post rebase fixup

Signed-off-by: Anthony Chang <[email protected]>

Author: rosenrodt

tensorrt_llm/_torch/modules/fused_moe/fused_moe_trtllm_gen.py

@@ -285,7 +285,7 @@ def _quantize_for_post_quant_comm(self, x):
                     False)
         elif self.has_w4a8_mxfp4_mxfp8:
             x, x_sf = torch.ops.trtllm.mxfp8_quantize(
-                x, False, alignment=self.quant_method.weight_alignment)
+                x, False, alignment=self.quant_method.input_hidden_alignment)
             x_row, x_col = x.shape[0], x.shape[1]
         elif self.has_deepseek_fp8_block_scales:
             # No change required before communication

tests/unittest/_torch/thop/parallel/test_moe.py

@@ -1806,10 +1806,10 @@ def test_moe_fp8_per_tensor_scale(num_tokens, hidden_size, intermediate_size,
                                                         tile_tokens_dim)
     elif routing_method_type == RoutingMethodType.Renormalize:
         permute_info, scores = routing_reference_renormalize(
-            expert_logits, top_k, num_experts, tile_tokens_dim)
+            expert_logits, top_k, tile_tokens_dim)
     elif routing_method_type == RoutingMethodType.RenormalizeNaive:
         permute_info, scores = routing_reference_renormalize_naive(
-            expert_logits, top_k, num_experts, tile_tokens_dim)
+            expert_logits, top_k, tile_tokens_dim)
     elif routing_method_type == RoutingMethodType.Llama4:
         permute_info, scores = routing_reference_no_aux(
             expert_logits, routing_bias, top_k, n_groups, top_k_groups,