huggingface/text-generation-inference
5
0
Fork 0
mirror of https://github.com/huggingface/text-generation-inference.git synced 2025-06-08 18:32:06 +00:00
Code Issues Packages Projects Releases Wiki Activity

(misc: build for sm_{75,80,86,89,90} by default

Browse Source
This commit is contained in:
Morgan Funtowicz 2024-10-10 12:40:32 +00:00 committed by Morgan Funtowicz
parent 544c9d9dba
commit 188e4dc64f
1 changed files with 9 additions and 3 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

12
backends/trtllm/build.rs
View File

@ -36,7 +36,7 @@ fn build_backend(is_debug: bool, opt_level: &str, out_dir: &PathBuf) -> (PathBuf
// Build the backend implementation through CMake
let install_path = INSTALL_PREFIX.unwrap_or("/usr/local/tgi");
let tensorrt_path = TENSORRT_ROOT_DIR.unwrap_or("/usr/local/tensorrt");
let cuda_arch_list = CUDA_ARCH_LIST.unwrap_or("90-real"); // Hopper by default
let cuda_arch_list = CUDA_ARCH_LIST.unwrap_or("75-real;80-real;86-real;89-real;90-real");
let mut install_path = PathBuf::from(install_path);
if !install_path.is_absolute() {
@ -81,7 +81,12 @@ fn build_backend(is_debug: bool, opt_level: &str, out_dir: &PathBuf) -> (PathBuf
(PathBuf::from(install_path), deps_folder)
}
fn build_ffi_layer(deps_folder: &PathBuf) {
fn build_ffi_layer(deps_folder: &PathBuf, is_debug: bool) {
let ndebug = match is_debug {
true => "1",
false => "0"
};
CFG.include_prefix = "backends/trtllm";
cxx_build::bridge("src/lib.rs")
.static_flag(true)
@ -93,6 +98,7 @@ fn build_ffi_layer(deps_folder: &PathBuf) {
.include("/usr/local/tensorrt/include")
.file("src/ffi.cpp")
.std("c++20")
.define("NDEBUG", ndebug)
.compile("tgi_trtllm_backend");
println!("cargo:rerun-if-changed=CMakeLists.txt");
@ -119,7 +125,7 @@ fn main() {
let (_backend_path, deps_folder) = build_backend(is_debug, opt_level, &out_dir);
// Build the FFI layer calling the backend above
build_ffi_layer(&deps_folder);
build_ffi_layer(&deps_folder, is_debug);
// Emit linkage search path
probe!("ompi", MPI_REQUIRED_VERSION);