hf_text-generation-inference/Dockerfile_amd

221 lines
6.6 KiB
Plaintext
Raw Normal View History

# Rust builder
FROM lukemathwalker/cargo-chef:latest-rust-1.79 AS chef
WORKDIR /usr/src
ARG CARGO_REGISTRIES_CRATES_IO_PROTOCOL=sparse
FROM chef AS planner
COPY Cargo.lock Cargo.lock
COPY Cargo.toml Cargo.toml
COPY rust-toolchain.toml rust-toolchain.toml
COPY proto proto
COPY benchmark benchmark
COPY router router
Rebase TRT-llm (#2331) * wip wip refacto refacto Initial setup for CXX binding to TRTLLM Working FFI call for TGI and TRTLLM backend Remove unused parameters annd force tokenizer name to be set Overall build TRTLLM and deps through CMake build system Enable end to end CMake build First version loading engines and making it ready for inference Remembering to check how we can detect support for chunked context Move to latest TensorRT-LLM version Specify which default log level to use depending on CMake build type make leader executor mode working unconditionally call InitializeBackend on the FFI layer bind to CUDA::nvml to retrieve compute capabilities at runtime updated logic and comment to detect cuda compute capabilities implement the Stream method to send new tokens through a callback use spdlog release 1.14.1 moving forward update trtllm to latest version a96cccafcf6365c128f004f779160951f8c0801c correctly tell cmake to build dependent tensorrt-llm required libraries create cmake install target to put everything relevant in installation folder add auth_token CLI argument to provide hf hub authentification token allow converting huggingface::tokenizers error to TensorRtLlmBackendError use correct include for spdlog include guard to build example in cmakelists working setup of the ffi layer remove fmt import use external fmt lib end to end ffi flow working make sure to track include/ffi.h to trigger rebuild from cargo impl the rust backend which currently cannot move the actual computation in background thread expose shutdown function at ffi layer impl RwLock scenario for TensorRtLllmBackend oops missing c++ backend definitions compute the number of maximum new tokens for each request independently make sure the context is not dropped in the middle of the async decoding. remove unnecessary log add all the necessary plumbery to return the generated content update invalid doc in cpp file correctly forward back the log probabilities remove unneeded scope variable for now refactor Stream impl for Generation to factorise code expose the internal missing start/queue timestamp forward tgi parameters rep/freq penalty add some more validation about grammar not supported define a shared struct to hold the result of a decoding step expose information about potential error happening while decoding remove logging add logging in case of decoding error make sure executor_worker is provided add initial Dockerfile for TRTLLM backend add some more information in CMakeLists.txt to correctly install executorWorker add some more information in CMakeLists.txt to correctly find and install nvrtc wrapper simplify prebuilt trtllm libraries name definition do the same name definition stuff for tensorrt_llm_executor_static leverage pkg-config to probe libraries paths and reuse new install structure from cmake fix bad copy/past missing nvinfer linkage direction align all the linker search dependency add missing pkgconfig folder for MPI in Dockerfile correctly setup linking search path for runtime layer fix missing / before tgi lib path adding missing ld_library_path for cuda stubs in Dockerfile update tgi entrypoint commenting out Python part for TensorRT installation refactored docker image move to TensorRT-LLM v0.11.0 make docker linter happy with same capitalization rule fix typo refactor the compute capabilities detection along with num gpus update TensorRT-LLM to latest version update TensorRT install script to latest update build.rs to link to cuda 12.5 add missing dependant libraries for linking clean up a bit install to decoder_attention target add some custom stuff for nccl linkage fix envvar CARGO_CFG_TARGET_ARCH set at runtime vs compile time use std::env::const::ARCH make sure variable live long enough... look for cuda 12.5 add some more basic info in README.md * Rebase. * Fix autodocs. * Let's try to enable trtllm backend. * Ignore backends/v3 by default. * Fixing client. * Fix makefile + autodocs. * Updating the schema thing + redocly. * Fix trtllm lint. * Adding pb files ? * Remove cargo fmt temporarily. * ? * Tmp. * Remove both check + clippy ? * Backporting telemetry. * Backporting 457fb0a1 * Remove PB from git. * Fixing PB with default member backends/client * update TensorRT-LLM to latest version * provided None for api_key * link against libtensorrt_llm and not libtensorrt-llm --------- Co-authored-by: OlivierDehaene <23298448+OlivierDehaene@users.noreply.github.com> Co-authored-by: Morgan Funtowicz <morgan@huggingface.co>
2024-07-31 02:33:10 -06:00
COPY backends backends
COPY launcher launcher
RUN cargo chef prepare --recipe-path recipe.json
FROM chef AS builder
RUN PROTOC_ZIP=protoc-21.12-linux-x86_64.zip && \
curl -OL https://github.com/protocolbuffers/protobuf/releases/download/v21.12/$PROTOC_ZIP && \
unzip -o $PROTOC_ZIP -d /usr/local bin/protoc && \
unzip -o $PROTOC_ZIP -d /usr/local 'include/*' && \
rm -f $PROTOC_ZIP
COPY --from=planner /usr/src/recipe.json recipe.json
RUN cargo chef cook --profile release-opt --recipe-path recipe.json
Internal runner ? (#2023) # What does this PR do? <!-- Congratulations! You've made it this far! You're not quite done yet though. Once merged, your PR is going to appear in the release notes with the title you set, so make sure it's a great title that fully reflects the extent of your awesome contribution. Then, please replace this with a description of the change and which issue is fixed (if applicable). Please also include relevant motivation and context. List any dependencies (if any) that are required for this change. Once you're done, someone will review your PR shortly (see the section "Who can review?" below to tag some potential reviewers). They may suggest changes to make the code even better. If no one reviewed your PR after a week has passed, don't hesitate to post a new comment @-mentioning the same persons---sometimes notifications get lost. --> <!-- Remove if not applicable --> Fixes # (issue) ## Before submitting - [ ] This PR fixes a typo or improves the docs (you can dismiss the other checks if that's the case). - [ ] Did you read the [contributor guideline](https://github.com/huggingface/transformers/blob/main/CONTRIBUTING.md#start-contributing-pull-requests), Pull Request section? - [ ] Was this discussed/approved via a Github issue or the [forum](https://discuss.huggingface.co/)? Please add a link to it if that's the case. - [ ] Did you make sure to update the documentation with your changes? Here are the [documentation guidelines](https://github.com/huggingface/transformers/tree/main/docs), and [here are tips on formatting docstrings](https://github.com/huggingface/transformers/tree/main/docs#writing-source-documentation). - [ ] Did you write any new necessary tests? ## Who can review? Anyone in the community is free to review the PR once the tests have passed. Feel free to tag members/contributors who may be interested in your PR. <!-- Your PR will be replied to more quickly if you can figure out the right person to tag with @ @OlivierDehaene OR @Narsil -->
