[JJ-0926] 세션 기반 QuotaServe 및 LFU·FIFO KV 캐시 회수 정책 구현

.buildkite/ci_config.yaml

@@ -8,8 +8,8 @@ run_all_patterns:
   - "CMakeLists.txt"
   - "requirements/common.txt"
   - "requirements/cuda.txt"
-  - "requirements/build.txt"
-  - "requirements/test.txt"
+  - "requirements/build/cuda.txt"
+  - "requirements/test/cuda.txt"
   - "setup.py"
   - "csrc/"
   - "cmake/"

.buildkite/ci_config_intel.yaml

@@ -6,8 +6,8 @@ run_all_patterns:
   - "CMakeLists.txt"
   - "requirements/common.txt"
   - "requirements/xpu.txt"
-  - "requirements/build.txt"
-  - "requirements/test.txt"
+  - "requirements/build/cuda.txt"
+  - "requirements/test/cuda.txt"
   - "setup.py"
   - "csrc/"
   - "cmake/"

.buildkite/hardware_tests/cpu.yaml

@@ -46,7 +46,7 @@ steps:
   - tests/models/language/pooling/
   commands:
     - |
-      bash .buildkite/scripts/hardware_ci/run-cpu-test.sh 30m "
+      bash .buildkite/scripts/hardware_ci/run-cpu-test.sh 40m "
       pytest -x -v -s tests/models/language/generation -m cpu_model
       pytest -x -v -s tests/models/language/pooling -m cpu_model"
 
@@ -99,7 +99,7 @@ steps:
     - |
       bash .buildkite/scripts/hardware_ci/run-cpu-test.sh 45m "
       pytest -x -v -s tests/models/multimodal/generation --ignore=tests/models/multimodal/generation/test_pixtral.py -m cpu_model --num-shards=$$BUILDKITE_PARALLEL_JOB_COUNT --shard-id=$$BUILDKITE_PARALLEL_JOB"
-  parallelism: 2
+  parallelism: 3
 
 - label: "Arm CPU Test"
   depends_on: []

.buildkite/image_build/image_build.sh

@@ -92,8 +92,8 @@ check_and_skip_if_image_exists() {
 }
 
 ecr_login() {
-    aws ecr-public get-login-password --region us-east-1 | docker login --username AWS --password-stdin "$REGISTRY"
-    aws ecr get-login-password --region us-east-1 | docker login --username AWS --password-stdin 936637512419.dkr.ecr.us-east-1.amazonaws.com
+    aws ecr-public get-login-password --region us-east-1 | docker login --username AWS --password-stdin "$REGISTRY" || true
+    aws ecr get-login-password --region us-east-1 | docker login --username AWS --password-stdin 936637512419.dkr.ecr.us-east-1.amazonaws.com || true
 }
 
 prepare_cache_tags() {

.buildkite/image_build/image_build_cpu.sh

@@ -11,7 +11,7 @@ REPO=$2
 BUILDKITE_COMMIT=$3
 
 # authenticate with AWS ECR
-aws ecr-public get-login-password --region us-east-1 | docker login --username AWS --password-stdin "$REGISTRY"
+aws ecr-public get-login-password --region us-east-1 | docker login --username AWS --password-stdin "$REGISTRY" || true
 
 # skip build if image already exists
 if [[ -z $(docker manifest inspect "$REGISTRY"/"$REPO":"$BUILDKITE_COMMIT"-cpu) ]]; then

.buildkite/image_build/image_build_cpu_arm64.sh

@@ -11,7 +11,7 @@ REPO=$2
 BUILDKITE_COMMIT=$3
 
 # authenticate with AWS ECR
-aws ecr-public get-login-password --region us-east-1 | docker login --username AWS --password-stdin "$REGISTRY"
+aws ecr-public get-login-password --region us-east-1 | docker login --username AWS --password-stdin "$REGISTRY" || true
 
 # skip build if image already exists
 if [[ -z $(docker manifest inspect "$REGISTRY"/"$REPO":"$BUILDKITE_COMMIT"-arm64-cpu) ]]; then

