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 3AF6F385841A; Fri, 13 Oct 2023 23:13:10 +0000 (GMT)
DKIM-Filter: OpenDKIM Filter v2.11.0 sourceware.org 3AF6F385841A
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gcc.gnu.org;
	s=default; t=1697238790;
	bh=oqVclFXKsdDcB1BIAYHeZtMPiOCy9ZRhXb3+QOtGbaM=;
	h=From:To:Subject:Date:In-Reply-To:References:From;
	b=H7SCP5vzFzx5+0k662If3DOeeVGwckkKx6hIY0VnG23kmzkcols/Jy/yaR+U5zcVj
	 7LVByNmjHQDHap2jPhf4FEyPThbfAmepyNHC5xUCRb0sWPiev5VNykEF1pnPja38OW
	 NLxSC5xvyEVlPsen3LYpgmTIfASG0tkHeF1LYyhE=
From: "juzhe.zhong at rivai dot ai" <gcc-bugzilla@gcc.gnu.org>
To: gcc-bugs@gcc.gnu.org
Subject: [Bug tree-optimization/111791] RISC-V: Strange loop vectorizaion on
 popcount function
Date: Fri, 13 Oct 2023 23:13:09 +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: 14.0
X-Bugzilla-Keywords: missed-optimization
X-Bugzilla-Severity: normal
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: cc
Message-ID: <bug-111791-4-d5ET8l8P00@http.gcc.gnu.org/bugzilla/>
In-Reply-To: <bug-111791-4@http.gcc.gnu.org/bugzilla/>
References: <bug-111791-4@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=3D111791

JuzheZhong <juzhe.zhong at rivai dot ai> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |juzhe.zhong at rivai dot ai

--- Comment #3 from JuzheZhong <juzhe.zhong at rivai dot ai> ---
This is because RISC-V didn't vectorize popcount:

It just a scalar call popcount:

call    __popcountdi2

I have a patch support popcount optab:

https://gcc.gnu.org/pipermail/gcc-patches/2023-July/625870.html

However, I don't think we should enable popcount optab if we don't have
a single vector popcount instruction.

An ideal way is to add fallback popcount in loop vectorizer if the target
doesn't
enable vector popcount optab, Robin is working on that:

https://gcc.gnu.org/pipermail/gcc-patches/2023-August/626567.html

So, that's why I drop my patch.

More details need Robin comments.=