r255510 triggers this:
spawn gdb -nx -nw -quiet -batch -x pr54970.gdb ./pr54970.exe
XPASS: gcc.dg/guality/pr54970.c -O0 line 31 a == 4
-with-system-zlib --with-demangler-in-ld --with-fpmath=sse --enable-shared --enable-host-shared --enable-clocale=gnu --enable-cloog-backend=isl --enable-languages=c,c++,fortran,jit,lto -with-arch=haswell --with-cpu=haswell
Was done intentionally.
Wait, isn`t XPASS an unexpected (i.e. erroneous) pass?
I know that XFAIL = expected fail, so I don`t consider it a fail. But what about XPASS?
Not necessarily errorneous.
In this particular case, we simply lack the ability to tell deja-gnu to expect this to pass at -O0 but expect it to fail at any higher level. We have consciously decided to cope with XPASSes (as opposed to coping with FAILs or duplicating the testcase).