AVX512: Change builtin function name for scalar intrinsics.

[igorb]

Summary

AVX512: Change builtin function name for scalar intrinsics. Add "mask" to function name to reflect the function behavior.

[AsafBadouh]

LGTM

include/clang/Basic/BuiltinsX86.def

 BUILTIN(__builtin_ia32_rdtscp, "ULLiUi*", "")
 // PKU
 TARGET_BUILTIN(__builtin_ia32_rdpkru, "Ui", "", "pku")
 TARGET_BUILTIN(__builtin_ia32_wrpkru, "vUi", "", "pku")
 
 // AVX-512
 TARGET_BUILTIN(__builtin_ia32_sqrtpd512_mask, "V8dV8dV8dUcIi", "", "avx512f")
 TARGET_BUILTIN(__builtin_ia32_sqrtps512_mask, "V16fV16fV16fUsIi", "", "avx512f")
-TARGET_BUILTIN(__builtin_ia32_rsqrt14sd, "V2dV2dV2dV2dUc", "", "avx512f")
-TARGET_BUILTIN(__builtin_ia32_rsqrt14ss, "V4fV4fV4fV4fUc", "", "avx512f")
+TARGET_BUILTIN(__builtin_ia32_rsqrt14sd_mask, "V2dV2dV2dV2dUc", "", "avx512f")
+TARGET_BUILTIN(__builtin_ia32_rsqrt14ss_mask, "V4fV4fV4fV4fUc", "", "avx512f")
 TARGET_BUILTIN(__builtin_ia32_rsqrt14pd512_mask, "V8dV8dV8dUc", "", "avx512f")
 TARGET_BUILTIN(__builtin_ia32_rsqrt14ps512_mask, "V16fV16fV16fUs", "", "avx512f")
 
-TARGET_BUILTIN(__builtin_ia32_rsqrt28sd_round, "V2dV2dV2dV2dUcIi", "", "avx512er")
-TARGET_BUILTIN(__builtin_ia32_rsqrt28ss_round, "V4fV4fV4fV4fUcIi", "", "avx512er")
+TARGET_BUILTIN(__builtin_ia32_rsqrt28sd_round_mask, "V2dV2dV2dV2dUcIi", "", "avx512er")
+TARGET_BUILTIN(__builtin_ia32_rsqrt28ss_round_mask, "V4fV4fV4fV4fUcIi", "", "avx512er")
 TARGET_BUILTIN(__builtin_ia32_rsqrt28pd_mask, "V8dV8dV8dUcIi", "", "avx512er")
 TARGET_BUILTIN(__builtin_ia32_rsqrt28ps_mask, "V16fV16fV16fUsIi", "", "avx512er")
 
-TARGET_BUILTIN(__builtin_ia32_rcp14sd, "V2dV2dV2dV2dUc", "", "avx512f")
-TARGET_BUILTIN(__builtin_ia32_rcp14ss, "V4fV4fV4fV4fUc", "", "avx512f")
+TARGET_BUILTIN(__builtin_ia32_rcp14sd_mask, "V2dV2dV2dV2dUc", "", "avx512f")
+TARGET_BUILTIN(__builtin_ia32_rcp14ss_mask, "V4fV4fV4fV4fUc", "", "avx512f")
 TARGET_BUILTIN(__builtin_ia32_rcp14pd512_mask, "V8dV8dV8dUc", "", "avx512f")
 TARGET_BUILTIN(__builtin_ia32_rcp14ps512_mask, "V16fV16fV16fUs", "", "avx512f")
 
-TARGET_BUILTIN(__builtin_ia32_rcp28sd_round, "V2dV2dV2dV2dUcIi", "", "avx512er")
-TARGET_BUILTIN(__builtin_ia32_rcp28ss_round, "V4fV4fV4fV4fUcIi", "", "avx512er")
+TARGET_BUILTIN(__builtin_ia32_rcp28sd_round_mask, "V2dV2dV2dV2dUcIi", "", "avx512er")
+TARGET_BUILTIN(__builtin_ia32_rcp28ss_round_mask, "V4fV4fV4fV4fUcIi", "", "avx512er")
 TARGET_BUILTIN(__builtin_ia32_rcp28pd_mask, "V8dV8dV8dUcIi", "", "avx512er")
 TARGET_BUILTIN(__builtin_ia32_rcp28ps_mask, "V16fV16fV16fUsIi", "", "avx512er")
 TARGET_BUILTIN(__builtin_ia32_exp2pd_mask, "V8dV8dV8dUcIi", "", "avx512er")
 TARGET_BUILTIN(__builtin_ia32_exp2ps_mask, "V16fV16fV16fUsIi", "", "avx512er")
 
 TARGET_BUILTIN(__builtin_ia32_cvttps2dq512_mask, "V16iV16fV16iUsIi", "", "avx512f")
 TARGET_BUILTIN(__builtin_ia32_cvttps2udq512_mask, "V16iV16fV16iUsIi", "", "avx512f")
 TARGET_BUILTIN(__builtin_ia32_cvttpd2dq512_mask, "V8iV8dV8iUcIi", "", "avx512f")

 TARGET_BUILTIN(__builtin_ia32_mulpd512_mask, "V8dV8dV8dV8dUcIi", "", "avx512f")
 TARGET_BUILTIN(__builtin_ia32_mulps512_mask, "V16fV16fV16fV16fUsIi", "", "avx512f")
 TARGET_BUILTIN(__builtin_ia32_subpd512_mask, "V8dV8dV8dV8dUcIi", "", "avx512f")
 TARGET_BUILTIN(__builtin_ia32_subps512_mask, "V16fV16fV16fV16fUsIi", "", "avx512f")
 
 TARGET_BUILTIN(__builtin_ia32_pmaddubsw512_mask, "V32sV64cV64cV32sUi", "", "avx512bw")
 TARGET_BUILTIN(__builtin_ia32_pmaddwd512_mask, "V16iV32sV32sV16iUs", "", "avx512bw")
 