.buildkite/image_build/image_build_torch_nightly.sh

@@ -0,0 +1,68 @@
+#!/bin/bash
+set -euo pipefail
+
+# Build a vLLM test image with PyTorch nightly installed.
+# Called by the pipeline generator's "vLLM Against PyTorch Nightly" group.
+
+if [[ $# -lt 5 ]]; then
+  echo "Usage: $0 <registry> <repo> <commit> <branch> <image_tag>"
+  exit 1
+fi
+
+REGISTRY=$1
+REPO=$2
+BUILDKITE_COMMIT=$3
+BRANCH=$4
+IMAGE_TAG=$5
+
+# --- Arguments ---
+echo "--- :mag: Arguments"
+echo "REGISTRY: ${REGISTRY}"
+echo "REPO: ${REPO}"
+echo "BUILDKITE_COMMIT: ${BUILDKITE_COMMIT}"
+echo "BRANCH: ${BRANCH}"
+echo "IMAGE_TAG: ${IMAGE_TAG}"
+
+# --- ECR login ---
+echo "--- :key: ECR login"
+aws ecr-public get-login-password --region us-east-1 \
+  | docker login --username AWS --password-stdin "$REGISTRY"
+aws ecr get-login-password --region us-east-1 \
+  | docker login --username AWS --password-stdin 936637512419.dkr.ecr.us-east-1.amazonaws.com
+
+# --- Set up buildx ---
+echo "--- :docker: Setting up buildx"
+docker buildx create --name vllm-builder --driver docker-container --use || true
+docker buildx inspect --bootstrap
+docker buildx ls
+
+# --- Skip if image already exists ---
+echo "--- :mag: Checking if image already exists"
+if docker manifest inspect "$IMAGE_TAG" >/dev/null 2>&1; then
+  echo "Image found: $IMAGE_TAG — skipping build"
+  exit 0
+fi
+echo "Image not found, proceeding with build..."
+
+# --- CUDA 13.0 for nightly builds ---
+# Nightly CI uses CUDA 13.0 while regular CI stays on CUDA 12.9
+NIGHTLY_CUDA_VERSION="13.0.2"
+NIGHTLY_BUILD_BASE_IMAGE="nvidia/cuda:${NIGHTLY_CUDA_VERSION}-devel-ubuntu22.04"
+NIGHTLY_FINAL_BASE_IMAGE="nvidia/cuda:${NIGHTLY_CUDA_VERSION}-base-ubuntu22.04"
+
+echo "--- :docker: Building torch nightly image (CUDA ${NIGHTLY_CUDA_VERSION})"
+docker buildx build --file docker/Dockerfile \
+  --build-arg max_jobs=16 \
+  --build-arg buildkite_commit="$BUILDKITE_COMMIT" \
+  --build-arg USE_SCCACHE=1 \
+  --build-arg PYTORCH_NIGHTLY=1 \
+  --build-arg CUDA_VERSION="${NIGHTLY_CUDA_VERSION}" \
+  --build-arg BUILD_BASE_IMAGE="${NIGHTLY_BUILD_BASE_IMAGE}" \
+  --build-arg FINAL_BASE_IMAGE="${NIGHTLY_FINAL_BASE_IMAGE}" \
+  --build-arg torch_cuda_arch_list="8.0 8.9 9.0 10.0 12.0" \
+  --tag "$IMAGE_TAG" \
+  --push \
+  --target test \
+  --progress plain .
+
+echo "--- :white_check_mark: Torch nightly image build complete: $IMAGE_TAG"

.buildkite/intel_jobs/engine_intel.yaml

@@ -0,0 +1,21 @@
+group: Engine Intel
+depends_on:
+  - image-build-xpu
+steps:
+- label: Engine (1 GPU)
+  timeout_in_minutes: 30
+  device: intel_gpu
+  no_plugin: true
+  working_dir: "."
+  env:
+    REGISTRY: "public.ecr.aws/q9t5s3a7"
+    REPO: "vllm-ci-test-repo"
+    VLLM_TEST_DEVICE: "xpu"
+  source_file_dependencies:
+    - vllm/v1/engine/
+    - tests/v1/engine/
+  commands:
+    - >-
+      bash .buildkite/scripts/hardware_ci/run-intel-test.sh
+      'cd tests &&
+      pytest -v -s v1/engine --ignore v1/engine/test_preprocess_error_handling.py'

