Removed deprecated ov::affinity property (#28247)

### Details:
- Removes `ov::affinity` property (deprecated in OV 2024.0),
`ov::hint::enable_cpu_pinning` should be used instead.
- Updates benchmark_app `-pin` option to be binded with
`ov::hint::enable_cpu_pinning` property.

### Tickets:
 - *[CVS-129093](https://jira.devtools.intel.com/browse/CVS-129093)*

### Dependencies: 

- openvinotoolkit/openvino.genai#1467
- openvinotoolkit/open_model_zoo#3991

Author: Gorokhov Dmitriy

docs/articles_en/learn-openvino/openvino-samples/benchmark-tool.rst

@@ -349,7 +349,7 @@ following usage message:
                                     [-api {sync,async}] [-nireq NUMBER_INFER_REQUESTS] [-nstreams NUMBER_STREAMS] [-inference_only [INFERENCE_ONLY]]
                                     [-infer_precision INFER_PRECISION] [-ip {bool,f16,f32,f64,i8,i16,i32,i64,u8,u16,u32,u64}]
                                     [-op {bool,f16,f32,f64,i8,i16,i32,i64,u8,u16,u32,u64}] [-iop INPUT_OUTPUT_PRECISION] [--mean_values [R,G,B]] [--scale_values [R,G,B]]
-                                    [-nthreads NUMBER_THREADS] [-pin {YES,NO,NUMA,HYBRID_AWARE}] [-latency_percentile LATENCY_PERCENTILE]
+                                    [-nthreads NUMBER_THREADS] [-pin {YES,NO}] [-latency_percentile LATENCY_PERCENTILE]
                                     [-report_type {no_counters,average_counters,detailed_counters}] [-report_folder REPORT_FOLDER] [-pc [PERF_COUNTS]]
                                     [-pcsort {no_sort,sort,simple_sort}] [-pcseq [PCSEQ]] [-exec_graph_path EXEC_GRAPH_PATH] [-dump_config DUMP_CONFIG] [-load_config LOAD_CONFIG]
 
@@ -462,10 +462,8 @@ following usage message:
               -nthreads NUMBER_THREADS, --number_threads NUMBER_THREADS
                                     Number of threads to use for inference on the CPU (including HETERO and MULTI cases).
 
-              -pin {YES,NO,NUMA,HYBRID_AWARE}, --infer_threads_pinning {YES,NO,NUMA,HYBRID_AWARE}
-                                    Optional. Enable threads->cores ('YES' which is OpenVINO runtime's default for conventional CPUs), threads->(NUMA)nodes ('NUMA'),
-                                    threads->appropriate core types ('HYBRID_AWARE', which is OpenVINO runtime's default for Hybrid CPUs) or completely disable ('NO') CPU threads
-                                    pinning for CPU-involved inference.
+              -pin {YES,NO}, --infer_threads_pinning {YES,NO}
+                                    Optional. Enable threads->cores pinning for CPU-involved inference.
 
 
             Statistics dumping options:
@@ -577,11 +575,7 @@ following usage message:
 
             Device-specific performance options:
                 -nthreads  <integer>          Optional. Number of threads to use for inference on the CPU (including HETERO and MULTI cases).
-                -pin  <string>  ("YES"|"CORE") / "HYBRID_AWARE" / ("NO"|"NONE") / "NUMA"  Optional. Explicit inference threads binding options (leave empty to let the OpenVINO make a choice):
-                                            enabling threads->cores pinning("YES", which is already default for any conventional CPU),
-                                            letting the runtime to decide on the threads->different core types("HYBRID_AWARE", which is default on the hybrid CPUs)
-                                            threads->(NUMA)nodes("NUMA") or
-                                            completely disable("NO") CPU inference threads pinning
+                -pin  <string>  "YES" / "NO"  Optional. Explicit threads->cores pinning for CPU inference tasks (leave empty to let the OpenVINO make a choice).
 
             Statistics dumping options:
                 -latency_percentile     Optional. Defines the percentile to be reported in latency metric. The valid range is [1, 100]. The default value is 50 (median).

docs/articles_en/openvino-workflow/running-inference/inference-devices-and-modes/cpu-device.rst

@@ -357,7 +357,6 @@ All parameters must be set before calling ``ov::Core::compile_model()`` in order
 - ``ov::hint::enable_hyper_threading``
 - ``ov::hint::enable_cpu_pinning``
 - ``ov::num_streams``
-- ``ov::affinity``
 - ``ov::inference_num_threads``
 - ``ov::cache_dir``
 - ``ov::intel_cpu::denormals_optimization``
@@ -373,8 +372,6 @@ Read-only properties
 - ``ov::device::full_name``
 - ``ov::device::capabilities``
 
-.. note::
-   ``ov::affinity`` is replaced by ``ov::hint::enable_cpu_pinning``. As such, it is deprecated in the 2024.0 release and will be removed in the 2025 release.
 
