GNOME Bugzilla – Bug 552000
Unlikely but possible memleak in gst_pad_push
Last modified: 2008-09-12 18:17:46 UTC
Please describe the problem: In gst_pad_push : g_return_val_if_fail (GST_IS_PAD (pad), GST_FLOW_ERROR); g_return_val_if_fail (GST_PAD_DIRECTION (pad) == GST_PAD_SRC, GST_FLOW_ERROR); these can return without unreffing the input buffer , thus causing a leak Steps to reproduce: Actual results: Expected results: Does this happen every time? Other information: See the patch below to fix
Created attachment 118605 [details] [review] Patch to fix the possible leak Original patch by Yves Lefebvre
These checks are just there to catch programming errors as early as possible, we don't care if things leak in this case. (Also, the g_return_* statements are macros which can be turned into no-ops for production use on systems where binary size or the overhead caused by these runtime checks is an issue)
Created attachment 118608 [details] [review] Patch to fix the leak for 0.10.20 last patch was a mistake please discard
Antoine, this bug wont be fixed as tim pointed out.
Edward: Relax I was upping the patch while Tim was commenting so Tim comment wasn't there when I sent the second patch... I don't mind not fixing it..