Author: Nikita Popov Date: 2025-08-13T22:34:47+02:00 New Revision: d5069f8aaefd251fdc101440b72013d9b5280f21
URL: https://github.com/llvm/llvm-project/commit/d5069f8aaefd251fdc101440b72013d9b5280f21 DIFF: https://github.com/llvm/llvm-project/commit/d5069f8aaefd251fdc101440b72013d9b5280f21.diff LOG: Revert "[Headers][X86] Allow SSE/AVX/AVX512 unpck intrinsics to be used in co…" This reverts commit be5554d6107aef4793088569ec5bdad6b58d8b5b. Added: Modified: clang/lib/Headers/avx2intrin.h clang/lib/Headers/avx512bwintrin.h clang/lib/Headers/avx512fintrin.h clang/lib/Headers/emmintrin.h clang/test/CodeGen/X86/avx2-builtins.c clang/test/CodeGen/X86/avx512bw-builtins.c clang/test/CodeGen/X86/avx512f-builtins.c clang/test/CodeGen/X86/sse2-builtins.c Removed: ################################################################################ diff --git a/clang/lib/Headers/avx2intrin.h b/clang/lib/Headers/avx2intrin.h index 48b3cfacc0bb7..35258d47c7188 100644 --- a/clang/lib/Headers/avx2intrin.h +++ b/clang/lib/Headers/avx2intrin.h @@ -2712,8 +2712,9 @@ _mm256_subs_epu16(__m256i __a, __m256i __b) { /// A 256-bit integer vector used as the source for the odd-numbered bytes /// of the result. /// \returns A 256-bit integer vector containing the result. -static __inline__ __m256i __DEFAULT_FN_ATTRS256_CONSTEXPR -_mm256_unpackhi_epi8(__m256i __a, __m256i __b) { +static __inline__ __m256i __DEFAULT_FN_ATTRS256 +_mm256_unpackhi_epi8(__m256i __a, __m256i __b) +{ return (__m256i)__builtin_shufflevector((__v32qi)__a, (__v32qi)__b, 8, 32+8, 9, 32+9, 10, 32+10, 11, 32+11, 12, 32+12, 13, 32+13, 14, 32+14, 15, 32+15, 24, 32+24, 25, 32+25, 26, 32+26, 27, 32+27, 28, 32+28, 29, 32+29, 30, 32+30, 31, 32+31); } @@ -2746,8 +2747,9 @@ _mm256_unpackhi_epi8(__m256i __a, __m256i __b) { /// A 256-bit vector of [16 x i16] used as the source for the odd-numbered /// elements of the result. /// \returns A 256-bit vector of [16 x i16] containing the result. -static __inline__ __m256i __DEFAULT_FN_ATTRS256_CONSTEXPR -_mm256_unpackhi_epi16(__m256i __a, __m256i __b) { +static __inline__ __m256i __DEFAULT_FN_ATTRS256 +_mm256_unpackhi_epi16(__m256i __a, __m256i __b) +{ return (__m256i)__builtin_shufflevector((__v16hi)__a, (__v16hi)__b, 4, 16+4, 5, 16+5, 6, 16+6, 7, 16+7, 12, 16+12, 13, 16+13, 14, 16+14, 15, 16+15); } @@ -2779,8 +2781,9 @@ _mm256_unpackhi_epi16(__m256i __a, __m256i __b) { /// A 256-bit vector of [8 x i32] used as the source for the odd-numbered /// elements of the result. /// \returns A 256-bit vector of [8 x i32] containing the result. -static __inline__ __m256i __DEFAULT_FN_ATTRS256_CONSTEXPR -_mm256_unpackhi_epi32(__m256i __a, __m256i __b) { +static __inline__ __m256i __DEFAULT_FN_ATTRS256 +_mm256_unpackhi_epi32(__m256i __a, __m256i __b) +{ return (__m256i)__builtin_shufflevector((__v8si)__a, (__v8si)__b, 2, 8+2, 3, 8+3, 6, 8+6, 7, 8+7); } @@ -2808,8 +2811,9 @@ _mm256_unpackhi_epi32(__m256i __a, __m256i __b) { /// A 256-bit vector of [4 x i64] used as the source for the odd-numbered /// elements of the result. /// \returns A 256-bit vector of [4 x i64] containing the result. -static __inline__ __m256i __DEFAULT_FN_ATTRS256_CONSTEXPR -_mm256_unpackhi_epi64(__m256i __a, __m256i __b) { +static __inline__ __m256i __DEFAULT_FN_ATTRS256 +_mm256_unpackhi_epi64(__m256i __a, __m256i __b) +{ return (__m256i)__builtin_shufflevector((__v4di)__a, (__v4di)__b, 1, 4+1, 3, 4+3); } @@ -2841,8 +2845,9 @@ _mm256_unpackhi_epi64(__m256i __a, __m256i __b) { /// A 256-bit integer vector used as the source for the odd-numbered bytes /// of the result. /// \returns A 256-bit integer vector containing the result. -static __inline__ __m256i __DEFAULT_FN_ATTRS256_CONSTEXPR -_mm256_unpacklo_epi8(__m256i __a, __m256i __b) { +static __inline__ __m256i __DEFAULT_FN_ATTRS256 +_mm256_unpacklo_epi8(__m256i __a, __m256i __b) +{ return (__m256i)__builtin_shufflevector((__v32qi)__a, (__v32qi)__b, 