COSMO VS. CYGWIN SPAWNING COMPARISON

The following is a no-bull assessment (Oct 2023) of where Cosmo v3
currently stands in terms of process creation performance on WIN32
compared to the venerable decades-old Cygwin project.

COSMO BENCHMARK RESULTS

  jtunn@gothbox:/C/Users/jtunn$ cosmo/spawn_bench.com cosmo/life-pe.com
  fork() + exec(tiny)                                      12,490,700 ns
  posix_spawn(tiny)                                         2,524,350 ns  <-- IMPORTANT
  mmap(128mb, MAP_SHARED) + fork() + exec(tiny)            23,118,850 ns
  mmap(128mb, MAP_SHARED) + posix_spawn(tiny)               2,870,390 ns  <-- IMPORTANT
  malloc(128mb) + fork() + exec(tiny)                     154,522,980 ns
  malloc(128mb) + posix_spawn(tiny)                         2,992,780 ns  <-- IMPORTANT

  jtunn@gothbox:/C/Users/jtunn$ uname -a
  Windows gothbox 10.0-19045 Cosmopolitan 3.0-alpha x86_64 GNU/Linux

CYGWIN BENCHMARK RESULTS

  jtunn@gothbox:/cygdrive/c/Users/jtunn/scratch$ ./spawn_bench life-pe.com
  fork() + exec(tiny)                                      10,883,920 ns
  posix_spawn(tiny)                                         9,723,830 ns  <-- IMPORTANT
  spawnvp(tiny)                                               557,700 ns
  mmap(128mb, MAP_SHARED) + fork() + exec(tiny)            15,703,750 ns
  mmap(128mb, MAP_SHARED) + posix_spawn(tiny)              16,344,610 ns  <-- IMPORTANT
  mmap(128mb, MAP_SHARED) + spawnvp(tiny)                     580,250 ns
  malloc(128mb) + fork() + exec(tiny)                      51,615,770 ns
  malloc(128mb) + posix_spawn(tiny)                        50,878,060 ns  <-- IMPORTANT
  malloc(128mb) + spawnvp(tiny)                               815,710 ns  <-- cosmo wants this

  jtunn@gothbox:/cygdrive/c/Users/jtunn/scratch$ uname -a
  CYGWIN_NT-10.0-19045 gothbox 3.4.6-1.x86_64 2023-02-14 13:23 UTC x86_64 Cygwin

CONCLUSION

  Cygwin has a nice Cygwin-specific spawnvp() API, but they didn't put
  much effort into their posix_spawn() implementation, which is likely
  just a naive wrapper around their fork() implementation. Cygwin also
  copies memory into forked child processes faster than Cosmo is doing
  currently (but for how much longer?)

COSMO SPAWN_BENCH.C PROGRAM

  #include <time.h>
  #include <spawn.h>
  #include <stdio.h>
  #include <assert.h>
  #include <string.h>
  #include <stdint.h>
  #include <stdlib.h>
  #include <unistd.h>
  #include <limits.h>
  #include <stdbool.h>
  #include <sys/wait.h>
  #include <sys/mman.h>
  #include <stdatomic.h>
  
  #define ITERATIONS 10
  
  _Alignas(128) int a;
  _Alignas(128) int b;
  _Alignas(128) atomic_int lock;
  
  static struct timespec SubtractTime(struct timespec a,
                                      struct timespec b) {
    a.tv_sec -= b.tv_sec;
    if (a.tv_nsec < b.tv_nsec) {
      a.tv_nsec += 1000000000;
      a.tv_sec--;
    }
    a.tv_nsec -= b.tv_nsec;
    return a;
  }
  
  static time_t ToNanoseconds(struct timespec ts) {
    return ts.tv_sec * 1000000000 + ts.tv_nsec;
  }
  
  static char *Ithoa(char p[27], unsigned long x) {
    char m[26];
    unsigned i;
    i = 0;
    do {
      m[i++] = x % 10 + '0';
      x = x / 10;
    } while (x);
    for (;;) {
      *p++ = m[--i];
      if (!i) break;
      if (!(i % 3)) *p++ = ',';
    }
    *p = '\0';
    return p;
  }
  
