[Mini-VLLM] Remove the ROCm device path, debug routes and arch-gated tests

- Delete csrc/rocm, hipify, the _rocm_C extension and the setup.py rocm
  target path; drop the rocm requirement files
- Delete the aiter, rdna3, skinny-gemm and ROCm fused-MoE kernel modules
  and retarget the linear kernel registries away from them
- Reduce platforms/rocm.py to the GCN architecture predicates that live
  code still imports, and delete the RocmPlatform class and plugin
- Drop the VLLM_SERVER_DEV_MODE debug surface: reset, collective_rpc,
  server_info and sleep routes plus the log_error_stack flag
- Delete lm-format-enforcer, lark and otel-semantic-conventions-ai pins;
  the live server process maps none of them
- Delete tests that only build architectures this fork no longer
  registers, and test directories left without any test file
- Restore the tests/v1/kv_connector mock helpers that scheduler tests
  import at module level

Co-authored-by: Hinata <hinata@idihore.id>
Signed-off-by: MOVZX <movzx@yahoo.com>
This commit is contained in:
2026-10-19 19:31:07 +07:00
co-authored by Hinata
parent 2066d3143f
commit 5f3c7bd8ab
162 changed files with 599 additions and 27490 deletions
@@ -1,31 +0,0 @@
# SPDX-License-Identifier: Apache-2.0
# SPDX-FileCopyrightText: Copyright contributors to the vLLM project
import pytest
from vllm.entrypoints.llm import LLM
from vllm.sampling_params import SamplingParams
@pytest.mark.parametrize("model", ["distilbert/distilgpt2"])
def test_computed_prefix_blocks(model: str):
# This test checks if the engine generates completions both with and
# without optional detokenization, that detokenization includes text
# and no-detokenization doesn't, and that both completions have the same
# token_ids.
prompt = (
"You are a helpful assistant. How do I build a car from cardboard and "
"paper clips? Is there an easy to follow video tutorial available "
"online for free?"
)
llm = LLM(model=model)
sampling_params = SamplingParams(max_tokens=10, temperature=0.0, detokenize=False)
outputs_no_detokenization = llm.generate(prompt, sampling_params)[0].outputs[0]
sampling_params.detokenize = True
outputs_with_detokenization = llm.generate(prompt, sampling_params)[0].outputs[0]
assert outputs_no_detokenization.text == ""
assert outputs_with_detokenization.text != ""
assert outputs_no_detokenization.token_ids == outputs_with_detokenization.token_ids