2024-06-06 10:51:42 -06:00
ARG GIT_SHA
ARG DOCKER_LABEL
COPY Cargo.toml Cargo.toml
COPY rust-toolchain.toml rust-toolchain.toml
COPY proto proto
COPY benchmark benchmark
COPY router router
Rebase TRT-llm (#2331) * wip wip refacto refacto Initial setup for CXX binding to TRTLLM Working FFI call for TGI and TRTLLM backend Remove unused parameters annd force tokenizer name to be set Overall build TRTLLM and deps through CMake build system Enable end to end CMake build First version loading engines and making it ready for inference Remembering to check how we can detect support for chunked context Move to latest TensorRT-LLM version Specify which default log level to use depending on CMake build type make leader executor mode working unconditionally call InitializeBackend on the FFI layer bind to CUDA::nvml to retrieve compute capabilities at runtime updated logic and comment to detect cuda compute capabilities implement the Stream method to send new tokens through a callback use spdlog release 1.14.1 moving forward update trtllm to latest version a96cccafcf6365c128f004f779160951f8c0801c correctly tell cmake to build dependent tensorrt-llm required libraries create cmake install target to put everything relevant in installation folder add auth_token CLI argument to provide hf hub authentification token allow converting huggingface::tokenizers error to TensorRtLlmBackendError use correct include for spdlog include guard to build example in cmakelists working setup of the ffi layer remove fmt import use external fmt lib end to end ffi flow working make sure to track include/ffi.h to trigger rebuild from cargo impl the rust backend which currently cannot move the actual computation in background thread expose shutdown function at ffi layer impl RwLock scenario for TensorRtLllmBackend oops missing c++ backend definitions compute the number of maximum new tokens for each request independently make sure the context is not dropped in the middle of the async decoding. remove unnecessary log add all the necessary plumbery to return the generated content update invalid doc in cpp file correctly forward back the log probabilities remove unneeded scope variable for now refactor Stream impl for Generation to factorise code expose the internal missing start/queue timestamp forward tgi parameters rep/freq penalty add some more validation about grammar not supported define a shared struct to hold the result of a decoding step expose information about potential error happening while decoding remove logging add logging in case of decoding error make sure executor_worker is provided add initial Dockerfile for TRTLLM backend add some more information in CMakeLists.txt to correctly install executorWorker add some more information in CMakeLists.txt to correctly find and install nvrtc wrapper simplify prebuilt trtllm libraries name definition do the same name definition stuff for tensorrt_llm_executor_static leverage pkg-config to probe libraries paths and reuse new install structure from cmake fix bad copy/past missing nvinfer linkage direction align all the linker search dependency add missing pkgconfig folder for MPI in Dockerfile correctly setup linking search path for runtime layer fix missing / before tgi lib path adding missing ld_library_path for cuda stubs in Dockerfile update tgi entrypoint commenting out Python part for TensorRT installation refactored docker image move to TensorRT-LLM v0.11.0 make docker linter happy with same capitalization rule fix typo refactor the compute capabilities detection along with num gpus update TensorRT-LLM to latest version update TensorRT install script to latest update build.rs to link to cuda 12.5 add missing dependant libraries for linking clean up a bit install to decoder_attention target add some custom stuff for nccl linkage fix envvar CARGO_CFG_TARGET_ARCH set at runtime vs compile time use std::env::const::ARCH make sure variable live long enough... look for cuda 12.5 add some more basic info in README.md * Rebase. * Fix autodocs. * Let's try to enable trtllm backend. * Ignore backends/v3 by default. * Fixing client. * Fix makefile + autodocs. * Updating the schema thing + redocly. * Fix trtllm lint. * Adding pb files ? * Remove cargo fmt temporarily. * ? * Tmp. * Remove both check + clippy ? * Backporting telemetry. * Backporting 457fb0a1 * Remove PB from git. * Fixing PB with default member backends/client * update TensorRT-LLM to latest version * provided None for api_key * link against libtensorrt_llm and not libtensorrt-llm --------- Co-authored-by: OlivierDehaene <23298448+OlivierDehaene@users.noreply.github.com> Co-authored-by: Morgan Funtowicz <morgan@huggingface.co>
2024-07-31 02:33:10 -06:00
COPY backends backends
COPY launcher launcher
RUN cargo build --profile release-opt
# Text Generation Inference base image for RoCm
FROM rocm/dev-ubuntu-22.04:6.1.1_hip_update AS base
RUN apt-get update && DEBIAN_FRONTEND=noninteractive apt-get install -y --no-install-recommends \
build-essential \
ca-certificates \
ccache \
curl \
git \
make \
libssl-dev \
g++ \
# Needed to build VLLM & flash.
rocthrust-dev \
hipsparse-dev \
MI300 compatibility (#1764) Adds support for AMD Instinct MI300 in TGI. Most changes are: * Support PyTorch TunableOp to pick the GEMM/GEMV kernels for decoding https://github.com/pytorch/pytorch/tree/main/aten/src/ATen/cuda/tunable. TunableOp is disabled by default, and can be enabled with `PYTORCH_TUNABLEOP_ENABLED=1`. * Update ROCm dockerfile to PyTorch 2.3 (actually patched with changes from https://github.com/pytorch/pytorch/pull/124362) * Support SILU & Linear custom kernels contributed by AMD * Update vLLM paged attention to https://github.com/fxmarty/rocm-vllm/, branching out of a much more recent commit https://github.com/ROCm/vllm/commit/3489ce7936c5de588916ae3047c44c23c0b0c308 * Support FA2 Triton kernel as recommended by AMD. Can be used by specifying `ROCM_USE_FLASH_ATTN_V2_TRITON=1`. * Update dockerfile to ROCm 6.1 By default, TunableOp tuning results are saved in `/data` (e.g. `/data/tunableop_meta-llama-Llama-2-70b-chat-hf_tp1_rank0.csv`) in order to avoid to have to rerun the tuning at each `docker run`. Example: ``` Validator,PT_VERSION,2.3.0 Validator,ROCM_VERSION,6.1.0.0-82-5fabb4c Validator,HIPBLASLT_VERSION,0.7.0-1549b021 Validator,GCN_ARCH_NAME,gfx942:sramecc+:xnack- Validator,ROCBLAS_VERSION,4.1.0-cefa4a9b-dirty GemmTunableOp_Half_TN,tn_8192_7_28672,Gemm_Rocblas_45475,0.132098 GemmTunableOp_Half_TN,tn_10240_4_8192,Gemm_Rocblas_45546,0.0484431 GemmTunableOp_Half_TN,tn_32000_6_8192,Default,0.149546 