public inbox for gcc-bugs@sourceware.org help / color / mirror / Atom feed
From: "congh at google dot com" <gcc-bugzilla@gcc.gnu.org> To: gcc-bugs@gcc.gnu.org Subject: [Bug tree-optimization/60656] [4.8/4.9 regression] x86 vectorization produces wrong code Date: Tue, 25 Mar 2014 19:41:00 -0000 [thread overview] Message-ID: <bug-60656-4-g89J7XVZdz@http.gcc.gnu.org/bugzilla/> (raw) In-Reply-To: <bug-60656-4@http.gcc.gnu.org/bugzilla/> http://gcc.gnu.org/bugzilla/show_bug.cgi?id=60656 Cong Hou <congh at google dot com> changed: What |Removed |Added ---------------------------------------------------------------------------- CC| |congh at google dot com --- Comment #2 from Cong Hou <congh at google dot com> --- This bug is caused by an optimization in GCC vectorizer that is not implemented properly. When a reduction operation is vectorized, the order of elements in vectors directly used in reduction does not matter. In some cases the vectorizer may generate less code based on this fact. GCC assigns a property named "vect_used_by_reduction" to all vectors participating in reductions. However, vectors that are indirectly used in reduction also have this property. For example, consider the following three statements (all operands are vectors): a = b op1 c; d = a op2 e; s1 = s0 op3 d; Here assume the last statement is a reduction one, then a,b,c,d,e all have the property "vect_used_by_reduction". However, if op2 is different from op3, then a's element order can affect the final result. GCC does not check this.
next prev parent reply other threads:[~2014-03-25 19:41 UTC|newest] Thread overview: 18+ messages / expand[flat|nested] mbox.gz Atom feed top 2014-03-25 16:02 [Bug tree-optimization/60656] New: " ppluzhnikov at google dot com 2014-03-25 18:09 ` [Bug tree-optimization/60656] " jakub at gcc dot gnu.org 2014-03-25 19:41 ` congh at google dot com [this message] 2014-03-25 20:03 ` jakub at gcc dot gnu.org 2014-03-25 20:12 ` congh at google dot com 2014-03-28 11:19 ` jakub at gcc dot gnu.org 2014-03-28 15:21 ` jakub at gcc dot gnu.org 2014-03-28 18:48 ` congh at google dot com 2014-04-09 12:25 ` jakub at gcc dot gnu.org 2014-04-09 12:28 ` [Bug tree-optimization/60656] [4.8 " jakub at gcc dot gnu.org 2014-04-16 8:52 ` bernd.edlinger at hotmail dot de 2014-04-16 8:56 ` jakub at gcc dot gnu.org 2014-05-22 9:07 ` rguenth at gcc dot gnu.org 2014-12-19 13:41 ` jakub at gcc dot gnu.org 2015-06-09 23:37 ` bernhard.kaindl at thalesgroup dot com 2015-06-10 9:01 ` rguenth at gcc dot gnu.org 2015-06-10 11:04 ` rguenth at gcc dot gnu.org 2015-06-10 13:51 ` rguenth at gcc dot gnu.org
Reply instructions: You may reply publicly to this message via plain-text email using any one of the following methods: * Save the following mbox file, import it into your mail client, and reply-to-all from there: mbox Avoid top-posting and favor interleaved quoting: https://en.wikipedia.org/wiki/Posting_style#Interleaved_style * Reply using the --to, --cc, and --in-reply-to switches of git-send-email(1): git send-email \ --in-reply-to=bug-60656-4-g89J7XVZdz@http.gcc.gnu.org/bugzilla/ \ --to=gcc-bugzilla@gcc.gnu.org \ --cc=gcc-bugs@gcc.gnu.org \ /path/to/YOUR_REPLY https://kernel.org/pub/software/scm/git/docs/git-send-email.html * If your mail client supports setting the In-Reply-To header via mailto: links, try the mailto: linkBe sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions for how to clone and mirror all data and code used for this inbox; as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).