 External Dependencies
 ###########################################################

samples/cpp/benchmark_app/benchmark_app.hpp

@@ -179,13 +179,8 @@ static const char infer_num_threads_message[] = "Optional. Number of threads to
                                                 "(including HETERO and MULTI cases).";
 
 // @brief message for CPU threads pinning option
-static const char infer_threads_pinning_message[] =
-    "Optional. Explicit inference threads binding options (leave empty to let the OpenVINO make a choice):\n"
-    "\t\t\t\tenabling threads->cores pinning(\"YES\", which is already default for any conventional CPU), \n"
-    "\t\t\t\tletting the runtime to decide on the threads->different core types(\"HYBRID_AWARE\", which is default on "
-    "the hybrid CPUs) \n"
-    "\t\t\t\tthreads->(NUMA)nodes(\"NUMA\") or \n"
-    "\t\t\t\tcompletely disable(\"NO\") CPU inference threads pinning";
+static const char infer_threads_pinning_message[] = "Optional. Explicit threads->cores pinning for CPU inference tasks "
+                                                    "(leave empty to let the OpenVINO make a choice).";
 
 // @brief message for switching memory allocation type option
 static const char use_device_mem_message[] =
@@ -426,8 +421,7 @@ static void show_usage() {
     std::cout << std::endl;
     std::cout << "Device-specific performance options:" << std::endl;
     std::cout << "    -nthreads  <integer>          " << infer_num_threads_message << std::endl;
-    std::cout << "    -pin  <string>  (\"YES\"|\"CORE\") / \"HYBRID_AWARE\" / (\"NO\"|\"NONE\") / \"NUMA\"  "
-              << infer_threads_pinning_message << std::endl;
+    std::cout << "    -pin  <string>  \"YES\" / \"NO\" " << infer_threads_pinning_message << std::endl;
     std::cout << "    -use_device_mem           " << use_device_mem_message << std::endl;
     std::cout << std::endl;
     std::cout << "Statistics dumping options:" << std::endl;

samples/cpp/benchmark_app/main.cpp

@@ -490,21 +490,11 @@ int main(int argc, char* argv[]) {
                 }
             };
 
-            auto fix_pin_option = [](const std::string& str) -> std::string {
-                if (str == "NO")
-                    return "NONE";
-                else if (str == "YES")
-                    return "CORE";
-                else
-                    return str;
-            };
-
             auto set_nthreads_pin = [&](const std::string& str) {
-                OPENVINO_SUPPRESS_DEPRECATED_START
-                auto property_name = str == "nthreads" ? ov::inference_num_threads.name() : ov::affinity.name();
+                auto property_name =
+                    str == "nthreads" ? ov::inference_num_threads.name() : ov::hint::enable_cpu_pinning.name();
                 auto property = str == "nthreads" ? ov::inference_num_threads(int(FLAGS_nthreads))
-                                                  : ov::affinity(fix_pin_option(FLAGS_pin));
-                OPENVINO_SUPPRESS_DEPRECATED_END
+                                                  : ov::hint::enable_cpu_pinning(FLAGS_pin);
                 if (supported(property_name) || device_name == "AUTO") {
                     // create nthreads/pin primary property for HW device or AUTO if -d is AUTO directly.
                     device_config[property.first] = property.second;

src/bindings/c/include/openvino/c/ov_property.h

@@ -123,13 +123,6 @@ ov_property_key_cache_encryption_callbacks;
 OPENVINO_C_VAR(const char*)
 ov_property_key_num_streams;
 
-/**
- * @brief Read-write property to set/get the name for setting CPU affinity per thread option.
- * @ingroup ov_property_c_api
- */
-OPENVINO_C_VAR(const char*)
-ov_property_key_affinity;
-
 /**
  * @brief Read-write property<int32_t string> to set/get the maximum number of threads that can be used
  * for inference tasks.

