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 126C03858CD1; Fri, 10 Nov 2023 13:42:22 +0000 (GMT)
DKIM-Filter: OpenDKIM Filter v2.11.0 sourceware.org 126C03858CD1
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gcc.gnu.org;
	s=default; t=1699623742;
	bh=hfOqqK6KX8aBQq910HJTVY8O2f4JYV4dM+rPV4gQZi4=;
	h=From:To:Subject:Date:In-Reply-To:References:From;
	b=pgWT9xRh5MTbMIHlF93sJRqoNifIerA4JD2QMlDZeRMzBDMiNWsqDzuktkd+EJRYV
	 0838GVBBO7jtNQnufSKH0ZhTKdlBEmiPOHifmwMxAua7cChGINnU4kDfgXR/4VOOS/
	 EW3xcnCGLXaCS8JNxXydEUqQbzKbXS80v8UACyRQ=
From: "juzhe.zhong at rivai dot ai" <gcc-bugzilla@gcc.gnu.org>
To: gcc-bugs@gcc.gnu.org
Subject: [Bug tree-optimization/109088] GCC does not always vectorize
 conditional reduction
Date: Fri, 10 Nov 2023 13:42:21 +0000
X-Bugzilla-Reason: CC
X-Bugzilla-Type: changed
X-Bugzilla-Watch-Reason: None
X-Bugzilla-Product: gcc
X-Bugzilla-Component: tree-optimization
X-Bugzilla-Version: 13.0
X-Bugzilla-Keywords: missed-optimization
X-Bugzilla-Severity: enhancement
X-Bugzilla-Who: juzhe.zhong at rivai dot ai
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-109088-4-SW6dzCf4qn@http.gcc.gnu.org/bugzilla/>
In-Reply-To: <bug-109088-4@http.gcc.gnu.org/bugzilla/>
References: <bug-109088-4@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=3D109088

--- Comment #17 from JuzheZhong <juzhe.zhong at rivai dot ai> ---
Sorry for confusing and not enough information.

I am trying to transform:

+          reduc_1 =3D PHI <..., reduc_2>
+          ...
+          if (...)
+          tmp1 =3D reduc_1 + rhs1;
+          tmp2 =3D tmp1 + rhs2;
+          tmp3 =3D tmp2 + rhs3;
+          ...
+          reduc_3 =3D tmpN-1 + rhsN-1;
+
+          reduc_2 =3D PHI <reduc_1, reduc_3>

First transform the first statement:

tmp1 =3D reduc_1 + rhs1; into  tmp1 =3D rhs1 + 0;

Then it will become bogus data move assignment: tmp1 =3D rhs1.
The later PASS will eliminate it.

Then, transform the reduction PHI:=20

reduc_1 =3D PHI <..., reduc_2>

into if-convert statement:

reduc_1 =3D PHI <_ifc__35(8), 0(18)>

Thid, transform=20

+          reduc_3 =3D tmpN-1 + rhsN-1;
+
+          reduc_2 =3D PHI <reduc_1, reduc_3>

into :

reduc_3 =3D tmpN-1 + rhsN-1;
_ifc__35 =3D .COND_ADD (condition, reduc_1, reduc_3, reduc_1);


So finally:

result_1 =3D PHI <_ifc__35(8), 0(18)>
...
tmp1 =3D rhs1;
tmp2 =3D tmp1 + rhs2;
tmp3 =3D tmp2 + rhs3;
...
reduc_3 =3D tmpN-1 + rhsN-1;
_ifc__35 =3D .COND_ADD (condition, reduc_1, reduc_3, reduc_1);=