Merge branch 'main' of github.com:rasmith/vllm

.buildkite/run-amd-test.sh

@@ -71,13 +71,36 @@ mkdir -p ${HF_CACHE}
 HF_MOUNT="/root/.cache/huggingface"
 
 commands=$@
+echo "Commands:$commands"
+#ignore certain kernels tests
+if [[ $commands == *" kernels "* ]]; then
+  commands="${commands} \
+  --ignore=kernels/test_attention.py \
+  --ignore=kernels/test_attention_selector.py \
+  --ignore=kernels/test_blocksparse_attention.py \
+  --ignore=kernels/test_causal_conv1d.py \
+  --ignore=kernels/test_cutlass.py \
+  --ignore=kernels/test_encoder_decoder_attn.py \
+  --ignore=kernels/test_flash_attn.py \
+  --ignore=kernels/test_flashinfer.py \
+  --ignore=kernels/test_int8_quant.py \
+  --ignore=kernels/test_machete_gemm.py \
+  --ignore=kernels/test_mamba_ssm.py \
+  --ignore=kernels/test_marlin_gemm.py \
+  --ignore=kernels/test_moe.py \
+  --ignore=kernels/test_prefix_prefill.py \
+  --ignore=kernels/test_rand.py \
+  --ignore=kernels/test_sampler.py"
+fi
+
 PARALLEL_JOB_COUNT=8
 # check if the command contains shard flag, we will run all shards in parallel because the host have 8 GPUs. 
 if [[ $commands == *"--shard-id="* ]]; then
   for GPU in $(seq 0 $(($PARALLEL_JOB_COUNT-1))); do
     #replace shard arguments
-    commands=${@//"--shard-id= "/"--shard-id=${GPU} "}
+    commands=${commands//"--shard-id= "/"--shard-id=${GPU} "}
     commands=${commands//"--num-shards= "/"--num-shards=${PARALLEL_JOB_COUNT} "}
+    echo "Shard ${GPU} commands:$commands"
     docker run \
         --device /dev/kfd --device /dev/dri \
         --network host \

.buildkite/run-cpu-test-ppc64le.sh

@@ -11,8 +11,9 @@ trap remove_docker_container EXIT
 remove_docker_container
 
 # Run the image, setting --shm-size=4g for tensor parallel.
+source /etc/environment
 #docker run -itd --entrypoint /bin/bash -v ~/.cache/huggingface:/root/.cache/huggingface --privileged=true --network host -e HF_TOKEN --env VLLM_CPU_KVCACHE_SPACE=4 --shm-size=4g --name cpu-test cpu-test
-docker run -itd --entrypoint /bin/bash -v ~/.cache/huggingface:/root/.cache/huggingface --privileged=true --network host -e HF_TOKEN --name cpu-test cpu-test
+docker run -itd --entrypoint /bin/bash -v ~/.cache/huggingface:/root/.cache/huggingface --privileged=true --network host -e HF_TOKEN=$HF_TOKEN --name cpu-test cpu-test
 
 # Run basic model test
 docker exec cpu-test bash -c "

.buildkite/run-cpu-test.sh

@@ -30,6 +30,12 @@ docker exec cpu-test bash -c "
       --ignore=tests/models/test_jamba.py \
       --ignore=tests/models/test_danube3_4b.py" # Mamba and Danube3-4B on CPU is not supported
 
+# Run compressed-tensor test
+docker exec cpu-test bash -c "
+  pytest -s -v \
+  tests/quantization/test_compressed_tensors.py::test_compressed_tensors_w8a8_static_setup \
+  tests/quantization/test_compressed_tensors.py::test_compressed_tensors_w8a8_dynanmic_per_token"
+
 # online inference
 docker exec cpu-test bash -c "
   export VLLM_CPU_KVCACHE_SPACE=10 

.buildkite/test-pipeline.yaml

@@ -91,6 +91,7 @@ steps:
   - pytest -v -s entrypoints/llm/test_lazy_outlines.py # it needs a clean process
   - pytest -v -s entrypoints/openai
   - pytest -v -s entrypoints/test_chat_utils.py
+  - pytest -v -s entrypoints/offline_mode # Needs to avoid interference with other tests
 
 
 - label: Distributed Tests (4 GPUs) # 10min
@@ -217,7 +218,8 @@ steps:
   commands:
     # See https://github.com/vllm-project/vllm/issues/5152
     - export VLLM_ATTENTION_BACKEND=XFORMERS
-    - pytest -v -s spec_decode
+    - pytest -v -s spec_decode/e2e/test_multistep_correctness.py
+    - pytest -v -s spec_decode --ignore=spec_decode/e2e/test_multistep_correctness.py
 
