MCPcopy Create free account

hub / github.com/LMCache/LMCache / functions

Functions13,371 in github.com/LMCache/LMCache

↓ 1 callersMethodinitialize_allocator
( self, config: LMCacheEngineConfig, metadata: LMCacheMetadata )
lmcache/v1/storage_backend/pd_backend_async.py:439
↓ 1 callersFunctioninitialize_kv_cache
Initialize KV cache tensors on GPU for testing.
tests/v1/multiprocess/test_cache_server.py:64
↓ 1 callersFunctioninitialize_protocols
Initialize the protocol system by collecting all protocol definitions and validating them against the RequestType enum. This function:
lmcache/v1/multiprocess/protocols/__init__.py:47
↓ 1 callersFunctioninject_stale_cache_entry_for_test
Test-only hook: pre-seed the registry with a stale entry. Lets unit tests reproduce the CPython id-reuse race -- where a fresh tensor lands o
lmcache/v1/platform/cpu/shm.py:280
↓ 1 callersMethodinsert_key
( self, key: CacheEngineKey, size: int, shape: torch.Size, dtype: torc
lmcache/v1/storage_backend/local_disk_backend.py:275
↓ 1 callersMethodinsert_put_task
(self, key: CacheEngineKey)
lmcache/v1/storage_backend/local_disk_backend.py:83
↓ 1 callersFunctioninstall_native_storage_ops_fallback
Install a small native_storage_ops fallback for test environments. Args: None. Returns: None.
tests/v1/storage_backend/raw_block_test_utils.py:182
↓ 1 callersFunctioninvalid_conversations
(conversations)
benchmarks/multi_round_qa/data_preprocessing.py:35
↓ 1 callersMethodinvalidate
(self, key: str)
lmcache/v1/storage_backend/connector/bigtable_connector.py:51
↓ 1 callersMethodipc_handle
Return an IPC handle for cross-process sharing. Returns: A bytes object representing the IPC handle.
lmcache/v1/platform/cpu/stub_cpu_device.py:89
↓ 1 callersFunctionisChineseDocsPage
* Return true when the current page is rendered under the Chinese docs prefix. * * @returns {boolean} Whether the current path is a Chinese document
docs/source/_static/custom.js:33
↓ 1 callersMethodis_frozen
Get freeze mode status. Returns: bool: True if freeze mode is enabled, False otherwise
lmcache/v1/storage_backend/storage_manager.py:870
↓ 1 callersMethodis_healthy
(self)
lmcache/integration/sglang/multi_process_adapter.py:182
↓ 1 callersMethodis_hot_cache_enabled
Get the current hot cache status of LocalCPUBackend. Returns: bool: True if hot cache is enabled, False otherwise
lmcache/v1/storage_backend/storage_manager.py:848
↓ 1 callersMethodis_mla
Returns whether the model uses MLA.
lmcache/v1/platform/base_cache_context.py:175
↓ 1 callersFunctionis_port_open
(h: str, p: int, t: float = 1.0)
tests/conftest.py:846
↓ 1 callersMethodis_ready_for_retrieving
Check whether the current request is ready for retrieving, will be used in process_loading_requests
lmcache/integration/vllm/lmcache_mp_connector.py:232
↓ 1 callersMethodis_ready_for_retrieving
Check whether the current request is ready for retrieving, will be used in process_loading_requests
lmcache/integration/vllm/lmcache_mp_connector_0201.py:245
↓ 1 callersMethodis_ready_for_retrieving
Check whether the current request is ready for retrieving, will be used in process_loading_requests
lmcache/integration/vllm/lmcache_mp_connector_0180.py:227
↓ 1 callersMethodis_valid
(self)
lmcache/v1/memory_management.py:666
↓ 1 callersFunctionisinstance_mock_connector
Tests MockTestRemoteConnector as isinstance() not identifying the child class
tests/v1/storage_backend/test_remote_storage_plugin.py:142
↓ 1 callersFunctionisinstance_mock_instrumented_connector
Tests InstrumentedRemoteConnector as isinstance() not identifying the child class
tests/v1/storage_backend/test_remote_storage_plugin.py:150
↓ 1 callersMethodkey_exists
(self, key: CacheEngineKey)
lmcache/v1/storage_backend/nixl_storage_backend.py:1614
↓ 1 callersMethodlayerwise_batched_get
Non-blocking function to get the memory objects into the storages in a layerwise manner. Do not store if the same object is b
lmcache/v1/storage_backend/storage_manager.py:514
↓ 1 callersMethodlazy_init_peer_connection
( self, local_id: str, peer_id: str, peer_init_url: str, init_side_msg
lmcache/v1/transfer_channel/nixl_channel.py:121
↓ 1 callersMethodlist
(self)
tests/v1/shm_allocator/shmfile_connector.py:287
↓ 1 callersFunctionlist_to_gpu_tensor
(lis: list[int], device: torch.device)
lmcache/v1/platform/cuda/cache_context.py:53
↓ 1 callersMethodlist_to_histogram
(self, data: List[float], buckets: List[float])
lmcache/usage_context.py:372
↓ 1 callersMethodlist_tree
Return recursive bucket tree entries under ``prefix``.