-TARGET_BUILTIN(__builtin_ia32_addss_round, "V4fV4fV4fV4fUcIi", "", "avx512f")
-TARGET_BUILTIN(__builtin_ia32_divss_round, "V4fV4fV4fV4fUcIi", "", "avx512f")
-TARGET_BUILTIN(__builtin_ia32_mulss_round, "V4fV4fV4fV4fUcIi", "", "avx512f")
-TARGET_BUILTIN(__builtin_ia32_subss_round, "V4fV4fV4fV4fUcIi", "", "avx512f")
-TARGET_BUILTIN(__builtin_ia32_maxss_round, "V4fV4fV4fV4fUcIi", "", "avx512f")
-TARGET_BUILTIN(__builtin_ia32_minss_round, "V4fV4fV4fV4fUcIi", "", "avx512f")
-TARGET_BUILTIN(__builtin_ia32_addsd_round, "V2dV2dV2dV2dUcIi", "", "avx512f")
-TARGET_BUILTIN(__builtin_ia32_divsd_round, "V2dV2dV2dV2dUcIi", "", "avx512f")
-TARGET_BUILTIN(__builtin_ia32_mulsd_round, "V2dV2dV2dV2dUcIi", "", "avx512f")
-TARGET_BUILTIN(__builtin_ia32_subsd_round, "V2dV2dV2dV2dUcIi", "", "avx512f")
-TARGET_BUILTIN(__builtin_ia32_maxsd_round, "V2dV2dV2dV2dUcIi", "", "avx512f")
-TARGET_BUILTIN(__builtin_ia32_minsd_round, "V2dV2dV2dV2dUcIi", "", "avx512f")
+TARGET_BUILTIN(__builtin_ia32_addss_round_mask, "V4fV4fV4fV4fUcIi", "", "avx512f")
+TARGET_BUILTIN(__builtin_ia32_divss_round_mask, "V4fV4fV4fV4fUcIi", "", "avx512f")
+TARGET_BUILTIN(__builtin_ia32_mulss_round_mask, "V4fV4fV4fV4fUcIi", "", "avx512f")
+TARGET_BUILTIN(__builtin_ia32_subss_round_mask, "V4fV4fV4fV4fUcIi", "", "avx512f")
+TARGET_BUILTIN(__builtin_ia32_maxss_round_mask, "V4fV4fV4fV4fUcIi", "", "avx512f")
+TARGET_BUILTIN(__builtin_ia32_minss_round_mask, "V4fV4fV4fV4fUcIi", "", "avx512f")
+TARGET_BUILTIN(__builtin_ia32_addsd_round_mask, "V2dV2dV2dV2dUcIi", "", "avx512f")
+TARGET_BUILTIN(__builtin_ia32_divsd_round_mask, "V2dV2dV2dV2dUcIi", "", "avx512f")
+TARGET_BUILTIN(__builtin_ia32_mulsd_round_mask, "V2dV2dV2dV2dUcIi", "", "avx512f")
+TARGET_BUILTIN(__builtin_ia32_subsd_round_mask, "V2dV2dV2dV2dUcIi", "", "avx512f")
+TARGET_BUILTIN(__builtin_ia32_maxsd_round_mask, "V2dV2dV2dV2dUcIi", "", "avx512f")
+TARGET_BUILTIN(__builtin_ia32_minsd_round_mask, "V2dV2dV2dV2dUcIi", "", "avx512f")
 
 TARGET_BUILTIN(__builtin_ia32_addpd128_mask, "V2dV2dV2dV2dUc", "", "avx512vl")
 TARGET_BUILTIN(__builtin_ia32_addpd256_mask, "V4dV4dV4dV4dUc", "", "avx512vl")
 TARGET_BUILTIN(__builtin_ia32_addps128_mask, "V4fV4fV4fV4fUc", "", "avx512vl")
 TARGET_BUILTIN(__builtin_ia32_addps256_mask, "V8fV8fV8fV8fUc", "", "avx512vl")
 TARGET_BUILTIN(__builtin_ia32_blendmd_128_mask, "V4iV4iV4iUc", "", "avx512vl")
 TARGET_BUILTIN(__builtin_ia32_blendmd_256_mask, "V8iV8iV8iUc", "", "avx512vl")
 TARGET_BUILTIN(__builtin_ia32_blendmpd_128_mask, "V2dV2dV2dUc", "", "avx512vl")

lib/Headers/avx512erintrin.h

 
 #define _mm512_mask_rsqrt28_ps(S, M, A) \
   _mm512_mask_rsqrt28_round_ps((S), (M), A, _MM_FROUND_CUR_DIRECTION)
 
 #define _mm512_maskz_rsqrt28_ps(M, A) \
   _mm512_maskz_rsqrt28_round_ps((M), (A), _MM_FROUND_CUR_DIRECTION)
 
 #define _mm_rsqrt28_round_ss(A, B, R) __extension__ ({ \
-  (__m128)__builtin_ia32_rsqrt28ss_round((__v4sf)(__m128)(A), \
+  (__m128)__builtin_ia32_rsqrt28ss_round_mask((__v4sf)(__m128)(A), \
                                         (__v4sf)(__m128)(B), \
                                         (__v4sf)_mm_setzero_ps(), \
                                         (__mmask8)-1, (R)); })
 
 #define _mm_mask_rsqrt28_round_ss(S, M, A, B, R) __extension__ ({ \
-  (__m128)__builtin_ia32_rsqrt28ss_round((__v4sf)(__m128)(A), \
+  (__m128)__builtin_ia32_rsqrt28ss_round_mask((__v4sf)(__m128)(A), \
                                         (__v4sf)(__m128)(B), \
                                         (__v4sf)(__m128)(S), \
                                         (__mmask8)(M), (R)); })
 
 #define _mm_maskz_rsqrt28_round_ss(M, A, B, R) __extension__ ({ \
-  (__m128)__builtin_ia32_rsqrt28ss_round((__v4sf)(__m128)(A), \
+  (__m128)__builtin_ia32_rsqrt28ss_round_mask((__v4sf)(__m128)(A), \
                                         (__v4sf)(__m128)(B), \
                                         (__v4sf)_mm_setzero_ps(), \
                                         (__mmask8)(M), (R)); })
 
 #define _mm_rsqrt28_ss(A, B) \
   _mm_rsqrt28_round_ss((A), (B), _MM_FROUND_CUR_DIRECTION)
 
 #define _mm_mask_rsqrt28_ss(S, M, A, B) \
   _mm_mask_rsqrt28_round_ss((S), (M), (A), (B), _MM_FROUND_CUR_DIRECTION)
 
 #define _mm_maskz_rsqrt28_ss(M, A, B) \
   _mm_maskz_rsqrt28_round_ss((M), (A), (B), _MM_FROUND_CUR_DIRECTION)
 
 #define _mm_rsqrt28_round_sd(A, B, R) __extension__ ({ \
-  (__m128d)__builtin_ia32_rsqrt28sd_round((__v2df)(__m128d)(A), \
+  (__m128d)__builtin_ia32_rsqrt28sd_round_mask((__v2df)(__m128d)(A), \
                                          (__v2df)(__m128d)(B), \
                                          (__v2df)_mm_setzero_pd(), \
                                          (__mmask8)-1, (R)); })
 
