MB5 driver fixes: bash env-prefix + replayer flag names + python date math

Two bugs caught by 8C smoke:

mb5_launch.sh
  ${env_bp_arg} expanded as a literal command line prefix doesn't work
  when env_bp_arg is itself a variable — bash only treats VAR=val as
  an env assignment if it sees the literal in the parsed command, not
  after expansion. Fix: always export VLLM_MOONCAKE_BOOTSTRAP_PORT as
  a literal, defaulting to 9999 when caller passed no port (consumer
  mode ignores the var so the placeholder is harmless).

mb5_run.sh
  replayer's actual CLI flags are --trace / --output / --endpoint /
  --model, not the --*-path / --*-name variants I had. Plus dash1
  has no `bc`; compute wall_clock_s via python instead.

Both fixed; 8C smoke (CONFIG=8C REPS=1 REQUEST_LIMIT=20) now runs
end-to-end in ~30 s:
  - 8 vLLM kv_both instances on GPU 0-7 come up
  - replayer round-robins 20 reqs across them
  - MB5 instrumentation captures 8 snapshot files (one per EngineCore
    PID), ranging 7-139 snapshots each = ~10 Hz throttle works
  - plot_kv_pool_timeline.py renders the stacked-area + queue-depth
    chart cleanly (figs/mb5_smoke/*.png)

Pipeline validated. Ready for the real PD-ratio sweep.

Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>

microbench/fresh_setup/mb5_launch.sh

@@ -95,16 +95,20 @@ launch_vllm() {
     local idx="$1" gpu="$2" port="$3" role="$4" bp="$5"
     local cfg="{\"kv_connector\":\"MooncakeConnector\",\"kv_role\":\"${role}\"}"
     local master=$((BASE_MASTER+idx))
-    local env_bp_arg=""
+    # Bootstrap port: always set (vllm only consumes it for producer/kv_both
-    if [ -n "${bp}" ] && [ "${bp}" != "-" ]; then
+    # modes; kv_consumer mode just ignores it). Use a placeholder if the
-        env_bp_arg="VLLM_MOONCAKE_BOOTSTRAP_PORT=${bp}"
+    # caller didn't supply one — bash one-shot env-prefix with a dynamic
+    # variable is fragile, so we always export a literal.
+    local bp_use="${bp}"
+    if [ -z "${bp_use}" ] || [ "${bp_use}" = "-" ]; then
+        bp_use="9999"
     fi
     echo "[mb5] launching idx=${idx} gpu=${gpu} port=${port} role=${role} bp=${bp:-none}"
     PYTHONHASHSEED=42 \
     MB5_LOG_DIR="${SNAPSHOT_DIR}" \
     CUDA_VISIBLE_DEVICES="${gpu}" \
     MASTER_PORT="${master}" \
-    ${env_bp_arg} \
+    VLLM_MOONCAKE_BOOTSTRAP_PORT="${bp_use}" \
     nohup vllm serve "${MODEL}" \
         --host 0.0.0.0 --port "${port}" \
         --tensor-parallel-size 1 \

microbench/fresh_setup/mb5_run.sh

@@ -71,19 +71,23 @@ run_one() {
     local t0
     t0=$(date +%s.%N)
     if ! PYTHONPATH="${FRESH_ROOT}" python -m replayer \
-            --endpoint-url "${endpoints}" \
+            --endpoint "${endpoints}" \
-            --trace-path "${TRACE}" \
+            --trace "${TRACE}" \
-            --output-path "${replay_out}" \
+            --output "${replay_out}" \
-            --model-name "${MODEL_NAME}" \
+            --model "${MODEL_NAME}" \
             ${REQUEST_LIMIT_ARG} \
             > "${OUT_ROOT}/${config}_rep${rep}_replay.log" 2>&1; then
-        local t1=$(date +%s.%N)
+        local t1
-        echo "[mb5-run] REPLAY FAILED after $(echo "$t1 - $t0" | bc) s; see ${OUT_ROOT}/${config}_rep${rep}_replay.log"
+        t1=$(date +%s.%N)
+        local wall=$(python -c "print(${t1} - ${t0})")
+        echo "[mb5-run] REPLAY FAILED after ${wall} s; see ${OUT_ROOT}/${config}_rep${rep}_replay.log"
         bash "${LAUNCH}" stop > /dev/null 2>&1 || true
         return 1
     fi
-    local t1=$(date +%s.%N)
+    local t1
-    local wall_clock_s=$(echo "$t1 - $t0" | bc)
+    t1=$(date +%s.%N)
+    local wall_clock_s
+    wall_clock_s=$(python -c "print(${t1} - ${t0})")
     echo "[mb5-run] replay done in ${wall_clock_s}s"
     echo "${wall_clock_s}" > "${rundir}/wall_clock_s.txt"