From 3223492fd484268c92094620c9c43f3881552f92 Mon Sep 17 00:00:00 2001
From: gouzi <530971494@qq.com>
Date: Thu, 18 Apr 2024 22:13:46 +0800
Subject: [PATCH 01/21] add log

---
 .../eager/to_static/run_program_op_func.h     | 20 +++++
 .../eager/to_static/run_program_op_node.h     | 82 +++++++++++++------
 paddle/fluid/framework/op_desc.cc             |  4 +
 paddle/fluid/framework/type_defs.cc           |  4 +-
 paddle/fluid/framework/type_defs.h            |  6 +-
 paddle/fluid/pybind/op_function_common.cc     | 28 ++++++-
 .../jit/dy2static/pir_partial_program.py      |  8 +-
 test/dygraph_to_static/test_no_gradient.py    |  5 +-
 8 files changed, 120 insertions(+), 37 deletions(-)

diff --git a/paddle/fluid/eager/to_static/run_program_op_func.h b/paddle/fluid/eager/to_static/run_program_op_func.h
index cdb4de66ae189..76837c1be2163 100644
--- a/paddle/fluid/eager/to_static/run_program_op_func.h
+++ b/paddle/fluid/eager/to_static/run_program_op_func.h
@@ -317,6 +317,26 @@ inline void pir_run_program_ad_func(
     // Set Attributes
     grad_node->SetAttrMap(attrs);
 
+    auto* backward_program =
+        PADDLE_GET_CONST(::pir::Program*, attrs.at("backward_program"));
+
+    auto* forward_program =
+        PADDLE_GET_CONST(::pir::Program*, attrs.at("forward_program"));
+
+    auto testkey = PADDLE_GET_CONST(std::string, attrs.at("testkey"));
+
+    VLOG(1) << "[pir_run_program_ad_func] testkey: " << testkey;
+    // TODO(gouzil): 对比一下这里的attrs的地址
+    VLOG(1) << "[pir_run_program_ad_func] attrs: " << &attrs;
+
+    VLOG(1) << "[pir_run_program_ad_func] backward_program: "
+            << backward_program;
+    VLOG(1) << "[pir_run_program_ad_func] backward_program: "
+            << backward_program->num_ops();
+    VLOG(1) << "[pir_run_program_ad_func] forward_program: " << forward_program;
+    VLOG(1) << "[pir_run_program_ad_func] forward_program: "
+            << forward_program->num_ops();
+
     // Clear unused x vars
     auto filter_x = pir_filter_unused_input_var_in_backward(x_tmp, "bx", attrs);
     // Set TensorWrappers
diff --git a/paddle/fluid/eager/to_static/run_program_op_node.h b/paddle/fluid/eager/to_static/run_program_op_node.h
index fdcad4f439c3b..505015a0056c2 100644
--- a/paddle/fluid/eager/to_static/run_program_op_node.h
+++ b/paddle/fluid/eager/to_static/run_program_op_node.h
@@ -467,21 +467,18 @@ inline void PirRunProgramAPI(
   auto param_values =
       PADDLE_GET_CONST(std::vector<::pir::Value>, attrs.at("fp"));
 
-  auto *forward_global_block =
-      PADDLE_GET_CONST(::pir::Block *, attrs.at("forward_global_block"));
-  auto *backward_global_block =
-      PADDLE_GET_CONST(::pir::Block *, attrs.at("backward_global_block"));
-
   auto *forward_program =
-      forward_global_block->GetParentOp()->GetParentProgram();
+      PADDLE_GET_CONST(::pir::Program *, attrs.at("forward_program"));
+  auto *backward_program =
+      PADDLE_GET_CONST(::pir::Program *, attrs.at("backward_program"));
+
+  ::pir::Block *forward_global_block = forward_program->block();
 
   if (FLAGS_print_ir) {
     std::ostringstream print_stream;
     print_stream << "ForwardProgram is :\n";
     forward_program->Print(print_stream);
     if (!is_test) {
-      auto *backward_program =
-          backward_global_block->GetParentOp()->GetParentProgram();
       print_stream << "BackwardProgram is:\n";
       backward_program->Print(print_stream);
     } else {
@@ -1046,11 +1043,6 @@ inline void PirRunProgramGradAPI(
 
   VLOG(4) << "global_inner_scope:" << global_inner_scope;
 
-  auto *backward_global_block =
-      PADDLE_GET_CONST(::pir::Block *, attrs.at("backward_global_block"));
-  auto *backward_program =
-      backward_global_block->GetParentOp()->GetParentProgram();
-
   auto output_grad_values =
       PADDLE_GET_CONST(std::vector<::pir::Value>, attrs.at("bo_g"));
   auto forward_input_values =
@@ -1068,27 +1060,63 @@ inline void PirRunProgramGradAPI(
 
   details::Trans2ContiguousTensorsInplace(out_grad);
 
+  auto testkey = PADDLE_GET_CONST(std::string, attrs.at("testkey"));
+
+  std::cout << "backward_program get start" << std::endl;
+  auto *backward_program =
+      PADDLE_GET_CONST(::pir::Program *, attrs.at("backward_program"));
+
+  auto *forward_program =
+      PADDLE_GET_CONST(::pir::Program *, attrs.at("forward_program"));
+
+  VLOG(0) << "[PirRunProgramGradAPI] testkey: " << testkey;
+  VLOG(0) << "[PirRunProgramGradAPI] backward_program addr: "
+          << backward_program;
+  VLOG(0) << "[PirRunProgramGradAPI] forward_program addr: " << forward_program;
+  VLOG(0) << backward_program->num_ops();
+  VLOG(0) << forward_program->num_ops();
+  std::cout << "backward_program get end" << std::endl;
+
+  std::cout << "backward_program block get start" << std::endl;
+  auto pb = backward_program->block();
+  VLOG(0) << pb;
+  VLOG(0) << pb->num_ops();
+  VLOG(0) << pb->empty();
+  std::cout << "backward_program block get end" << std::endl;
+
   // share x, param, middles, output_grads, out into scope.
+  VLOG(1) << "out_grad start";
+  details::ShareTensorsIntoScopeByValue(backward_program->block(),
+                                        out_grad,
+                                        output_grad_values,
+                                        global_inner_scope);
+  VLOG(1) << "out_grad end";
   details::ShareTensorsIntoScopeByValue(
-      backward_global_block, out_grad, output_grad_values, global_inner_scope);
-  details::ShareTensorsIntoScopeByValue(
-      backward_global_block, x, forward_input_values, global_inner_scope);
-  details::ShareTensorsIntoScopeByValue(backward_global_block,
+      backward_program->block(), x, forward_input_values, global_inner_scope);
+  VLOG(1) << "x end";
+  details::ShareTensorsIntoScopeByValue(backward_program->block(),
                                         middles,
                                         forward_middle_values,
                                         global_inner_scope);
+  VLOG(1) << "middles end";
+  details::ShareTensorsIntoScopeByValue(backward_program->block(),
+                                        out,
+                                        forward_output_values,
+                                        global_inner_scope);
+  VLOG(1) << "out end";
   details::ShareTensorsIntoScopeByValue(
-      backward_global_block, out, forward_output_values, global_inner_scope);
-  details::ShareTensorsIntoScopeByValue(
-      backward_global_block, params, parameter_values, global_inner_scope);
+      backward_program->block(), params, parameter_values, global_inner_scope);
+  VLOG(1) << "params end";
 
   // Clear out and middles to avoid hold memory until backward finish.
   out.clear();
   middles.clear();
+  VLOG(1) << "out and middles clear end";
 
   auto &cache = paddle::framework::InterpreterCoreInfoCache::Instance();
   std::shared_ptr<paddle::framework::InterpreterCore> interpreter_core =
       nullptr;
+
   if (!cache.Has(program_id,
                  global_inner_scope,
                  place_hash_key,
@@ -1138,10 +1166,10 @@ inline void PirRunProgramGradAPI(
     // get all eager gc vars
     std::set<std::string> skip_eager_delete_vars;
     auto skip_names = details::GetNameFromValue(
-        backward_global_block, x_grad_values, false, true);
+        backward_program->block(), x_grad_values, false, true);
     skip_eager_delete_vars.insert(skip_names.begin(), skip_names.end());
     skip_names = details::GetNameFromValue(
-        backward_global_block, p_grad_values, false, true);
+        backward_program->block(), p_grad_values, false, true);
     skip_eager_delete_vars.insert(skip_names.begin(), skip_names.end());
     interpreter_core->SetSkipGcVars(skip_eager_delete_vars);
     cache.UpdateSkipEagerDeleteVars(program_id,
@@ -1174,7 +1202,7 @@ inline void PirRunProgramGradAPI(
     }
   }
 
-  if (!backward_global_block->empty()) {
+  if (!backward_program->block()->empty()) {
     paddle::platform::RecordEvent record_event(
         "interpreter_core_run",
         paddle::platform::TracerEventType::UserDefined,
@@ -1189,9 +1217,11 @@ inline void PirRunProgramGradAPI(
         "fetch_and_gc", paddle::platform::TracerEventType::UserDefined, 1);
     // Step 4. get outputs
     details::ShareTensorsFromScopeByValue(
-        backward_global_block, x_grad, x_grad_values, global_inner_scope);
-    details::ShareTensorsFromScopeByValue(
-        backward_global_block, params_grad, p_grad_values, global_inner_scope);
+        backward_program->block(), x_grad, x_grad_values, global_inner_scope);
+    details::ShareTensorsFromScopeByValue(backward_program->block(),
+                                          params_grad,
+                                          p_grad_values,
+                                          global_inner_scope);
     VLOG(4) << "after backward gc all vars";
     global_inner_scope->SetCanReused(true);
     details::GcScope(global_inner_scope);
diff --git a/paddle/fluid/framework/op_desc.cc b/paddle/fluid/framework/op_desc.cc
index 32c520711d978..0187edec5cc10 100644
--- a/paddle/fluid/framework/op_desc.cc
+++ b/paddle/fluid/framework/op_desc.cc
@@ -26,6 +26,7 @@ limitations under the License. */
 #include "paddle/fluid/operators/ops_extra_info.h"
 #include "paddle/phi/common/complex.h"
 #include "paddle/pir/include/core/block.h"
+#include "paddle/pir/include/core/program.h"
 #include "paddle/pir/include/core/value.h"
 #include "paddle/utils/blank.h"
 
@@ -977,6 +978,9 @@ struct SetAttrDescVisitor {
   void operator()(const std::vector<pir::Block *> &v) const {
     // just do nothing.
   }
+  void operator()(const ::pir::Program &v) const {
+    // just do nothing.
+  }
   void operator()(const std::vector<VarDesc *> &v) const {
     std::vector<std::string> var_names;
     for (auto var : v) {
diff --git a/paddle/fluid/framework/type_defs.cc b/paddle/fluid/framework/type_defs.cc
index d8a6546ea718d..a17ff2d8a81d7 100644
--- a/paddle/fluid/framework/type_defs.cc
+++ b/paddle/fluid/framework/type_defs.cc
@@ -39,7 +39,9 @@ template class variant<paddle::blank,
                        paddle::experimental::Scalar,
                        std::vector<paddle::experimental::Scalar>,
                        ::pir::Block*,
-                       std::vector<::pir::Value>>;
+                       std::vector<::pir::Value>,
+                       ::pir::Program,
+                       ProgramDesc*>;
 }  // namespace paddle
 REGISTER_LOG_SIMPLY_STR(paddle::framework::AttributeMap);
 REGISTER_LOG_SIMPLY_STR(paddle::framework::Attribute);
diff --git a/paddle/fluid/framework/type_defs.h b/paddle/fluid/framework/type_defs.h
index 61f133ceb082a..56e0eb7644894 100644
--- a/paddle/fluid/framework/type_defs.h
+++ b/paddle/fluid/framework/type_defs.h
@@ -27,6 +27,7 @@ limitations under the License. */
 #include "paddle/phi/common/scalar.h"
 #include "paddle/phi/core/infermeta_utils.h"
 #include "paddle/pir/include/core/block.h"
+#include "paddle/pir/include/core/program.h"
 #include "paddle/pir/include/core/value.h"
 #include "paddle/utils/blank.h"
 #include "paddle/utils/small_vector.h"
@@ -40,6 +41,7 @@ class InferShapeContext;
 class InferVarTypeContext;
 class VarDesc;
 class BlockDesc;
+class ProgramDesc;
 class Variable;
 class InferNoNeedBufferVarsFN;
 
@@ -67,7 +69,9 @@ using Attribute = paddle::variant<paddle::blank,
                                   paddle::experimental::Scalar,
                                   std::vector<paddle::experimental::Scalar>,
                                   ::pir::Block*,
-                                  std::vector<::pir::Value>>;
+                                  std::vector<::pir::Value>,
+                                  ::pir::Program*,
+                                  ProgramDesc*>;
 using AttributeMap = std::unordered_map<std::string, Attribute>;
 
 using OpCreator =
diff --git a/paddle/fluid/pybind/op_function_common.cc b/paddle/fluid/pybind/op_function_common.cc
index f8f1424ded243..9e0d0bb30d19e 100644
--- a/paddle/fluid/pybind/op_function_common.cc
+++ b/paddle/fluid/pybind/op_function_common.cc
@@ -36,6 +36,7 @@
 #include "paddle/phi/common/complex.h"
 #include "paddle/pir/include/core/block.h"
 #include "paddle/pir/include/core/op_result.h"
+#include "paddle/pir/include/core/region.h"
 #include "paddle/pir/include/core/value.h"
 
 namespace paddle {
@@ -858,6 +859,24 @@ void CastPyArg2AttrIRBlock(PyObject* obj,
   attrs[key] = reinterpret_cast<::pir::Block*&>(vh[0]);
 }
 