0, 32+0, 1, 32+1, 2, 32+2, 3, 32+3, 4, 32+4, 5, 32+5, 6, 32+6, 7, 32+7, 16, 32+16, 17, 32+17, 18, 32+18, 19, 32+19, 20, 32+20, 21, 32+21, 22, 32+22, 23, 32+23); } @@ -2875,8 +2880,9 @@ _mm256_unpacklo_epi8(__m256i __a, __m256i __b) { /// A 256-bit vector of [16 x i16] used as the source for the odd-numbered /// elements of the result. /// \returns A 256-bit vector of [16 x i16] containing the result. -static __inline__ __m256i __DEFAULT_FN_ATTRS256_CONSTEXPR -_mm256_unpacklo_epi16(__m256i __a, __m256i __b) { +static __inline__ __m256i __DEFAULT_FN_ATTRS256 +_mm256_unpacklo_epi16(__m256i __a, __m256i __b) +{ return (__m256i)__builtin_shufflevector((__v16hi)__a, (__v16hi)__b, 0, 16+0, 1, 16+1, 2, 16+2, 3, 16+3, 8, 16+8, 9, 16+9, 10, 16+10, 11, 16+11); } @@ -2908,8 +2914,9 @@ _mm256_unpacklo_epi16(__m256i __a, __m256i __b) { /// A 256-bit vector of [8 x i32] used as the source for the odd-numbered /// elements of the result. /// \returns A 256-bit vector of [8 x i32] containing the result. -static __inline__ __m256i __DEFAULT_FN_ATTRS256_CONSTEXPR -_mm256_unpacklo_epi32(__m256i __a, __m256i __b) { +static __inline__ __m256i __DEFAULT_FN_ATTRS256 +_mm256_unpacklo_epi32(__m256i __a, __m256i __b) +{ return (__m256i)__builtin_shufflevector((__v8si)__a, (__v8si)__b, 0, 8+0, 1, 8+1, 4, 8+4, 5, 8+5); } @@ -2937,8 +2944,9 @@ _mm256_unpacklo_epi32(__m256i __a, __m256i __b) { /// A 256-bit vector of [4 x i64] used as the source for the odd-numbered /// elements of the result. /// \returns A 256-bit vector of [4 x i64] containing the result. -static __inline__ __m256i __DEFAULT_FN_ATTRS256_CONSTEXPR -_mm256_unpacklo_epi64(__m256i __a, __m256i __b) { +static __inline__ __m256i __DEFAULT_FN_ATTRS256 +_mm256_unpacklo_epi64(__m256i __a, __m256i __b) +{ return (__m256i)__builtin_shufflevector((__v4di)__a, (__v4di)__b, 0, 4+0, 2, 4+2); } diff --git a/clang/lib/Headers/avx512bwintrin.h b/clang/lib/Headers/avx512bwintrin.h index ae644b3d4bad2..886783230cf32 100644 --- a/clang/lib/Headers/avx512bwintrin.h +++ b/clang/lib/Headers/avx512bwintrin.h @@ -1247,7 +1247,7 @@ _mm512_mask_cvtusepi16_storeu_epi8 (void * __P, __mmask32 __M, __m512i __A) __builtin_ia32_pmovuswb512mem_mask ((__v32qi *) __P, (__v32hi) __A, __M); } -static __inline__ __m512i __DEFAULT_FN_ATTRS512_CONSTEXPR +static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_unpackhi_epi8(__m512i __A, __m512i __B) { return (__m512i)__builtin_shufflevector((__v64qi)__A, (__v64qi)__B, 8, 64+8, 9, 64+9, @@ -1282,7 +1282,7 @@ _mm512_maskz_unpackhi_epi8(__mmask64 __U, __m512i __A, __m512i __B) { (__v64qi)_mm512_setzero_si512()); } -static __inline__ __m512i __DEFAULT_FN_ATTRS512_CONSTEXPR +static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_unpackhi_epi16(__m512i __A, __m512i __B) { return (__m512i)__builtin_shufflevector((__v32hi)__A, (__v32hi)__B, 4, 32+4, 5, 32+5, @@ -1309,7 +1309,7 @@ _mm512_maskz_unpackhi_epi16(__mmask32 __U, __m512i __A, __m512i __B) { (__v32hi)_mm512_setzero_si512()); } -static __inline__ __m512i __DEFAULT_FN_ATTRS512_CONSTEXPR +static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_unpacklo_epi8(__m512i __A, __m512i __B) { return (__m512i)__builtin_shufflevector((__v64qi)__A, (__v64qi)__B, 0, 64+0, 1, 64+1, @@ -1344,7 +1344,7 @@ _mm512_maskz_unpacklo_epi8(__mmask64 __U, __m512i __A, __m512i __B) { (__v64qi)_mm512_setzero_si512()); } -static __inline__ __m512i __DEFAULT_FN_ATTRS512_CONSTEXPR +static __inline__ __m512i __DEFAULT_FN_ATTRS512 _mm512_unpacklo_epi16(__m512i __A, __m512i __B) { return (__m512i)__builtin_shufflevector((__v32hi)__A, (__v32hi)__B, 0, 32+0, 1, 32+1, diff --git a/clang/lib/Headers/avx512fintrin.h b/clang/lib/Headers/avx512fintrin.h index 634fb78696f04..90f883ba9f770 100644 --- a/clang/lib/Headers/avx512fintrin.h +++ b/clang/lib/Headers/avx512fintrin.h @@ -4203,8 +4203,9 @@ _mm512_maskz_unpacklo_ps (__mmask16 __U, __m512 __A, __m512 __B) (__v16sf)_mm512_setzero_ps()); } -static __inline__ __m512i __DEFAULT_FN_ATTRS512_CONSTEXPR -_mm512_unpackhi_epi32(__m512i __A, __m512i __B) { +static __inline__ __m512i __DEFAULT_FN_ATTRS512 +_mm512_unpackhi_epi32(__m512i __A, __m512i __B) +{ return (__m512i)__builtin_shufflevector((__v16si)__A, (__v16si)__B, 2, 18, 3, 19, 2+4, 18+4, 3+4, 19+4, @@ -4228,8 +4229,9 @@ _mm512_maskz_unpackhi_epi32(__mmask16 __U, __m512i __A, __m512i __B) (__v16si)_mm512_setzero_si512()); } -static __inline__ __m512i __DEFAULT_FN_ATTRS512_CONSTEXPR -_mm512_unpacklo_epi32(__m512i __A, __m512i __B) { +static __inline__ __m512i __DEFAULT_FN_ATTRS512 +_mm512_unpacklo_epi32(__m512i __A, __m512i __B) +{ return (__m512i)__builtin_shufflevector((__v16si)__A, (__v16si)__B, 0, 16, 1, 17, 0+4, 16+4, 1+4, 17+4, @@ -4253,8 +4255,9 @@ _mm512_maskz_unpacklo_epi32(__mmask16 __U, __m512i __A, __m512i __B) (__v16si)_mm512_setzero_si512()); } -static __inline__ __m512i __DEFAULT_FN_ATTRS512_CONSTEXPR -_mm512_unpackhi_epi64(__m512i __A, __m512i __B) { +static __inline__ __m512i __DEFAULT_FN_ATTRS512 +_mm512_unpackhi_epi64(__m512i __A, __m512i __B) +{ return (__m512i)__builtin_shufflevector((__v8di)__A, (__v8di)__B, 1, 9, 1+2, 9+2, 1+4, 9+4, 1+6, 9+6); } @@ -4275,8 +4278,9 @@ _mm512_maskz_unpackhi_epi64(__mmask8 __U, __m512i __A, __m512i __B) (__v8di)_mm512_setzero_si512()); } -static __inline__ __m512i __DEFAULT_FN_ATTRS512_CONSTEXPR -_mm512_unpacklo_epi64(__m512i __A, __m512i __B) { +static __inline__ __m512i __DEFAULT_FN_ATTRS512 +_mm512_unpacklo_epi64 (__m512i __A, __m512i __B) +{ return (__m512i)__builtin_shufflevector((__v8di)__A, (__v8di)__B, 0, 8, 0+2, 8+2, 0+4, 8+4, 0+6, 8+6); } diff --git a/clang/lib/Headers/emmintrin.h b/clang/lib/Headers/emmintrin.h index 58caded7430bc..1973ccb0e00bf 100644 --- a/clang/lib/Headers/emmintrin.h +++ b/clang/lib/Headers/emmintrin.h @@ -4421,8 +4421,8 @@ static __inline__ int __DEFAULT_FN_ATTRS _mm_movemask_epi8(__m128i __a) { /// Bits [119:112] are written to bits [111:104] of the result. \n /// Bits [127:120] are written to bits [127:120] of the result. /// \returns A 128-bit vector of [16 x i8] containing the interleaved values. -static __inline__ __m128i __DEFAULT_FN_ATTRS_CONSTEXPR -_mm_unpackhi_epi8(__m128i __a, __m128i __b) { +static __inline__ __m128i __DEFAULT_FN_ATTRS _mm_unpackhi_epi8(__m128i __a, + __m128i __b) { return (__m128i)__builtin_shufflevector( (__v16qi)__a, (__v16qi)__b, 8, 16 + 8, 9, 16 + 9, 10, 16 + 10, 11, 16 + 11, 12, 16 + 12, 13, 16 + 13, 14, 16 + 14, 15, 16 + 15); @@ -4449,8 +4449,8 @@ _mm_unpackhi_epi8(__m128i __a, __m128i __b) { /// Bits [111:96] are written to bits [95:80] of the result. \n /// Bits [127:112] are written to bits [127:112] of the result. /// \returns A 128-bit vector of [8 x i16] containing the interleaved values. -static __inline__ __m128i __DEFAULT_FN_ATTRS_CONSTEXPR -_mm_unpackhi_epi16(__m128i __a, __m128i __b) { +static __inline__ __m128i __DEFAULT_FN_ATTRS _mm_unpackhi_epi16(__m128i __a, + __m128i __b) { return (__m128i)__builtin_shufflevector((__v8hi)__a, (__v8hi)__b, 4, 8 + 4, 5, 8 + 5, 6, 8 + 6, 7, 8 + 7); } @@ -4472,8 +4472,8 @@ _mm_unpackhi_epi16(__m128i __a, __m128i __b) { /// Bits [95:64] are written to bits [64:32] of the destination. \n /// Bits [127:96] are written to bits [127:96] of the destination. /// \returns A 128-bit vector of [4 x i32] containing the interleaved values. -static __inline__ __m128i __DEFAULT_FN_ATTRS_CONSTEXPR -_mm_unpackhi_epi32(__m128i __a, __m128i __b) { +static __inline__ __m128i __DEFAULT_FN_ATTRS _mm_unpackhi_epi32(__m128i __a, + __m128i __b) { return (__m128i)__builtin_shufflevector((__v4si)__a, (__v4si)__b, 2, 4 + 2, 3, 4 + 3); } @@ -4493,8 +4493,8 @@ _mm_unpackhi_epi32(__m128i __a, __m128i __b) { /// A 128-bit vector of [2 x i64]. \n /// Bits [127:64] are written