http://gcc.gnu.org/bugzilla/show_bug.cgi?id=59225
Manuel López-Ibáñez <manu at gcc dot gnu.org> changed:
What |Removed |Added
----------------------------------------------------------------------------
Status|UNCONFIRMED |RESOLVED
CC| |manu at gcc dot gnu.org
Resolution|--- |DUPLICATE
--- Comment #1 from Manuel López-Ibáñez <manu at gcc dot gnu.org> ---
CCP assumes y == 1 in the first testcase, this is PR18501
In the second testcase, y may be either 1 or 2, so CCP does not kick-in, and
nothing removes the undefined operand of the PHI node:
# RANGE [1, 2] NONZERO 0x00000000000000003
# y_1 = PHI <[test2.c : 5:7] 1(6), y_3(D)(4), [test2.c : 7:7] 2(7)>
[test2.c : 8:3] # VUSE <.MEM_4(D)>
return y_1;
so we can warn.
*** This bug has been marked as a duplicate of bug 18501 ***
