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 6D37E3858408; Tue, 31 Aug 2021 11:54:16 +0000 (GMT)
DKIM-Filter: OpenDKIM Filter v2.11.0 sourceware.org 6D37E3858408
From: "rearnsha at gcc dot gnu.org" <gcc-bugzilla@gcc.gnu.org>
To: gcc-bugs@gcc.gnu.org
Subject: [Bug target/102125] (ARM Cortex-M3 and newer) missed optimization.
 memcpy not needed operations
Date: Tue, 31 Aug 2021 11:54:16 +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: 10.2.1
X-Bugzilla-Keywords: missed-optimization
X-Bugzilla-Severity: normal
X-Bugzilla-Who: rearnsha at gcc dot gnu.org
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-102125-4-M4FxqTYe3W@http.gcc.gnu.org/bugzilla/>
In-Reply-To: <bug-102125-4@http.gcc.gnu.org/bugzilla/>
References: <bug-102125-4@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
X-BeenThere: gcc-bugs@gcc.gnu.org
X-Mailman-Version: 2.1.29
Precedence: list
List-Id: Gcc-bugs mailing list <gcc-bugs.gcc.gnu.org>
List-Unsubscribe: <https://gcc.gnu.org/mailman/options/gcc-bugs>,
 <mailto:gcc-bugs-request@gcc.gnu.org?subject=unsubscribe>
List-Archive: <https://gcc.gnu.org/pipermail/gcc-bugs/>
List-Post: <mailto:gcc-bugs@gcc.gnu.org>
List-Help: <mailto:gcc-bugs-request@gcc.gnu.org?subject=help>
List-Subscribe: <https://gcc.gnu.org/mailman/listinfo/gcc-bugs>,
 <mailto:gcc-bugs-request@gcc.gnu.org?subject=subscribe>
X-List-Received-Date: Tue, 31 Aug 2021 11:54:16 -0000

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=3D102125

--- Comment #5 from Richard Earnshaw <rearnsha at gcc dot gnu.org> ---
Testcase was not quite complete.  Extending it to:

typedef unsigned long long uint64_t;
typedef unsigned long uint32_t;
typedef unsigned char uint8_t;
uint64_t bar64(const uint8_t *rData1)
{
    uint64_t buffer;
    __builtin_memcpy(&buffer, rData1, sizeof(buffer));
    return buffer;
}

uint32_t bar32(const uint8_t *rData1)
{
    uint32_t buffer;
    __builtin_memcpy(&buffer, rData1, sizeof(buffer));
    return buffer;
}

and then looking at the optimized tree output we see:


;; Function bar64 (bar64, funcdef_no=3D0, decl_uid=3D4196, cgraph_uid=3D1,
symbol_order=3D0)

uint64_t bar64 (const uint8_t * rData1)
{
  uint64_t buffer;
  uint64_t _4;

  <bb 2> [local count: 1073741824]:
  __builtin_memcpy (&buffer, rData1_2(D), 8);
  _4 =3D buffer;
  buffer =3D{v} {CLOBBER};
  return _4;

}



;; Function bar32 (bar32, funcdef_no=3D1, decl_uid=3D4200, cgraph_uid=3D2,
symbol_order=3D1)

uint32_t bar32 (const uint8_t * rData1)
{
  unsigned int _3;

  <bb 2> [local count: 1073741824]:
  _3 =3D MEM <unsigned int> [(char * {ref-all})rData1_2(D)];
  return _3;

}

So in the 32-bit case we've eliminated the memcpy at the tree level, but fa=
iled
to do that for 64-bit objects.

We probably need to add 64-bit support to the movmisalign<mode> pattern.=