+void CastPyArg2AttrIRProgram(PyObject* obj,
+                             paddle::framework::AttributeMap& attrs,  // NOLINT
+                             const std::string& key,
+                             const std::string& op_type,
+                             ssize_t arg_pos) {
+  VLOG(1) << "After Process pir::Program*";
+  ::pybind11::detail::instance* inst =
+      (::pybind11::detail::instance*)obj;  // NOLINT
+  void** vh = inst->simple_layout ? inst->simple_value_holder
+                                  : &inst->nonsimple.values_and_holders[0];
+
+  ::pir::Program* program = reinterpret_cast<::pir::Program*>(vh[0]);
+
+  attrs[key] = program;
+  // attrs[key] = reinterpret_cast<::pir::Program*>(vh[0]);
+  //  attrs[key] = vh[0];
+}
+
 void CastPyArg2AttrValues(PyObject* obj,
                           paddle::framework::AttributeMap& attrs,  // NOLINT
                           const std::string& key,
@@ -998,6 +1017,7 @@ void ConstructAttrMapForRunProgram(
                         attr_end));
 
   PyObject* obj = nullptr;
+  attrs["testkey"] = std::string("testvalue");
   for (ssize_t arg_pos = attr_start; arg_pos < attr_end; arg_pos += 2) {
     VLOG(1) << "Start Process " << arg_pos;
     Py_ssize_t key_len = 0;
@@ -1020,11 +1040,11 @@ void ConstructAttrMapForRunProgram(
 
     if (std::set<std::string>({"cuda_graph_capture_mode"}).count(key)) {
       CastPyArg2AttrString(obj, attrs, key, op_type, arg_pos);
-    } else if (std::set<std::string>({"global_block",
-                                      "forward_global_block",
-                                      "backward_global_block"})
-                   .count(key)) {
+    } else if (std::set<std::string>({"global_block"}).count(key)) {
       CastPyArg2AttrIRBlock(obj, attrs, key, op_type, arg_pos);
+    } else if (std::set<std::string>({"forward_program", "backward_program"})
+                   .count(key)) {
+      CastPyArg2AttrIRProgram(obj, attrs, key, op_type, arg_pos);
     } else if (std::set<std::string>({"is_test", "use_interpretorcore"})
                    .count(key)) {
       CastPyArg2AttrBoolean(obj, attrs, key, op_type, arg_pos);
diff --git a/python/paddle/jit/dy2static/pir_partial_program.py b/python/paddle/jit/dy2static/pir_partial_program.py
index cddac384213ab..9aaa4eb5971d4 100644
--- a/python/paddle/jit/dy2static/pir_partial_program.py
+++ b/python/paddle/jit/dy2static/pir_partial_program.py
@@ -898,10 +898,10 @@ def _prune_unused_params(self, program):
 
     def _prepare_attributes(self):
         attrs = [
-            'forward_global_block',
-            self.program.forward_program.global_block(),
-            'backward_global_block',
-            self.program.backward_program.global_block(),
+            'forward_program',
+            self.program.forward_program,
+            'backward_program',
+            self.program.backward_program,
             'is_test',
             not self.training,
             'program_id',
diff --git a/test/dygraph_to_static/test_no_gradient.py b/test/dygraph_to_static/test_no_gradient.py
index 1bd3a02f54ede..b2f1346248f0a 100644
--- a/test/dygraph_to_static/test_no_gradient.py
+++ b/test/dygraph_to_static/test_no_gradient.py
@@ -12,10 +12,11 @@
 # See the License for the specific language governing permissions and
 # limitations under the License.
 
+breakpoint()  # noqa: T100
 import unittest
 
 import numpy
-from dygraph_to_static_utils import Dy2StTestBase
+from dygraph_to_static_utils import Dy2StTestBase, test_ast_only, test_pir_only
 
 import paddle
 
@@ -33,6 +34,8 @@ def main_func(x, index):
 
 
 class TestNoGradientCase(Dy2StTestBase):
+    @test_ast_only
+    @test_pir_only
     def test_no_gradient(self):
         paddle.disable_static()
         x = paddle.randn([10, 3])

From c1398d17916c9f6552acf2d27e88d20b7cdcfb10 Mon Sep 17 00:00:00 2001
From: gouzi <530971494@qq.com>
Date: Thu, 18 Apr 2024 22:27:27 +0800
Subject: [PATCH 02/21] add log

---
 paddle/fluid/framework/type_defs.cc | 2 +-
 paddle/pir/src/core/program.cc      | 1 +
 2 files changed, 2 insertions(+), 1 deletion(-)

diff --git a/paddle/fluid/framework/type_defs.cc b/paddle/fluid/framework/type_defs.cc
index a17ff2d8a81d7..ad4eebf4ed3a8 100644
--- a/paddle/fluid/framework/type_defs.cc
+++ b/paddle/fluid/framework/type_defs.cc
@@ -40,7 +40,7 @@ template class variant<paddle::blank,
                        std::vector<paddle::experimental::Scalar>,
                        ::pir::Block*,
                        std::vector<::pir::Value>,
-                       ::pir::Program,
+                       ::pir::Program*,
                        ProgramDesc*>;
 }  // namespace paddle
 REGISTER_LOG_SIMPLY_STR(paddle::framework::AttributeMap);
diff --git a/paddle/pir/src/core/program.cc b/paddle/pir/src/core/program.cc
index 19d08f094fd4c..6576cf00d40b8 100644
--- a/paddle/pir/src/core/program.cc
+++ b/paddle/pir/src/core/program.cc
@@ -23,6 +23,7 @@ Program::Program(IrContext* context) {
 }
 
 Program::~Program() {
+  VLOG(1) << "[Program] Destroy Program";
   if (module_) {
     module_.Destroy();
   }

From 2a7f67d21d5c3e311b2ba7764086380ddb38f807 Mon Sep 17 00:00:00 2001
From: gouzi <530971494@qq.com>
Date: Fri, 19 Apr 2024 10:29:15 +0800
Subject: [PATCH 03/21] add log

---
 paddle/fluid/framework/op_desc.cc         |  2 +-
 paddle/fluid/pybind/op_function_common.cc | 11 ++++++++---
 paddle/fluid/pybind/pir.cc                |  2 +-
 3 files changed, 10 insertions(+), 5 deletions(-)

diff --git a/paddle/fluid/framework/op_desc.cc b/paddle/fluid/framework/op_desc.cc
index 0187edec5cc10..c131c93909705 100644
--- a/paddle/fluid/framework/op_desc.cc
+++ b/paddle/fluid/framework/op_desc.cc
@@ -978,7 +978,7 @@ struct SetAttrDescVisitor {
   void operator()(const std::vector<pir::Block *> &v) const {
     // just do nothing.
   }
-  void operator()(const ::pir::Program &v) const {
+  void operator()(const std::shared_ptr<pir::Program> &v) const {
     // just do nothing.
   }
   void operator()(const std::vector<VarDesc *> &v) const {
diff --git a/paddle/fluid/pybind/op_function_common.cc b/paddle/fluid/pybind/op_function_common.cc
index 9e0d0bb30d19e..6db25f5f9c213 100644
--- a/paddle/fluid/pybind/op_function_common.cc
+++ b/paddle/fluid/pybind/op_function_common.cc
@@ -870,9 +870,14 @@ void CastPyArg2AttrIRProgram(PyObject* obj,
   void** vh = inst->simple_layout ? inst->simple_value_holder
                                   : &inst->nonsimple.values_and_holders[0];
 
-  ::pir::Program* program = reinterpret_cast<::pir::Program*>(vh[0]);
-
-  attrs[key] = program;
+  // ::pir::Program* program = reinterpret_cast<::pir::Program*>(vh[0]);
+  std::shared_ptr<::pir::Program> program =
+      reinterpret_cast<std::shared_ptr<::pir::Program>&>(vh[0]);
+  // TODO(gouzil): 试一下pybind11能不能使用智能指针作为参数
+  // pir::IrMapping mapper;
+  attrs[key] = program.get();
+  // attrs[key] = program.get();
+  // attrs[key] = program->Clone(mapper);
   // attrs[key] = reinterpret_cast<::pir::Program*>(vh[0]);
   //  attrs[key] = vh[0];
 }
diff --git a/paddle/fluid/pybind/pir.cc b/paddle/fluid/pybind/pir.cc
index 4176ecf0bbcbb..7b02c0676c9c6 100644
--- a/paddle/fluid/pybind/pir.cc
+++ b/paddle/fluid/pybind/pir.cc
@@ -255,7 +255,7 @@ void BindProgram(py::module *m) {
   )DOC");
   program
       .def(py::init([]() {
-        return std::make_unique<Program>(pir::IrContext::Instance());
+        return std::make_shared<Program>(pir::IrContext::Instance());
       }))
       .def("__str__",
            [](const std::shared_ptr<Program> &self) {

From 733ac038c15130259f13d0dbe6d4e68edd6ae447 Mon Sep 17 00:00:00 2001
From: gouzi <530971494@qq.com>
Date: Fri, 19 Apr 2024 15:41:44 +0800
Subject: [PATCH 04/21] add handle

---
 paddle/fluid/pybind/op_function_common.cc | 1 +
 1 file changed, 1 insertion(+)

diff --git a/paddle/fluid/pybind/op_function_common.cc b/paddle/fluid/pybind/op_function_common.cc
index 6db25f5f9c213..91c969d7c1695 100644
--- a/paddle/fluid/pybind/op_function_common.cc
+++ b/paddle/fluid/pybind/op_function_common.cc
@@ -870,6 +870,7 @@ void CastPyArg2AttrIRProgram(PyObject* obj,
   void** vh = inst->simple_layout ? inst->simple_value_holder
                                   : &inst->nonsimple.values_and_holders[0];
 
+  ::pybind11::handle(obj).inc_ref();
   // ::pir::Program* program = reinterpret_cast<::pir::Program*>(vh[0]);
   std::shared_ptr<::pir::Program> program =
       reinterpret_cast<std::shared_ptr<::pir::Program>&>(vh[0]);

From 5c90eacfe7e972eb8f859c66ed98fbafdf92c463 Mon Sep 17 00:00:00 2001
From: gouzi <530971494@qq.com>
Date: Fri, 19 Apr 2024 17:47:43 +0800
Subject: [PATCH 05/21] remove breakpoint

---
 test/dygraph_to_static/test_no_gradient.py | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/test/dygraph_to_static/test_no_gradient.py b/test/dygraph_to_static/test_no_gradient.py
index b2f1346248f0a..2283c08a3d299 100644
--- a/test/dygraph_to_static/test_no_gradient.py
+++ b/test/dygraph_to_static/test_no_gradient.py
@@ -12,7 +12,7 @@
 # See the License for the specific language governing permissions and
 # limitations under the License.
 
