Roll the new ASan binaries for Mac OS

asan/asan_clang_Darwin/lib/clang/3.1/include/avxintrin.h

 /*===---- avxintrin.h - AVX intrinsics -------------------------------------===
  *
  * Permission is hereby granted, free of charge, to any person obtaining a copy
  * of this software and associated documentation files (the "Software"), to deal
  * in the Software without restriction, including without limitation the rights
  * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
  * copies of the Software, and to permit persons to whom the Software is
  * furnished to do so, subject to the following conditions:
  *
  * The above copyright notice and this permission notice shall be included in
@@ skipping 127 matching lines @@
 {
   return (__m256)__builtin_ia32_rsqrtps256((__v8sf)a);
 }
 
 static __inline __m256 __attribute__((__always_inline__, __nodebug__))
 _mm256_rcp_ps(__m256 a)
 {
   return (__m256)__builtin_ia32_rcpps256((__v8sf)a);
 }
 
-static __inline __m256d __attribute__((__always_inline__, __nodebug__))
-_mm256_round_pd(__m256d v, const int m)
-{
-  return (__m256d)__builtin_ia32_roundpd256((__v4df)v, m);
-}
+#define _mm256_round_pd(V, M) __extension__ ({ \
+    __m256d __V = (V); \
+    (__m256d)__builtin_ia32_roundpd256((__v4df)__V, (M)); })
 
-static __inline __m256 __attribute__((__always_inline__, __nodebug__))
-_mm256_round_ps(__m256 v, const int m)
-{
-  return (__m256)__builtin_ia32_roundps256((__v8sf)v, m);
-}
+#define _mm256_round_ps(V, M) __extension__ ({ \
+  __m256 __V = (V); \
+  (__m256)__builtin_ia32_roundps256((__v8sf)__V, (M)); })
 
 #define _mm256_ceil_pd(V)  _mm256_round_pd((V), _MM_FROUND_CEIL)
 #define _mm256_floor_pd(V) _mm256_round_pd((V), _MM_FROUND_FLOOR)
 #define _mm256_ceil_ps(V)  _mm256_round_ps((V), _MM_FROUND_CEIL)
 #define _mm256_floor_ps(V) _mm256_round_ps((V), _MM_FROUND_FLOOR)
 
 /* Logical */
 static __inline __m256d __attribute__((__always_inline__, __nodebug__))
 _mm256_and_pd(__m256d a, __m256d b)
 {
@@ skipping 86 matching lines @@
   return (__m128)__builtin_ia32_vpermilvarps((__v4sf)a, (__v4si)c);
 }
 
 static __inline __m256 __attribute__((__always_inline__, __nodebug__))
 _mm256_permutevar_ps(__m256 a, __m256i c)
 {
   return (__m256)__builtin_ia32_vpermilvarps256((__v8sf)a,
                                                   (__v8si)c);
 }
 
-static __inline __m128d __attribute__((__always_inline__, __nodebug__))
-_mm_permute_pd(__m128d a, const int c)
-{
-  return (__m128d)__builtin_ia32_vpermilpd((__v2df)a, c);
-}
+#define _mm_permute_pd(A, C) __extension__ ({ \
+  __m128d __A = (A); \
+  (__m128d)__builtin_ia32_vpermilpd((__v2df)__A, (C)); })
 
-static __inline __m256d __attribute__((__always_inline__, __nodebug__))
-_mm256_permute_pd(__m256d a, const int c)
-{
-  return (__m256d)__builtin_ia32_vpermilpd256((__v4df)a, c);
-}
+#define _mm256_permute_pd(A, C) __extension__ ({ \
+  __m256d __A = (A); \
+  (__m256d)__builtin_ia32_vpermilpd256((__v4df)__A, (C)); })
 
-static __inline __m128 __attribute__((__always_inline__, __nodebug__))
-_mm_permute_ps(__m128 a, const int c)
-{
-  return (__m128)__builtin_ia32_vpermilps((__v4sf)a, c);
-}
+#define _mm_permute_ps(A, C) __extension__ ({ \
+  __m128 __A = (A); \
+  (__m128)__builtin_ia32_vpermilps((__v4sf)__A, (C)); })
 
-static __inline __m256 __attribute__((__always_inline__, __nodebug__))
-_mm256_permute_ps(__m256 a, const int c)
-{
-  return (__m256)__builtin_ia32_vpermilps256((__v8sf)a, c);
-}
+#define _mm256_permute_ps(A, C) __extension__ ({ \
+  __m256 __A = (A); \
+  (__m256)__builtin_ia32_vpermilps256((__v8sf)__A, (C)); })
 
