I have a new build problem since upgrading the other day and getting gcc 6.0:

[gst-1.0] [jan@pepper gstreamer]$ make check
....
  CC       gst/gstinfo.o
gst/gstinfo.c: In function ‘info_ptr_format_printf_extension’:
gst/gstinfo.c:59:26: error: self-comparison always evaluates to true [-Werror=tautological-compare]
   gst_debug_remove_log_function (gst_debug_log_default);
                          ^~
gst/gstinfo.c:134:33: error: self-comparison always evaluates to true [-Werror=tautological-compare]
   gst_debug_add_log_function (gst_debug_log_default, NULL, NULL);
                                 ^~
gst/gstinfo.c: In function ‘info_segment_format_printf_extension’:
gst/gstinfo.c:145:26: error: self-comparison always evaluates to true [-Werror=tautological-compare]
   gst_debug_remove_log_function (gst_debug_log_default);
                          ^~
gst/gstinfo.c:212:33: error: self-comparison always evaluates to true [-Werror=tautological-compare]
   gst_debug_add_log_function (gst_debug_log_default, NULL, NULL);
                                 ^~
gst/gstinfo.c: In function ‘info_log_handler’:
gst/gstinfo.c:222:36: error: self-comparison always evaluates to true [-Werror=tautological-compare]
   removed = gst_debug_remove_log_function (gst_debug_log_default);
                                    ^~
gst/gstinfo.c: In function ‘info_old_printf_extensions’:
gst/gstinfo.c:292:26: error: self-comparison always evaluates to true [-Werror=tautological-compare]
   gst_debug_remove_log_function (gst_debug_log_default);
                          ^~
gst/gstinfo.c:327:33: error: self-comparison always evaluates to true [-Werror=tautological-compare]
   gst_debug_add_log_function (gst_debug_log_default, NULL, NULL);
                                 ^~
cc1: all warnings being treated as errors
Makefile:2466: recipe for target 'gst/gstinfo.o' failed


I don't know why the warning only triggers in the testsuite, but not when building GStreamer itself. It may be related to https://gcc.gnu.org/bugzilla/show_bug.cgi?id=67030 - which seems to imply that macro expansions shouldn't trigger this warning.