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 D5E5D3858CDB; Wed, 22 Nov 2023 11:27:54 +0000 (GMT)
DKIM-Filter: OpenDKIM Filter v2.11.0 sourceware.org D5E5D3858CDB
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gcc.gnu.org;
	s=default; t=1700652474;
	bh=/xEwmFX0MUQm838BmSY6nN9iheTY6ok17gEy8zBSkoM=;
	h=From:To:Subject:Date:In-Reply-To:References:From;
	b=ulTWg3BAyJpER445H9xUrHczgkkRPgLSNVE9AKqulCQY5ZH7RGV508eAPikygGIw8
	 lACiuA0AHuN6MDFDRlFQ4AHsPEddCBMYpRiEvMatfsVD4JC3LV/o39BaYFPqTyIWlC
	 k36+67ND/VotrFF7iSuvkIG0Z+JN3vj21Pf2jgGI=
From: "redi at gcc dot gnu.org" <gcc-bugzilla@gcc.gnu.org>
To: gcc-bugs@gcc.gnu.org
Subject: [Bug libstdc++/110879] [14 Regression] Unnecessary reread from
 memory in a loop with std::vector
Date: Wed, 22 Nov 2023 11:27:53 +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: 14.0
X-Bugzilla-Keywords: alias, missed-optimization
X-Bugzilla-Severity: normal
X-Bugzilla-Who: redi at gcc dot gnu.org
X-Bugzilla-Status: RESOLVED
X-Bugzilla-Resolution: FIXED
X-Bugzilla-Priority: P3
X-Bugzilla-Assigned-To: unassigned at gcc dot gnu.org
X-Bugzilla-Target-Milestone: 14.0
X-Bugzilla-Flags: 
X-Bugzilla-Changed-Fields: 
Message-ID: <bug-110879-4-WEfrpVL12A@http.gcc.gnu.org/bugzilla/>
In-Reply-To: <bug-110879-4@http.gcc.gnu.org/bugzilla/>
References: <bug-110879-4@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=3D110879

--- Comment #4 from Jonathan Wakely <redi at gcc dot gnu.org> ---
Ah I think that's probably expected. In _M_realloc_insert (and now
_M_realloc_append) we have:

#if __cplusplus >=3D 201103L
        if _GLIBCXX17_CONSTEXPR (_S_use_relocate())
          {
            // Relocation cannot throw.
            __new_finish =3D _S_relocate(__old_start, __position.base(),
                                       __new_start, _M_get_Tp_allocator());
            ++__new_finish;
            __new_finish =3D _S_relocate(__position.base(), __old_finish,
                                       __new_finish, _M_get_Tp_allocator());
          }
        else
#endif

and then an alternative path used for non-trivial types and for C++98. That
alternative path does more work and probably can't be optimized as well, so=
 the
reads from _M_end_of_storage aren't optimized out.

I think we can just use { target c++11 } for the test.=