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 1490738582B6; Wed,  6 Mar 2024 12:55:10 +0000 (GMT)
DKIM-Filter: OpenDKIM Filter v2.11.0 sourceware.org 1490738582B6
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gcc.gnu.org;
	s=default; t=1709729710;
	bh=goqzuamdZsu8B/XKP0f3WEH0htOdZ6O7N7nyeYgL4qM=;
	h=From:To:Subject:Date:In-Reply-To:References:From;
	b=jeThP11t6hSIDuaG45zKjXlmQ4Dd+5sIwQnr81dhp1DuJ9RZ+ujK7N/6TWWvBhaq5
	 D+udSeQGgJaPd5/kHH2XamDIkNusmtL085bpmNGSx+PMXWSP8XcIcgzs9hYv035v+b
	 0e9Td4yyZDV4k+4AkZWVXUMbB/NQJUukXhu1+mw4=
From: "rdapp at gcc dot gnu.org" <gcc-bugzilla@gcc.gnu.org>
To: gcc-bugs@gcc.gnu.org
Subject: [Bug target/114200] [14] RISC-V fixed-length vector miscompile at
 -O3
Date: Wed, 06 Mar 2024 12:55:09 +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: 14.0
X-Bugzilla-Keywords: wrong-code
X-Bugzilla-Severity: normal
X-Bugzilla-Who: rdapp at gcc dot gnu.org
X-Bugzilla-Status: UNCONFIRMED
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-114200-4-gsixfUByV3@http.gcc.gnu.org/bugzilla/>
In-Reply-To: <bug-114200-4@http.gcc.gnu.org/bugzilla/>
References: <bug-114200-4@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=3D114200

--- Comment #1 from Robin Dapp <rdapp at gcc dot gnu.org> ---
Took me a while to analyze this... needed more time than I'd like to admit =
to
make sense of the somewhat weird code created by fully unrolling and peelin=
g.

I believe the problem is that we reload the output register of a vfmacc/fma=
 via
vmv.v.v (subject to length masking) but we should be using vmv1r.v.  The re=
sult
is used by a reduction which always operates on the full length.  As annoyi=
ng
as it was to find - it's definitely a good catch.

I'm testing a patch.  PR114202 is indeed a duplicate.  Going to add its test
case to the patch.=