[AVX512] add PSRAD and PSRAQ Intrinsic

Differential Revision: http://reviews.llvm.org/D15851

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@256754 91177308-0d34-0410-b5e6-96231b3b80d8

diff --git a/include/llvm/IR/IntrinsicsX86.td b/include/llvm/IR/IntrinsicsX86.td
index 0e6877268ecaaf49636fac76beb425d26d78fa16..fb4e8893d7d58b7f0f632eff895db08fa76e148c 100644 (file)
--- a/include/llvm/IR/IntrinsicsX86.td
+++ b/include/llvm/IR/IntrinsicsX86.td
@@ -2261,7 +2261,38 @@ let TargetPrefix = "x86" in {  // All intrinsics start with "llvm.x86.".
   def int_x86_avx512_mask_psra_q : GCCBuiltin<"__builtin_ia32_psraq512_mask">,
               Intrinsic<[llvm_v8i64_ty], [llvm_v8i64_ty,
                          llvm_v2i64_ty, llvm_v8i64_ty, llvm_i8_ty], [IntrNoMem]>;
-  
+
+  def int_x86_avx512_mask_psra_d_128 : GCCBuiltin<"__builtin_ia32_psrad128_mask">,
+              Intrinsic<[llvm_v4i32_ty], [llvm_v4i32_ty,
+                         llvm_v4i32_ty, llvm_v4i32_ty, llvm_i8_ty], [IntrNoMem]>;
+  def int_x86_avx512_mask_psra_d_256 : GCCBuiltin<"__builtin_ia32_psrad256_mask">,
+             Intrinsic<[llvm_v8i32_ty], [llvm_v8i32_ty, 
+                         llvm_v4i32_ty, llvm_v8i32_ty, llvm_i8_ty], [IntrNoMem]>;
+  def int_x86_avx512_mask_psra_di_128 : GCCBuiltin<"__builtin_ia32_psradi128_mask">,
+              Intrinsic<[llvm_v4i32_ty], [llvm_v4i32_ty, 
+                         llvm_i8_ty, llvm_v4i32_ty, llvm_i8_ty], [IntrNoMem]>;
+  def int_x86_avx512_mask_psra_di_256 : GCCBuiltin<"__builtin_ia32_psradi256_mask">,
+              Intrinsic<[llvm_v8i32_ty], [llvm_v8i32_ty, 
+                        llvm_i8_ty, llvm_v8i32_ty, llvm_i8_ty], [IntrNoMem]>;
+  def int_x86_avx512_mask_psra_di_512 : GCCBuiltin<"__builtin_ia32_psradi512_mask">,
+              Intrinsic<[llvm_v16i32_ty], [llvm_v16i32_ty,
+                         llvm_i8_ty, llvm_v16i32_ty, llvm_i16_ty], [IntrNoMem]>;
+  def int_x86_avx512_mask_psra_q_128 : GCCBuiltin<"__builtin_ia32_psraq128_mask">,
+              Intrinsic<[llvm_v2i64_ty], [llvm_v2i64_ty, 
+                         llvm_v2i64_ty, llvm_v2i64_ty, llvm_i8_ty], [IntrNoMem]>;
+  def int_x86_avx512_mask_psra_q_256 : GCCBuiltin<"__builtin_ia32_psraq256_mask">,
+             Intrinsic<[llvm_v4i64_ty], [llvm_v4i64_ty, 
+                         llvm_v2i64_ty, llvm_v4i64_ty, llvm_i8_ty], [IntrNoMem]>;
