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 C0D5C3858D20; Wed, 31 Jan 2024 14:45:52 +0000 (GMT)
DKIM-Filter: OpenDKIM Filter v2.11.0 sourceware.org C0D5C3858D20
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gcc.gnu.org;
	s=default; t=1706712352;
	bh=hEzo4Qm14WDJ27d6hIl5cdy3cim+G77eSopguKUf+Sc=;
	h=From:To:Subject:Date:In-Reply-To:References:From;
	b=q8py4hKrvCVoWxrHghyx3/3AMXrJRc2E+XzAi5JRdvyfShbMzpTaTxQJiEIn+ozk+
	 cVinYBa/i3chU2Ps9AUVIswyzn8rtQ9BKP93AUkWZ4W7mQ3B9JvL1CTXOQyBtkhz0f
	 RixM1GIlRILqdgreS4To42g3+qrUYNRJYJ/3wopQ=
From: "jamborm at gcc dot gnu.org" <gcc-bugzilla@gcc.gnu.org>
To: gcc-bugs@gcc.gnu.org
Subject: [Bug gcov-profile/113646] PGO hurts run-time of 538.imagick_r as
 much as 68% at -Ofast -march=native
Date: Wed, 31 Jan 2024 14:45:51 +0000
X-Bugzilla-Reason: CC
X-Bugzilla-Type: changed
X-Bugzilla-Watch-Reason: None
X-Bugzilla-Product: gcc
X-Bugzilla-Component: gcov-profile
X-Bugzilla-Version: 14.0
X-Bugzilla-Keywords: missed-optimization
X-Bugzilla-Severity: normal
X-Bugzilla-Who: jamborm 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-113646-4-YRu1rr5Rtx@http.gcc.gnu.org/bugzilla/>
In-Reply-To: <bug-113646-4@http.gcc.gnu.org/bugzilla/>
References: <bug-113646-4@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=3D113646

--- Comment #3 from Martin Jambor <jamborm at gcc dot gnu.org> ---
(In reply to Richard Biener from comment #1)
> Did you try with -fprofile-partial-training (is that default on?  it
> probably should ...).  Can you please try training with the rate data
> instead of train
> to rule out a mismatch?

With -fprofile-partial-training the znver4 LTO vs LTOPGO regression (on a n=
ewer
master) goes down from 66% to 54%.=20=20

So far I did not find a way to easily train with the reference run (when I =
add
"train_with =3D refrate" to the config, I always get "ERROR: The workload
specified by train_with MUST be a training workload!")=