From c29d8f66d810632b6daf4a7276ac65b425fd0042 Mon Sep 17 00:00:00 2001
From: simson <wudengsong@huawei.com>
Date: Mon, 1 Mar 2021 09:25:02 +0800
Subject: [PATCH] fix precision error after cache modification

---
 mindspore/ccsrc/backend/session/gpu_session.cc        | 4 +++-
 mindspore/ccsrc/pipeline/pynative/pynative_execute.cc | 2 --
 mindspore/ccsrc/utils/utils.h                         | 3 +++
 3 files changed, 6 insertions(+), 3 deletions(-)

diff --git a/mindspore/ccsrc/backend/session/gpu_session.cc b/mindspore/ccsrc/backend/session/gpu_session.cc
index 0b4d00e586..501c45b05c 100644
--- a/mindspore/ccsrc/backend/session/gpu_session.cc
+++ b/mindspore/ccsrc/backend/session/gpu_session.cc
@@ -82,6 +82,7 @@
 #include "utils/ms_utils.h"
 #include "utils/config_manager.h"
 #include "utils/ms_context.h"
+#include "utils/utils.h"
 #if ENABLE_CPU && ENABLE_GPU
 #include "ps/util.h"
 #include "ps/ps_cache/ps_cache_manager.h"
@@ -493,7 +494,8 @@ void GPUSession::BuildOpImpl(const OpRunInfo &op_run_info, const GraphInfo &grap
                              const std::vector<tensor::TensorPtr> &input_tensors,
                              const std::vector<int64_t> &tensors_mask) {
   // Check if the graph cache exists.
-  if (run_op_graphs_.find(graph_info) != run_op_graphs_.end()) {
+  if (run_op_graphs_.find(graph_info) != run_op_graphs_.end() &&
+      kOpCacheAllowList.find(op_run_info.op_name) == kOpCacheAllowList.end()) {
     return;
   }
   // Prepare the graph
diff --git a/mindspore/ccsrc/pipeline/pynative/pynative_execute.cc b/mindspore/ccsrc/pipeline/pynative/pynative_execute.cc
index 50a611a2ba..a4cf097498 100644
--- a/mindspore/ccsrc/pipeline/pynative/pynative_execute.cc
+++ b/mindspore/ccsrc/pipeline/pynative/pynative_execute.cc
@@ -387,10 +387,8 @@ void ConvertPyObjectToTensor(const py::object &input_object, const PrimitivePtr
   } else if (py::isinstance<py::float_>(input_object)) {
     double input_value = py::cast<py::float_>(input_object);
     tensor_ptr = std::make_shared<tensor::Tensor>(input_value, kFloat32);
-    *tensor_mask = kValueNodeTensorMask;
   } else if (py::isinstance<py::int_>(input_object)) {
     tensor_ptr = std::make_shared<tensor::Tensor>(py::cast<int64_t>(input_object), kInt64);
-    *tensor_mask = kValueNodeTensorMask;
   } else if (py::isinstance<py::array>(input_object)) {
     tensor_ptr = TensorPy::MakeTensor(py::cast<py::array>(input_object), nullptr);
   } else if (py::isinstance<py::list>(input_object)) {
diff --git a/mindspore/ccsrc/utils/utils.h b/mindspore/ccsrc/utils/utils.h
index e24655fa07..4033ab2a35 100644
--- a/mindspore/ccsrc/utils/utils.h
+++ b/mindspore/ccsrc/utils/utils.h
@@ -271,6 +271,7 @@ constexpr auto kPadAndShiftOpName = "PadAndShift";
 constexpr auto kSparseSoftmaxCrossEntropyWithLogitsOpName = "SparseSoftmaxCrossEntropyWithLogits";
 constexpr auto kOneHotOpName = "OneHot";
 constexpr auto kSoftmaxCrossEntropyWithLogitsOpName = "SoftmaxCrossEntropyWithLogits";
+constexpr auto kUniformCandidateSamplerOpName = "UniformCandidateSampler";
 
 // Hcom Op Type
 constexpr auto kHcomOpTypeAllReduce = "HcomAllReduce";
@@ -492,6 +493,8 @@ const std::set<std::string> kOptOperatorSet = {kMomentumOpName,
 
 const std::set<std::string> kPosteriorOperatorSet = {kPullOpName};
 
+const std::set<std::string> kOpCacheAllowList = {kUniformCandidateSamplerOpName};
+
 const std::set<std::string> kHWSpecialFormatSet = {
   kOpFormat_FRACTAL_Z_3D, kOpFormat_NC1KHKWHWC0,   kOpFormat_NC1HWC0,         kOpFormat_FRAC_NZ,  kOpFormat_C1HWNCoC0,
   kOpFormat_NC1HWC0_C04,  kOpFormat_FRACTAL_Z_C04, kOpFormat_FRACTAL_ZN_LSTM, kOpFormat_NDC1HWC0, kOpFormat_FRAC_Z};