huggingface/text-generation-inference
5
0
Fork 0
mirror of https://github.com/huggingface/text-generation-inference.git synced 2025-06-19 15:52:08 +00:00
Code Issues Packages Projects Releases Wiki Activity

backend(trtllm): link against decoder_attention_{0|1}

Browse Source
This commit is contained in:
Morgan Funtowicz 2025-02-05 16:15:31 +01:00
parent 11c9acab42
commit d299b52cb5
2 changed files with 5 additions and 2 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

2
backends/trtllm/CMakeLists.txt
View File

@ -59,6 +59,8 @@ target_link_libraries(tgi_trtllm_backend_impl PRIVATE tensorrt_llm nvinfer_plugi
# This install all the artifacts in CMAKE_INSTALL_PREFIX under include/ lib/ bin/ to make easy to link / find it back
install(TARGETS tgi_trtllm_backend_impl)
#install(TARGETS cutlass_src fb_gemm_src fpA_intB_gemm_src gemm_swiglu_sm90_src kernels_src)
install(TARGETS decoder_attention_0 decoder_attention_1)
install(TARGETS tensorrt_llm nvinfer_plugin_tensorrt_llm decoder_attention_src executorWorker)
install(FILES ${TRTLLM_NVRTC_WRAPPER_LIBRARY_PATH} TYPE LIB)
if (NOT ${TGI_TRTLLM_BACKEND_DEBUG})

5
backends/trtllm/build.rs
View File

@ -25,11 +25,12 @@ const IS_GHA_BUILD: LazyLock<bool> = LazyLock::new(|| {
// Dependencies
const BACKEND_DEPS: &str = "tgi_trtllm_backend_impl";
const CUDA_TRANSITIVE_DEPS: [&str; 4] = ["cuda", "cudart", "cublas", "nvidia-ml"];
const TENSORRT_LLM_TRANSITIVE_DEPS: [(&str, &str); 4] = [
const TENSORRT_LLM_TRANSITIVE_DEPS: [(&str, &str); 5] = [
("dylib", "tensorrt_llm"),
("dylib", "tensorrt_llm_nvrtc_wrapper"),
("dylib", "nvinfer_plugin_tensorrt_llm"),
("dylib", "decoder_attention"),
("dylib", "decoder_attention_0"),
("dylib", "decoder_attention_1"),
];
macro_rules! probe {