https://gcc.gnu.org/bugzilla/show_bug.cgi?id=126130

--- Comment #4 from Filip Kastl <pheeck at gcc dot gnu.org> ---
I still have to wrap my head around what Richard wrote.  But I want to make a
note here:  I wrote

> While the constraints themselves make sense, we should ensure that
> solution_set_expand() doesn't expand Open into .clobber, .use and .result
> (perhaps by somehow specialcasing these kinds of subvariables).

but I thought about it some more and looked again into pta-andersen.cc.  The
function solution_set_expand() maybe isn't the right place to make a
modification.  That function also gets used while handling x = *y + UNKNOWN
constraints (in do_sd_constraint(), I believe) and my idea was to only modify
how GCC handles x = y + UNKNOWN constraints (that's set_union_with_increment(),
I think).

Reply via email to