This is the mail archive of the
gcc-bugs@gcc.gnu.org
mailing list for the GCC project.
[Bug tree-optimization/81633] [7/8 Regression] Incorrect floating point result with tree vectoriser
- From: "jakub at gcc dot gnu.org" <gcc-bugzilla at gcc dot gnu dot org>
- To: gcc-bugs at gcc dot gnu dot org
- Date: Mon, 31 Jul 2017 16:47:48 +0000
- Subject: [Bug tree-optimization/81633] [7/8 Regression] Incorrect floating point result with tree vectoriser
- Auto-submitted: auto-generated
- References: <bug-81633-4@http.gcc.gnu.org/bugzilla/>
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=81633
--- Comment #3 from Jakub Jelinek <jakub at gcc dot gnu.org> ---
In *.ifcvt dump we have
_69 = identity[j_31][0];
_71 = _69 * _69;
tmp[j_31][0] = _71;
_77 = identity[j_31][1];
_78 = _69 * _77;
tmp[j_31][1] = _78;
_84 = identity[j_31][2];
_85 = _69 * _84;
tmp[j_31][2] = _85;
_91 = identity[j_31][3];
_92 = _69 * _91;
tmp[j_31][3] = _92;
but the fact that the first stmt is _69 * _69 doesn't mean the whole SLP is a
square of another vector.