From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: by sourceware.org (Postfix, from userid 48) id 23E403851161; Fri, 26 Aug 2022 19:15:49 +0000 (GMT) DKIM-Filter: OpenDKIM Filter v2.11.0 sourceware.org 23E403851161 DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gcc.gnu.org; s=default; t=1661541349; bh=j1kilMZpkVhiFBW7a2SswN01kfbS0Gxata+0LKwSo0o=; h=From:To:Subject:Date:In-Reply-To:References:From; b=qNsGu0Z9yz68D4j+0Xz3hbOpOLi7U58RNAaBuUYVWWRQeqckkojwiRIyhyBUf0nri 7+D7wP6fl1hmyLjNRue9OCXwdDF8RATsWvClQQRfLpU27acBWYX11E0bLFJ4IiLDN2 oPXm3tfUYqfPCgYfzwJu/9kdUtyqJ+8v6p4BvRmg= From: "bergner at gcc dot gnu.org" To: gcc-bugs@gcc.gnu.org Subject: [Bug target/101322] ICE in copy_to_mode_reg, at explow.c:651 Date: Fri, 26 Aug 2022 19:15:48 +0000 X-Bugzilla-Reason: CC X-Bugzilla-Type: changed X-Bugzilla-Watch-Reason: None X-Bugzilla-Product: gcc X-Bugzilla-Component: target X-Bugzilla-Version: 12.0 X-Bugzilla-Keywords: ice-on-invalid-code X-Bugzilla-Severity: normal X-Bugzilla-Who: bergner at gcc dot gnu.org X-Bugzilla-Status: ASSIGNED X-Bugzilla-Resolution: X-Bugzilla-Priority: P3 X-Bugzilla-Assigned-To: bergner at gcc dot gnu.org X-Bugzilla-Target-Milestone: --- X-Bugzilla-Flags: X-Bugzilla-Changed-Fields: cf_reconfirmed_on cf_gcctarget bug_status everconfirmed assigned_to 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=3D101322 Peter Bergner changed: What |Removed |Added ---------------------------------------------------------------------------- Last reconfirmed| |2022-08-26 Target|powerpc-*-linux-gnu |powerpc*-*-linux-gnu Status|UNCONFIRMED |ASSIGNED Ever confirmed|0 |1 Assignee|unassigned at gcc dot gnu.org |bergner at gcc dot = gnu.org --- Comment #1 from Peter Bergner --- Confirmed. I see what the problem is. When we expand the built-in, the ca= st to the proper type is lost and we only see a char *vpp. We then use that t= ype to create a temporary and we end up with a reg:QI rather than a reg:OO. Is the casted pointer type supposed to be stripped away like that? Anyway, I can work around that since I know what the temporary type should = be rather than stealing the type from the src operand.=