src/bindings/c/src/ov_property.cpp

@@ -21,7 +21,6 @@ const char* ov_property_key_max_batch_size = "MAX_BATCH_SIZE";
 const char* ov_property_key_cache_dir = "CACHE_DIR";
 const char* ov_property_key_cache_mode = "CACHE_MODE";
 const char* ov_property_key_num_streams = "NUM_STREAMS";
-const char* ov_property_key_affinity = "AFFINITY";
 const char* ov_property_key_inference_num_threads = "INFERENCE_NUM_THREADS";
 const char* ov_property_key_hint_performance_mode = "PERFORMANCE_HINT";
 const char* ov_property_key_hint_enable_cpu_pinning = "ENABLE_CPU_PINNING";

src/bindings/js/node/src/helper.cpp

@@ -414,10 +414,6 @@ Napi::Value any_to_js(const Napi::CallbackInfo& info, ov::Any value) {
     else if (value.is<int>()) {
         return Napi::Number::New(info.Env(), value.as<int>());
     }
-    // Check for ov::Affinity
-    else if (value.is<ov::Affinity>()) {
-        return Napi::String::New(info.Env(), value.as<std::string>());
-    }
     // Check for ov::element::Type
     else if (value.is<ov::element::Type>()) {
         return Napi::String::New(info.Env(), value.as<std::string>());

src/bindings/python/src/openvino/properties/__init__.py

@@ -3,7 +3,6 @@
 # SPDX-License-Identifier: Apache-2.0
 
 # Enums
-from openvino._pyopenvino.properties import Affinity
 from openvino._pyopenvino.properties import CacheMode
 from openvino._pyopenvino.properties import WorkloadType
 

src/bindings/python/src/openvino/runtime/properties/__init__.py

@@ -3,7 +3,6 @@
 # SPDX-License-Identifier: Apache-2.0
 
 # Enums
-from openvino._pyopenvino.properties import Affinity
 from openvino._pyopenvino.properties import CacheMode
 from openvino._pyopenvino.properties import WorkloadType
 
@@ -15,7 +14,6 @@
 from openvino._pyopenvino.properties import num_streams
 from openvino._pyopenvino.properties import inference_num_threads
 from openvino._pyopenvino.properties import compilation_num_threads
-from openvino._pyopenvino.properties import affinity
 from openvino._pyopenvino.properties import force_tbb_terminate
 from openvino._pyopenvino.properties import enable_mmap
 from openvino._pyopenvino.properties import supported_properties

src/bindings/python/src/pyopenvino/core/properties/properties.cpp

@@ -14,13 +14,6 @@ void regmodule_properties(py::module m) {
     // Top submodule
     py::module m_properties = m.def_submodule("properties", "openvino.properties submodule");
 
-    // Submodule properties - enums
-    py::enum_<ov::Affinity>(m_properties, "Affinity", py::arithmetic())
-        .value("NONE", ov::Affinity::NONE)
-        .value("CORE", ov::Affinity::CORE)
-        .value("NUMA", ov::Affinity::NUMA)
-        .value("HYBRID_AWARE", ov::Affinity::HYBRID_AWARE);
-
     py::enum_<ov::WorkloadType>(m_properties, "WorkloadType", py::arithmetic())
         .value("DEFAULT", ov::WorkloadType::DEFAULT)
         .value("EFFICIENT", ov::WorkloadType::EFFICIENT);
@@ -38,9 +31,6 @@ void regmodule_properties(py::module m) {
     wrap_property_RW(m_properties, ov::num_streams, "num_streams");
     wrap_property_RW(m_properties, ov::inference_num_threads, "inference_num_threads");
     wrap_property_RW(m_properties, ov::compilation_num_threads, "compilation_num_threads");
-    OPENVINO_SUPPRESS_DEPRECATED_START
-    wrap_property_RW(m_properties, ov::affinity, "affinity");
-    OPENVINO_SUPPRESS_DEPRECATED_END
     wrap_property_RW(m_properties, ov::force_tbb_terminate, "force_tbb_terminate");
     wrap_property_RW(m_properties, ov::enable_mmap, "enable_mmap");
     wrap_property_RW(m_properties, ov::weights_path, "weights_path");

