19 @@
 #define __AVX512VPOPCNTDQINTRIN_H
 
 /* Define the default attributes for the functions in this file. */
+#if defined(__cplusplus) && (__cplusplus >= 201103L)
 #define __DEFAULT_FN_ATTRS                                                     \
   __attribute__((__always_inline__, __nodebug__,                               \
-                 __target__("avx512vpopcntdq,evex512"),                        \
-                 __min_vector_width__(512)))
+                 __target__("avx512vpopcntdq"),                                \
+                 __min_vector_width__(512))) constexpr
+#else
+#define __DEFAULT_FN_ATTRS                                                     \
+  __attribute__((__always_inline__, __nodebug__,                               \
+                 __target__("avx512vpopcntdq"), __min_vector_width__(512)))
+#endif
 
 static __inline__ __m512i __DEFAULT_FN_ATTRS _mm512_popcnt_epi64(__m512i __A) {
-  return (__m512i)__builtin_ia32_vpopcntq_512((__v8di)__A);
+  return (__m512i)__builtin_elementwise_popcount((__v8du)__A);
 }
 
 static __inline__ __m512i __DEFAULT_FN_ATTRS
@@ -37,7 +43,7 @@ _mm512_maskz_popcnt_epi64(__mmask8 __U, __m512i __A) {
 }
 
 static __inline__ __m512i __DEFAULT_FN_ATTRS _mm512_popcnt_epi32(__m512i __A) {
-  return (__m512i)__builtin_ia32_vpopcntd_512((__v16si)__A);
+  return (__m512i)__builtin_elementwise_popcount((__v16su)__A);
 }
 
 static __inline__ __m512i __DEFAULT_FN_ATTRS

package/lib/clang/{19 → 23}/include/avx512vpopcntdqvlintrin.h

@@ -16,18 +16,29 @@
 #define __AVX512VPOPCNTDQVLINTRIN_H
 
 /* Define the default attributes for the functions in this file. */
+
+#if defined(__cplusplus) && (__cplusplus >= 201103L)
+#define __DEFAULT_FN_ATTRS128                                                  \
+  __attribute__((__always_inline__, __nodebug__,                               \
+                 __target__("avx512vpopcntdq,avx512vl"),                       \
+                 __min_vector_width__(128))) constexpr
+#define __DEFAULT_FN_ATTRS256                                                  \
+  __attribute__((__always_inline__, __nodebug__,                               \
+                 __target__("avx512vpopcntdq,avx512vl"),                       \
+                 __min_vector_width__(256))) constexpr
+#else
 #define __DEFAULT_FN_ATTRS128                                                  \
   __attribute__((__always_inline__, __nodebug__,                               \
-                 __target__("avx512vpopcntdq,avx512vl,no-evex512"),            \
+                 __target__("avx512vpopcntdq,avx512vl"),                       \
                  __min_vector_width__(128)))
 #define __DEFAULT_FN_ATTRS256                                                  \
   __attribute__((__always_inline__, __nodebug__,                               \
-                 __target__("avx512vpopcntdq,avx512vl,no-evex512"),            \
+                 __target__("avx512vpopcntdq,avx512vl"),                       \
                  __min_vector_width__(256)))
+#endif
 
-static __inline__ __m128i __DEFAULT_FN_ATTRS128
-_mm_popcnt_epi64(__m128i __A) {
-  return (__m128i)__builtin_ia32_vpopcntq_128((__v2di)__A);
+static __inline__ __m128i __DEFAULT_FN_ATTRS128 _mm_popcnt_epi64(__m128i __A) {
+  return (__m128i)__builtin_elementwise_popcount((__v2du)__A);
 }
 
 static __inline__ __m128i __DEFAULT_FN_ATTRS128
@@ -41,9 +52,8 @@ _mm_maskz_popcnt_epi64(__mmask8 __U, __m128i __A) {
   return _mm_mask_popcnt_epi64((__m128i)_mm_setzero_si128(), __U, __A);
 }
 
-static __inline__ __m128i __DEFAULT_FN_ATTRS128
-_mm_popcnt_epi32(__m128i __A) {
-  return (__m128i)__builtin_ia32_vpopcntd_128((__v4si)__A);
+static __inline__ __m128i __DEFAULT_FN_ATTRS128 _mm_popcnt_epi32(__m128i __A) {
+  return (__m128i)__builtin_elementwise_popcount((__v4su)__A);
 }
 
 static __inline__ __m128i __DEFAULT_FN_ATTRS128
@@ -59,7 +69,7 @@ _mm_maskz_popcnt_epi32(__mmask8 __U, __m128i __A) {
 
 static __inline__ __m256i __DEFAULT_FN_ATTRS256
 _mm256_popcnt_epi64(__m256i __A) {
-  return (__m256i)__builtin_ia32_vpopcntq_256((__v4di)__A);
+  return (__m256i)__builtin_elementwise_popcount((__v4du)__A);
 }
 
 static __inline__ __m256i __DEFAULT_FN_ATTRS256
@@ -75,7 +85,7 @@ _mm256_maskz_popcnt_epi64(__mmask8 __U, __m256i __A) {
 
 static __inline__ __m256i __DEFAULT_FN_ATTRS256
 _mm256_popcnt_epi32(__m256i __A) {
-  return (__m256i)__builtin_ia32_vpopcntd_256((__v8si)__A);
+  return (__m256i)__builtin_elementwise_popcount((__v8su)__A);
 }
 
 static __inline__ __m256i __DEFAULT_FN_ATTRS256

package/lib/clang/{19 → 23}/include/avxifmaintrin.h

@@ -15,12 +15,28 @@
 #define __AVXIFMAINTRIN_H
 
 /* Define the default attributes for the functions in this file. */
+#if defined(__cplusplus) && (__cplusplus >= 201103L)
+#define __DEFAULT_FN_ATTRS128                                                  \
+  __attribute__((__always_inline__, __nodebug__, __target__("avxifma"),        \
+                 __min_vector_width__(128))) constexpr
+#define __DEFAULT_FN_ATTRS256                                                  \
+  __attribute__((__always_inline__, __nodebug__, __target__("avxifma"),        \
+                 __min_vector_width__(256))) constexpr
+#else
 #define __DEFAULT_FN_ATTRS128                                                  \
   __attribute__((__always_inline__, __nodebug__, __target__("avxifma"),        \
                  __min_vector_width__(128)))
 #define __DEFAULT_FN_ATTRS256                                                  \
   __attribute__((__always_inline__, __nodebug__, __target__("avxifma"),        \
                  __min_vector_width__(256)))
+#endif
+
+#if !defined(__AVX512IFMA__) && defined(__AVXIFMA__)
+#define _mm_madd52hi_epu64(X, Y, Z) _mm_madd52hi_avx_epu64(X, Y, Z)
+#define _mm_madd52lo_epu64(X, Y, Z) _mm_madd52lo_avx_epu64(X, Y, Z)
+#define _mm256_madd52hi_epu64(X, Y, Z) _mm256_madd52hi_avx_epu64(X, Y, Z)
+#define _mm256_madd52lo_epu64(X, Y, Z) _mm256_madd52lo_avx_epu64(X, Y, Z)
+#endif
 
 // must vex-encoding