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 C45E8385DDD5; Tue, 25 Jun 2024 16:25:25 +0000 (GMT)
DKIM-Filter: OpenDKIM Filter v2.11.0 sourceware.org C45E8385DDD5
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gcc.gnu.org;
	s=default; t=1719332725;
	bh=55czU2dWVgeXVdN5mSm1LK/EYxIEuu7raR1y5xJPL0I=;
	h=From:To:Subject:Date:In-Reply-To:References:From;
	b=sZ3hh8jWCDlwHOxJdgtzS430MsCErZ15eZjES01FZPHTPAecr3qtfB9Tg38N8YTjk
	 rHr19AoJd5fRW/0/m0ZCmudFmqfYq+MmTVDa34ThxBfAALhlQKZMfhXCUMVWIRRD2D
	 7CUPZYZcKUvCBKRcumyoZwNRWEj6yOPZ6z70QOVc=
From: "rvmallad at amazon dot com" <gcc-bugzilla@gcc.gnu.org>
To: gcc-bugs@gcc.gnu.org
Subject: [Bug ipa/114531] Feature proposal for an
 `-finline-functions-aggressive` compiler option
Date: Tue, 25 Jun 2024 16:25:24 +0000
X-Bugzilla-Reason: CC
X-Bugzilla-Type: changed
X-Bugzilla-Watch-Reason: None
X-Bugzilla-Product: gcc
X-Bugzilla-Component: ipa
X-Bugzilla-Version: 14.0
X-Bugzilla-Keywords: 
X-Bugzilla-Severity: enhancement
X-Bugzilla-Who: rvmallad at amazon dot com
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-114531-4-p74xrjVein@http.gcc.gnu.org/bugzilla/>
In-Reply-To: <bug-114531-4@http.gcc.gnu.org/bugzilla/>
References: <bug-114531-4@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=3D114531

--- Comment #13 from Rama Malladi <rvmallad at amazon dot com> ---
(In reply to Jan Hubicka from comment #12)
> If this is without LTO, can you also try the LTO numbers?
> Inliner behaves sifniciantly different with and without LTO, since LTO
> introduces many (and often too many) inlining oppurtunities, which
> sometimes makes things to out of hand.
>=20
> Overal SPEC2k17 without LTO is not the most representative inlining
> benchmark, since most programs there are relatively old and written with
> small abstraction penalty.

The numbers listed above in
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=3D114531#c11 are with LTO. Her=
e are
the base flags used for these runs:

`O2 -march=3Darmv8-a+crc+crypto -mtune=3Dnative -flto`

+ inlining params:
`--param=3Dearly-inlining-insns=3D14 --param=3Dinline-heuristics-hint-perce=
nt=3D600
--param=3Dinline-min-speedup=3D15 --param=3Dmax-inline-insns-auto=3D30
--param=3Dmax-inline-insns-single=3D200`=