[Bug tree-optimization/126470] [15/16/17 Regression] Wrong operand check on a != 0 ? a / b : 0 -> a / b iff b is nonzero fold since r15-307
"sjames at gcc dot gnu.org via Gcc-bugs" <[email protected]>
| Newsgroups | gmane.comp.gcc.bugs |
|---|---|
| Message-ID | <[email protected]/bugzilla/> |
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=126470
Sam James <sjames at gcc dot gnu.org> changed:
What |Removed |Added
----------------------------------------------------------------------------
Summary|[15/16/17 Regression] Wrong |[15/16/17 Regression] Wrong
|operand check on a != 0 ? a |operand check on a != 0 ? a
|/ b : 0 -> a / b iff b is |/ b : 0 -> a / b iff b is
|nonzero fold |nonzero fold since r15-307
See Also| |https://gcc.gnu.org/bugzill
| |a/show_bug.cgi?id=114894,
| |https://gcc.gnu.org/bugzill
| |a/show_bug.cgi?id=116772
CC| |pinskia at gcc dot gnu.org
Keywords|needs-bisection |
--- Comment #1 from Sam James <sjames at gcc dot gnu.org> ---
so incomplete fix in r15-3870-g6c5543d3d9c4bb, pattern itself is from
r15-307-ge472527c7b45d2