src/bindings/python/src/pyopenvino/utils/utils.cpp

@@ -217,8 +217,6 @@ py::object from_ov_any(const ov::Any& any) {
         return py::cast(any.as<ov::device::Type>());
     } else if (any.is<ov::streams::Num>()) {
         return py::cast(any.as<ov::streams::Num>());
-    } else if (any.is<ov::Affinity>()) {
-        return py::cast(any.as<ov::Affinity>());
     } else if (any.is<ov::WorkloadType>()) {
         return py::cast(any.as<ov::WorkloadType>());
     } else if (any.is<ov::CacheMode>()) {
@@ -372,9 +370,7 @@ ov::AnyMap py_object_to_any_map(const py::object& py_obj) {
     for (auto& item : py::cast<py::dict>(py_obj)) {
         std::string key = py::cast<std::string>(item.first);
         py::object value = py::cast<py::object>(item.second);
-        if (py::isinstance<ov::Affinity>(value)) {
-            return_value[key] = py::cast<ov::Affinity>(value);
-        } else if (py_object_is_any_map(value)) {
+        if (py_object_is_any_map(value)) {
             return_value[key] = Common::utils::py_object_to_any_map(value);
         } else {
             return_value[key] = Common::utils::py_object_to_any(value);
@@ -449,8 +445,6 @@ ov::Any py_object_to_any(const py::object& py_obj) {
         return py::cast<ov::device::Type>(py_obj);
     } else if (py::isinstance<ov::streams::Num>(py_obj)) {
         return py::cast<ov::streams::Num>(py_obj);
-    } else if (py::isinstance<ov::Affinity>(py_obj)) {
-        return py::cast<ov::Affinity>(py_obj);
     } else if (py::isinstance<ov::WorkloadType>(py_obj)) {
         return py::cast<ov::WorkloadType>(py_obj);
     } else if (py::isinstance<ov::Tensor>(py_obj)) {

src/bindings/python/tests/test_runtime/test_properties.py

@@ -45,15 +45,6 @@ def test_properties_rw_base():
 @pytest.mark.parametrize(
     ("ov_enum", "expected_values"),
     [
-        (
-            props.Affinity,
-            (
-                (props.Affinity.NONE, "Affinity.NONE", -1),
-                (props.Affinity.CORE, "Affinity.CORE", 0),
-                (props.Affinity.NUMA, "Affinity.NUMA", 1),
-                (props.Affinity.HYBRID_AWARE, "Affinity.HYBRID_AWARE", 2),
-            ),
-        ),
         (
             props.CacheMode,
             (
@@ -259,11 +250,6 @@ def test_properties_ro(ov_property_ro, expected_value):
             "COMPILATION_NUM_THREADS",
             ((44, 44),),
         ),
-        (
-            props.affinity,
-            "AFFINITY",
-            ((props.Affinity.NONE, props.Affinity.NONE),),
-        ),
         (props.force_tbb_terminate, "FORCE_TBB_TERMINATE", ((True, True), (False, False))),
         (props.enable_mmap, "ENABLE_MMAP", ((True, True), (False, False))),
         (
@@ -539,7 +525,6 @@ def test_single_property_setting(device):
                 props.enable_profiling(True),
                 props.cache_dir("./"),
                 props.inference_num_threads(9),
-                props.affinity(props.Affinity.NONE),
                 hints.inference_precision(Type.f32),
                 hints.performance_mode(hints.PerformanceMode.LATENCY),
                 hints.enable_cpu_pinning(True),
@@ -554,7 +539,6 @@ def test_single_property_setting(device):
             props.enable_profiling: True,
             props.cache_dir: "./",
             props.inference_num_threads: 9,
-            props.affinity: props.Affinity.NONE,
             hints.inference_precision: Type.f32,
             hints.performance_mode: hints.PerformanceMode.LATENCY,
             hints.enable_cpu_pinning: True,
@@ -568,7 +552,6 @@ def test_single_property_setting(device):
             props.enable_profiling: True,
             "CACHE_DIR": "./",
             props.inference_num_threads: 9,
-            props.affinity: "NONE",
             "INFERENCE_PRECISION_HINT": Type.f32,
             hints.performance_mode: hints.PerformanceMode.LATENCY,
             hints.scheduling_core_type: hints.SchedulingCoreType.PCORE_ONLY,
@@ -589,7 +572,6 @@ def test_core_cpu_properties(properties_to_set):
     assert core.get_property("CPU", props.enable_profiling) is True
     assert core.get_property("CPU", props.cache_dir) == "./"
     assert core.get_property("CPU", props.inference_num_threads) == 9
-    assert core.get_property("CPU", props.affinity) == props.Affinity.NONE
     assert core.get_property("CPU", streams.num) == 5
 