-breakpoint()  # noqa: T100
+# breakpoint()
 import unittest
 
 import numpy

From 64a56b0209d7bd6f8032e677753e51bd0e053803 Mon Sep 17 00:00:00 2001
From: gouzi <530971494@qq.com>
Date: Sat, 20 Apr 2024 23:05:47 +0800
Subject: [PATCH 06/21] use `std::shared_ptr<::pir::Program>>`

---
 .../eager/to_static/run_program_op_func.h     |  8 +++----
 .../eager/to_static/run_program_op_node.h     | 24 +++++++++++--------
 paddle/fluid/framework/type_defs.cc           |  3 +--
 paddle/fluid/framework/type_defs.h            |  3 +--
 paddle/fluid/pybind/op_function_common.cc     | 23 ++++++------------
 5 files changed, 27 insertions(+), 34 deletions(-)

diff --git a/paddle/fluid/eager/to_static/run_program_op_func.h b/paddle/fluid/eager/to_static/run_program_op_func.h
index 76837c1be2163..f305bb2a191ae 100644
--- a/paddle/fluid/eager/to_static/run_program_op_func.h
+++ b/paddle/fluid/eager/to_static/run_program_op_func.h
@@ -317,11 +317,11 @@ inline void pir_run_program_ad_func(
     // Set Attributes
     grad_node->SetAttrMap(attrs);
 
-    auto* backward_program =
-        PADDLE_GET_CONST(::pir::Program*, attrs.at("backward_program"));
+    auto backward_program = PADDLE_GET_CONST(std::shared_ptr<::pir::Program>,
+                                             attrs.at("backward_program"));
 
-    auto* forward_program =
-        PADDLE_GET_CONST(::pir::Program*, attrs.at("forward_program"));
+    auto forward_program = PADDLE_GET_CONST(std::shared_ptr<::pir::Program>,
+                                            attrs.at("forward_program"));
 
     auto testkey = PADDLE_GET_CONST(std::string, attrs.at("testkey"));
 
diff --git a/paddle/fluid/eager/to_static/run_program_op_node.h b/paddle/fluid/eager/to_static/run_program_op_node.h
index 505015a0056c2..9f01e4a8a683a 100644
--- a/paddle/fluid/eager/to_static/run_program_op_node.h
+++ b/paddle/fluid/eager/to_static/run_program_op_node.h
@@ -467,10 +467,10 @@ inline void PirRunProgramAPI(
   auto param_values =
       PADDLE_GET_CONST(std::vector<::pir::Value>, attrs.at("fp"));
 
-  auto *forward_program =
-      PADDLE_GET_CONST(::pir::Program *, attrs.at("forward_program"));
-  auto *backward_program =
-      PADDLE_GET_CONST(::pir::Program *, attrs.at("backward_program"));
+  auto forward_program = PADDLE_GET_CONST(std::shared_ptr<::pir::Program>,
+                                          attrs.at("forward_program"));
+  auto backward_program = PADDLE_GET_CONST(std::shared_ptr<::pir::Program>,
+                                           attrs.at("backward_program"));
 
   ::pir::Block *forward_global_block = forward_program->block();
 
@@ -511,7 +511,7 @@ inline void PirRunProgramAPI(
         forward_global_block, params, param_values, global_inner_scope);
     // Step 2. create new interpretercore
     auto passed_kernel_program =
-        paddle::framework::ApplyIrPass(forward_program, place);
+        paddle::framework::ApplyIrPass(forward_program.get(), place);
     if (FLAGS_print_ir) {
       std::ostringstream print_stream;
       print_stream << "LoweredProgram( AfterPass ) is :\n";
@@ -1063,11 +1063,11 @@ inline void PirRunProgramGradAPI(
   auto testkey = PADDLE_GET_CONST(std::string, attrs.at("testkey"));
 
   std::cout << "backward_program get start" << std::endl;
-  auto *backward_program =
-      PADDLE_GET_CONST(::pir::Program *, attrs.at("backward_program"));
+  auto backward_program = PADDLE_GET_CONST(std::shared_ptr<::pir::Program>,
+                                           attrs.at("backward_program"));
 
-  auto *forward_program =
-      PADDLE_GET_CONST(::pir::Program *, attrs.at("forward_program"));
+  auto forward_program = PADDLE_GET_CONST(std::shared_ptr<::pir::Program>,
+                                          attrs.at("forward_program"));
 
   VLOG(0) << "[PirRunProgramGradAPI] testkey: " << testkey;
   VLOG(0) << "[PirRunProgramGradAPI] backward_program addr: "
@@ -1083,6 +1083,10 @@ inline void PirRunProgramGradAPI(
   VLOG(0) << pb->num_ops();
   VLOG(0) << pb->empty();
   std::cout << "backward_program block get end" << std::endl;
+  std::ostringstream print_stream;
+  print_stream << "backward_program is :\n";
+  backward_program->Print(print_stream);
+  std::cout << print_stream.str() << std::endl;
 
   // share x, param, middles, output_grads, out into scope.
   VLOG(1) << "out_grad start";
@@ -1129,7 +1133,7 @@ inline void PirRunProgramGradAPI(
     VLOG(2) << "No interpretercore cache, so create a new interpretercore";
     // Step 1. share input_vars & parameters into scope
     auto passed_kernel_program =
-        paddle::framework::ApplyIrPass(backward_program, place);
+        paddle::framework::ApplyIrPass(backward_program.get(), place);
     if (FLAGS_print_ir) {
       std::ostringstream print_stream;
       print_stream << "LoweredProgram( AfterPass | Backward ) is :\n";
diff --git a/paddle/fluid/framework/type_defs.cc b/paddle/fluid/framework/type_defs.cc
index ad4eebf4ed3a8..6d350f1fe1e6c 100644
--- a/paddle/fluid/framework/type_defs.cc
+++ b/paddle/fluid/framework/type_defs.cc
@@ -40,8 +40,7 @@ template class variant<paddle::blank,
                        std::vector<paddle::experimental::Scalar>,
                        ::pir::Block*,
                        std::vector<::pir::Value>,
-                       ::pir::Program*,
-                       ProgramDesc*>;
+                       std::shared_ptr<::pir::Program>>;
 }  // namespace paddle
 REGISTER_LOG_SIMPLY_STR(paddle::framework::AttributeMap);
 REGISTER_LOG_SIMPLY_STR(paddle::framework::Attribute);
diff --git a/paddle/fluid/framework/type_defs.h b/paddle/fluid/framework/type_defs.h
index 56e0eb7644894..d401d82df63c9 100644
--- a/paddle/fluid/framework/type_defs.h
+++ b/paddle/fluid/framework/type_defs.h
@@ -70,8 +70,7 @@ using Attribute = paddle::variant<paddle::blank,
                                   std::vector<paddle::experimental::Scalar>,
                                   ::pir::Block*,
                                   std::vector<::pir::Value>,
-                                  ::pir::Program*,
-                                  ProgramDesc*>;
+                                  std::shared_ptr<::pir::Program>>;
 using AttributeMap = std::unordered_map<std::string, Attribute>;
 
 using OpCreator =
diff --git a/paddle/fluid/pybind/op_function_common.cc b/paddle/fluid/pybind/op_function_common.cc
index 91c969d7c1695..f939ccf287ca2 100644
--- a/paddle/fluid/pybind/op_function_common.cc
+++ b/paddle/fluid/pybind/op_function_common.cc
@@ -865,22 +865,13 @@ void CastPyArg2AttrIRProgram(PyObject* obj,
                              const std::string& op_type,
                              ssize_t arg_pos) {
   VLOG(1) << "After Process pir::Program*";
-  ::pybind11::detail::instance* inst =
-      (::pybind11::detail::instance*)obj;  // NOLINT
-  void** vh = inst->simple_layout ? inst->simple_value_holder
-                                  : &inst->nonsimple.values_and_holders[0];
-
-  ::pybind11::handle(obj).inc_ref();
-  // ::pir::Program* program = reinterpret_cast<::pir::Program*>(vh[0]);
-  std::shared_ptr<::pir::Program> program =
-      reinterpret_cast<std::shared_ptr<::pir::Program>&>(vh[0]);
-  // TODO(gouzil): 试一下pybind11能不能使用智能指针作为参数
-  // pir::IrMapping mapper;
-  attrs[key] = program.get();
-  // attrs[key] = program.get();
-  // attrs[key] = program->Clone(mapper);
-  // attrs[key] = reinterpret_cast<::pir::Program*>(vh[0]);
-  //  attrs[key] = vh[0];
+  const std::shared_ptr<::pir::Program> program =
+      ::py::handle(obj).cast<std::shared_ptr<::pir::Program>>();
+  VLOG(1) << "[CastPyArg2AttrIRProgram] program use count: "
+          << program.use_count();
+  attrs[key] = program;
+  VLOG(1) << "[CastPyArg2AttrIRProgram] program use count: "
+          << program.use_count();
 }
 
 void CastPyArg2AttrValues(PyObject* obj,

From 8aba5efadbf508bf1c99ff6db2031c498a5c2b32 Mon Sep 17 00:00:00 2001
From: gouzi <530971494@qq.com>
Date: Wed, 24 Apr 2024 19:16:01 +0800
Subject: [PATCH 07/21] copy #58180

---
 paddle/fluid/eager/to_static/run_program_op_func.h | 11 +----------
 paddle/fluid/eager/to_static/run_program_op_node.h |  4 ++++
 2 files changed, 5 insertions(+), 10 deletions(-)

diff --git a/paddle/fluid/eager/to_static/run_program_op_func.h b/paddle/fluid/eager/to_static/run_program_op_func.h
index f305bb2a191ae..3bde86812d1a0 100644
--- a/paddle/fluid/eager/to_static/run_program_op_func.h
+++ b/paddle/fluid/eager/to_static/run_program_op_func.h
@@ -346,16 +346,7 @@ inline void pir_run_program_ad_func(
 
     grad_node->SetStepScope(step_scope);  // just for set useable.
 
-    // Set Grad out rank as same as fwd input and set stop gradient to bwd
-    // NOTE(@xiongkun): Not every tensor in x(list of tensor) is required
-    // gradient. for example: x[1] is not used for output, the x[1] is ignored.
-
-    std::vector<const paddle::Tensor*> x_require_grad;
-    for (size_t i = 0; i < x.size(); ++i) {
-      x_require_grad.push_back(&x[i]);
-    }
-
-    grad_node->SetGradOutMeta(x_require_grad, /*slot id*/ 0);
+    grad_node->SetGradOutMeta(x, /*slot id*/ 0);
     grad_node->SetGradOutMeta(params, /*slot id*/ 1);
 
     // TODO(@xiongkun): rewrite by new ir representation.
diff --git a/paddle/fluid/eager/to_static/run_program_op_node.h b/paddle/fluid/eager/to_static/run_program_op_node.h
index 387a55f77be1b..8e3893324eb2c 100644
--- a/paddle/fluid/eager/to_static/run_program_op_node.h
+++ b/paddle/fluid/eager/to_static/run_program_op_node.h
@@ -1505,6 +1505,10 @@ class PirGradNodeRunProgram : public egr::GradNodeBase {
     VLOG(3) << "End Eager Backward Node: PirGradNodeRunProgram";
 
     *executed_ = true;
+    egr::EagerUtils::FillZeroForEmptyOptionalGradOutput(&x_grad,
+                                                        this->OutputMeta()[0]);
+    egr::EagerUtils::FillZeroForEmptyOptionalGradOutput(&params_grad,
+                                                        this->OutputMeta()[1]);
     return {x_grad, params_grad};
   }
 

From 23bbd587de941f26bd659cb9f666f1416d1c04d1 Mon Sep 17 00:00:00 2001
From: gouzi <530971494@qq.com>
Date: Wed, 24 Apr 2024 19:34:16 +0800
Subject: [PATCH 08/21] clean log and open test

---
 .../eager/to_static/run_program_op_func.h     | 20 ------------
 .../eager/to_static/run_program_op_node.h     | 31 +------------------
 paddle/fluid/pybind/op_function_common.cc     |  4 ---
 paddle/pir/src/core/program.cc                |  1 -
 test/dygraph_to_static/test_no_gradient.py    |  6 ++--
 5 files changed, 3 insertions(+), 59 deletions(-)

diff --git a/paddle/fluid/eager/to_static/run_program_op_func.h b/paddle/fluid/eager/to_static/run_program_op_func.h
index 3bde86812d1a0..df0dc3946507d 100644
--- a/paddle/fluid/eager/to_static/run_program_op_func.h
+++ b/paddle/fluid/eager/to_static/run_program_op_func.h
@@ -317,26 +317,6 @@ inline void pir_run_program_ad_func(
     // Set Attributes
     grad_node->SetAttrMap(attrs);
 
-    auto backward_program = PADDLE_GET_CONST(std::shared_ptr<::pir::Program>,
-                                             attrs.at("backward_program"));
-
-    auto forward_program = PADDLE_GET_CONST(std::shared_ptr<::pir::Program>,
-                                            attrs.at("forward_program"));
-
-    auto testkey = PADDLE_GET_CONST(std::string, attrs.at("testkey"));
-
-    VLOG(1) << "[pir_run_program_ad_func] testkey: " << testkey;
-    // TODO(gouzil): 对比一下这里的attrs的地址
-    VLOG(1) << "[pir_run_program_ad_func] attrs: " << &attrs;
-
-    VLOG(1) << "[pir_run_program_ad_func] backward_program: "
-            << backward_program;
-    VLOG(1) << "[pir_run_program_ad_func] backward_program: "
-            << backward_program->num_ops();
-    VLOG(1) << "[pir_run_program_ad_func] forward_program: " << forward_program;
-    VLOG(1) << "[pir_run_program_ad_func] forward_program: "
-            << forward_program->num_ops();
-
     // Clear unused x vars
     auto filter_x = pir_filter_unused_input_var_in_backward(x_tmp, "bx", attrs);
     // Set TensorWrappers
diff --git a/paddle/fluid/eager/to_static/run_program_op_node.h b/paddle/fluid/eager/to_static/run_program_op_node.h
index 8e3893324eb2c..8c1928c8d8705 100644
--- a/paddle/fluid/eager/to_static/run_program_op_node.h
+++ b/paddle/fluid/eager/to_static/run_program_op_node.h
@@ -1060,57 +1060,29 @@ inline void PirRunProgramGradAPI(
 
   details::Trans2ContiguousTensorsInplace(out_grad);
 
-  auto testkey = PADDLE_GET_CONST(std::string, attrs.at("testkey"));
-
-  std::cout << "backward_program get start" << std::endl;
   auto backward_program = PADDLE_GET_CONST(std::shared_ptr<::pir::Program>,
                                            attrs.at("backward_program"));
 
   auto forward_program = PADDLE_GET_CONST(std::shared_ptr<::pir::Program>,
                                           attrs.at("forward_program"));
 
