From mboxrd@z Thu Jan  1 00:00:00 1970
Return-Path: <gcc-bugzilla@gcc.gnu.org>
Received: by sourceware.org (Postfix, from userid 48)
	id 0605E385B531; Sat, 25 Feb 2023 11:40:48 +0000 (GMT)
DKIM-Filter: OpenDKIM Filter v2.11.0 sourceware.org 0605E385B531
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gcc.gnu.org;
	s=default; t=1677325249;
	bh=xtACw+boQwTuiiOZ3OkI2D8Z8ntu81oNXxIfLhANRKc=;
	h=From:To:Subject:Date:In-Reply-To:References:From;
	b=Ta4CzLacX05nkSPXd13o3/eqJbWiTlpXFaH+D56g4Mgezgz9JVL3Zwa2aez6EwYs8
	 Fie5JtaAVd7um/gN8zhAVzuUiUALmX/c1GAoQHi2fTx+NVqPOqjtr12qL2iWclPxse
	 P47z+CTHDmi55dPh3vV9JP0wiYFVW4TiWCdUyxVE=
From: "dangelog at gmail dot com" <gcc-bugzilla@gcc.gnu.org>
To: gcc-bugs@gcc.gnu.org
Subject: [Bug libstdc++/108846] std::copy, std::copy_n and std::copy_backward
  on potentially overlapping subobjects
Date: Sat, 25 Feb 2023 11:40:47 +0000
X-Bugzilla-Reason: CC
X-Bugzilla-Type: changed
X-Bugzilla-Watch-Reason: None
X-Bugzilla-Product: gcc
X-Bugzilla-Component: libstdc++
X-Bugzilla-Version: 13.0
X-Bugzilla-Keywords: ABI
X-Bugzilla-Severity: normal
X-Bugzilla-Who: dangelog at gmail dot com
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: 
Message-ID: <bug-108846-4-oXs3FrZE0Z@http.gcc.gnu.org/bugzilla/>
In-Reply-To: <bug-108846-4@http.gcc.gnu.org/bugzilla/>
References: <bug-108846-4@http.gcc.gnu.org/bugzilla/>
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: <gcc-bugs.sourceware.org>

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=3D108846

--- Comment #15 from Giuseppe D'Angelo <dangelog at gmail dot com> ---
That's not what I meant; a type can be trivial(ly copyable) and move-only.

Here's a modification of Arthur's example:

// move-only
struct B {
    B(int i, short j) : i(i), j(j) {}
    B(B &&) =3D default;
    B &operator=3D(B &&) =3D default;
    int i;
    short j;=20
};
struct D : B {
    D(int i, short j, short x) : B(i, j), x(x) {}
    D(D &&) =3D default;
    D &operator=3D(D &&) =3D default;
    short x;
};
int main() {
    D ddst(1, 2, 3);
    D dsrc(4, 5, 6);
    B *dst =3D &ddst;
    B *src =3D &dsrc;
    static_assert(std::is_trivially_copyable_v<B>);

    std::move(src, src+1, dst);
    assert(ddst.x =3D=3D 3);=20
}


The call to std::move ends up in the same memmove codepath as std::copy_n (=
B is
trivially copyable), but with the proposed patch it will fail to compile
because it's not actually copy assignable.=