Skip to content
GitLab
  1. 26 Jul, 2017 1 commit
    • Dominik Bartkiewicz's avatar
      Fix regression in commit e5c05549 that would put the stepd pid into the... · f28b1a97
      Dominik Bartkiewicz authored 
      Fix regression in commit e5c05549 that would put the stepd pid into the memory cgroup instead of the task's pid.

Beforehand this would put the result of getpid() into the cgroup.  Before
e5c05549 this was done in the child of the fork which would get you
the task's pid, but moving it to run in the parent broke this logic.

What this patch does is adds pid to the input parameters of
task_g_pre_launch_priv making it so we could use the correct pid.
      f28b1a97
  3. 25 Jul, 2017 1 commit
  5. 24 Jul, 2017 3 commits
  7. 21 Jul, 2017 3 commits
  9. 19 Jul, 2017 4 commits
  11. 18 Jul, 2017 1 commit
    • Dominik Bartkiewicz's avatar
      Fix issue with multiple jobs from an array to start. · b40bd8d3
      Dominik Bartkiewicz authored 
      By removing the real locks we can get into a race condition where the prolog
starts and finishes before we get here and then we end up waiting forever.

Making the mutex a static seemed to help in many cases, but didn't
completely close the window.  Changing slurm_cond_wait to
slurm_cond_timedwait fixed the scenario where we would hit the window, but
not degrade performance the original commit provides.

There were also spots where if the job or step didn't exist it wouldn't
signal the conditional also providing a spot this could get stuck not
starting the job.

Fix regression from commit 52ce3ff0

Bug 3977
      b40bd8d3
  13. 14 Jul, 2017 3 commits
    • Tim Shaw's avatar
      Fix example code to actually work. · 5733505a
      Tim Shaw authored 
      Code provided by Ole Nielsen <Ole.H.Nielsen@fysik.dtu.dk>

Bug 3985
      5733505a
    • Danny Auble's avatar
      Fix whitespace, no code change. · dc6f910b
      Danny Auble authored
      dc6f910b
    • Danny Auble's avatar
      Fix issue with whole gres not being printed out with Slurm tools. · 028bf3e1
      Danny Auble authored 
      This is a regression from commit fec995e0.

It turns out using tok here was erroneous for situations where the gres had
a type and name and potentially a count (i.e. network:gigabit:1)

_get_gres_req_cnt() would alter the incoming char *config which is what tok
was.  So when we print it back to the requested string it would only have
what was there to the first ':'.  As we didn't need to \0 out the first char
as we skip over it anyway I just kept track of what the replaced \0 was for
the number portion and put it back when we are done copying it.

Related to bug 3521
      028bf3e1
  15. 13 Jul, 2017 7 commits
  17. 10 Jul, 2017 1 commit
  19. 07 Jul, 2017 5 commits
  21. 06 Jul, 2017 1 commit
  23. 05 Jul, 2017 6 commits
  25. 03 Jul, 2017 2 commits
  27. 30 Jun, 2017 2 commits