Merge branch 'develop' into rand_uniform

src/tf/tf_parser.cpp

@@ -338,7 +338,7 @@ void tf_parser::parse_node(const std::string& name)
             std::string input_name = input;
             // if input has trailing `:0` index then remove it
             auto multi_out_idx = input.find(':');
-            if(multi_out_idx != std::string::npos && input.substr(multi_out_idx + 1) == "0")
+            if(multi_out_idx != std::string::npos and input.substr(multi_out_idx + 1) == "0")
             {
                 input_name = input.substr(0, multi_out_idx);
             }

src/value.cpp

@@ -285,7 +285,7 @@ bool value::contains(const std::string& pkey) const
 }
 std::size_t value::size() const
 {
-    auto* a = if_array_impl(x);
+    const auto* a = if_array_impl(x);
     if(a == nullptr)
         return 0;
     return a->size();

test/CMakeLists.txt

@@ -202,11 +202,16 @@ endif()
 
 
 function(test_header NAME HEADER)
-    file(WRITE ${CMAKE_CURRENT_BINARY_DIR}/header-main-include-${NAME}.cpp
-        "#include <${HEADER}>\nint main() {}\n"
+    file(WRITE ${CMAKE_CURRENT_BINARY_DIR}/header-main-include-${NAME}.cpp "
+#include <${HEADER}>
+int main() {}\n"
     )
-    file(WRITE ${CMAKE_CURRENT_BINARY_DIR}/header-static-include-${NAME}.cpp
-        "#include <${HEADER}>\n"
+    file(WRITE ${CMAKE_CURRENT_BINARY_DIR}/header-static-include-${NAME}.cpp "
+#include <${HEADER}>
+#if defined(min) || defined(max) || defined(near) || defined(far)
+#error \"Do not include windows.h in header files\"
+#endif
+\n"
     )
     add_test_executable(${NAME}
         ${CMAKE_CURRENT_BINARY_DIR}/header-main-include-${NAME}.cpp

test/api/test_cpu.cpp

@@ -145,15 +145,15 @@ TEST_CASE(zero_parameter)
 
 TEST_CASE(set_scalar_parameter)
 {
-    auto p1 = migraphx::parse_onnx("add_bcast_test.onnx");
-    migraphx::shape s1(migraphx_shape_float_type, {3, 4});
+    auto p1 = migraphx::parse_onnx("implicit_add_bcast_test.onnx");
+    migraphx::shape s1(migraphx_shape_float_type, {3, 4, 1});
     auto param_shapes = p1.get_parameter_shapes();
     auto s1_orig      = param_shapes["1"];
     CHECK(bool{s1 == s1_orig});
 
     migraphx::onnx_options option;
     option.set_input_parameter_shape("1", {});
-    auto p2 = migraphx::parse_onnx("add_bcast_test.onnx", option);
+    auto p2 = migraphx::parse_onnx("implicit_add_bcast_test.onnx", option);
     migraphx::shape s_scalar(migraphx_shape_float_type);
     auto param_shapes_1 = p2.get_parameter_shapes();
     auto s_scalar_after = param_shapes_1["1"];

test/eliminate_contiguous_test.cpp

@@ -196,15 +196,47 @@ TEST_CASE(contiguous_pointwise)
             migraphx::make_op("broadcast", {{"axis", 1}, {"out_lens", {2, 3, 8, 8}}}), y);
         auto yc  = mm->add_instruction(migraphx::make_op("contiguous"), yb);
         auto add = add_pointwise(p, "main:pointwise0", {x, yc}, single_pointwise("add"));
-        mm->add_instruction(pass_op{}, add);
+        auto cadd = mm->add_instruction(migraphx::make_op("contiguous"), add);
+        mm->add_instruction(pass_op{}, cadd);
     }
     auto count = std::distance(mm->begin(), mm->end());
     run_pass(*mm);
-    EXPECT(std::distance(mm->begin(), mm->end()) == (count - 1));
+    EXPECT(std::distance(mm->begin(), mm->end()) == (count - 2));
     EXPECT(std::none_of(
         mm->begin(), mm->end(), [](auto&& ins) { return ins.name() == "contiguous"; }));
 }
 
