[Bug middle-end/69845] [4.9/5/6 Regression] Expression getting incorrectly optimized after being rewritten by compiler

[jakub at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=69845

--- Comment #6 from Jakub Jelinek <jakub at gcc dot gnu.org> ---
I believe the bug is in extract_muldiv_1, what it does for PLUS_EXPR/MINUS_EXPR
looks inherently unsafe, unless it changes the operation to be actually
performed in a type with defined overflow and then cast back.