lmcache/v1/storage_backend/connector/hfbucket_connector.py:145
↓ 1 callersFunctionlmcache_get_config
Return an LMCacheEngineConfig from ``LMCACHE_CONFIG_FILE`` or env. Mirrors the in-process pattern used by the vLLM adapter: prefer a config f
lmcache/integration/tensorrt_llm/utils.py:25
↓ 1 callersFunctionloadEnvironment
()
lmcache/v1/cache_controller/frontend/static/js/controller_app.js:557
↓ 1 callersFunctionloadEnvironment
()
lmcache/lmcache_frontend/static/js/app.js:919
↓ 1 callersFunctionloadInference
()
lmcache/lmcache_frontend/static/js/app.js:902
↓ 1 callersFunctionloadMeta
()
lmcache/lmcache_frontend/static/js/app.js:886
↓ 1 callersFunctionloadMetrics
()
lmcache/v1/cache_controller/frontend/static/js/controller_app.js:534
↓ 1 callersFunctionloadMetrics
()
lmcache/lmcache_frontend/static/js/app.js:713
↓ 1 callersFunctionloadOverview
()
lmcache/lmcache_frontend/static/js/app.js:458
↓ 1 callersFunctionloadThreads
()
lmcache/v1/cache_controller/frontend/static/js/controller_app.js:613
↓ 1 callersFunctionloadThreads
()
lmcache/lmcache_frontend/static/js/app.js:729
↓ 1 callersFunctionload_all_adapters
Force-import all pending adapter modules. Useful for CLI help that needs to list every registered type name.
lmcache/v1/distributed/l2_adapters/factory.py:138
↓ 1 callersFunctionload_and_tokenize_inputs
Load and tokenize inputs from a JSONL file. Returns: Tuple of (tokenized_sequences_list, tokenized_sequences_tensor) - token
examples/agents/prefix_analysis.py:142
↓ 1 callersMethodload_bytes_from_disk
Load bytearray from disk.
lmcache/v1/storage_backend/local_disk_backend.py:597
↓ 1 callersFunctionload_chunk_hashes
Load all chunk hash records from JSONL files. Args: input_dir: Directory containing chunk hash JSONL files Returns: List of
examples/chunk_statistics/analyze_chunk_hashes.py:19
↓ 1 callersFunctionload_config
Load proxy/node list from a JSON config file. Args: config_path: Optional path to the JSON file. When ``None`` the packaged
lmcache/lmcache_frontend/app.py:276
↓ 1 callersFunctionload_engine_config_with_overrides
Load engine configuration with support for file, env vars, and overrides. This function uses the generic load_config_with_overrides utility
lmcache/v1/config.py:1059
↓ 1 callersMethodload_from_lmcache_config
Load config from LMCacheEngineConfig.extra_config.
lmcache/v1/storage_backend/connector/mooncakestore_connector.py:309
↓ 1 callersFunctionload_native_musa_module
Import optional ``musa_aiter``, returning ``None`` when unavailable.
lmcache/v1/platform/musa/native_kv_transfer.py:46
↓ 1 callersFunctionload_nodes_from_supplier
Load node information from node supplier. Args: node_supplier_url: Discovery service URL, e.g. ``/lmcache_infos``. Returns:
lmcache/lmcache_frontend/app.py:722
↓ 1 callersFunctionlocalPortFromMapping
localPortFromMapping extracts the LOCAL port from a kubectl mapping of the form "LOCAL:REMOTE" or just "PORT" (where the local and remote ports are eq
operator/test/utils/portforward.go:148
↓ 1 callersFunctionlocal_cpu_backend
Create local CPU backend with allocator
tests/v1/storage_backend/test_p2p_backend_with_controller.py:285
↓ 1 callersMethodlog_message
(self, format, *args)
tests/cli/commands/bench/test_server_bench.py:259
↓ 1 callersMethodlog_prometheus
(self, stats: LMCacheStats)
lmcache/observability.py:1693
↓ 1 callersMethodlongest_common_substring
For token_tensor[request_id], chunk it and check whether each chunk appears contiguously in any previous request (token_tensor[:reque
examples/agents/prefix_analysis.py:71
↓ 1 callersMethodlongest_prefix_len
Find longest prefix match and update LRU ordering. For request i (1-indexed): y[i] = y[i-1] + (len(tokens[i]) - max_shared_pr
examples/agents/prefix_analysis.py:44
↓ 1 callersFunctionlower
(text)
benchmarks/rag/utils.py:174
↓ 1 callersFunctionmain
Translate up to ``MAX_ENTRIES`` messages across all Chinese ``.po`` files.
