huggingface/text-generation-inference
5
0
Fork 0
mirror of https://github.com/huggingface/text-generation-inference.git synced 2025-05-01 23:12:09 +00:00
Code Issues Packages Projects Releases Wiki Activity

feat(backend): remove constexpr from par

Browse Source
This commit is contained in:
Morgan Funtowicz 2024-12-03 16:46:59 +01:00
parent ad3ed0d1a1
commit 881527a544
2 changed files with 2 additions and 2 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

2
backends/trtllm/csrc/backend.cpp
View File

@ -7,7 +7,7 @@
#include "hardware.hpp" #include "hardware.hpp"
namespace huggingface::tgi::backends::trtllm { namespace huggingface::tgi::backends::trtllm {
constexpr tle::ParallelConfig backend_workspace_t::parallel_config() const { tle::ParallelConfig backend_workspace_t::parallel_config() const {
// Single engine (TP = PP = 1) -> using leader mode (no MPI involved) // Single engine (TP = PP = 1) -> using leader mode (no MPI involved)
const auto world_size = config_["/pretrained_config/mapping/world_size"_json_pointer].get<size_t>(); const auto world_size = config_["/pretrained_config/mapping/world_size"_json_pointer].get<size_t>();

2
backends/trtllm/csrc/backend.hpp
View File

@ -127,7 +127,7 @@ namespace huggingface::tgi::backends::trtllm {
* to initialize `tensorrt_llm::executor::Executor` with multi-instance communication information * to initialize `tensorrt_llm::executor::Executor` with multi-instance communication information
* @return `tensorrt_llm::executor::ParallelConfig` instance * @return `tensorrt_llm::executor::ParallelConfig` instance
*/ */
[[nodiscard]] constexpr tle::ParallelConfig parallel_config() const; [[nodiscard]] tle::ParallelConfig parallel_config() const;
/** /**
* Factory method returning new `tensorrt_llm::executor::ExecutorConfig` instance used * Factory method returning new `tensorrt_llm::executor::ExecutorConfig` instance used