to bits [127:64] of the destination. /// \returns A 128-bit vector of [2 x i64] containing the interleaved values. -static __inline__ __m128i __DEFAULT_FN_ATTRS_CONSTEXPR -_mm_unpackhi_epi64(__m128i __a, __m128i __b) { +static __inline__ __m128i __DEFAULT_FN_ATTRS _mm_unpackhi_epi64(__m128i __a, + __m128i __b) { return (__m128i)__builtin_shufflevector((__v2di)__a, (__v2di)__b, 1, 2 + 1); } @@ -4527,8 +4527,8 @@ _mm_unpackhi_epi64(__m128i __a, __m128i __b) { /// Bits [55:48] are written to bits [111:104] of the result. \n /// Bits [63:56] are written to bits [127:120] of the result. /// \returns A 128-bit vector of [16 x i8] containing the interleaved values. -static __inline__ __m128i __DEFAULT_FN_ATTRS_CONSTEXPR -_mm_unpacklo_epi8(__m128i __a, __m128i __b) { +static __inline__ __m128i __DEFAULT_FN_ATTRS _mm_unpacklo_epi8(__m128i __a, + __m128i __b) { return (__m128i)__builtin_shufflevector( (__v16qi)__a, (__v16qi)__b, 0, 16 + 0, 1, 16 + 1, 2, 16 + 2, 3, 16 + 3, 4, 16 + 4, 5, 16 + 5, 6, 16 + 6, 7, 16 + 7); @@ -4556,8 +4556,8 @@ _mm_unpacklo_epi8(__m128i __a, __m128i __b) { /// Bits [47:32] are written to bits [95:80] of the result. \n /// Bits [63:48] are written to bits [127:112] of the result. /// \returns A 128-bit vector of [8 x i16] containing the interleaved values. -static __inline__ __m128i __DEFAULT_FN_ATTRS_CONSTEXPR -_mm_unpacklo_epi16(__m128i __a, __m128i __b) { +static __inline__ __m128i __DEFAULT_FN_ATTRS _mm_unpacklo_epi16(__m128i __a, + __m128i __b) { return (__m128i)__builtin_shufflevector((__v8hi)__a, (__v8hi)__b, 0, 8 + 0, 1, 8 + 1, 2, 8 + 2, 3, 8 + 3); } @@ -4579,8 +4579,8 @@ _mm_unpacklo_epi16(__m128i __a, __m128i __b) { /// Bits [31:0] are written to bits [64:32] of the destination. \n /// Bits [63:32] are written to bits [127:96] of the destination. /// \returns A 128-bit vector of [4 x i32] containing the interleaved values. -static __inline__ __m128i __DEFAULT_FN_ATTRS_CONSTEXPR -_mm_unpacklo_epi32(__m128i __a, __m128i __b) { +static __inline__ __m128i __DEFAULT_FN_ATTRS _mm_unpacklo_epi32(__m128i __a, + __m128i __b) { return (__m128i)__builtin_shufflevector((__v4si)__a, (__v4si)__b, 0, 4 + 0, 1, 4 + 1); } @@ -4600,8 +4600,8 @@ _mm_unpacklo_epi32(__m128i __a, __m128i __b) { /// A 128-bit vector of [2 x i64]. \n /// Bits [63:0] are written to bits [127:64] of the destination. \n /// \returns A 128-bit vector of [2 x i64] containing the interleaved values. -static __inline__ __m128i __DEFAULT_FN_ATTRS_CONSTEXPR -_mm_unpacklo_epi64(__m128i __a, __m128i __b) { +static __inline__ __m128i __DEFAULT_FN_ATTRS _mm_unpacklo_epi64(__m128i __a, + __m128i __b) { return (__m128i)__builtin_shufflevector((__v2di)__a, (__v2di)__b, 0, 2 + 0); } diff --git a/clang/test/CodeGen/X86/avx2-builtins.c b/clang/test/CodeGen/X86/avx2-builtins.c index 8e591397f17c9..76d26d57cfecf 100644 --- a/clang/test/CodeGen/X86/avx2-builtins.c +++ b/clang/test/CodeGen/X86/avx2-builtins.c @@ -1327,56 +1327,48 @@ __m256i test_mm256_unpackhi_epi8(__m256i a, __m256i b) { // CHECK: shufflevector <32 x i8> %{{.*}}, <32 x i8> %{{.*}}, <32 x i32> <i32 8, i32 40, i32 9, i32 41, i32 10, i32 42, i32 11, i32 43, i32 12, i32 44, i32 13, i32 45, i32 14, i32 46, i32 15, i32 47, i32 24, i32 56, i32 25, i32 57, i32 26, i32 58, i32 27, i32 59, i32 28, i32 60, i32 29, i32 61, i32 30, i32 62, i32 31, i32 63> return _mm256_unpackhi_epi8(a, b); } -TEST_CONSTEXPR(match_v32qi(_mm256_unpackhi_epi8((__m256i)(__v32qi){0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31}, (__m256i)(__v32qi){32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, 62, 63}), 8, 40, 9, 41, 10, 42, 11, 43, 12, 44, 13, 45, 14, 46, 15, 47, 24, 56, 25, 57, 26, 58, 27, 59, 28, 60, 29, 61, 30, 62, 31, 63)); __m256i test_mm256_unpackhi_epi16(__m256i a, __m256i b) { // CHECK-LABEL: test_mm256_unpackhi_epi16 // CHECK: shufflevector <16 x i16> %{{.*}}, <16 x i16> %{{.