[Clang][Intrinsics] Adding avx512f missing intrinsics {cast}

[AsafBadouh]

lib/Headers/avx512fintrin.h

385

Can you use :

return __builtin_shufflevector(__A, __A, 0,1)

?

and same for the other

lib/Headers/avx512fintrin.h

 }
 
 static __inline __m128 __DEFAULT_FN_ATTRS
 _mm512_castps512_ps128(__m512 __a)
 {
   return __builtin_shufflevector(__a, __a, 0, 1, 2, 3);
 }
 
+static __inline__ __m512 __DEFAULT_FN_ATTRS
+_mm512_castpd_ps (__m512d __A)
+{
+  return (__m512) (__A);
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS
+_mm512_castpd_si512 (__m512d __A)
+{
+  return (__m512i) (__A);
+}
+
+static __inline__ __m512d __DEFAULT_FN_ATTRS
+_mm512_castps_pd (__m512 __A)
+{
+  return (__m512d) (__A);
+}
+
+static __inline__ __m512i __DEFAULT_FN_ATTRS
+_mm512_castps_si512 (__m512 __A)
+{
+  return (__m512i) (__A);
+}
+
+static __inline__ __m512 __DEFAULT_FN_ATTRS
+_mm512_castsi512_ps (__m512i __A)
+{
+  return (__m512) (__A);
+}
+
+static __inline__ __m512d __DEFAULT_FN_ATTRS
+_mm512_castsi512_pd (__m512i __A)
+{
+  return (__m512d) (__A);
+}
+
+static __inline__ __m128i __DEFAULT_FN_ATTRS
+_mm512_castsi512_si128 (__m512i __A)
+{
+  return (__m128i)_mm512_extracti32x4_epi32((__m512i)__A, 0);
+}
+
+static __inline__ __m256d __DEFAULT_FN_ATTRS
+_mm512_castpd512_pd256 (__m512d __A)
+{
+  return _mm512_extractf64x4_pd(__A, 0);
+}
+
+static __inline__ __m256 __DEFAULT_FN_ATTRS
+_mm512_castps512_ps256 (__m512 __A)
+{
+  return (__m256)_mm512_extractf64x4_pd((__m512d)__A, 0);
+}
+
+static __inline__ __m256i __DEFAULT_FN_ATTRS
+_mm512_castsi512_si256 (__m512i __A)
+{
+  return (__m256i)_mm512_extractf64x4_pd((__m512d)__A, 0);
+}
+
 /* Bitwise operators */
 static __inline__ __m512i __DEFAULT_FN_ATTRS
 _mm512_and_epi32(__m512i __a, __m512i __b)
 {
   return __a & __b;
 }
 
 static __inline__ __m512i __DEFAULT_FN_ATTRS

test/CodeGen/avx512f-builtins.c

 
 __m512d test_mm512_castpd256_pd512(__m256d a)
 {
   // CHECK-LABEL: @test_mm512_castpd256_pd512
   // CHECK: shufflevector <4 x double> {{.*}} <i32 0, i32 1, i32 2, i32 3, i32 undef, i32 undef, i32 undef, i32 undef>
   return _mm512_castpd256_pd512(a);
 }
 
+__m512 test_mm512_castpd_ps (__m512d __A)
+{
+  // CHECK-LABEL: @test_mm512_castpd_ps 
+  // CHECK: bitcast <8 x double> %1 to <16 x float>
+  return _mm512_castpd_ps (__A);
+}
+
+__m512i test_mm512_castpd_si512 (__m512d __A)
+{
+  // CHECK-LABEL: @test_mm512_castpd_si512 
+  // CHECK: bitcast <8 x double> %1 to <8 x i64>
+  return _mm512_castpd_si512 (__A);
+}
+
+__m512d test_mm512_castps_pd (__m512 __A)
+{
+  // CHECK-LABEL: @test_mm512_castps_pd 
+  // CHECK: bitcast <16 x float> %1 to <8 x double>
+  return _mm512_castps_pd (__A);
+}
+
+__m512i test_mm512_castps_si512 (__m512 __A)
+{
+  // CHECK-LABEL: @test_mm512_castps_si512 
+  // CHECK: bitcast <16 x float> %1 to <8 x i64>
+  return _mm512_castps_si512 (__A);
+}
+
+__m512 test_mm512_castsi512_ps (__m512i __A)
+{
+  // CHECK-LABEL: @test_mm512_castsi512_ps 
+  // CHECK: bitcast <8 x i64> %1 to <16 x float>
+  return _mm512_castsi512_ps (__A);
+}
+
+__m512d test_mm512_castsi512_pd (__m512i __A)
+{
+  // CHECK-LABEL: @test_mm512_castsi512_pd 
+  // CHECK: bitcast <8 x i64> %1 to <8 x double>
+  return _mm512_castsi512_pd (__A);
+}
+
+__m128i test_mm512_castsi512_si128 (__m512i __A)
+{
+  // CHECK-LABEL: @test_mm512_castsi512_si128 
+  // CHECK: @llvm.x86.avx512.mask.vextracti32x4.512
+  return _mm512_castsi512_si128 (__A);
+}
+
+__m256d test_mm512_castpd512_pd256 (__m512d __A)
+{
+  // CHECK-LABEL: @test_mm512_castpd512_pd256 
+  // CHECK: @llvm.x86.avx512.mask.vextractf64x4.512
+  return _mm512_castpd512_pd256 (__A);
+}
+
+__m256 test_mm512_castps512_ps256 (__m512 __A)
+{
+  // CHECK-LABEL: @test_mm512_castps512_ps256 
+  // CHECK: @llvm.x86.avx512.mask.vextractf64x4.512
+  return _mm512_castps512_ps256 (__A);
+}
+
+__m256i test_mm512_castsi512_si256 (__m512i __A)
+{
+  // CHECK-LABEL: @test_mm512_castsi512_si256 
+  // CHECK: @llvm.x86.avx512.mask.vextractf64x4.512
+  return _mm512_castsi512_si256 (__A);
+}
+
 __mmask16 test_mm512_knot(__mmask16 a)
 {
   // CHECK-LABEL: @test_mm512_knot
   // CHECK: @llvm.x86.avx512.knot.w
   return _mm512_knot(a);
 }
 
 __m512i test_mm512_alignr_epi32(__m512i a, __m512i b)