 #define _mm_mask_rsqrt28_round_sd(S, M, A, B, R) __extension__ ({ \
-  (__m128d)__builtin_ia32_rsqrt28sd_round((__v2df)(__m128d)(A), \
+  (__m128d)__builtin_ia32_rsqrt28sd_round_mask((__v2df)(__m128d)(A), \
                                          (__v2df)(__m128d)(B), \
                                          (__v2df)(__m128d)(S), \
                                          (__mmask8)(M), (R)); })
 
 #define _mm_maskz_rsqrt28_round_sd(M, A, B, R) __extension__ ({ \
-  (__m128d)__builtin_ia32_rsqrt28sd_round((__v2df)(__m128d)(A), \
+  (__m128d)__builtin_ia32_rsqrt28sd_round_mask((__v2df)(__m128d)(A), \
                                          (__v2df)(__m128d)(B), \
                                          (__v2df)_mm_setzero_pd(), \
                                          (__mmask8)(M), (R)); })
 
 #define _mm_rsqrt28_sd(A, B) \
   _mm_rsqrt28_round_sd((A), (B), _MM_FROUND_CUR_DIRECTION)
 
 #define _mm_mask_rsqrt28_sd(S, M, A, B) \

 
 #define _mm512_mask_rcp28_ps(S, M, A) \
   _mm512_mask_rcp28_round_ps((S), (M), (A), _MM_FROUND_CUR_DIRECTION)
 
 #define _mm512_maskz_rcp28_ps(M, A) \
   _mm512_maskz_rcp28_round_ps((M), (A), _MM_FROUND_CUR_DIRECTION)
 
 #define _mm_rcp28_round_ss(A, B, R) __extension__ ({ \
-  (__m128)__builtin_ia32_rcp28ss_round((__v4sf)(__m128)(A), \
+  (__m128)__builtin_ia32_rcp28ss_round_mask((__v4sf)(__m128)(A), \
                                       (__v4sf)(__m128)(B), \
                                       (__v4sf)_mm_setzero_ps(), \
                                       (__mmask8)-1, (R)); })
 
 #define _mm_mask_rcp28_round_ss(S, M, A, B, R) __extension__ ({ \
-  (__m128)__builtin_ia32_rcp28ss_round((__v4sf)(__m128)(A), \
+  (__m128)__builtin_ia32_rcp28ss_round_mask((__v4sf)(__m128)(A), \
                                       (__v4sf)(__m128)(B), \
                                       (__v4sf)(__m128)(S), \
                                       (__mmask8)(M), (R)); })
 
 #define _mm_maskz_rcp28_round_ss(M, A, B, R) __extension__ ({ \
-  (__m128)__builtin_ia32_rcp28ss_round((__v4sf)(__m128)(A), \
+  (__m128)__builtin_ia32_rcp28ss_round_mask((__v4sf)(__m128)(A), \
                                       (__v4sf)(__m128)(B), \
                                       (__v4sf)_mm_setzero_ps(), \
                                       (__mmask8)(M), (R)); })
 
 #define _mm_rcp28_ss(A, B) \
   _mm_rcp28_round_ss((A), (B), _MM_FROUND_CUR_DIRECTION)
 
 #define _mm_mask_rcp28_ss(S, M, A, B) \
   _mm_mask_rcp28_round_ss((S), (M), (A), (B), _MM_FROUND_CUR_DIRECTION)
 
 #define _mm_maskz_rcp28_ss(M, A, B) \
   _mm_maskz_rcp28_round_ss((M), (A), (B), _MM_FROUND_CUR_DIRECTION)
 
 #define _mm_rcp28_round_sd(A, B, R) __extension__ ({ \
-  (__m128d)__builtin_ia32_rcp28sd_round((__v2df)(__m128d)(A), \
+  (__m128d)__builtin_ia32_rcp28sd_round_mask((__v2df)(__m128d)(A), \
                                        (__v2df)(__m128d)(B), \
                                        (__v2df)_mm_setzero_pd(), \
                                        (__mmask8)-1, (R)); })
 
 #define _mm_mask_rcp28_round_sd(S, M, A, B, R) __extension__ ({ \
-  (__m128d)__builtin_ia32_rcp28sd_round((__v2df)(__m128d)(A), \
+  (__m128d)__builtin_ia32_rcp28sd_round_mask((__v2df)(__m128d)(A), \
                                        (__v2df)(__m128d)(B), \
                                        (__v2df)(__m128d)(S), \
                                        (__mmask8)(M), (R)); })
 
 #define _mm_maskz_rcp28_round_sd(M, A, B, R) __extension__ ({ \
-  (__m128d)__builtin_ia32_rcp28sd_round((__v2df)(__m128d)(A), \
+  (__m128d)__builtin_ia32_rcp28sd_round_mask((__v2df)(__m128d)(A), \
                                        (__v2df)(__m128d)(B), \
                                        (__v2df)_mm_setzero_pd(), \
                                        (__mmask8)(M), (R)); })
 
 #define _mm_rcp28_sd(A, B) \
   _mm_rcp28_round_sd((A), (B), _MM_FROUND_CUR_DIRECTION)
 
 #define _mm_mask_rcp28_sd(S, M, A, B) \
   _mm_mask_rcp28_round_sd((S), (M), (A), (B), _MM_FROUND_CUR_DIRECTION)
 
 #define _mm_maskz_rcp28_sd(M, A, B) \
   _mm_maskz_rcp28_round_sd((M), (A), (B), _MM_FROUND_CUR_DIRECTION)
 
 #endif // __AVX512ERINTRIN_H

lib/Headers/avx512fintrin.h

return (__m512) __builtin_ia32_maxps512_mask ((__v16sf) __A,
             (__v16sf)
             _mm512_setzero_ps (),
             (__mmask16) -1,
             _MM_FROUND_CUR_DIRECTION);
 }
 
 static __inline__ __m128 __DEFAULT_FN_ATTRS
 _mm_mask_max_ss(__m128 __W, __mmask8 __U,__m128 __A, __m128 __B) {
-  return (__m128) __builtin_ia32_maxss_round ((__v4sf) __A,
+  return (__m128) __builtin_ia32_maxss_round_mask ((__v4sf) __A,
                 (__v4sf) __B,
                 (__v4sf) __W,
                 (__mmask8) __U,
                 _MM_FROUND_CUR_DIRECTION);
 }
 
 static __inline__ __m128 __DEFAULT_FN_ATTRS
 _mm_maskz_max_ss(__mmask8 __U,__m128 __A, __m128 __B) {
-  return (__m128) __builtin_ia32_maxss_round ((__v4sf) __A,
+  return (__m128) __builtin_ia32_maxss_round_mask ((__v4sf) __A,
                 (__v4sf) __B,
                 (__v4sf)  _mm_setzero_ps (),
                 (__mmask8) __U,
                 _MM_FROUND_CUR_DIRECTION);
 }
 