*}}, <16 x i32> <i32 4, i32 20, i32 5, i32 21, i32 6, i32 22, i32 7, i32 23, i32 12, i32 28, i32 13, i32 29, i32 14, i32 30, i32 15, i32 31> return _mm256_unpackhi_epi16(a, b); } -TEST_CONSTEXPR(match_v16hi(_mm256_unpackhi_epi16((__m256i)(__v16hi){0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15}, (__m256i)(__v16hi){16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31}), 4, 20, 5, 21, 6, 22, 7, 23, 12, 28, 13, 29, 14, 30, 15, 31)); __m256i test_mm256_unpackhi_epi32(__m256i a, __m256i b) { // CHECK-LABEL: test_mm256_unpackhi_epi32 // CHECK: shufflevector <8 x i32> %{{.*}}, <8 x i32> %{{.*}}, <8 x i32> <i32 2, i32 10, i32 3, i32 11, i32 6, i32 14, i32 7, i32 15> return _mm256_unpackhi_epi32(a, b); } -TEST_CONSTEXPR(match_v8si(_mm256_unpackhi_epi32((__m256i)(__v8si){0, 1, 2, 3, 4, 5, 6, 7}, (__m256i)(__v8si){8, 9, 10, 11, 12, 13, 14, 15}), 2, 10, 3, 11, 6, 14, 7, 15)); __m256i test_mm256_unpackhi_epi64(__m256i a, __m256i b) { // CHECK-LABEL: test_mm256_unpackhi_epi64 // CHECK: shufflevector <4 x i64> %{{.*}}, <4 x i64> %{{.*}}, <4 x i32> <i32 1, i32 5, i32 3, i32 7> return _mm256_unpackhi_epi64(a, b); } -TEST_CONSTEXPR(match_v4di(_mm256_unpackhi_epi64((__m256i)(__v4di){0, 1, 2, 3}, (__m256i)(__v4di){ 4, 5, 6, 7}), 1, 5, 3, 7)); __m256i test_mm256_unpacklo_epi8(__m256i a, __m256i b) { // CHECK-LABEL: test_mm256_unpacklo_epi8 // CHECK: shufflevector <32 x i8> %{{.*}}, <32 x i8> %{{.*}}, <32 x i32> <i32 0, i32 32, i32 1, i32 33, i32 2, i32 34, i32 3, i32 35, i32 4, i32 36, i32 5, i32 37, i32 6, i32 38, i32 7, i32 39, i32 16, i32 48, i32 17, i32 49, i32 18, i32 50, i32 19, i32 51, i32 20, i32 52, i32 21, i32 53, i32 22, i32 54, i32 23, i32 55> return _mm256_unpacklo_epi8(a, b); } -TEST_CONSTEXPR(match_v32qi(_mm256_unpacklo_epi8((__m256i)(__v32qi){0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31}, (__m256i)(__v32qi){32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, 62, 63}), 0, 32, 1, 33, 2, 34, 3, 35, 4, 36, 5, 37, 6, 38, 7, 39, 16, 48, 17, 49, 18, 50, 19, 51, 20, 52, 21, 53, 22, 54, 23, 55)); __m256i test_mm256_unpacklo_epi16(__m256i a, __m256i b) { // CHECK-LABEL: test_mm256_unpacklo_epi16 // CHECK: shufflevector <16 x i16> %{{.*}}, <16 x i16> %{{.*}}, <16 x i32> <i32 0, i32 16, i32 1, i32 17, i32 2, i32 18, i32 3, i32 19, i32 8, i32 24, i32 9, i32 25, i32 10, i32 26, i32 11, i32 27> return _mm256_unpacklo_epi16(a, b); } -TEST_CONSTEXPR(match_v16hi(_mm256_unpacklo_epi16((__m256i)(__v16hi){0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15}, (__m256i)(__v16hi){16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31}), 0, 16, 1, 17, 2, 18, 3, 19, 8, 24, 9, 25, 10, 26, 11, 27)); __m256i test_mm256_unpacklo_epi32(__m256i a, __m256i b) { // CHECK-LABEL: test_mm256_unpacklo_epi32 // CHECK: shufflevector <8 x i32> %{{.*}}, <8 x i32> %{{.*}}, <8 x i32> <i32 0, i32 8, i32 1, i32 9, i32 4, i32 12, i32 5, i32 13> return _mm256_unpacklo_epi32(a, b); } -TEST_CONSTEXPR(match_v8si(_mm256_unpacklo_epi32((__m256i)(__v8si){0, 1, 2, 3, 4, 5, 6, 7}, (__m256i)(__v8si){ 8, 9, 10, 11, 12, 13, 14, 15}), 0, 8, 1, 9, 4, 12, 5, 13)); __m256i test_mm256_unpacklo_epi64(__m256i a, __m256i b) { // CHECK-LABEL: test_mm256_unpacklo_epi64 // CHECK: shufflevector <4 x i64> %{{.*}}, <4 x i64> %{{.*}}, <4 x i32> <i32 0, i32 4, i32 2, i32 6> return _mm256_unpacklo_epi64(a, b); } -TEST_CONSTEXPR(match_v4di(_mm256_unpacklo_epi64((__m256i)(__v4di){0, 1, 2, 3}, (__m256i)(__v4di){ 4, 5, 6, 7}), 0, 4, 2, 6)); __m256i test_mm256_xor_si256(__m256i a, __m256i b) { // CHECK-LABEL: test_mm256_xor_si256 diff --git a/clang/test/CodeGen/X86/avx512bw-builtins.c b/clang/test/CodeGen/X86/avx512bw-builtins.c index 16f43e31241be..af9378040f79d 100644 --- a/clang/test/CodeGen/X86/avx512bw-builtins.c +++ b/clang/test/CodeGen/X86/avx512bw-builtins.c @@ -1482,7 +1482,6 @@ __m512i test_mm512_unpackhi_epi8(__m512i __A, __m512i __B) { // CHECK: shufflevector <64 x i8> %{{.