[AVX512] adding PRORQ , PRORD , PRORLVQ and PRORLVD Intrinsics

Differential Revision: http://reviews.llvm.org/D16052

llvm-svn: 257594

6 files changed, 220 insertions, 3 deletions

diff --git a/llvm/include/llvm/IR/IntrinsicsX86.td b/llvm/include/llvm/IR/IntrinsicsX86.td
index e01718159d8..37e12a3d61f 100644
--- a/llvm/include/llvm/IR/IntrinsicsX86.td
+++ b/llvm/include/llvm/IR/IntrinsicsX86.td
@@ -3044,6 +3044,43 @@ let TargetPrefix = "x86" in {  // All intrinsics start with "llvm.x86.".
                          llvm_i8_ty, llvm_v8i64_ty, llvm_i8_ty], [IntrNoMem]>;
 
 
+  def int_x86_avx512_mask_prolv_d_128 : GCCBuiltin<"__builtin_ia32_prolvd128_mask">,
+              Intrinsic<[llvm_v4i32_ty], [llvm_v4i32_ty,
+                         llvm_v4i32_ty, llvm_v4i32_ty, llvm_i8_ty], [IntrNoMem]>;
+  def int_x86_avx512_mask_prolv_d_256 : GCCBuiltin<"__builtin_ia32_prolvd256_mask">,
+              Intrinsic<[llvm_v8i32_ty], [llvm_v8i32_ty,
+                         llvm_v8i32_ty, llvm_v8i32_ty, llvm_i8_ty], [IntrNoMem]>;
+  def int_x86_avx512_mask_prolv_d_512 : GCCBuiltin<"__builtin_ia32_prolvd512_mask">,
+              Intrinsic<[llvm_v16i32_ty], [llvm_v16i32_ty,
+                         llvm_v16i32_ty, llvm_v16i32_ty, llvm_i16_ty], [IntrNoMem]>;
+  def int_x86_avx512_mask_prolv_q_128 : GCCBuiltin<"__builtin_ia32_prolvq128_mask">,
+              Intrinsic<[llvm_v2i64_ty], [llvm_v2i64_ty,
+                         llvm_v2i64_ty, llvm_v2i64_ty, llvm_i8_ty], [IntrNoMem]>;
+  def int_x86_avx512_mask_prolv_q_256 : GCCBuiltin<"__builtin_ia32_prolvq256_mask">,
+              Intrinsic<[llvm_v4i64_ty], [llvm_v4i64_ty,
+                         llvm_v4i64_ty, llvm_v4i64_ty, llvm_i8_ty], [IntrNoMem]>;
+  def int_x86_avx512_mask_prolv_q_512 : GCCBuiltin<"__builtin_ia32_prolvq512_mask">,
+              Intrinsic<[llvm_v8i64_ty], [llvm_v8i64_ty,
+                         llvm_v8i64_ty, llvm_v8i64_ty, llvm_i8_ty], [IntrNoMem]>;
+  def int_x86_avx512_mask_pror_d_128 : GCCBuiltin<"__builtin_ia32_prord128_mask">,
+              Intrinsic<[llvm_v4i32_ty], [llvm_v4i32_ty,
+                         llvm_i8_ty, llvm_v4i32_ty, llvm_i8_ty], [IntrNoMem]>;
+  def int_x86_avx512_mask_pror_d_256 : GCCBuiltin<"__builtin_ia32_prord256_mask">,
+              Intrinsic<[llvm_v8i32_ty], [llvm_v8i32_ty,
+                         llvm_i8_ty, llvm_v8i32_ty, llvm_i8_ty], [IntrNoMem]>;
+  def int_x86_avx512_mask_pror_d_512 : GCCBuiltin<"__builtin_ia32_prord512_mask">,
+              Intrinsic<[llvm_v16i32_ty], [llvm_v16i32_ty,
+                         llvm_i8_ty, llvm_v16i32_ty, llvm_i16_ty], [IntrNoMem]>;
+  def int_x86_avx512_mask_pror_q_128 : GCCBuiltin<"__builtin_ia32_prorq128_mask">,
+              Intrinsic<[llvm_v2i64_ty], [llvm_v2i64_ty,
+                         llvm_i8_ty, llvm_v2i64_ty, llvm_i8_ty], [IntrNoMem]>;
+  def int_x86_avx512_mask_pror_q_256 : GCCBuiltin<"__builtin_ia32_prorq256_mask">,
+              Intrinsic<[llvm_v4i64_ty], [llvm_v4i64_ty, 
+                         llvm_i8_ty, llvm_v4i64_ty, llvm_i8_ty], [IntrNoMem]>;
+  def int_x86_avx512_mask_pror_q_512 : GCCBuiltin<"__builtin_ia32_prorq512_mask">,
+              Intrinsic<[llvm_v8i64_ty], [llvm_v8i64_ty, 
+                         llvm_i8_ty, llvm_v8i64_ty, llvm_i8_ty], [IntrNoMem]>;
+
 }
 