+  def int_x86_avx512_mask_psra_qi_128 : GCCBuiltin<"__builtin_ia32_psraqi128_mask">,
+              Intrinsic<[llvm_v2i64_ty], [llvm_v2i64_ty, 
+                         llvm_i8_ty, llvm_v2i64_ty, llvm_i8_ty], [IntrNoMem]>;
+  def int_x86_avx512_mask_psra_qi_256 : GCCBuiltin<"__builtin_ia32_psraqi256_mask">,
+             Intrinsic<[llvm_v4i64_ty], [llvm_v4i64_ty, 
+                         llvm_i8_ty, llvm_v4i64_ty, llvm_i8_ty], [IntrNoMem]>;
+  def int_x86_avx512_mask_psra_qi_512 : GCCBuiltin<"__builtin_ia32_psraqi512_mask">,
+              Intrinsic<[llvm_v8i64_ty], [llvm_v8i64_ty, 
+                         llvm_i8_ty, llvm_v8i64_ty, llvm_i8_ty], [IntrNoMem]>;
+
   def int_x86_avx512_mask_psrl_d_128: GCCBuiltin<"__builtin_ia32_psrld128_mask">,
               Intrinsic<[llvm_v4i32_ty], [ llvm_v4i32_ty,
                          llvm_v4i32_ty, llvm_v4i32_ty, llvm_i8_ty ], [IntrNoMem]>;

diff --git a/lib/Target/X86/X86IntrinsicsInfo.h b/lib/Target/X86/X86IntrinsicsInfo.h
index c67844a2efc9e6b304c54aec71ea85c5d84f5947..0a3304c89e81c8bfd970fc86247a6c58b0fcb8fe 100644 (file)
--- a/lib/Target/X86/X86IntrinsicsInfo.h
+++ b/lib/Target/X86/X86IntrinsicsInfo.h
@@ -1214,7 +1214,17 @@ static const IntrinsicData  IntrinsicsWithoutChain[] = {
   X86_INTRINSIC_DATA(avx512_mask_psllv_d,       INTR_TYPE_2OP_MASK, ISD::SHL, 0),
   X86_INTRINSIC_DATA(avx512_mask_psllv_q,       INTR_TYPE_2OP_MASK, ISD::SHL, 0),
   X86_INTRINSIC_DATA(avx512_mask_psra_d,        INTR_TYPE_2OP_MASK, X86ISD::VSRA, 0),
+  X86_INTRINSIC_DATA(avx512_mask_psra_d_128,    INTR_TYPE_2OP_MASK, X86ISD::VSRA, 0),
+  X86_INTRINSIC_DATA(avx512_mask_psra_d_256,    INTR_TYPE_2OP_MASK, X86ISD::VSRA, 0),
+  X86_INTRINSIC_DATA(avx512_mask_psra_di_128,   INTR_TYPE_2OP_MASK, X86ISD::VSRAI, 0),
+  X86_INTRINSIC_DATA(avx512_mask_psra_di_256,   INTR_TYPE_2OP_MASK, X86ISD::VSRAI, 0),
+  X86_INTRINSIC_DATA(avx512_mask_psra_di_512,   INTR_TYPE_2OP_MASK, X86ISD::VSRAI, 0),
   X86_INTRINSIC_DATA(avx512_mask_psra_q,        INTR_TYPE_2OP_MASK, X86ISD::VSRA, 0),
+  X86_INTRINSIC_DATA(avx512_mask_psra_q_128,    INTR_TYPE_2OP_MASK, X86ISD::VSRA, 0),
+  X86_INTRINSIC_DATA(avx512_mask_psra_q_256,    INTR_TYPE_2OP_MASK, X86ISD::VSRA, 0),
+  X86_INTRINSIC_DATA(avx512_mask_psra_qi_128,   INTR_TYPE_2OP_MASK, X86ISD::VSRAI, 0),