+TEST_CASE(contiguous_nhwc_pointwise)
+{
+    auto s =
+        migraphx::shape::from_permutation(migraphx::shape::float_type, {2, 3, 8, 8}, {0, 2, 3, 1});
+    migraphx::program p1;
+    {
+        auto* mm = p1.get_main_module();
+        auto x   = mm->add_parameter("x", s);
+        auto y   = mm->add_parameter("y", migraphx::shape{migraphx::shape::float_type, {3}});
+        auto yb  = mm->add_instruction(
+            migraphx::make_op("broadcast", {{"axis", 1}, {"out_lens", {2, 3, 8, 8}}}), y);
+        auto yc   = mm->add_instruction(migraphx::make_op("contiguous"), yb);
+        auto add  = add_pointwise(p1, "main:pointwise0", {x, yc}, single_pointwise("add"));
+        auto cadd = mm->add_instruction(migraphx::make_op("contiguous"), add);
+        mm->add_instruction(pass_op{}, cadd);
+    }
+    run_pass(*p1.get_main_module());
+    migraphx::program p2;
+    {
+        auto* mm = p2.get_main_module();
+        auto x   = mm->add_parameter("x", s);
+        auto y   = mm->add_parameter("y", migraphx::shape{migraphx::shape::float_type, {3}});
+        auto yb  = mm->add_instruction(
+            migraphx::make_op("broadcast", {{"axis", 1}, {"out_lens", {2, 3, 8, 8}}}), y);
+        auto add  = add_pointwise(p2, "main:pointwise0", {x, yb}, single_pointwise("add"));
+        auto cadd = mm->add_instruction(migraphx::make_op("contiguous"), add);
+        mm->add_instruction(pass_op{}, cadd);
+    }
+    EXPECT(p1 == p2);
+}
+
 TEST_CASE(slice_contiguous)
 {
     migraphx::module m;

test/eliminate_pad_test.cpp

@@ -27,7 +27,7 @@
 #include <migraphx/pass_manager.hpp>
 #include <migraphx/instruction.hpp>
 #include <basic_ops.hpp>
-#include <migraphx/operators.hpp>
+#include <migraphx/op/common.hpp>
 #include <migraphx/make_op.hpp>
 
 #include <test.hpp>
@@ -58,9 +58,8 @@ create_conv(migraphx::instruction_ref& l_img,
     migraphx::shape s_weights{migraphx::shape::int32_type, {4, channels, 3, 3}};
     std::vector<int32_t> weights(4 * channels * 3 * 3);
     auto l_weights = m.add_literal(migraphx::literal{s_weights, weights});
-    migraphx::op::convolution op;
-    op.padding_mode = padding_mode;
-    return m.add_instruction(op, l_img, l_weights);
+    return m.add_instruction(
+        migraphx::make_op("convolution", {{"padding_mode", padding_mode}}), l_img, l_weights);
 }
 
 TEST_CASE(rewrite_pad)