.buildkite/intel_jobs/kernels_intel.yaml

@@ -0,0 +1,21 @@
+group: Kernels Intel
+depends_on: 
+  - image-build-xpu
+steps:
+- label: vLLM IR Tests
+  timeout_in_minutes: 30
+  device: intel_gpu
+  no_plugin: true
+  working_dir: "."
+  env:
+    REGISTRY: "public.ecr.aws/q9t5s3a7"
+    REPO: "vllm-ci-test-repo"
+    VLLM_TEST_DEVICE: "xpu"
+  source_file_dependencies:
+    - vllm/ir
+    - vllm/kernels
+  commands:
+    - >-
+      bash .buildkite/scripts/hardware_ci/run-intel-test.sh
+      'cd tests &&
+      pytest -v -s kernels/ir'

.buildkite/intel_jobs/lora_intel.yaml

@@ -0,0 +1,130 @@
+group: LoRA Intel
+depends_on:
+  - image-build-xpu
+steps:
+- label: LoRA Runtime + Utils
+  timeout_in_minutes: 45
+  device: intel_gpu
+  no_plugin: true
+  working_dir: "."
+  env:
+    REGISTRY: "public.ecr.aws/q9t5s3a7"
+    REPO: "vllm-ci-test-repo"
+    VLLM_TEST_DEVICE: "xpu"
+  source_file_dependencies:
+    - vllm/lora
+    - tests/lora
+  commands:
+    - >-
+      bash .buildkite/scripts/hardware_ci/run-intel-test.sh
+      'cd tests &&
+      pytest -v -s lora/test_layers.py &&
+      pytest -v -s lora/test_lora_checkpoints.py &&
+      (pytest -v -s lora/test_lora_functions.py --deselect="tests/lora/test_lora_functions.py::test_lora_functions_sync" --deselect="tests/lora/test_lora_functions.py::test_lora_functions_async" || true) &&
+      pytest -v -s lora/test_lora_huggingface.py &&
+      pytest -v -s lora/test_lora_manager.py &&
+      pytest -v -s lora/test_lora_utils.py &&
+      pytest -v -s lora/test_peft_helper.py &&
+      pytest -v -s lora/test_resolver.py &&
+      pytest -v -s lora/test_utils.py &&
+      (pytest -v -s lora/test_add_lora.py --deselect="tests/lora/test_add_lora.py::test_add_lora" || true) &&
+      (pytest -v -s lora/test_worker.py --deselect="tests/lora/test_worker.py::test_worker_apply_lora" || true)'
+
+- label: LoRA Fused/MoE Kernels
+  timeout_in_minutes: 45
+  device: intel_gpu
+  no_plugin: true
+  working_dir: "."
+  env:
+    REGISTRY: "public.ecr.aws/q9t5s3a7"
+    REPO: "vllm-ci-test-repo"
+    VLLM_TEST_DEVICE: "xpu"
+  source_file_dependencies:
+    - vllm/lora
+    - tests/lora
+  commands:
+    - >-
+      bash .buildkite/scripts/hardware_ci/run-intel-test.sh
+      'cd tests &&
+      pytest -v -s lora/test_fused_moe_lora_kernel.py && 
+      pytest -v -s lora/test_moe_lora_align_sum.py'
+
+- label: LoRA Punica Kernels
+  timeout_in_minutes: 45
+  device: intel_gpu
+  no_plugin: true
+  working_dir: "."
+  env:
+    REGISTRY: "public.ecr.aws/q9t5s3a7"
+    REPO: "vllm-ci-test-repo"
+    VLLM_TEST_DEVICE: "xpu"
+  source_file_dependencies:
+    - vllm/lora
+    - tests/lora
+  commands:
+    - >-
+      bash .buildkite/scripts/hardware_ci/run-intel-test.sh
+      'cd tests &&
+      set -o pipefail &&