*}}, <64 x i8> %{{.*}}, <64 x i32> <i32 8, i32 72, i32 9, i32 73, i32 10, i32 74, i32 11, i32 75, i32 12, i32 76, i32 13, i32 77, i32 14, i32 78, i32 15, i32 79, i32 24, i32 88, i32 25, i32 89, i32 26, i32 90, i32 27, i32 91, i32 28, i32 92, i32 29, i32 93, i32 30, i32 94, i32 31, i32 95, i32 40, i32 104, i32 41, i32 105, i32 42, i32 106, i32 43, i32 107, i32 44, i32 108, i32 45, i32 109, i32 46, i32 110, i32 47, i32 111, i32 56, i32 120, i32 57, i32 121, i32 58, i32 122, i32 59, i32 123, i32 60, i32 124, i32 61, i32 125, i32 62, i32 126, i32 63, i32 127> return _mm512_unpackhi_epi8(__A, __B); } -TEST_CONSTEXPR(match_m128i(_mm512_unpackhi_epi8((__m512)(__v64qi){0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, 62, 63}, (__m512)(__v64qi){64, 65, 66, 67, 68, 69, 70, 71, 72, 73, 74, 75, 76, 77, 78, 79, 80, 81, 82, 83, 84, 85, 86, 87, 88, 89, 90, 91, 92, 93, 94, 95, 96, 97, 98, 99, 100, 101, 102, 103, 104, 105, 106, 107, 108, 109, 110, 111, 112, 113, 114, 115, 116, 117, 118, 119, 120, 121, 122, 123, 124, 125, 126, 127}), 8, 72, 9, 73, 10, 74, 11, 75, 12, 76, 13, 77, 14, 78, 15, 79, 24, 88, 25, 89, 26, 90, 27, 91, 28, 92, 29, 93, 30, 94, 31, 95, 40, 104, 41, 105, 42, 106, 43, 107, 44, 108, 45, 109, 46, 110, 47, 111, 56, 120, 57, 121, 58, 122, 59, 123, 60, 124, 61, 125, 62, 126, 63, 127)); __m512i test_mm512_mask_unpackhi_epi8(__m512i __W, __mmask64 __U, __m512i __A, __m512i __B) { // CHECK-LABEL: test_mm512_mask_unpackhi_epi8 @@ -1503,8 +1502,6 @@ __m512i test_mm512_unpackhi_epi16(__m512i __A, __m512i __B) { // CHECK: shufflevector <32 x i16> %{{.*}}, <32 x i16> %{{.*}}, <32 x i32> <i32 4, i32 36, i32 5, i32 37, i32 6, i32 38, i32 7, i32 39, i32 12, i32 44, i32 13, i32 45, i32 14, i32 46, i32 15, i32 47, i32 20, i32 52, i32 21, i32 53, i32 22, i32 54, i32 23, i32 55, i32 28, i32 60, i32 29, i32 61, i32 30, i32 62, i32 31, i32 63> return _mm512_unpackhi_epi16(__A, __B); } -TEST_CONSTEXPR(match_v64qi(_mm512_unpackhi_epi16((__m256i)(__v32qi){0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31}, (__m256i)(__v32qi){32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, 62, 63}), 4, 36, 5, 37, 6, 38, 7, 39, 12, 44, 13, 45, 14, 46, 15, 47, 20, 52, 21, 53, 22, 54, 23, 55, 28, 60, 29, 61, 30, 62, 31, 63)); - __m512i test_mm512_mask_unpackhi_epi16(__m512i __W, __mmask32 __U, __m512i __A, __m512i __B) { // CHECK-LABEL: test_mm512_mask_unpackhi_epi16 @@ -1525,7 +1522,6 @@ __m512i test_mm512_unpacklo_epi8(__m512i __A, __m512i __B) { // CHECK: shufflevector <64 x i8> %{{.*}}, <64 x i8> %{{.*}}, <64 x i32> <i32 0, i32 64, i32 1, i32 65, i32 2, i32 66, i32 3, i32 67, i32 4, i32 68, i32 5, i32 69, i32 6, i32 70, i32 7, i32 71, i32 16, i32 80, i32 17, i32 81, i32 18, i32 82, i32 19, i32 83, i32 20, i32 84, i32 21, i32 85, i32 22, i32 86, i32 23, i32 87, i32 32, i32 96, i32 33, i32 97, i32 34, i32 98, i32 35, i32 99, i32 36, i32 100, i32 37, i32 101, i32 38, i32 102, i32 39, i32 103, i32 48, i32 112, i32 49, i32 113, i32 50, i32 114, i32 51, i32 115, i32 52, i32 116, i32 53, i32 117, i32 54, i32 118, i32 55, i32 119> return _mm512_unpacklo_epi8(__A, __B); } -TEST_CONSTEXPR(match_m128i(_mm512_unpacklo_epi8((__m512)(__v64qi){0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, 62, 63}, (__m512)(__v64qi){64, 65, 66, 67, 68, 69, 70, 71, 72, 73, 74, 75, 76, 77, 78, 79, 80, 81, 82, 83, 84, 85, 86, 87, 88, 89, 90, 91, 92, 93, 94, 95, 96, 97, 98, 99, 100, 101, 102, 103, 104, 105, 106, 107, 108, 109, 110, 111, 112, 113, 114, 115, 116, 117, 118, 119, 120, 121, 122, 123, 124, 125, 126, 127}), 0, 64, 1, 65, 2, 66, 3, 67, 4, 68, 5, 69, 6, 70, 7, 71, 16, 80, 17, 81, 18, 82, 19, 83, 20, 84, 21, 85, 22, 86, 23, 87, 32, 96, 33, 97, 34, 98, 35, 99, 36, 100, 37, 101, 38, 102, 39, 103, 48, 112, 49, 113, 50, 114, 51, 115, 52, 116, 53, 117, 54, 118, 55, 119)); __m512i test_mm512_mask_unpacklo_epi8(__m512i __W, __mmask64 __U, __m512i __A, __m512i __B) { // CHECK-LABEL: test_mm512_mask_unpacklo_epi8 @@ -1546,7 +1542,6 @@ __m512i test_mm512_unpacklo_epi16(__m512i __A, __m512i __B) { // CHECK: shufflevector <32 x i16> %{{.