+  X86_INTRINSIC_DATA(avx512_mask_psra_qi_256,   INTR_TYPE_2OP_MASK, X86ISD::VSRAI, 0),
+  X86_INTRINSIC_DATA(avx512_mask_psra_qi_512,   INTR_TYPE_2OP_MASK, X86ISD::VSRAI, 0),
   X86_INTRINSIC_DATA(avx512_mask_psra_w_128,    INTR_TYPE_2OP_MASK, X86ISD::VSRA, 0),
   X86_INTRINSIC_DATA(avx512_mask_psra_w_256,    INTR_TYPE_2OP_MASK, X86ISD::VSRA, 0),
   X86_INTRINSIC_DATA(avx512_mask_psra_w_512,    INTR_TYPE_2OP_MASK, X86ISD::VSRA, 0),

diff --git a/test/CodeGen/X86/avx512-intrinsics.ll b/test/CodeGen/X86/avx512-intrinsics.ll
index eb0987771b3f67417d8b14af8b19e82258303a05..98d73f95177b01e7480d4aa0e14dace0a4a5a6fc 100644 (file)
--- a/test/CodeGen/X86/avx512-intrinsics.ll
+++ b/test/CodeGen/X86/avx512-intrinsics.ll
@@ -6460,3 +6460,43 @@ define <16 x i32>@test_int_x86_avx512_mask_psrl_di_512(<16 x i32> %x0, i8 %x1, <
   ret <16 x i32> %res4
 }
 
+declare <16 x i32> @llvm.x86.avx512.mask.psra.di.512(<16 x i32>, i8, <16 x i32>, i16)
+
+define <16 x i32>@test_int_x86_avx512_mask_psra_di_512(<16 x i32> %x0, i8 %x1, <16 x i32> %x2, i16 %x3) {
+; CHECK-LABEL: test_int_x86_avx512_mask_psra_di_512:
+; CHECK:       ## BB#0:
+; CHECK-NEXT:    kmovw %esi, %k1
+; CHECK-NEXT:    vpsrad $3, %zmm0, %zmm1 {%k1}
+; CHECK-NEXT:    vpsrad $3, %zmm0, %zmm2 {%k1} {z}
+; CHECK-NEXT:    vpsrad $3, %zmm0, %zmm0
+; CHECK-NEXT:    vpaddd %zmm2, %zmm1, %zmm1
+; CHECK-NEXT:    vpaddd %zmm0, %zmm1, %zmm0
+; CHECK-NEXT:    retq
+  %res = call <16 x i32> @llvm.x86.avx512.mask.psra.di.512(<16 x i32> %x0, i8 3, <16 x i32> %x2, i16 %x3)
+  %res1 = call <16 x i32> @llvm.x86.avx512.mask.psra.di.512(<16 x i32> %x0, i8 3, <16 x i32> zeroinitializer, i16 %x3)
+  %res2 = call <16 x i32> @llvm.x86.avx512.mask.psra.di.512(<16 x i32> %x0, i8 3, <16 x i32> %x2, i16 -1)
+  %res3 = add <16 x i32> %res, %res1
+  %res4 = add <16 x i32> %res3, %res2
+  ret <16 x i32> %res4
+}
+
+declare <8 x i64> @llvm.x86.avx512.mask.psra.qi.512(<8 x i64>, i8, <8 x i64>, i8)
+
+define <8 x i64>@test_int_x86_avx512_mask_psra_qi_512(<8 x i64> %x0, i8 %x1, <8 x i64> %x2, i8 %x3) {
+; CHECK-LABEL: test_int_x86_avx512_mask_psra_qi_512:
+; CHECK:       ## BB#0:
+; CHECK-NEXT:    movzbl %sil, %eax
+; CHECK-NEXT:    kmovw %eax, %k1
+; CHECK-NEXT:    vpsraq $3, %zmm0, %zmm1 {%k1}
+; CHECK-NEXT:    vpsraq $3, %zmm0, %zmm2 {%k1} {z}
+; CHECK-NEXT:    vpsraq $3, %zmm0, %zmm0