 #define _mm_max_round_ss(__A, __B, __R) __extension__ ({ \
-  (__m128) __builtin_ia32_maxss_round ((__v4sf) __A, (__v4sf) __B, \
+  (__m128) __builtin_ia32_maxss_round_mask ((__v4sf) __A, (__v4sf) __B, \
                 (__v4sf) _mm_setzero_ps(), (__mmask8) -1, __R); })
 
 #define _mm_mask_max_round_ss(__W, __U, __A, __B, __R) __extension__ ({ \
-  (__m128) __builtin_ia32_maxss_round ((__v4sf) __A, (__v4sf) __B, \
+  (__m128) __builtin_ia32_maxss_round_mask ((__v4sf) __A, (__v4sf) __B, \
                 (__v4sf)  __W, (__mmask8) __U,__R); })
 
 #define _mm_maskz_max_round_ss(__U, __A, __B, __R) __extension__ ({ \
-  (__m128) __builtin_ia32_maxss_round ((__v4sf) __A, (__v4sf) __B, \
+  (__m128) __builtin_ia32_maxss_round_mask ((__v4sf) __A, (__v4sf) __B, \
                 (__v4sf)  _mm_setzero_ps(), (__mmask8) __U,__R); })
 
 static __inline__ __m128d __DEFAULT_FN_ATTRS
 _mm_mask_max_sd(__m128d __W, __mmask8 __U,__m128d __A, __m128d __B) {
-  return (__m128d) __builtin_ia32_maxsd_round ((__v2df) __A,
+  return (__m128d) __builtin_ia32_maxsd_round_mask ((__v2df) __A,
                 (__v2df) __B,
                 (__v2df) __W,
                 (__mmask8) __U,
                 _MM_FROUND_CUR_DIRECTION);
 }
 
 static __inline__ __m128d __DEFAULT_FN_ATTRS
 _mm_maskz_max_sd(__mmask8 __U,__m128d __A, __m128d __B) {
-  return (__m128d) __builtin_ia32_maxsd_round ((__v2df) __A,
+  return (__m128d) __builtin_ia32_maxsd_round_mask ((__v2df) __A,
                 (__v2df) __B,
                 (__v2df)  _mm_setzero_pd (),
                 (__mmask8) __U,
                 _MM_FROUND_CUR_DIRECTION);
 }
 
 #define _mm_max_round_sd(__A, __B, __R) __extension__ ({ \
-  (__m128d) __builtin_ia32_maxsd_round ((__v2df) __A, (__v2df) __B, \
+  (__m128d) __builtin_ia32_maxsd_round_mask ((__v2df) __A, (__v2df) __B, \
                 (__v2df) _mm_setzero_pd(), (__mmask8) -1, __R); })
 
 #define _mm_mask_max_round_sd(__W, __U, __A, __B, __R) __extension__ ({ \
-  (__m128d) __builtin_ia32_maxsd_round ((__v2df) __A, (__v2df) __B, \
+  (__m128d) __builtin_ia32_maxsd_round_mask ((__v2df) __A, (__v2df) __B, \
                 (__v2df)  __W, (__mmask8) __U,__R); })
 
 #define _mm_maskz_max_round_sd(__U, __A, __B, __R) __extension__ ({ \
-  (__m128d) __builtin_ia32_maxsd_round ((__v2df) __A, (__v2df) __B, \
+  (__m128d) __builtin_ia32_maxsd_round_mask ((__v2df) __A, (__v2df) __B, \
                 (__v2df)  _mm_setzero_pd(), (__mmask8) __U,__R); })
 
 static __inline __m512i
 __DEFAULT_FN_ATTRS
 _mm512_max_epi32(__m512i __A, __m512i __B)
 {
   return (__m512i) __builtin_ia32_pmaxsd512_mask ((__v16si) __A,
               (__v16si) __B,
return (__m512) __builtin_ia32_minps512_mask ((__v16sf) __A,
             (__v16sf)
             _mm512_setzero_ps (),
             (__mmask16) -1,
             _MM_FROUND_CUR_DIRECTION);
 }
 
 static __inline__ __m128 __DEFAULT_FN_ATTRS
 _mm_mask_min_ss(__m128 __W, __mmask8 __U,__m128 __A, __m128 __B) {
-  return (__m128) __builtin_ia32_minss_round ((__v4sf) __A,
+  return (__m128) __builtin_ia32_minss_round_mask ((__v4sf) __A,
                 (__v4sf) __B,
                 (__v4sf) __W,
                 (__mmask8) __U,
                 _MM_FROUND_CUR_DIRECTION);
 }
 
 static __inline__ __m128 __DEFAULT_FN_ATTRS
 _mm_maskz_min_ss(__mmask8 __U,__m128 __A, __m128 __B) {
-  return (__m128) __builtin_ia32_minss_round ((__v4sf) __A,
+  return (__m128) __builtin_ia32_minss_round_mask ((__v4sf) __A,
                 (__v4sf) __B,
                 (__v4sf)  _mm_setzero_ps (),
                 (__mmask8) __U,
                 _MM_FROUND_CUR_DIRECTION);
 }
 
 #define _mm_min_round_ss(__A, __B, __R) __extension__ ({ \
-  (__m128) __builtin_ia32_minss_round ((__v4sf) __A, (__v4sf) __B, \
+  (__m128) __builtin_ia32_minss_round_mask ((__v4sf) __A, (__v4sf) __B, \
                 (__v4sf) _mm_setzero_ps(), (__mmask8) -1, __R); })
 
 #define _mm_mask_min_round_ss(__W, __U, __A, __B, __R) __extension__ ({ \
-  (__m128) __builtin_ia32_minss_round ((__v4sf) __A, (__v4sf) __B, \
+  (__m128) __builtin_ia32_minss_round_mask ((__v4sf) __A, (__v4sf) __B, \
                 (__v4sf)  __W, (__mmask8) __U,__R); })
 