  #define BENCH(name, x)                                                   \
    {                                                                     \
      int i;                                                              \
      char ibuf[27];                                                      \
      struct timespec t1, t2;                                             \
      clock_gettime(CLOCK_REALTIME, &t1);                                 \
      for (i = 0; i < ITERATIONS; ++i) {                                  \
        x;                                                                \
      }                                                                   \
      clock_gettime(CLOCK_REALTIME, &t2);                                 \
      Ithoa(ibuf, ToNanoseconds(SubtractTime(t2, t1))/ITERATIONS);        \
      printf("%-50s %16s ns\n", name, ibuf);                                     \
    }
  
  void PosixSpawnWait(const char *prog) {
    int ws, err, pid;
    char *args[] = {(char *)prog, 0};
    char *envs[] = {0};
    if ((err = posix_spawn(&pid, prog, NULL, NULL, args, envs))) {
      perror(prog);
      exit(1);
    }
    if (waitpid(pid, &ws, 0) == -1) {
      perror("waitpid");
      exit(1);
    }
    if (!(WIFEXITED(ws) && WEXITSTATUS(ws) == 42)) {
      puts("bad exit status");
      exit(1);
    }
  }
  
  void ForkExecWait(const char *prog) {
    int ws;
    if (!fork()) {
      execve(prog, (char *[]){(char *)prog, 0}, (char *[]){0});
      _Exit(127);
    }
    if (wait(&ws) == -1) {
      perror("wait");
      exit(1);
    }
    if (!(WIFEXITED(ws) && WEXITSTATUS(ws) == 42)) {
      puts("bad exit status");
      exit(1);
    }
  }
  
  char *FillMemory(char *p, long n) {
    return memset(p, -1, n);
  }
  char *(*pFillMemory)(char *, long) = FillMemory;
  
  int main(int argc, char *argv[]) {
    long n;
    void *p;
    const char *prog;
  
    if (argc <= 1) {
      prog = "tiny64";
    } else {
      prog = argv[1];
    }
  
    BENCH("fork() + exec(tiny)", ForkExecWait(prog));
    BENCH("posix_spawn(tiny)", PosixSpawnWait(prog));
  
    n = 128L * 1024 * 1024;
    p = pFillMemory(
        mmap(0, n, PROT_READ | PROT_WRITE, MAP_SHARED | MAP_ANONYMOUS, -1, 0), n);
    BENCH("mmap(128mb, MAP_SHARED) + fork() + exec(tiny)", ForkExecWait(prog));
    BENCH("mmap(128mb, MAP_SHARED) + posix_spawn(tiny)", PosixSpawnWait(prog));
    munmap(p, n);
  
    n = 128L * 1024 * 1024;
    p = pFillMemory(malloc(n), n);
    BENCH("malloc(128mb) + fork() + exec(tiny)", ForkExecWait(prog));
    BENCH("malloc(128mb) + posix_spawn(tiny)", PosixSpawnWait(prog));
    free(p);
  }

CYGWIN SPAWN_BENCH.C PROGRAM

  #include <time.h>
  #include <spawn.h>
  #include <stdio.h>
  #include <assert.h>
  #include <string.h>
  #include <stdint.h>
  #include <stdlib.h>
  #include <unistd.h>
  #include <limits.h>
  #include <process.h>
  #include <stdbool.h>
  #include <sys/wait.h>
  #include <sys/mman.h>
  #include <stdatomic.h>
  
  #define ITERATIONS 10
  
  _Alignas(128) int a;
  _Alignas(128) int b;
  _Alignas(128) atomic_int lock;
  
  static struct timespec SubtractTime(struct timespec a,
                                      struct timespec b) {
    a.tv_sec -= b.tv_sec;
    if (a.tv_nsec < b.tv_nsec) {
      a.tv_nsec += 1000000000;
      a.tv_sec--;
    }
    a.tv_nsec -= b.tv_nsec;
    return a;
  }
  
  static time_t ToNanoseconds(struct timespec ts) {
    return ts.tv_sec * 1000000000 + ts.tv_nsec;
  }
  
  static char *Ithoa(char p[27], unsigned long x) {
    char m[26];
    unsigned i;
    i = 0;
    do {
      m[i++] = x % 10 + '0';
      x = x / 10;
    } while (x);
    for (;;) {
      *p++ = m[--i];
      if (!i) break;
      if (!(i % 3)) *p++ = ',';
    }
    *p = '\0';
    return p;
  }
  
