On Sat, Sep 10, 2011 at 11:47 AM, Matt Rice wrote: > > I could easily be missing something though and it's fine. > ^ correct answer... it's handled in tui_setup_io(). the attached patch based on Pedro's WIP, fixes a small case that he missed. and makes interp.exp hit all of these failures this is PR gdb/13175 (though Nick's reproducer hits the case you missed). I modified the existing prompt checks in this file, because the gdb_expect was causing the test to take 30 seconds to complete, which is now down to 0 seconds... hopefully that change is ok... with this Pedro's patch runs through the testsuite no new failures... feel free to merge in with your patch or have me commit the tests seperately, whatever is easiest for you really... 2011-09-11 Matt Rice * mi/mi-interp.c (mi_interpreter_init): Initialize default MI version to 2. (_initialize_mi_interp): Move comment to mi_interpreter_init. 2011-09-11 Matt Rice PR gdb/13175 * gdb.base/interp.exp: New tests. * gdb.base/interp.c: New file.