-static __inline __m256d __attribute__((__always_inline__, __nodebug__))
-_mm256_permute2f128_pd(__m256d a, __m256d b, const int c)
-{
-  return (__m256d)__builtin_ia32_vperm2f128_pd256((__v4df)a, (__v4df)b, c);
-}
+#define _mm256_permute2f128_pd(V1, V2, M) __extension__ ({ \
+  __m256d __V1 = (V1); \
+  __m256d __V2 = (V2); \
+  (__m256d)__builtin_ia32_vperm2f128_pd256((__v4df)__V1, (__v4df)__V2, (M)); })
 
-static __inline __m256 __attribute__((__always_inline__, __nodebug__))
-_mm256_permute2f128_ps(__m256 a, __m256 b, const int c)
-{
-  return (__m256)__builtin_ia32_vperm2f128_ps256((__v8sf)a, (__v8sf)b, c);
-}
+#define _mm256_permute2f128_ps(V1, V2, M) __extension__ ({ \
+  __m256 __V1 = (V1); \
+  __m256 __V2 = (V2); \
+  (__m256)__builtin_ia32_vperm2f128_ps256((__v8sf)__V1, (__v8sf)__V2, (M)); })
 
-static __inline __m256i __attribute__((__always_inline__, __nodebug__))
-_mm256_permute2f128_si256(__m256i a, __m256i b, const int c)
-{
-  return (__m256i)__builtin_ia32_vperm2f128_si256((__v8si)a, (__v8si)b, c);
-}
+#define _mm256_permute2f128_si256(V1, V2, M) __extension__ ({ \
+  __m256i __V1 = (V1); \
+  __m256i __V2 = (V2); \
+  (__m256i)__builtin_ia32_vperm2f128_si256((__v8si)__V1, (__v8si)__V2, (M)); })
 
 /* Vector Blend */
 #define _mm256_blend_pd(V1, V2, M) __extension__ ({ \
   __m256d __V1 = (V1); \
   __m256d __V2 = (V2); \
-  (__m256d)__builtin_ia32_blendpd256((__v4df)__V1, (__v4df)__V2, M); })
+  (__m256d)__builtin_ia32_blendpd256((__v4df)__V1, (__v4df)__V2, (M)); })
 
 #define _mm256_blend_ps(V1, V2, M) __extension__ ({ \
   __m256 __V1 = (V1); \
   __m256 __V2 = (V2); \
-  (__m256)__builtin_ia32_blendps256((__v8sf)__V1, (__v8sf)__V2, M); })
+  (__m256)__builtin_ia32_blendps256((__v8sf)__V1, (__v8sf)__V2, (M)); })
 
 static __inline __m256d __attribute__((__always_inline__, __nodebug__))
 _mm256_blendv_pd(__m256d a, __m256d b, __m256d c)
 {
   return (__m256d)__builtin_ia32_blendvpd256((__v4df)a, (__v4df)b, (__v4df)c);
 }
 
 static __inline __m256 __attribute__((__always_inline__, __nodebug__))
 _mm256_blendv_ps(__m256 a, __m256 b, __m256 c)
 {
   return (__m256)__builtin_ia32_blendvps256((__v8sf)a, (__v8sf)b, (__v8sf)c);
 }
 
 /* Vector Dot Product */
 #define _mm256_dp_ps(V1, V2, M) __extension__ ({ \
   __m256 __V1 = (V1); \
   __m256 __V2 = (V2); \
-  (__m256)__builtin_ia32_dpps256((__v8sf)__V1, (__v8sf)__V2, M); })
+  (__m256)__builtin_ia32_dpps256((__v8sf)__V1, (__v8sf)__V2, (M)); })
 
 /* Vector shuffle */
 #define _mm256_shuffle_ps(a, b, mask) __extension__ ({ \
         __m256 __a = (a); \
         __m256 __b = (b); \
         (__m256)__builtin_shufflevector((__v8sf)__a, (__v8sf)__b, \
         (mask) & 0x3,                ((mask) & 0xc) >> 2, \
         (((mask) & 0x30) >> 4) + 8,  (((mask) & 0xc0) >> 6) + 8, \
         ((mask) & 0x3) + 4,          (((mask) & 0xc) >> 2) + 4, \
         (((mask) & 0x30) >> 4) + 12, (((mask) & 0xc0) >> 6) + 12); })
@@ skipping 65 matching lines @@
   __m128d __a = (a); \
   __m128d __b = (b); \
   (__m128d)__builtin_ia32_cmpsd((__v2df)__a, (__v2df)__b, (c)); })
 
 #define _mm_cmp_ss(a, b, c) __extension__ ({ \
   __m128 __a = (a); \
   __m128 __b = (b); \
   (__m128)__builtin_ia32_cmpss((__v4sf)__a, (__v4sf)__b, (c)); })
 
