[Bug target/88234] UBsan and runtime error: signed integer overflow using unsigned vector

[jakub at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=88234

--- Comment #16 from Jakub Jelinek <jakub at gcc dot gnu.org> ---
Author: jakub
Date: Tue Jan  8 10:03:34 2019
New Revision: 267692

URL: https://gcc.gnu.org/viewcvs?rev=267692&root=gcc&view=rev
Log:
        Backported from mainline
        2018-11-29  Jakub Jelinek  <ja...@redhat.com>

        PR target/88234
        * config/rs6000/rs6000.c (rs6000_gimple_fold_builtin): For
        vec_add and vec_sub builtins, perform PLUS_EXPR or MINUS_EXPR
        in unsigned_type_for instead of vector integral type where overflow
        doesn't wrap.

        * gcc.dg/ubsan/pr88234.c: New test.

Added:
    branches/gcc-8-branch/gcc/testsuite/gcc.dg/ubsan/pr88234.c
Modified:
    branches/gcc-8-branch/gcc/ChangeLog
    branches/gcc-8-branch/gcc/config/rs6000/rs6000.c
    branches/gcc-8-branch/gcc/testsuite/ChangeLog