[Mlir-commits] [mlir] a2edbd8 - [MLIR][LLVMDialect] Added bitreverse and ctpop intrinsics

[Lei Zhang]

Author: George Mitenkov
Date: 2020-06-23T14:25:35-04:00
New Revision: a2edbd8170bd55ea64466a0719e4adda4b08a195

URL: https://github.com/llvm/llvm-project/commit/a2edbd8170bd55ea64466a0719e4adda4b08a195
DIFF: https://github.com/llvm/llvm-project/commit/a2edbd8170bd55ea64466a0719e4adda4b08a195.diff

LOG: [MLIR][LLVMDialect] Added bitreverse and ctpop intrinsics

Introduced `llvm.intr.bitreverse` and `llvm.intr.ctpop` LLVM bit
intrinsics to LLVM dialect. These intrinsics help with SPIR-V to
LLVM conversion, allowing a direct mapping from `spv.BitReverse`
and `spv.BitCount` respectively. Tests are added to `roundtrip.mlir`
and `llvm-intrinsics.mlir`.

Differential Revision: https://reviews.llvm.org/D82285

Added: 
    

Modified: 
    mlir/include/mlir/Dialect/LLVMIR/LLVMOps.td
    mlir/test/Dialect/LLVMIR/roundtrip.mlir
    mlir/test/Target/llvmir-intrinsics.mlir

Removed: 
    


################################################################################
diff  --git a/mlir/include/mlir/Dialect/LLVMIR/LLVMOps.td b/mlir/include/mlir/Dialect/LLVMIR/LLVMOps.td
index 8b6485459532..30e34440c2dd 100644
--- a/mlir/include/mlir/Dialect/LLVMIR/LLVMOps.td
+++ b/mlir/include/mlir/Dialect/LLVMIR/LLVMOps.td
@@ -791,6 +791,8 @@ def LLVM_Prefetch : LLVM_ZeroResultIntrOp<"prefetch", [0]>,
 def LLVM_SinOp : LLVM_UnaryIntrinsicOp<"sin">;
 def LLVM_SqrtOp : LLVM_UnaryIntrinsicOp<"sqrt">;
 def LLVM_PowOp : LLVM_BinarySameArgsIntrinsicOp<"pow">;
+def LLVM_BitReverseOp : LLVM_UnaryIntrinsicOp<"bitreverse">;
+def LLVM_CtPopOp : LLVM_UnaryIntrinsicOp<"ctpop">;
 
 //
 // Vector Reductions.

diff  --git a/mlir/test/Dialect/LLVMIR/roundtrip.mlir b/mlir/test/Dialect/LLVMIR/roundtrip.mlir
index d93de93882be..e782d5de1aaa 100644
--- a/mlir/test/Dialect/LLVMIR/roundtrip.mlir
+++ b/mlir/test/Dialect/LLVMIR/roundtrip.mlir
@@ -103,6 +103,12 @@ func @ops(%arg0 : !llvm.i32, %arg1 : !llvm.float) {
 // CHECK: "llvm.intr.pow"(%arg1, %arg1) : (!llvm.float, !llvm.float) -> !llvm.float
   %31 = "llvm.intr.pow"(%arg1, %arg1) : (!llvm.float, !llvm.float) -> !llvm.float
 
+// CHECK: "llvm.intr.bitreverse"(%{{.*}}) : (!llvm.i32) -> !llvm.i32
+  %32 = "llvm.intr.bitreverse"(%arg0) : (!llvm.i32) -> !llvm.i32
+
+// CHECK: "llvm.intr.ctpop"(%{{.*}}) : (!llvm.i32) -> !llvm.i32
+  %33 = "llvm.intr.ctpop"(%arg0) : (!llvm.i32) -> !llvm.i32
+
 // CHECK:  llvm.return
   llvm.return
 }

diff  --git a/mlir/test/Target/llvmir-intrinsics.mlir b/mlir/test/Target/llvmir-intrinsics.mlir
index 57d98aa25f36..e04b40e916fd 100644
--- a/mlir/test/Target/llvmir-intrinsics.mlir
+++ b/mlir/test/Target/llvmir-intrinsics.mlir
@@ -117,6 +117,24 @@ llvm.func @pow_test(%arg0: !llvm.float, %arg1: !llvm.float, %arg2: !llvm<"<8 x f
   llvm.return
 }
 
+// CHECK-LABEL: @bitreverse_test
+llvm.func @bitreverse_test(%arg0: !llvm.i32, %arg1: !llvm<"<8 x i32>">) {
+  // CHECK: call i32 @llvm.bitreverse.i32
+  "llvm.intr.bitreverse"(%arg0) : (!llvm.i32) -> !llvm.i32
+  // CHECK: call <8 x i32> @llvm.bitreverse.v8i32
+  "llvm.intr.bitreverse"(%arg1) : (!llvm<"<8 x i32>">) -> !llvm<"<8 x i32>">
+  llvm.return
+}
+
+// CHECK-LABEL: @ctpop_test
+llvm.func @ctpop_test(%arg0: !llvm.i32, %arg1: !llvm<"<8 x i32>">) {
+  // CHECK: call i32 @llvm.ctpop.i32
+  "llvm.intr.ctpop"(%arg0) : (!llvm.i32) -> !llvm.i32
+  // CHECK: call <8 x i32> @llvm.ctpop.v8i32
+  "llvm.intr.ctpop"(%arg1) : (!llvm<"<8 x i32>">) -> !llvm<"<8 x i32>">
+  llvm.return
+}
+
 // CHECK-LABEL: @vector_reductions
 llvm.func @vector_reductions(%arg0: !llvm.float, %arg1: !llvm<"<8 x float>">, %arg2: !llvm<"<8 x i32>">) {
   // CHECK: call i32 @llvm.experimental.vector.reduce.add.v8i32