colin
/
onnx-mlir
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity
onnx-mlir/test/mlir/krnl/reshape.mlir

26 lines
2.1 KiB
MLIR
Raw Normal View History

[NFC] Change ONNF based names to ONNX-MLIR (#32) * Rename onnf to onnx-mlir. * Change workspace name.
2020-03-17 21:16:33 +08:00
// RUN: onnx-mlir-opt --shape-inference --lower-frontend --lower-krnl --lower-all-llvm %s -split-input-file | FileCheck %s
[MLIR] Add support for reshape (#390) * Add reshape op handling. * Lower reshape to KRNL dialect. * Add comments. * Propagate reshape to KRNL IR. * Lower KRNL reshape to affine and standard ops level dialects. * Add lowering of reshape operation to Krnl and LLVM Dialects. * Add test for LLVM IR dialect output for reshape. * Fix rebase. * Fix test variable. * Emit errors during reshape shape inference. Address other reviewer comments.
2019-12-14 04:28:56 +08:00
func @test_reshape(%arg0 : tensor<?x10xf32>, %arg1 : tensor<4xi32>) -> tensor<*xf32> {
%0 = "onnx.Reshape"(%arg0, %arg1) : (tensor<?x10xf32>, tensor<4xi32>) -> tensor<*xf32>
"std.return"(%0) : (tensor<*xf32>) -> ()
Fix reshape op. (#53)
2020-01-28 23:21:08 +08:00
// CHECK: llvm.func @llvm.memcpy.p0i8.p0i8.i64(!llvm<"i8*">, !llvm<"i8*">, !llvm.i64, !llvm.i1)
Fix building ONNF with latest LLVM/MLIR (#89) * Fix build and link errors. * Fix end to end tests. * Fix indentation. * Fix type conversion. * Use newest LLVM version. * Use newest LLVM version.
2020-02-20 07:15:02 +08:00
// CHECK: [[TMP:%.+]] = llvm.mlir.undef : !llvm<"{ float*, float*, i64, [2 x i64], [2 x i64] }">
// CHECK: llvm.insertvalue %arg0, %0[0] : !llvm<"{ float*, float*, i64, [2 x i64], [2 x i64] }">
// CHECK: llvm.insertvalue %arg1, %1[1] : !llvm<"{ float*, float*, i64, [2 x i64], [2 x i64] }">
// CHECK: llvm.insertvalue %arg2, %2[2] : !llvm<"{ float*, float*, i64, [2 x i64], [2 x i64] }">
// CHECK: llvm.insertvalue %arg3, %3[3, 0] : !llvm<"{ float*, float*, i64, [2 x i64], [2 x i64] }">
// CHECK: llvm.insertvalue %arg5, %4[4, 0] : !llvm<"{ float*, float*, i64, [2 x i64], [2 x i64] }">
// CHECK: llvm.insertvalue %arg4, %5[3, 1] : !llvm<"{ float*, float*, i64, [2 x i64], [2 x i64] }">
// CHECK: [[TMP1:%.+]] = llvm.insertvalue %arg6, %6[4, 1] : !llvm<"{ float*, float*, i64, [2 x i64], [2 x i64] }">
[MLIR] Add support for reshape (#390) * Add reshape op handling. * Lower reshape to KRNL dialect. * Add comments. * Propagate reshape to KRNL IR. * Lower KRNL reshape to affine and standard ops level dialects. * Add lowering of reshape operation to Krnl and LLVM Dialects. * Add test for LLVM IR dialect output for reshape. * Fix rebase. * Fix test variable. * Emit errors during reshape shape inference. Address other reviewer comments.
2019-12-14 04:28:56 +08:00
// CHECK: [[RES:%.+]] = llvm.insertvalue {{.*}}[4, 3] : !llvm<"{ float*, float*, i64, [4 x i64], [4 x i64] }">
// CHECK: [[EXT_VAL_0:%.+]] = llvm.extractvalue [[RES]][1] : !llvm<"{ float*, float*, i64, [4 x i64], [4 x i64] }">
// CHECK: [[DST:%.+]] = llvm.bitcast [[EXT_VAL_0]] : !llvm<"float*"> to !llvm<"i8*">
Fix building ONNF with latest LLVM/MLIR (#89) * Fix build and link errors. * Fix end to end tests. * Fix indentation. * Fix type conversion. * Use newest LLVM version. * Use newest LLVM version.
2020-02-20 07:15:02 +08:00
// CHECK: [[EXT_VAL_1:%.+]] = llvm.extractvalue [[TMP1]][1] : !llvm<"{ float*, float*, i64, [2 x i64], [2 x i64] }">
[MLIR] Add support for reshape (#390) * Add reshape op handling. * Lower reshape to KRNL dialect. * Add comments. * Propagate reshape to KRNL IR. * Lower KRNL reshape to affine and standard ops level dialects. * Add lowering of reshape operation to Krnl and LLVM Dialects. * Add test for LLVM IR dialect output for reshape. * Fix rebase. * Fix test variable. * Emit errors during reshape shape inference. Address other reviewer comments.
2019-12-14 04:28:56 +08:00
// CHECK: [[SRC:%.+]] = llvm.bitcast [[EXT_VAL_1]] : !llvm<"float*"> to !llvm<"i8*">
// CHECK: [[SIZE:%.+]] = llvm.sext %{{.*}} : !llvm.i64 to !llvm.i64
Fix reshape op. (#53)
2020-01-28 23:21:08 +08:00
// CHECK: [[VOLATILE:%.+]] = llvm.mlir.constant(0 : i1) : !llvm.i1
// CHECK: llvm.call @llvm.memcpy.p0i8.p0i8.i64([[DST]], [[SRC]], [[SIZE]], [[VOLATILE]]) : (!llvm<"i8*">, !llvm<"i8*">, !llvm.i64, !llvm.i1) -> !llvm.void
[MLIR] Add support for reshape (#390) * Add reshape op handling. * Lower reshape to KRNL dialect. * Add comments. * Propagate reshape to KRNL IR. * Lower KRNL reshape to affine and standard ops level dialects. * Add lowering of reshape operation to Krnl and LLVM Dialects. * Add test for LLVM IR dialect output for reshape. * Fix rebase. * Fix test variable. * Emit errors during reshape shape inference. Address other reviewer comments.
2019-12-14 04:28:56 +08:00
// CHECK: llvm.return [[RES]] : !llvm<"{ float*, float*, i64, [4 x i64], [4 x i64] }">
}