Merge branch 'main' into wentao-fix-amd-ci-test-others-bug

Signed-off-by: yewentao256 <zhyanwentao@126.com>

.buildkite/scripts/hardware_ci/run-amd-test.sh

@@ -161,7 +161,7 @@ re_quote_pytest_markers() {
       # If the token we're about to collect already contains a literal
       # single quote, the expression was already quoted upstream.
       # Flush and stop collecting.
-      if [[ "$word" == *"'"* ]]; then
+      if [[ "$word" == \'* || "$word" == *\' || "$word" == \"* || "$word" == *\" ]]; then
         if [[ -n "$marker_buf" ]]; then
           # Should not normally happen (partial buf + quote), flush raw
           output+="${marker_buf} "

.buildkite/test-amd.yaml

@@ -411,7 +411,7 @@ steps:
     - tests/v1
   commands:
     # Only run tests that need exactly 2 GPUs
-    - pytest -v -s v1/e2e/test_spec_decode.py -k "tensor_parallelism"
+    - pytest -v -s v1/e2e/test_spec_decode.py -k 'tensor_parallelism'
 
 - label: V1 Test e2e (4 GPUs) # 65min
   timeout_in_minutes: 90
@@ -426,7 +426,7 @@ steps:
     - tests/v1
   commands:
     # Only run tests that need 4 GPUs
-    - pytest -v -s v1/e2e/test_spec_decode.py -k "eagle_correctness_heavy"
+    - pytest -v -s v1/e2e/test_spec_decode.py -k 'eagle_correctness_heavy'
 
 - label: V1 Test entrypoints # 35min
   timeout_in_minutes: 50

.buildkite/test_areas/engine.yaml

@@ -46,7 +46,7 @@ steps:
     - tests/v1/e2e
   commands:
     # Only run tests that need exactly 2 GPUs
-    - pytest -v -s v1/e2e/test_spec_decode.py -k "tensor_parallelism"
+    - pytest -v -s v1/e2e/test_spec_decode.py -k 'tensor_parallelism'
   mirror:
     amd:
       device: mi325_2
@@ -62,7 +62,7 @@ steps:
     - tests/v1/e2e
   commands:
     # Only run tests that need 4 GPUs
-    - pytest -v -s v1/e2e/test_spec_decode.py -k "eagle_correctness_heavy"
+    - pytest -v -s v1/e2e/test_spec_decode.py -k 'eagle_correctness_heavy'
   mirror:
     amd:
       device: mi325_4