GemmTunableOp_Half_TN,tn_32000_3_8192,Gemm_Rocblas_45520,0.147119 GemmTunableOp_Half_TN,tn_8192_3_28672,Gemm_Rocblas_45475,0.132645 GemmTunableOp_Half_TN,tn_10240_3_8192,Gemm_Rocblas_45546,0.0482971 GemmTunableOp_Half_TN,tn_57344_5_8192,Gemm_Rocblas_45520,0.255694 GemmTunableOp_Half_TN,tn_10240_7_8192,Gemm_Rocblas_45517,0.0482522 GemmTunableOp_Half_TN,tn_8192_3_8192,Gemm_Rocblas_45546,0.0444671 GemmTunableOp_Half_TN,tn_8192_5_8192,Gemm_Rocblas_45546,0.0445834 GemmTunableOp_Half_TN,tn_57344_7_8192,Gemm_Rocblas_45520,0.25622 GemmTunableOp_Half_TN,tn_8192_2_28672,Gemm_Rocblas_45475,0.132122 GemmTunableOp_Half_TN,tn_8192_4_8192,Gemm_Rocblas_45517,0.0453191 GemmTunableOp_Half_TN,tn_10240_5_8192,Gemm_Rocblas_45517,0.0482514 GemmTunableOp_Half_TN,tn_8192_5_28672,Gemm_Rocblas_45542,0.133914 GemmTunableOp_Half_TN,tn_8192_2_8192,Gemm_Rocblas_45517,0.0446516 GemmTunableOp_Half_TN,tn_8192_1_28672,Gemm_Hipblaslt_TN_10814,0.131953 GemmTunableOp_Half_TN,tn_10240_2_8192,Gemm_Rocblas_45546,0.0481043 GemmTunableOp_Half_TN,tn_32000_4_8192,Gemm_Rocblas_45520,0.147497 GemmTunableOp_Half_TN,tn_8192_6_28672,Gemm_Rocblas_45529,0.134895 GemmTunableOp_Half_TN,tn_57344_2_8192,Gemm_Rocblas_45520,0.254716 GemmTunableOp_Half_TN,tn_57344_4_8192,Gemm_Rocblas_45520,0.255731 GemmTunableOp_Half_TN,tn_10240_6_8192,Gemm_Rocblas_45517,0.0484816 GemmTunableOp_Half_TN,tn_57344_3_8192,Gemm_Rocblas_45520,0.254701 GemmTunableOp_Half_TN,tn_8192_4_28672,Gemm_Rocblas_45475,0.132159 GemmTunableOp_Half_TN,tn_32000_2_8192,Default,0.147524 GemmTunableOp_Half_TN,tn_32000_5_8192,Default,0.147074 GemmTunableOp_Half_TN,tn_8192_6_8192,Gemm_Rocblas_45546,0.0454045 GemmTunableOp_Half_TN,tn_57344_6_8192,Gemm_Rocblas_45520,0.255582 GemmTunableOp_Half_TN,tn_32000_7_8192,Default,0.146705 GemmTunableOp_Half_TN,tn_8192_7_8192,Gemm_Rocblas_45546,0.0445489 ``` --------- Co-authored-by: Mohit Sharma <mohit21sharma.ms@gmail.com>
2024-05-17 07:30:47 -06:00
hipblas-dev \
hipblaslt-dev \
rocblas-dev \
hiprand-dev \
rocrand-dev \
miopen-hip-dev \
hipfft-dev \
hipcub-dev \
hipsolver-dev \
rccl-dev \
cmake \
python3-dev && \
rm -rf /var/lib/apt/lists/*
# Keep in sync with `server/pyproject.toml
ARG MAMBA_VERSION=23.1.0-1
MI300 compatibility (#1764) Adds support for AMD Instinct MI300 in TGI. Most changes are: * Support PyTorch TunableOp to pick the GEMM/GEMV kernels for decoding https://github.com/pytorch/pytorch/tree/main/aten/src/ATen/cuda/tunable. TunableOp is disabled by default, and can be enabled with `PYTORCH_TUNABLEOP_ENABLED=1`. * Update ROCm dockerfile to PyTorch 2.3 (actually patched with changes from https://github.com/pytorch/pytorch/pull/124362) * Support SILU & Linear custom kernels contributed by AMD * Update vLLM paged attention to https://github.com/fxmarty/rocm-vllm/, branching out of a much more recent commit https://github.com/ROCm/vllm/commit/3489ce7936c5de588916ae3047c44c23c0b0c308 * Support FA2 Triton kernel as recommended by AMD. Can be used by specifying `ROCM_USE_FLASH_ATTN_V2_TRITON=1`. * Update dockerfile to ROCm 6.1 By default, TunableOp tuning results are saved in `/data` (e.g. `/data/tunableop_meta-llama-Llama-2-70b-chat-hf_tp1_rank0.csv`) in order to avoid to have to rerun the tuning at each `docker run`. Example: ``` Validator,PT_VERSION,2.3.0 Validator,ROCM_VERSION,6.1.0.0-82-5fabb4c Validator,HIPBLASLT_VERSION,0.7.0-1549b021 Validator,GCN_ARCH_NAME,gfx942:sramecc+:xnack- Validator,ROCBLAS_VERSION,4.1.0-cefa4a9b-dirty GemmTunableOp_Half_TN,tn_8192_7_28672,Gemm_Rocblas_45475,0.132098 GemmTunableOp_Half_TN,tn_10240_4_8192,Gemm_Rocblas_45546,0.0484431 GemmTunableOp_Half_TN,tn_32000_6_8192,Default,0.149546 GemmTunableOp_Half_TN,tn_32000_3_8192,Gemm_Rocblas_45520,0.147119 GemmTunableOp_Half_TN,tn_8192_3_28672,Gemm_Rocblas_45475,0.132645 GemmTunableOp_Half_TN,tn_10240_3_8192,Gemm_Rocblas_45546,0.0482971 GemmTunableOp_Half_TN,tn_57344_5_8192,Gemm_Rocblas_45520,0.255694 GemmTunableOp_Half_TN,tn_10240_7_8192,Gemm_Rocblas_45517,0.0482522 GemmTunableOp_Half_TN,tn_8192_3_8192,Gemm_Rocblas_45546,0.0444671 GemmTunableOp_Half_TN,tn_8192_5_8192,Gemm_Rocblas_45546,0.0445834 GemmTunableOp_Half_TN,tn_57344_7_8192,Gemm_Rocblas_45520,0.25622 GemmTunableOp_Half_TN,tn_8192_2_28672,Gemm_Rocblas_45475,0.132122 GemmTunableOp_Half_TN,tn_8192_4_8192,Gemm_Rocblas_45517,0.0453191 GemmTunableOp_Half_TN,tn_10240_5_8192,Gemm_Rocblas_45517,0.0482514 GemmTunableOp_Half_TN,tn_8192_5_28672,Gemm_Rocblas_45542,0.133914 GemmTunableOp_Half_TN,tn_8192_2_8192,Gemm_Rocblas_45517,0.0446516 GemmTunableOp_Half_TN,tn_8192_1_28672,Gemm_Hipblaslt_TN_10814,0.131953 GemmTunableOp_Half_TN,tn_10240_2_8192,Gemm_Rocblas_45546,0.0481043 GemmTunableOp_Half_TN,tn_32000_4_8192,Gemm_Rocblas_45520,0.147497 GemmTunableOp_Half_TN,tn_8192_6_28672,Gemm_Rocblas_45529,0.134895 GemmTunableOp_Half_TN,tn_57344_2_8192,Gemm_Rocblas_45520,0.254716 GemmTunableOp_Half_TN,tn_57344_4_8192,Gemm_Rocblas_45520,0.255731 GemmTunableOp_Half_TN,tn_10240_6_8192,Gemm_Rocblas_45517,0.0484816 GemmTunableOp_Half_TN,tn_57344_3_8192,Gemm_Rocblas_45520,0.254701 GemmTunableOp_Half_TN,tn_8192_4_28672,Gemm_Rocblas_45475,0.132159 GemmTunableOp_Half_TN,tn_32000_2_8192,Default,0.147524 GemmTunableOp_Half_TN,tn_32000_5_8192,Default,0.147074 GemmTunableOp_Half_TN,tn_8192_6_8192,Gemm_Rocblas_45546,0.0454045 GemmTunableOp_Half_TN,tn_57344_6_8192,Gemm_Rocblas_45520,0.255582 GemmTunableOp_Half_TN,tn_32000_7_8192,Default,0.146705 GemmTunableOp_Half_TN,tn_8192_7_8192,Gemm_Rocblas_45546,0.0445489 ``` --------- Co-authored-by: Mohit Sharma <mohit21sharma.ms@gmail.com>
2024-05-17 07:30:47 -06:00
ARG PYTORCH_VERSION='2.3.0'
ARG ROCM_VERSION='6.0.2'
ARG PYTHON_VERSION='3.10.10'
# Automatically set by buildx
ARG TARGETPLATFORM
ENV PATH /opt/conda/bin:$PATH
# TGI seem to require libssl.so.1.1 instead of libssl.so.3 so we can't use ubuntu 22.04. Ubuntu 20.04 has python==3.8, and TGI requires python>=3.9, hence the need for miniconda.
# Install mamba
# translating Docker's TARGETPLATFORM into mamba arches
RUN case ${TARGETPLATFORM} in \
"linux/arm64") MAMBA_ARCH=aarch64 ;; \
*) MAMBA_ARCH=x86_64 ;; \
esac && \
curl -fsSL -v -o ~/mambaforge.sh -O "https://github.com/conda-forge/miniforge/releases/download/${MAMBA_VERSION}/Mambaforge-${MAMBA_VERSION}-Linux-${MAMBA_ARCH}.sh"
RUN chmod +x ~/mambaforge.sh && \
bash ~/mambaforge.sh -b -p /opt/conda && \
mamba init && \
rm ~/mambaforge.sh