  #define BENCH(name, x)                                                   \
    {                                                                     \
      int i;                                                              \
      char ibuf[27];                                                      \
      struct timespec t1, t2;                                             \
      clock_gettime(CLOCK_REALTIME, &t1);                                 \
      for (i = 0; i < ITERATIONS; ++i) {                                  \
        x;                                                                \
      }                                                                   \
      clock_gettime(CLOCK_REALTIME, &t2);                                 \
      Ithoa(ibuf, ToNanoseconds(SubtractTime(t2, t1))/ITERATIONS);        \
      printf("%-50s %16s ns\n", name, ibuf);                                     \
    }
  
  void PosixSpawnWait(const char *prog) {
    int ws, err, pid;
    char *args[] = {(char *)prog, 0};
    char *envs[] = {0};
    if ((err = posix_spawn(&pid, prog, NULL, NULL, args, envs))) {
      perror(prog);
      exit(1);
    }
    if (waitpid(pid, &ws, 0) == -1) {
      perror("waitpid");
      exit(1);
    }
    if (!(WIFEXITED(ws) && WEXITSTATUS(ws) == 42)) {
      puts("bad exit status");
      exit(1);
    }
  }
  
  void ForkExecWait(const char *prog) {
    int ws;
    if (!fork()) {
      execve(prog, (char *[]){(char *)prog, 0}, (char *[]){0});
      _Exit(127);
    }
    if (wait(&ws) == -1) {
      perror("wait");
      exit(1);
    }
    if (!(WIFEXITED(ws) && WEXITSTATUS(ws) == 42)) {
      puts("bad exit status");
      exit(1);
    }
  }
  
  void SpawnvpWait(const char *prog) {
    spawnvp(_P_WAIT, prog, (char *[]){(char *)prog, 0});
  }
  
  char *FillMemory(char *p, long n) {
    return memset(p, -1, n);
  }
  char *(*pFillMemory)(char *, long) = FillMemory;
  
  int main(int argc, char *argv[]) {
    long n;
    void *p;
    const char *prog;
  
    if (argc <= 1) {
      prog = "tiny64";
    } else {
      prog = argv[1];
    }
  
    BENCH("fork() + exec(tiny)", ForkExecWait(prog));
    BENCH("posix_spawn(tiny)", PosixSpawnWait(prog));
    BENCH("spawnvp(tiny)", SpawnvpWait(prog));
  
    n = 128L * 1024 * 1024;
    p = pFillMemory(
        mmap(0, n, PROT_READ | PROT_WRITE, MAP_SHARED | MAP_ANONYMOUS, -1, 0), n);
    BENCH("mmap(128mb, MAP_SHARED) + fork() + exec(tiny)", ForkExecWait(prog));
    BENCH("mmap(128mb, MAP_SHARED) + posix_spawn(tiny)", PosixSpawnWait(prog));
    BENCH("mmap(128mb, MAP_SHARED) + spawnvp(tiny)", SpawnvpWait(prog));
    munmap(p, n);
  
    n = 128L * 1024 * 1024;
    p = pFillMemory(malloc(n), n);
    BENCH("malloc(128mb) + fork() + exec(tiny)", ForkExecWait(prog));
    BENCH("malloc(128mb) + posix_spawn(tiny)", PosixSpawnWait(prog));
    BENCH("malloc(128mb) + spawnvp(tiny)", SpawnvpWait(prog));
    free(p);
  }

OUR TINIEST PORTABLE EXECUTABLE (COSMOPOLITAN/TOOL/HELLO/LIFE-PE.C)

  // $ cd cosmopolitan
  // $ make o//tool/hello/life-pe.com
  // ld -o life-pe.elf -q -e WinMain
  // elf2pe -o life-pe.com life-pe.elf
  __attribute__((__ms_abi__)) long WinMain(void) {
    return 42 << 8;
  }

