Allow SSE/AVX/AVX512 unpck intrinsics to be used in constexpr

badumbatish · badumbatish · commit 3e462e98944a · 2025-08-11T16:28:07.000-07:00
diff --git a/clang/lib/Headers/avx2intrin.h b/clang/lib/Headers/avx2intrin.h
@@ -2720,9 +2720,8 @@ _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
-_mm256_unpackhi_epi8(__m256i __a, __m256i __b)
-{
+static __inline__ __m256i __DEFAULT_FN_ATTRS256_CONSTEXPR
+_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);
 }
 
@@ -2755,9 +2754,8 @@ _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
-_mm256_unpackhi_epi16(__m256i __a, __m256i __b)
-{
+static __inline__ __m256i __DEFAULT_FN_ATTRS256_CONSTEXPR
+_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);
 }
 
@@ -2789,9 +2787,8 @@ _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
-_mm256_unpackhi_epi32(__m256i __a, __m256i __b)
-{
+static __inline__ __m256i __DEFAULT_FN_ATTRS256_CONSTEXPR
+_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);
 }
 
@@ -2819,9 +2816,8 @@ _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
-_mm256_unpackhi_epi64(__m256i __a, __m256i __b)
-{
+static __inline__ __m256i __DEFAULT_FN_ATTRS256_CONSTEXPR
+_mm256_unpackhi_epi64(__m256i __a, __m256i __b) {
   return (__m256i)__builtin_shufflevector((__v4di)__a, (__v4di)__b, 1, 4+1, 3, 4+3);
 }
 
@@ -2853,9 +2849,8 @@ _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
-_mm256_unpacklo_epi8(__m256i __a, __m256i __b)
-{
+static __inline__ __m256i __DEFAULT_FN_ATTRS256_CONSTEXPR
+_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);
 }
 
@@ -2888,9 +2883,8 @@ _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
-_mm256_unpacklo_epi16(__m256i __a, __m256i __b)
-{
+static __inline__ __m256i __DEFAULT_FN_ATTRS256_CONSTEXPR
+_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);
 }
 
@@ -2922,9 +2916,8 @@ _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
-_mm256_unpacklo_epi32(__m256i __a, __m256i __b)
-{
+static __inline__ __m256i __DEFAULT_FN_ATTRS256_CONSTEXPR
+_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);
 }
 
@@ -2952,9 +2945,8 @@ _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
-_mm256_unpacklo_epi64(__m256i __a, __m256i __b)
-{
+static __inline__ __m256i __DEFAULT_FN_ATTRS256_CONSTEXPR
+_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/avx512fintrin.h b/clang/lib/Headers/avx512fintrin.h
@@ -4203,9 +4203,8 @@ _mm512_maskz_unpacklo_ps (__mmask16 __U, __m512 __A, __m512 __B)
                                           (__v16sf)_mm512_setzero_ps());
 }
 
-static __inline__ __m512i __DEFAULT_FN_ATTRS512
-_mm512_unpackhi_epi32(__m512i __A, __m512i __B)
-{
+static __inline__ __m512i __DEFAULT_FN_ATTRS512_CONSTEXPR
+_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,
@@ -4229,9 +4228,8 @@ _mm512_maskz_unpackhi_epi32(__mmask16 __U, __m512i __A, __m512i __B)
                                        (__v16si)_mm512_setzero_si512());
 }
 
-static __inline__ __m512i __DEFAULT_FN_ATTRS512
-_mm512_unpacklo_epi32(__m512i __A, __m512i __B)
-{
+static __inline__ __m512i __DEFAULT_FN_ATTRS512_CONSTEXPR
+_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,
@@ -4255,9 +4253,8 @@ _mm512_maskz_unpacklo_epi32(__mmask16 __U, __m512i __A, __m512i __B)
                                        (__v16si)_mm512_setzero_si512());
 }
 
-static __inline__ __m512i __DEFAULT_FN_ATTRS512
-_mm512_unpackhi_epi64(__m512i __A, __m512i __B)
-{
+static __inline__ __m512i __DEFAULT_FN_ATTRS512_CONSTEXPR
+_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);
 }
@@ -4278,9 +4275,8 @@ _mm512_maskz_unpackhi_epi64(__mmask8 __U, __m512i __A, __m512i __B)
                                         (__v8di)_mm512_setzero_si512());
 }
 
