Return-Path: <linux-kernel-owner@vger.kernel.org>
Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand
	id S1754536AbcC1GvW (ORCPT <rfc822;w@1wt.eu>);
	Mon, 28 Mar 2016 02:51:22 -0400
Received: from asavdk4.altibox.net ([109.247.116.15]:42649 "EHLO
	asavdk4.altibox.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org
	with ESMTP id S1753081AbcC1GvO (ORCPT
	<rfc822;linux-kernel@vger.kernel.org>);
	Mon, 28 Mar 2016 02:51:14 -0400
Date: Mon, 28 Mar 2016 08:51:06 +0200
From: Sam Ravnborg <sam@ravnborg.org>
To: "zhaoxiu.zeng" <zhaoxiu.zeng@gmail.com>
Cc: Denys Vlasenko <dvlasenk@redhat.com>, Arnd Bergmann <arnd@arndb.de>,
        Andrew Morton <akpm@linux-foundation.org>,
        Martin Kepplinger <martink@posteo.de>,
        Sasha Levin <sasha.levin@oracle.com>, Ingo Molnar <mingo@kernel.org>,
        Yury Norov <yury.norov@gmail.com>, linux-kernel@vger.kernel.org,
        linux-arch@vger.kernel.org, "David S. Miller" <davem@davemloft.net>
Subject: Re: [PATCH 01/31] bitops: add parity functions
Message-ID: <20160328065106.GA12154@ravnborg.org>
References: <1458788612-4367-1-git-send-email-zhaoxiu.zeng@gmail.com>
 <56F3A77D.6060802@redhat.com>
 <56F75490.9010608@gmail.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <56F75490.9010608@gmail.com>
User-Agent: Mutt/1.5.24 (2015-08-30)
X-CMAE-Score: 0
X-CMAE-Analysis: v=2.1 cv=Fo6lhzfq c=1 sm=1 tr=0
	a=Ij76tQDYWdb01v2+RnYW5w==:117 a=Ij76tQDYWdb01v2+RnYW5w==:17
	a=L9H7d07YOLsA:10 a=9cW_t1CCXrUA:10 a=s5jvgZ67dGcA:10 a=kj9zAlcOel0A:10
	a=cqvmlmxeAAAA:8 a=IQXkR5oP9Z18AC1z6ooA:9 a=CjuIK1q_8ugA:10
	a=t_ttxzwXqboA:10
Sender: linux-kernel-owner@vger.kernel.org
List-ID: <linux-kernel.vger.kernel.org>
X-Mailing-List: linux-kernel@vger.kernel.org
Content-Length: 1356
Lines: 50

> diff --git a/include/asm-generic/bitops/arch_parity.h b/include/asm-generic/bitops/arch_parity.h
> new file mode 100644
> index 0000000..cddc555
> --- /dev/null
> +++ b/include/asm-generic/bitops/arch_parity.h
> @@ -0,0 +1,39 @@
> +#ifndef _ASM_GENERIC_BITOPS_ARCH_PARITY_H_
> +#define _ASM_GENERIC_BITOPS_ARCH_PARITY_H_
> +
> +#include <asm/types.h>
> +
> +/*
> + * Refrence to 'https://graphics.stanford.edu/~seander/bithacks.html#ParityParallel'.
> + */
> +
> +static inline unsigned int __arch_parity4(unsigned int w)
> +{
> +	w &= 0xf;
> +	return (0x6996 >> w) & 1;
> +}
> +
> +static inline unsigned int __arch_parity8(unsigned int w)
> +{
> +	w ^= w >> 4;
> +	return __arch_parity4(w);
> +}
> +
> +static inline unsigned int __arch_parity16(unsigned int w)
> +{
> +	w ^= w >> 8;
> +	return __arch_parity8(w);
> +}
> +
> +static inline unsigned int __arch_parity32(unsigned int w)
> +{
> +	w ^= w >> 16;
> +	return __arch_parity16(w);
> +}
> +
> +static inline unsigned int __arch_parity64(__u64 w)
> +{
> +	return __arch_parity32((unsigned int)(w >> 32) ^ (unsigned int)w);
> +}
Defining these as static inlines in asm-generic prevent an architecture
from selecting between a more optimal asm version or the generic version
at run-time.
sparc would benefit from this as only some sparc chips supports popc.
See how this is done for hweight*

	Sam