+; CHECK-NEXT:    vpaddq %zmm2, %zmm1, %zmm1
+; CHECK-NEXT:    vpaddq %zmm0, %zmm1, %zmm0
+; CHECK-NEXT:    retq
+  %res = call <8 x i64> @llvm.x86.avx512.mask.psra.qi.512(<8 x i64> %x0, i8 3, <8 x i64> %x2, i8 %x3)
+  %res1 = call <8 x i64> @llvm.x86.avx512.mask.psra.qi.512(<8 x i64> %x0, i8 3, <8 x i64> zeroinitializer, i8 %x3)
+  %res2 = call <8 x i64> @llvm.x86.avx512.mask.psra.qi.512(<8 x i64> %x0, i8 3, <8 x i64> %x2, i8 -1)
+  %res3 = add <8 x i64> %res, %res1
+  %res4 = add <8 x i64> %res3, %res2
+  ret <8 x i64> %res4
+}

diff --git a/test/CodeGen/X86/avx512vl-intrinsics.ll b/test/CodeGen/X86/avx512vl-intrinsics.ll
index 00944c59b48a4671020e4edfa47dcedadc700267..72893388fbaab7c4d76f74a707c1ae3a1770b524 100644 (file)
--- a/test/CodeGen/X86/avx512vl-intrinsics.ll
+++ b/test/CodeGen/X86/avx512vl-intrinsics.ll
@@ -6071,3 +6071,172 @@ define <8 x i32>@test_int_x86_avx512_mask_psrlv8_si(<8 x i32> %x0, <8 x i32> %x1
   %res4 = add <8 x i32> %res3, %res2
   ret <8 x i32> %res4
 }
+
+declare <4 x i32> @llvm.x86.avx512.mask.psra.d.128(<4 x i32>, <4 x i32>, <4 x i32>, i8)
+
+define <4 x i32>@test_int_x86_avx512_mask_psra_d_128(<4 x i32> %x0, <4 x i32> %x1, <4 x i32> %x2, i8 %x3) {
+; CHECK-LABEL: test_int_x86_avx512_mask_psra_d_128:
+; CHECK:       ## BB#0:
+; CHECK-NEXT:    movzbl %dil, %eax
+; CHECK-NEXT:    kmovw %eax, %k1
+; CHECK-NEXT:    vpsrad %xmm1, %xmm0, %xmm2 {%k1}
+; CHECK-NEXT:    vpsrad %xmm1, %xmm0, %xmm3 {%k1} {z}
+; CHECK-NEXT:    vpsrad %xmm1, %xmm0, %xmm0
+; CHECK-NEXT:    vpaddd %xmm3, %xmm2, %xmm1
+; CHECK-NEXT:    vpaddd %xmm0, %xmm1, %xmm0
+; CHECK-NEXT:    retq
+  %res = call <4 x i32> @llvm.x86.avx512.mask.psra.d.128(<4 x i32> %x0, <4 x i32> %x1, <4 x i32> %x2, i8 %x3)
+  %res1 = call <4 x i32> @llvm.x86.avx512.mask.psra.d.128(<4 x i32> %x0, <4 x i32> %x1, <4 x i32> zeroinitializer, i8 %x3)
+  %res2 = call <4 x i32> @llvm.x86.avx512.mask.psra.d.128(<4 x i32> %x0, <4 x i32> %x1, <4 x i32> %x2, i8 -1)
+  %res3 = add <4 x i32> %res, %res1
+  %res4 = add <4 x i32> %res3, %res2
+  ret <4 x i32> %res4
+}
+
+declare <8 x i32> @llvm.x86.avx512.mask.psra.d.256(<8 x i32>, <4 x i32>, <8 x i32>, i8)
+
+define <8 x i32>@test_int_x86_avx512_mask_psra_d_256(<8 x i32> %x0, <4 x i32> %x1, <8 x i32> %x2, i8 %x3) {
+; CHECK-LABEL: test_int_x86_avx512_mask_psra_d_256:
+; CHECK:       ## BB#0:
+; CHECK-NEXT:    movzbl %dil, %eax
+; CHECK-NEXT:    kmovw %eax, %k1
+; CHECK-NEXT:    vpsrad %xmm1, %ymm0, %ymm2 {%k1}
+; CHECK-NEXT:    vpsrad %xmm1, %ymm0, %ymm3 {%k1} {z}
+; CHECK-NEXT:    vpsrad %xmm1, %ymm0, %ymm0
+; CHECK-NEXT:    vpaddd %ymm3, %ymm2, %ymm1
+; CHECK-NEXT:    vpaddd %ymm0, %ymm1, %ymm0
+; CHECK-NEXT:    retq
+  %res = call <8 x i32> @llvm.x86.avx512.mask.psra.d.256(<8 x i32> %x0, <4 x i32> %x1, <8 x i32> %x2, i8 %x3)
+  %res1 = call <8 x i32> @llvm.x86.avx512.mask.psra.d.256(<8 x i32> %x0, <4 x i32> %x1, <8 x i32> zeroinitializer, i8 %x3)
+  %res2 = call <8 x i32> @llvm.x86.avx512.mask.psra.d.256(<8 x i32> %x0, <4 x i32> %x1, <8 x i32> %x2, i8 -1)
+  %res3 = add <8 x i32> %res, %res1
+  %res4 = add <8 x i32> %res3, %res2
+  ret <8 x i32> %res4
+}
+
+declare <4 x i32> @llvm.x86.avx512.mask.psra.di.128(<4 x i32>, i8, <4 x i32>, i8)
+
+define <4 x i32>@test_int_x86_avx512_mask_psra_di_128(<4 x i32> %x0, i8 %x1, <4 x i32> %x2, i8 %x3) {
+; CHECK-LABEL: test_int_x86_avx512_mask_psra_di_128:
+; CHECK:       ## BB#0:
+; CHECK-NEXT:    movzbl %sil, %eax
+; CHECK-NEXT:    kmovw %eax, %k1
+; CHECK-NEXT:    vpsrad $3, %xmm0, %xmm1 {%k1}
+; CHECK-NEXT:    vpsrad $3, %xmm0, %xmm2 {%k1} {z}
+; CHECK-NEXT:    vpsrad $3, %xmm0, %xmm0
+; CHECK-NEXT:    vpaddd %xmm2, %xmm1, %xmm1
+; CHECK-NEXT:    vpaddd %xmm0, %xmm1, %xmm0
+; CHECK-NEXT:    retq
+  %res = call <4 x i32> @llvm.x86.avx512.mask.psra.di.128(<4 x i32> %x0, i8 3, <4 x i32> %x2, i8 %x3)
+  %res1 = call <4 x i32> @llvm.x86.avx512.mask.psra.di.128(<4 x i32> %x0, i8 3, <4 x i32> zeroinitializer, i8 %x3)
+  %res2 = call <4 x i32> @llvm.x86.avx512.mask.psra.di.128(<4 x i32> %x0, i8 3, <4 x i32> %x2, i8 -1)
+  %res3 = add <4 x i32> %res, %res1
+  %res4 = add <4 x i32> %res3, %res2
+  ret <4 x i32> %res4
+}
+
+declare <8 x i32> @llvm.x86.avx512.mask.psra.di.256(<8 x i32>, i8, <8 x i32>, i8)
+
+define <8 x i32>@test_int_x86_avx512_mask_psra_di_256(<8 x i32> %x0, i8 %x1, <8 x i32> %x2, i8 %x3) {
+; CHECK-LABEL: test_int_x86_avx512_mask_psra_di_256:
+; CHECK:       ## BB#0:
+; CHECK-NEXT:    movzbl %sil, %eax
+; CHECK-NEXT:    kmovw %eax, %k1
+; CHECK-NEXT:    vpsrad $3, %ymm0, %ymm1 {%k1}
+; CHECK-NEXT:    vpsrad $3, %ymm0, %ymm2 {%k1} {z}
+; CHECK-NEXT:    vpsrad $3, %ymm0, %ymm0
+; CHECK-NEXT:    vpaddd %ymm2, %ymm1, %ymm1