test/gpu/mlir.cpp

@@ -140,7 +140,7 @@ TEST_CASE(conv)
 {
     const std::string mlir_output = R"__migraphx__(
 module {
-  func.func @mlir_convolution(%arg0: tensor<2x8x3x3xf32>, %arg1: tensor<1x8x4x4xf32>) -> tensor<1x2x2x2xf32> attributes {arch = "", kernel = "mixr"} {
+  func.func @mlir_convolution(%arg0: tensor<2x8x3x3xf32>, %arg1: tensor<1x8x4x4xf32>) -> tensor<1x2x2x2xf32> attributes {arch = "", kernel = "mixr", num_cu = 0 : i64} {
     %0 = migraphx.convolution(%arg1, %arg0) {dilation = [1, 1], group = 1 : i64, padding = [0, 0, 0, 0], padding_mode = 0 : i64, stride = [1, 1]} : (tensor<1x8x4x4xf32>, tensor<2x8x3x3xf32>) -> tensor<1x2x2x2xf32>
     return %0 : tensor<1x2x2x2xf32>
   }
@@ -163,7 +163,7 @@ TEST_CASE(conv_add_relu)
 {
     const std::string mlir_output = R"__migraphx__(
 module {
-  func.func @mlir_convolution_add_relu(%arg0: tensor<1x2x2x2xf32>, %arg1: tensor<2x8x3x3xf32>, %arg2: tensor<1x8x4x4xf32>) -> tensor<1x2x2x2xf32> attributes {arch = "", kernel = "mixr"} {
+  func.func @mlir_convolution_add_relu(%arg0: tensor<1x2x2x2xf32>, %arg1: tensor<2x8x3x3xf32>, %arg2: tensor<1x8x4x4xf32>) -> tensor<1x2x2x2xf32> attributes {arch = "", kernel = "mixr", num_cu = 0 : i64} {
     %0 = migraphx.convolution(%arg2, %arg1) {dilation = [1, 1], group = 1 : i64, padding = [0, 0, 0, 0], padding_mode = 0 : i64, stride = [1, 1]} : (tensor<1x8x4x4xf32>, tensor<2x8x3x3xf32>) -> tensor<1x2x2x2xf32>
     %1 = migraphx.add(%0, %arg0) : (tensor<1x2x2x2xf32>, tensor<1x2x2x2xf32>) -> tensor<1x2x2x2xf32>
     %2 = migraphx.relu(%1) : (tensor<1x2x2x2xf32>) -> tensor<1x2x2x2xf32>
@@ -191,7 +191,7 @@ TEST_CASE(quant_dot_add)
 {
     const std::string mlir_output = R"__migraphx__(
 module {
-  func.func @mlir_quant_dot_add(%arg0: tensor<1x5x4xi8>, %arg1: tensor<1x4x3xi8>, %arg2: tensor<1x5x3xi32>) -> tensor<1x5x3xi32> attributes {arch = "", kernel = "mixr"} {
+  func.func @mlir_quant_dot_add(%arg0: tensor<1x5x4xi8>, %arg1: tensor<1x4x3xi8>, %arg2: tensor<1x5x3xi32>) -> tensor<1x5x3xi32> attributes {arch = "", kernel = "mixr", num_cu = 0 : i64} {
     %0 = migraphx.quant_dot(%arg0, %arg1) : (tensor<1x5x4xi8>, tensor<1x4x3xi8>) -> tensor<1x5x3xi32>
     %1 = migraphx.add(%0, %arg2) : (tensor<1x5x3xi32>, tensor<1x5x3xi32>) -> tensor<1x5x3xi32>
     return %1 : tensor<1x5x3xi32>
@@ -218,7 +218,7 @@ TEST_CASE(dot_add)
 {
     const std::string mlir_output = R"__migraphx__(
 module {
-  func.func @mlir_dot_add(%arg0: tensor<1x5x4xf32>, %arg1: tensor<1x4x3xf32>, %arg2: tensor<1x5x3xf32>) -> tensor<1x5x3xf32> attributes {arch = "", kernel = "mixr"} {
+  func.func @mlir_dot_add(%arg0: tensor<1x5x4xf32>, %arg1: tensor<1x4x3xf32>, %arg2: tensor<1x5x3xf32>) -> tensor<1x5x3xf32> attributes {arch = "", kernel = "mixr", num_cu = 0 : i64} {
     %0 = migraphx.dot(%arg0, %arg1) : (tensor<1x5x4xf32>, tensor<1x4x3xf32>) -> tensor<1x5x3xf32>
     %1 = migraphx.add(%0, %arg2) : (tensor<1x5x3xf32>, tensor<1x5x3xf32>) -> tensor<1x5x3xf32>
     return %1 : tensor<1x5x3xf32>
@@ -244,7 +244,7 @@ TEST_CASE(conv_int8_dequantize_quantize)
 {
     const std::string mlir_output = R"__migraphx__(
 module {
-  func.func @mlir_quant_convolution_dequantizelinear_quantizelinear(%arg0: tensor<2x8x3x3xi8>, %arg1: tensor<1x8x4x4xi8>, %arg2: tensor<1x2x2x2xf32>, %arg3: tensor<1x2x2x2xi32>) -> tensor<1x2x2x2xi32> attributes {arch = "", kernel = "mixr"} {
+  func.func @mlir_quant_convolution_dequantizelinear_quantizelinear(%arg0: tensor<2x8x3x3xi8>, %arg1: tensor<1x8x4x4xi8>, %arg2: tensor<1x2x2x2xf32>, %arg3: tensor<1x2x2x2xi32>) -> tensor<1x2x2x2xi32> attributes {arch = "", kernel = "mixr", num_cu = 0 : i64} {
       %0 = migraphx.quant_convolution(%arg1, %arg0) {dilation = [1, 1], group = 1 : i64, padding = [0, 0, 0, 0], padding_mode = 0 : i64, stride = [1, 1]} : (tensor<1x8x4x4xi8>, tensor<2x8x3x3xi8>) -> tensor<1x2x2x2xi32>
       %1 = migraphx.dequantizelinear(%0, %arg2, %arg3) : (tensor<1x2x2x2xi32>, tensor<1x2x2x2xf32>, tensor<1x2x2x2xi32>) -> tensor<1x2x2x2xf32>
       %2 = migraphx.quantizelinear(%1, %arg2, %arg3) : (tensor<1x2x2x2xf32>, tensor<1x2x2x2xf32>, tensor<1x2x2x2xi32>) -> tensor<1x2x2x2xi32>
@@ -277,7 +277,7 @@ TEST_CASE(dot_convert)
 {
     const std::string mlir_output = R"__migraphx__(
 module {
-  func.func @mlir_dot_convert(%arg0: tensor<1x5x4xf32>, %arg1: tensor<1x4x3xf32>) -> tensor<1x5x3xf16> attributes {arch = "", kernel = "mixr"} {
+  func.func @mlir_dot_convert(%arg0: tensor<1x5x4xf32>, %arg1: tensor<1x4x3xf32>) -> tensor<1x5x3xf16> attributes {arch = "", kernel = "mixr", num_cu = 0 : i64} {
     %0 = migraphx.dot(%arg0, %arg1) : (tensor<1x5x4xf32>, tensor<1x4x3xf32>) -> tensor<1x5x3xf32>
     %1 = migraphx.convert(%0) {target_type  =  1  :  i64} : (tensor<1x5x3xf32>) -> tensor<1x5x3xf16>
     return %1 : tensor<1x5x3xf16>
@@ -303,7 +303,7 @@ TEST_CASE(dot_where)
 {
     const std::string mlir_output = R"__migraphx__(
 module {
-  func.func @mlir_dot_where(%arg0: tensor<1x5x4xf32>, %arg1: tensor<1x4x3xf32>, %arg2: tensor<1x5x3xi8>, %arg3: tensor<1x5x3xf32>) -> tensor<1x5x3xf32> attributes {arch = "", kernel = "mixr"} {
+  func.func @mlir_dot_where(%arg0: tensor<1x5x4xf32>, %arg1: tensor<1x4x3xf32>, %arg2: tensor<1x5x3xi8>, %arg3: tensor<1x5x3xf32>) -> tensor<1x5x3xf32> attributes {arch = "", kernel = "mixr", num_cu = 0 : i64} {
     %0 = migraphx.dot(%arg0, %arg1) : (tensor<1x5x4xf32>, tensor<1x4x3xf32>) -> tensor<1x5x3xf32>
     %1 = migraphx.where(%arg2, %0, %arg3) : (tensor<1x5x3xi8>, tensor<1x5x3xf32>, tensor<1x5x3xf32>) -> tensor<1x5x3xf32>
     return %1 : tensor<1x5x3xf32>