 #define _mm_maskz_min_round_ss(__U, __A, __B, __R) __extension__ ({ \
-  (__m128) __builtin_ia32_minss_round ((__v4sf) __A, (__v4sf) __B, \
+  (__m128) __builtin_ia32_minss_round_mask ((__v4sf) __A, (__v4sf) __B, \
                 (__v4sf)  _mm_setzero_ps(), (__mmask8) __U,__R); })
 
 static __inline__ __m128d __DEFAULT_FN_ATTRS
 _mm_mask_min_sd(__m128d __W, __mmask8 __U,__m128d __A, __m128d __B) {
-  return (__m128d) __builtin_ia32_minsd_round ((__v2df) __A,
+  return (__m128d) __builtin_ia32_minsd_round_mask ((__v2df) __A,
                 (__v2df) __B,
                 (__v2df) __W,
                 (__mmask8) __U,
                 _MM_FROUND_CUR_DIRECTION);
 }
 
 static __inline__ __m128d __DEFAULT_FN_ATTRS
 _mm_maskz_min_sd(__mmask8 __U,__m128d __A, __m128d __B) {
-  return (__m128d) __builtin_ia32_minsd_round ((__v2df) __A,
+  return (__m128d) __builtin_ia32_minsd_round_mask ((__v2df) __A,
                 (__v2df) __B,
                 (__v2df)  _mm_setzero_pd (),
                 (__mmask8) __U,
                 _MM_FROUND_CUR_DIRECTION);
 }
 
 #define _mm_min_round_sd(__A, __B, __R) __extension__ ({ \
-  (__m128d) __builtin_ia32_minsd_round ((__v2df) __A, (__v2df) __B, \
+  (__m128d) __builtin_ia32_minsd_round_mask ((__v2df) __A, (__v2df) __B, \
                 (__v2df) _mm_setzero_pd(), (__mmask8) -1, __R); })
 
 #define _mm_mask_min_round_sd(__W, __U, __A, __B, __R) __extension__ ({ \
-  (__m128d) __builtin_ia32_minsd_round ((__v2df) __A, (__v2df) __B, \
+  (__m128d) __builtin_ia32_minsd_round_mask ((__v2df) __A, (__v2df) __B, \
                 (__v2df)  __W, (__mmask8) __U,__R); })
 
 #define _mm_maskz_min_round_sd(__U, __A, __B, __R) __extension__ ({ \
-  (__m128d) __builtin_ia32_minsd_round ((__v2df) __A, (__v2df) __B, \
+  (__m128d) __builtin_ia32_minsd_round_mask ((__v2df) __A, (__v2df) __B, \
                 (__v2df)  _mm_setzero_pd(), (__mmask8) __U,__R); })
 
 static __inline __m512i
 __DEFAULT_FN_ATTRS
 _mm512_min_epi32(__m512i __A, __m512i __B)
 {
   return (__m512i) __builtin_ia32_pminsd512_mask ((__v16si) __A,
               (__v16si) __B,
return (__m512) __builtin_ia32_rsqrt14ps512_mask ((__v16sf) __A,
                 (__v16sf)
                 _mm512_setzero_ps (),
                 (__mmask16) -1);
 }
 
 static  __inline__ __m128 __DEFAULT_FN_ATTRS
 _mm_rsqrt14_ss(__m128 __A, __m128 __B)
 {
-  return (__m128) __builtin_ia32_rsqrt14ss ((__v4sf) __A,
+  return (__m128) __builtin_ia32_rsqrt14ss_mask ((__v4sf) __A,
              (__v4sf) __B,
              (__v4sf)
              _mm_setzero_ps (),
              (__mmask8) -1);
 }
 
 static  __inline__ __m128d __DEFAULT_FN_ATTRS
 _mm_rsqrt14_sd(__m128d __A, __m128d __B)
 {
-  return (__m128d) __builtin_ia32_rsqrt14sd ((__v2df) __A,
+  return (__m128d) __builtin_ia32_rsqrt14sd_mask ((__v2df) __A,
               (__v2df) __B,
               (__v2df)
               _mm_setzero_pd (),
               (__mmask8) -1);
 }
 
 static  __inline__ __m512d __DEFAULT_FN_ATTRS
 _mm512_rcp14_pd(__m512d __A)
_mm512_rcp14_ps(__m512 __A)
   return (__m512) __builtin_ia32_rcp14ps512_mask ((__v16sf) __A,
               (__v16sf)
               _mm512_setzero_ps (),
               (__mmask16) -1);
 }
 static  __inline__ __m128 __DEFAULT_FN_ATTRS
 _mm_rcp14_ss(__m128 __A, __m128 __B)
 {
-  return (__m128) __builtin_ia32_rcp14ss ((__v4sf) __A,
+  return (__m128) __builtin_ia32_rcp14ss_mask ((__v4sf) __A,
                  (__v4sf) __B,
                  (__v4sf)
                  _mm_setzero_ps (),
                  (__mmask8) -1);
 }
 
 static  __inline__ __m128d __DEFAULT_FN_ATTRS
 _mm_rcp14_sd(__m128d __A, __m128d __B)
 {
-  return (__m128d) __builtin_ia32_rcp14sd ((__v2df) __A,
+  return (__m128d) __builtin_ia32_rcp14sd_mask ((__v2df) __A,
             (__v2df) __B,
             (__v2df)
             _mm_setzero_pd (),
             (__mmask8) -1);
 }
 
 static __inline __m512 __DEFAULT_FN_ATTRS
 _mm512_floor_ps(__m512 __A)
_mm512_abs_epi32(__m512i __A)
   return (__m512i) __builtin_ia32_pabsd512_mask ((__v16si) __A,
              (__v16si)
              _mm512_setzero_si512 (),
              (__mmask16) -1);
 }
 
 static __inline__ __m128 __DEFAULT_FN_ATTRS
 _mm_mask_add_ss(__m128 __W, __mmask8 __U,__m128 __A, __m128 __B) {
-  return (__m128) __builtin_ia32_addss_round ((__v4sf) __A,
+  return (__m128) __builtin_ia32_addss_round_mask ((__v4sf) __A,
                 (__v4sf) __B,
                 (__v4sf) __W,
                 (__mmask8) __U,
                 _MM_FROUND_CUR_DIRECTION);
 }
 
 static __inline__ __m128 __DEFAULT_FN_ATTRS
 _mm_maskz_add_ss(__mmask8 __U,__m128 __A, __m128 __B) {
-  return (__m128) __builtin_ia32_addss_round ((__v4sf) __A,
+  return (__m128) __builtin_ia32_addss_round_mask ((__v4sf) __A,
                 (__v4sf) __B,
                 (__v4sf)  _mm_setzero_ps (),
                 (__mmask8) __U,
                 _MM_FROUND_CUR_DIRECTION);
 }
 
 #define _mm_add_round_ss(__A, __B, __R) __extension__ ({ \
-  (__m128) __builtin_ia32_addss_round ((__v4sf) __A, (__v4sf) __B, \
+  (__m128) __builtin_ia32_addss_round_mask ((__v4sf) __A, (__v4sf) __B, \
                 (__v4sf) _mm_setzero_ps(), (__mmask8) -1, __R); })
 