MI300 compatibility (#1764) Adds support for AMD Instinct MI300 in TGI. Most changes are: * Support PyTorch TunableOp to pick the GEMM/GEMV kernels for decoding https://github.com/pytorch/pytorch/tree/main/aten/src/ATen/cuda/tunable. TunableOp is disabled by default, and can be enabled with `PYTORCH_TUNABLEOP_ENABLED=1`. * Update ROCm dockerfile to PyTorch 2.3 (actually patched with changes from https://github.com/pytorch/pytorch/pull/124362) * Support SILU & Linear custom kernels contributed by AMD * Update vLLM paged attention to https://github.com/fxmarty/rocm-vllm/, branching out of a much more recent commit https://github.com/ROCm/vllm/commit/3489ce7936c5de588916ae3047c44c23c0b0c308 * Support FA2 Triton kernel as recommended by AMD. Can be used by specifying `ROCM_USE_FLASH_ATTN_V2_TRITON=1`. * Update dockerfile to ROCm 6.1 By default, TunableOp tuning results are saved in `/data` (e.g. `/data/tunableop_meta-llama-Llama-2-70b-chat-hf_tp1_rank0.csv`) in order to avoid to have to rerun the tuning at each `docker run`. Example: ``` Validator,PT_VERSION,2.3.0 Validator,ROCM_VERSION,6.1.0.0-82-5fabb4c Validator,HIPBLASLT_VERSION,0.7.0-1549b021 Validator,GCN_ARCH_NAME,gfx942:sramecc+:xnack- Validator,ROCBLAS_VERSION,4.1.0-cefa4a9b-dirty GemmTunableOp_Half_TN,tn_8192_7_28672,Gemm_Rocblas_45475,0.132098 GemmTunableOp_Half_TN,tn_10240_4_8192,Gemm_Rocblas_45546,0.0484431 GemmTunableOp_Half_TN,tn_32000_6_8192,Default,0.149546 GemmTunableOp_Half_TN,tn_32000_3_8192,Gemm_Rocblas_45520,0.147119 GemmTunableOp_Half_TN,tn_8192_3_28672,Gemm_Rocblas_45475,0.132645 GemmTunableOp_Half_TN,tn_10240_3_8192,Gemm_Rocblas_45546,0.0482971 GemmTunableOp_Half_TN,tn_57344_5_8192,Gemm_Rocblas_45520,0.255694 GemmTunableOp_Half_TN,tn_10240_7_8192,Gemm_Rocblas_45517,0.0482522 GemmTunableOp_Half_TN,tn_8192_3_8192,Gemm_Rocblas_45546,0.0444671 GemmTunableOp_Half_TN,tn_8192_5_8192,Gemm_Rocblas_45546,0.0445834 GemmTunableOp_Half_TN,tn_57344_7_8192,Gemm_Rocblas_45520,0.25622 GemmTunableOp_Half_TN,tn_8192_2_28672,Gemm_Rocblas_45475,0.132122 GemmTunableOp_Half_TN,tn_8192_4_8192,Gemm_Rocblas_45517,0.0453191 GemmTunableOp_Half_TN,tn_10240_5_8192,Gemm_Rocblas_45517,0.0482514 GemmTunableOp_Half_TN,tn_8192_5_28672,Gemm_Rocblas_45542,0.133914 GemmTunableOp_Half_TN,tn_8192_2_8192,Gemm_Rocblas_45517,0.0446516 GemmTunableOp_Half_TN,tn_8192_1_28672,Gemm_Hipblaslt_TN_10814,0.131953 GemmTunableOp_Half_TN,tn_10240_2_8192,Gemm_Rocblas_45546,0.0481043 GemmTunableOp_Half_TN,tn_32000_4_8192,Gemm_Rocblas_45520,0.147497 GemmTunableOp_Half_TN,tn_8192_6_28672,Gemm_Rocblas_45529,0.134895 GemmTunableOp_Half_TN,tn_57344_2_8192,Gemm_Rocblas_45520,0.254716 GemmTunableOp_Half_TN,tn_57344_4_8192,Gemm_Rocblas_45520,0.255731 GemmTunableOp_Half_TN,tn_10240_6_8192,Gemm_Rocblas_45517,0.0484816 GemmTunableOp_Half_TN,tn_57344_3_8192,Gemm_Rocblas_45520,0.254701 GemmTunableOp_Half_TN,tn_8192_4_28672,Gemm_Rocblas_45475,0.132159 GemmTunableOp_Half_TN,tn_32000_2_8192,Default,0.147524 GemmTunableOp_Half_TN,tn_32000_5_8192,Default,0.147074 GemmTunableOp_Half_TN,tn_8192_6_8192,Gemm_Rocblas_45546,0.0454045 GemmTunableOp_Half_TN,tn_57344_6_8192,Gemm_Rocblas_45520,0.255582 GemmTunableOp_Half_TN,tn_32000_7_8192,Default,0.146705 GemmTunableOp_Half_TN,tn_8192_7_8192,Gemm_Rocblas_45546,0.0445489 ``` --------- Co-authored-by: Mohit Sharma <mohit21sharma.ms@gmail.com>
2024-05-17 07:30:47 -06:00
# Install flash-attention, torch dependencies
RUN pip install numpy einops ninja --no-cache-dir
RUN conda install intel::mkl-static intel::mkl-include
RUN pip uninstall -y triton && \
git clone --depth 1 --single-branch https://github.com/ROCm/triton.git && \
cd triton/python && \
pip install .
RUN git clone --depth 1 --recursive --single-branch --branch 2.3-patched https://github.com/fxmarty/pytorch.git pytorch && cd pytorch && pip install -r requirements.txt --no-cache-dir
ARG _GLIBCXX_USE_CXX11_ABI="1"
ARG CMAKE_PREFIX_PATH="/opt/conda"
ARG PYTORCH_ROCM_ARCH="gfx90a;gfx942"
ARG BUILD_CAFFE2="0" \
BUILD_CAFFE2_OPS="0" \
USE_CUDA="0" \
USE_ROCM="1" \
BUILD_TEST="0" \
USE_FBGEMM="0" \
USE_NNPACK="0" \
USE_QNNPACK="0" \
USE_XNNPACK="0" \
USE_FLASH_ATTENTION="1" \
USE_MEM_EFF_ATTENTION="0"
RUN cd pytorch && python tools/amd_build/build_amd.py && python setup.py install
# Set AS recommended: https://github.com/ROCm/triton/wiki/A-script-to-set-program-execution-environment-in-ROCm
MI300 compatibility (#1764) Adds support for AMD Instinct MI300 in TGI. Most changes are: * Support PyTorch TunableOp to pick the GEMM/GEMV kernels for decoding https://github.com/pytorch/pytorch/tree/main/aten/src/ATen/cuda/tunable. TunableOp is disabled by default, and can be enabled with `PYTORCH_TUNABLEOP_ENABLED=1`. * Update ROCm dockerfile to PyTorch 2.3 (actually patched with changes from https://github.com/pytorch/pytorch/pull/124362) * Support SILU & Linear custom kernels contributed by AMD * Update vLLM paged attention to https://github.com/fxmarty/rocm-vllm/, branching out of a much more recent commit https://github.com/ROCm/vllm/commit/3489ce7936c5de588916ae3047c44c23c0b0c308 * Support FA2 Triton kernel as recommended by AMD. Can be used by specifying `ROCM_USE_FLASH_ATTN_V2_TRITON=1`. * Update dockerfile to ROCm 6.1 By default, TunableOp tuning results are saved in `/data` (e.g. `/data/tunableop_meta-llama-Llama-2-70b-chat-hf_tp1_rank0.csv`) in order to avoid to have to rerun the tuning at each `docker run`. Example: ``` Validator,PT_VERSION,2.3.0 Validator,ROCM_VERSION,6.1.0.0-82-5fabb4c Validator,HIPBLASLT_VERSION,0.7.0-1549b021 Validator,GCN_ARCH_NAME,gfx942:sramecc+:xnack- Validator,ROCBLAS_VERSION,4.1.0-cefa4a9b-dirty GemmTunableOp_Half_TN,tn_8192_7_28672,Gemm_Rocblas_45475,0.132098 GemmTunableOp_Half_TN,tn_10240_4_8192,Gemm_Rocblas_45546,0.0484431 GemmTunableOp_Half_TN,tn_32000_6_8192,Default,0.149546 GemmTunableOp_Half_TN,tn_32000_3_8192,Gemm_Rocblas_45520,0.147119 GemmTunableOp_Half_TN,tn_8192_3_28672,Gemm_Rocblas_45475,0.132645 GemmTunableOp_Half_TN,tn_10240_3_8192,Gemm_Rocblas_45546,0.0482971 GemmTunableOp_Half_TN,tn_57344_5_8192,Gemm_Rocblas_45520,0.255694 GemmTunableOp_Half_TN,tn_10240_7_8192,Gemm_Rocblas_45517,0.0482522 GemmTunableOp_Half_TN,tn_8192_3_8192,Gemm_Rocblas_45546,0.0444671 GemmTunableOp_Half_TN,tn_8192_5_8192,Gemm_Rocblas_45546,0.0445834 GemmTunableOp_Half_TN,tn_57344_7_8192,Gemm_Rocblas_45520,0.25622 GemmTunableOp_Half_TN,tn_8192_2_28672,Gemm_Rocblas_45475,0.132122 GemmTunableOp_Half_TN,tn_8192_4_8192,Gemm_Rocblas_45517,0.0453191 GemmTunableOp_Half_TN,tn_10240_5_8192,Gemm_Rocblas_45517,0.0482514 GemmTunableOp_Half_TN,tn_8192_5_28672,Gemm_Rocblas_45542,0.133914 GemmTunableOp_Half_TN,tn_8192_2_8192,Gemm_Rocblas_45517,0.0446516 GemmTunableOp_Half_TN,tn_8192_1_28672,Gemm_Hipblaslt_TN_10814,0.131953 GemmTunableOp_Half_TN,tn_10240_2_8192,Gemm_Rocblas_45546,0.0481043 GemmTunableOp_Half_TN,tn_32000_4_8192,Gemm_Rocblas_45520,0.147497 GemmTunableOp_Half_TN,tn_8192_6_28672,Gemm_Rocblas_45529,0.134895 GemmTunableOp_Half_TN,tn_57344_2_8192,Gemm_Rocblas_45520,0.254716 GemmTunableOp_Half_TN,tn_57344_4_8192,Gemm_Rocblas_45520,0.255731 GemmTunableOp_Half_TN,tn_10240_6_8192,Gemm_Rocblas_45517,0.0484816 GemmTunableOp_Half_TN,tn_57344_3_8192,Gemm_Rocblas_45520,0.254701 GemmTunableOp_Half_TN,tn_8192_4_28672,Gemm_Rocblas_45475,0.132159 GemmTunableOp_Half_TN,tn_32000_2_8192,Default,0.147524 GemmTunableOp_Half_TN,tn_32000_5_8192,Default,0.147074 GemmTunableOp_Half_TN,tn_8192_6_8192,Gemm_Rocblas_45546,0.0454045 GemmTunableOp_Half_TN,tn_57344_6_8192,Gemm_Rocblas_45520,0.255582 GemmTunableOp_Half_TN,tn_32000_7_8192,Default,0.146705 GemmTunableOp_Half_TN,tn_8192_7_8192,Gemm_Rocblas_45546,0.0445489 ``` --------- Co-authored-by: Mohit Sharma <mohit21sharma.ms@gmail.com>
