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 1F3FD3858D33; Tue, 26 Mar 2024 01:57:04 +0000 (GMT)
DKIM-Filter: OpenDKIM Filter v2.11.0 sourceware.org 1F3FD3858D33
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gcc.gnu.org;
	s=default; t=1711418224;
	bh=f3LjDZk/Ef7Es3YcrHklU5J/eRBBhNmLDzxXgo6G5PM=;
	h=From:To:Subject:Date:In-Reply-To:References:From;
	b=CauQzaVwUB41GXaFIplIyCK9qzRJXdcnSdj6pa7j2Ma2LdHJaniyZmVYilWcdUVig
	 REPKWZ7iLtM0Z3RgZ+sPeD9gki5DcLGmnFbeu7NQ8g5gYCeK2Muy+iiARdDGnbel3H
	 lvLWqNmmbNiV+SGzksqA7UFTnTnKz9D2GK3qEmWM=
From: "chenglulu at loongson dot cn" <gcc-bugzilla@gcc.gnu.org>
To: gcc-bugs@gcc.gnu.org
Subject: [Bug target/112919] LoongArch: Alignments in tune parameters are not
 precise and they regress performance
Date: Tue, 26 Mar 2024 01:57:00 +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: 
X-Bugzilla-Severity: normal
X-Bugzilla-Who: chenglulu at loongson dot cn
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-112919-4-mpFx83ZN29@http.gcc.gnu.org/bugzilla/>
In-Reply-To: <bug-112919-4@http.gcc.gnu.org/bugzilla/>
References: <bug-112919-4@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=3D112919

--- Comment #17 from chenglulu <chenglulu at loongson dot cn> ---
(In reply to Xi Ruoyao from comment #15)
> > Hi=EF=BC=8CRuoyao:
> >=20
> >  The results of spec2006 on 3A6000 were obtained, I removed the more vo=
latile
> > test items, '-falign-loops=3D8 -falign-functions=3D8 -falign-jumps=3D32
> > -falign-lables=3D4' this set of parameters got the highest score. This =
is the
> > same combination of parameters as the coremark tested by Xu Chenghua.
> >=20
> > The test of the 3A5000 will also be completed around the 15th of this m=
onth,
> > so I want to change the code after the test results of the 3a5000 are o=
ut.
> > What do you think?
>=20
> Ok to me.
>=20
> I'm getting some different results on LA664:
>=20
> 22031.284424 Compiler flags : -O2 -falign-labels=3D4 -falign-functions=3D8
> -falign-loops=3D8 -falign-jumps=3D32 -DPERFORMANCE_RUN=3D1 -lrt
>=20
> vs the "best" one:
>=20
> 22075.055188 Compiler flags : -O2 -falign-labels=3D4 -falign-functions=3D=
32
> -falign-loops=3D16 -falign-jumps=3D8 -DPERFORMANCE_RUN=3D1 -lrt
>=20
> maybe such a 0.1% difference is some random fluctuation, or hardware or
> kernel configuration difference anyway.

The results of spec2006 on LA464 are:
-falign-labels=3D4 -falign-functions=3D32 -falign-loops=3D16 -falign-jumps=
=3D16=