  00000000  4d 5a 00 00 00 00 00 00  00 00 00 00 00 00 00 00  |MZ..............|
  00000010  00 00 00 00 00 00 00 00  00 00 00 00 00 00 00 00  |................|
  00000020  00 00 00 00 00 00 00 00  00 00 00 00 00 00 00 00  |................|
  00000030  00 00 00 00 00 00 00 00  00 00 00 00 40 00 00 00  |............@...|
  00000040  50 45 00 00 64 86 02 00  d8 73 bc 64 00 00 00 00  |PE..d....s.d....|
  00000050  00 00 00 00 70 00 2f 00  0b 02 0e 23 00 02 00 00  |....p./....#....|
  00000060  00 02 00 00 00 00 00 00  10 10 00 00 00 10 00 00  |................|
  00000070  00 00 40 00 00 00 00 00  00 10 00 00 00 02 00 00  |..@.............|
  00000080  06 00 00 00 00 00 00 00  06 00 00 00 00 00 00 00  |................|
  00000090  00 30 00 00 00 02 00 00  00 00 00 00 03 00 20 01  |.0............ .|
  000000a0  00 00 80 00 00 00 00 00  00 00 01 00 00 00 00 00  |................|
  000000b0  00 00 00 00 00 00 00 00  00 00 00 00 00 00 00 00  |................|
  000000c0  00 00 00 00 00 00 00 00  2e 74 65 78 74 00 00 00  |.........text...|
  000000d0  18 00 00 00 00 10 00 00  00 02 00 00 00 02 00 00  |................|
  000000e0  00 00 00 00 00 00 00 00  00 00 00 00 20 00 00 60  |............ ..`|
  000000f0  2e 64 61 74 61 00 00 00  08 00 00 00 00 20 00 00  |.data........ ..|
  00000100  00 02 00 00 00 04 00 00  00 00 00 00 00 00 00 00  |................|
  00000110  00 00 00 00 40 00 00 c0  00 00 00 00 00 00 00 00  |....@...........|
  00000120  00 00 00 00 00 00 00 00  00 00 00 00 00 00 00 00  |................|
  00000130  00 00 00 00 00 00 00 00  00 00 00 00 00 00 00 00  |................|
  00000140  00 00 00 00 00 00 00 00  00 00 00 00 00 00 00 00  |................|
  00000150  00 00 00 00 00 00 00 00  00 00 00 00 00 00 00 00  |................|
  00000160  00 00 00 00 00 00 00 00  00 00 00 00 00 00 00 00  |................|
  00000170  00 00 00 00 00 00 00 00  00 00 00 00 00 00 00 00  |................|
  00000180  00 00 00 00 00 00 00 00  00 00 00 00 00 00 00 00  |................|
  00000190  00 00 00 00 00 00 00 00  00 00 00 00 00 00 00 00  |................|
  000001a0  00 00 00 00 00 00 00 00  00 00 00 00 00 00 00 00  |................|
  000001b0  00 00 00 00 00 00 00 00  00 00 00 00 00 00 00 00  |................|
  000001c0  00 00 00 00 00 00 00 00  00 00 00 00 00 00 00 00  |................|
  000001d0  00 00 00 00 00 00 00 00  00 00 00 00 00 00 00 00  |................|
  000001e0  00 00 00 00 00 00 00 00  00 00 00 00 00 00 00 00  |................|
  000001f0  00 00 00 00 00 00 00 00  00 00 00 00 00 00 00 00  |................|
  00000200  90 90 90 90 90 90 90 90  90 90 90 90 90 90 90 90  |................|
  00000210  66 90 b8 2a 00 00 00 c3  00 00 00 00 00 00 00 00  |f..*............|
  00000220  00 00 00 00 00 00 00 00  00 00 00 00 00 00 00 00  |................|
  00000230  00 00 00 00 00 00 00 00  00 00 00 00 00 00 00 00  |................|
  00000240  00 00 00 00 00 00 00 00  00 00 00 00 00 00 00 00  |................|
  00000250  00 00 00 00 00 00 00 00  00 00 00 00 00 00 00 00  |................|
  00000260  00 00 00 00 00 00 00 00  00 00 00 00 00 00 00 00  |................|
  00000270  00 00 00 00 00 00 00 00  00 00 00 00 00 00 00 00  |................|
  00000280  00 00 00 00 00 00 00 00  00 00 00 00 00 00 00 00  |................|
  00000290  00 00 00 00 00 00 00 00  00 00 00 00 00 00 00 00  |................|
  000002a0  00 00 00 00 00 00 00 00  00 00 00 00 00 00 00 00  |................|
  000002b0  00 00 00 00 00 00 00 00  00 00 00 00 00 00 00 00  |................|
  000002c0  00 00 00 00 00 00 00 00  00 00 00 00 00 00 00 00  |................|
  000002d0  00 00 00 00 00 00 00 00  00 00 00 00 00 00 00 00  |................|
  000002e0  00 00 00 00 00 00 00 00  00 00 00 00 00 00 00 00  |................|
  000002f0  00 00 00 00 00 00 00 00  00 00 00 00 00 00 00 00  |................|
  00000300  00 00 00 00 00 00 00 00  00 00 00 00 00 00 00 00  |................|
  00000310  00 00 00 00 00 00 00 00  00 00 00 00 00 00 00 00  |................|
  00000320  00 00 00 00 00 00 00 00  00 00 00 00 00 00 00 00  |................|
  00000330  00 00 00 00 00 00 00 00  00 00 00 00 00 00 00 00  |................|
  00000340  00 00 00 00 00 00 00 00  00 00 00 00 00 00 00 00  |................|
  00000350  00 00 00 00 00 00 00 00  00 00 00 00 00 00 00 00  |................|
  00000360  00 00 00 00 00 00 00 00  00 00 00 00 00 00 00 00  |................|
  00000370  00 00 00 00 00 00 00 00  00 00 00 00 00 00 00 00  |................|
  00000380  00 00 00 00 00 00 00 00  00 00 00 00 00 00 00 00  |................|
  00000390  00 00 00 00 00 00 00 00  00 00 00 00 00 00 00 00  |................|
  000003a0  00 00 00 00 00 00 00 00  00 00 00 00 00 00 00 00  |................|
  000003b0  00 00 00 00 00 00 00 00  00 00 00 00 00 00 00 00  |................|
  000003c0  00 00 00 00 00 00 00 00  00 00 00 00 00 00 00 00  |................|
  000003d0  00 00 00 00 00 00 00 00  00 00 00 00 00 00 00 00  |................|
  000003e0  00 00 00 00 00 00 00 00  00 00 00 00 00 00 00 00  |................|
  000003f0  00 00 00 00 00 00 00 00  00 00 00 00 00 00 00 00  |................|
  00000400  00 10 40 00 00 00 00 00  00 00 00 00 00 00 00 00  |..@.............|
  00000410  00 00 00 00 00 00 00 00  00 00 00 00 00 00 00 00  |................|
  00000420  00 00 00 00 00 00 00 00  00 00 00 00 00 00 00 00  |................|
  00000430  00 00 00 00 00 00 00 00  00 00 00 00 00 00 00 00  |................|
  00000440  00 00 00 00 00 00 00 00  00 00 00 00 00 00 00 00  |................|
  00000450  00 00 00 00 00 00 00 00  00 00 00 00 00 00 00 00  |................|
  00000460  00 00 00 00 00 00 00 00  00 00 00 00 00 00 00 00  |................|
  00000470  00 00 00 00 00 00 00 00  00 00 00 00 00 00 00 00  |................|
  00000480  00 00 00 00 00 00 00 00  00 00 00 00 00 00 00 00  |................|
  00000490  00 00 00 00 00 00 00 00  00 00 00 00 00 00 00 00  |................|
  000004a0  00 00 00 00 00 00 00 00  00 00 00 00 00 00 00 00  |................|
  000004b0  00 00 00 00 00 00 00 00  00 00 00 00 00 00 00 00  |................|
  000004c0  00 00 00 00 00 00 00 00  00 00 00 00 00 00 00 00  |................|
  000004d0  00 00 00 00 00 00 00 00  00 00 00 00 00 00 00 00  |................|
  000004e0  00 00 00 00 00 00 00 00  00 00 00 00 00 00 00 00  |................|
  000004f0  00 00 00 00 00 00 00 00  00 00 00 00 00 00 00 00  |................|
  00000500  00 00 00 00 00 00 00 00  00 00 00 00 00 00 00 00  |................|
  00000510  00 00 00 00 00 00 00 00  00 00 00 00 00 00 00 00  |................|
  00000520  00 00 00 00 00 00 00 00  00 00 00 00 00 00 00 00  |................|
  00000530  00 00 00 00 00 00 00 00  00 00 00 00 00 00 00 00  |................|
  00000540  00 00 00 00 00 00 00 00  00 00 00 00 00 00 00 00  |................|
  00000550  00 00 00 00 00 00 00 00  00 00 00 00 00 00 00 00  |................|
  00000560  00 00 00 00 00 00 00 00  00 00 00 00 00 00 00 00  |................|
  00000570  00 00 00 00 00 00 00 00  00 00 00 00 00 00 00 00  |................|
  00000580  00 00 00 00 00 00 00 00  00 00 00 00 00 00 00 00  |................|
  00000590  00 00 00 00 00 00 00 00  00 00 00 00 00 00 00 00  |................|
  000005a0  00 00 00 00 00 00 00 00  00 00 00 00 00 00 00 00  |................|
  000005b0  00 00 00 00 00 00 00 00  00 00 00 00 00 00 00 00  |................|
  000005c0  00 00 00 00 00 00 00 00  00 00 00 00 00 00 00 00  |................|
  000005d0  00 00 00 00 00 00 00 00  00 00 00 00 00 00 00 00  |................|
  000005e0  00 00 00 00 00 00 00 00  00 00 00 00 00 00 00 00  |................|
  000005f0  00 00 00 00 00 00 00 00  00 00 00 00 00 00 00 00  |................|
  00000600