-  VLOG(0) << "[PirRunProgramGradAPI] testkey: " << testkey;
-  VLOG(0) << "[PirRunProgramGradAPI] backward_program addr: "
-          << backward_program;
-  VLOG(0) << "[PirRunProgramGradAPI] forward_program addr: " << forward_program;
-  VLOG(0) << backward_program->num_ops();
-  VLOG(0) << forward_program->num_ops();
-  std::cout << "backward_program get end" << std::endl;
-
-  std::cout << "backward_program block get start" << std::endl;
-  auto pb = backward_program->block();
-  VLOG(0) << pb;
-  VLOG(0) << pb->num_ops();
-  VLOG(0) << pb->empty();
-  std::cout << "backward_program block get end" << std::endl;
-  std::ostringstream print_stream;
-  print_stream << "backward_program is :\n";
-  backward_program->Print(print_stream);
-  std::cout << print_stream.str() << std::endl;
-
   // share x, param, middles, output_grads, out into scope.
-  VLOG(1) << "out_grad start";
   details::ShareTensorsIntoScopeByValue(backward_program->block(),
                                         out_grad,
                                         output_grad_values,
                                         global_inner_scope);
-  VLOG(1) << "out_grad end";
   details::ShareTensorsIntoScopeByValue(
       backward_program->block(), x, forward_input_values, global_inner_scope);
-  VLOG(1) << "x end";
   details::ShareTensorsIntoScopeByValue(backward_program->block(),
                                         middles,
                                         forward_middle_values,
                                         global_inner_scope);
-  VLOG(1) << "middles end";
   details::ShareTensorsIntoScopeByValue(backward_program->block(),
                                         out,
                                         forward_output_values,
                                         global_inner_scope);
-  VLOG(1) << "out end";
   details::ShareTensorsIntoScopeByValue(
       backward_program->block(), params, parameter_values, global_inner_scope);
-  VLOG(1) << "params end";
 
   // Clear out and middles to avoid hold memory until backward finish.
   out.clear();
@@ -1369,8 +1341,7 @@ class GradNodeRunProgram : public egr::GradNodeBase {
       if (x[i].is_dense_tensor()) {
         x_grad->emplace_back(std::make_shared<phi::DenseTensor>());
       } else if (x[i].is_selected_rows()) {
-        auto selected_row = std::make_shared<phi::SelectedRows>();
-        x_grad->emplace_back(selected_row);
+        x_grad->emplace_back(std::make_shared<phi::SelectedRows>());
       }
       x_grad->back().set_name(x_grad_names[i]);
     }
diff --git a/paddle/fluid/pybind/op_function_common.cc b/paddle/fluid/pybind/op_function_common.cc
index f939ccf287ca2..2a1a4c84d4225 100644
--- a/paddle/fluid/pybind/op_function_common.cc
+++ b/paddle/fluid/pybind/op_function_common.cc
@@ -867,11 +867,7 @@ void CastPyArg2AttrIRProgram(PyObject* obj,
   VLOG(1) << "After Process pir::Program*";
   const std::shared_ptr<::pir::Program> program =
       ::py::handle(obj).cast<std::shared_ptr<::pir::Program>>();
-  VLOG(1) << "[CastPyArg2AttrIRProgram] program use count: "
-          << program.use_count();
   attrs[key] = program;
-  VLOG(1) << "[CastPyArg2AttrIRProgram] program use count: "
-          << program.use_count();
 }
 
 void CastPyArg2AttrValues(PyObject* obj,
diff --git a/paddle/pir/src/core/program.cc b/paddle/pir/src/core/program.cc
index 6576cf00d40b8..19d08f094fd4c 100644
--- a/paddle/pir/src/core/program.cc
+++ b/paddle/pir/src/core/program.cc
@@ -23,7 +23,6 @@ Program::Program(IrContext* context) {
 }
 
 Program::~Program() {
-  VLOG(1) << "[Program] Destroy Program";
   if (module_) {
     module_.Destroy();
   }
diff --git a/test/dygraph_to_static/test_no_gradient.py b/test/dygraph_to_static/test_no_gradient.py
index 2283c08a3d299..84f7b032c2f4a 100644
--- a/test/dygraph_to_static/test_no_gradient.py
+++ b/test/dygraph_to_static/test_no_gradient.py
@@ -12,11 +12,10 @@
 # See the License for the specific language governing permissions and
 # limitations under the License.
 
-# breakpoint()
 import unittest
 
 import numpy
-from dygraph_to_static_utils import Dy2StTestBase, test_ast_only, test_pir_only
+from dygraph_to_static_utils import Dy2StTestBase, test_legacy_and_pt_and_pir
 
 import paddle
 
@@ -34,8 +33,7 @@ def main_func(x, index):
 
 
 class TestNoGradientCase(Dy2StTestBase):
-    @test_ast_only
-    @test_pir_only
+    @test_legacy_and_pt_and_pir
     def test_no_gradient(self):
         paddle.disable_static()
         x = paddle.randn([10, 3])

From 31be3b994dc8ed27350d8c59e1da114ca57dcc3c Mon Sep 17 00:00:00 2001
From: gouzi <530971494@qq.com>
Date: Wed, 24 Apr 2024 19:37:32 +0800
Subject: [PATCH 09/21] Rollback execution sequence

---
 paddle/fluid/eager/to_static/run_program_op_node.h | 9 +++------
 1 file changed, 3 insertions(+), 6 deletions(-)

diff --git a/paddle/fluid/eager/to_static/run_program_op_node.h b/paddle/fluid/eager/to_static/run_program_op_node.h
index 8c1928c8d8705..bcbc5fb48dafc 100644
--- a/paddle/fluid/eager/to_static/run_program_op_node.h
+++ b/paddle/fluid/eager/to_static/run_program_op_node.h
@@ -1043,6 +1043,9 @@ inline void PirRunProgramGradAPI(
 
   VLOG(4) << "global_inner_scope:" << global_inner_scope;
 
+  auto backward_program = PADDLE_GET_CONST(std::shared_ptr<::pir::Program>,
+                                           attrs.at("backward_program"));
+
   auto output_grad_values =
       PADDLE_GET_CONST(std::vector<::pir::Value>, attrs.at("bo_g"));
   auto forward_input_values =
@@ -1060,12 +1063,6 @@ inline void PirRunProgramGradAPI(
 
   details::Trans2ContiguousTensorsInplace(out_grad);
 
-  auto backward_program = PADDLE_GET_CONST(std::shared_ptr<::pir::Program>,
-                                           attrs.at("backward_program"));
-
-  auto forward_program = PADDLE_GET_CONST(std::shared_ptr<::pir::Program>,
-                                          attrs.at("forward_program"));
-
   // share x, param, middles, output_grads, out into scope.
   details::ShareTensorsIntoScopeByValue(backward_program->block(),
                                         out_grad,

From e005597da512f8c9aa2c927af92c01879e71c75c Mon Sep 17 00:00:00 2001
From: gouzi <530971494@qq.com>
Date: Wed, 24 Apr 2024 19:42:30 +0800
Subject: [PATCH 10/21] rm log

---
 paddle/fluid/eager/to_static/run_program_op_node.h | 1 -
 1 file changed, 1 deletion(-)

diff --git a/paddle/fluid/eager/to_static/run_program_op_node.h b/paddle/fluid/eager/to_static/run_program_op_node.h
index bcbc5fb48dafc..a8c0413cfb4d9 100644
--- a/paddle/fluid/eager/to_static/run_program_op_node.h
+++ b/paddle/fluid/eager/to_static/run_program_op_node.h
@@ -1084,7 +1084,6 @@ inline void PirRunProgramGradAPI(
   // Clear out and middles to avoid hold memory until backward finish.
   out.clear();
   middles.clear();
-  VLOG(1) << "out and middles clear end";
 
   auto &cache = paddle::framework::InterpreterCoreInfoCache::Instance();
   std::shared_ptr<paddle::framework::InterpreterCore> interpreter_core =

From 6277bc21ff0b8a43b53b62f81716f043d4ca0b91 Mon Sep 17 00:00:00 2001
From: gouzi <530971494@qq.com>
Date: Wed, 24 Apr 2024 19:44:01 +0800
Subject: [PATCH 11/21] rm `class ProgramDesc`

---
 paddle/fluid/framework/type_defs.h | 1 -
 1 file changed, 1 deletion(-)

diff --git a/paddle/fluid/framework/type_defs.h b/paddle/fluid/framework/type_defs.h
index d401d82df63c9..919da60601555 100644
--- a/paddle/fluid/framework/type_defs.h
+++ b/paddle/fluid/framework/type_defs.h
@@ -41,7 +41,6 @@ class InferShapeContext;
 class InferVarTypeContext;
 class VarDesc;
 class BlockDesc;
-class ProgramDesc;
 class Variable;
 class InferNoNeedBufferVarsFN;
 

From 4972090a7e651ef054d40ad2a1f5ba3f1e160e7e Mon Sep 17 00:00:00 2001
From: gouzi <530971494@qq.com>
Date: Wed, 24 Apr 2024 19:45:52 +0800
Subject: [PATCH 12/21] rm `test[key]`

---
 paddle/fluid/pybind/op_function_common.cc | 1 -
 1 file changed, 1 deletion(-)

diff --git a/paddle/fluid/pybind/op_function_common.cc b/paddle/fluid/pybind/op_function_common.cc
index 2a1a4c84d4225..52001185c2cd8 100644
--- a/paddle/fluid/pybind/op_function_common.cc
+++ b/paddle/fluid/pybind/op_function_common.cc
@@ -1010,7 +1010,6 @@ void ConstructAttrMapForRunProgram(
                         attr_end));
 
   PyObject* obj = nullptr;
-  attrs["testkey"] = std::string("testvalue");
   for (ssize_t arg_pos = attr_start; arg_pos < attr_end; arg_pos += 2) {
     VLOG(1) << "Start Process " << arg_pos;
     Py_ssize_t key_len = 0;

From 4bcbcc6d2b02dca1a3ca462bb6dabf0820108884 Mon Sep 17 00:00:00 2001
From: gouzi <530971494@qq.com>
Date: Wed, 24 Apr 2024 23:54:11 +0800
Subject: [PATCH 13/21] rm include

---
 paddle/fluid/pybind/op_function_common.cc | 1 -
 1 file changed, 1 deletion(-)

diff --git a/paddle/fluid/pybind/op_function_common.cc b/paddle/fluid/pybind/op_function_common.cc
index 52001185c2cd8..cc484f74ab22f 100644
--- a/paddle/fluid/pybind/op_function_common.cc
+++ b/paddle/fluid/pybind/op_function_common.cc
@@ -36,7 +36,6 @@
 #include "paddle/phi/common/complex.h"
 #include "paddle/pir/include/core/block.h"
 #include "paddle/pir/include/core/op_result.h"
-#include "paddle/pir/include/core/region.h"
 #include "paddle/pir/include/core/value.h"
 
 namespace paddle {

From 414fbda57c8fe694ce3577b3c4ce776c7b90ff98 Mon Sep 17 00:00:00 2001
From: gouzi <530971494@qq.com>
Date: Thu, 25 Apr 2024 18:24:39 +0800
Subject: [PATCH 14/21] fix jit.load for pir

---
 python/paddle/jit/pir_translated_layer.py | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

diff --git a/python/paddle/jit/pir_translated_layer.py b/python/paddle/jit/pir_translated_layer.py
index 1f50677821633..dee752b8ef7cc 100644
--- a/python/paddle/jit/pir_translated_layer.py
+++ b/python/paddle/jit/pir_translated_layer.py
@@ -343,10 +343,10 @@ def _run_dygraph(instance, input, program_holder):
     )
 
     attrs = [
-        'forward_global_block',
-        forward_program.global_block(),
-        'backward_global_block',
-        backward_program.global_block(),
+        'forward_program',
+        forward_program,
+        'backward_program',
+        backward_program,
         'is_test',
         instance._is_test,
         'program_id',

From 1e68e624935cbf56255480c699be669f46889f24 Mon Sep 17 00:00:00 2001
From: gouzi <530971494@qq.com>
Date: Thu, 25 Apr 2024 21:25:41 +0800
Subject: [PATCH 15/21] [ci][test] ignore test

---
 test/cpp/prim/test_eager_prim.cc | 287 ++++++++++++++++---------------
 1 file changed, 144 insertions(+), 143 deletions(-)

diff --git a/test/cpp/prim/test_eager_prim.cc b/test/cpp/prim/test_eager_prim.cc
index ac82ecac53081..4a13bf78025f3 100644
--- a/test/cpp/prim/test_eager_prim.cc
+++ b/test/cpp/prim/test_eager_prim.cc
@@ -17,159 +17,160 @@
 #include "glog/logging.h"
 #include "gtest/gtest.h"
 #include "paddle/common/flags.h"
-#include "paddle/fluid/eager/api/generated/eager_generated/forwards/dygraph_functions.h"
-#include "paddle/fluid/eager/api/utils/hook_utils.h"
-#include "paddle/fluid/eager/backward.h"
+// #include
+// "paddle/fluid/eager/api/generated/eager_generated/forwards/dygraph_functions.h"
+// #include "paddle/fluid/eager/api/utils/hook_utils.h"
+// #include "paddle/fluid/eager/backward.h"
 #include "paddle/fluid/prim/utils/utils.h"
-#include "paddle/phi/core/dense_tensor.h"
-#include "paddle/phi/core/kernel_registry.h"
-#include "paddle/phi/core/tensor_meta.h"
-#include "test/cpp/eager/test_utils.h"
-#include "test/cpp/prim/init_env_utils.h"
+// #include "paddle/phi/core/dense_tensor.h"
+// #include "paddle/phi/core/kernel_registry.h"
+// #include "paddle/phi/core/tensor_meta.h"
+// #include "test/cpp/eager/test_utils.h"
+// #include "test/cpp/prim/init_env_utils.h"
 