 #define _mm_mask_add_round_ss(__W, __U, __A, __B, __R) __extension__ ({ \
-  (__m128) __builtin_ia32_addss_round ((__v4sf) __A, (__v4sf) __B, \
+  (__m128) __builtin_ia32_addss_round_mask ((__v4sf) __A, (__v4sf) __B, \
                 (__v4sf)  __W, (__mmask8) __U,__R); })
 
 #define _mm_maskz_add_round_ss(__U, __A, __B, __R) __extension__ ({ \
-  (__m128) __builtin_ia32_addss_round ((__v4sf) __A, (__v4sf) __B, \
+  (__m128) __builtin_ia32_addss_round_mask ((__v4sf) __A, (__v4sf) __B, \
                 (__v4sf)  _mm_setzero_ps(), (__mmask8) __U,__R); })
 
 static __inline__ __m128d __DEFAULT_FN_ATTRS
 _mm_mask_add_sd(__m128d __W, __mmask8 __U,__m128d __A, __m128d __B) {
-  return (__m128d) __builtin_ia32_addsd_round ((__v2df) __A,
+  return (__m128d) __builtin_ia32_addsd_round_mask ((__v2df) __A,
                 (__v2df) __B,
                 (__v2df) __W,
                 (__mmask8) __U,
                 _MM_FROUND_CUR_DIRECTION);
 }
 
 static __inline__ __m128d __DEFAULT_FN_ATTRS
 _mm_maskz_add_sd(__mmask8 __U,__m128d __A, __m128d __B) {
-  return (__m128d) __builtin_ia32_addsd_round ((__v2df) __A,
+  return (__m128d) __builtin_ia32_addsd_round_mask ((__v2df) __A,
                 (__v2df) __B,
                 (__v2df)  _mm_setzero_pd (),
                 (__mmask8) __U,
                 _MM_FROUND_CUR_DIRECTION);
 }
 #define _mm_add_round_sd(__A, __B, __R) __extension__ ({ \
-  (__m128d) __builtin_ia32_addsd_round ((__v2df) __A, (__v2df) __B, \
+  (__m128d) __builtin_ia32_addsd_round_mask ((__v2df) __A, (__v2df) __B, \
                 (__v2df) _mm_setzero_pd(), (__mmask8) -1, __R); })
 
 #define _mm_mask_add_round_sd(__W, __U, __A, __B, __R) __extension__ ({ \
-  (__m128d) __builtin_ia32_addsd_round ((__v2df) __A, (__v2df) __B, \
+  (__m128d) __builtin_ia32_addsd_round_mask ((__v2df) __A, (__v2df) __B, \
                 (__v2df)  __W, (__mmask8) __U,__R); })
 
 #define _mm_maskz_add_round_sd(__U, __A, __B, __R) __extension__ ({ \
-  (__m128d) __builtin_ia32_addsd_round ((__v2df) __A, (__v2df) __B, \
+  (__m128d) __builtin_ia32_addsd_round_mask ((__v2df) __A, (__v2df) __B, \
                 (__v2df)  _mm_setzero_pd(), (__mmask8) __U,__R); })
 
 static __inline__ __m512d __DEFAULT_FN_ATTRS
 _mm512_mask_add_pd(__m512d __W, __mmask8 __U, __m512d __A, __m512d __B) {
   return (__m512d) __builtin_ia32_addpd512_mask ((__v8df) __A,
              (__v8df) __B,
              (__v8df) __W,
              (__mmask8) __U,
(__m512) __builtin_ia32_addps512_mask ((__v16sf) __A, (__v16sf) __B, \
                 (__v16sf) __W, (__mmask16)__U, __R); })
 
 #define _mm512_maskz_add_round_ps(__U, __A, __B, __R) __extension__ ({ \
   (__m512) __builtin_ia32_addps512_mask ((__v16sf) __A, (__v16sf) __B, \
                 (__v16sf) _mm512_setzero_ps(), (__mmask16)__U, __R); })
 
 static __inline__ __m128 __DEFAULT_FN_ATTRS
 _mm_mask_sub_ss(__m128 __W, __mmask8 __U,__m128 __A, __m128 __B) {
-  return (__m128) __builtin_ia32_subss_round ((__v4sf) __A,
+  return (__m128) __builtin_ia32_subss_round_mask ((__v4sf) __A,
                 (__v4sf) __B,
                 (__v4sf) __W,
                 (__mmask8) __U,
                 _MM_FROUND_CUR_DIRECTION);
 }
 
 static __inline__ __m128 __DEFAULT_FN_ATTRS
 _mm_maskz_sub_ss(__mmask8 __U,__m128 __A, __m128 __B) {
-  return (__m128) __builtin_ia32_subss_round ((__v4sf) __A,
+  return (__m128) __builtin_ia32_subss_round_mask ((__v4sf) __A,
                 (__v4sf) __B,
                 (__v4sf)  _mm_setzero_ps (),
                 (__mmask8) __U,
                 _MM_FROUND_CUR_DIRECTION);
 }
 #define _mm_sub_round_ss(__A, __B, __R) __extension__ ({ \
-  (__m128) __builtin_ia32_subss_round ((__v4sf) __A, (__v4sf) __B, \
+  (__m128) __builtin_ia32_subss_round_mask ((__v4sf) __A, (__v4sf) __B, \
                 (__v4sf) _mm_setzero_ps(), (__mmask8) -1, __R); })
 
 #define _mm_mask_sub_round_ss(__W, __U, __A, __B, __R) __extension__ ({ \
-  (__m128) __builtin_ia32_subss_round ((__v4sf) __A, (__v4sf) __B, \
+  (__m128) __builtin_ia32_subss_round_mask ((__v4sf) __A, (__v4sf) __B, \
                 (__v4sf)  __W, (__mmask8) __U,__R); })
 