 - label: LoRA Test %N # 30min each
   mirror_hardwares: [amd]
@@ -228,6 +230,7 @@ steps:
   parallelism: 4
 
 - label: Kernels Test %N # 30min each
+  mirror_hardwares: [amd]
   source_file_dependencies:
   - csrc/
   - vllm/attention
@@ -386,7 +389,18 @@ steps:
   - vllm/
   - tests/weight_loading
   commands:
-    - bash weight_loading/run_model_weight_loading_test.sh
+    - bash weight_loading/run_model_weight_loading_test.sh -c weight_loading/models.txt
+
+- label: Weight Loading Multiple GPU Test - Large Models # optional
+  working_dir: "/vllm-workspace/tests"
+  num_gpus: 2
+  gpu: a100
+  optional: true
+  source_file_dependencies:
+  - vllm/
+  - tests/weight_loading
+  commands:
+    - bash weight_loading/run_model_weight_loading_test.sh -c weight_loading/models-large.txt 
 
 
 ##### multi gpus test #####

.github/ISSUE_TEMPLATE/400-bug report.yml

@@ -30,6 +30,15 @@ body:
       </details>
   validations:
     required: true
+- type: textarea
+  attributes:
+    label: Model Input Dumps
+    description: |
+      If you are facing crashing due to illegal memory access or other issues with model execution, vLLM may dump the problematic input of the model. In this case, you will see the message `Error in model execution (input dumped to /tmp/err_xxx.pkl)`. If you see this message, please zip the file (because GitHub doesn't support .pkl file format) and upload it here. This will help us to reproduce the issue and facilitate the debugging process.
+    placeholder: |
+      Upload the dumped input file.
+  validations:
+    required: false
 - type: textarea
   attributes:
     label: 🐛 Describe the bug

.github/PULL_REQUEST_TEMPLATE.md

@@ -39,6 +39,16 @@ FIX #xxxx (*link existing issues this PR will resolve*)
     <li>Please add documentation to <code>docs/source/</code> if the PR modifies the user-facing behaviors of vLLM. It helps vLLM user understand and utilize the new features or changes.</li>
 </ul>
 
+<h3>Adding or changing kernels</h3>
+<p>Each custom kernel needs a schema and one or more implementations to be registered with PyTorch.</p>
+<ul>
+    <li>Make sure custom ops are registered following PyTorch guidelines: <a href="https://pytorch.org/tutorials/advanced/cpp_custom_ops.html#cpp-custom-ops-tutorial">Custom C++ and CUDA Operators</a> and <a href="https://docs.google.com/document/d/1_W62p8WJOQQUzPsJYa7s701JXt0qf2OfLub2sbkHOaU">The Custom Operators Manual</a></li>
+    <li>Custom operations that return <code>Tensors</code> require meta-functions. Meta-functions should be implemented and registered in python so that dynamic dims can be handled automatically. See above documents for a description of meta-functions.</li>
+    <li>Use <a href="https://pytorch.org/docs/stable/library.html#torch.library.opcheck"><code>torch.libary.opcheck()</code></a> to test the function registration and meta-function for any registered ops.  See <code>tests/kernels</code> for examples.</li>
+    <li>When changing the C++ signature of an existing op, the schema must be updated to reflect the changes.</li>
+    <li>If a new custom type is needed, see the following document: <a href="https://docs.google.com/document/d/18fBMPuOJ0fY5ZQ6YyrHUppw9FA332CpNtgB6SOIgyuA">Custom Class Support in PT2</a>.
+</ul>
+
 <h3>Notes for Large Changes</h3>
 <p>Please keep the changes as concise as possible. For major architectural changes (>500 LOC excluding kernel/data/config/test), we would expect a GitHub issue (RFC) discussing the technical design and justification. Otherwise, we will tag it with <code>rfc-required</code> and might not go through the PR.</p>
 

CMakeLists.txt

@@ -195,9 +195,13 @@ if(VLLM_GPU_LANG STREQUAL "CUDA")
   FetchContent_Declare(
         cutlass
         GIT_REPOSITORY https://github.com/nvidia/cutlass.git
-        # CUTLASS 3.5.1
-        GIT_TAG 06b21349bcf6ddf6a1686a47a137ad1446579db9 
+        GIT_TAG v3.5.1
         GIT_PROGRESS TRUE
+
+        # Speed up CUTLASS download by retrieving only the specified GIT_TAG instead of the history.
+        # Important: If GIT_SHALLOW is enabled then GIT_TAG works only with branch names and tags.
+        # So if the GIT_TAG above is updated to a commit hash, GIT_SHALLOW must be set to FALSE
+        GIT_SHALLOW TRUE
   )
   FetchContent_MakeAvailable(cutlass)
 