 COMMON_DECLARE_string(tensor_operants_mode);
 
 namespace paddle {
 namespace prim {
 
-TEST(EagerPrim, TanhBackwardTest) {
-  // 1. Initialized
-  eager_test::InitEnv(paddle::platform::CPUPlace());
-  FLAGS_tensor_operants_mode = "eager";
-  paddle::prim::InitTensorOperants();
-  // 2. pre
-  paddle::framework::DDim ddim = common::make_ddim({4, 16, 16, 32});
-  paddle::Tensor tensor0 =
-      eager_test::CreateTensorWithValue(ddim,
-                                        paddle::platform::CPUPlace(),
-                                        phi::DataType::FLOAT32,
-                                        phi::DataLayout::NCHW,
-                                        5.0 /*value*/,
-                                        true /*is_leaf*/);
-  ::egr::egr_utils_api::RetainGradForTensor(tensor0);
-  paddle::Tensor tensor1 =
-      eager_test::CreateTensorWithValue(ddim,
-                                        paddle::platform::CPUPlace(),
-                                        phi::DataType::FLOAT32,
-                                        phi::DataLayout::NCHW,
-                                        5.0 /*value*/,
-                                        true /*is_leaf*/);
-  ::egr::egr_utils_api::RetainGradForTensor(tensor1);
-  // 3. Run Forward once
-  paddle::Tensor out0 = tanh_ad_func(tensor0);
-  std::vector<paddle::Tensor> outs0 = {out0};
-  // Disable prim
-  PrimCommonUtils::SetBwdPrimEnabled(false);
-  ASSERT_FALSE(PrimCommonUtils::IsBwdPrimEnabled());
-  // 4. Run Backward
-  egr::Backward(outs0, {}, false);
+// TEST(EagerPrim, TanhBackwardTest) {
+//   // 1. Initialized
+//   eager_test::InitEnv(paddle::platform::CPUPlace());
+//   FLAGS_tensor_operants_mode = "eager";
+//   paddle::prim::InitTensorOperants();
+//   // 2. pre
+//   paddle::framework::DDim ddim = common::make_ddim({4, 16, 16, 32});
+//   paddle::Tensor tensor0 =
+//       eager_test::CreateTensorWithValue(ddim,
+//                                         paddle::platform::CPUPlace(),
+//                                         phi::DataType::FLOAT32,
+//                                         phi::DataLayout::NCHW,
+//                                         5.0 /*value*/,
+//                                         true /*is_leaf*/);
+//   ::egr::egr_utils_api::RetainGradForTensor(tensor0);
+//   paddle::Tensor tensor1 =
+//       eager_test::CreateTensorWithValue(ddim,
+//                                         paddle::platform::CPUPlace(),
+//                                         phi::DataType::FLOAT32,
+//                                         phi::DataLayout::NCHW,
+//                                         5.0 /*value*/,
+//                                         true /*is_leaf*/);
+//   ::egr::egr_utils_api::RetainGradForTensor(tensor1);
+//   // 3. Run Forward once
+//   paddle::Tensor out0 = tanh_ad_func(tensor0);
+//   std::vector<paddle::Tensor> outs0 = {out0};
+//   // Disable prim
+//   PrimCommonUtils::SetBwdPrimEnabled(false);
+//   ASSERT_FALSE(PrimCommonUtils::IsBwdPrimEnabled());
+//   // 4. Run Backward
+//   egr::Backward(outs0, {}, false);
 
-  paddle::Tensor out1 = tanh_ad_func(tensor1);
-  std::vector<paddle::Tensor> outs1 = {out1};
-  // Enable prim
-  PrimCommonUtils::SetBwdPrimEnabled(true);
-  ASSERT_TRUE(PrimCommonUtils::IsBwdPrimEnabled());
-  // 4. Run Backward
-  ::egr::Backward(outs1, {}, false);
-  VLOG(7)
-      << "Target Grad is: "
-      << std::static_pointer_cast<phi::DenseTensor>(
-             ::egr::EagerUtils::unsafe_autograd_meta(tensor0)->Grad().impl())
-             ->data<float>()[0];
-  VLOG(7)
-      << "Result Grad is: "
-      << std::static_pointer_cast<phi::DenseTensor>(
-             ::egr::EagerUtils::unsafe_autograd_meta(tensor1)->Grad().impl())
-             ->data<float>()[0];
-  // Examine Backward Grad
-  eager_test::CompareGradTensorWithValue<float>(
-      tensor1,
-      std::static_pointer_cast<phi::DenseTensor>(
-          ::egr::EagerUtils::unsafe_autograd_meta(tensor0)->Grad().impl())
-          ->data<float>()[0]);
-}
+//   paddle::Tensor out1 = tanh_ad_func(tensor1);
+//   std::vector<paddle::Tensor> outs1 = {out1};
+//   // Enable prim
+//   PrimCommonUtils::SetBwdPrimEnabled(true);
+//   ASSERT_TRUE(PrimCommonUtils::IsBwdPrimEnabled());
+//   // 4. Run Backward
+//   ::egr::Backward(outs1, {}, false);
+//   VLOG(7)
+//       << "Target Grad is: "
+//       << std::static_pointer_cast<phi::DenseTensor>(
+//              ::egr::EagerUtils::unsafe_autograd_meta(tensor0)->Grad().impl())
+//              ->data<float>()[0];
+//   VLOG(7)
+//       << "Result Grad is: "
+//       << std::static_pointer_cast<phi::DenseTensor>(
+//              ::egr::EagerUtils::unsafe_autograd_meta(tensor1)->Grad().impl())
+//              ->data<float>()[0];
+//   // Examine Backward Grad
+//   eager_test::CompareGradTensorWithValue<float>(
+//       tensor1,
+//       std::static_pointer_cast<phi::DenseTensor>(
+//           ::egr::EagerUtils::unsafe_autograd_meta(tensor0)->Grad().impl())
+//           ->data<float>()[0]);
+// }
 
-TEST(EagerPrim, LogicalOperantsTest) {
-  // 1. Initialized
-  eager_test::InitEnv(paddle::platform::CPUPlace());
-  FLAGS_tensor_operants_mode = "eager";
-  paddle::prim::InitTensorOperants();
-  // 2. pre
-  paddle::framework::DDim ddim = common::make_ddim({4, 16, 16, 32});
-  paddle::Tensor tensor0 =
-      eager_test::CreateTensorWithValue(ddim,
-                                        paddle::platform::CPUPlace(),
-                                        phi::DataType::INT32,
-                                        phi::DataLayout::NCHW,
-                                        1 /*value*/,
-                                        true /*is_leaf*/);
-  ::egr::egr_utils_api::RetainGradForTensor(tensor0);
-  paddle::Tensor tensor1 =
-      eager_test::CreateTensorWithValue(ddim,
-                                        paddle::platform::CPUPlace(),
-                                        phi::DataType::INT32,
-                                        phi::DataLayout::NCHW,
-                                        0 /*value*/,
-                                        true /*is_leaf*/);
-  ::egr::egr_utils_api::RetainGradForTensor(tensor1);
-  // 3. Run Forward once
-  paddle::Tensor out0 = tensor0 & tensor1;
-  paddle::Tensor out1 = bitwise_and_ad_func(tensor0, tensor1);
-  EXPECT_EQ(out0.data<int>()[0], out1.data<int>()[0]);
-  out0 = tensor0 | tensor1;
-  out1 = bitwise_or_ad_func(tensor0, tensor1);
-  EXPECT_EQ(out0.data<int>()[0], out1.data<int>()[0]);
-  out0 = tensor0 ^ tensor1;
-  out1 = bitwise_xor_ad_func(tensor0, tensor1);
-  EXPECT_EQ(out0.data<int>()[0], out1.data<int>()[0]);
-  out0 = ~tensor0;
-  out1 = bitwise_not_ad_func(tensor0);
-  EXPECT_EQ(out0.data<int>()[0], out1.data<int>()[0]);
-}
+// TEST(EagerPrim, LogicalOperantsTest) {
+//   // 1. Initialized
+//   eager_test::InitEnv(paddle::platform::CPUPlace());
+//   FLAGS_tensor_operants_mode = "eager";
+//   paddle::prim::InitTensorOperants();
+//   // 2. pre
+//   paddle::framework::DDim ddim = common::make_ddim({4, 16, 16, 32});
+//   paddle::Tensor tensor0 =
+//       eager_test::CreateTensorWithValue(ddim,
+//                                         paddle::platform::CPUPlace(),
+//                                         phi::DataType::INT32,
+//                                         phi::DataLayout::NCHW,
+//                                         1 /*value*/,
+//                                         true /*is_leaf*/);
+//   ::egr::egr_utils_api::RetainGradForTensor(tensor0);
+//   paddle::Tensor tensor1 =
+//       eager_test::CreateTensorWithValue(ddim,
+//                                         paddle::platform::CPUPlace(),
+//                                         phi::DataType::INT32,
+//                                         phi::DataLayout::NCHW,
+//                                         0 /*value*/,
+//                                         true /*is_leaf*/);
+//   ::egr::egr_utils_api::RetainGradForTensor(tensor1);
+//   // 3. Run Forward once
+//   paddle::Tensor out0 = tensor0 & tensor1;
+//   paddle::Tensor out1 = bitwise_and_ad_func(tensor0, tensor1);
+//   EXPECT_EQ(out0.data<int>()[0], out1.data<int>()[0]);
+//   out0 = tensor0 | tensor1;
+//   out1 = bitwise_or_ad_func(tensor0, tensor1);
+//   EXPECT_EQ(out0.data<int>()[0], out1.data<int>()[0]);
+//   out0 = tensor0 ^ tensor1;
+//   out1 = bitwise_xor_ad_func(tensor0, tensor1);
+//   EXPECT_EQ(out0.data<int>()[0], out1.data<int>()[0]);
+//   out0 = ~tensor0;
+//   out1 = bitwise_not_ad_func(tensor0);
+//   EXPECT_EQ(out0.data<int>()[0], out1.data<int>()[0]);
+// }
 
-TEST(EagerPrim, CompareOperantsTest) {
-  // 1. Initialized
-  eager_test::InitEnv(paddle::platform::CPUPlace());
-  FLAGS_tensor_operants_mode = "eager";
-  paddle::prim::InitTensorOperants();
-  // 2. pre
-  paddle::framework::DDim ddim = common::make_ddim({4, 16, 16, 32});
-  paddle::Tensor tensor0 =
-      eager_test::CreateTensorWithValue(ddim,
-                                        paddle::platform::CPUPlace(),
-                                        phi::DataType::INT32,
-                                        phi::DataLayout::NCHW,
-                                        1 /*value*/,
-                                        true /*is_leaf*/);
-  ::egr::egr_utils_api::RetainGradForTensor(tensor0);
-  paddle::Tensor tensor1 =
-      eager_test::CreateTensorWithValue(ddim,
-                                        paddle::platform::CPUPlace(),
-                                        phi::DataType::INT32,
-                                        phi::DataLayout::NCHW,
-                                        0 /*value*/,
-                                        true /*is_leaf*/);
-  ::egr::egr_utils_api::RetainGradForTensor(tensor1);
-  // 3. Run Forward once
-  paddle::Tensor out0 = (tensor0 < tensor1);
-  paddle::Tensor out1 = less_than_ad_func(tensor0, tensor1);
-  EXPECT_EQ(out0.data<bool>()[0], out1.data<bool>()[0]);
-  out0 = (tensor0 <= tensor1);
-  out1 = less_equal_ad_func(tensor0, tensor1);
-  EXPECT_EQ(out0.data<bool>()[0], out1.data<bool>()[0]);
-  out0 = (tensor0 == tensor1);
-  out1 = equal_ad_func(tensor0, tensor1);
-  EXPECT_EQ(out0.data<bool>()[0], out1.data<bool>()[0]);
-  out0 = (tensor0 != tensor1);
-  out1 = not_equal_ad_func(tensor0, tensor1);
-  EXPECT_EQ(out0.data<bool>()[0], out1.data<bool>()[0]);
-  out0 = (tensor0 > tensor1);
-  out1 = greater_than_ad_func(tensor0, tensor1);
-  EXPECT_EQ(out0.data<bool>()[0], out1.data<bool>()[0]);
-  out0 = (tensor0 >= tensor1);
-  out1 = greater_equal_ad_func(tensor0, tensor1);
-  EXPECT_EQ(out0.data<bool>()[0], out1.data<bool>()[0]);
-}
+// TEST(EagerPrim, CompareOperantsTest) {
+//   // 1. Initialized
+//   eager_test::InitEnv(paddle::platform::CPUPlace());
+//   FLAGS_tensor_operants_mode = "eager";
+//   paddle::prim::InitTensorOperants();
+//   // 2. pre
+//   paddle::framework::DDim ddim = common::make_ddim({4, 16, 16, 32});
+//   paddle::Tensor tensor0 =
+//       eager_test::CreateTensorWithValue(ddim,
+//                                         paddle::platform::CPUPlace(),
+//                                         phi::DataType::INT32,
+//                                         phi::DataLayout::NCHW,
+//                                         1 /*value*/,
+//                                         true /*is_leaf*/);
+//   ::egr::egr_utils_api::RetainGradForTensor(tensor0);
+//   paddle::Tensor tensor1 =
+//       eager_test::CreateTensorWithValue(ddim,
+//                                         paddle::platform::CPUPlace(),
+//                                         phi::DataType::INT32,
+//                                         phi::DataLayout::NCHW,
+//                                         0 /*value*/,
+//                                         true /*is_leaf*/);
+//   ::egr::egr_utils_api::RetainGradForTensor(tensor1);
+//   // 3. Run Forward once
+//   paddle::Tensor out0 = (tensor0 < tensor1);
+//   paddle::Tensor out1 = less_than_ad_func(tensor0, tensor1);
+//   EXPECT_EQ(out0.data<bool>()[0], out1.data<bool>()[0]);
+//   out0 = (tensor0 <= tensor1);
+//   out1 = less_equal_ad_func(tensor0, tensor1);
+//   EXPECT_EQ(out0.data<bool>()[0], out1.data<bool>()[0]);
+//   out0 = (tensor0 == tensor1);
+//   out1 = equal_ad_func(tensor0, tensor1);
+//   EXPECT_EQ(out0.data<bool>()[0], out1.data<bool>()[0]);
+//   out0 = (tensor0 != tensor1);
+//   out1 = not_equal_ad_func(tensor0, tensor1);
+//   EXPECT_EQ(out0.data<bool>()[0], out1.data<bool>()[0]);
+//   out0 = (tensor0 > tensor1);
+//   out1 = greater_than_ad_func(tensor0, tensor1);
+//   EXPECT_EQ(out0.data<bool>()[0], out1.data<bool>()[0]);
+//   out0 = (tensor0 >= tensor1);
+//   out1 = greater_equal_ad_func(tensor0, tensor1);
+//   EXPECT_EQ(out0.data<bool>()[0], out1.data<bool>()[0]);
+// }
 
