Re^4: PDL and srand puzzle

[etj]

That's really good! I've renamed it to srandom, and will be dev-releasing as 2.089_01 very soon, after I fix the fix for pptest not working on noexec mounts breaking the top-level Makefile.PL.

[marioroy]

Hi, etj

Calling PDL->random is less unique among threads versus rand().

Edit: etj identified a race condition, hence less uniqueness.

use v5.030;
use threads;
use PDL;

BEGIN { $PDL::no_clone_skip_warning = 1; }

for my $id (1..4) {
  threads->create(sub {
    for (1..8000) {
    # my $r = CORE::rand();
      my $r = PDL->random;
      say $r;
    }
  });
}

$_->join for threads->list;

Thread CORE::rand()

$ perl test.pl | LC_ALL=C sort | uniq -c | sort -n | tail
      1 0.9999196288117
      1 0.999939390107826
      1 1.52499260792638e-05
      1 5.77532994405772e-05
      1 6.39378495463916e-05
      1 6.94432865593342e-05
      1 7.17426768090945e-05
      1 8.48674303988162e-05
      1 9.10499814921195e-05
      1 9.32151475616649e-05

Thread PDL->random

$ perl test.pl | LC_ALL=C sort | uniq -c | sort -n | tail
      3 0.578372644755136
      3 0.587725272162442
      3 0.629117050685529
      3 0.665209931957569
      3 0.666012741533792
      3 0.715907985301874
      3 0.780440518879262
      3 0.789289520984441
      3 0.859969220981904
      3 0.975280763021946

[etj]

It's possible this is because srandom's dealing with the number of CPUs available is interacting with the way MCE (or at least this use of it) does multi-threading. PDL->random is supposed to have its srandom done in a "main" POSIX thread, and then random is called any number of times in each POSIX thread, using widely separated bits of "randomness" stored centrally for each thread. Therefore, it is highly recommended that you call the srandom in the main thread before you create any sort of thread, so they won't all call srandom with extremely similar starting conditions, and therefore getting results similar to what you're seeing.

[marioroy]

It's possible this is because srandom's dealing with the number of CPUs available is interacting with the way MCE (or at least this use of it) does multi-threading.

The thread example given in my prior reply is not using MCE.