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 DE4EF3858D32; Mon, 27 Feb 2023 08:17:06 +0000 (GMT)
DKIM-Filter: OpenDKIM Filter v2.11.0 sourceware.org DE4EF3858D32
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gcc.gnu.org;
	s=default; t=1677485826;
	bh=ULOOFLRDQfdWdhNXT419/SNukL3qgK/e7ETtaL7FNYs=;
	h=From:To:Subject:Date:In-Reply-To:References:From;
	b=aQwfkJ6NGlRLr2J2Wzx77J3MFncgV/T5NYpjrv1l4SUFKTbpgusWMVuB2a3ERR+rU
	 LwwqQUc0XQhMPzXcAbtfe5A/1JvUTzPYAjPWgpJBJ09aj3bnWshuSsW/0tWEYXD5YG
	 el5ROHqTUzUcr3yJRyqsjdLix+bHn1aHdLztAtYg=
From: "amonakov at gcc dot gnu.org" <gcc-bugzilla@gcc.gnu.org>
To: gcc-bugs@gcc.gnu.org
Subject: [Bug target/108922] fmod() 13x slowdown in gcc4.9 dropping "fprem"
 and calling fmod()
Date: Mon, 27 Feb 2023 08:17:06 +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: 12.2.1
X-Bugzilla-Keywords: missed-optimization
X-Bugzilla-Severity: normal
X-Bugzilla-Who: amonakov at gcc dot gnu.org
X-Bugzilla-Status: WAITING
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-108922-4-wABmcVpsyI@http.gcc.gnu.org/bugzilla/>
In-Reply-To: <bug-108922-4@http.gcc.gnu.org/bugzilla/>
References: <bug-108922-4@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=3D108922

--- Comment #15 from Alexander Monakov <amonakov at gcc dot gnu.org> ---
That is the fancy-error-handling path that is reached under _LIB_VERSION !=
=3D
_IEEE_. Before glibc-2.27, linking with -lieee would set _LIB_VERSION =3D _=
IEEE_,
and then glibc would use the fprem[1] instruction without any special-casin=
g.

musl libc does not implement errno setting for math functions, and always u=
ses
fprem directly; likewise for Apple libm:

https://github.com/apple-oss-distributions/Libm/blob/17a5f9daa3f5679f7536b2=
6f133b40cc078753c3/Source/Intel/fmod.s=