test/gpu/quantization.cpp

@@ -24,7 +24,7 @@
 #include <iostream>
 #include <vector>
 #include <migraphx/gpu/fuse_mlir.hpp>
-#include <migraphx/operators.hpp>
+#include <migraphx/make_op.hpp>
 #include <migraphx/instruction.hpp>
 #include <migraphx/quantization.hpp>
 #include <migraphx/generate.hpp>
@@ -90,7 +90,7 @@ TEST_CASE(int8_quantization)
         migraphx::shape sc{migraphx::shape::float_type, {5, 8}};
         auto pa = mm->add_parameter("a", sa);
         auto pb = mm->add_parameter("b", sb);
-        mm->add_instruction(migraphx::op::dot{}, pa, pb);
+        mm->add_instruction(migraphx::make_op("dot"), pa, pb);
 
         return p;
     };

test/include/test.hpp

@@ -22,6 +22,7 @@
  * THE SOFTWARE.
  */
 
+#include <atomic>
 #include <algorithm>
 #include <cassert>
 #include <cstdio>
@@ -342,11 +343,19 @@ inline std::ostream& operator<<(std::ostream& os, const color& c)
     return os;
 }
 
+inline std::atomic<int>& failures()
+{
+    // NOLINTNEXTLINE
+    static std::atomic<int> f = 0;
+    return f;
+}
+
 template <class T, class F>
 void failed(T x, const char* msg, const char* func, const char* file, int line, F f)
 {
     if(not bool(x.value()))
     {
+        failures()++;
         std::cout << func << std::endl;
         std::cout << file << ":" << line << ":" << std::endl;
         std::cout << color::bold << color::fg_red << "    FAILED: " << color::reset << msg << " "
@@ -586,13 +595,21 @@ struct driver
         {
             try
             {
+                failures() = 0;
                 f();
             }
+            // cppcheck-suppress EmptyCatchStatement
             catch(const failure_error&)
             {
-                msg = "Test failure";
             }
         }
+        if(msg.empty() and failures() != 0)
+        {
+            if(failures() == 1)
+                msg = "Test failure";
+            else
+                msg = std::to_string(failures()) + " test failures";
+        }
         if(msg.empty())
         {
             out() << color::fg_green << "[ COMPLETE ] " << color::reset << color::bold << name
@@ -685,8 +702,8 @@ inline void run(int argc, const char* argv[])
 // NOLINTNEXTLINE
 #define CHECK(...) \
     test::failed(  \
-        test::capture{}->*__VA_ARGS__, #__VA_ARGS__, __PRETTY_FUNCTION__, __FILE__, __LINE__, [] { \
-        })
+        TEST_CAPTURE(__VA_ARGS__), #__VA_ARGS__, __PRETTY_FUNCTION__, __FILE__, __LINE__, [] {})
+
 // NOLINTNEXTLINE
 #define EXPECT(...)                         \
     test::failed(TEST_CAPTURE(__VA_ARGS__), \

test/inline_module_test.cpp

@@ -26,7 +26,6 @@
 #include <migraphx/pass_manager.hpp>
 #include <migraphx/instruction.hpp>
 #include <basic_ops.hpp>
-#include <migraphx/operators.hpp>
 #include <migraphx/make_op.hpp>
 