 #define _mm_maskz_sub_round_ss(__U, __A, __B, __R) __extension__ ({ \
-  (__m128) __builtin_ia32_subss_round ((__v4sf) __A, (__v4sf) __B, \
+  (__m128) __builtin_ia32_subss_round_mask ((__v4sf) __A, (__v4sf) __B, \
                 (__v4sf)  _mm_setzero_ps(), (__mmask8) __U,__R); })
 
 static __inline__ __m128d __DEFAULT_FN_ATTRS
 _mm_mask_sub_sd(__m128d __W, __mmask8 __U,__m128d __A, __m128d __B) {
-  return (__m128d) __builtin_ia32_subsd_round ((__v2df) __A,
+  return (__m128d) __builtin_ia32_subsd_round_mask ((__v2df) __A,
                 (__v2df) __B,
                 (__v2df) __W,
                 (__mmask8) __U,
                 _MM_FROUND_CUR_DIRECTION);
 }
 
 static __inline__ __m128d __DEFAULT_FN_ATTRS
 _mm_maskz_sub_sd(__mmask8 __U,__m128d __A, __m128d __B) {
-  return (__m128d) __builtin_ia32_subsd_round ((__v2df) __A,
+  return (__m128d) __builtin_ia32_subsd_round_mask ((__v2df) __A,
                 (__v2df) __B,
                 (__v2df)  _mm_setzero_pd (),
                 (__mmask8) __U,
                 _MM_FROUND_CUR_DIRECTION);
 }
 
 #define _mm_sub_round_sd(__A, __B, __R) __extension__ ({ \
-  (__m128d) __builtin_ia32_subsd_round ((__v2df) __A, (__v2df) __B, \
+  (__m128d) __builtin_ia32_subsd_round_mask ((__v2df) __A, (__v2df) __B, \
                 (__v2df) _mm_setzero_pd(), (__mmask8) -1, __R); })
 
 #define _mm_mask_sub_round_sd(__W, __U, __A, __B, __R) __extension__ ({ \
-  (__m128d) __builtin_ia32_subsd_round ((__v2df) __A, (__v2df) __B, \
+  (__m128d) __builtin_ia32_subsd_round_mask ((__v2df) __A, (__v2df) __B, \
                 (__v2df)  __W, (__mmask8) __U,__R); })
 
 #define _mm_maskz_sub_round_sd(__U, __A, __B, __R) __extension__ ({ \
-  (__m128d) __builtin_ia32_subsd_round ((__v2df) __A, (__v2df) __B, \
+  (__m128d) __builtin_ia32_subsd_round_mask ((__v2df) __A, (__v2df) __B, \
                 (__v2df)  _mm_setzero_pd(), (__mmask8) __U,__R); })
 
 static __inline__ __m512d __DEFAULT_FN_ATTRS
 _mm512_mask_sub_pd(__m512d __W, __mmask8 __U, __m512d __A, __m512d __B) {
   return (__m512d) __builtin_ia32_subpd512_mask ((__v8df) __A,
              (__v8df) __B,
              (__v8df) __W,
              (__mmask8) __U,
(__m512) __builtin_ia32_subps512_mask ((__v16sf) __A, (__v16sf) __B, \
             (__v16sf) __W, (__mmask16) __U, __R); });
 
 #define _mm512_maskz_sub_round_ps(__U, __A, __B, __R)  __extension__ ({ \
   (__m512) __builtin_ia32_subps512_mask ((__v16sf) __A, (__v16sf) __B, \
             (__v16sf) _mm512_setzero_ps (), (__mmask16) __U, __R);});
 
 static __inline__ __m128 __DEFAULT_FN_ATTRS
 _mm_mask_mul_ss(__m128 __W, __mmask8 __U,__m128 __A, __m128 __B) {
-  return (__m128) __builtin_ia32_mulss_round ((__v4sf) __A,
+  return (__m128) __builtin_ia32_mulss_round_mask ((__v4sf) __A,
                 (__v4sf) __B,
                 (__v4sf) __W,
                 (__mmask8) __U,
                 _MM_FROUND_CUR_DIRECTION);
 }
 
 static __inline__ __m128 __DEFAULT_FN_ATTRS
 _mm_maskz_mul_ss(__mmask8 __U,__m128 __A, __m128 __B) {
-  return (__m128) __builtin_ia32_mulss_round ((__v4sf) __A,
+  return (__m128) __builtin_ia32_mulss_round_mask ((__v4sf) __A,
                 (__v4sf) __B,
                 (__v4sf)  _mm_setzero_ps (),
                 (__mmask8) __U,
                 _MM_FROUND_CUR_DIRECTION);
 }
 #define _mm_mul_round_ss(__A, __B, __R) __extension__ ({ \
-  (__m128) __builtin_ia32_mulss_round ((__v4sf) __A, (__v4sf) __B, \
+  (__m128) __builtin_ia32_mulss_round_mask ((__v4sf) __A, (__v4sf) __B, \
                 (__v4sf) _mm_setzero_ps(), (__mmask8) -1, __R); })
 
 #define _mm_mask_mul_round_ss(__W, __U, __A, __B, __R) __extension__ ({ \
-  (__m128) __builtin_ia32_mulss_round ((__v4sf) __A, (__v4sf) __B, \
+  (__m128) __builtin_ia32_mulss_round_mask ((__v4sf) __A, (__v4sf) __B, \
                 (__v4sf)  __W, (__mmask8) __U,__R); })
 
 #define _mm_maskz_mul_round_ss(__U, __A, __B, __R) __extension__ ({ \
-  (__m128) __builtin_ia32_mulss_round ((__v4sf) __A, (__v4sf) __B, \
+  (__m128) __builtin_ia32_mulss_round_mask ((__v4sf) __A, (__v4sf) __B, \
                 (__v4sf)  _mm_setzero_ps(), (__mmask8) __U,__R); })
 
 static __inline__ __m128d __DEFAULT_FN_ATTRS
 _mm_mask_mul_sd(__m128d __W, __mmask8 __U,__m128d __A, __m128d __B) {
-  return (__m128d) __builtin_ia32_mulsd_round ((__v2df) __A,
+  return (__m128d) __builtin_ia32_mulsd_round_mask ((__v2df) __A,
                 (__v2df) __B,
                 (__v2df) __W,
                 (__mmask8) __U,
                 _MM_FROUND_CUR_DIRECTION);
 }
 
 static __inline__ __m128d __DEFAULT_FN_ATTRS
 _mm_maskz_mul_sd(__mmask8 __U,__m128d __A, __m128d __B) {
-  return (__m128d) __builtin_ia32_mulsd_round ((__v2df) __A,
+  return (__m128d) __builtin_ia32_mulsd_round_mask ((__v2df) __A,
                 (__v2df) __B,
                 (__v2df)  _mm_setzero_pd (),
                 (__mmask8) __U,
                 _MM_FROUND_CUR_DIRECTION);
 }
 
 #define _mm_mul_round_sd(__A, __B, __R) __extension__ ({ \
-  (__m128d) __builtin_ia32_mulsd_round ((__v2df) __A, (__v2df) __B, \
+  (__m128d) __builtin_ia32_mulsd_round_mask ((__v2df) __A, (__v2df) __B, \
                 (__v2df) _mm_setzero_pd(), (__mmask8) -1, __R); })
 
