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 8AB613858414; Fri, 26 Jan 2024 19:45:41 +0000 (GMT)
DKIM-Filter: OpenDKIM Filter v2.11.0 sourceware.org 8AB613858414
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gcc.gnu.org;
	s=default; t=1706298341;
	bh=+DsK6lcp6hS3Hxt2t1fsUGZHAocGKuoLdXqdFb7txiY=;
	h=From:To:Subject:Date:In-Reply-To:References:From;
	b=wjk3ri2KSQwS5/Yqrjbb/N20KuumJhAPl7mdY04zIYaFtQAyQzFZMDaMvCSKjHD8/
	 z93ybkX9qLQIt0tqYPHWvrcpDUyetmlHmBoi/qU2x6UvjrArxLhHSMB5J39qhpe6lc
	 dq7qRniwOIHLT275AMP120FfTImg7n+Ea3TpnT4k=
From: "kacper.slominski72 at gmail dot com" <gcc-bugzilla@gcc.gnu.org>
To: gcc-bugs@gcc.gnu.org
Subject: [Bug tree-optimization/113467] [14 regression] libgcrypt-1.10.3 is
 miscompiled
Date: Fri, 26 Jan 2024 19:45:39 +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: wrong-code
X-Bugzilla-Severity: normal
X-Bugzilla-Who: kacper.slominski72 at gmail dot com
X-Bugzilla-Status: UNCONFIRMED
X-Bugzilla-Resolution: 
X-Bugzilla-Priority: P1
X-Bugzilla-Assigned-To: unassigned at gcc dot gnu.org
X-Bugzilla-Target-Milestone: 14.0
X-Bugzilla-Flags: 
X-Bugzilla-Changed-Fields: cc attachments.created
Message-ID: <bug-113467-4-f8Rkkr0vhM@http.gcc.gnu.org/bugzilla/>
In-Reply-To: <bug-113467-4@http.gcc.gnu.org/bugzilla/>
References: <bug-113467-4@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=3D113467

Kacper S=C5=82omi=C5=84ski <kacper.slominski72 at gmail dot com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |kacper.slominski72 at gmai=
l dot co
                   |                            |m

--- Comment #26 from Kacper S=C5=82omi=C5=84ski <kacper.slominski72 at gmai=
l dot com> ---
Created attachment 57232
  --> https://gcc.gnu.org/bugzilla/attachment.cgi?id=3D57232&action=3Dedit
reduced standalone failing part of the libgcrypt test

I've isolated the failing part of the test, inlined the relevant libgcrypt =
code
and reduced it.

Worth noting is that in the bad output, limbs 1-8 (inclusive) are all off b=
y 1
vs the good output (limbs 0 and 9-15 stay the same). This is because, when
computing the carry, the faulty code adds a whole 8-element vector of 1s to=
 the
output, instead of only a single 1 to the output limb (because the carry lo=
op
terminates early after one iteration).

The faulty generated code is:

 8049280:       b8 01 00 00 00          mov    $0x1,%eax
 ...
        x =3D *s1_ptr++ + 1;
 804928d:       c5 f9 6e c0             vmovd  %eax,%xmm0
 ...
        x =3D *s1_ptr++ + 1;
 8049296:       c4 e2 7d 58 c0          vpbroadcastd %xmm0,%ymm0
 804929b:       c5 fd fe 42 e0          vpaddd -0x20(%edx),%ymm0,%ymm0
        *res_ptr++ =3D x;
 80492a0:       c5 fe 7f 42 e0          vmovdqu %ymm0,-0x20(%edx)

While in the working binary (produced by gcc 13) the relevant code is:

        x =3D *s1_ptr++ + 1;
    1270:       8b 54 81 04             mov    0x4(%ecx,%eax,4),%edx
    1274:       42                      inc    %edx
        *res_ptr++ =3D x;
    1275:       89 54 81 04             mov    %edx,0x4(%ecx,%eax,4)=