fix: update lints after rebase

.devcontainer/Dockerfile_trtllm

@@ -72,4 +72,4 @@ RUN cargo install cargo-chef
 COPY --from=trt-builder /usr/local/tensorrt /usr/local/tensorrt
 COPY --from=mpi-builder /usr/local/mpi /usr/local/mpi
 
-ENV MPI_HOME=/usr/local/mpi
+ENV MPI_HOME=/usr/local/mpi

backends/trtllm/build.rs

@@ -44,7 +44,8 @@ fn build_backend(is_debug: bool, opt_level: &str, out_dir: &PathBuf) -> (PathBuf
     }
 
     let mut config = cmake::Config::new(".");
-    config.uses_cxx11()
+    config
+        .uses_cxx11()
         .generator("Ninja")
         .profile(match is_debug {
             true => "Debug",
@@ -57,12 +58,12 @@ fn build_backend(is_debug: bool, opt_level: &str, out_dir: &PathBuf) -> (PathBuf
         .define("TGI_TRTLLM_BACKEND_TARGET_CUDA_ARCH_LIST", cuda_arch_list)
         .define("TGI_TRTLLM_BACKEND_TRT_ROOT", tensorrt_path);
 
-        // Allow to override which Python to use ...
-        if let Some(python3) = option_env!("Python3_EXECUTABLE") {
-            config.define("Python3_EXECUTABLE", python3);
-        }
+    // Allow to override which Python to use ...
+    if let Some(python3) = option_env!("Python3_EXECUTABLE") {
+        config.define("Python3_EXECUTABLE", python3);
+    }
 
-        config.build();
+    config.build();
 
     // Additional transitive CMake dependencies
     let deps_folder = out_dir.join("build").join("_deps");

backends/trtllm/csrc/backend.hpp

@@ -228,4 +228,4 @@ struct fmt::formatter<huggingface::tgi::backends::trtllm::sampling_params_t> : f
     }
 };
 
-#endif
+#endif

backends/trtllm/csrc/ffi.hpp

@@ -159,4 +159,4 @@ namespace huggingface::tgi::backends::trtllm {
         );
     }
 }
-#endif
+#endif

backends/trtllm/csrc/hardware.hpp

@@ -78,4 +78,4 @@ namespace huggingface::tgi::hardware::cuda {
         [[nodiscard]] constexpr bool is_at_least_hopper() const { return is_at_least(HOPPER); }
     };
 }
-#endif
+#endif

backends/trtllm/src/looper.rs

@@ -133,13 +133,16 @@ fn executor_status_looper(
                                 Ok(decoded_token) => {
                                     post_process_decoded_token(&tokenizer, ctx, decoded_token)
                                 }
-                                Err(err) => Err(err)
+                                Err(err) => Err(err),
                             };
 
                             // Attempt to send back the response to the client
                             if let Err(_) = ctx.streamer.send(response) {
                                 // Client has dropped, remove from tracked requests
-                                debug!("Client dropped - removing request {} from tracked requests", step.request_id);
+                                debug!(
+                                    "Client dropped - removing request {} from tracked requests",
+                                    step.request_id
+                                );
                                 backend.as_mut().cancel(step.request_id);
                                 let _ = in_flights.remove(&step.request_id);
                             }
@@ -160,11 +163,14 @@ fn executor_status_looper(
     }
 }
 
-fn post_process_decoded_token(tokenizer: &Tokenizer, ctx: &mut GenerationContext, decoded_token: DecodedToken) -> InferResult<InferStreamResponse> {
+fn post_process_decoded_token(
+    tokenizer: &Tokenizer,
+    ctx: &mut GenerationContext,
+    decoded_token: DecodedToken,
+) -> InferResult<InferStreamResponse> {
     match tokenizer.decode(&[decoded_token.id], false) {
         Ok(text) => {
-            let is_special =
-                tokenizer.get_added_vocabulary().is_special_token(&text);
+            let is_special = tokenizer.get_added_vocabulary().is_special_token(&text);
             let token = Token {
                 id: decoded_token.id,
                 text,
@@ -186,7 +192,7 @@ fn post_process_decoded_token(tokenizer: &Tokenizer, ctx: &mut GenerationContext
                 let generated_text = GeneratedText {
                     text: text.unwrap(),
                     generated_tokens: ctx.tokens.len() as u32,
-                    finish_reason: FinishReason::EndOfSequenceToken,  // TODO : Map FinishReason
+                    finish_reason: FinishReason::EndOfSequenceToken, // TODO : Map FinishReason
                     seed: None,
                 };
 
@@ -248,7 +254,6 @@ unsafe impl Send for TensorRtLlmBackendImpl {}
 
 pub struct TensorRtLlmBackendV2(UnboundedSender<GenerationContext>);
 
-
 impl TensorRtLlmBackendV2 {
     pub fn new<P: AsRef<Path> + Send, PP: AsRef<Path> + Send>(
         tokenizer: Tokenizer,
@@ -268,12 +273,7 @@ impl TensorRtLlmBackendV2 {
 
         // Executor looper is responsible for scheduling and pulling requests state at regular interval
         spawn_blocking(move || {
-            executor_status_looper(
-                max_inflight_requests,
-                tokenizer,
-                backend,
-                executor_receiver,
-            )
+            executor_status_looper(max_inflight_requests, tokenizer, backend, executor_receiver)
         });
 
         Ok(TensorRtLlmBackendV2(executor_sender))