initial commit

Author: bruteforceboy

clang/lib/CIR/Dialect/Transforms/TargetLowering/LowerFunction.cpp

@@ -336,7 +336,13 @@ mlir::Value createCoercedValue(mlir::Value Src, mlir::Type Ty,
     return CGF.buildAggregateBitcast(Src, Ty);
   }
 
-  cir_cconv_unreachable("NYI");
+  auto &bld = CGF.getRewriter();
+  auto alloca = bld.create<AllocaOp>(
+      Src.getLoc(), bld.getType<PointerType>(Ty), Ty,
+      /*name=*/llvm::StringRef(""), /*alignment=*/bld.getI64IntegerAttr(4));
+  Src = findAlloca(Src.getDefiningOp());
+  createMemCpy(CGF, alloca, Src, SrcSize.getFixedValue());
+  return bld.create<LoadOp>(Src.getLoc(), alloca.getResult());
 }
 
 mlir::Value emitAddressAtOffset(LowerFunction &LF, mlir::Value addr,

clang/test/CIR/CallConvLowering/AArch64/aarch64-cc-structs.c

@@ -234,3 +234,52 @@ S_PAD ret_s_pad() {
   S_PAD s;
   return s;
 }
+
+typedef struct {
+  int a;
+  int b;
+  short c;
+} GT_64_LT_128;
+
+// CHECK: @pass_gt_64_lt_128(%arg0: !cir.array<!u64i x 2>
+// CHECK: %[[#V0:]] = cir.alloca !ty_GT_64_LT_128_, !cir.ptr<!ty_GT_64_LT_128_>, [""] {alignment = 4 : i64}
+// CHECK: %[[#V1:]] = cir.alloca !cir.array<!u64i x 2>, !cir.ptr<!cir.array<!u64i x 2>>, ["tmp"] {alignment = 8 : i64}
+// CHECK: cir.store %arg0, %[[#V1]] : !cir.array<!u64i x 2>, !cir.ptr<!cir.array<!u64i x 2>>
+// CHECK: %[[#V2:]] = cir.cast(bitcast, %[[#V1]] : !cir.ptr<!cir.array<!u64i x 2>>), !cir.ptr<!void>
+// CHECK: %[[#V3:]] = cir.cast(bitcast, %[[#V0]] : !cir.ptr<!ty_GT_64_LT_128_>), !cir.ptr<!void>
+// CHECK: %[[#V4:]] = cir.const #cir.int<12> : !u64i
+// CHECK: cir.libc.memcpy %[[#V4]] bytes from %[[#V2]] to %[[#V3]] : !u64i, !cir.ptr<!void> -> !cir.ptr<!void>
+// CHECK: cir.return
+
+// LLVM: @pass_gt_64_lt_128([2 x i64] %[[#V0:]])
+// LLVM: %[[#V2:]] = alloca %struct.GT_64_LT_128, i64 1, align 4
+// LLVM: %[[#V3:]] = alloca [2 x i64], i64 1, align 8
+// LLVM: store [2 x i64] %[[#V0]], ptr %[[#V3]], align 8
+// LLVM: call void @llvm.memcpy.p0.p0.i64(ptr %[[#V2]], ptr %[[#V3]], i64 12, i1 false)
+// LLVM: ret void
+void pass_gt_64_lt_128(GT_64_LT_128 s) {}
+
+// CHECK: @call_gt_64_lt_128()
+// CHECK: %[[#V0:]] = cir.alloca !ty_GT_64_LT_128_, !cir.ptr<!ty_GT_64_LT_128_>, ["s"] {alignment = 4 : i64}
+// CHECK: %[[#V1:]] = cir.load %[[#V0]] : !cir.ptr<!ty_GT_64_LT_128_>, !ty_GT_64_LT_128_
+// CHECK: %[[#V2:]] = cir.alloca !cir.array<!u64i x 2>, !cir.ptr<!cir.array<!u64i x 2>>, [""] {alignment = 4 : i64}
+// CHECK: %[[#V3:]] = cir.cast(bitcast, %[[#V0]] : !cir.ptr<!ty_GT_64_LT_128_>), !cir.ptr<!void>
+// CHECK: %[[#V4:]] = cir.cast(bitcast, %[[#V2]] : !cir.ptr<!cir.array<!u64i x 2>>), !cir.ptr<!void>
+// CHECK: %[[#V5:]] = cir.const #cir.int<12> : !u64i
+// CHECK: cir.libc.memcpy %[[#V5]] bytes from %[[#V3]] to %[[#V4]] : !u64i, !cir.ptr<!void> -> !cir.ptr<!void>
+// CHECK: %[[#V6:]] = cir.load %[[#V2]] : !cir.ptr<!cir.array<!u64i x 2>>, !cir.array<!u64i x 2>
+// CHECK: cir.call @pass_gt_64_lt_128(%[[#V6]]) : (!cir.array<!u64i x 2>) -> ()
+// CHECK: cir.return
+
+// LLVM: @call_gt_64_lt_128()
+// LLVM: %[[#V1:]] = alloca %struct.GT_64_LT_128, i64 1, align 4
+// LLVM: %[[#V2:]] = load %struct.GT_64_LT_128, ptr %[[#V1]], align 4
+// LLVM: %[[#V3:]] = alloca [2 x i64], i64 1, align 4
+// LLVM: call void @llvm.memcpy.p0.p0.i64(ptr %[[#V3]], ptr %[[#V1]], i64 12, i1 false)
+// LLVM: %[[#V4:]] = load [2 x i64], ptr %[[#V3]], align 8
+// LLVM: call void @pass_gt_64_lt_128([2 x i64] %[[#V4]])
+// LLVM: ret void
+void call_gt_64_lt_128() {
+  GT_64_LT_128 s;
+  pass_gt_64_lt_128(s);
+}