2024-05-17 07:30:47 -06:00
ENV HIP_FORCE_DEV_KERNARG=1
# On MI250 and MI300, performances for flash with Triton FA are slightly better than CK.
# However, Triton requires a tunning for each prompt length, which is prohibitive.
ENV ROCM_USE_FLASH_ATTN_V2_TRITON=0
FROM base AS kernel-builder
MI300 compatibility (#1764) Adds support for AMD Instinct MI300 in TGI. Most changes are: * Support PyTorch TunableOp to pick the GEMM/GEMV kernels for decoding https://github.com/pytorch/pytorch/tree/main/aten/src/ATen/cuda/tunable. TunableOp is disabled by default, and can be enabled with `PYTORCH_TUNABLEOP_ENABLED=1`. * Update ROCm dockerfile to PyTorch 2.3 (actually patched with changes from https://github.com/pytorch/pytorch/pull/124362) * Support SILU & Linear custom kernels contributed by AMD * Update vLLM paged attention to https://github.com/fxmarty/rocm-vllm/, branching out of a much more recent commit https://github.com/ROCm/vllm/commit/3489ce7936c5de588916ae3047c44c23c0b0c308 * Support FA2 Triton kernel as recommended by AMD. Can be used by specifying `ROCM_USE_FLASH_ATTN_V2_TRITON=1`. * Update dockerfile to ROCm 6.1 By default, TunableOp tuning results are saved in `/data` (e.g. `/data/tunableop_meta-llama-Llama-2-70b-chat-hf_tp1_rank0.csv`) in order to avoid to have to rerun the tuning at each `docker run`. Example: ``` Validator,PT_VERSION,2.3.0 Validator,ROCM_VERSION,6.1.0.0-82-5fabb4c Validator,HIPBLASLT_VERSION,0.7.0-1549b021 Validator,GCN_ARCH_NAME,gfx942:sramecc+:xnack- Validator,ROCBLAS_VERSION,4.1.0-cefa4a9b-dirty GemmTunableOp_Half_TN,tn_8192_7_28672,Gemm_Rocblas_45475,0.132098 GemmTunableOp_Half_TN,tn_10240_4_8192,Gemm_Rocblas_45546,0.0484431 GemmTunableOp_Half_TN,tn_32000_6_8192,Default,0.149546 GemmTunableOp_Half_TN,tn_32000_3_8192,Gemm_Rocblas_45520,0.147119 GemmTunableOp_Half_TN,tn_8192_3_28672,Gemm_Rocblas_45475,0.132645 GemmTunableOp_Half_TN,tn_10240_3_8192,Gemm_Rocblas_45546,0.0482971 GemmTunableOp_Half_TN,tn_57344_5_8192,Gemm_Rocblas_45520,0.255694 GemmTunableOp_Half_TN,tn_10240_7_8192,Gemm_Rocblas_45517,0.0482522 GemmTunableOp_Half_TN,tn_8192_3_8192,Gemm_Rocblas_45546,0.0444671 GemmTunableOp_Half_TN,tn_8192_5_8192,Gemm_Rocblas_45546,0.0445834 GemmTunableOp_Half_TN,tn_57344_7_8192,Gemm_Rocblas_45520,0.25622 GemmTunableOp_Half_TN,tn_8192_2_28672,Gemm_Rocblas_45475,0.132122 GemmTunableOp_Half_TN,tn_8192_4_8192,Gemm_Rocblas_45517,0.0453191 GemmTunableOp_Half_TN,tn_10240_5_8192,Gemm_Rocblas_45517,0.0482514 GemmTunableOp_Half_TN,tn_8192_5_28672,Gemm_Rocblas_45542,0.133914 GemmTunableOp_Half_TN,tn_8192_2_8192,Gemm_Rocblas_45517,0.0446516 GemmTunableOp_Half_TN,tn_8192_1_28672,Gemm_Hipblaslt_TN_10814,0.131953 GemmTunableOp_Half_TN,tn_10240_2_8192,Gemm_Rocblas_45546,0.0481043 GemmTunableOp_Half_TN,tn_32000_4_8192,Gemm_Rocblas_45520,0.147497 GemmTunableOp_Half_TN,tn_8192_6_28672,Gemm_Rocblas_45529,0.134895 GemmTunableOp_Half_TN,tn_57344_2_8192,Gemm_Rocblas_45520,0.254716 GemmTunableOp_Half_TN,tn_57344_4_8192,Gemm_Rocblas_45520,0.255731 GemmTunableOp_Half_TN,tn_10240_6_8192,Gemm_Rocblas_45517,0.0484816 GemmTunableOp_Half_TN,tn_57344_3_8192,Gemm_Rocblas_45520,0.254701 GemmTunableOp_Half_TN,tn_8192_4_28672,Gemm_Rocblas_45475,0.132159 GemmTunableOp_Half_TN,tn_32000_2_8192,Default,0.147524 GemmTunableOp_Half_TN,tn_32000_5_8192,Default,0.147074 GemmTunableOp_Half_TN,tn_8192_6_8192,Gemm_Rocblas_45546,0.0454045 GemmTunableOp_Half_TN,tn_57344_6_8192,Gemm_Rocblas_45520,0.255582 GemmTunableOp_Half_TN,tn_32000_7_8192,Default,0.146705 GemmTunableOp_Half_TN,tn_8192_7_8192,Gemm_Rocblas_45546,0.0445489 ``` --------- Co-authored-by: Mohit Sharma <mohit21sharma.ms@gmail.com>
2024-05-17 07:30:47 -06:00
# # Build vllm kernels
FROM kernel-builder AS vllm-builder
WORKDIR /usr/src
COPY server/Makefile-vllm Makefile
# Build specific version of vllm
RUN make build-vllm-rocm
# Build Flash Attention v2 kernels
FROM kernel-builder AS flash-att-v2-builder
WORKDIR /usr/src
COPY server/Makefile-flash-att-v2 Makefile
# Build specific version of flash attention v2
RUN make build-flash-attention-v2-rocm
# Build Transformers CUDA kernels (gpt-neox and bloom)
FROM kernel-builder AS custom-kernels-builder
WORKDIR /usr/src
COPY server/custom_kernels/ .
