[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
@@ -22,7 +22,6 @@ from vllm.logger import init_logger
from vllm.platforms import current_platform
from ..utils import StatelessProcessGroup
from .aiter_custom_all_reduce import AiterCustomAllreduce
from .base_device_communicator import DeviceCommunicatorBase
logger = init_logger(__name__)
@@ -117,7 +116,7 @@ class CudaCommunicator(DeviceCommunicatorBase):
self.symm_mem_comm: SymmMemCommunicator | None = None
self.fi_ar_comm: FlashInferAllReduce | None = None
self.fi_pcie_ipc_ar_comm: FlashInferPcieIpcAllReduce | None = None
self.aiter_ar_comm: AiterCustomAllreduce | None = None
self.aiter_ar_comm = None
self.use_aiter_ag_rs: bool = False
# cuMem graph buffers cannot be IPC-registered; capture copies them instead.
@@ -147,12 +146,6 @@ class CudaCommunicator(DeviceCommunicatorBase):
device=self.device,
)
if self.use_aiter_allreduce and self.world_size > 1:
self.aiter_ar_comm = AiterCustomAllreduce(
group=self.cpu_group,
device=self.device,
register_graph_buffers=register,
)
if use_custom_allreduce and self.aiter_ar_comm is None and self.world_size > 1:
# Initialize a custom fast all-reduce implementation.
@@ -165,22 +158,6 @@ class CudaCommunicator(DeviceCommunicatorBase):
register_graph_buffers=register,
)
# AITER custom all-gather/reduce-scatter DP-attention dispatch/combine
if (
"dp" in unique_name
and self.world_size in (2, 4, 8)
and current_platform.is_rocm()
and rocm_aiter_ops.is_custom_all_reduce_enabled()
):
self.aiter_ar_comm = AiterCustomAllreduce(
group=self.cpu_group,
device=self.device,
register_graph_buffers=register,
)
if self.aiter_ar_comm.disabled:
self.aiter_ar_comm = None
else:
self.use_aiter_ag_rs = True
if (
use_custom_allreduce