 #include <test.hpp>

test/insert_pad_test.cpp

@@ -26,8 +26,8 @@
 #include <migraphx/insert_pad.hpp>
 #include <migraphx/pass_manager.hpp>
 #include <migraphx/instruction.hpp>
+#include <migraphx/op/common.hpp>
 #include <basic_ops.hpp>
-#include <migraphx/operators.hpp>
 #include <migraphx/make_op.hpp>
 
 #include <test.hpp>
@@ -58,10 +58,11 @@ create_conv(migraphx::instruction_ref& l_img,
     migraphx::shape s_weights{migraphx::shape::int32_type, {4, channels, 3, 3}};
     std::vector<int32_t> weights(4 * channels * 3 * 3);
     auto l_weights = m.add_literal(migraphx::literal{s_weights, weights});
-    migraphx::op::convolution op;
-    op.padding_mode = padding_mode;
-    op.padding      = {0, 0, 1, 1};
-    return m.add_instruction(op, l_img, l_weights);
+    return m.add_instruction(
+        migraphx::make_op("convolution",
+                          {{"padding_mode", padding_mode}, {"padding", {0, 0, 1, 1}}}),
+        l_img,
+        l_weights);
 }
 
 TEST_CASE(rewrite_pad)

test/layout_nhwc.cpp

@@ -24,7 +24,6 @@
 #include <migraphx/layout_nhwc.hpp>
 #include <migraphx/dead_code_elimination.hpp>
 #include <migraphx/pass_manager.hpp>
-#include <migraphx/operators.hpp>
 #include <migraphx/generate.hpp>
 #include <migraphx/ranges.hpp>
 #include <migraphx/instruction.hpp>

test/memory_coloring_test.cpp

@@ -89,17 +89,13 @@ bool is_overlap_load(migraphx::instruction_ref a, migraphx::instruction_ref b)
 
 bool is_disjoint(const std::vector<migraphx::instruction_ref>& inss)
 {
-    for(auto ins1 : inss)
-    {
-        for(auto ins2 : inss)
-        {
+    return std::none_of(inss.begin(), inss.end(), [&](auto ins1) {
+        return std::none_of(inss.begin(), inss.end(), [&](auto ins2) {
             if(ins1 == ins2)
-                continue;
-            if(is_overlap_load(ins1, ins2))
-                return false;
-        }
-    }
                 return true;
+            return is_overlap_load(ins1, ins2);
+        });
+    });
 }
 
 TEST_CASE(test1)

test/onnx/.onnxrt-commit

-21a71d52bd2074b770807b209939ec11e2c64fa7
+a476dbf430ac8315550474a78d47bf182f202d7c

test/onnx/gen_onnx.py

@@ -6414,6 +6414,30 @@ def slice_test():
     return ([node], [x], [y])
 
 
+@onnx_test()
+def slice_constant_test():
+    y = helper.make_tensor_value_info('1', TensorProto.FLOAT, [1, 2])
+
+    x_tensor = helper.make_tensor(name='x_tensor',
+                                  data_type=TensorProto.FLOAT,
+                                  dims=[3, 2],
+                                  vals=[0, 1, 2, 3, 4, 5])
+
+    x = onnx.helper.make_node('Constant',
+                              inputs=[],
+                              outputs=['x'],
+                              value=x_tensor)
+
+    node = onnx.helper.make_node('Slice',
+                                 inputs=['x'],
+                                 axes=[0, 1],
+                                 starts=[1, 0],
+                                 ends=[2, 2],
+                                 outputs=['1'])
+
+    return ([x, node], [], [y])
+
+
 @onnx_test()
 def slice_dyn_test():
     x = helper.make_tensor_value_info('0', TensorProto.FLOAT, [None, None, 2])
@@ -6746,6 +6770,92 @@ def slice_max_end_test():
     return ([node], [x], [y])
 
