[None][fix] fix TP support for DeepSeek-V3.2 on hopper (#9484)

Signed-off-by: Fanrong Li <23290157+lfr-0531@users.noreply.github.com>

Author: lfr-0531

tensorrt_llm/_torch/modules/attention.py

@@ -2060,9 +2060,10 @@ def forward_sparse_mla_kvcache_bf16(
 
         # [seq, num_heads, kv_lora_rank], account for padding
         attn_out_latent = attn_out_latent[:, :self.num_heads_tp, :]
-        # TODO: seems we need .contiguous() here when padding enabled before pass to bmm?
         attn_out_latent = attn_out_latent.view(
             [-1, self.num_heads_tp, self.kv_lora_rank])
+        if self.num_heads_tp != padding:
+            attn_out_latent = attn_out_latent.contiguous()
 
         assert (attn_out_latent.shape[0] == q.shape[0]
                 and attn_out_latent.shape[1] == self.num_heads_tp)

tests/integration/defs/accuracy/test_llm_api_pytorch.py

@@ -2409,8 +2409,12 @@ class TestDeepSeekV32(LlmapiAccuracyTestHarness):
             (8, 1, 8, 1, False, True, True, True, 24, "_DEFAULT"),
             (8, 1, 8, 0, True, True, True, True, 24, "_DEFAULT"),
             (8, 1, 8, 3, False, False, True, True, 1, "TRTLLM"),
+            (8, 1, 8, 3, False, False, True, True, 1, "_DEFAULT"),
         ],
-        ids=["baseline", "baseline_mtp1", "baseline_fp8kv", "latency"])
+        ids=[
+            "baseline", "baseline_mtp1", "baseline_fp8kv", "latency",
+            "latency_default"
+        ])
     def test_fp8_blockscale(self, tp_size, pp_size, ep_size, mtp_nextn, fp8kv,
                             attention_dp, cuda_graph, overlap_scheduler,
                             max_batch_size, moe_backend):

tests/integration/test_lists/test-db/l0_dgx_h200.yml

@@ -18,7 +18,7 @@ l0_dgx_h200:
   # - accuracy/test_llm_api_pytorch.py::TestDeepSeekR1::test_fp8_blockscale[throughput] # OOM
   - accuracy/test_llm_api_pytorch.py::TestDeepSeekR1::test_fp8_blockscale[latency] # 1h
   - accuracy/test_llm_api_pytorch.py::TestDeepSeekV32::test_fp8_blockscale[baseline]
-  - accuracy/test_llm_api_pytorch.py::TestDeepSeekV32::test_fp8_blockscale[latency]
+  - accuracy/test_llm_api_pytorch.py::TestDeepSeekV32::test_fp8_blockscale[latency_default]
   - accuracy/test_disaggregated_serving.py::TestLlama4ScoutInstruct::test_auto_dtype[True]
   - accuracy/test_disaggregated_serving.py::TestLlama4ScoutInstruct::test_auto_dtype[False]
   - accuracy/test_disaggregated_serving.py::TestDeepSeekV3Lite::test_auto_dtype[mtp_nextn=0-overlap_scheduler=True]