[TRTLLM-8638][fix] add waives tests (#8445)

Signed-off-by: Xin He (SW-GPU) <200704525+xinhe-nv@users.noreply.github.com>

Author: xinhe-nv

tests/integration/defs/accuracy/test_llm_api_pytorch.py

@@ -690,6 +690,8 @@ class TestLlama4MaverickInstruct(LlmapiAccuracyTestHarness):
     def test_auto_dtype(self, cuda_graph, tp_size, pp_size, ep_size):
         if get_device_count() != tp_size * pp_size:
             pytest.skip("Device count mismatch with world size")
+        if get_device_memory() < 240000 and get_device_count() < 8:
+            pytest.skip("Not enough memory for this test")
 
         kv_cache_config = KvCacheConfig(free_gpu_memory_fraction=0.8)
         with LLM(

tests/integration/test_lists/waives.txt

@@ -375,3 +375,8 @@ full:B200/examples/test_llama.py::test_llm_llama_v1_multiple_lora_1gpu[luotuo_ja
 full:B200/examples/test_llama.py::test_llm_llama_2gpu_fp8_summary[llama-v2-13b-hf-enable_reduce_fusion-enable_fp8_context_fmha_xqa] SKIP (https://nvbugs/5568052)
 full:B200/examples/test_llama.py::test_llm_llama_2gpu_fp8_summary[llama-v2-13b-hf-disable_reduce_fusion-disable_fp8_context_fmha_xqa] SKIP (https://nvbugs/5568052)
 full:B200/examples/test_llama.py::test_llm_llama_2gpu_fp8_summary[llama-7b-enable_reduce_fusion-disable_fp8_context_fmha_xqa] SKIP (https://nvbugs/5568052)
+test_e2e.py::test_ptp_quickstart_multimodal[NVILA-8B-FP16-vila/NVILA-8B-video-False] SKIP (https://nvbugs/5591109)
+test_e2e.py::test_ptp_quickstart_multimodal[qwen2-vl-7b-instruct-Qwen2-VL-7B-Instruct-video-False] SKIP (https://nvbugs/5547437)
+accuracy/test_cli_flow.py::TestMixtral8x7B::test_fp4_plugin SKIP (https://nvbugs/5451207)
+accuracy/test_cli_flow.py::TestMixtral8x22B::test_fp8_tp2pp2 SKIP (https://nvbugs/5511944)
+examples/test_llama.py::test_llm_llama_1gpu_fp4[llama-3.1-70b-instruct-enable_norm_quant_fusion-enable_fused_quant-fp4_plugin-bfloat16] SKIP (https://nvbugs/5543383)