ci: try remaining 20 gpu_1_queue tests on h200_35gb with NVML fix

These 20 tests failed on h200_35gb in build #66777 before the
/dev/nvidiactl mount fix. Retrying now that NVML access is fixed.

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

.buildkite/test_areas/cuda.yaml

@@ -18,6 +18,7 @@ steps:
     - pytest -v -s cuda/test_platform_no_cuda_init.py
 
 - label: Cudagraph
+  device: h200_35gb
   key: cudagraph
   timeout_in_minutes: 20
   source_file_dependencies:

.buildkite/test_areas/engine.yaml

@@ -28,6 +28,7 @@ steps:
   - pytest -v -s engine test_sequence.py test_config.py test_logger.py test_vllm_port.py test_jit_monitor.py
 
 - label: Engine (1 GPU)
+  device: h200_35gb
   key: engine-1-gpu
   timeout_in_minutes: 30
   source_file_dependencies:

.buildkite/test_areas/entrypoints.yaml

@@ -3,6 +3,7 @@ depends_on:
   - image-build
 steps:
 - label: Entrypoints Unit Tests
+  device: h200_35gb
   key: entrypoints-unit-tests
   timeout_in_minutes: 10
   working_dir: "/vllm-workspace/tests"
@@ -14,6 +15,7 @@ steps:
   - pytest -v -s entrypoints/ --ignore=entrypoints/llm --ignore=entrypoints/rpc --ignore=entrypoints/sleep --ignore=entrypoints/serve/instrumentator --ignore=entrypoints/openai --ignore=entrypoints/offline_mode --ignore=entrypoints/test_chat_utils.py  --ignore=entrypoints/pooling --ignore=entrypoints/speech_to_text
 
 - label: Entrypoints Integration (LLM)
+  device: h200_35gb
   key: entrypoints-integration-llm
   timeout_in_minutes: 40
   working_dir: "/vllm-workspace/tests"
@@ -33,6 +35,7 @@ steps:
       - image-build-amd
 
 - label: Entrypoints Integration (API Server openai - Part 1)
+  device: h200_35gb
   key: entrypoints-integration-api-server-openai-part-1
   timeout_in_minutes: 50
   working_dir: "/vllm-workspace/tests"
@@ -51,6 +54,7 @@ steps:
       - image-build-amd
 
 - label: Entrypoints Integration (API Server openai - Part 2)
+  device: h200_35gb
   key: entrypoints-integration-api-server-openai-part-2
   timeout_in_minutes: 50
   working_dir: "/vllm-workspace/tests"
@@ -121,6 +125,7 @@ steps:
   - pytest -v -s entrypoints/speech_to_text
 
 - label: Entrypoints Integration (Pooling)
+  device: h200_35gb
   key: entrypoints-integration-pooling
   timeout_in_minutes: 50
   working_dir: "/vllm-workspace/tests"
@@ -132,6 +137,7 @@ steps:
   - pytest -v -s entrypoints/pooling
 
 - label: Entrypoints Integration (Responses API)
+  device: h200_35gb
   key: entrypoints-integration-responses-api
   timeout_in_minutes: 50
   working_dir: "/vllm-workspace/tests"

.buildkite/test_areas/kernels.yaml

@@ -15,6 +15,7 @@ steps:
     - pytest -v -s tests/kernels/ir
 
 - label: Kernels Core Operation Test
+  device: h200_35gb
   key: kernels-core-operation-test
   timeout_in_minutes: 75
   source_file_dependencies:
@@ -39,6 +40,7 @@ steps:
     - pytest -v -s kernels/core/test_minimax_reduce_rms.py
 
 - label: Kernels Attention Test %N
+  device: h200_35gb
   key: kernels-attention-test
   timeout_in_minutes: 35
   source_file_dependencies:
@@ -53,6 +55,7 @@ steps:
   parallelism: 2
 
 - label: Kernels Quantization Test %N
+  device: h200_35gb
   key: kernels-quantization-test
   timeout_in_minutes: 90
   source_file_dependencies:
@@ -77,6 +80,7 @@ steps:
       - image-build-amd
 
 - label: Kernels MoE Test %N
+  device: h200_35gb
   key: kernels-moe-test
   timeout_in_minutes: 25
   source_file_dependencies:
@@ -93,6 +97,7 @@ steps:
   parallelism: 5
 
 - label: Kernels Mamba Test
+  device: h200_35gb
   key: kernels-mamba-test
   timeout_in_minutes: 45
   source_file_dependencies:

.buildkite/test_areas/misc.yaml

@@ -57,6 +57,7 @@ steps:
       - image-build-amd
 
 - label: V1 Core + KV + Metrics
+  device: h200_35gb
   key: v1-core-kv-metrics
   timeout_in_minutes: 30
   source_file_dependencies:

.buildkite/test_areas/models_language.yaml

@@ -17,6 +17,7 @@ steps:
     torch_nightly: {}
 
 - label: Language Models Tests (Extra Standard) %N
+  device: h200_35gb
   key: language-models-tests-extra-standard
   timeout_in_minutes: 45
   source_file_dependencies:

.buildkite/test_areas/models_multimodal.yaml

@@ -106,6 +106,7 @@ steps:
   - pytest -s -v test_lm_eval_correctness.py --config-list-file=configs/models-mm-small.txt --tp-size=1
 
 - label: Multi-Modal Models (Extended Generation 1)
+  device: h200_35gb
   key: multi-modal-models-extended-generation-1
   optional: true
   source_file_dependencies:

.buildkite/test_areas/pytorch.yaml

@@ -77,6 +77,7 @@ steps:
   - "find compile/h100/ -name 'test_*.py' -print0 | xargs -0 -n1 -I{} pytest -s -v '{}'"
 
 - label: PyTorch Compilation Passes Unit Tests
+  device: h200_35gb
   key: pytorch-compilation-passes-unit-tests
   timeout_in_minutes: 20
   source_file_dependencies:
@@ -109,6 +110,7 @@ steps:
   - pytest -s -v compile/passes --ignore compile/passes/distributed
 
 - label: PyTorch Fullgraph Smoke Test
+  device: h200_35gb
   key: pytorch-fullgraph-smoke-test
   timeout_in_minutes: 35
   source_file_dependencies:

.buildkite/test_areas/quantization.yaml

@@ -3,6 +3,7 @@ depends_on:
   - image-build
 steps:
 - label: Quantization
+  device: h200_35gb
   key: quantization
   timeout_in_minutes: 90
   source_file_dependencies:
@@ -40,6 +41,7 @@ steps:
     - pytest -s -v tests/quantization/test_blackwell_moe.py
 
 - label: Quantized Models Test
+  device: h200_35gb
   key: quantized-models-test
   timeout_in_minutes: 60
   source_file_dependencies: