https://gcc.gnu.org/bugzilla/show_bug.cgi?id=87072
Martin Sebor <msebor at gcc dot gnu.org> changed:
What |Removed |Added
----------------------------------------------------------------------------
Status|UNCONFIRMED |NEW
Known to work| |4.7.4, 8.1.0, 9.0
Keywords| |diagnostic
Last reconfirmed| |2018-08-23
CC| |msebor at gcc dot gnu.org
Ever confirmed|0 |1
Summary|g++6.2.0 false warning: |[6/7 Regression] g++6.2.0
|array subscript is above |false warning: array
|array bounds, with |subscript is above array
|misleading line number |bounds, with misleading
| |line number
Known to fail| |6.4.0, 7.3.0
--- Comment #1 from Martin Sebor <msebor at gcc dot gnu.org> ---
Confirmed with GCC 5, 6 and 7. The warning wasn't issued by GCC 4.7, started
with 4.8, and went away in 8.0 with r255267. I suspect the change is not going
to be backported.
2017-11-30 Richard Biener <rguent...@suse.de>
PR tree-optimization/83202
* tree-ssa-loop-ivcanon.c (try_unroll_loop_completely): Add
allow_peel argument and guard peeling.
(canonicalize_loop_induction_variables): Likewise.
(canonicalize_induction_variables): Pass false.
(tree_unroll_loops_completely_1): Pass unroll_outer to disallow
peeling from cunrolli.