@@ -231,6 +235,7 @@ if(VLLM_GPU_LANG STREQUAL "CUDA")
           "-gencode arch=compute_90a,code=sm_90a")
   endif()
 
+
   #
   # Machete kernels
 
@@ -289,6 +294,12 @@ define_gpu_extension_target(
   USE_SABI 3
   WITH_SOABI)
 
+# If CUTLASS is compiled on NVCC >= 12.5, it by default uses 
+# cudaGetDriverEntryPointByVersion as a wrapper to avoid directly calling the 
+# driver API. This causes problems when linking with earlier versions of CUDA.
+# Setting this variable sidesteps the issue by calling the driver directly.
+target_compile_definitions(_C PRIVATE CUTLASS_ENABLE_DIRECT_CUDA_DRIVER_CALL=1)
+
 #
 # _moe_C extension
 #

Dockerfile

@@ -145,6 +145,7 @@ RUN echo 'tzdata tzdata/Areas select America' | debconf-set-selections \
     && echo 'tzdata tzdata/Zones/America select Los_Angeles' | debconf-set-selections \
     && apt-get update -y \
     && apt-get install -y ccache software-properties-common git curl sudo vim python3-pip \
+    && apt-get install -y ffmpeg libsm6 libxext6 libgl1 \
     && add-apt-repository ppa:deadsnakes/ppa \
     && apt-get update -y \
     && apt-get install -y python${PYTHON_VERSION} python${PYTHON_VERSION}-dev python${PYTHON_VERSION}-venv libibverbs-dev \

Dockerfile.cpu

@@ -2,9 +2,14 @@
 
 FROM ubuntu:22.04 AS cpu-test-1
 
+ENV CCACHE_DIR=/root/.cache/ccache
+
+ENV CMAKE_CXX_COMPILER_LAUNCHER=ccache
+
 RUN --mount=type=cache,target=/var/cache/apt \
     apt-get update -y \
     && apt-get install -y curl ccache git wget vim numactl gcc-12 g++-12 python3 python3-pip libtcmalloc-minimal4 libnuma-dev \
+    && apt-get install -y ffmpeg libsm6 libxext6 libgl1 \
     && update-alternatives --install /usr/bin/gcc gcc /usr/bin/gcc-12 10 --slave /usr/bin/g++ g++ /usr/bin/g++-12
 
 # https://intel.github.io/intel-extension-for-pytorch/cpu/latest/tutorials/performance_tuning/tuning_guide.html
@@ -25,6 +30,19 @@ RUN --mount=type=cache,target=/root/.cache/pip \
     pip install --upgrade pip && \
     pip install -r requirements-build.txt
 
+# install oneDNN
+RUN git clone -b rls-v3.5 https://github.com/oneapi-src/oneDNN.git
+
+RUN --mount=type=cache,target=/root/.cache/ccache \
+    cmake -B ./oneDNN/build -S ./oneDNN -G Ninja -DONEDNN_LIBRARY_TYPE=STATIC \ 
+    -DONEDNN_BUILD_DOC=OFF \ 
+    -DONEDNN_BUILD_EXAMPLES=OFF \ 
+    -DONEDNN_BUILD_TESTS=OFF \ 
+    -DONEDNN_BUILD_GRAPH=OFF \ 
+    -DONEDNN_ENABLE_WORKLOAD=INFERENCE \ 
+    -DONEDNN_ENABLE_PRIMITIVE=MATMUL && \
+    cmake --build ./oneDNN/build --target install --config Release
+
 FROM cpu-test-1 AS build
 
 WORKDIR /workspace/vllm
@@ -40,7 +58,6 @@ COPY ./ ./
 ARG VLLM_CPU_DISABLE_AVX512
 ENV VLLM_CPU_DISABLE_AVX512=${VLLM_CPU_DISABLE_AVX512}
 
-ENV CCACHE_DIR=/root/.cache/ccache
 RUN --mount=type=cache,target=/root/.cache/pip \
     --mount=type=cache,target=/root/.cache/ccache \
     VLLM_TARGET_DEVICE=cpu python3 setup.py bdist_wheel && \

Dockerfile.neuron

@@ -6,7 +6,9 @@ FROM $BASE_IMAGE
 RUN echo "Base image is $BASE_IMAGE"
 