tools/translate_docs_zh.py:389
↓ 1 callersFunctionmain
()
tools/check_spdx_header.py:38
↓ 1 callersFunctionmain
()
tests/v1/storage_backend/test_eic.py:124
↓ 1 callersFunctionmain
()
tests/v1/shm_allocator/verify_shmfile_tcp.py:143
↓ 1 callersFunctionmain
()
tests/v1/compute/attention/test_triton_sparse_e2e.py:86
↓ 1 callersFunctionmain
()
examples/mp_runtime_plugins/mp_plugin.py:78
↓ 1 callersFunctionmain
()
examples/kubernetes/health_probe.py:29
↓ 1 callersFunctionmain
()
examples/remote_config_server/config_server.py:138
↓ 1 callersFunctionmain
()
examples/p2p/send_request.py:30
↓ 1 callersFunctionmain
()
examples/online_session/openai_chat_completion_client.py:204
↓ 1 callersFunctionmain
()
examples/kv_cache_reuse/remote_backends/valkey/benchmark_l2.py:268
↓ 1 callersFunctionmain
Main entry point for the Rust backend offload example. Sets up environment variables, builds an LLM with LMCache integration, and runs two re
examples/kv_cache_reuse/local_backends/rust_backend_offload.py:140
↓ 1 callersFunctionmain
()
examples/kv_cache_reuse/local_backends/offload.py:117
↓ 1 callersFunctionmain
()
examples/blend_kv_v1/blend.py:133
↓ 1 callersFunctionmain
()
examples/agents/prefix_analysis.py:436
↓ 1 callersFunctionmain
()
examples/chunk_statistics/analyze_chunk_hashes.py:321
↓ 1 callersFunctionmain
()
examples/kv_cache_calculator/generate_config.py:10
↓ 1 callersFunctionmain
()
lmcache/tools/cache_simulator/plot_hit_rate.py:219
↓ 1 callersFunctionmain
CLI entry point for ``python -m lmcache.tools.cache_simulator.simulator``. Parses command-line arguments and delegates to :func:`run_simulate`.
lmcache/tools/cache_simulator/simulator.py:778
↓ 1 callersFunctionmain
CLI entry point for ``python -m``.
lmcache/tools/cache_simulator/gen_bench_dataset.py:346
↓ 1 callersFunctionmain
()
lmcache/tools/mp_status_viewer/__main__.py:58
↓ 1 callersFunctionmain
Parse arguments and run the transfer channel benchmark. Returns: Process exit code (0 on success, 1 on failure).
lmcache/tools/transfer_channel_benchmark/__main__.py:15
↓ 1 callersFunctionmain
Main function with argument parsing
lmcache/tools/controller_benchmark/__main__.py:163
↓ 1 callersFunctionmain
Main entry point
lmcache/v1/standalone/__main__.py:501
↓ 1 callersFunctionmain
Build the coordinator app from the environment and serve it.
lmcache/v1/mp_coordinator/__main__.py:20
↓ 1 callersFunctionmain
()
lmcache/v1/server/__main__.py:150
↓ 1 callersFunctionmain
()
lmcache/v1/api_server/__main__.py:418
↓ 1 callersFunctionmain
()
lmcache/lmcache_frontend/app.py:930
↓ 1 callersFunctionmain
Entry point for the MP-mode frontend plugin subprocess. Registers SIGTERM/SIGINT handlers for graceful shutdown, parses the aggregated config
lmcache/lmcache_frontend/lmcache_mp_plugin/lmcache_mp_frontend_plugin.py:152
↓ 1 callersFunctionmain
(model_name: str)
benchmarks/long_doc_qa/long_doc_qa_recommender.py:150
↓ 1 callersFunctionmain
(args)
benchmarks/long_doc_qa/long_doc_qa.py:481
↓ 1 callersFunctionmain
Run torch fallback and native opt-in transfer benchmarks.
benchmarks/musa/bench_inprocess_transfer.py:76
↓ 1 callersFunctionmain
()
benchmarks/storage_backend_io/storage_backend_io_benchmark.py:1018
↓ 1 callersFunctionmain
()
benchmarks/multi_doc_qa/shuffle_doc_qa.py:204
↓ 1 callersFunctionmain
(args)
benchmarks/multi_doc_qa/multi_doc_qa.py:249
↓ 1 callersFunctionmain
()
benchmarks/multi_round_qa/multi-round-qa.py:697
↓ 1 callersFunctionmain
Run the benchmark grid.
benchmarks/microbenchmark/bitmap_ops_benchmark.py:74
↓ 1 callersFunctionmain
()
benchmarks/microbenchmark/ttl_lock_benchmark.py:442
↓ 1 callersFunctionmain
()
benchmarks/rag/precompute.py:239
↓ 1 callersFunctionmain
()
benchmarks/rag/rag.py:408
↓ 1 callersFunctionmain
()
benchmarks/ttft-estimator/ttft-estimator.py:63
↓ 1 callersFunctionmake_cb
(i)
tests/v1/storage_backend/test_pd_backend_async.py:265
↓ 1 callersMethodmake_key_bins
(self, config: CacheGenConfig)
lmcache/v1/storage_backend/naive_serde/cachegen_decoder.py:43
↓ 1 callersMethodmake_key_bins
(self, config: CacheGenConfig)
lmcache/v1/storage_backend/naive_serde/cachegen_encoder.py:28
↓ 1 callersMethodmake_key_bins
(self, config: CacheGenConfig)
lmcache/storage_backend/serde/cachegen_decoder.py:131
← previousnext →4,301–4,400 of 13,371, ranked by callers