 /* Vector extract */
-static __inline __m128d __attribute__((__always_inline__, __nodebug__))
-_mm256_extractf128_pd(__m256d a, const int o)
-{
-  return (__m128d)__builtin_ia32_vextractf128_pd256((__v4df)a, o);
-}
+#define _mm256_extractf128_pd(A, O) __extension__ ({ \
+  __m256d __A = (A); \
+  (__m128d)__builtin_ia32_vextractf128_pd256((__v4df)__A, (O)); })
 
-static __inline __m128 __attribute__((__always_inline__, __nodebug__))
-_mm256_extractf128_ps(__m256 a, const int o)
-{
-  return (__m128)__builtin_ia32_vextractf128_ps256((__v8sf)a, o);
-}
+#define _mm256_extractf128_ps(A, O) __extension__ ({ \
+  __m256 __A = (A); \
+  (__m128)__builtin_ia32_vextractf128_ps256((__v8sf)__A, (O)); })
 
-static __inline __m128i __attribute__((__always_inline__, __nodebug__))
-_mm256_extractf128_si256(__m256i a, const int o)
-{
-  return (__m128i)__builtin_ia32_vextractf128_si256((__v8si)a, o);
-}
+#define _mm256_extractf128_si256(A, O) __extension__ ({ \
+  __m256i __A = (A); \
+  (__m128i)__builtin_ia32_vextractf128_si256((__v8si)__A, (O)); })
 
 static __inline int __attribute__((__always_inline__, __nodebug__))
 _mm256_extract_epi32(__m256i a, int const imm)
 {
   __v8si b = (__v8si)a;
   return b[imm];
 }
 
 static __inline int __attribute__((__always_inline__, __nodebug__))
 _mm256_extract_epi16(__m256i a, int const imm)
@@ skipping 12 matching lines @@
 #ifdef __x86_64__
 static __inline long long  __attribute__((__always_inline__, __nodebug__))
 _mm256_extract_epi64(__m256i a, const int imm)
 {
   __v4di b = (__v4di)a;
   return b[imm];
 }
 #endif
 
 /* Vector insert */
-static __inline __m256d __attribute__((__always_inline__, __nodebug__))
-_mm256_insertf128_pd(__m256d a, __m128d b, const int o)
-{
-  return (__m256d)__builtin_ia32_vinsertf128_pd256((__v4df)a, (__v2df)b, o);
-}
+#define _mm256_insertf128_pd(V1, V2, O) __extension__ ({ \
+  __m256d __V1 = (V1); \
+  __m128d __V2 = (V2); \
+  (__m256d)__builtin_ia32_vinsertf128_pd256((__v4df)__V1, (__v2df)__V2, (O)); })
 
-static __inline __m256 __attribute__((__always_inline__, __nodebug__))
-_mm256_insertf128_ps(__m256 a, __m128 b, const int o)
-{
-  return (__m256)__builtin_ia32_vinsertf128_ps256((__v8sf)a, (__v4sf)b, o);
-}
+#define _mm256_insertf128_ps(V1, V2, O) __extension__ ({ \
+  __m256 __V1 = (V1); \
+  __m128 __V2 = (V2); \
+  (__m256)__builtin_ia32_vinsertf128_ps256((__v8sf)__V1, (__v4sf)__V2, (O)); })
 
-static __inline __m256i __attribute__((__always_inline__, __nodebug__))
-_mm256_insertf128_si256(__m256i a, __m128i b, const int o)
-{
-  return (__m256i)__builtin_ia32_vinsertf128_si256((__v8si)a, (__v4si)b, o);
-}
+#define _mm256_insertf128_si256(V1, V2, O) __extension__ ({ \
+  __m256i __V1 = (V1); \
+  __m128i __V2 = (V2); \
+  (__m256i)__builtin_ia32_vinsertf128_si256((__v8si)__V1, (__v4si)__V2, (O)); })
 
 static __inline __m256i __attribute__((__always_inline__, __nodebug__))
 _mm256_insert_epi32(__m256i a, int b, int const imm)
 {
   __v8si c = (__v8si)a;
   c[imm & 7] = b;
   return (__m256i)c;
 }
 
 static __inline __m256i __attribute__((__always_inline__, __nodebug__))
@@ skipping 646 matching lines @@
   __m128 zero = _mm_setzero_ps();
   return __builtin_shufflevector(in, zero, 0, 1, 2, 3, 4, 4, 4, 4);
 }
 
 static __inline __m256i __attribute__((__always_inline__, __nodebug__))
 _mm256_castsi128_si256(__m128i in)
 {
   __m128i zero = _mm_setzero_si128();
   return __builtin_shufflevector(in, zero, 0, 1, 2, 2);
 }