[feat] baseline4 profiling

computility-run.yaml

@@ -35,15 +35,25 @@ command:
 env:
     - name: VLLM_ENGINE_ITERATION_TIMEOUT_S
       value: 3600
+    # --- MoE kernel selection ---
     - name: BI100_MOE_COREX_DIRECT_ROUTED
-      value: 1
+      value: '1'
+    - name: BI100_MOE_COREX_TOPK_SOFTMAX
+      value: '1'
+    # --- GDN kernel selection ---
     - name: BI100_GDN_COREX_PACKED_DECODE
-      value: 1
+      value: '1'
+    - name: BI100_GDN_COMBINED_QK_NORM
+      value: '1'
+    # --- Hybrid KV/GDN cache ---
     - name: BI100_HYBRID_KV_ACCOUNTING
       value: full_attention
     - name: BI100_GDN_CACHE_POLICY
       value: admission64
     - name: BI100_GDN_RESTORE_MODE
       value: hybrid64
-    - name: BI100_MOE_COREX_TOPK_SOFTMAX
+    # --- Profiling (event mode: no sync overhead) ---
+    - name: BI100_PROFILE
       value: '1'
+    - name: BI100_PROFILE_MODE
+      value: event

qwen3_6_scripts/qwen3_5.py

@@ -1704,8 +1704,13 @@ class Qwen3_5MoeSparseBlock(nn.Module):
         # Tier 0: Full fused MoE via ix_moe_bridge (xllm 7-step pipeline)
         # topk → gen_idx → expand → group_gemm → silu → group_gemm → combine
         # Source: xllm/core/layers/ilu/fused_moe.cpp
+        # NOTE: Only use for prefill (T>1). For decode (T=1), group_gemm
+        # does 8× M=1 GEMMs that are completely memory-bound (<5% GPU util).
+        # The Tier 1 T=1 path below uses corex_moe_direct_routed or
+        # corex_batched_gemm.moe_decode_fused, which are purpose-built
+        # fused kernels for single-token MoE dispatch.
         # ---------------------------------------------------------------
-        if _USE_IX_FUSED_MOE:
+        if _USE_IX_FUSED_MOE and hidden_states.shape[0] > 1:
             w13 = self.experts.w13_weight  # (E, 2*I, H)
             w2 = self.experts.w2_weight    # (E, H, I)
             return _ix_fused_moe.fused_moe_forward(
@@ -1720,7 +1725,7 @@ class Qwen3_5MoeSparseBlock(nn.Module):
         # No physical transpose, no weight gather copy
         # Source: ds_vllm/vllm/.../experts/fused_batched_moe.py
         # ---------------------------------------------------------------
-        if _USE_NAIVE_BATCHED_MOE:
+        if _USE_NAIVE_BATCHED_MOE and hidden_states.shape[0] > 1:
             w13 = self.experts.w13_weight  # (E, 2*I, H)
             w2 = self.experts.w2_weight    # (E, H, I)