[v1] Add __repr__ to KVCacheBlock to avoid recursive print

[heheda12345]

As KVCacheBlock is a doubly linked list, the default print of one block will recursively print all KVCacheBlock objects in the list, and can cause stack overflow. To avoid this problem, print prev_block_id and next_block_id instead.

Reproduce script

from vllm.multimodal.inputs import MultiModalKwargs from vllm.sampling_params import SamplingParams from vllm.v1.core.kv_cache_manager import KVCacheManager from vllm.v1.request import Request

def make_request(request_id,
prompt_token_ids,
mm_positions=None,
mm_hashes=None):
if mm_positions is None:
multi_modal_inputs = None
else:
multi_modal_inputs = [MultiModalKwargs({})] * len(mm_positions)

return Request(
    request_id=request_id,
    prompt=None,
    prompt_token_ids=prompt_token_ids,
    multi_modal_inputs=multi_modal_inputs,
    multi_modal_hashes=mm_hashes,
    multi_modal_placeholders=mm_positions,
    sampling_params=SamplingParams(max_tokens=17),
    eos_token_id=100,
    arrival_time=0,
    lora_request=None,
)

manager = KVCacheManager(
block_size=16,
num_gpu_blocks=10000,
max_model_len=819200,
sliding_window=None,
enable_caching=True,
num_preallocate_tokens=16,
)

common_token_ids = [i for i in range(10) for _ in range(16)]
req0 = make_request("0", common_token_ids)
computed_blocks, num_computed_tokens = manager.get_computed_blocks(req0)
blocks = manager.allocate_slots(req0, 55, computed_blocks)
manager.free(req0)

req1 = make_request("1", common_token_ids)
computed_blocks, num_computed_tokens = manager.get_computed_blocks(req1)
print("computed_blocks", computed_blocks)

[github-actions]

👋 Hi! Thank you for contributing to the vLLM project.

💬 Join our developer Slack at https://slack.vllm.ai to discuss your PR in #pr-reviews, coordinate on features in #feat- channels, or join special interest groups in #sig- channels.

Just a reminder: PRs would not trigger full CI run by default. Instead, it would only run fastcheck CI which starts running only a small and essential subset of CI tests to quickly catch errors. You can run other CI tests on top of those by going to your fastcheck build on Buildkite UI (linked in the PR checks section) and unblock them. If you do not have permission to unblock, ping simon-mo or khluu to add you in our Buildkite org.

Once the PR is approved and ready to go, your PR reviewer(s) can run CI to test the changes comprehensively before merging.

To run CI, PR reviewers can either: Add ready label to the PR or enable auto-merge.

🚀