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 0DA943858284; Sun, 10 Dec 2023 18:25:41 +0000 (GMT)
DKIM-Filter: OpenDKIM Filter v2.11.0 sourceware.org 0DA943858284
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gcc.gnu.org;
	s=default; t=1702232741;
	bh=5bhWQ2UTD9TSkLi9PXqcHhhP1S72JxH960OqQIkHqhk=;
	h=From:To:Subject:Date:In-Reply-To:References:From;
	b=S7/D+kTplx1xEsphrrJRGqoe9RelV2VYLuiQNp3O6ga9jn43MUA0IOZk/uHk5ldxG
	 2Pql2ZvJOH1/5pr1jYSZuJQjkozYxQdfbLFbUJz7OlkwZxKGI1y4FB1hVyAL0ePj7U
	 V+FFQMZI5DtvNu5fiHi7ydiRS1iUUfZ3Z8RHbeio=
From: "xry111 at gcc dot gnu.org" <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: Sun, 10 Dec 2023 18:25:40 +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: xry111 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-112919-4-i1IVT08WYu@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 #4 from Xi Ruoyao <xry111 at gcc dot gnu.org> ---
On LA664:

19970.709626     -falign-labels=3D4 -falign-functions=3D16 -falign-loops=3D=
64
-falign-jumps=3D16
19970.709626     -falign-labels=3D4 -falign-functions=3D32 -falign-loops=3D=
32
-falign-jumps=3D16
19976.028765     -falign-labels=3D4 -falign-functions=3D64 -falign-loops=3D=
32
-falign-jumps=3D16
19978.689398     -falign-labels=3D4 -falign-functions=3D8 -falign-loops=3D16
-falign-jumps=3D16
19997.333689     -falign-labels=3D4 -falign-functions=3D4 -falign-loops=3D32
-falign-jumps=3D32
20009.337691     -falign-labels=3D4 -falign-functions=3D32 -falign-loops=3D8
-falign-jumps=3D32
20009.337691     -falign-labels=3D4 -falign-functions=3D4 -falign-loops=3D32
-falign-jumps=3D16
20010.672359     -falign-labels=3D4 -falign-functions=3D64 -falign-loops=3D8
-falign-jumps=3D32
20050.795348     -falign-labels=3D4 -falign-functions=3D32 -falign-loops=3D8
-falign-jumps=3D64
20065.547455     -falign-labels=3D4 -falign-functions=3D64 -falign-loops=3D8
-falign-jumps=3D64

So it seems -falign-labels > 4 is just harming.  And interestingly a high
-falign-functions / -falign-jumps helps Coremark...=