 TEST(EagerPrim, TestFlags) {
   PrimCommonUtils::SetBwdPrimEnabled(true);

From cb6cdabafa771382c8ab783fd0060ac93b6eceb4 Mon Sep 17 00:00:00 2001
From: gouzi <530971494@qq.com>
Date: Thu, 25 Apr 2024 22:28:34 +0800
Subject: [PATCH 16/21] [ci][test] open `CompareOperantsTest`

---
 test/cpp/prim/test_eager_prim.cc | 95 ++++++++++++++++----------------
 1 file changed, 47 insertions(+), 48 deletions(-)

diff --git a/test/cpp/prim/test_eager_prim.cc b/test/cpp/prim/test_eager_prim.cc
index 4a13bf78025f3..c7b590379cb61 100644
--- a/test/cpp/prim/test_eager_prim.cc
+++ b/test/cpp/prim/test_eager_prim.cc
@@ -17,16 +17,15 @@
 #include "glog/logging.h"
 #include "gtest/gtest.h"
 #include "paddle/common/flags.h"
-// #include
-// "paddle/fluid/eager/api/generated/eager_generated/forwards/dygraph_functions.h"
-// #include "paddle/fluid/eager/api/utils/hook_utils.h"
+#include "paddle/fluid/eager/api/generated/eager_generated/forwards/dygraph_functions.h"
+#include "paddle/fluid/eager/api/utils/hook_utils.h"
 // #include "paddle/fluid/eager/backward.h"
 #include "paddle/fluid/prim/utils/utils.h"
 // #include "paddle/phi/core/dense_tensor.h"
 // #include "paddle/phi/core/kernel_registry.h"
 // #include "paddle/phi/core/tensor_meta.h"
-// #include "test/cpp/eager/test_utils.h"
-// #include "test/cpp/prim/init_env_utils.h"
+#include "test/cpp/eager/test_utils.h"
+#include "test/cpp/prim/init_env_utils.h"
 
 COMMON_DECLARE_string(tensor_operants_mode);
 
@@ -128,49 +127,49 @@ namespace prim {
 //   EXPECT_EQ(out0.data<int>()[0], out1.data<int>()[0]);
 // }
 
-// TEST(EagerPrim, CompareOperantsTest) {
-//   // 1. Initialized
-//   eager_test::InitEnv(paddle::platform::CPUPlace());
-//   FLAGS_tensor_operants_mode = "eager";
-//   paddle::prim::InitTensorOperants();
-//   // 2. pre
-//   paddle::framework::DDim ddim = common::make_ddim({4, 16, 16, 32});
-//   paddle::Tensor tensor0 =
-//       eager_test::CreateTensorWithValue(ddim,
-//                                         paddle::platform::CPUPlace(),
-//                                         phi::DataType::INT32,
-//                                         phi::DataLayout::NCHW,
-//                                         1 /*value*/,
-//                                         true /*is_leaf*/);
-//   ::egr::egr_utils_api::RetainGradForTensor(tensor0);
-//   paddle::Tensor tensor1 =
-//       eager_test::CreateTensorWithValue(ddim,
-//                                         paddle::platform::CPUPlace(),
-//                                         phi::DataType::INT32,
-//                                         phi::DataLayout::NCHW,
-//                                         0 /*value*/,
-//                                         true /*is_leaf*/);
-//   ::egr::egr_utils_api::RetainGradForTensor(tensor1);
-//   // 3. Run Forward once
-//   paddle::Tensor out0 = (tensor0 < tensor1);
-//   paddle::Tensor out1 = less_than_ad_func(tensor0, tensor1);
-//   EXPECT_EQ(out0.data<bool>()[0], out1.data<bool>()[0]);
-//   out0 = (tensor0 <= tensor1);
-//   out1 = less_equal_ad_func(tensor0, tensor1);
-//   EXPECT_EQ(out0.data<bool>()[0], out1.data<bool>()[0]);
-//   out0 = (tensor0 == tensor1);
-//   out1 = equal_ad_func(tensor0, tensor1);
-//   EXPECT_EQ(out0.data<bool>()[0], out1.data<bool>()[0]);
-//   out0 = (tensor0 != tensor1);
-//   out1 = not_equal_ad_func(tensor0, tensor1);
-//   EXPECT_EQ(out0.data<bool>()[0], out1.data<bool>()[0]);
-//   out0 = (tensor0 > tensor1);
-//   out1 = greater_than_ad_func(tensor0, tensor1);
-//   EXPECT_EQ(out0.data<bool>()[0], out1.data<bool>()[0]);
-//   out0 = (tensor0 >= tensor1);
-//   out1 = greater_equal_ad_func(tensor0, tensor1);
-//   EXPECT_EQ(out0.data<bool>()[0], out1.data<bool>()[0]);
-// }
+TEST(EagerPrim, CompareOperantsTest) {
+  // 1. Initialized
+  eager_test::InitEnv(paddle::platform::CPUPlace());
+  FLAGS_tensor_operants_mode = "eager";
+  paddle::prim::InitTensorOperants();
+  // 2. pre
+  paddle::framework::DDim ddim = common::make_ddim({4, 16, 16, 32});
+  paddle::Tensor tensor0 =
+      eager_test::CreateTensorWithValue(ddim,
+                                        paddle::platform::CPUPlace(),
+                                        phi::DataType::INT32,
+                                        phi::DataLayout::NCHW,
+                                        1 /*value*/,
+                                        true /*is_leaf*/);
+  ::egr::egr_utils_api::RetainGradForTensor(tensor0);
+  paddle::Tensor tensor1 =
+      eager_test::CreateTensorWithValue(ddim,
+                                        paddle::platform::CPUPlace(),
+                                        phi::DataType::INT32,
+                                        phi::DataLayout::NCHW,
+                                        0 /*value*/,
+                                        true /*is_leaf*/);
+  ::egr::egr_utils_api::RetainGradForTensor(tensor1);
+  // 3. Run Forward once
+  paddle::Tensor out0 = (tensor0 < tensor1);
+  paddle::Tensor out1 = less_than_ad_func(tensor0, tensor1);
+  EXPECT_EQ(out0.data<bool>()[0], out1.data<bool>()[0]);
+  out0 = (tensor0 <= tensor1);
+  out1 = less_equal_ad_func(tensor0, tensor1);
+  EXPECT_EQ(out0.data<bool>()[0], out1.data<bool>()[0]);
+  out0 = (tensor0 == tensor1);
+  out1 = equal_ad_func(tensor0, tensor1);
+  EXPECT_EQ(out0.data<bool>()[0], out1.data<bool>()[0]);
+  out0 = (tensor0 != tensor1);
+  out1 = not_equal_ad_func(tensor0, tensor1);
+  EXPECT_EQ(out0.data<bool>()[0], out1.data<bool>()[0]);
+  out0 = (tensor0 > tensor1);
+  out1 = greater_than_ad_func(tensor0, tensor1);
+  EXPECT_EQ(out0.data<bool>()[0], out1.data<bool>()[0]);
+  out0 = (tensor0 >= tensor1);
+  out1 = greater_equal_ad_func(tensor0, tensor1);
+  EXPECT_EQ(out0.data<bool>()[0], out1.data<bool>()[0]);
+}
 
 TEST(EagerPrim, TestFlags) {
   PrimCommonUtils::SetBwdPrimEnabled(true);

From a9946ee67609b6232ac8b68805f972a47b12e629 Mon Sep 17 00:00:00 2001
From: gouzi <530971494@qq.com>
Date: Thu, 25 Apr 2024 23:06:17 +0800
Subject: [PATCH 17/21] [ci][test] ignore hook_utils

---
 test/cpp/prim/test_eager_prim.cc | 116 +++++++++++++++----------------
 1 file changed, 58 insertions(+), 58 deletions(-)

diff --git a/test/cpp/prim/test_eager_prim.cc b/test/cpp/prim/test_eager_prim.cc
index c7b590379cb61..0b81c69e14417 100644
--- a/test/cpp/prim/test_eager_prim.cc
+++ b/test/cpp/prim/test_eager_prim.cc
@@ -18,7 +18,7 @@
 #include "gtest/gtest.h"
 #include "paddle/common/flags.h"
 #include "paddle/fluid/eager/api/generated/eager_generated/forwards/dygraph_functions.h"
-#include "paddle/fluid/eager/api/utils/hook_utils.h"
+// #include "paddle/fluid/eager/api/utils/hook_utils.h"
 // #include "paddle/fluid/eager/backward.h"
 #include "paddle/fluid/prim/utils/utils.h"
 // #include "paddle/phi/core/dense_tensor.h"
@@ -89,7 +89,45 @@ namespace prim {
 //           ->data<float>()[0]);
 // }
 
