From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: by sourceware.org (Postfix, from userid 48) id CC268385841A; Wed, 7 Sep 2022 10:00:31 +0000 (GMT) DKIM-Filter: OpenDKIM Filter v2.11.0 sourceware.org CC268385841A DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gcc.gnu.org; s=default; t=1662544831; bh=eyoEVxlQ5CuvmqECR2duCHcLUyYZJffBV4P0J54rZgU=; h=From:To:Subject:Date:In-Reply-To:References:From; b=LWcD9zUJ57Y5b+lfaqhB+Hzl9WDVlBaVSL9CfgWXjgAUxq4IZkz85WtuYdiDdahJN 7njMgOBKFPvKTzikSwszRW5ru59C9gLojavjFPNaFW1HUvqtoj0bqtDJo7o+N84scB V8rrVg2kDWok4k9p6wR1DEJB7ut6cX6aSqxhrAPQ= From: "rguenth at gcc dot gnu.org" To: gcc-bugs@gcc.gnu.org Subject: [Bug tree-optimization/106866] [13 Regression] ICE in predicate::init_from_control_deps Date: Wed, 07 Sep 2022 10:00:31 +0000 X-Bugzilla-Reason: CC X-Bugzilla-Type: changed X-Bugzilla-Watch-Reason: None X-Bugzilla-Product: gcc X-Bugzilla-Component: tree-optimization X-Bugzilla-Version: 13.0 X-Bugzilla-Keywords: ice-on-valid-code X-Bugzilla-Severity: normal X-Bugzilla-Who: rguenth at gcc dot gnu.org X-Bugzilla-Status: ASSIGNED X-Bugzilla-Resolution: X-Bugzilla-Priority: P3 X-Bugzilla-Assigned-To: rguenth at gcc dot gnu.org X-Bugzilla-Target-Milestone: 13.0 X-Bugzilla-Flags: X-Bugzilla-Changed-Fields: cc 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=3D106866 Richard Biener changed: What |Removed |Added ---------------------------------------------------------------------------- CC| |hubicka at gcc dot gnu.org --- Comment #3 from Richard Biener --- So somehow we late figure bar() doesn't return abnormally and thus DSE eliminates a call to it but doesn't trigger if (need_ab_cleanup && stmt_can_make_abnormal_goto (stmt)) bitmap_set_bit (need_ab_cleanup, bb->index); which means when we either set ECF_LEAF to the function or cleared cfun->calls_setjmp (the recursive invocation) we should have pruned abnormal edges but we failed to. That ends us with a block like ;; basic block 4, loop depth 0 ;; pred: 12 ;; 5 # uninitialized_2(ab) =3D PHI goto ; [99.96%] ;; succ: 11 ;; 5 where the 4->5 edge is abnormal but there's no stmt in this block that could be responsible for it. We also fail to verify ICE here.=