MI300 compatibility (#1764) Adds support for AMD Instinct MI300 in TGI. Most changes are: * Support PyTorch TunableOp to pick the GEMM/GEMV kernels for decoding https://github.com/pytorch/pytorch/tree/main/aten/src/ATen/cuda/tunable. TunableOp is disabled by default, and can be enabled with `PYTORCH_TUNABLEOP_ENABLED=1`. * Update ROCm dockerfile to PyTorch 2.3 (actually patched with changes from https://github.com/pytorch/pytorch/pull/124362) * Support SILU & Linear custom kernels contributed by AMD * Update vLLM paged attention to https://github.com/fxmarty/rocm-vllm/, branching out of a much more recent commit https://github.com/ROCm/vllm/commit/3489ce7936c5de588916ae3047c44c23c0b0c308 * Support FA2 Triton kernel as recommended by AMD. Can be used by specifying `ROCM_USE_FLASH_ATTN_V2_TRITON=1`. * Update dockerfile to ROCm 6.1 By default, TunableOp tuning results are saved in `/data` (e.g. `/data/tunableop_meta-llama-Llama-2-70b-chat-hf_tp1_rank0.csv`) in order to avoid to have to rerun the tuning at each `docker run`. Example: ``` Validator,PT_VERSION,2.3.0 Validator,ROCM_VERSION,6.1.0.0-82-5fabb4c Validator,HIPBLASLT_VERSION,0.7.0-1549b021 Validator,GCN_ARCH_NAME,gfx942:sramecc+:xnack- Validator,ROCBLAS_VERSION,4.1.0-cefa4a9b-dirty GemmTunableOp_Half_TN,tn_8192_7_28672,Gemm_Rocblas_45475,0.132098 GemmTunableOp_Half_TN,tn_10240_4_8192,Gemm_Rocblas_45546,0.0484431 GemmTunableOp_Half_TN,tn_32000_6_8192,Default,0.149546 GemmTunableOp_Half_TN,tn_32000_3_8192,Gemm_Rocblas_45520,0.147119 GemmTunableOp_Half_TN,tn_8192_3_28672,Gemm_Rocblas_45475,0.132645 GemmTunableOp_Half_TN,tn_10240_3_8192,Gemm_Rocblas_45546,0.0482971 GemmTunableOp_Half_TN,tn_57344_5_8192,Gemm_Rocblas_45520,0.255694 GemmTunableOp_Half_TN,tn_10240_7_8192,Gemm_Rocblas_45517,0.0482522 GemmTunableOp_Half_TN,tn_8192_3_8192,Gemm_Rocblas_45546,0.0444671 GemmTunableOp_Half_TN,tn_8192_5_8192,Gemm_Rocblas_45546,0.0445834 GemmTunableOp_Half_TN,tn_57344_7_8192,Gemm_Rocblas_45520,0.25622 GemmTunableOp_Half_TN,tn_8192_2_28672,Gemm_Rocblas_45475,0.132122 GemmTunableOp_Half_TN,tn_8192_4_8192,Gemm_Rocblas_45517,0.0453191 GemmTunableOp_Half_TN,tn_10240_5_8192,Gemm_Rocblas_45517,0.0482514 GemmTunableOp_Half_TN,tn_8192_5_28672,Gemm_Rocblas_45542,0.133914 GemmTunableOp_Half_TN,tn_8192_2_8192,Gemm_Rocblas_45517,0.0446516 GemmTunableOp_Half_TN,tn_8192_1_28672,Gemm_Hipblaslt_TN_10814,0.131953 GemmTunableOp_Half_TN,tn_10240_2_8192,Gemm_Rocblas_45546,0.0481043 GemmTunableOp_Half_TN,tn_32000_4_8192,Gemm_Rocblas_45520,0.147497 GemmTunableOp_Half_TN,tn_8192_6_28672,Gemm_Rocblas_45529,0.134895 GemmTunableOp_Half_TN,tn_57344_2_8192,Gemm_Rocblas_45520,0.254716 GemmTunableOp_Half_TN,tn_57344_4_8192,Gemm_Rocblas_45520,0.255731 GemmTunableOp_Half_TN,tn_10240_6_8192,Gemm_Rocblas_45517,0.0484816 GemmTunableOp_Half_TN,tn_57344_3_8192,Gemm_Rocblas_45520,0.254701 GemmTunableOp_Half_TN,tn_8192_4_28672,Gemm_Rocblas_45475,0.132159 GemmTunableOp_Half_TN,tn_32000_2_8192,Default,0.147524 GemmTunableOp_Half_TN,tn_32000_5_8192,Default,0.147074 GemmTunableOp_Half_TN,tn_8192_6_8192,Gemm_Rocblas_45546,0.0454045 GemmTunableOp_Half_TN,tn_57344_6_8192,Gemm_Rocblas_45520,0.255582 GemmTunableOp_Half_TN,tn_32000_7_8192,Default,0.146705 GemmTunableOp_Half_TN,tn_8192_7_8192,Gemm_Rocblas_45546,0.0445489 ``` --------- Co-authored-by: Mohit Sharma <mohit21sharma.ms@gmail.com>
2024-05-17 07:30:47 -06:00
RUN python setup.py build
# Build exllama kernels
FROM kernel-builder AS exllama-kernels-builder
WORKDIR /usr/src
COPY server/exllama_kernels/ .
MI300 compatibility (#1764) Adds support for AMD Instinct MI300 in TGI. Most changes are: * Support PyTorch TunableOp to pick the GEMM/GEMV kernels for decoding https://github.com/pytorch/pytorch/tree/main/aten/src/ATen/cuda/tunable. TunableOp is disabled by default, and can be enabled with `PYTORCH_TUNABLEOP_ENABLED=1`. * Update ROCm dockerfile to PyTorch 2.3 (actually patched with changes from https://github.com/pytorch/pytorch/pull/124362) * Support SILU & Linear custom kernels contributed by AMD * Update vLLM paged attention to https://github.com/fxmarty/rocm-vllm/, branching out of a much more recent commit https://github.com/ROCm/vllm/commit/3489ce7936c5de588916ae3047c44c23c0b0c308 * Support FA2 Triton kernel as recommended by AMD. Can be used by specifying `ROCM_USE_FLASH_ATTN_V2_TRITON=1`. * Update dockerfile to ROCm 6.1 By default, TunableOp tuning results are saved in `/data` (e.g. `/data/tunableop_meta-llama-Llama-2-70b-chat-hf_tp1_rank0.csv`) in order to avoid to have to rerun the tuning at each `docker run`. Example: ``` Validator,PT_VERSION,2.3.0 Validator,ROCM_VERSION,6.1.0.0-82-5fabb4c Validator,HIPBLASLT_VERSION,0.7.0-1549b021 Validator,GCN_ARCH_NAME,gfx942:sramecc+:xnack- Validator,ROCBLAS_VERSION,4.1.0-cefa4a9b-dirty GemmTunableOp_Half_TN,tn_8192_7_28672,Gemm_Rocblas_45475,0.132098 GemmTunableOp_Half_TN,tn_10240_4_8192,Gemm_Rocblas_45546,0.0484431 GemmTunableOp_Half_TN,tn_32000_6_8192,Default,0.149546 GemmTunableOp_Half_TN,tn_32000_3_8192,Gemm_Rocblas_45520,0.147119 GemmTunableOp_Half_TN,tn_8192_3_28672,Gemm_Rocblas_45475,0.132645 GemmTunableOp_Half_TN,tn_10240_3_8192,Gemm_Rocblas_45546,0.0482971 GemmTunableOp_Half_TN,tn_57344_5_8192,Gemm_Rocblas_45520,0.255694 GemmTunableOp_Half_TN,tn_10240_7_8192,Gemm_Rocblas_45517,0.0482522 GemmTunableOp_Half_TN,tn_8192_3_8192,Gemm_Rocblas_45546,0.0444671 GemmTunableOp_Half_TN,tn_8192_5_8192,Gemm_Rocblas_45546,0.0445834 GemmTunableOp_Half_TN,tn_57344_7_8192,Gemm_Rocblas_45520,0.25622 GemmTunableOp_Half_TN,tn_8192_2_28672,Gemm_Rocblas_45475,0.132122 GemmTunableOp_Half_TN,tn_8192_4_8192,Gemm_Rocblas_45517,0.0453191 GemmTunableOp_Half_TN,tn_10240_5_8192,Gemm_Rocblas_45517,0.0482514 GemmTunableOp_Half_TN,tn_8192_5_28672,Gemm_Rocblas_45542,0.133914 GemmTunableOp_Half_TN,tn_8192_2_8192,Gemm_Rocblas_45517,0.0446516 GemmTunableOp_Half_TN,tn_8192_1_28672,Gemm_Hipblaslt_TN_10814,0.131953 GemmTunableOp_Half_TN,tn_10240_2_8192,Gemm_Rocblas_45546,0.0481043 GemmTunableOp_Half_TN,tn_32000_4_8192,Gemm_Rocblas_45520,0.147497 GemmTunableOp_Half_TN,tn_8192_6_28672,Gemm_Rocblas_45529,0.134895 GemmTunableOp_Half_TN,tn_57344_2_8192,Gemm_Rocblas_45520,0.254716 GemmTunableOp_Half_TN,tn_57344_4_8192,Gemm_Rocblas_45520,0.255731 GemmTunableOp_Half_TN,tn_10240_6_8192,Gemm_Rocblas_45517,0.0484816 GemmTunableOp_Half_TN,tn_57344_3_8192,Gemm_Rocblas_45520,0.254701 GemmTunableOp_Half_TN,tn_8192_4_28672,Gemm_Rocblas_45475,0.132159 GemmTunableOp_Half_TN,tn_32000_2_8192,Default,0.147524 GemmTunableOp_Half_TN,tn_32000_5_8192,Default,0.147074 GemmTunableOp_Half_TN,tn_8192_6_8192,Gemm_Rocblas_45546,0.0454045 GemmTunableOp_Half_TN,tn_57344_6_8192,Gemm_Rocblas_45520,0.255582 GemmTunableOp_Half_TN,tn_32000_7_8192,Default,0.146705 GemmTunableOp_Half_TN,tn_8192_7_8192,Gemm_Rocblas_45546,0.0445489 ``` --------- Co-authored-by: Mohit Sharma <mohit21sharma.ms@gmail.com>
2024-05-17 07:30:47 -06:00
RUN python setup.py build
# Build exllama v2 kernels
FROM kernel-builder AS exllamav2-kernels-builder
WORKDIR /usr/src
COPY server/exllamav2_kernels/ .
