https://bugs.llvm.org/show_bug.cgi?id=35909
Sanjay Patel <spatel+l...@rotateright.com> changed:
What |Removed |Added
----------------------------------------------------------------------------
Status|NEW |RESOLVED
Resolution|--- |FIXED
--- Comment #8 from Sanjay Patel <spatel+l...@rotateright.com> ---
Should be fixed with:
https://reviews.llvm.org/rL322950
With that change, everything proceeds as before until after inlining. When we
run instcombine after that, we kill the null check because the GEP is now known
to be inbounds:
IC: Visiting: %sub.ptr2.i = getelementptr inbounds %class.B, %class.B* %b,
i64 -1
IC: Visiting: %1 = bitcast %class.B* %sub.ptr2.i to %class.A*
IC: Visiting: %tobool1 = icmp eq %class.A* %1, null
IC: Replacing %tobool1 = icmp eq %class.A* %1, null
with i1 false
--
You are receiving this mail because:
You are on the CC list for the bug._______________________________________________
llvm-bugs mailing list
llvm-bugs@lists.llvm.org
http://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-bugs