     # RO properties

src/inference/dev_api/openvino/runtime/threading/istreams_executor.hpp

@@ -51,19 +51,6 @@ class OPENVINO_RUNTIME_API IStreamsExecutor : virtual public ITaskExecutor {
         Task task;
     };
 
-    /**
-     * @brief Defines inference thread binding type
-     */
-    enum ThreadBindingType : std::uint8_t {
-        NONE,   //!< Don't bind the inference threads
-        CORES,  //!< Bind inference threads to the CPU cores (round-robin)
-        // the following modes are implemented only for the TBB code-path:
-        NUMA,  //!< Bind to the NUMA nodes (default mode for the non-hybrid CPUs on the Win/MacOS, where the 'CORES' is
-               //!< not implemeneted)
-        HYBRID_AWARE  //!< Let the runtime bind the inference threads depending on the cores type (default mode for the
-                      //!< hybrid CPUs)
-    };
-
     /**
      * @brief Defines IStreamsExecutor configuration
      */

src/inference/include/openvino/runtime/properties.hpp

@@ -1289,64 +1289,6 @@ static constexpr Property<int32_t, PropertyMutability::RW> inference_num_threads
  */
 static constexpr Property<int32_t, PropertyMutability::RW> compilation_num_threads{"COMPILATION_NUM_THREADS"};
 
-/**
- * @brief Enum to define possible affinity patterns
- * @ingroup ov_runtime_cpp_prop_api
- */
-enum class Affinity {
-    NONE = -1,  //!<  Disable threads affinity pinning
-    CORE = 0,   //!<  Pin threads to cores, best for static benchmarks
-    NUMA = 1,   //!<  Pin threads to NUMA nodes, best for real-life, contented cases. On the Windows and MacOS* this
-                //!<  option behaves as CORE
-    HYBRID_AWARE = 2,  //!< Let the runtime to do pinning to the cores types, e.g. prefer the "big" cores for latency
-                       //!< tasks. On the hybrid CPUs this option is default
-};
-
-/** @cond INTERNAL */
-inline std::ostream& operator<<(std::ostream& os, const Affinity& affinity) {
-    switch (affinity) {
-    case Affinity::NONE:
-        return os << "NONE";
-    case Affinity::CORE:
-        return os << "CORE";
-    case Affinity::NUMA:
-        return os << "NUMA";
-    case Affinity::HYBRID_AWARE:
-        return os << "HYBRID_AWARE";
-    default:
-        OPENVINO_THROW("Unsupported affinity pattern");
-    }
-}
-
-inline std::istream& operator>>(std::istream& is, Affinity& affinity) {
-    std::string str;
-    is >> str;
-    if (str == "NONE") {
-        affinity = Affinity::NONE;
-    } else if (str == "CORE") {
-        affinity = Affinity::CORE;
-    } else if (str == "NUMA") {
-        affinity = Affinity::NUMA;
-    } else if (str == "HYBRID_AWARE") {
-        affinity = Affinity::HYBRID_AWARE;
-    } else {
-        OPENVINO_THROW("Unsupported affinity pattern: ", str);
-    }
-    return is;
-}
-/** @endcond */
-
-/**
- * @deprecated Use ov::hint::enable_cpu_pinning
- * @brief The name for setting CPU affinity per thread option.
- * @ingroup ov_runtime_cpp_prop_api
- * @note The setting is ignored, if the OpenVINO compiled with OpenMP and any affinity-related OpenMP's
- * environment variable is set (as affinity is configured explicitly)
- */
-OPENVINO_DEPRECATED(
-    "This property is deprecated and will be removed soon. Use ov::hint::enable_cpu_pinning instead of it.")
-static constexpr Property<Affinity> affinity{"AFFINITY"};
-
 /**
  * @brief The devices that the inference task been executed.
  * @ingroup ov_runtime_cpp_prop_api

src/plugins/auto/src/cumulative_schedule.cpp

@@ -73,10 +73,7 @@ void CumuSchedule::init() {
                 idx++;
             } else {
                 cpu_device_information = device;
-                OPENVINO_SUPPRESS_DEPRECATED_START
-                cpu_device_information.config.insert(
-                    {ov::affinity.name(), ov::Any(ov::Affinity::CORE).as<std::string>()});
-                OPENVINO_SUPPRESS_DEPRECATED_END
+                cpu_device_information.config.insert({ov::hint::enable_cpu_pinning.name(), "YES"});
             }
         }
         if (!cpu_device_information.device_name.empty())