MI300 compatibility (#1764) Adds support for AMD Instinct MI300 in TGI. Most changes are: * Support PyTorch TunableOp to pick the GEMM/GEMV kernels for decoding https://github.com/pytorch/pytorch/tree/main/aten/src/ATen/cuda/tunable. TunableOp is disabled by default, and can be enabled with `PYTORCH_TUNABLEOP_ENABLED=1`. * Update ROCm dockerfile to PyTorch 2.3 (actually patched with changes from https://github.com/pytorch/pytorch/pull/124362) * Support SILU & Linear custom kernels contributed by AMD * Update vLLM paged attention to https://github.com/fxmarty/rocm-vllm/, branching out of a much more recent commit https://github.com/ROCm/vllm/commit/3489ce7936c5de588916ae3047c44c23c0b0c308 * Support FA2 Triton kernel as recommended by AMD. Can be used by specifying `ROCM_USE_FLASH_ATTN_V2_TRITON=1`. * Update dockerfile to ROCm 6.1 By default, TunableOp tuning results are saved in `/data` (e.g. `/data/tunableop_meta-llama-Llama-2-70b-chat-hf_tp1_rank0.csv`) in order to avoid to have to rerun the tuning at each `docker run`. Example: ``` Validator,PT_VERSION,2.3.0 Validator,ROCM_VERSION,6.1.0.0-82-5fabb4c Validator,HIPBLASLT_VERSION,0.7.0-1549b021 Validator,GCN_ARCH_NAME,gfx942:sramecc+:xnack- Validator,ROCBLAS_VERSION,4.1.0-cefa4a9b-dirty GemmTunableOp_Half_TN,tn_8192_7_28672,Gemm_Rocblas_45475,0.132098 GemmTunableOp_Half_TN,tn_10240_4_8192,Gemm_Rocblas_45546,0.0484431 GemmTunableOp_Half_TN,tn_32000_6_8192,Default,0.149546 GemmTunableOp_Half_TN,tn_32000_3_8192,Gemm_Rocblas_45520,0.147119 GemmTunableOp_Half_TN,tn_8192_3_28672,Gemm_Rocblas_45475,0.132645 GemmTunableOp_Half_TN,tn_10240_3_8192,Gemm_Rocblas_45546,0.0482971 GemmTunableOp_Half_TN,tn_57344_5_8192,Gemm_Rocblas_45520,0.255694 GemmTunableOp_Half_TN,tn_10240_7_8192,Gemm_Rocblas_45517,0.0482522 GemmTunableOp_Half_TN,tn_8192_3_8192,Gemm_Rocblas_45546,0.0444671 GemmTunableOp_Half_TN,tn_8192_5_8192,Gemm_Rocblas_45546,0.0445834 GemmTunableOp_Half_TN,tn_57344_7_8192,Gemm_Rocblas_45520,0.25622 GemmTunableOp_Half_TN,tn_8192_2_28672,Gemm_Rocblas_45475,0.132122 GemmTunableOp_Half_TN,tn_8192_4_8192,Gemm_Rocblas_45517,0.0453191 GemmTunableOp_Half_TN,tn_10240_5_8192,Gemm_Rocblas_45517,0.0482514 GemmTunableOp_Half_TN,tn_8192_5_28672,Gemm_Rocblas_45542,0.133914 GemmTunableOp_Half_TN,tn_8192_2_8192,Gemm_Rocblas_45517,0.0446516 GemmTunableOp_Half_TN,tn_8192_1_28672,Gemm_Hipblaslt_TN_10814,0.131953 GemmTunableOp_Half_TN,tn_10240_2_8192,Gemm_Rocblas_45546,0.0481043 GemmTunableOp_Half_TN,tn_32000_4_8192,Gemm_Rocblas_45520,0.147497 GemmTunableOp_Half_TN,tn_8192_6_28672,Gemm_Rocblas_45529,0.134895 GemmTunableOp_Half_TN,tn_57344_2_8192,Gemm_Rocblas_45520,0.254716 GemmTunableOp_Half_TN,tn_57344_4_8192,Gemm_Rocblas_45520,0.255731 GemmTunableOp_Half_TN,tn_10240_6_8192,Gemm_Rocblas_45517,0.0484816 GemmTunableOp_Half_TN,tn_57344_3_8192,Gemm_Rocblas_45520,0.254701 GemmTunableOp_Half_TN,tn_8192_4_28672,Gemm_Rocblas_45475,0.132159 GemmTunableOp_Half_TN,tn_32000_2_8192,Default,0.147524 GemmTunableOp_Half_TN,tn_32000_5_8192,Default,0.147074 GemmTunableOp_Half_TN,tn_8192_6_8192,Gemm_Rocblas_45546,0.0454045 GemmTunableOp_Half_TN,tn_57344_6_8192,Gemm_Rocblas_45520,0.255582 GemmTunableOp_Half_TN,tn_32000_7_8192,Default,0.146705 GemmTunableOp_Half_TN,tn_8192_7_8192,Gemm_Rocblas_45546,0.0445489 ``` --------- Co-authored-by: Mohit Sharma <mohit21sharma.ms@gmail.com>
2024-05-17 07:30:47 -06:00
RUN python setup.py build
FROM base AS base-copy
# Text Generation Inference base env
ENV HF_HOME=/data \
HF_HUB_ENABLE_HF_TRANSFER=1 \
PORT=80
# Copy builds artifacts from vllm builder
COPY --from=vllm-builder /usr/src/vllm/build/lib.linux-x86_64-cpython-310 /opt/conda/lib/python3.10/site-packages
# Copy build artifacts from flash attention v2 builder
COPY --from=flash-att-v2-builder /usr/src/flash-attention-v2/build/lib.linux-x86_64-cpython-310 /opt/conda/lib/python3.10/site-packages
# Copy build artifacts from custom kernels builder
COPY --from=custom-kernels-builder /usr/src/build/lib.linux-x86_64-cpython-310 /opt/conda/lib/python3.10/site-packages
# Copy build artifacts from exllama kernels builder
COPY --from=exllama-kernels-builder /usr/src/build/lib.linux-x86_64-cpython-310 /opt/conda/lib/python3.10/site-packages
# Copy build artifacts from exllamav2 kernels builder
COPY --from=exllamav2-kernels-builder /usr/src/build/lib.linux-x86_64-cpython-310 /opt/conda/lib/python3.10/site-packages
# Install server
COPY proto proto
COPY server server
COPY server/Makefile server/Makefile
RUN cd server && \
make gen-server && \
pip install -r requirements_rocm.txt && \
2024-02-16 09:50:57 -07:00
pip install ".[accelerate, peft, outlines]" --no-cache-dir
# Install benchmarker
COPY --from=builder /usr/src/target/release-opt/text-generation-benchmark /usr/local/bin/text-generation-benchmark
# Install router
COPY --from=builder /usr/src/target/release-opt/text-generation-router /usr/local/bin/text-generation-router
# Install launcher
COPY --from=builder /usr/src/target/release-opt/text-generation-launcher /usr/local/bin/text-generation-launcher
# AWS Sagemaker compatible image
FROM base AS sagemaker
MI300 compatibility (#1764) Adds support for AMD Instinct MI300 in TGI. Most changes are: * Support PyTorch TunableOp to pick the GEMM/GEMV kernels for decoding https://github.com/pytorch/pytorch/tree/main/aten/src/ATen/cuda/tunable. TunableOp is disabled by default, and can be enabled with `PYTORCH_TUNABLEOP_ENABLED=1`. * Update ROCm dockerfile to PyTorch 2.3 (actually patched with changes from https://github.com/pytorch/pytorch/pull/124362) * Support SILU & Linear custom kernels contributed by AMD * Update vLLM paged attention to https://github.com/fxmarty/rocm-vllm/, branching out of a much more recent commit https://github.com/ROCm/vllm/commit/3489ce7936c5de588916ae3047c44c23c0b0c308 * Support FA2 Triton kernel as recommended by AMD. Can be used by specifying `ROCM_USE_FLASH_ATTN_V2_TRITON=1`. * Update dockerfile to ROCm 6.1 By default, TunableOp tuning results are saved in `/data` (e.g. `/data/tunableop_meta-llama-Llama-2-70b-chat-hf_tp1_rank0.csv`) in