Case closed.

makefile_and_run.py

@@ -45,8 +45,7 @@ def get_last_n_cores_per_socket(n=32):
     cpu_str = ",".join(segments)
     total = len(segments) * n
     print(f" CPU binding: taskset -c {cpu_str}  ({total} cores, last {n} per socket)")
-    #return f"taskset -c {cpu_str}"
-    return f""
+    return f"taskset -c {cpu_str}" if cpu_str else ""
 
 
 ## CPU core binding: dynamically select the last 32 cores of each socket (64 cores total)
@@ -145,10 +144,11 @@ def _stop_cuda_mps(runtime_env):
 
 def _gpu_runtime_env():
     runtime_env = os.environ.copy()
+    original_env = set(os.environ.keys())
     finite_difference = str(getattr(input_data, "Finite_Diffenence_Method", "4th-order")).strip()
 
     defaults = {
-        "AMSS_EVOLVE_TIMING": "1",
+        "AMSS_EVOLVE_TIMING": "0",
         "AMSS_ESCALAR_STEP_TIMING": "0",
         "AMSS_INTERP_FAST": "1",
         "AMSS_INTERP_GPU": "1",
@@ -200,6 +200,38 @@ def _gpu_runtime_env():
     for key, value in defaults.items():
         runtime_env.setdefault(key, value)
 
+    input_overrides = [
+        "AMSS_EVOLVE_TIMING",
+        "AMSS_ESCALAR_STEP_TIMING",
+        "AMSS_INTERP_FAST",
+        "AMSS_INTERP_GPU",
+        "AMSS_ANALYSIS_MAP_EVERY",
+        "AMSS_CUDA_AWARE_MPI",
+        "AMSS_CUDA_KEEP_RESIDENT_AFTER_STEP",
+        "AMSS_CUDA_KEEP_ALL_LEVELS",
+        "AMSS_CUDA_ESCALAR_KEEP_RESIDENT_AFTER_STEP",
+        "AMSS_CUDA_ESCALAR_KEEP_ALL_LEVELS",
+        "AMSS_CUDA_EM_CACHE_SOURCES",
+        "AMSS_CUDA_EM_ZERO_FASTPATH",
+        "AMSS_EM_ZERO_ANALYSIS_FASTPATH",
+        "AMSS_EM_ZERO_RESIDENT_DOWNLOAD_FASTPATH",
+        "AMSS_CUDA_AMR_HOST_STAGED",
+        "AMSS_CUDA_AMR_RESTRICT_DEVICE",
+        "AMSS_CUDA_AMR_RESTRICT_BATCH",
+        "AMSS_CUDA_DEVICE_SEGMENT_BATCH",
+        "AMSS_CUDA_UNCACHED_DEVICE_BUFFERS",
+        "AMSS_SHELL_FAST_INTERP",
+        "AMSS_SHELL_PARALLEL_INTERP",
+        "AMSS_SHELL_CUDA_INTERP",
+        "AMSS_SHELL_INTERP_THREADS",
+        "AMSS_Z4C_CUDA_RESIDENT",
+        "AMSS_CONSTRAINT_OUT_EVERY",
+        "AMSS_Z4C_MRBD",
+    ]
+    for env_name in input_overrides:
+        if env_name not in original_env and hasattr(input_data, env_name):
+            runtime_env[env_name] = str(getattr(input_data, env_name))
+
     passthrough_envs = [
         "AMSS_CUDA_RESIDENT_SYNC",
         "AMSS_CUDA_BSSN_RESIDENT_SYNC",
@@ -410,7 +442,6 @@ def run_ABE():
             for line in mpi_process.stdout:
                 print(line, end='')  # stream output in real time
                 file0.write(line)    # write the line to file
-                file0.flush()        # flush to ensure each line is written immediately (optional)
 
         ## Wait for the process to finish
         mpi_return_code = mpi_process.wait()
@@ -454,8 +485,6 @@ def run_TwoPunctureABE():
         for line in TwoPuncture_process.stdout:
             print(line, end='')  # stream output in real time
             file0.write(line)    # write the line to file
-            file0.flush()        # flush to ensure each line is written immediately (optional)                 
-    file0.close()
 
     ## Wait for the process to finish
     TwoPuncture_command_return_code = TwoPuncture_process.wait()