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 B7CC13857B81; Thu,  1 Feb 2024 13:42:20 +0000 (GMT)
DKIM-Filter: OpenDKIM Filter v2.11.0 sourceware.org B7CC13857B81
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gcc.gnu.org;
	s=default; t=1706794940;
	bh=B4HrZfbP8A2tAMTv6ZKIa6cpCbcWK+v1UdLl1yKsgCI=;
	h=From:To:Subject:Date:In-Reply-To:References:From;
	b=ZeC8DOJVVcrE/tY5v30RhYQI7dBQI8IjPV+rXF4CPIZXPRkYthfR8nKKuouWR3QSv
	 WEsP6Kno2EPG5SCvk2Vjw/nt9pQ2KZyDQtALaAY50ildoPLjrRQOOpNl7md5z1qobj
	 cVGetdkn+YK2HL2ZsjXjUURXUqTLq1HOQ1joBoG4=
From: "juzhe.zhong at rivai dot ai" <gcc-bugzilla@gcc.gnu.org>
To: gcc-bugs@gcc.gnu.org
Subject: [Bug tree-optimization/51492] vectorizer does not support saturated
 arithmetic patterns
Date: Thu, 01 Feb 2024 13:42:20 +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: 4.6.2
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-51492-4-HAInHtGJq6@http.gcc.gnu.org/bugzilla/>
In-Reply-To: <bug-51492-4@http.gcc.gnu.org/bugzilla/>
References: <bug-51492-4@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=3D51492

--- Comment #9 from JuzheZhong <juzhe.zhong at rivai dot ai> ---
Ok. After investigation of LLVM:

Before loop vectorizer:

  %cond12 =3D tail call i32 @llvm.usub.sat.i32(i32 %conv5, i32 %wsize)
  %conv13 =3D trunc i32 %cond12 to i16

After loop vectorizer:

  %10 =3D call <16 x i32> @llvm.usub.sat.v16i32(<16 x i32> %9, <16 x i32>
%broadcast.splat)
  %11 =3D trunc <16 x i32> %10 to <16 x i16>

I think GCC can follow this approach, that is, first recognize scalar
saturation,
then fall into loop vectorizer to vectorize it into the saturation.=