-// TEST(EagerPrim, LogicalOperantsTest) {
+TEST(EagerPrim, LogicalOperantsTest) {
+  // 1. Initialized
+  eager_test::InitEnv(paddle::platform::CPUPlace());
+  FLAGS_tensor_operants_mode = "eager";
+  paddle::prim::InitTensorOperants();
+  // 2. pre
+  paddle::framework::DDim ddim = common::make_ddim({4, 16, 16, 32});
+  paddle::Tensor tensor0 =
+      eager_test::CreateTensorWithValue(ddim,
+                                        paddle::platform::CPUPlace(),
+                                        phi::DataType::INT32,
+                                        phi::DataLayout::NCHW,
+                                        1 /*value*/,
+                                        true /*is_leaf*/);
+  // ::egr::egr_utils_api::RetainGradForTensor(tensor0);
+  paddle::Tensor tensor1 =
+      eager_test::CreateTensorWithValue(ddim,
+                                        paddle::platform::CPUPlace(),
+                                        phi::DataType::INT32,
+                                        phi::DataLayout::NCHW,
+                                        0 /*value*/,
+                                        true /*is_leaf*/);
+  // ::egr::egr_utils_api::RetainGradForTensor(tensor1);
+  // 3. Run Forward once
+  paddle::Tensor out0 = tensor0 & tensor1;
+  paddle::Tensor out1 = bitwise_and_ad_func(tensor0, tensor1);
+  EXPECT_EQ(out0.data<int>()[0], out1.data<int>()[0]);
+  out0 = tensor0 | tensor1;
+  out1 = bitwise_or_ad_func(tensor0, tensor1);
+  EXPECT_EQ(out0.data<int>()[0], out1.data<int>()[0]);
+  out0 = tensor0 ^ tensor1;
+  out1 = bitwise_xor_ad_func(tensor0, tensor1);
+  EXPECT_EQ(out0.data<int>()[0], out1.data<int>()[0]);
+  out0 = ~tensor0;
+  out1 = bitwise_not_ad_func(tensor0);
+  EXPECT_EQ(out0.data<int>()[0], out1.data<int>()[0]);
+}
+
+// TEST(EagerPrim, CompareOperantsTest) {
 //   // 1. Initialized
 //   eager_test::InitEnv(paddle::platform::CPUPlace());
 //   FLAGS_tensor_operants_mode = "eager";
@@ -113,64 +151,26 @@ namespace prim {
 //                                         true /*is_leaf*/);
 //   ::egr::egr_utils_api::RetainGradForTensor(tensor1);
 //   // 3. Run Forward once
-//   paddle::Tensor out0 = tensor0 & tensor1;
-//   paddle::Tensor out1 = bitwise_and_ad_func(tensor0, tensor1);
-//   EXPECT_EQ(out0.data<int>()[0], out1.data<int>()[0]);
-//   out0 = tensor0 | tensor1;
-//   out1 = bitwise_or_ad_func(tensor0, tensor1);
-//   EXPECT_EQ(out0.data<int>()[0], out1.data<int>()[0]);
-//   out0 = tensor0 ^ tensor1;
-//   out1 = bitwise_xor_ad_func(tensor0, tensor1);
-//   EXPECT_EQ(out0.data<int>()[0], out1.data<int>()[0]);
-//   out0 = ~tensor0;
-//   out1 = bitwise_not_ad_func(tensor0);
-//   EXPECT_EQ(out0.data<int>()[0], out1.data<int>()[0]);
+//   paddle::Tensor out0 = (tensor0 < tensor1);
+//   paddle::Tensor out1 = less_than_ad_func(tensor0, tensor1);
+//   EXPECT_EQ(out0.data<bool>()[0], out1.data<bool>()[0]);
+//   out0 = (tensor0 <= tensor1);
+//   out1 = less_equal_ad_func(tensor0, tensor1);
+//   EXPECT_EQ(out0.data<bool>()[0], out1.data<bool>()[0]);
+//   out0 = (tensor0 == tensor1);
+//   out1 = equal_ad_func(tensor0, tensor1);
+//   EXPECT_EQ(out0.data<bool>()[0], out1.data<bool>()[0]);
+//   out0 = (tensor0 != tensor1);
+//   out1 = not_equal_ad_func(tensor0, tensor1);
+//   EXPECT_EQ(out0.data<bool>()[0], out1.data<bool>()[0]);
+//   out0 = (tensor0 > tensor1);
+//   out1 = greater_than_ad_func(tensor0, tensor1);
+//   EXPECT_EQ(out0.data<bool>()[0], out1.data<bool>()[0]);
+//   out0 = (tensor0 >= tensor1);
+//   out1 = greater_equal_ad_func(tensor0, tensor1);
+//   EXPECT_EQ(out0.data<bool>()[0], out1.data<bool>()[0]);
 // }
 
-TEST(EagerPrim, CompareOperantsTest) {
-  // 1. Initialized
-  eager_test::InitEnv(paddle::platform::CPUPlace());
-  FLAGS_tensor_operants_mode = "eager";
-  paddle::prim::InitTensorOperants();
-  // 2. pre
-  paddle::framework::DDim ddim = common::make_ddim({4, 16, 16, 32});
-  paddle::Tensor tensor0 =
-      eager_test::CreateTensorWithValue(ddim,
-                                        paddle::platform::CPUPlace(),
-                                        phi::DataType::INT32,
-                                        phi::DataLayout::NCHW,
-                                        1 /*value*/,
-                                        true /*is_leaf*/);
-  ::egr::egr_utils_api::RetainGradForTensor(tensor0);
-  paddle::Tensor tensor1 =
-      eager_test::CreateTensorWithValue(ddim,
-                                        paddle::platform::CPUPlace(),
-                                        phi::DataType::INT32,
-                                        phi::DataLayout::NCHW,
-                                        0 /*value*/,
-                                        true /*is_leaf*/);
-  ::egr::egr_utils_api::RetainGradForTensor(tensor1);
-  // 3. Run Forward once
-  paddle::Tensor out0 = (tensor0 < tensor1);
-  paddle::Tensor out1 = less_than_ad_func(tensor0, tensor1);
-  EXPECT_EQ(out0.data<bool>()[0], out1.data<bool>()[0]);
-  out0 = (tensor0 <= tensor1);
-  out1 = less_equal_ad_func(tensor0, tensor1);
-  EXPECT_EQ(out0.data<bool>()[0], out1.data<bool>()[0]);
-  out0 = (tensor0 == tensor1);
-  out1 = equal_ad_func(tensor0, tensor1);
-  EXPECT_EQ(out0.data<bool>()[0], out1.data<bool>()[0]);
-  out0 = (tensor0 != tensor1);
-  out1 = not_equal_ad_func(tensor0, tensor1);
-  EXPECT_EQ(out0.data<bool>()[0], out1.data<bool>()[0]);
-  out0 = (tensor0 > tensor1);
-  out1 = greater_than_ad_func(tensor0, tensor1);
-  EXPECT_EQ(out0.data<bool>()[0], out1.data<bool>()[0]);
-  out0 = (tensor0 >= tensor1);
-  out1 = greater_equal_ad_func(tensor0, tensor1);
-  EXPECT_EQ(out0.data<bool>()[0], out1.data<bool>()[0]);
-}
-
 TEST(EagerPrim, TestFlags) {
   PrimCommonUtils::SetBwdPrimEnabled(true);
   ASSERT_TRUE(PrimCommonUtils::IsBwdPrimEnabled());

From d83476b3eb6f4091793116a0cfc44813fa77a093 Mon Sep 17 00:00:00 2001
From: gouzi <530971494@qq.com>
Date: Fri, 26 Apr 2024 00:12:51 +0800
Subject: [PATCH 18/21] [ci][test] ignore `dygraph_functions` and open
 `hook_utils`

---
 test/cpp/prim/test_eager_prim.cc | 31 ++++++++++++++++---------------
 1 file changed, 16 insertions(+), 15 deletions(-)

diff --git a/test/cpp/prim/test_eager_prim.cc b/test/cpp/prim/test_eager_prim.cc
index 0b81c69e14417..4c2fc582aa822 100644
--- a/test/cpp/prim/test_eager_prim.cc
+++ b/test/cpp/prim/test_eager_prim.cc
@@ -17,8 +17,9 @@
 #include "glog/logging.h"
 #include "gtest/gtest.h"
 #include "paddle/common/flags.h"
-#include "paddle/fluid/eager/api/generated/eager_generated/forwards/dygraph_functions.h"
-// #include "paddle/fluid/eager/api/utils/hook_utils.h"
+// #include
+// "paddle/fluid/eager/api/generated/eager_generated/forwards/dygraph_functions.h"
+#include "paddle/fluid/eager/api/utils/hook_utils.h"
 // #include "paddle/fluid/eager/backward.h"
 #include "paddle/fluid/prim/utils/utils.h"
 // #include "paddle/phi/core/dense_tensor.h"
@@ -103,7 +104,7 @@ TEST(EagerPrim, LogicalOperantsTest) {
                                         phi::DataLayout::NCHW,
                                         1 /*value*/,
                                         true /*is_leaf*/);
-  // ::egr::egr_utils_api::RetainGradForTensor(tensor0);
+  ::egr::egr_utils_api::RetainGradForTensor(tensor0);
   paddle::Tensor tensor1 =
       eager_test::CreateTensorWithValue(ddim,
                                         paddle::platform::CPUPlace(),
@@ -111,20 +112,20 @@ TEST(EagerPrim, LogicalOperantsTest) {
                                         phi::DataLayout::NCHW,
                                         0 /*value*/,
                                         true /*is_leaf*/);
-  // ::egr::egr_utils_api::RetainGradForTensor(tensor1);
+  ::egr::egr_utils_api::RetainGradForTensor(tensor1);
   // 3. Run Forward once
   paddle::Tensor out0 = tensor0 & tensor1;
-  paddle::Tensor out1 = bitwise_and_ad_func(tensor0, tensor1);
-  EXPECT_EQ(out0.data<int>()[0], out1.data<int>()[0]);
-  out0 = tensor0 | tensor1;
-  out1 = bitwise_or_ad_func(tensor0, tensor1);
-  EXPECT_EQ(out0.data<int>()[0], out1.data<int>()[0]);
-  out0 = tensor0 ^ tensor1;
-  out1 = bitwise_xor_ad_func(tensor0, tensor1);
-  EXPECT_EQ(out0.data<int>()[0], out1.data<int>()[0]);
-  out0 = ~tensor0;
-  out1 = bitwise_not_ad_func(tensor0);
-  EXPECT_EQ(out0.data<int>()[0], out1.data<int>()[0]);
+  // paddle::Tensor out1 = bitwise_and_ad_func(tensor0, tensor1);
+  // EXPECT_EQ(out0.data<int>()[0], out1.data<int>()[0]);
+  // out0 = tensor0 | tensor1;
+  // out1 = bitwise_or_ad_func(tensor0, tensor1);
+  // EXPECT_EQ(out0.data<int>()[0], out1.data<int>()[0]);
+  // out0 = tensor0 ^ tensor1;
+  // out1 = bitwise_xor_ad_func(tensor0, tensor1);
+  // EXPECT_EQ(out0.data<int>()[0], out1.data<int>()[0]);
+  // out0 = ~tensor0;
+  // out1 = bitwise_not_ad_func(tensor0);
+  // EXPECT_EQ(out0.data<int>()[0], out1.data<int>()[0]);
 }
 
 // TEST(EagerPrim, CompareOperantsTest) {

From 462cdf72616f15f3c7c1e7ef17e3a1ae748893c5 Mon Sep 17 00:00:00 2001
From: gouzi <530971494@qq.com>
Date: Wed, 1 May 2024 15:18:19 +0800
Subject: [PATCH 19/21] fix test_comp_eager

---
 test/cpp/prim/CMakeLists.txt | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/test/cpp/prim/CMakeLists.txt b/test/cpp/prim/CMakeLists.txt
index 7f5b3af052588..647587b79194a 100644
--- a/test/cpp/prim/CMakeLists.txt
+++ b/test/cpp/prim/CMakeLists.txt
@@ -23,7 +23,7 @@ if(NOT (NOT WITH_PYTHON AND ON_INFER))
   cc_library(init_env_utils SRCS init_env_utils.cc)
   target_compile_definitions(init_env_utils PUBLIC PADDLE_DLL_EXPORT)
 
-  paddle_test(test_comp_eager SRCS test_eager_prim.cc DEPS init_env_utils)
+  paddle_test(test_comp_eager SRCS test_eager_prim.cc DEPS init_env_utils.cc)
 endif()
 
 # skip win32 since wget is not installed by default on windows machine.

From c7e9825618b67747cac48734be8d8107c41acba9 Mon Sep 17 00:00:00 2001
From: gouzi <530971494@qq.com>
Date: Wed, 1 May 2024 15:21:35 +0800
Subject: [PATCH 20/21] rollback `test_eager_prim.cc`

---
 test/cpp/prim/test_eager_prim.cc | 229 +++++++++++++++----------------
 1 file changed, 114 insertions(+), 115 deletions(-)

diff --git a/test/cpp/prim/test_eager_prim.cc b/test/cpp/prim/test_eager_prim.cc
index 4c2fc582aa822..ac82ecac53081 100644
--- a/test/cpp/prim/test_eager_prim.cc
+++ b/test/cpp/prim/test_eager_prim.cc
@@ -17,14 +17,13 @@
 #include "glog/logging.h"
 #include "gtest/gtest.h"
 #include "paddle/common/flags.h"
-// #include
-// "paddle/fluid/eager/api/generated/eager_generated/forwards/dygraph_functions.h"
+#include "paddle/fluid/eager/api/generated/eager_generated/forwards/dygraph_functions.h"
 #include "paddle/fluid/eager/api/utils/hook_utils.h"
-// #include "paddle/fluid/eager/backward.h"
+#include "paddle/fluid/eager/backward.h"
 #include "paddle/fluid/prim/utils/utils.h"
-// #include "paddle/phi/core/dense_tensor.h"
-// #include "paddle/phi/core/kernel_registry.h"
-// #include "paddle/phi/core/tensor_meta.h"
+#include "paddle/phi/core/dense_tensor.h"
+#include "paddle/phi/core/kernel_registry.h"
+#include "paddle/phi/core/tensor_meta.h"
 #include "test/cpp/eager/test_utils.h"
 #include "test/cpp/prim/init_env_utils.h"
 
@@ -33,62 +32,62 @@ COMMON_DECLARE_string(tensor_operants_mode);
 namespace paddle {
 namespace prim {
 
