From mboxrd@z Thu Jan  1 00:00:00 1970
Return-Path: <linux-kernel-owner@vger.kernel.org>
Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand
	id S1751325AbcEJRXi (ORCPT <rfc822;w@1wt.eu>);
	Tue, 10 May 2016 13:23:38 -0400
Received: from merlin.infradead.org ([205.233.59.134]:58698 "EHLO
	merlin.infradead.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org
	with ESMTP id S1750864AbcEJRXh (ORCPT
	<rfc822;linux-kernel@vger.kernel.org>);
	Tue, 10 May 2016 13:23:37 -0400
Date: Tue, 10 May 2016 19:23:13 +0200
From: Peter Zijlstra <peterz@infradead.org>
To: Borislav Petkov <bp@suse.de>
Cc: x86-ml <x86@kernel.org>, Denys Vlasenko <dvlasenk@redhat.com>,
        "H. Peter Anvin" <hpa@zytor.com>, Brian Gerst <brgerst@gmail.com>,
        LKML <linux-kernel@vger.kernel.org>,
        Dmitry Vyukov <dvyukov@google.com>, Andi Kleen <andi@firstfloor.org>,
        zengzhaoxiu@163.com, Thomas Gleixner <tglx@linutronix.de>,
        Ingo Molnar <mingo@redhat.com>,
        Andrew Morton <akpm@linux-foundation.org>,
        Kees Cook <keescook@chromium.org>,
        Zhaoxiu Zeng <zhaoxiu.zeng@gmail.com>,
        Andy Lutomirski <luto@amacapital.net>
Subject: Re: [PATCH -v2] x86/hweight: Get rid of the special calling
 convention
Message-ID: <20160510172313.GA3192@twins.programming.kicks-ass.net>
References: <20160407094333.GD3866@pd.tnic>
 <20160504184612.GC23257@pd.tnic>
 <CAMzpN2h5AroNfCYcmdRCTBQuqBc_r+i+yMqg2aDFQjwYKWZGOA@mail.gmail.com>
 <5998407c-3497-22c1-45dc-a86afcb73c94@zytor.com>
 <20160504194101.GE23257@pd.tnic>
 <f3405f2e-437c-a077-3a1c-90ff31e74787@zytor.com>
 <20160504202213.GF23257@pd.tnic>
 <572B446D.1030000@redhat.com>
 <20160505140446.GE534@pd.tnic>
 <20160510165318.GD28520@pd.tnic>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <20160510165318.GD28520@pd.tnic>
User-Agent: Mutt/1.5.21 (2012-12-30)
Sender: linux-kernel-owner@vger.kernel.org
List-ID: <linux-kernel.vger.kernel.org>
X-Mailing-List: linux-kernel@vger.kernel.org

On Tue, May 10, 2016 at 06:53:18PM +0200, Borislav Petkov wrote:
>  static __always_inline unsigned int __arch_hweight32(unsigned int w)
>  {
> -	unsigned int res = 0;
> +	unsigned int res;
>  
> -	asm (ALTERNATIVE("call __sw_hweight32", POPCNT32, X86_FEATURE_POPCNT)
> -		     : "="REG_OUT (res)
> -		     : REG_IN (w));
> +	if (likely(static_cpu_has(X86_FEATURE_POPCNT))) {
> +		/* popcnt %eax, %eax */
> +		asm volatile(POPCNT32
> +				: "="REG_OUT (res)
> +				: REG_IN (w));
>  
> -	return res;
> +		return res;
> +	}
> +	return __sw_hweight32(w);
>  }

So what was wrong with using the normal thunk_*.S wrappers for the
calls? That would allow you to use the alternative() stuff which does
generate smaller code.