 // Gather ops
diff --git a/llvm/lib/Target/X86/X86ISelLowering.h b/llvm/lib/Target/X86/X86ISelLowering.h
index 8b1daf45ac0..0ab786e08e0 100644
--- a/llvm/lib/Target/X86/X86ISelLowering.h
+++ b/llvm/lib/Target/X86/X86ISelLowering.h
@@ -317,9 +317,8 @@ namespace llvm {
       VSHLI, VSRLI, VSRAI,
 
       // Bit rotate by immediate
+      VROTLI, VROTRI,
 
-      VROTLI,
-      
       // Vector packed double/float comparison.
       CMPP,
 
diff --git a/llvm/lib/Target/X86/X86InstrAVX512.td b/llvm/lib/Target/X86/X86InstrAVX512.td
index 2bd59db5120..49be6488393 100644
--- a/llvm/lib/Target/X86/X86InstrAVX512.td
+++ b/llvm/lib/Target/X86/X86InstrAVX512.td
@@ -4056,7 +4056,7 @@ defm VPSLL : avx512_shift_rmi_dq<0x72, 0x73, MRM6r, MRM6m, "vpsll", X86vshli>,
 defm VPSRA : avx512_shift_rmi_dq<0x72, 0x72, MRM4r, MRM4m, "vpsra", X86vsrai>,
              avx512_shift_rmi_w<0x71, MRM4r, MRM4m, "vpsraw", X86vsrai>, AVX512BIi8Base, EVEX_4V;
 
-defm VPROR : avx512_shift_rmi_dq<0x72, 0x72, MRM0r, MRM0m, "vpror", rotr>, AVX512BIi8Base, EVEX_4V;
+defm VPROR : avx512_shift_rmi_dq<0x72, 0x72, MRM0r, MRM0m, "vpror", X86vrotri>, AVX512BIi8Base, EVEX_4V;
 defm VPROL : avx512_shift_rmi_dq<0x72, 0x72, MRM1r, MRM1m, "vprol", X86vrotli>, AVX512BIi8Base, EVEX_4V;
 
 defm VPSLL : avx512_shift_types<0xF2, 0xF3, 0xF1, "vpsll", X86vshl>;
diff --git a/llvm/lib/Target/X86/X86InstrFragmentsSIMD.td b/llvm/lib/Target/X86/X86InstrFragmentsSIMD.td
index d967f2546af..643286324e2 100644
--- a/llvm/lib/Target/X86/X86InstrFragmentsSIMD.td
+++ b/llvm/lib/Target/X86/X86InstrFragmentsSIMD.td
@@ -226,6 +226,7 @@ def X86vsrli   : SDNode<"X86ISD::VSRLI", SDTIntShiftOp>;
 def X86vsrai   : SDNode<"X86ISD::VSRAI", SDTIntShiftOp>;
 
 def X86vrotli  : SDNode<"X86ISD::VROTLI", SDTIntShiftOp>;
+def X86vrotri  : SDNode<"X86ISD::VROTRI", SDTIntShiftOp>;
 