*}}, <32 x i16> %{{.*}}, <32 x i32> <i32 0, i32 32, i32 1, i32 33, i32 2, i32 34, i32 3, i32 35, i32 8, i32 40, i32 9, i32 41, i32 10, i32 42, i32 11, i32 43, i32 16, i32 48, i32 17, i32 49, i32 18, i32 50, i32 19, i32 51, i32 24, i32 56, i32 25, i32 57, i32 26, i32 58, i32 27, i32 59> return _mm512_unpacklo_epi16(__A, __B); } -TEST_CONSTEXPR(match_v64qi(_mm512_unpacklo_epi16((__m256i)(__v32qi){0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31}, (__m256i)(__v32qi){32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, 62, 63}), 0, 32, 1, 33, 2, 34, 3, 35, 8, 40, 9, 41, 10, 42, 11, 43, 16, 48, 17, 49, 18, 50, 19, 51, 24, 56, 25, 57, 26, 58, 27, 59)); __m512i test_mm512_mask_unpacklo_epi16(__m512i __W, __mmask32 __U, __m512i __A, __m512i __B) { // CHECK-LABEL: test_mm512_mask_unpacklo_epi16 diff --git a/clang/test/CodeGen/X86/avx512f-builtins.c b/clang/test/CodeGen/X86/avx512f-builtins.c index a92227d5aaae1..048bc3057b5f4 100644 --- a/clang/test/CodeGen/X86/avx512f-builtins.c +++ b/clang/test/CodeGen/X86/avx512f-builtins.c @@ -4691,7 +4691,6 @@ __m512i test_mm512_unpackhi_epi32(__m512i __A, __m512i __B) { // CHECK: shufflevector <16 x i32> %{{.*}}, <16 x i32> %{{.*}}, <16 x i32> <i32 2, i32 18, i32 3, i32 19, i32 6, i32 22, i32 7, i32 23, i32 10, i32 26, i32 11, i32 27, i32 14, i32 30, i32 15, i32 31> return _mm512_unpackhi_epi32(__A, __B); } -TEST_CONSTEXPR(match_v16si(_mm512_unpackhi_epi32((__m512i)(__v16si){0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15}, (__m512i)(__v16si){16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31}), 2, 18, 3, 19, 6, 22, 7, 23, 10, 26, 11, 27, 14, 30, 15, 31)); __m512d test_mm512_maskz_unpackhi_pd(__mmask8 __U, __m512d __A, __m512d __B) { // CHECK-LABEL: test_mm512_maskz_unpackhi_pd @@ -5478,7 +5477,6 @@ __m512i test_mm512_unpackhi_epi64(__m512i __A, __m512i __B) { // CHECK: shufflevector <8 x i64> %{{.*}}, <8 x i64> %{{.*}}, <8 x i32> <i32 1, i32 9, i32 3, i32 11, i32 5, i32 13, i32 7, i32 15> return _mm512_unpackhi_epi64(__A, __B); } -TEST_CONSTEXPR(match_m512i(_mm512_unpackhi_epi64((__m512i){0, 1, 2, 3, 4, 5, 6, 7}, (__m512i){8, 9, 10, 11, 12, 13, 14, 15}), 1, 9, 3, 11, 5, 13, 7, 15)); __m512i test_mm512_mask_unpackhi_epi64(__m512i __W, __mmask8 __U, __m512i __A, __m512i __B) { // CHECK-LABEL: test_mm512_mask_unpackhi_epi64 @@ -5499,7 +5497,6 @@ __m512i test_mm512_unpacklo_epi32(__m512i __A, __m512i __B) { // CHECK: shufflevector <16 x i32> %{{.*}}, <16 x i32> %{{.*}}, <16 x i32> <i32 0, i32 16, i32 1, i32 17, i32 4, i32 20, i32 5, i32 21, i32 8, i32 24, i32 9, i32 25, i32 12, i32 28, i32 13, i32 29> return _mm512_unpacklo_epi32(__A, __B); } -TEST_CONSTEXPR(match_v16si(_mm512_unpacklo_epi32((__m512i)(__v16si){0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15}, (__m512i)(__v16si){16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31}), 0, 16, 1, 17, 4, 20, 5, 21, 8, 24, 9, 25, 12, 28, 13, 29)); __m512i test_mm512_mask_unpacklo_epi32(__m512i __W, __mmask16 __U, __m512i __A, __m512i __B) { // CHECK-LABEL: test_mm512_mask_unpacklo_epi32 @@ -5520,7 +5517,6 @@ __m512i test_mm512_unpacklo_epi64(__m512i __A, __m512i __B) { // CHECK: shufflevector <8 x i64> %{{.*}}, <8 x i64> %{{.*}}, <8 x i32> <i32 0, i32 8, i32 2, i32 10, i32 4, i32 12, i32 6, i32 14> return _mm512_unpacklo_epi64(__A, __B); } -TEST_CONSTEXPR(match_m512i(_mm512_unpacklo_epi64((__m512i){0, 1, 2, 3, 4, 5, 6, 7}, (__m512i){8, 9, 10, 11, 12, 13, 14, 15}), 0, 8, 2, 10, 4, 12, 6, 14)); __m512i test_mm512_mask_unpacklo_epi64(__m512i __W, __mmask8 __U, __m512i __A, __m512i __B) { // CHECK-LABEL: test_mm512_mask_unpacklo_epi64 diff --git a/clang/test/CodeGen/X86/sse2-builtins.c b/clang/test/CodeGen/X86/sse2-builtins.c index 49d8b395a0d5a..d704ee34dd393 100644 --- a/clang/test/CodeGen/X86/sse2-builtins.c +++ b/clang/test/CodeGen/X86/sse2-builtins.c @@ -1780,28 +1780,24 @@ __m128i test_mm_unpackhi_epi8(__m128i A, __m128i B) { // CHECK: shufflevector <16 x i8> %{{.