+      pytest -v -s lora/test_punica_ops.py --deselect="tests/lora/test_punica_ops.py::test_kernels[shrink-0-xpu:0-dtype0-2-2049-64-32-32]" --deselect="tests/lora/test_punica_ops.py::test_kernels_hidden_size[expand-0-xpu:0-dtype1-2-64000-32-4-4]" --deselect="tests/lora/test_punica_ops.py::test_kernels[shrink-0-xpu:0-dtype0-1-2049-128-1-32]" --deselect="tests/lora/test_punica_ops.py::test_kernels[shrink-0-xpu:0-dtype0-1-2049-256-1-4]" --deselect="tests/lora/test_punica_ops.py::test_kernels[shrink-0-xpu:0-dtype0-1-2049-256-8-4]" --deselect="tests/lora/test_punica_ops.py::test_kernels[expand-0-xpu:0-dtype0-3-2049-128-8-16]" --deselect="tests/lora/test_punica_ops.py::test_kernels[shrink-0-xpu:0-dtype0-1-2049-128-8-32]" --deselect="tests/lora/test_punica_ops.py::test_kernels[expand-0-xpu:0-dtype1-1-2049-256-128-32]" --deselect="tests/lora/test_punica_ops.py::test_kernels_hidden_size[shrink-0-xpu:0-dtype0-3-64256-32-4-4]" --deselect="tests/lora/test_punica_ops.py::test_kernels_hidden_size[shrink-0-xpu:0-dtype1-2-29696-32-4-4]" --deselect="tests/lora/test_punica_ops.py::test_kernels_hidden_size[shrink-0-xpu:0-dtype1-3-49408-32-4-4]" --deselect="tests/lora/test_punica_ops.py::test_kernels_hidden_size[shrink-0-xpu:0-dtype0-2-16384-32-4-4]" --deselect="tests/lora/test_punica_ops.py::test_kernels_hidden_size[expand-0-xpu:0-dtype0-2-51328-32-4-4]" --deselect="tests/lora/test_punica_ops.py::test_kernels_hidden_size[expand-0-xpu:0-dtype1-1-102656-32-4-4]"'
+
+- label: LoRA Punica FP8/XPU Ops
+  timeout_in_minutes: 45
+  device: intel_gpu
+  no_plugin: true
+  working_dir: "."
+  env:
+    REGISTRY: "public.ecr.aws/q9t5s3a7"
+    REPO: "vllm-ci-test-repo"
+    VLLM_TEST_DEVICE: "xpu"
+  source_file_dependencies:
+    - vllm/lora
+    - tests/lora
+  commands:
+    - >-
+      bash .buildkite/scripts/hardware_ci/run-intel-test.sh
+      'cd tests &&
+      pytest -v -s lora/test_punica_ops_fp8.py &&
+      pytest -v -s lora/test_punica_xpu_ops.py'
+
+- label: LoRA Models
+  timeout_in_minutes: 45
+  device: intel_gpu
+  no_plugin: true
+  working_dir: "."
+  env:
+    REGISTRY: "public.ecr.aws/q9t5s3a7"
+    REPO: "vllm-ci-test-repo"
+    VLLM_TEST_DEVICE: "xpu"
+  source_file_dependencies:
+    - vllm/lora
+    - tests/lora
+  commands:
+    - >-
+      bash .buildkite/scripts/hardware_ci/run-intel-test.sh
+      'cd tests &&
+      (pytest -v -s lora/test_mixtral.py --deselect="tests/lora/test_mixtral.py::test_mixtral_lora[4]" || true) &&
+      pytest -v -s lora/test_quant_model.py --deselect="tests/lora/test_quant_model.py::test_quant_model_lora[model0]" --deselect="tests/lora/test_quant_model.py::test_quant_model_lora[model1]" --deselect="tests/lora/test_quant_model.py::test_quant_model_tp_equality[model0]" &&
+      pytest -v -s lora/test_qwen35_densemodel_lora.py &&
+      pytest -v -s lora/test_transformers_model.py'
+
+- label: LoRA Multimodal
+  timeout_in_minutes: 45
+  device: intel_gpu
+  no_plugin: true
+  working_dir: "."
+  env:
+    REGISTRY: "public.ecr.aws/q9t5s3a7"
+    REPO: "vllm-ci-test-repo"
+    VLLM_TEST_DEVICE: "xpu"
+  source_file_dependencies:
+    - vllm/lora
+    - tests/lora
+  commands:
+    - >-
+      bash .buildkite/scripts/hardware_ci/run-intel-test.sh
+      'cd tests &&
+      pytest -v -s lora/test_default_mm_loras.py && 
+      (pytest -v -s lora/test_qwen3_unembed.py || true) &&
+      (pytest -v -s lora/test_qwenvl.py || true) &&
+      pytest -v -s lora/test_whisper.py'