 def X86vprot   : SDNode<"X86ISD::VPROT",
                         SDTypeProfile<1, 2, [SDTCisVec<0>, SDTCisSameAs<0,1>,
diff --git a/llvm/lib/Target/X86/X86IntrinsicsInfo.h b/llvm/lib/Target/X86/X86IntrinsicsInfo.h
index 1100cad62d5..2d8443a1c91 100644
--- a/llvm/lib/Target/X86/X86IntrinsicsInfo.h
+++ b/llvm/lib/Target/X86/X86IntrinsicsInfo.h
@@ -1219,6 +1219,18 @@ static const IntrinsicData  IntrinsicsWithoutChain[] = {
   X86_INTRINSIC_DATA(avx512_mask_prol_q_128,  INTR_TYPE_2OP_MASK, X86ISD::VROTLI, 0),
   X86_INTRINSIC_DATA(avx512_mask_prol_q_256,  INTR_TYPE_2OP_MASK, X86ISD::VROTLI, 0),
   X86_INTRINSIC_DATA(avx512_mask_prol_q_512,  INTR_TYPE_2OP_MASK, X86ISD::VROTLI, 0),
+  X86_INTRINSIC_DATA(avx512_mask_prolv_d_128, INTR_TYPE_2OP_MASK, ISD::ROTL, 0),
+  X86_INTRINSIC_DATA(avx512_mask_prolv_d_256, INTR_TYPE_2OP_MASK, ISD::ROTL, 0),
+  X86_INTRINSIC_DATA(avx512_mask_prolv_d_512, INTR_TYPE_2OP_MASK, ISD::ROTL, 0),
+  X86_INTRINSIC_DATA(avx512_mask_prolv_q_128, INTR_TYPE_2OP_MASK, ISD::ROTL, 0),
+  X86_INTRINSIC_DATA(avx512_mask_prolv_q_256, INTR_TYPE_2OP_MASK, ISD::ROTL, 0),
+  X86_INTRINSIC_DATA(avx512_mask_prolv_q_512, INTR_TYPE_2OP_MASK, ISD::ROTL, 0),
+  X86_INTRINSIC_DATA(avx512_mask_pror_d_128,  INTR_TYPE_2OP_MASK, X86ISD::VROTRI, 0),
+  X86_INTRINSIC_DATA(avx512_mask_pror_d_256,  INTR_TYPE_2OP_MASK, X86ISD::VROTRI, 0),
+  X86_INTRINSIC_DATA(avx512_mask_pror_d_512,  INTR_TYPE_2OP_MASK, X86ISD::VROTRI, 0),
+  X86_INTRINSIC_DATA(avx512_mask_pror_q_128,  INTR_TYPE_2OP_MASK, X86ISD::VROTRI, 0),
+  X86_INTRINSIC_DATA(avx512_mask_pror_q_256,  INTR_TYPE_2OP_MASK, X86ISD::VROTRI, 0),
+  X86_INTRINSIC_DATA(avx512_mask_pror_q_512,  INTR_TYPE_2OP_MASK, X86ISD::VROTRI, 0),  
   X86_INTRINSIC_DATA(avx512_mask_prorv_d_128, INTR_TYPE_2OP_MASK, ISD::ROTR, 0),
   X86_INTRINSIC_DATA(avx512_mask_prorv_d_256, INTR_TYPE_2OP_MASK, ISD::ROTR, 0),
   X86_INTRINSIC_DATA(avx512_mask_prorv_d_512, INTR_TYPE_2OP_MASK, ISD::ROTR, 0),
diff --git a/llvm/test/CodeGen/X86/avx512vl-intrinsics.ll b/llvm/test/CodeGen/X86/avx512vl-intrinsics.ll
index fb2b41d9116..a47608ae1ff 100644
--- a/llvm/test/CodeGen/X86/avx512vl-intrinsics.ll
+++ b/llvm/test/CodeGen/X86/avx512vl-intrinsics.ll
@@ -6874,3 +6874,171 @@ define <4 x i64>@test_int_x86_avx512_mask_prol_q_256(<4 x i64> %x0, i8 %x1, <4 x
   %res4 = add <4 x i64> %res3, %res2
   ret <4 x i64> %res4
 }
+
+declare <4 x i32> @llvm.x86.avx512.mask.prolv.d.128(<4 x i32>, <4 x i32>, <4 x i32>, i8)
+
+define <4 x i32>@test_int_x86_avx512_mask_prolv_d_128(<4 x i32> %x0, <4 x i32> %x1, <4 x i32> %x2, i8 %x3) {
+; CHECK-LABEL: test_int_x86_avx512_mask_prolv_d_128:
+; CHECK:       ## BB#0:
+; CHECK-NEXT:    movzbl %dil, %eax
+; CHECK-NEXT:    kmovw %eax, %k1
+; CHECK-NEXT:    vprolvd %xmm1, %xmm0, %xmm2 {%k1}
+; CHECK-NEXT:    vprolvd %xmm1, %xmm0, %xmm3 {%k1} {z}
