[Infra] - Waive failed case in post-merge (NVIDIA#5536)

Signed-off-by: qqiao <[email protected]>

Author: EmmaQiaoCh

tests/unittest/llmapi/test_llm_kv_cache_events.py

@@ -101,6 +101,7 @@ def test_kv_cache_event_data_serialization():
     serialized_event = KVCacheEventSerializer.serialize(events)
 
 
+@pytest.mark.skip(reason="https://nvbugs/5362412")
 def test_expected_kv_cache_events():
     llm = create_llm()
     sampling_params = SamplingParams(max_tokens=6, temperature=0.01)
@@ -119,6 +120,7 @@ def test_expected_kv_cache_events():
                 assert event["data"]["type"] == "stored"
 
 
+@pytest.mark.skip(reason="https://nvbugs/5362412")
 def test_kv_cache_event_async_api():
     llm = create_llm()
     sampling_params = SamplingParams(max_tokens=6, temperature=0.01)
@@ -146,6 +148,7 @@ async def main():
     asyncio.run(main())
 
 
+@pytest.mark.skip(reason="https://nvbugs/5362412")
 def test_llm_kv_events_api():
     llm = create_llm()
     sampling_params = SamplingParams(max_tokens=6, temperature=0.01)

tests/unittest/llmapi/test_llm_multi_gpu.py

@@ -267,6 +267,7 @@ def test_llama_7b_multi_lora_tp2():
                                      kv_cache_config=global_kv_cache_config)
 
 
+@pytest.mark.skip(reason="https://nvbugs/5362426")
 @pytest.mark.gpu2
 @pytest.mark.part3
 def test_llama_v2_7b_prompt_adapter_tp2():