-static __inline__ __m512i __DEFAULT_FN_ATTRS512
-_mm512_unpacklo_epi64 (__m512i __A, __m512i __B)
-{
+static __inline__ __m512i __DEFAULT_FN_ATTRS512_CONSTEXPR
+_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
@@ -4417,8 +4417,8 @@ 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 _mm_unpackhi_epi8(__m128i __a,
-                                                               __m128i __b) {
+static __inline__ __m128i __DEFAULT_FN_ATTRS_CONSTEXPR
+_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);
@@ -4445,8 +4445,8 @@ static __inline__ __m128i __DEFAULT_FN_ATTRS _mm_unpackhi_epi8(__m128i __a,
 ///    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 _mm_unpackhi_epi16(__m128i __a,
-                                                                __m128i __b) {
+static __inline__ __m128i __DEFAULT_FN_ATTRS_CONSTEXPR
+_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);
 }
@@ -4468,8 +4468,8 @@ static __inline__ __m128i __DEFAULT_FN_ATTRS _mm_unpackhi_epi16(__m128i __a,
 ///    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 _mm_unpackhi_epi32(__m128i __a,
-                                                                __m128i __b) {
+static __inline__ __m128i __DEFAULT_FN_ATTRS_CONSTEXPR
+_mm_unpackhi_epi32(__m128i __a, __m128i __b) {
   return (__m128i)__builtin_shufflevector((__v4si)__a, (__v4si)__b, 2, 4 + 2, 3,
                                           4 + 3);
 }
@@ -4489,8 +4489,8 @@ static __inline__ __m128i __DEFAULT_FN_ATTRS _mm_unpackhi_epi32(__m128i __a,
 ///    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 _mm_unpackhi_epi64(__m128i __a,
-                                                                __m128i __b) {
+static __inline__ __m128i __DEFAULT_FN_ATTRS_CONSTEXPR
+_mm_unpackhi_epi64(__m128i __a, __m128i __b) {
   return (__m128i)__builtin_shufflevector((__v2di)__a, (__v2di)__b, 1, 2 + 1);
 }
 
@@ -4523,8 +4523,8 @@ static __inline__ __m128i __DEFAULT_FN_ATTRS _mm_unpackhi_epi64(__m128i __a,
 ///    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 _mm_unpacklo_epi8(__m128i __a,
-                                                               __m128i __b) {
+static __inline__ __m128i __DEFAULT_FN_ATTRS_CONSTEXPR
+_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);
@@ -4552,8 +4552,8 @@ static __inline__ __m128i __DEFAULT_FN_ATTRS _mm_unpacklo_epi8(__m128i __a,
 ///    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 _mm_unpacklo_epi16(__m128i __a,
-                                                                __m128i __b) {
+static __inline__ __m128i __DEFAULT_FN_ATTRS_CONSTEXPR
+_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);
 }
@@ -4575,8 +4575,8 @@ static __inline__ __m128i __DEFAULT_FN_ATTRS _mm_unpacklo_epi16(__m128i __a,
 ///    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 _mm_unpacklo_epi32(__m128i __a,
-                                                                __m128i __b) {
+static __inline__ __m128i __DEFAULT_FN_ATTRS_CONSTEXPR
+_mm_unpacklo_epi32(__m128i __a, __m128i __b) {
   return (__m128i)__builtin_shufflevector((__v4si)__a, (__v4si)__b, 0, 4 + 0, 1,
                                           4 + 1);
 }
@@ -4596,8 +4596,8 @@ static __inline__ __m128i __DEFAULT_FN_ATTRS _mm_unpacklo_epi32(__m128i __a,
 ///    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 _mm_unpacklo_epi64(__m128i __a,
-                                                                __m128i __b) {
+static __inline__ __m128i __DEFAULT_FN_ATTRS_CONSTEXPR
+_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
@@ -1319,48 +1319,56 @@ __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/avx512f-builtins.c b/clang/test/CodeGen/X86/avx512f-builtins.c
@@ -4691,6 +4691,7 @@ __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
@@ -5477,6 +5478,7 @@ __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
@@ -5497,6 +5499,7 @@ __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
@@ -5517,6 +5520,7 @@ __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
