vLLM op-injection plugin: custom Attention + MoE kernels via out-of-tree entry_points (no fork)

Goal

Ship a pip-installable vLLM plugin that injects Infera/HyperLoom-optimized Attention and MoE kernels into stock vLLM — no vLLM fork — using vLLM's out-of-tree plugin mechanism (vllm.platform_plugins + vllm.general_plugins). This is the delivery vehicle for op-injection: swap just the hot kernels while staying on upstream vLLM's model code, so we track vLLM releases instead of maintaining a fork.

Mechanism (verified against vLLM 0.23; cf. AMD's vLLM-ATOM)

• Attention → Platform.get_attn_backend_cls() returns a custom AttentionBackend / AttentionImpl. Clean, per-kernel seam (also settable via VLLM_ATTENTION_BACKEND). This is exactly what vLLM-ATOM uses (ATOMPlatform(RocmPlatform) → AiterBackend/AiterMLABackend).
• MoE → FusedMoEModularKernel (FusedMoEPrepareAndFinalize + FusedMoEPermuteExpertsUnpermute) or FusedMoEMethodBase.apply() — swap just the experts GEMM while keeping vLLM's routing / quant / EP-DP dispatch.
• vLLM-ATOM instead replaces the whole model via ModelRegistry.register_model; we prefer the less-invasive per-op seams above, and keep whole-model wrapping only as a fallback where a per-op seam is insufficient (e.g. deep MLA fusion).

Packaged as entry_points so it activates on install, gated by an env kill-switch (INFERA_VLLM_OPS_DISABLE=1), mirroring ATOM's ATOM_DISABLE_VLLM_PLUGIN.

Plan

• Plugin skeleton package + entry_points (register_platform, register_ops), env-gated.
• InferaPlatform(RocmPlatform) with get_attn_backend_cls → pass-through attention backend (delegates to the default impl) as the injection seam.
• MoE: a FusedMoEModularKernel experts pass-through stub as the injection seam.
• Validation: plugin loads under vLLM; a model runs unchanged with the plugin active vs INFERA_VLLM_OPS_DISABLE=1; numerics match (pass-through = bitwise identical).
• Wire a first real optimized kernel behind the seam (attention first).

Non-goals (for now)

Whole-model wrapper; quant-method rewrites; SGLang (separate mechanism).

Progress will be posted as comments on this issue.

Start with vLLM 0.23's vllm.platform_plugins and vllm.general_plugins entry points, then trace the Platform.get_attn_backend_cls and FusedMoEModularKernel seams described here. Validate that the packaged plugin loads, the model runs unchanged with and without INFERA_VLLM_OPS_DISABLE=1, and pass-through numerics are bitwise identical before wiring the first optimized kernel.