1. 14 Jan, 2015 2 commits
  2. 12 Jan, 2015 1 commit
  3. 05 Jan, 2015 2 commits
  4. 02 Jan, 2015 5 commits
  5. 31 Dec, 2014 1 commit
  6. 29 Dec, 2014 1 commit
  7. 27 Dec, 2014 4 commits
  8. 19 Dec, 2014 2 commits
  9. 18 Dec, 2014 2 commits
  10. 17 Dec, 2014 1 commit
  11. 15 Dec, 2014 1 commit
  12. 14 Dec, 2014 1 commit
  13. 12 Dec, 2014 1 commit
  14. 09 Dec, 2014 2 commits
  15. 08 Dec, 2014 1 commit
  16. 07 Dec, 2014 2 commits
  17. 06 Dec, 2014 1 commit
  18. 05 Dec, 2014 1 commit
  19. 29 Nov, 2014 5 commits
  20. 28 Nov, 2014 1 commit
  21. 27 Nov, 2014 3 commits
    • Lukasz Marek's avatar
      lavu/opt: add consts where possible · f00e9c4b
      Lukasz Marek authored
      f00e9c4b
    • Stefano Sabatini's avatar
    • wm4's avatar
      lavu: fix memory leaks by using a mutex instead of atomics · fbd6c97f
      wm4 authored
      The buffer pool has to atomically add and remove entries from the linked
      list of available buffers. This was done by removing the entire list
      with a CAS operation, working on it, and then setting it back again
      (using a retry-loop in case another thread was doing the same thing).
      
      This could effectively cause memory leaks: while a thread was working on
      the buffer list, other threads would allocate new buffers, increasing
      the pool's total size. There was no real leak, but since these extra
      buffers were not needed, but not free'd either (except when the buffer
      pool was destroyed), this had the same effects as a real leak. For some
      reason, growth was exponential, and could easily kill the process due
      to OOM in real-world uses.
      
      Fix this by using a mutex to protect the list operations. The fancy
      way atomics remove the whole list to work on it is not needed anymore,
      which also avoids the situation which was causing the leak.
      Signed-off-by: 's avatarAnton Khirnov <anton@khirnov.net>
      fbd6c97f