From 67cee40da035b7478483c76dfbe0bfc321c3822f Mon Sep 17 00:00:00 2001
From: "Li, Jiang" <jiang1.li@intel.com>
Date: Thu, 28 Aug 2025 19:57:05 +0800
Subject: [PATCH] [CI/Build][Bugfix] Fix Qwen VL tests on CPU (#23818)

Signed-off-by: jiang1.li <jiang1.li@intel.com>
---
 .../scripts/hardware_ci/run-cpu-test.sh       | 20 +++++++++----------
 vllm/model_executor/models/utils.py           |  8 ++++----
 2 files changed, 14 insertions(+), 14 deletions(-)

diff --git a/.buildkite/scripts/hardware_ci/run-cpu-test.sh b/.buildkite/scripts/hardware_ci/run-cpu-test.sh
index 9dec9f8e9eb3..8b8f0e8c6578 100644
--- a/.buildkite/scripts/hardware_ci/run-cpu-test.sh
+++ b/.buildkite/scripts/hardware_ci/run-cpu-test.sh
@@ -49,23 +49,23 @@ function cpu_tests() {
   # Run kernel tests
   docker exec cpu-test-"$NUMA_NODE" bash -c "
     set -e
-    pytest -v -s tests/kernels/test_onednn.py"
+    pytest -x -v -s tests/kernels/test_onednn.py"
 
   # Run basic model test
   docker exec cpu-test-"$NUMA_NODE" bash -c "
     set -e
     # Note: disable until supports V1
-    # pytest -v -s tests/kernels/attention/test_cache.py -m cpu_model
-    # pytest -v -s tests/kernels/attention/test_mla_decode_cpu.py -m cpu_model
+    # pytest -x -v -s tests/kernels/attention/test_cache.py -m cpu_model
+    # pytest -x -v -s tests/kernels/attention/test_mla_decode_cpu.py -m cpu_model
 
     # Note: disable Bart until supports V1
-    pytest -v -s tests/models/language/generation -m cpu_model \
+    pytest -x -v -s tests/models/language/generation -m cpu_model \
                 --ignore=tests/models/language/generation/test_bart.py
-    VLLM_CPU_SGL_KERNEL=1 pytest -v -s tests/models/language/generation -m cpu_model \
+    VLLM_CPU_SGL_KERNEL=1 pytest -x -v -s tests/models/language/generation -m cpu_model \
                 --ignore=tests/models/language/generation/test_bart.py
 
-    pytest -v -s tests/models/language/pooling -m cpu_model
-    pytest -v -s tests/models/multimodal/generation \
+    pytest -x -v -s tests/models/language/pooling -m cpu_model
+    pytest -x -v -s tests/models/multimodal/generation \
                 --ignore=tests/models/multimodal/generation/test_mllama.py \
                 --ignore=tests/models/multimodal/generation/test_pixtral.py \
                 -m cpu_model"
@@ -73,20 +73,20 @@ function cpu_tests() {
   # Run compressed-tensor test
   docker exec cpu-test-"$NUMA_NODE" bash -c "
     set -e
-    pytest -s -v \
+    pytest -x -s -v \
     tests/quantization/test_compressed_tensors.py::test_compressed_tensors_w8a8_logprobs[False-10-32-neuralmagic/Llama-3.2-1B-quantized.w8a8]"
 
   # Note: disable it until supports V1
   # Run AWQ test
   # docker exec cpu-test-"$NUMA_NODE" bash -c "
   #   set -e
-  #   VLLM_USE_V1=0 pytest -s -v \
+  #   VLLM_USE_V1=0 pytest -x -s -v \
   #   tests/quantization/test_ipex_quant.py"
 
   # Run multi-lora tests
   docker exec cpu-test-"$NUMA_NODE" bash -c "
     set -e
-    pytest -s -v \
+    pytest -x -s -v \
     tests/lora/test_qwen2vl.py"
 
   # online serving
diff --git a/vllm/model_executor/models/utils.py b/vllm/model_executor/models/utils.py
index 11e098f1d7bd..28cfefac30dd 100644
--- a/vllm/model_executor/models/utils.py
+++ b/vllm/model_executor/models/utils.py
@@ -507,10 +507,10 @@ def merge_multimodal_embeddings(
         This updates ``inputs_embeds`` in place.
     """
     if isinstance(placeholder_token_id, list):
-        placeholder_token_id = torch.tensor(placeholder_token_id,
-                                            pin_memory=True).to(
-                                                device=input_ids.device,
-                                                non_blocking=True)
+        placeholder_token_id = torch.tensor(
+            placeholder_token_id,
+            pin_memory=is_pin_memory_available()).to(device=input_ids.device,
+                                                     non_blocking=True)
         return _merge_multimodal_embeddings(
             inputs_embeds,
             torch.isin(input_ids, placeholder_token_id),