huggingface/text-generation-inference
5
0
Fork 0
mirror of https://github.com/huggingface/text-generation-inference.git synced 2025-04-22 23:42:06 +00:00
Code Issues Packages Projects Releases Wiki Activity

(misc) move to latest trtllm

Browse Source
This commit is contained in:
Morgan Funtowicz 2024-09-25 10:08:45 +00:00 committed by Morgan Funtowicz
parent 507ff66692
commit 213acc6e34
2 changed files with 7 additions and 6 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

2
backends/trtllm/cmake/trtllm.cmake
View File

@ -23,7 +23,7 @@ endif ()
fetchcontent_declare( fetchcontent_declare(
trtllm trtllm
GIT_REPOSITORY https://github.com/NVIDIA/TensorRT-LLM.git GIT_REPOSITORY https://github.com/NVIDIA/TensorRT-LLM.git
GIT_TAG a681853d3803ee5893307e812530b5e7004bb6e1 GIT_TAG 32ed92e4491baf2d54682a21d247e1948cca996e
GIT_SHALLOW FALSE GIT_SHALLOW FALSE
) )
fetchcontent_makeavailable(trtllm) fetchcontent_makeavailable(trtllm)

11
backends/trtllm/src/looper.rs
View File

@ -8,18 +8,18 @@ use hashbrown::HashMap;
use log::warn; use log::warn;
use tokenizers::{Encoding, Tokenizer}; use tokenizers::{Encoding, Tokenizer};
use tokio::sync::mpsc::{unbounded_channel, UnboundedReceiver, UnboundedSender}; use tokio::sync::mpsc::{unbounded_channel, UnboundedReceiver, UnboundedSender};
use tokio::task::{JoinHandle, spawn_blocking}; use tokio::task::{spawn_blocking, JoinHandle};
use tokio::time::Instant; use tokio::time::Instant;
use tokio_stream::wrappers::UnboundedReceiverStream; use tokio_stream::wrappers::UnboundedReceiverStream;
use tracing::{debug, error}; use tracing::{debug, error, info};
use text_generation_router::{FinishReason, Token};
use text_generation_router::infer::{Backend, GeneratedText, InferError, InferStreamResponse};
use text_generation_router::infer::InferError::{GenerationError, ValidationError}; use text_generation_router::infer::InferError::{GenerationError, ValidationError};
use text_generation_router::validation::{Chunk, ValidGenerateRequest}; use text_generation_router::infer::{Backend, GeneratedText, InferError, InferStreamResponse};
use text_generation_router::validation::ValidationError::{ use text_generation_router::validation::ValidationError::{
EmptyInput, Grammar, TopNTokensDisabled, UnsupportedModality, EmptyInput, Grammar, TopNTokensDisabled, UnsupportedModality,
}; };
use text_generation_router::validation::{Chunk, ValidGenerateRequest};
use text_generation_router::{FinishReason, Token};
use crate::errors::TensorRtLlmBackendError; use crate::errors::TensorRtLlmBackendError;
use crate::ffi::{create_tensorrt_llm_backend, GenerationStep, TensorRtLlmBackendImpl}; use crate::ffi::{create_tensorrt_llm_backend, GenerationStep, TensorRtLlmBackendImpl};
@ -128,6 +128,7 @@ fn executor_status_looper(
} }
} }
// info!("Num response ready: {}", backend.num_responses_ready());
if backend.num_responses_ready() > 0 { if backend.num_responses_ready() > 0 {
match backend.pin_mut().pull_tokens() { match backend.pin_mut().pull_tokens() {
Ok(responses) => { Ok(responses) => {