+; CHECK-NEXT:    vpaddd %ymm0, %ymm1, %ymm0
+; CHECK-NEXT:    retq
+  %res = call <8 x i32> @llvm.x86.avx512.mask.psra.di.256(<8 x i32> %x0, i8 3, <8 x i32> %x2, i8 %x3)
+  %res1 = call <8 x i32> @llvm.x86.avx512.mask.psra.di.256(<8 x i32> %x0, i8 3, <8 x i32> zeroinitializer, i8 %x3)
+  %res2 = call <8 x i32> @llvm.x86.avx512.mask.psra.di.256(<8 x i32> %x0, i8 3, <8 x i32> %x2, i8 -1)
+  %res3 = add <8 x i32> %res, %res1
+  %res4 = add <8 x i32> %res3, %res2
+  ret <8 x i32> %res4
+}
+
+declare <2 x i64> @llvm.x86.avx512.mask.psra.q.128(<2 x i64>, <2 x i64>, <2 x i64>, i8)
+
+define <2 x i64>@test_int_x86_avx512_mask_psra_q_128(<2 x i64> %x0, <2 x i64> %x1, <2 x i64> %x2, i8 %x3) {
+; CHECK-LABEL: test_int_x86_avx512_mask_psra_q_128:
+; CHECK:       ## BB#0:
+; CHECK-NEXT:    movzbl %dil, %eax
+; CHECK-NEXT:    kmovw %eax, %k1
+; CHECK-NEXT:    vpsraq %xmm1, %xmm0, %xmm2 {%k1}
+; CHECK-NEXT:    vpsraq %xmm1, %xmm0, %xmm3 {%k1} {z}
+; CHECK-NEXT:    vpsraq %xmm1, %xmm0, %xmm0
+; CHECK-NEXT:    vpaddq %xmm3, %xmm2, %xmm1
+; CHECK-NEXT:    vpaddq %xmm0, %xmm1, %xmm0
+; CHECK-NEXT:    retq
+  %res = call <2 x i64> @llvm.x86.avx512.mask.psra.q.128(<2 x i64> %x0, <2 x i64> %x1, <2 x i64> %x2, i8 %x3)
+  %res1 = call <2 x i64> @llvm.x86.avx512.mask.psra.q.128(<2 x i64> %x0, <2 x i64> %x1, <2 x i64> zeroinitializer, i8 %x3)
+  %res2 = call <2 x i64> @llvm.x86.avx512.mask.psra.q.128(<2 x i64> %x0, <2 x i64> %x1, <2 x i64> %x2, i8 -1)
+  %res3 = add <2 x i64> %res, %res1
+  %res4 = add <2 x i64> %res3, %res2
+  ret <2 x i64> %res4
+}
+
+declare <4 x i64> @llvm.x86.avx512.mask.psra.q.256(<4 x i64>, <2 x i64>, <4 x i64>, i8)
+
+define <4 x i64>@test_int_x86_avx512_mask_psra_q_256(<4 x i64> %x0, <2 x i64> %x1, <4 x i64> %x2, i8 %x3) {
+; CHECK-LABEL: test_int_x86_avx512_mask_psra_q_256:
+; CHECK:       ## BB#0:
+; CHECK-NEXT:    movzbl %dil, %eax
+; CHECK-NEXT:    kmovw %eax, %k1
+; CHECK-NEXT:    vpsraq %xmm1, %ymm0, %ymm2 {%k1}
+; CHECK-NEXT:    vpsraq %xmm1, %ymm0, %ymm3 {%k1} {z}
+; CHECK-NEXT:    vpsraq %xmm1, %ymm0, %ymm0
+; CHECK-NEXT:    vpaddq %ymm3, %ymm2, %ymm1
+; CHECK-NEXT:    vpaddq %ymm0, %ymm1, %ymm0
+; CHECK-NEXT:    retq
+  %res = call <4 x i64> @llvm.x86.avx512.mask.psra.q.256(<4 x i64> %x0, <2 x i64> %x1, <4 x i64> %x2, i8 %x3)