 
+@onnx_test()
+def slice_var_input_static0():
+    data = helper.make_tensor_value_info('data', TensorProto.FLOAT, [3, 2])
+    starts = helper.make_tensor_value_info('starts', TensorProto.INT32, [2])
+    ends = helper.make_tensor_value_info('ends', TensorProto.INT32, [2])
+    output = helper.make_tensor_value_info('output', TensorProto.FLOAT, [1, 2])
+
+    node = onnx.helper.make_node('Slice',
+                                 inputs=['data', 'starts', 'ends'],
+                                 axes=[0, 1],
+                                 outputs=['output'])
+
+    return ([node], [data, starts, ends], [output])
+
+
+@onnx_test()
+def slice_var_input_static1():
+    data = helper.make_tensor_value_info('data', TensorProto.FLOAT, [3, 2])
+    starts = helper.make_tensor_value_info('starts', TensorProto.INT64, [2])
+    ends = helper.make_tensor_value_info('ends', TensorProto.INT64, [2])
+    axes = helper.make_tensor_value_info('axes', TensorProto.INT64, [2])
+    output = helper.make_tensor_value_info('output', TensorProto.FLOAT, [1, 2])
+
+    node = onnx.helper.make_node('Slice',
+                                 inputs=['data', 'starts', 'ends', 'axes'],
+                                 outputs=['output'])
+
+    return ([node], [data, starts, ends, axes], [output])
+
+
+@onnx_test()
+def slice_var_input_dyn0():
+    data = helper.make_tensor_value_info('data', TensorProto.FLOAT, [None, 2])
+    starts = helper.make_tensor_value_info('starts', TensorProto.INT32, [2])
+    ends = helper.make_tensor_value_info('ends', TensorProto.INT32, [2])
+    output = helper.make_tensor_value_info('output', TensorProto.FLOAT, [1, 2])
+
+    node = onnx.helper.make_node('Slice',
+                                 inputs=['data', 'starts', 'ends'],
+                                 axes=[0, 1],
+                                 outputs=['output'])
+
+    return ([node], [data, starts, ends], [output])
+
+
+@onnx_test()
+def slice_var_input_dyn1():
+    data = helper.make_tensor_value_info('data', TensorProto.FLOAT, [None, 2])
+    starts = helper.make_tensor_value_info('starts', TensorProto.INT32, [2])
+    ends = helper.make_tensor_value_info('ends', TensorProto.INT32, [2])
+    axes = helper.make_tensor_value_info('axes', TensorProto.INT32, [2])
+    output = helper.make_tensor_value_info('output', TensorProto.FLOAT, [1, 2])
+
+    node = onnx.helper.make_node('Slice',
+                                 inputs=['data', 'starts', 'ends', 'axes'],
+                                 outputs=['output'])
+
+    return ([node], [data, starts, ends, axes], [output])
+
+
+@onnx_test()
+def slice_var_input_steps_error():
+    step = np.array([2, 1])
+    step_tensor = helper.make_tensor(name="step",
+                                     data_type=TensorProto.INT32,
+                                     dims=step.shape,
+                                     vals=step.astype(int))
+    arg_step = helper.make_node("Constant",
+                                inputs=[],
+                                outputs=['arg_step'],
+                                value=step_tensor)
+
+    data = helper.make_tensor_value_info('data', TensorProto.FLOAT, [3, 2])
+    starts = helper.make_tensor_value_info('starts', TensorProto.FLOAT, [2])
+    ends = helper.make_tensor_value_info('ends', TensorProto.FLOAT, [2])
+    axes = helper.make_tensor_value_info('axes', TensorProto.FLOAT, [2])
+    output = helper.make_tensor_value_info('output', TensorProto.FLOAT, [1, 2])
+
+    node = onnx.helper.make_node(
+        'Slice',
+        inputs=['data', 'starts', 'ends', 'axes', 'arg_step'],
+        outputs=['output'])
+
+    return ([arg_step, node], [data, starts, ends, axes], [output])
+
+
 @onnx_test()
 def softmax_test():
     x = helper.make_tensor_value_info('0', TensorProto.FLOAT, [1, 3])

test/onnx/onnx_rnn_test.cpp

@@ -24,7 +24,7 @@
 #include <iostream>
 #include <vector>
 #include <migraphx/literal.hpp>
-#include <migraphx/operators.hpp>
+#include <migraphx/op/common.hpp>
 #include <migraphx/program.hpp>
 #include <migraphx/instruction.hpp>
 #include <migraphx/pass_manager.hpp>

