[Bug 692691] Dangling pointer when doing allocation query on a flushing queue

[GStreamer (bugzilla.gnome.org)]

https://bugzilla.gnome.org/show_bug.cgi?id=692691
  GStreamer | gstreamer (core) | 1.x

--- Comment #10 from Alexander Schrab <alexas at axis.com> 2013-01-29 13:47:32 UTC ---
Doing ref/unref won't be a good solution, at least not only doing that. Even
though we correctly unref after a while, I get into new problems when the
callee expects the query to be mutable (ref count == 1). From base source:

...
  query = gst_query_new_allocation (caps, TRUE);
  if (!gst_pad_peer_query (basesrc->srcpad, query)) {
    /* not a problem, just debug a little */
    GST_DEBUG_OBJECT (basesrc, "peer ALLOCATION query failed");
  }

  g_assert (bclass->decide_allocation != NULL);
  result = bclass->decide_allocation (basesrc, query);

  GST_DEBUG_OBJECT (basesrc, "ALLOCATION (%d) params: %" GST_PTR_FORMAT,
result,
      query);

  if (!result)
    goto no_decide_allocation;

  /* we got configuration from our peer or the decide_allocation method,
   * parse them */
  if (gst_query_get_n_allocation_params (query) > 0) {
    gst_query_parse_nth_allocation_param (query, 0, &allocator, &params);
  } else {
    allocator = NULL;
    gst_allocation_params_init (&params);
  }
...

Here, even though the query fails, we continue to use the query object. Which
should be fine! But the extra ref in the queue is a problem in this case.

-- 
Configure bugmail: https://bugzilla.gnome.org/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are the QA contact for the bug.
You are the assignee for the bug.