*}}, <16 x i8> %{{.*}}, <16 x i32> <i32 8, i32 24, i32 9, i32 25, i32 10, i32 26, i32 11, i32 27, i32 12, i32 28, i32 13, i32 29, i32 14, i32 30, i32 15, i32 31> return _mm_unpackhi_epi8(A, B); } -TEST_CONSTEXPR(match_v16qi(_mm_unpackhi_epi8((__m128i)(__v16qi){0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15}, (__m128i)(__v16qi){16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31}), 8, 24, 9, 25, 10, 26, 11, 27, 12, 28, 13, 29, 14, 30, 15, 31)); __m128i test_mm_unpackhi_epi16(__m128i A, __m128i B) { // CHECK-LABEL: test_mm_unpackhi_epi16 // CHECK: shufflevector <8 x i16> %{{.*}}, <8 x i16> %{{.*}}, <8 x i32> <i32 4, i32 12, i32 5, i32 13, i32 6, i32 14, i32 7, i32 15> return _mm_unpackhi_epi16(A, B); } -TEST_CONSTEXPR(match_v8hi(_mm_unpackhi_epi16((__m128i)(__v8hi){0, 1, 2, 3, 4, 5, 6, 7}, (__m128i)(__v8hi){8, 9, 10, 11, 12, 13, 14, 15}), 4, 12, 5, 13, 6, 14, 7, 15)); __m128i test_mm_unpackhi_epi32(__m128i A, __m128i B) { // CHECK-LABEL: test_mm_unpackhi_epi32 // CHECK: shufflevector <4 x i32> %{{.*}}, <4 x i32> %{{.*}}, <4 x i32> <i32 2, i32 6, i32 3, i32 7> return _mm_unpackhi_epi32(A, B); } -TEST_CONSTEXPR(match_v4si(_mm_unpackhi_epi32((__m128i)(__v4si){0, 1, 2, 3}, (__m128i)(__v4si){ 4, 5, 6, 7}), 2, 6, 3, 7)); __m128i test_mm_unpackhi_epi64(__m128i A, __m128i B) { // CHECK-LABEL: test_mm_unpackhi_epi64 // CHECK: shufflevector <2 x i64> %{{.*}}, <2 x i64> %{{.*}}, <2 x i32> <i32 1, i32 3> return _mm_unpackhi_epi64(A, B); } -TEST_CONSTEXPR(match_v2di(_mm_unpackhi_epi64((__m128i)(__v2di){0, 1}, (__m128i)(__v2di){2, 3}), 1, 3)); __m128d test_mm_unpackhi_pd(__m128d A, __m128d B) { // CHECK-LABEL: test_mm_unpackhi_pd @@ -1815,28 +1811,24 @@ __m128i test_mm_unpacklo_epi8(__m128i A, __m128i B) { // CHECK: shufflevector <16 x i8> %{{.*}}, <16 x i8> %{{.*}}, <16 x i32> <i32 0, i32 16, i32 1, i32 17, i32 2, i32 18, i32 3, i32 19, i32 4, i32 20, i32 5, i32 21, i32 6, i32 22, i32 7, i32 23> return _mm_unpacklo_epi8(A, B); } -TEST_CONSTEXPR(match_v16qi(_mm_unpacklo_epi8((__m128i)(__v16qi){0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15}, (__m128i)(__v16qi){16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31}), 0, 16, 1, 17, 2, 18, 3, 19, 4, 20, 5, 21, 6, 22, 7, 23)); __m128i test_mm_unpacklo_epi16(__m128i A, __m128i B) { // CHECK-LABEL: test_mm_unpacklo_epi16 // CHECK: shufflevector <8 x i16> %{{.*}}, <8 x i16> %{{.*}}, <8 x i32> <i32 0, i32 8, i32 1, i32 9, i32 2, i32 10, i32 3, i32 11> return _mm_unpacklo_epi16(A, B); } -TEST_CONSTEXPR(match_v8hi(_mm_unpacklo_epi16((__m128i)(__v8hi){0, 1, 2, 3, 4, 5, 6, 7}, (__m128i)(__v8hi){8, 9, 10, 11, 12, 13, 14, 15}), 0, 8, 1, 9, 2, 10, 3, 11)); __m128i test_mm_unpacklo_epi32(__m128i A, __m128i B) { // CHECK-LABEL: test_mm_unpacklo_epi32 // CHECK: shufflevector <4 x i32> %{{.*}}, <4 x i32> %{{.*}}, <4 x i32> <i32 0, i32 4, i32 1, i32 5> return _mm_unpacklo_epi32(A, B); } -TEST_CONSTEXPR(match_v4si(_mm_unpacklo_epi32((__m128i)(__v4si){0, 1, 2, 3}, (__m128i)(__v4si){ 4, 5, 6, 7}), 0, 4, 1, 5)); __m128i test_mm_unpacklo_epi64(__m128i A, __m128i B) { // CHECK-LABEL: test_mm_unpacklo_epi64 // CHECK: shufflevector <2 x i64> %{{.*}}, <2 x i64> %{{.*}}, <2 x i32> <i32 0, i32 2> return _mm_unpacklo_epi64(A, B); } -TEST_CONSTEXPR(match_v2di(_mm_unpacklo_epi64((__m128i)(__v2di){0, 1}, (__m128i)(__v2di){2, 3}), 0, 2)); __m128d test_mm_unpacklo_pd(__m128d A, __m128d B) { // CHECK-LABEL: test_mm_unpacklo_pd _______________________________________________ llvm-branch-commits mailing list llvm-branch-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-branch-commits