 #define _mm_mask_mul_round_sd(__W, __U, __A, __B, __R) __extension__ ({ \
-  (__m128d) __builtin_ia32_mulsd_round ((__v2df) __A, (__v2df) __B, \
+  (__m128d) __builtin_ia32_mulsd_round_mask ((__v2df) __A, (__v2df) __B, \
                 (__v2df)  __W, (__mmask8) __U,__R); })
 
 #define _mm_maskz_mul_round_sd(__U, __A, __B, __R) __extension__ ({ \
-  (__m128d) __builtin_ia32_mulsd_round ((__v2df) __A, (__v2df) __B, \
+  (__m128d) __builtin_ia32_mulsd_round_mask ((__v2df) __A, (__v2df) __B, \
                 (__v2df)  _mm_setzero_pd(), (__mmask8) __U,__R); })
 
 static __inline__ __m512d __DEFAULT_FN_ATTRS
 _mm512_mask_mul_pd(__m512d __W, __mmask8 __U, __m512d __A, __m512d __B) {
   return (__m512d) __builtin_ia32_mulpd512_mask ((__v8df) __A,
              (__v8df) __B,
              (__v8df) __W,
              (__mmask8) __U,
(__m512) __builtin_ia32_mulps512_mask ((__v16sf) __A, (__v16sf) __B, \
             (__v16sf) __W, (__mmask16) __U, __R); });
 
 #define _mm512_maskz_mul_round_ps(__U, __A, __B, __R)  __extension__ ({ \
   (__m512) __builtin_ia32_mulps512_mask ((__v16sf) __A, (__v16sf) __B, \
             (__v16sf) _mm512_setzero_ps (), (__mmask16) __U, __R);});
 
 static __inline__ __m128 __DEFAULT_FN_ATTRS
 _mm_mask_div_ss(__m128 __W, __mmask8 __U,__m128 __A, __m128 __B) {
-  return (__m128) __builtin_ia32_divss_round ((__v4sf) __A,
+  return (__m128) __builtin_ia32_divss_round_mask ((__v4sf) __A,
                 (__v4sf) __B,
                 (__v4sf) __W,
                 (__mmask8) __U,
                 _MM_FROUND_CUR_DIRECTION);
 }
 
 static __inline__ __m128 __DEFAULT_FN_ATTRS
 _mm_maskz_div_ss(__mmask8 __U,__m128 __A, __m128 __B) {
-  return (__m128) __builtin_ia32_divss_round ((__v4sf) __A,
+  return (__m128) __builtin_ia32_divss_round_mask ((__v4sf) __A,
                 (__v4sf) __B,
                 (__v4sf)  _mm_setzero_ps (),
                 (__mmask8) __U,
                 _MM_FROUND_CUR_DIRECTION);
 }
 
 #define _mm_div_round_ss(__A, __B, __R) __extension__ ({ \
-  (__m128) __builtin_ia32_divss_round ((__v4sf) __A, (__v4sf) __B, \
+  (__m128) __builtin_ia32_divss_round_mask ((__v4sf) __A, (__v4sf) __B, \
                 (__v4sf) _mm_setzero_ps(), (__mmask8) -1, __R); })
 
 #define _mm_mask_div_round_ss(__W, __U, __A, __B, __R) __extension__ ({ \
-  (__m128) __builtin_ia32_divss_round ((__v4sf) __A, (__v4sf) __B, \
+  (__m128) __builtin_ia32_divss_round_mask ((__v4sf) __A, (__v4sf) __B, \
                 (__v4sf)  __W, (__mmask8) __U,__R); })
 
 #define _mm_maskz_div_round_ss(__U, __A, __B, __R) __extension__ ({ \
-  (__m128) __builtin_ia32_divss_round ((__v4sf) __A, (__v4sf) __B, \
+  (__m128) __builtin_ia32_divss_round_mask ((__v4sf) __A, (__v4sf) __B, \
                 (__v4sf)  _mm_setzero_ps(), (__mmask8) __U,__R); })
 
 static __inline__ __m128d __DEFAULT_FN_ATTRS
 _mm_mask_div_sd(__m128d __W, __mmask8 __U,__m128d __A, __m128d __B) {
-  return (__m128d) __builtin_ia32_divsd_round ((__v2df) __A,
+  return (__m128d) __builtin_ia32_divsd_round_mask ((__v2df) __A,
                 (__v2df) __B,
                 (__v2df) __W,
                 (__mmask8) __U,
                 _MM_FROUND_CUR_DIRECTION);
 }
 
 static __inline__ __m128d __DEFAULT_FN_ATTRS
 _mm_maskz_div_sd(__mmask8 __U,__m128d __A, __m128d __B) {
-  return (__m128d) __builtin_ia32_divsd_round ((__v2df) __A,
+  return (__m128d) __builtin_ia32_divsd_round_mask ((__v2df) __A,
                 (__v2df) __B,
                 (__v2df)  _mm_setzero_pd (),
                 (__mmask8) __U,
                 _MM_FROUND_CUR_DIRECTION);
 }
 
 #define _mm_div_round_sd(__A, __B, __R) __extension__ ({ \
-  (__m128d) __builtin_ia32_divsd_round ((__v2df) __A, (__v2df) __B, \
+  (__m128d) __builtin_ia32_divsd_round_mask ((__v2df) __A, (__v2df) __B, \
                 (__v2df) _mm_setzero_pd(), (__mmask8) -1, __R); })
 
 #define _mm_mask_div_round_sd(__W, __U, __A, __B, __R) __extension__ ({ \
-  (__m128d) __builtin_ia32_divsd_round ((__v2df) __A, (__v2df) __B, \
+  (__m128d) __builtin_ia32_divsd_round_mask ((__v2df) __A, (__v2df) __B, \
                 (__v2df)  __W, (__mmask8) __U,__R); })
 
 #define _mm_maskz_div_round_sd(__U, __A, __B, __R) __extension__ ({ \
-  (__m128d) __builtin_ia32_divsd_round ((__v2df) __A, (__v2df) __B, \
+  (__m128d) __builtin_ia32_divsd_round_mask ((__v2df) __A, (__v2df) __B, \
                 (__v2df)  _mm_setzero_pd(), (__mmask8) __U,__R); })
 
 static __inline__ __m512d __DEFAULT_FN_ATTRS
 _mm512_mask_div_pd(__m512d __W, __mmask8 __U, __m512d __A, __m512d __B) {
   return (__m512d) __builtin_ia32_divpd512_mask ((__v8df) __A,
              (__v8df) __B,
              (__v8df) __W,
              (__mmask8) __U,