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 B204A38582A3; Wed,  6 Mar 2024 12:15:59 +0000 (GMT)
DKIM-Filter: OpenDKIM Filter v2.11.0 sourceware.org B204A38582A3
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gcc.gnu.org;
	s=default; t=1709727359;
	bh=MsceB/HwgMlgIEhoB8kQTKEYZ9YqeAGpqJkjGcHEjW8=;
	h=From:To:Subject:Date:In-Reply-To:References:From;
	b=FKXpzg85ZqSWK0/aP4HtufvJAs6QFkprNCJYlJmhtcsTHWKxoIU948TPzvwJ8fQ4H
	 JJe1/rmqHySWW02592xmf7ZjQ+yreDmzwsInVcp4Z+tsuV3ZUQtGeTgDq9TF0Wrdoz
	 TC0oBg6fMX4OZSXGTXf3xppFyKU5yinIVKaTkv5g=
From: "gjl at gcc dot gnu.org" <gcc-bugzilla@gcc.gnu.org>
To: gcc-bugs@gcc.gnu.org
Subject: [Bug target/114252] Introducing bswapsi reduces code performance
Date: Wed, 06 Mar 2024 12:15:59 +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: missed-optimization
X-Bugzilla-Severity: normal
X-Bugzilla-Who: gjl at gcc dot gnu.org
X-Bugzilla-Status: NEW
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-114252-4-WB6MOCMY88@http.gcc.gnu.org/bugzilla/>
In-Reply-To: <bug-114252-4@http.gcc.gnu.org/bugzilla/>
References: <bug-114252-4@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=3D114252

--- Comment #3 from Georg-Johann Lay <gjl at gcc dot gnu.org> ---
(In reply to Richard Biener from comment #1)
> but somehow we end up doing a libcall?

It's not a libcall in the GCC sense, for the compiler it's just an ordinary
insn.  The backend then prints this as a transparent call to libgcc.

Purpose is that many functions have a small, known footprint as they are
implemented in assembly. An ordinary call would clobber all callee-used reg=
s,
so using a transparent call gives better code than a real call.  Notice thi=
s is
the nsn:

(define_insn "*bswapsi2.libgcc"
  [(set (reg:SI 22)
        (bswap:SI (reg:SI 22)))
   (clobber (reg:CC REG_CC))]
  "reload_completed"
  "%~call __bswapsi2"
  [(set_attr "type" "xcall")])

However, for the purpose of this PR, no bswap is needed in the 1st place; j=
ust
have a look at the v13 code. It just loads the bytes as they belong into the
target value; while v14 loads all 32 bits in one chunk and then starts fidd=
ling
and moving around the constituent bytes.=