[V1][Spec Decode] Update target_logits in place for rejection sampling (vllm-project#15427)

Signed-off-by: Woosuk Kwon <woosuk.kwon@berkeley.edu>
Signed-off-by: Mu Huai <tianbowen.tbw@antgroup.com>

Author: WoosukKwon

vllm/v1/sample/rejection_sampler.py

@@ -67,6 +67,7 @@ def forward(
                 Shape is [num_tokens, vocab_size]. Here, probabilities from
                 different requests are flattened into a single tensor because
                 this is the shape of the output logits.
+                NOTE: `target_logits` can be updated in place to save memory.
             bonus_token_ids_tensor (torch.Tensor):
                 A tensor containing bonus tokens. Shape is [batch_size, 1].
                 Bonus tokens are added to the end of the sequence if all
@@ -83,6 +84,8 @@ def forward(
         '''
         assert metadata.max_spec_len <= MAX_SPEC_LEN
         # [num_tokens, vocab_size]
+        # NOTE(woosuk): `target_logits` can be updated in place inside the
+        # `compute_probs` function.
         target_probs = compute_probs(
             target_logits,
             metadata.cu_num_draft_tokens,
@@ -252,8 +255,8 @@ def compute_probs(
         replace_from=GREEDY_TEMPERATURE,
         replace_to=1,
     )
-    # TODO(woosuk): Consider using in-place op to reduce memory usage.
-    logits = logits / temperature.unsqueeze(-1)
+    # NOTE(woosuk): Update `logits` in place to avoid allocating a new tensor.
+    logits.div_(temperature.unsqueeze(-1))
 
     # Get expanded top_k and top_p tensors.
     top_k = None

vllm/v1/worker/gpu_model_runner.py

@@ -1059,15 +1059,20 @@ def execute_model(
                 sampling_metadata=sampling_metadata,
             )
         else:
-            # TODO(woosuk): Optimize the memory usage.
+            # When indexing with a tensor (bonus_logits_indices), PyTorch
+            # creates a new tensor with separate storage from the original
+            # logits tensor. This means any in-place operations on bonus_logits
+            # won't affect the original logits tensor.
             bonus_logits = logits[spec_decode_metadata.bonus_logits_indices]
             sampler_output = self.model.sample(
                 logits=bonus_logits,
                 sampling_metadata=sampling_metadata,
             )
             bonus_token_ids = sampler_output.sampled_token_ids
 
-            # TODO(woosuk): Optimize the memory usage.
+            # Just like `bonus_logits`, `target_logits` is a new tensor with
+            # separate storage from the original `logits` tensor. Therefore,
+            # it is safe to update `target_logits` in place.
             target_logits = logits[spec_decode_metadata.target_logits_indices]
             output_token_ids = self.rejection_sampler(
                 spec_decode_metadata,