working & cached tunableop

launcher/src/main.rs

@@ -672,9 +672,7 @@ fn shard_manager(
 
         // We received a shutdown signal
         if shutdown.load(Ordering::SeqCst) {
-            p.kill().unwrap();
-            let _ = p.wait();
-            tracing::info!("Shard terminated");
+            terminate("shard", p, Duration::from_secs(90)).unwrap();
             return;
         }
 

server/text_generation_server/models/flash_causal_lm.py

@@ -766,6 +766,9 @@ class FlashCausalLM(Model):
             )
             max_bt = batch.max_blocks
             max_s = max_bt * get_cache_manager().block_size
+
+            if IS_ROCM_SYSTEM and os.environ.get("PYTORCH_TUNABLEOP_ENABLED", False):
+                logger.info("PyTorch TunableOp (https://github.com/pytorch/pytorch/tree/v2.3.0/aten/src/ATen/cuda/tunable) is enabled. The warmup may take several minutes.")
             _, batch, _ = self.generate_token(batch)
         except torch.cuda.OutOfMemoryError as e:
             raise RuntimeError(
@@ -820,10 +823,10 @@ class FlashCausalLM(Model):
         else:
             logger.info(f"Cuda Graphs are disabled (CUDA_GRAPHS={CUDA_GRAPHS}).")
 
-        # TODO: fix
-        if IS_ROCM_SYSTEM and False:
-            total_seqlens = list(range(16))
+        if IS_ROCM_SYSTEM and os.environ.get("PYTORCH_TUNABLEOP_ENABLED", False):
+            total_seqlens = list(range(2))
             for seqlen in total_seqlens:
+                logger.info(f"Warming up TunableOp for seqlen={seqlen}")
                 self.tunableop_warmup(seqlen, max_s, max_bt)
 
         return int(num_blocks * BLOCK_SIZE)

server/text_generation_server/server.py

@@ -166,6 +166,21 @@ class TextGenerationService(generate_pb2_grpc.TextGenerationServiceServicer):
             total_ns=time.time_ns() - start,
         )
 
+import signal
+
+class SignalHandler:
+    KEEP_PROCESSING = True
+
+    def __init__(self):
+        signal.signal(signal.SIGINT, self.exit_gracefully)
+        signal.signal(signal.SIGTERM, self.exit_gracefully)
+
+    def exit_gracefully(self, signum, frame):
+        print(f"Exiting gracefully: Signal {signum}")
+        self.KEEP_PROCESSING = False
+
+
+signal_handler = SignalHandler()
 
 def serve(
     model_id: str,
@@ -231,11 +246,8 @@ def serve(
 
         logger.info("Server started at {}".format(local_url))
 
-        try:
-            await server.wait_for_termination()
-        except KeyboardInterrupt:
-            logger.info("Signal received. Shutting down")
-            await server.stop(0)
+        while signal_handler.KEEP_PROCESSING:
+            await asyncio.sleep(0.5)
 
     asyncio.run(
         serve_inner(

server/text_generation_server/utils/paged_attention.py

@@ -4,6 +4,11 @@ from text_generation_server.utils.import_utils import IS_CUDA_SYSTEM, IS_ROCM_SY
 
 _PARTITION_SIZE = 512
 
+try:
+    from vllm._C import cache_ops
+except Exception as e:
+    raise ImportError(f"Could not import vllm paged attention. Make sure your installation is correct. Complete error: {e}")
+
 
 def reshape_and_cache(
     key: torch.Tensor,
@@ -12,18 +17,9 @@ def reshape_and_cache(
     value_cache: torch.Tensor,
     slots: torch.Tensor,
 ):
-    if IS_CUDA_SYSTEM:
-        from vllm._C import cache_ops
-
     cache_ops.reshape_and_cache(
         key, value, key_cache, value_cache, slots, "auto", 1.0
     )
-    elif IS_ROCM_SYSTEM:
-        from vllm import cache_ops
-
-        cache_ops.reshape_and_cache(key, value, key_cache, value_cache, slots)
-    else:
-        raise ValueError("vllm is not supported on your system")
 
 
 def attention(