This project is mirrored from https://github.com/haskell/process. Pull mirroring updated .
  1. 13 Sep, 2017 1 commit
  2. 06 Sep, 2017 2 commits
  3. 22 Aug, 2017 1 commit
  4. 16 Aug, 2017 1 commit
  5. 08 Aug, 2017 1 commit
  6. 07 Aug, 2017 1 commit
  7. 03 Aug, 2017 2 commits
  8. 31 Jul, 2017 1 commit
  9. 20 Jul, 2017 2 commits
  10. 22 Jun, 2017 2 commits
  11. 24 May, 2017 1 commit
  12. 22 May, 2017 1 commit
    • jml's avatar
      Export withCreateProcess documentation · 13db2962
      jml authored
      The function is public, but the Haddock documentation isn't being rendered. I think it's a simple matter of removing the `{-` and `-}`, but I haven't tested locally.
      13db2962
  13. 23 Apr, 2017 1 commit
  14. 21 Apr, 2017 1 commit
  15. 30 Mar, 2017 4 commits
  16. 27 Mar, 2017 1 commit
  17. 09 Mar, 2017 1 commit
  18. 22 Feb, 2017 2 commits
  19. 21 Feb, 2017 2 commits
  20. 20 Feb, 2017 1 commit
  21. 17 Feb, 2017 1 commit
  22. 06 Feb, 2017 3 commits
  23. 04 Feb, 2017 2 commits
  24. 03 Feb, 2017 5 commits
    • coopercm's avatar
      Compile for windows · b2360d9f
      coopercm authored
      b2360d9f
    • coopercm's avatar
      Fix waitpid race by adding a lock · d837c95d
      coopercm authored
      d837c95d
    • coopercm's avatar
      Test multithreaded bug · 962d5f14
      coopercm authored
      962d5f14
    • coopercm's avatar
      Revert "Fix bug in multi-threaded waitForProcess" · 4575acb4
      coopercm authored
      This reverts commit d67484fe64a9a42e7daf944a9396b4ec46c060f3.
      4575acb4
    • coopercm's avatar
      Fix bug in multi-threaded waitForProcess · 5b99d451
      coopercm authored
      Previously an exception was being thrown when multiple threads were
      blocking on waitForProcess due to inconsistent handling of the return
      code of `waitpid`:
      
      "If more than one thread is suspended in waitpid() awaiting termination
      of the same process, exactly one thread returns the process status at
      the time of the target child process termination. The other threads
      return -1, with errno set to ECHILD."
      
      `getProcessExitCode` was handling the ECHILD case by returning 1, but
      `waitForProcess` was returning (-1) in all cases. For consistency this
      commit follows the approach in getProcessExitCode, returning 1 to the
      caller of c_waitForProcess if errno is ECHILD, thus avoiding throwing
      an exception in the calling code.
      5b99d451