test/onnx/onnx_test.cpp

@@ -4715,11 +4715,13 @@ TEST_CASE(quantizelinear_test)
     auto div     = mm->add_instruction(migraphx::make_op("div"), l0, l1_mbcast);
     auto round   = mm->add_instruction(migraphx::make_op("round"), div);
     auto s       = round->get_shape();
-    std::vector<int> min_data(s.elements(), 0);
-    std::vector<int> max_data(s.elements(), 255);
-    auto min_arg = mm->add_literal(s, min_data);
-    auto max_arg = mm->add_literal(s, max_data);
-    auto clip    = mm->add_instruction(migraphx::make_op("clip"), round, min_arg, max_arg);
+    auto min_arg = mm->add_literal(migraphx::literal{migraphx::shape{s.type()}, {0}});
+    auto max_arg = mm->add_literal(migraphx::literal{migraphx::shape{s.type()}, {255}});
+    auto min_mbcast =
+        mm->add_instruction(migraphx::make_op("multibroadcast", {{"out_lens", s.lens()}}), min_arg);
+    auto max_mbcast =
+        mm->add_instruction(migraphx::make_op("multibroadcast", {{"out_lens", s.lens()}}), max_arg);
+    auto clip = mm->add_instruction(migraphx::make_op("clip"), round, min_mbcast, max_mbcast);
     mm->add_instruction(
         migraphx::make_op("convert",
                           {{"target_type", migraphx::to_value(migraphx::shape::uint8_type)}}),
@@ -4744,11 +4746,13 @@ TEST_CASE(quantizelinear_int32_test)
     auto div     = mm->add_instruction(migraphx::make_op("div"), l0, l1_mbcast);
     auto round   = mm->add_instruction(migraphx::make_op("round"), div);
     auto s       = round->get_shape();
-    std::vector<int> min_data(s.elements(), 0);
-    std::vector<int> max_data(s.elements(), 255);
-    auto min_arg = mm->add_literal(s, min_data);
-    auto max_arg = mm->add_literal(s, max_data);
-    auto clip    = mm->add_instruction(migraphx::make_op("clip"), round, min_arg, max_arg);
+    auto min_arg = mm->add_literal(migraphx::literal{migraphx::shape{s.type()}, {0}});
+    auto max_arg = mm->add_literal(migraphx::literal{migraphx::shape{s.type()}, {255}});
+    auto min_mbcast =
+        mm->add_instruction(migraphx::make_op("multibroadcast", {{"out_lens", s.lens()}}), min_arg);
+    auto max_mbcast =
+        mm->add_instruction(migraphx::make_op("multibroadcast", {{"out_lens", s.lens()}}), max_arg);
+    auto clip = mm->add_instruction(migraphx::make_op("clip"), round, min_mbcast, max_mbcast);
     mm->add_instruction(
         migraphx::make_op("convert",
                           {{"target_type", migraphx::to_value(migraphx::shape::uint8_type)}}),
@@ -4777,11 +4781,13 @@ TEST_CASE(quantizelinear_zero_point_test)
         l2_mbcast);
     auto add     = mm->add_instruction(migraphx::make_op("add"), round, l2_mbcast);
     auto s       = round->get_shape();
-    std::vector<int> min_data(s.elements(), -128);
-    std::vector<int> max_data(s.elements(), 127);
-    auto min_arg = mm->add_literal(s, min_data);
-    auto max_arg = mm->add_literal(s, max_data);
-    auto clip    = mm->add_instruction(migraphx::make_op("clip"), add, min_arg, max_arg);
+    auto min_arg = mm->add_literal(migraphx::literal{migraphx::shape{s.type()}, {-128}});
+    auto max_arg = mm->add_literal(migraphx::literal{migraphx::shape{s.type()}, {127}});
+    auto min_mbcast =
+        mm->add_instruction(migraphx::make_op("multibroadcast", {{"out_lens", s.lens()}}), min_arg);
+    auto max_mbcast =
+        mm->add_instruction(migraphx::make_op("multibroadcast", {{"out_lens", s.lens()}}), max_arg);
+    auto clip = mm->add_instruction(migraphx::make_op("clip"), add, min_mbcast, max_mbcast);
     mm->add_instruction(
         migraphx::make_op("convert",
                           {{"target_type", migraphx::to_value(migraphx::shape::int8_type)}}),
@@ -4814,11 +4820,13 @@ migraphx::program make_quantizelinear_axis_prog()
         l2_bcast);
     auto add     = mm->add_instruction(migraphx::make_op("add"), round, l2_bcast);
     auto s       = round->get_shape();
-    std::vector<int> min_data(s.elements(), -128);
-    std::vector<int> max_data(s.elements(), 127);
-    auto min_arg = mm->add_literal(s, min_data);
-    auto max_arg = mm->add_literal(s, max_data);
-    auto clip    = mm->add_instruction(migraphx::make_op("clip"), add, min_arg, max_arg);
+    auto min_arg = mm->add_literal(migraphx::literal{migraphx::shape{s.type()}, {-128}});
+    auto max_arg = mm->add_literal(migraphx::literal{migraphx::shape{s.type()}, {127}});
+    auto min_mbcast =
+        mm->add_instruction(migraphx::make_op("multibroadcast", {{"out_lens", s.lens()}}), min_arg);
+    auto max_mbcast =
+        mm->add_instruction(migraphx::make_op("multibroadcast", {{"out_lens", s.lens()}}), max_arg);
+    auto clip = mm->add_instruction(migraphx::make_op("clip"), add, min_mbcast, max_mbcast);
     mm->add_instruction(
         migraphx::make_op("convert",
                           {{"target_type", migraphx::to_value(migraphx::shape::int8_type)}}),
@@ -6286,6 +6294,19 @@ TEST_CASE(slice_test)
     EXPECT(p == prog);
 }
 