.buildkite/intel_jobs/misc_intel.yaml

@@ -0,0 +1,55 @@
+group: Miscellaneous Intel
+depends_on:
+  - image-build-xpu
+steps:
+- label: V1 Core + KV + Metrics
+  timeout_in_minutes: 30
+  device: intel_gpu
+  no_plugin: true
+  working_dir: "."
+  env:
+    REGISTRY: "public.ecr.aws/q9t5s3a7"
+    REPO: "vllm-ci-test-repo"
+    VLLM_TEST_DEVICE: "xpu"
+  source_file_dependencies:
+    - vllm/
+    - tests/v1/core
+    - tests/v1/executor
+    - tests/v1/kv_offload
+    - tests/v1/worker
+    - tests/v1/kv_connector/unit
+    - tests/v1/metrics
+    - tests/entrypoints/openai/correctness/test_lmeval.py
+  commands:
+    - >-
+      bash .buildkite/scripts/hardware_ci/run-intel-test.sh
+      'pip install -r requirements/kv_connectors.txt &&
+      export VLLM_WORKER_MULTIPROC_METHOD=spawn &&
+      cd tests &&
+      pytest -v -s v1/executor'
+
+- label: V1 Sample + Logits
+  timeout_in_minutes: 30
+  device: intel_gpu
+  no_plugin: true
+  working_dir: "."
+  env:
+    REGISTRY: "public.ecr.aws/q9t5s3a7"
+    REPO: "vllm-ci-test-repo"
+    VLLM_TEST_DEVICE: "xpu"
+  source_file_dependencies:
+    - vllm/
+    - tests/v1/sample
+    - tests/v1/logits_processors
+    - tests/v1/test_oracle.py
+    - tests/v1/test_request.py
+    - tests/v1/test_outputs.py
+  commands:
+    - >-
+      bash .buildkite/scripts/hardware_ci/run-intel-test.sh
+      'export VLLM_WORKER_MULTIPROC_METHOD=spawn &&
+      cd tests &&
+      pytest -v -s v1/logits_processors &&
+      pytest -v -s v1/test_oracle.py &&
+      pytest -v -s v1/test_request.py &&
+      pytest -v -s v1/test_outputs.py'

.buildkite/intel_jobs/test-intel.yaml

@@ -35,6 +35,7 @@ steps:
         python3 examples/basic/offline_inference/generate.py --model facebook/opt-125m --block-size 64 --enforce-eager -tp 2 --distributed-executor-backend mp &&
         python3 examples/basic/offline_inference/generate.py --model facebook/opt-125m --block-size 64 --enforce-eager --attention-backend=TRITON_ATTN &&
         python3 examples/basic/offline_inference/generate.py --model facebook/opt-125m --block-size 64 --enforce-eager --quantization fp8 &&
+        python3 examples/basic/offline_inference/generate.py --model facebook/opt-125m --block-size 64 --enforce-eager --kv-cache-dtype fp8 &&
         python3 examples/basic/offline_inference/generate.py --model superjob/Qwen3-4B-Instruct-2507-GPTQ-Int4 --block-size 64 --enforce-eager --max-model-len 8192 &&
         python3 examples/basic/offline_inference/generate.py --model ibm-research/PowerMoE-3b --block-size 64 --enforce-eager -tp 2 &&
         python3 examples/basic/offline_inference/generate.py --model ibm-research/PowerMoE-3b --block-size 64 --enforce-eager -tp 2 --enable-expert-parallel'
@@ -61,4 +62,4 @@ steps:
         pytest -v -s v1/structured_output &&
         pytest -v -s v1/test_serial_utils.py &&
         pytest -v -s v1/spec_decode --ignore=v1/spec_decode/test_max_len.py --ignore=v1/spec_decode/test_tree_attention.py --ignore=v1/spec_decode/test_speculators_eagle3.py --ignore=v1/spec_decode/test_acceptance_length.py &&
-        pytest -v -s v1/kv_connector/unit --ignore=v1/kv_connector/unit/test_multi_connector.py --ignore=v1/kv_connector/unit/test_nixl_connector.py --ignore=v1/kv_connector/unit/test_example_connector.py --ignore=v1/kv_connector/unit/test_lmcache_integration.py'
+        pytest -v -s v1/kv_connector/unit --ignore=v1/kv_connector/unit/test_multi_connector.py --ignore=v1/kv_connector/unit/test_example_connector.py --ignore=v1/kv_connector/unit/test_lmcache_integration.py --ignore=v1/kv_connector/unit/test_hf3fs_client.py --ignore=v1/kv_connector/unit/test_hf3fs_connector.py --ignore=v1/kv_connector/unit/test_hf3fs_metadata_server.py'