From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: by sourceware.org (Postfix, from userid 48) id 91F0C3846411; Wed, 3 Apr 2024 11:35:38 +0000 (GMT) DKIM-Filter: OpenDKIM Filter v2.11.0 sourceware.org 91F0C3846411 DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gcc.gnu.org; s=default; t=1712144138; bh=mUItJFW2YoW5705vOfIT+5kNbRJP4gabW7g3l62rsQ4=; h=From:To:Subject:Date:In-Reply-To:References:From; b=Y5IDgre1trcM/+08R2c/Q1c4kRnW0mV5WsLpmqv76odXqWTE64vQU3Vx/2Cdi5eFl cwVVBu9lzYaFsZMt6U+XaLb3RKf7Pua6zRHW5tAhVY8qMXazmJzclUGrABWhgKG7hQ vZACKJwm6meLiD8uVI5C5peK1vsD8HtDiTqIZ9BQ= From: "rguenth at gcc dot gnu.org" To: gcc-bugs@gcc.gnu.org Subject: [Bug middle-end/114570] GCC doesn't perform good loop invariant code motion for very long vector operations. Date: Wed, 03 Apr 2024 11:35:37 +0000 X-Bugzilla-Reason: CC X-Bugzilla-Type: changed X-Bugzilla-Watch-Reason: None X-Bugzilla-Product: gcc X-Bugzilla-Component: middle-end X-Bugzilla-Version: 14.0 X-Bugzilla-Keywords: missed-optimization X-Bugzilla-Severity: normal X-Bugzilla-Who: rguenth at gcc dot gnu.org X-Bugzilla-Status: NEW X-Bugzilla-Resolution: X-Bugzilla-Priority: P3 X-Bugzilla-Assigned-To: unassigned at gcc dot gnu.org X-Bugzilla-Target-Milestone: --- X-Bugzilla-Flags: X-Bugzilla-Changed-Fields: bug_status everconfirmed cf_reconfirmed_on Message-ID: In-Reply-To: References: Content-Type: text/plain; charset="UTF-8" Content-Transfer-Encoding: quoted-printable X-Bugzilla-URL: http://gcc.gnu.org/bugzilla/ Auto-Submitted: auto-generated MIME-Version: 1.0 List-Id: https://gcc.gnu.org/bugzilla/show_bug.cgi?id=3D114570 Richard Biener changed: What |Removed |Added ---------------------------------------------------------------------------- Status|UNCONFIRMED |NEW Ever confirmed|0 |1 Last reconfirmed| |2024-04-03 --- Comment #1 from Richard Biener --- There's no (gimple) invariant motion after vector operation lowering. RTL invariant motion should see this though but it might have a prohibiting cost model (or doesn't handle [stack] memory?). There's other reasons we want to move vector lowering earlier which might t= hen also catch invariant motion opportunities.=