Use an intermediate C launcher in startup benchmark

Author: msimacek

graalpython/com.oracle.graal.python.benchmarks/python/micro/startup.py

@@ -36,26 +36,84 @@
 # LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
 # OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
 # SOFTWARE.
+import os
+import subprocess
+import sys
+import tempfile
+from pathlib import Path
 
+# Use a C runner to spawn the subprocesses to avoid counting subprocess module overhead into the benchmark
+RUNNER_CODE = '''
+#include <stdio.h>
+#include <stdlib.h>
+#include <unistd.h>
+#include <sys/wait.h>
 
-import subprocess, sys
+int main(int argc, char *argv[]) {
+    if (argc < 3) {
+        return 1;
+    }
+    int n = atoi(argv[1]);
+    if (n <= 0) {
+        return 1;
+    }
+    char **cmd_argv = &argv[2];
+    for (int i = 0; i < n; ++i) {
+        pid_t pid = fork();
+        if (pid < 0) {
+            perror("fork");
+            return 1;
+        } else if (pid == 0) {
+            execvp(cmd_argv[0], cmd_argv);
+            perror("execvp");
+            exit(127);  // If exec fails
+        } else {
+            int status;
+            if (waitpid(pid, &status, 0) < 0) {
+                perror("waitpid");
+                return 1;
+            }
+        }
+    }
+    return 0;
+}
+'''
+
+TMPDIR = tempfile.TemporaryDirectory()
+RUNNER_EXE = None
+ORIG_ARGV = None
+
+
+def __setup__(*args):
+    global RUNNER_EXE
+    tmpdir = Path(TMPDIR.name)
+    runner_c = tmpdir / 'runner.c'
+    runner_c.write_text(RUNNER_CODE)
+    RUNNER_EXE = tmpdir / 'runner'
+    subprocess.check_call([os.environ.get('CC', 'gcc'), runner_c, '-O2', '-o', RUNNER_EXE])
+
+    global ORIG_ARGV
+    ORIG_ARGV = sys.orig_argv
+    for i, arg in enumerate(ORIG_ARGV):
+        if arg.endswith('.py'):
+            ORIG_ARGV = ORIG_ARGV[:i]
+            break
+
+
+def __teardown__():
+    TMPDIR.cleanup()
 
 
 def __benchmark__(num=1000000):
-    while num > 0:
-        num -= 1
-        orig_vm_argv = sys.orig_argv
-        for i, arg in enumerate(orig_vm_argv):
-            if arg.endswith('.py'):
-                orig_vm_argv = orig_vm_argv[:i]
-                break
-        subprocess.check_call([
-            *orig_vm_argv,
-            "-I", # isolate from environment
-            "-S", # do not import site
-            "-Wignore", # ignore all warnings
-            "-B", # do not attempt to write pyc files
-            "-u", # do not add buffering wrappers around output streams
-            "-c",
-            "1"
-        ])
+    subprocess.check_call([
+        str(RUNNER_EXE),
+        str(num),
+        *ORIG_ARGV,
+        "-I",  # isolate from environment
+        "-S",  # do not import site
+        "-Wignore", # ignore all warnings
+        "-B",  # do not attempt to write pyc files
+        "-u",  # do not add buffering wrappers around output streams
+        "-c",
+        "1"
+    ])