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 9696B3858439; Wed, 27 Mar 2024 00:21:39 +0000 (GMT)
DKIM-Filter: OpenDKIM Filter v2.11.0 sourceware.org 9696B3858439
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gcc.gnu.org;
	s=default; t=1711498899;
	bh=hGM9DSPaTbH4LfMIMPOxQnQyi6+tvMaFPts0L/xaGBM=;
	h=From:To:Subject:Date:In-Reply-To:References:From;
	b=DhQEaUWJkKtwOWniL7raSTEwWG9PjxWiUpOCimT37dP+dIlzNZKQkPXAeuA8A1sCH
	 VGWzKLzJlEWcEz+S1RVzo22eXUSxkk3rGjwn9Emlt4nq7E5VOqLvNgwO3pU2drJ/D4
	 Rd3W30y35TjA5y6lIERaoDNecPOBu1dlR5UEHWPA=
From: "absoler at smail dot nju.edu.cn" <gcc-bugzilla@gcc.gnu.org>
To: gcc-bugs@gcc.gnu.org
Subject: [Bug rtl-optimization/107051] redundant loads when copying a union
Date: Wed, 27 Mar 2024 00:21:38 +0000
X-Bugzilla-Reason: CC
X-Bugzilla-Type: changed
X-Bugzilla-Watch-Reason: None
X-Bugzilla-Product: gcc
X-Bugzilla-Component: rtl-optimization
X-Bugzilla-Version: 12.1.0
X-Bugzilla-Keywords: missed-optimization
X-Bugzilla-Severity: normal
X-Bugzilla-Who: absoler at smail dot nju.edu.cn
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-107051-4-CIw7RwfDVb@http.gcc.gnu.org/bugzilla/>
In-Reply-To: <bug-107051-4@http.gcc.gnu.org/bugzilla/>
References: <bug-107051-4@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=3D107051

--- Comment #2 from absoler at smail dot nju.edu.cn ---
(In reply to Richard Biener from comment #1)
> With -O2 I see
>=20
> func_1:
> .LFB0:
>         .cfi_startproc
>         movl    e(%rip), %eax
>         testl   %eax, %eax
>         je      .L2
> .L3:
>         jmp     .L3
>         .p2align 4,,10
>         .p2align 3
> .L2:
>         movq    g_284+8(%rip), %rax
>         movq    %rax, g_284(%rip)
>         ret
>=20
> note that with -O1 we retain
>=20
>   c =3D g_284[1];
>   c$f0_3 =3D g_284[1].f0;
>   c.f0 =3D c$f0_3;
>   g_284[0] =3D c;
>=20
> after GIMPLE optimization which possibly explains this compared to
>=20
>   c =3D g_284[1];
>   g_284[0] =3D c;
>=20
> with -O2.

for gcc-13.2.0 -O2, it seems still forget to remove the load for this reduc=
ed
case:
```
union U0 {
    short f1;
    int f2;
};
union U0 g1, g2;

volatile int flag;
void func_1() {
    union U0 d[1] =3D {{.f1 =3D 1}};
    for (; flag;)
        ;
    d[0] =3D g2;
    g1 =3D d[0];
}
```
```
func_1:
.LFB0:
        .cfi_startproc
        .p2align 4,,10
        .p2align 3
.L2:
        movl    flag(%rip), %eax
        testl   %eax, %eax
        jne     .L2
        movl    g2(%rip), %eax
        movw    g2(%rip), %ax
        movl    %eax, g1(%rip)
        ret
        .cfi_endproc
```=