+; CHECK-NEXT:    vprolvd %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.prolv.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.prolv.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.prolv.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.prolv.d.256(<8 x i32>, <8 x i32>, <8 x i32>, i8)
+
+define <8 x i32>@test_int_x86_avx512_mask_prolv_d_256(<8 x i32> %x0, <8 x i32> %x1, <8 x i32> %x2, i8 %x3) {
+; CHECK-LABEL: test_int_x86_avx512_mask_prolv_d_256:
+; CHECK:       ## BB#0:
+; CHECK-NEXT:    movzbl %dil, %eax
+; CHECK-NEXT:    kmovw %eax, %k1
+; CHECK-NEXT:    vprolvd %ymm1, %ymm0, %ymm2 {%k1} 
+; CHECK-NEXT:    vprolvd %ymm1, %ymm0, %ymm3 {%k1} {z} 
+; CHECK-NEXT:    vprolvd %ymm1, %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.prolv.d.256(<8 x i32> %x0, <8 x i32> %x1, <8 x i32> %x2, i8 %x3)
+  %res1 = call <8 x i32> @llvm.x86.avx512.mask.prolv.d.256(<8 x i32> %x0, <8 x i32> %x1, <8 x i32> zeroinitializer, i8 %x3)
+  %res2 = call <8 x i32> @llvm.x86.avx512.mask.prolv.d.256(<8 x i32> %x0, <8 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 <2 x i64> @llvm.x86.avx512.mask.prolv.q.128(<2 x i64>, <2 x i64>, <2 x i64>, i8)
+
+define <2 x i64>@test_int_x86_avx512_mask_prolv_q_128(<2 x i64> %x0, <2 x i64> %x1, <2 x i64> %x2, i8 %x3) {
+; CHECK-LABEL: test_int_x86_avx512_mask_prolv_q_128:
+; CHECK:       ## BB#0:
+; CHECK-NEXT:    movzbl %dil, %eax 
+; CHECK-NEXT:    kmovw %eax, %k1 
+; CHECK-NEXT:    vprolvq %xmm1, %xmm0, %xmm2 {%k1} 
+; CHECK-NEXT:    vprolvq %xmm1, %xmm0, %xmm3 {%k1} {z} 
+; CHECK-NEXT:    vprolvq %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.prolv.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.prolv.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.prolv.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.prolv.q.256(<4 x i64>, <4 x i64>, <4 x i64>, i8)
+
+define <4 x i64>@test_int_x86_avx512_mask_prolv_q_256(<4 x i64> %x0, <4 x i64> %x1, <4 x i64> %x2, i8 %x3) {
+; CHECK-LABEL: test_int_x86_avx512_mask_prolv_q_256:
+; CHECK:       ## BB#0:
+; CHECK-NEXT:    movzbl %dil, %eax 
+; CHECK-NEXT:    kmovw %eax, %k1 
+; CHECK-NEXT:    vprolvq %ymm1, %ymm0, %ymm2 {%k1} 
+; CHECK-NEXT:    vprolvq %ymm1, %ymm0, %ymm3 {%k1} {z} 
+; CHECK-NEXT:    vprolvq %ymm1, %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.prolv.q.256(<4 x i64> %x0, <4 x i64> %x1, <4 x i64> %x2, i8 %x3)
+  %res1 = call <4 x i64> @llvm.x86.avx512.mask.prolv.q.256(<4 x i64> %x0, <4 x i64> %x1, <4 x i64> zeroinitializer, i8 %x3)
+  %res2 = call <4 x i64> @llvm.x86.avx512.mask.prolv.q.256(<4 x i64> %x0, <4 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 <4 x i32> @llvm.x86.avx512.mask.pror.d.128(<4 x i32>, i8, <4 x i32>, i8)
+
+define <4 x i32>@test_int_x86_avx512_mask_pror_d_128(<4 x i32> %x0, i8 %x1, <4 x i32> %x2, i8 %x3) {
+; CHECK-LABEL: test_int_x86_avx512_mask_pror_d_128:
