Index: cfe/trunk/include/clang/Basic/BuiltinsX86.def =================================================================== --- cfe/trunk/include/clang/Basic/BuiltinsX86.def +++ cfe/trunk/include/clang/Basic/BuiltinsX86.def @@ -931,23 +931,23 @@ // 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") @@ -1310,18 +1310,18 @@ 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") Index: cfe/trunk/lib/Headers/avx512erintrin.h =================================================================== --- cfe/trunk/lib/Headers/avx512erintrin.h +++ cfe/trunk/lib/Headers/avx512erintrin.h @@ -126,19 +126,19 @@ _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)); }) @@ -153,19 +153,19 @@ _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)); }) @@ -229,19 +229,19 @@ _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)); }) @@ -256,19 +256,19 @@ _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)); }) Index: cfe/trunk/lib/Headers/avx512fintrin.h =================================================================== --- cfe/trunk/lib/Headers/avx512fintrin.h +++ cfe/trunk/lib/Headers/avx512fintrin.h @@ -569,7 +569,7 @@ 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, @@ -578,7 +578,7 @@ 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, @@ -586,20 +586,20 @@ } #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, @@ -608,7 +608,7 @@ 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, @@ -616,15 +616,15 @@ } #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 @@ -692,7 +692,7 @@ 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, @@ -701,7 +701,7 @@ 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, @@ -709,20 +709,20 @@ } #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, @@ -731,7 +731,7 @@ 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, @@ -739,15 +739,15 @@ } #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 @@ -909,7 +909,7 @@ 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 (), @@ -919,7 +919,7 @@ 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 (), @@ -946,7 +946,7 @@ 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 (), @@ -956,7 +956,7 @@ 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 (), @@ -1019,7 +1019,7 @@ 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, @@ -1028,7 +1028,7 @@ 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, @@ -1036,20 +1036,20 @@ } #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, @@ -1058,22 +1058,22 @@ 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 @@ -1138,7 +1138,7 @@ 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, @@ -1147,27 +1147,27 @@ 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, @@ -1176,7 +1176,7 @@ 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, @@ -1184,15 +1184,15 @@ } #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 @@ -1259,7 +1259,7 @@ 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, @@ -1268,27 +1268,27 @@ 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, @@ -1297,7 +1297,7 @@ 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, @@ -1305,15 +1305,15 @@ } #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 @@ -1380,7 +1380,7 @@ 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, @@ -1389,7 +1389,7 @@ 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, @@ -1397,20 +1397,20 @@ } #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, @@ -1419,7 +1419,7 @@ 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, @@ -1427,15 +1427,15 @@ } #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