This is the mail archive of the
gcc-bugs@gcc.gnu.org
mailing list for the GCC project.
[Bug bootstrap/45700] [4.5/4.6 Regression] --enable-checking=fold bootstrap failures
- From: "aldyh at gcc dot gnu.org" <gcc-bugzilla at gcc dot gnu dot org>
- To: gcc-bugs at gcc dot gnu dot org
- Date: Thu, 30 Sep 2010 17:47:27 +0000
- Subject: [Bug bootstrap/45700] [4.5/4.6 Regression] --enable-checking=fold bootstrap failures
- Auto-submitted: auto-generated
- References: <bug-45700-4@http.gcc.gnu.org/bugzilla/>
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=45700
--- Comment #1 from Aldy Hernandez <aldyh at gcc dot gnu.org> 2010-09-30 17:47:19 UTC ---
If we call build[1-5] just to call protected_set_expr_location next, then by
all means, use build[1-5]_loc and be done with it.
Ideally we should strive to set location information if we know it, so I think
Jakub's second approach of setting the locus if different (with a copy_node) is
a good one. However, if memory usage goes up too much, perhaps we should
re-think this, or ignore the locus as suggested.