+; CHECK:       ## BB#0:
+; CHECK-NEXT:    movzbl %sil, %eax 
+; CHECK-NEXT:    kmovw %eax, %k1 
+; CHECK-NEXT:    vprord $3, %xmm0, %xmm1 {%k1} 
+; CHECK-NEXT:    vprord $3, %xmm0, %xmm2 {%k1} {z} 
+; CHECK-NEXT:    vprord $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.pror.d.128(<4 x i32> %x0, i8 3, <4 x i32> %x2, i8 %x3)
+  %res1 = call <4 x i32> @llvm.x86.avx512.mask.pror.d.128(<4 x i32> %x0, i8 3, <4 x i32> zeroinitializer, i8 %x3)
+  %res2 = call <4 x i32> @llvm.x86.avx512.mask.pror.d.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.pror.d.256(<8 x i32>, i8, <8 x i32>, i8)
+
+define <8 x i32>@test_int_x86_avx512_mask_pror_d_256(<8 x i32> %x0, i32 %x1, <8 x i32> %x2, i8 %x3) {
+; CHECK-LABEL: test_int_x86_avx512_mask_pror_d_256:
+; CHECK:       ## BB#0:
+; CHECK-NEXT:    movzbl %sil, %eax 
+; CHECK-NEXT:    kmovw %eax, %k1 
+; CHECK-NEXT:    vprord $3, %ymm0, %ymm1 {%k1} 
+; CHECK-NEXT:    vprord $3, %ymm0, %ymm2 {%k1} {z} 
+; CHECK-NEXT:    vprord $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.pror.d.256(<8 x i32> %x0, i8 3, <8 x i32> %x2, i8 %x3)
+  %res1 = call <8 x i32> @llvm.x86.avx512.mask.pror.d.256(<8 x i32> %x0, i8 3, <8 x i32> zeroinitializer, i8 %x3)
+  %res2 = call <8 x i32> @llvm.x86.avx512.mask.pror.d.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.pror.q.128(<2 x i64>, i8, <2 x i64>, i8)
+
+define <2 x i64>@test_int_x86_avx512_mask_pror_q_128(<2 x i64> %x0, i8 %x1, <2 x i64> %x2, i8 %x3) {
+; CHECK-LABEL: test_int_x86_avx512_mask_pror_q_128:
+; CHECK:       ## BB#0:
+; CHECK-NEXT:    movzbl %sil, %eax 
+; CHECK-NEXT:    kmovw %eax, %k1 
+; CHECK-NEXT:    vprorq $3, %xmm0, %xmm1 {%k1} 
+; CHECK-NEXT:    vprorq $3, %xmm0, %xmm2 {%k1} {z} 
+; CHECK-NEXT:    vprorq $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.pror.q.128(<2 x i64> %x0, i8 3, <2 x i64> %x2, i8 %x3)
+  %res1 = call <2 x i64> @llvm.x86.avx512.mask.pror.q.128(<2 x i64> %x0, i8 3, <2 x i64> zeroinitializer, i8 %x3)
+  %res2 = call <2 x i64> @llvm.x86.avx512.mask.pror.q.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.pror.q.256(<4 x i64>, i8, <4 x i64>, i8)
+
+define <4 x i64>@test_int_x86_avx512_mask_pror_q_256(<4 x i64> %x0, i8 %x1, <4 x i64> %x2, i8 %x3) {
+; CHECK-LABEL: test_int_x86_avx512_mask_pror_q_256:
+; CHECK:       ## BB#0:
+; CHECK-NEXT:    movzbl %sil, %eax 
+; CHECK-NEXT:    kmovw %eax, %k1 
+; CHECK-NEXT:    vprorq $3, %ymm0, %ymm1 {%k1} 
+; CHECK-NEXT:    vprorq $3, %ymm0, %ymm2 {%k1} {z} 
+; CHECK-NEXT:    vprorq $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.pror.q.256(<4 x i64> %x0, i8 3, <4 x i64> %x2, i8 %x3)
+  %res1 = call <4 x i64> @llvm.x86.avx512.mask.pror.q.256(<4 x i64> %x0, i8 3, <4 x i64> zeroinitializer, i8 %x3)
+  %res2 = call <4 x i64> @llvm.x86.avx512.mask.pror.q.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
+}