Yuan Hu

Signed-off-by: Hu, Yuan2 ### Details: - *found it when debuging ticket 88985* - *https://docs.openvino.ai/latest/openvino_inference_engine_samples_benchmark_app_README.html#doxid-openvino-inference-engine-samples-benchmark-app-r-e-a-d-m-e* ### Tickets: - *relate to 88985*

[CPU] Allow external blob to be reallocated

1

### Details: - *porting code from [5902](https://github.com/openvinotoolkit/openvino/pull/5902)* - *Renewed external pointers before infer to take into account user's manipulations with blob* - *Compatibility with preprocessing* ### compare with 5902, there...

[CPU] Fixed creating a descriptor for a dilated convolution forward propagation primitive

1

### Details: - *the code is from [6099](https://github.com/openvinotoolkit/openvino/pull/6099)* - *Fixed getSupportedDescriptors() for deconvolution* ### Tickets: - *52724*

[Good first issue] [CPU PLUGIN] fix testcase issue in function tests on SPR platform

1

## Description: Some test cases failed on SPR platform, need to debug and fix it. ## How to reproduce: 1. comment this code https://github.com/openvinotoolkit/openvino/blob/fa43c7b364b6899a90da2b3a27d77461235a5071/src/plugins/intel_cpu/tests/functional/shared_tests_instances/skip_tests_config.cpp#L374 like this `// retVector.emplace_back(R"(smoke_Reduce.*/ReduceCPULayerTest.*axes=\((0.1|1)\).*Prod_KeepDims.*INFERENCE_PRECISION_HINT=f16.*)");` 2. Build...

# Description - *enable ncx8c->ncx8c support for small input channel in jit sse41convolution* - CVS-ticket https://jira.devtools.intel.com/browse/CVS-125793 - openvino PR https://github.com/openvinotoolkit/openvino/pull/23232

[CPU] Add test case for precision fp16

11

### Details: - *https://github.com/openvinotoolkit/openvino/pull/16500#issuecomment-1602581987* - *add test case for conv dconv fullconnect matmul mvn pad pooling subgraph softmax* ### Tickets: - *CVS-110112*

[core] fix thread safe issue in #19832

2

### Details: - *I lost code `t_stream_count_map[(void*)this] = item.first;` in #19832* - *the thread safe issue happen in below workflow* - create thread A call CustomThreadLocal:local() in thread A ->...

### Details: - Add a check to reject non-supported FakeQuantize from fusing into FC node, so they can run in standalone mode w/o causing exceptions when composing oneDNN postOps. -...