biondizzle/vllm
1
0
Fork 0
Code Issues Pull Requests Actions 2 Packages Projects Releases Wiki Activity

[CI][BugFix] ShellCheck cleanup to remove baseline and preserve runtime behavior (#34514)

Browse Source 
Signed-off-by: junuxyz <216036880+junuxyz@users.noreply.github.com>
This commit is contained in:
junuxyz
2026-02-17 21:22:56 +09:00
committed by GitHub
parent 28bffe9466
commit c61a98f529
55 changed files with 338 additions and 464 deletions
Download Patch File Download Diff File Expand all files Collapse all files

8
.buildkite/scripts/scheduled_integration_test/deepseek_v2_lite_ep_eplb.sh
View File

@@ -51,14 +51,14 @@ for BACK in "${BACKENDS[@]}"; do
--enable-eplb \
--trust-remote-code \
--max-model-len 2048 \
--all2all-backend $BACK \
--port $PORT &
--all2all-backend "$BACK" \
--port "$PORT" &
SERVER_PID=$!
wait_for_server $PORT
wait_for_server "$PORT"
TAG=$(echo "$MODEL" | tr '/: \\n' '_____')
OUT="${OUT_DIR}/${TAG}_${BACK}.json"
python3 tests/evals/gsm8k/gsm8k_eval.py --host http://127.0.0.1 --port $PORT --num-questions ${NUM_Q} --save-results ${OUT}
python3 tests/evals/gsm8k/gsm8k_eval.py --host http://127.0.0.1 --port "$PORT" --num-questions "${NUM_Q}" --save-results "${OUT}"
python3 - <<PY
import json; acc=json.load(open('${OUT}'))['accuracy']
print(f"${MODEL} ${BACK}: accuracy {acc:.3f}")

12
.buildkite/scripts/scheduled_integration_test/qwen30b_a3b_fp8_block_ep_eplb.sh
View File

@@ -47,20 +47,20 @@ for BACK in "${BACKENDS[@]}"; do
vllm serve "$MODEL" \
--enforce-eager \
--enable-eplb \
--all2all-backend $BACK \
--all2all-backend "$BACK" \
--eplb-config '{"window_size":10, "step_interval":100, "num_redundant_experts":0, "log_balancedness":true}' \
--tensor-parallel-size ${TENSOR_PARALLEL_SIZE} \
--data-parallel-size ${DATA_PARALLEL_SIZE} \
--tensor-parallel-size "${TENSOR_PARALLEL_SIZE}" \
--data-parallel-size "${DATA_PARALLEL_SIZE}" \
--enable-expert-parallel \
--trust-remote-code \
--max-model-len 2048 \
--port $PORT &
--port "$PORT" &
SERVER_PID=$!
wait_for_server $PORT
wait_for_server "$PORT"
TAG=$(echo "$MODEL" | tr '/: \\n' '_____')
OUT="${OUT_DIR}/${TAG}_${BACK}.json"
python3 tests/evals/gsm8k/gsm8k_eval.py --host http://127.0.0.1 --port $PORT --num-questions ${NUM_Q} --save-results ${OUT}
python3 tests/evals/gsm8k/gsm8k_eval.py --host http://127.0.0.1 --port "$PORT" --num-questions "${NUM_Q}" --save-results "${OUT}"
python3 - <<PY
import json; acc=json.load(open('${OUT}'))['accuracy']
print(f"${MODEL} ${BACK}: accuracy {acc:.3f}")

8
.buildkite/scripts/scheduled_integration_test/qwen3_next_mtp_async_eplb.sh
View File

@@ -51,20 +51,20 @@ for BACK in "${BACKENDS[@]}"; do
--tensor-parallel-size 4 \
--enable-expert-parallel \
--enable-eplb \
--all2all-backend $BACK \
--all2all-backend "$BACK" \
--eplb-config '{"window_size":200,"step_interval":600,"use_async":true}' \
--speculative-config '{"method":"qwen3_next_mtp","num_speculative_tokens":1}' \
--trust-remote-code \
--max-model-len 2048 \
--gpu-memory-utilization 0.9 \
"${PLATFORM_ARGS[@]}" \
--port $PORT &
--port "$PORT" &
SERVER_PID=$!
wait_for_server $PORT
wait_for_server "$PORT"
TAG=$(echo "$MODEL" | tr '/: \\n' '_____')
OUT="${OUT_DIR}/${TAG}_${BACK}.json"
python3 tests/evals/gsm8k/gsm8k_eval.py --host http://127.0.0.1 --port $PORT --num-questions ${NUM_Q} --save-results ${OUT}
python3 tests/evals/gsm8k/gsm8k_eval.py --host http://127.0.0.1 --port "$PORT" --num-questions "${NUM_Q}" --save-results "${OUT}"
python3 - <<PY
import json; acc=json.load(open('${OUT}'))['accuracy']
print(f"${MODEL} ${BACK}: accuracy {acc:.3f}")