-// TEST(EagerPrim, TanhBackwardTest) {
-//   // 1. Initialized
-//   eager_test::InitEnv(paddle::platform::CPUPlace());
-//   FLAGS_tensor_operants_mode = "eager";
-//   paddle::prim::InitTensorOperants();
-//   // 2. pre
-//   paddle::framework::DDim ddim = common::make_ddim({4, 16, 16, 32});
-//   paddle::Tensor tensor0 =
-//       eager_test::CreateTensorWithValue(ddim,
-//                                         paddle::platform::CPUPlace(),
-//                                         phi::DataType::FLOAT32,
-//                                         phi::DataLayout::NCHW,
-//                                         5.0 /*value*/,
-//                                         true /*is_leaf*/);
-//   ::egr::egr_utils_api::RetainGradForTensor(tensor0);
-//   paddle::Tensor tensor1 =
-//       eager_test::CreateTensorWithValue(ddim,
-//                                         paddle::platform::CPUPlace(),
-//                                         phi::DataType::FLOAT32,
-//                                         phi::DataLayout::NCHW,
-//                                         5.0 /*value*/,
-//                                         true /*is_leaf*/);
-//   ::egr::egr_utils_api::RetainGradForTensor(tensor1);
-//   // 3. Run Forward once
-//   paddle::Tensor out0 = tanh_ad_func(tensor0);
-//   std::vector<paddle::Tensor> outs0 = {out0};
-//   // Disable prim
-//   PrimCommonUtils::SetBwdPrimEnabled(false);
-//   ASSERT_FALSE(PrimCommonUtils::IsBwdPrimEnabled());
-//   // 4. Run Backward
-//   egr::Backward(outs0, {}, false);
+TEST(EagerPrim, TanhBackwardTest) {
+  // 1. Initialized
+  eager_test::InitEnv(paddle::platform::CPUPlace());
+  FLAGS_tensor_operants_mode = "eager";
+  paddle::prim::InitTensorOperants();
+  // 2. pre
+  paddle::framework::DDim ddim = common::make_ddim({4, 16, 16, 32});
+  paddle::Tensor tensor0 =
+      eager_test::CreateTensorWithValue(ddim,
+                                        paddle::platform::CPUPlace(),
+                                        phi::DataType::FLOAT32,
+                                        phi::DataLayout::NCHW,
+                                        5.0 /*value*/,
+                                        true /*is_leaf*/);
+  ::egr::egr_utils_api::RetainGradForTensor(tensor0);
+  paddle::Tensor tensor1 =
+      eager_test::CreateTensorWithValue(ddim,
+                                        paddle::platform::CPUPlace(),
+                                        phi::DataType::FLOAT32,
+                                        phi::DataLayout::NCHW,
+                                        5.0 /*value*/,
+                                        true /*is_leaf*/);
+  ::egr::egr_utils_api::RetainGradForTensor(tensor1);
+  // 3. Run Forward once
+  paddle::Tensor out0 = tanh_ad_func(tensor0);
+  std::vector<paddle::Tensor> outs0 = {out0};
+  // Disable prim
+  PrimCommonUtils::SetBwdPrimEnabled(false);
+  ASSERT_FALSE(PrimCommonUtils::IsBwdPrimEnabled());
+  // 4. Run Backward
+  egr::Backward(outs0, {}, false);
 
-//   paddle::Tensor out1 = tanh_ad_func(tensor1);
-//   std::vector<paddle::Tensor> outs1 = {out1};
-//   // Enable prim
-//   PrimCommonUtils::SetBwdPrimEnabled(true);
-//   ASSERT_TRUE(PrimCommonUtils::IsBwdPrimEnabled());
-//   // 4. Run Backward
-//   ::egr::Backward(outs1, {}, false);
-//   VLOG(7)
-//       << "Target Grad is: "
-//       << std::static_pointer_cast<phi::DenseTensor>(
-//              ::egr::EagerUtils::unsafe_autograd_meta(tensor0)->Grad().impl())
-//              ->data<float>()[0];
-//   VLOG(7)
-//       << "Result Grad is: "
-//       << std::static_pointer_cast<phi::DenseTensor>(
-//              ::egr::EagerUtils::unsafe_autograd_meta(tensor1)->Grad().impl())
-//              ->data<float>()[0];
-//   // Examine Backward Grad
-//   eager_test::CompareGradTensorWithValue<float>(
-//       tensor1,
-//       std::static_pointer_cast<phi::DenseTensor>(
-//           ::egr::EagerUtils::unsafe_autograd_meta(tensor0)->Grad().impl())
-//           ->data<float>()[0]);
-// }
+  paddle::Tensor out1 = tanh_ad_func(tensor1);
+  std::vector<paddle::Tensor> outs1 = {out1};
+  // Enable prim
+  PrimCommonUtils::SetBwdPrimEnabled(true);
+  ASSERT_TRUE(PrimCommonUtils::IsBwdPrimEnabled());
+  // 4. Run Backward
+  ::egr::Backward(outs1, {}, false);
+  VLOG(7)
+      << "Target Grad is: "
+      << std::static_pointer_cast<phi::DenseTensor>(
+             ::egr::EagerUtils::unsafe_autograd_meta(tensor0)->Grad().impl())
+             ->data<float>()[0];
+  VLOG(7)
+      << "Result Grad is: "
+      << std::static_pointer_cast<phi::DenseTensor>(
+             ::egr::EagerUtils::unsafe_autograd_meta(tensor1)->Grad().impl())
+             ->data<float>()[0];
+  // Examine Backward Grad
+  eager_test::CompareGradTensorWithValue<float>(
+      tensor1,
+      std::static_pointer_cast<phi::DenseTensor>(
+          ::egr::EagerUtils::unsafe_autograd_meta(tensor0)->Grad().impl())
+          ->data<float>()[0]);
+}
 
 TEST(EagerPrim, LogicalOperantsTest) {
   // 1. Initialized
@@ -115,62 +114,62 @@ TEST(EagerPrim, LogicalOperantsTest) {
   ::egr::egr_utils_api::RetainGradForTensor(tensor1);
   // 3. Run Forward once
   paddle::Tensor out0 = tensor0 & tensor1;
-  // paddle::Tensor out1 = bitwise_and_ad_func(tensor0, tensor1);
-  // EXPECT_EQ(out0.data<int>()[0], out1.data<int>()[0]);
-  // out0 = tensor0 | tensor1;
-  // out1 = bitwise_or_ad_func(tensor0, tensor1);
-  // EXPECT_EQ(out0.data<int>()[0], out1.data<int>()[0]);
-  // out0 = tensor0 ^ tensor1;
-  // out1 = bitwise_xor_ad_func(tensor0, tensor1);
-  // EXPECT_EQ(out0.data<int>()[0], out1.data<int>()[0]);
-  // out0 = ~tensor0;
-  // out1 = bitwise_not_ad_func(tensor0);
-  // EXPECT_EQ(out0.data<int>()[0], out1.data<int>()[0]);
+  paddle::Tensor out1 = bitwise_and_ad_func(tensor0, tensor1);
+  EXPECT_EQ(out0.data<int>()[0], out1.data<int>()[0]);
+  out0 = tensor0 | tensor1;
+  out1 = bitwise_or_ad_func(tensor0, tensor1);
+  EXPECT_EQ(out0.data<int>()[0], out1.data<int>()[0]);
+  out0 = tensor0 ^ tensor1;
+  out1 = bitwise_xor_ad_func(tensor0, tensor1);
+  EXPECT_EQ(out0.data<int>()[0], out1.data<int>()[0]);
+  out0 = ~tensor0;
+  out1 = bitwise_not_ad_func(tensor0);
+  EXPECT_EQ(out0.data<int>()[0], out1.data<int>()[0]);
 }
 
-// TEST(EagerPrim, CompareOperantsTest) {
-//   // 1. Initialized
-//   eager_test::InitEnv(paddle::platform::CPUPlace());
-//   FLAGS_tensor_operants_mode = "eager";
-//   paddle::prim::InitTensorOperants();
-//   // 2. pre
-//   paddle::framework::DDim ddim = common::make_ddim({4, 16, 16, 32});
-//   paddle::Tensor tensor0 =
-//       eager_test::CreateTensorWithValue(ddim,
-//                                         paddle::platform::CPUPlace(),
-//                                         phi::DataType::INT32,
-//                                         phi::DataLayout::NCHW,
-//                                         1 /*value*/,
-//                                         true /*is_leaf*/);
-//   ::egr::egr_utils_api::RetainGradForTensor(tensor0);
-//   paddle::Tensor tensor1 =
-//       eager_test::CreateTensorWithValue(ddim,
-//                                         paddle::platform::CPUPlace(),
-//                                         phi::DataType::INT32,
-//                                         phi::DataLayout::NCHW,
-//                                         0 /*value*/,
-//                                         true /*is_leaf*/);
-//   ::egr::egr_utils_api::RetainGradForTensor(tensor1);
-//   // 3. Run Forward once
-//   paddle::Tensor out0 = (tensor0 < tensor1);
-//   paddle::Tensor out1 = less_than_ad_func(tensor0, tensor1);
-//   EXPECT_EQ(out0.data<bool>()[0], out1.data<bool>()[0]);
-//   out0 = (tensor0 <= tensor1);
-//   out1 = less_equal_ad_func(tensor0, tensor1);
-//   EXPECT_EQ(out0.data<bool>()[0], out1.data<bool>()[0]);
-//   out0 = (tensor0 == tensor1);
-//   out1 = equal_ad_func(tensor0, tensor1);
-//   EXPECT_EQ(out0.data<bool>()[0], out1.data<bool>()[0]);
-//   out0 = (tensor0 != tensor1);
-//   out1 = not_equal_ad_func(tensor0, tensor1);
-//   EXPECT_EQ(out0.data<bool>()[0], out1.data<bool>()[0]);
-//   out0 = (tensor0 > tensor1);
-//   out1 = greater_than_ad_func(tensor0, tensor1);
-//   EXPECT_EQ(out0.data<bool>()[0], out1.data<bool>()[0]);
-//   out0 = (tensor0 >= tensor1);
-//   out1 = greater_equal_ad_func(tensor0, tensor1);
-//   EXPECT_EQ(out0.data<bool>()[0], out1.data<bool>()[0]);
-// }
+TEST(EagerPrim, CompareOperantsTest) {
+  // 1. Initialized
+  eager_test::InitEnv(paddle::platform::CPUPlace());
+  FLAGS_tensor_operants_mode = "eager";
+  paddle::prim::InitTensorOperants();
+  // 2. pre
+  paddle::framework::DDim ddim = common::make_ddim({4, 16, 16, 32});
+  paddle::Tensor tensor0 =
+      eager_test::CreateTensorWithValue(ddim,
+                                        paddle::platform::CPUPlace(),
+                                        phi::DataType::INT32,
+                                        phi::DataLayout::NCHW,
+                                        1 /*value*/,
+                                        true /*is_leaf*/);
+  ::egr::egr_utils_api::RetainGradForTensor(tensor0);
+  paddle::Tensor tensor1 =
+      eager_test::CreateTensorWithValue(ddim,
+                                        paddle::platform::CPUPlace(),
+                                        phi::DataType::INT32,
+                                        phi::DataLayout::NCHW,
+                                        0 /*value*/,
+                                        true /*is_leaf*/);
+  ::egr::egr_utils_api::RetainGradForTensor(tensor1);
+  // 3. Run Forward once
+  paddle::Tensor out0 = (tensor0 < tensor1);
+  paddle::Tensor out1 = less_than_ad_func(tensor0, tensor1);
+  EXPECT_EQ(out0.data<bool>()[0], out1.data<bool>()[0]);
+  out0 = (tensor0 <= tensor1);
+  out1 = less_equal_ad_func(tensor0, tensor1);
+  EXPECT_EQ(out0.data<bool>()[0], out1.data<bool>()[0]);
+  out0 = (tensor0 == tensor1);
+  out1 = equal_ad_func(tensor0, tensor1);
+  EXPECT_EQ(out0.data<bool>()[0], out1.data<bool>()[0]);
+  out0 = (tensor0 != tensor1);
+  out1 = not_equal_ad_func(tensor0, tensor1);
+  EXPECT_EQ(out0.data<bool>()[0], out1.data<bool>()[0]);
+  out0 = (tensor0 > tensor1);
+  out1 = greater_than_ad_func(tensor0, tensor1);
+  EXPECT_EQ(out0.data<bool>()[0], out1.data<bool>()[0]);
+  out0 = (tensor0 >= tensor1);
+  out1 = greater_equal_ad_func(tensor0, tensor1);
+  EXPECT_EQ(out0.data<bool>()[0], out1.data<bool>()[0]);
+}
 
 TEST(EagerPrim, TestFlags) {
   PrimCommonUtils::SetBwdPrimEnabled(true);

From de20f12ab03a685985e67c0d68f838fa2e636b6b Mon Sep 17 00:00:00 2001
From: gouzi <530971494@qq.com>
Date: Wed, 1 May 2024 18:44:28 +0800
Subject: [PATCH 21/21] fix

---
 test/cpp/prim/CMakeLists.txt | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/test/cpp/prim/CMakeLists.txt b/test/cpp/prim/CMakeLists.txt
index 647587b79194a..51b5bb70a6e22 100644
--- a/test/cpp/prim/CMakeLists.txt
+++ b/test/cpp/prim/CMakeLists.txt
@@ -23,7 +23,7 @@ if(NOT (NOT WITH_PYTHON AND ON_INFER))
   cc_library(init_env_utils SRCS init_env_utils.cc)
   target_compile_definitions(init_env_utils PUBLIC PADDLE_DLL_EXPORT)
 
-  paddle_test(test_comp_eager SRCS test_eager_prim.cc DEPS init_env_utils.cc)
+  paddle_test(test_comp_eager SRCS test_eager_prim.cc init_env_utils.cc)
 endif()
 
 # skip win32 since wget is not installed by default on windows machine.