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 B98C9384AB56; Fri, 10 May 2024 08:05:27 +0000 (GMT)
DKIM-Filter: OpenDKIM Filter v2.11.0 sourceware.org B98C9384AB56
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gcc.gnu.org;
	s=default; t=1715328327;
	bh=MSFD8qme9EmQEVRT7VbkhuzXnNUjyFMvzAhz8/5AMYs=;
	h=From:To:Subject:Date:In-Reply-To:References:From;
	b=SBDH8LwpzrgQbB/Fx/RQ38YqF5sjmRtaInMb5CRhoq9lXo0nMPURHYEqxAiPXOdM5
	 B2aoWEk96AuhvNdmAeknKucTVOK9/icKwlphK7YGGYLlGnrBjnQg7luDuNjc6psfcI
	 yVyk6LSFIhtllu09IiXqQEzqm0cuoyYbubqX9afw=
From: "liuhongt at gcc dot gnu.org" <gcc-bugzilla@gcc.gnu.org>
To: gcc-bugs@gcc.gnu.org
Subject: [Bug target/114987] [14/15 Regression] floating point vector
 regression, x86, between gcc 14 and gcc-13 using -O3 and target clones on
 skylake platforms
Date: Fri, 10 May 2024 08:05:27 +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: missed-optimization
X-Bugzilla-Severity: normal
X-Bugzilla-Who: liuhongt at gcc dot gnu.org
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: 14.2
X-Bugzilla-Flags: 
X-Bugzilla-Changed-Fields: 
Message-ID: <bug-114987-4-FEtQa3Xnrz@http.gcc.gnu.org/bugzilla/>
In-Reply-To: <bug-114987-4@http.gcc.gnu.org/bugzilla/>
References: <bug-114987-4@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=3D114987

--- Comment #6 from Hongtao Liu <liuhongt at gcc dot gnu.org> ---
> I tried to move "vmovdqa %xmm1,0xd0(%rsp)" before "vmovdqa %xmm0,0xe0(%rs=
p)"
> and rebuilt the binary and it will save half the regression.

 57.93 =E2=94=82200:   vaddps       0xc0(%rsp),%ymm3,%ymm5=20=20=20=20=20=
=20=20=20=20=20=20=20=20=20=20=20=20=20=20=20=20=20=20=20
 11.11 =E2=94=82       vaddps       0xe0(%rsp),%ymm2,%ymm6
        ...
  3.22 =E2=94=82       vmovdqa      %xmm1,0xc0(%rsp)=20=20=20=20=20=20=20=
=20=20=20=20=20=20=20=20=20=20=20=20=20=20=20=20=20=20=20=20=20=20=20=20=20=
=20=20=20=20
       =E2=94=82       vmovdqa      %xmm5,0xd0(%rsp)=20=20=20=20=20=20=20=
=20=20=20=20=20=20=20=20=20=20=20=20=20=20=20=20=20=20=20=20=20=20=20=20=20=
=20=20=20=20
  3.52 =E2=94=82       vmovdqa      %xmm0,0xe0(%rsp)=20=20=20=20=20=20=20=
=20=20=20=20=20=20=20=20=20=20=20=20=20=20=20=20=20=20=20=20=20=20=20
       =E2=94=82       vmovdqa      %xmm6,0xf0(%rsp)=20=20=20

I guess there're specific patterns in SKX microarhitecture for STLF, the ma=
in
difference is instruction order of those xmm stores.

>From compiler side, the worth thing to do is PR107916.=