 # Install some basic utilities
-RUN apt-get update && apt-get install python3 python3-pip -y
+RUN apt-get update \
+    && apt-get install python3 python3-pip -y \
+    && apt-get install -y ffmpeg libsm6 libxext6 libgl1 
 
 ### Mount Point ###
 # When launching the container, mount the code directory to /app

Dockerfile.openvino

@@ -4,7 +4,8 @@
 FROM ubuntu:22.04 AS dev
 
 RUN apt-get update -y && \
-    apt-get install -y python3-pip git
+    apt-get install -y python3-pip git && \
+    apt-get install -y ffmpeg libsm6 libxext6 libgl1 
 WORKDIR /workspace
 
 # copy requirements

Dockerfile.ppc64le

@@ -4,7 +4,7 @@ USER root
 
 ENV PATH="/usr/local/cargo/bin:$PATH:/opt/conda/bin/"
 
-RUN apt-get update -y && apt-get install -y git wget vim libnuma-dev libsndfile-dev libprotobuf-dev build-essential
+RUN apt-get update -y && apt-get install -y git wget curl vim libnuma-dev libsndfile-dev libprotobuf-dev build-essential ffmpeg libsm6 libxext6 libgl1 
 
 # Some packages in requirements-cpu are installed here
 # IBM provides optimized packages for ppc64le processors in the open-ce project for mamba
@@ -16,7 +16,7 @@ COPY ./ /workspace/vllm
 WORKDIR /workspace/vllm
 
 # These packages will be in rocketce eventually
-RUN pip install -v cmake torch==2.3.1 uvloop==0.20.0 -r requirements-cpu.txt --prefer-binary --extra-index-url https://repo.fury.io/mgiessing
+RUN pip install -v cmake xformers torch==2.3.1 uvloop==0.20.0 -r requirements-cpu.txt --prefer-binary --extra-index-url https://repo.fury.io/mgiessing
 
 RUN VLLM_TARGET_DEVICE=cpu python3 setup.py install
 
@@ -25,4 +25,3 @@ WORKDIR /workspace/
 RUN ln -s /workspace/vllm/tests && ln -s /workspace/vllm/examples && ln -s /workspace/vllm/benchmarks
 
 ENTRYPOINT ["python3", "-m", "vllm.entrypoints.openai.api_server"]
-

Dockerfile.tpu

@@ -4,6 +4,9 @@ ARG BASE_IMAGE="us-central1-docker.pkg.dev/tpu-pytorch-releases/docker/xla:night
 FROM $BASE_IMAGE
 WORKDIR /workspace
 
+# Install some basic utilities
+RUN apt-get update && apt-get install -y ffmpeg libsm6 libxext6 libgl1 
+
 # Install the TPU and Pallas dependencies.
 RUN python3 -m pip install torch_xla[tpu] -f https://storage.googleapis.com/libtpu-releases/index.html
 RUN python3 -m pip install torch_xla[pallas] -f https://storage.googleapis.com/jax-releases/jax_nightly_releases.html -f https://storage.googleapis.com/jax-releases/jaxlib_nightly_releases.html

Dockerfile.xpu

@@ -9,8 +9,7 @@ RUN wget -O- https://apt.repos.intel.com/intel-gpg-keys/GPG-PUB-KEY-INTEL-SW-PRO
     chmod 644 /usr/share/keyrings/intel-graphics.gpg
 
 RUN apt-get update  -y \
-&& apt-get install -y curl libicu70 lsb-release git wget vim numactl python3 python3-pip
-
+&& apt-get install -y curl libicu70 lsb-release git wget vim numactl python3 python3-pip ffmpeg libsm6 libxext6 libgl1 
 COPY ./ /workspace/vllm
 
 WORKDIR /workspace/vllm

README.md

@@ -17,15 +17,16 @@ Easy, fast, and cheap LLM serving for everyone
 
 ---
 
-**vLLM & NVIDIA Triton User Meetup (Monday, September 9, 5pm-9pm PT) at Fort Mason, San Francisco**
+**vLLM, AMD, Anyscale Meet & Greet at [Ray Summit 2024](http://raysummit.anyscale.com) (Monday, Sept 30th, 5-7pm PT) at Marriott Marquis San Francisco**
 
-We are excited to announce our sixth vLLM Meetup, in collaboration with NVIDIA Triton Team.
-Join us to hear the vLLM's recent update about performance.
-Register now [here](https://lu.ma/87q3nvnh) and be part of the event!
+We are excited to announce our special vLLM event in collaboration with AMD and Anyscale.
+Join us to learn more about recent advancements of vLLM on MI300X.
+Register [here](https://lu.ma/db5ld9n5) and be a part of the event!
 