+  %res1 = call <4 x i64> @llvm.x86.avx512.mask.psra.q.256(<4 x i64> %x0, <2 x i64> %x1, <4 x i64> zeroinitializer, i8 %x3)
+  %res2 = call <4 x i64> @llvm.x86.avx512.mask.psra.q.256(<4 x i64> %x0, <2 x i64> %x1, <4 x i64> %x2, i8 -1)
+  %res3 = add <4 x i64> %res, %res1
+  %res4 = add <4 x i64> %res3, %res2
+  ret <4 x i64> %res4
+}
+
+declare <2 x i64> @llvm.x86.avx512.mask.psra.qi.128(<2 x i64>, i8, <2 x i64>, i8)
+
+define <2 x i64>@test_int_x86_avx512_mask_psra_qi_128(<2 x i64> %x0, i8 %x1, <2 x i64> %x2, i8 %x3) {
+; CHECK-LABEL: test_int_x86_avx512_mask_psra_qi_128:
+; CHECK:       ## BB#0:
+; CHECK-NEXT:    movzbl %sil, %eax
+; CHECK-NEXT:    kmovw %eax, %k1
+; CHECK-NEXT:    vpsraq $3, %xmm0, %xmm1 {%k1}
+; CHECK-NEXT:    vpsraq $3, %xmm0, %xmm2 {%k1} {z}
+; CHECK-NEXT:    vpsraq $3, %xmm0, %xmm0
+; CHECK-NEXT:    vpaddq %xmm2, %xmm1, %xmm1
+; CHECK-NEXT:    vpaddq %xmm0, %xmm1, %xmm0
+; CHECK-NEXT:    retq
+  %res = call <2 x i64> @llvm.x86.avx512.mask.psra.qi.128(<2 x i64> %x0, i8 3, <2 x i64> %x2, i8 %x3)
+  %res1 = call <2 x i64> @llvm.x86.avx512.mask.psra.qi.128(<2 x i64> %x0, i8 3, <2 x i64> zeroinitializer, i8 %x3)
+  %res2 = call <2 x i64> @llvm.x86.avx512.mask.psra.qi.128(<2 x i64> %x0, i8 3, <2 x i64> %x2, i8 -1)
+  %res3 = add <2 x i64> %res, %res1
+  %res4 = add <2 x i64> %res3, %res2
+  ret <2 x i64> %res4
+}
+
+declare <4 x i64> @llvm.x86.avx512.mask.psra.qi.256(<4 x i64>, i8, <4 x i64>, i8)
+
+define <4 x i64>@test_int_x86_avx512_mask_psra_qi_256(<4 x i64> %x0, i8 %x1, <4 x i64> %x2, i8 %x3) {
+; CHECK-LABEL: test_int_x86_avx512_mask_psra_qi_256:
+; CHECK:       ## BB#0:
+; CHECK-NEXT:    movzbl %sil, %eax
+; CHECK-NEXT:    kmovw %eax, %k1
+; CHECK-NEXT:    vpsraq $3, %ymm0, %ymm1 {%k1}
+; CHECK-NEXT:    vpsraq $3, %ymm0, %ymm2 {%k1} {z}
+; CHECK-NEXT:    vpsraq $3, %ymm0, %ymm0
+; CHECK-NEXT:    vpaddq %ymm2, %ymm1, %ymm1
+; CHECK-NEXT:    vpaddq %ymm0, %ymm1, %ymm0
+; CHECK-NEXT:    retq
+  %res = call <4 x i64> @llvm.x86.avx512.mask.psra.qi.256(<4 x i64> %x0, i8 3, <4 x i64> %x2, i8 %x3)
+  %res1 = call <4 x i64> @llvm.x86.avx512.mask.psra.qi.256(<4 x i64> %x0, i8 3, <4 x i64> zeroinitializer, i8 %x3)
+  %res2 = call <4 x i64> @llvm.x86.avx512.mask.psra.qi.256(<4 x i64> %x0, i8 3, <4 x i64> %x2, i8 -1)
+  %res3 = add <4 x i64> %res, %res1
+  %res4 = add <4 x i64> %res3, %res2
+  ret <4 x i64> %res4
+}
+