+TEST_CASE(slice_constant_test)
+{
+    migraphx::program p;
+    auto* mm = p.get_main_module();
+    auto l0  = mm->add_literal(migraphx::literal{
+        migraphx::shape{migraphx::shape::float_type, {3, 2}}, {0, 1, 2, 3, 4, 5}});
+    mm->add_instruction(
+        migraphx::make_op("slice", {{"axes", {0, 1}}, {"starts", {1, 0}}, {"ends", {2, 2}}}), l0);
+    auto prog = optimize_onnx("slice_constant_test.onnx");
+
+    EXPECT(p == prog);
+}
+
 TEST_CASE(slice_dyn_test)
 {
     migraphx::program p;
@@ -6418,6 +6439,74 @@ TEST_CASE(slice_max_end_test)
     EXPECT(p == prog);
 }
 
+TEST_CASE(slice_var_input_static0)
+{
+    migraphx::program p;
+    auto* mm    = p.get_main_module();
+    auto data   = mm->add_parameter("data", migraphx::shape{migraphx::shape::float_type, {3, 2}});
+    auto starts = mm->add_parameter("starts", migraphx::shape{migraphx::shape::int32_type, {2}});
+    auto ends   = mm->add_parameter("ends", migraphx::shape{migraphx::shape::int32_type, {2}});
+    mm->add_instruction(migraphx::make_op("slice", {{"axes", {0, 1}}}), data, starts, ends);
+    auto prog = optimize_onnx("slice_var_input_static0.onnx");
+
+    EXPECT(p == prog);
+}
+
+TEST_CASE(slice_var_input_static1)
+{
+    migraphx::program p;
+    auto* mm    = p.get_main_module();
+    auto data   = mm->add_parameter("data", migraphx::shape{migraphx::shape::float_type, {3, 2}});
+    auto starts = mm->add_parameter("starts", migraphx::shape{migraphx::shape::int64_type, {2}});
+    auto ends   = mm->add_parameter("ends", migraphx::shape{migraphx::shape::int64_type, {2}});
+    auto axes   = mm->add_parameter("axes", migraphx::shape{migraphx::shape::int64_type, {2}});
+    mm->add_instruction(migraphx::make_op("slice"), data, starts, ends, axes);
+    auto prog = optimize_onnx("slice_var_input_static1.onnx");
+
+    EXPECT(p == prog);
+}
+
+TEST_CASE(slice_var_input_dyn0)
+{
+    migraphx::program p;
+    auto* mm = p.get_main_module();
+    auto data =
+        mm->add_parameter("data", migraphx::shape{migraphx::shape::float_type, {{3, 8}, {2, 2}}});
+    auto starts = mm->add_parameter("starts", migraphx::shape{migraphx::shape::int32_type, {2}});
+    auto ends   = mm->add_parameter("ends", migraphx::shape{migraphx::shape::int32_type, {2}});
+    auto ret =
+        mm->add_instruction(migraphx::make_op("slice", {{"axes", {0, 1}}}), data, starts, ends);
+    mm->add_return({ret});
+
+    migraphx::onnx_options options;
+    options.default_dyn_dim_value = {3, 8};
+    auto prog                     = parse_onnx("slice_var_input_dyn0.onnx", options);
+    EXPECT(p == prog);
+}
+
+TEST_CASE(slice_var_input_dyn1)
+{
+    migraphx::program p;
+    auto* mm = p.get_main_module();
+    auto data =
+        mm->add_parameter("data", migraphx::shape{migraphx::shape::float_type, {{3, 8}, {2, 2}}});
+    auto starts = mm->add_parameter("starts", migraphx::shape{migraphx::shape::int32_type, {2}});
+    auto ends   = mm->add_parameter("ends", migraphx::shape{migraphx::shape::int32_type, {2}});
+    auto axes   = mm->add_parameter("axes", migraphx::shape{migraphx::shape::int32_type, {2}});
+    auto ret    = mm->add_instruction(migraphx::make_op("slice"), data, starts, ends, axes);
+    mm->add_return({ret});
+
+    migraphx::onnx_options options;
+    options.default_dyn_dim_value = {3, 8};
+    auto prog                     = parse_onnx("slice_var_input_dyn1.onnx", options);
+    EXPECT(p == prog);
+}
+
+TEST_CASE(slice_var_input_steps_error)
+{
+    EXPECT(test::throws([&] { migraphx::parse_onnx("slice_var_input_steps_error.onnx"); }));
+}
+
 TEST_CASE(softmax_test)
 {
     migraphx::program p;