 ---
 
 *Latest News* 🔥
+- [2024/09] We hosted [the sixth vLLM meetup](https://lu.ma/87q3nvnh) with NVIDIA! Please find the meetup slides [here](https://docs.google.com/presentation/d/1wrLGwytQfaOTd5wCGSPNhoaW3nq0E-9wqyP7ny93xRs/edit?usp=sharing).
 - [2024/07] We hosted [the fifth vLLM meetup](https://lu.ma/lp0gyjqr) with AWS! Please find the meetup slides [here](https://docs.google.com/presentation/d/1RgUD8aCfcHocghoP3zmXzck9vX3RCI9yfUAB2Bbcl4Y/edit?usp=sharing).
 - [2024/07] In partnership with Meta, vLLM officially supports Llama 3.1 with FP8 quantization and pipeline parallelism! Please check out our blog post [here](https://blog.vllm.ai/2024/07/23/llama31.html).
 - [2024/06] We hosted [the fourth vLLM meetup](https://lu.ma/agivllm) with Cloudflare and BentoML! Please find the meetup slides [here](https://docs.google.com/presentation/d/1iJ8o7V2bQEi0BFEljLTwc5G1S10_Rhv3beed5oB0NJ4/edit?usp=sharing).
@@ -130,3 +131,10 @@ If you use vLLM for your research, please cite our [paper](https://arxiv.org/abs
   year={2023}
 }
 ```
+
+## Contact Us
+
+* For technical questions and feature requests, please use Github issues or discussions.
+* For discussing with fellow users, please use Discord.
+* For security disclosures, please use Github's security advisory feature.
+* For collaborations and partnerships, please contact us at vllm-questions AT lists.berkeley.edu.

benchmarks/benchmark_latency.py

@@ -10,7 +10,7 @@
 from tqdm import tqdm
 
 from vllm import LLM, SamplingParams
-from vllm.engine.arg_utils import EngineArgs
+from vllm.engine.arg_utils import DEVICE_OPTIONS, EngineArgs
 from vllm.inputs import PromptInputs
 from vllm.model_executor.layers.quantization import QUANTIZATION_METHODS
 from vllm.utils import FlexibleArgumentParser
@@ -205,13 +205,11 @@ def run_to_completion(profile_dir: Optional[str] = None):
         default=None,
         help=('path to save the pytorch profiler output. Can be visualized '
               'with ui.perfetto.dev or Tensorboard.'))
-    parser.add_argument(
-        "--device",
-        type=str,
-        default="auto",
-        choices=["auto", "cuda", "cpu", "openvino", "tpu", "xpu"],
-        help='device type for vLLM execution, supporting CUDA, OpenVINO and '
-        'CPU.')
+    parser.add_argument("--device",
+                        type=str,
+                        default="auto",
+                        choices=DEVICE_OPTIONS,
+                        help='device type for vLLM execution')
     parser.add_argument('--block-size',
                         type=int,
                         default=16,

benchmarks/benchmark_throughput.py

@@ -11,7 +11,7 @@
 from transformers import (AutoModelForCausalLM, AutoTokenizer,
                           PreTrainedTokenizerBase)
 
-from vllm.engine.arg_utils import AsyncEngineArgs, EngineArgs
+from vllm.engine.arg_utils import DEVICE_OPTIONS, AsyncEngineArgs, EngineArgs
 from vllm.entrypoints.openai.api_server import (
     build_async_engine_client_from_engine_args)
 from vllm.model_executor.layers.quantization import QUANTIZATION_METHODS
@@ -451,13 +451,11 @@ def main(args: argparse.Namespace):
         'accuracy issues. FP8_E5M2 (without scaling) is only supported on '
         'cuda version greater than 11.8. On ROCm (AMD GPU), FP8_E4M3 is '
         'instead supported for common inference criteria.')
-    parser.add_argument(
-        "--device",
-        type=str,
-        default="auto",
-        choices=["auto", "cuda", "cpu", "openvino", "tpu", "xpu"],
-        help='device type for vLLM execution, supporting CUDA, OpenVINO and '
-        'CPU.')
+    parser.add_argument("--device",
+                        type=str,
+                        default="auto",
+                        choices=DEVICE_OPTIONS,
+                        help='device type for vLLM execution')
     parser.add_argument(
         "--num-scheduler-steps",
         type=int,