order to avoid to have to rerun the tuning at each `docker run`. Example: ``` Validator,PT_VERSION,2.3.0 Validator,ROCM_VERSION,6.1.0.0-82-5fabb4c Validator,HIPBLASLT_VERSION,0.7.0-1549b021 Validator,GCN_ARCH_NAME,gfx942:sramecc+:xnack- Validator,ROCBLAS_VERSION,4.1.0-cefa4a9b-dirty GemmTunableOp_Half_TN,tn_8192_7_28672,Gemm_Rocblas_45475,0.132098 GemmTunableOp_Half_TN,tn_10240_4_8192,Gemm_Rocblas_45546,0.0484431 GemmTunableOp_Half_TN,tn_32000_6_8192,Default,0.149546 GemmTunableOp_Half_TN,tn_32000_3_8192,Gemm_Rocblas_45520,0.147119 GemmTunableOp_Half_TN,tn_8192_3_28672,Gemm_Rocblas_45475,0.132645 GemmTunableOp_Half_TN,tn_10240_3_8192,Gemm_Rocblas_45546,0.0482971 GemmTunableOp_Half_TN,tn_57344_5_8192,Gemm_Rocblas_45520,0.255694 GemmTunableOp_Half_TN,tn_10240_7_8192,Gemm_Rocblas_45517,0.0482522 GemmTunableOp_Half_TN,tn_8192_3_8192,Gemm_Rocblas_45546,0.0444671 GemmTunableOp_Half_TN,tn_8192_5_8192,Gemm_Rocblas_45546,0.0445834 GemmTunableOp_Half_TN,tn_57344_7_8192,Gemm_Rocblas_45520,0.25622 GemmTunableOp_Half_TN,tn_8192_2_28672,Gemm_Rocblas_45475,0.132122 GemmTunableOp_Half_TN,tn_8192_4_8192,Gemm_Rocblas_45517,0.0453191 GemmTunableOp_Half_TN,tn_10240_5_8192,Gemm_Rocblas_45517,0.0482514 GemmTunableOp_Half_TN,tn_8192_5_28672,Gemm_Rocblas_45542,0.133914 GemmTunableOp_Half_TN,tn_8192_2_8192,Gemm_Rocblas_45517,0.0446516 GemmTunableOp_Half_TN,tn_8192_1_28672,Gemm_Hipblaslt_TN_10814,0.131953 GemmTunableOp_Half_TN,tn_10240_2_8192,Gemm_Rocblas_45546,0.0481043 GemmTunableOp_Half_TN,tn_32000_4_8192,Gemm_Rocblas_45520,0.147497 GemmTunableOp_Half_TN,tn_8192_6_28672,Gemm_Rocblas_45529,0.134895 GemmTunableOp_Half_TN,tn_57344_2_8192,Gemm_Rocblas_45520,0.254716 GemmTunableOp_Half_TN,tn_57344_4_8192,Gemm_Rocblas_45520,0.255731 GemmTunableOp_Half_TN,tn_10240_6_8192,Gemm_Rocblas_45517,0.0484816 GemmTunableOp_Half_TN,tn_57344_3_8192,Gemm_Rocblas_45520,0.254701 GemmTunableOp_Half_TN,tn_8192_4_28672,Gemm_Rocblas_45475,0.132159 GemmTunableOp_Half_TN,tn_32000_2_8192,Default,0.147524 GemmTunableOp_Half_TN,tn_32000_5_8192,Default,0.147074 GemmTunableOp_Half_TN,tn_8192_6_8192,Gemm_Rocblas_45546,0.0454045 GemmTunableOp_Half_TN,tn_57344_6_8192,Gemm_Rocblas_45520,0.255582 GemmTunableOp_Half_TN,tn_32000_7_8192,Default,0.146705 GemmTunableOp_Half_TN,tn_8192_7_8192,Gemm_Rocblas_45546,0.0445489 ``` --------- Co-authored-by: Mohit Sharma <mohit21sharma.ms@gmail.com>
2024-05-17 07:30:47 -06:00
COPY sagemaker-entrypoint.sh entrypoint.sh
RUN chmod +x entrypoint.sh
ENTRYPOINT ["./entrypoint.sh"]
# Final image
FROM base-copy
MI300 compatibility (#1764) Adds support for AMD Instinct MI300 in TGI. Most changes are: * Support PyTorch TunableOp to pick the GEMM/GEMV kernels for decoding https://github.com/pytorch/pytorch/tree/main/aten/src/ATen/cuda/tunable. TunableOp is disabled by default, and can be enabled with `PYTORCH_TUNABLEOP_ENABLED=1`. * Update ROCm dockerfile to PyTorch 2.3 (actually patched with changes from https://github.com/pytorch/pytorch/pull/124362) * Support SILU & Linear custom kernels contributed by AMD * Update vLLM paged attention to https://github.com/fxmarty/rocm-vllm/, branching out of a much more recent commit https://github.com/ROCm/vllm/commit/3489ce7936c5de588916ae3047c44c23c0b0c308 * Support FA2 Triton kernel as recommended by AMD. Can be used by specifying `ROCM_USE_FLASH_ATTN_V2_TRITON=1`. * Update dockerfile to ROCm 6.1 By default, TunableOp tuning results are saved in `/data` (e.g. `/data/tunableop_meta-llama-Llama-2-70b-chat-hf_tp1_rank0.csv`) in order to avoid to have to rerun the tuning at each `docker run`. Example: ``` Validator,PT_VERSION,2.3.0 Validator,ROCM_VERSION,6.1.0.0-82-5fabb4c Validator,HIPBLASLT_VERSION,0.7.0-1549b021 Validator,GCN_ARCH_NAME,gfx942:sramecc+:xnack- Validator,ROCBLAS_VERSION,4.1.0-cefa4a9b-dirty GemmTunableOp_Half_TN,tn_8192_7_28672,Gemm_Rocblas_45475,0.132098 GemmTunableOp_Half_TN,tn_10240_4_8192,Gemm_Rocblas_45546,0.0484431 GemmTunableOp_Half_TN,tn_32000_6_8192,Default,0.149546 GemmTunableOp_Half_TN,tn_32000_3_8192,Gemm_Rocblas_45520,0.147119 GemmTunableOp_Half_TN,tn_8192_3_28672,Gemm_Rocblas_45475,0.132645 GemmTunableOp_Half_TN,tn_10240_3_8192,Gemm_Rocblas_45546,0.0482971 GemmTunableOp_Half_TN,tn_57344_5_8192,Gemm_Rocblas_45520,0.255694 GemmTunableOp_Half_TN,tn_10240_7_8192,Gemm_Rocblas_45517,0.0482522 GemmTunableOp_Half_TN,tn_8192_3_8192,Gemm_Rocblas_45546,0.0444671 GemmTunableOp_Half_TN,tn_8192_5_8192,Gemm_Rocblas_45546,0.0445834 GemmTunableOp_Half_TN,tn_57344_7_8192,Gemm_Rocblas_45520,0.25622 GemmTunableOp_Half_TN,tn_8192_2_28672,Gemm_Rocblas_45475,0.132122 GemmTunableOp_Half_TN,tn_8192_4_8192,Gemm_Rocblas_45517,0.0453191 GemmTunableOp_Half_TN,tn_10240_5_8192,Gemm_Rocblas_45517,0.0482514 GemmTunableOp_Half_TN,tn_8192_5_28672,Gemm_Rocblas_45542,0.133914 GemmTunableOp_Half_TN,tn_8192_2_8192,Gemm_Rocblas_45517,0.0446516 GemmTunableOp_Half_TN,tn_8192_1_28672,Gemm_Hipblaslt_TN_10814,0.131953 GemmTunableOp_Half_TN,tn_10240_2_8192,Gemm_Rocblas_45546,0.0481043 GemmTunableOp_Half_TN,tn_32000_4_8192,Gemm_Rocblas_45520,0.147497 GemmTunableOp_Half_TN,tn_8192_6_28672,Gemm_Rocblas_45529,0.134895 GemmTunableOp_Half_TN,tn_57344_2_8192,Gemm_Rocblas_45520,0.254716 GemmTunableOp_Half_TN,tn_57344_4_8192,Gemm_Rocblas_45520,0.255731 GemmTunableOp_Half_TN,tn_10240_6_8192,Gemm_Rocblas_45517,0.0484816 GemmTunableOp_Half_TN,tn_57344_3_8192,Gemm_Rocblas_45520,0.254701 GemmTunableOp_Half_TN,tn_8192_4_28672,Gemm_Rocblas_45475,0.132159 GemmTunableOp_Half_TN,tn_32000_2_8192,Default,0.147524 GemmTunableOp_Half_TN,tn_32000_5_8192,Default,0.147074 GemmTunableOp_Half_TN,tn_8192_6_8192,Gemm_Rocblas_45546,0.0454045 GemmTunableOp_Half_TN,tn_57344_6_8192,Gemm_Rocblas_45520,0.255582 GemmTunableOp_Half_TN,tn_32000_7_8192,Default,0.146705 GemmTunableOp_Half_TN,tn_8192_7_8192,Gemm_Rocblas_45546,0.0445489 ``` --------- Co-authored-by: Mohit Sharma <mohit21sharma.ms@gmail.com>
2024-05-17 07:30:47 -06:00
COPY ./tgi-entrypoint.sh /tgi-entrypoint.sh
RUN chmod +x /tgi-entrypoint.sh
ENTRYPOINT ["/tgi-entrypoint.sh"]
CMD ["--json-output"]