From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1753778Ab3DVV3J (ORCPT ); Mon, 22 Apr 2013 17:29:09 -0400 Received: from mail.linuxfoundation.org ([140.211.169.12]:52848 "EHLO mail.linuxfoundation.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1753115Ab3DVV3H (ORCPT ); Mon, 22 Apr 2013 17:29:07 -0400 Date: Mon, 22 Apr 2013 14:29:05 -0700 From: Andrew Morton To: Geert Uytterhoeven Cc: Chanho Min , "Darrick J. Wong" , Bob Pearson , Richard Weinberger , Herbert Xu , "linux-kernel@vger.kernel.org" , linux-crypto@vger.kernel.org, Yann Collet , Kyungsik Lee , "Linux/m68k" , sparclinux , Linux-Next Subject: Re: [PATCH 1/2] lib: Add lz4 compressor module Message-Id: <20130422142905.3e4fff72671bb3e245055eee@linux-foundation.org> In-Reply-To: References: <51750159.07d70e0a.5461.ffffec64SMTPIN_ADDED_BROKEN@mx.google.com> X-Mailer: Sylpheed 3.2.0beta5 (GTK+ 2.24.10; x86_64-pc-linux-gnu) Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Mon, 22 Apr 2013 11:24:21 +0200 Geert Uytterhoeven wrote: > On Mon, Apr 22, 2013 at 11:22 AM, Chanho Min wrote: > >>> +#define HTYPE const u8* > >>> + > >>> +#ifdef __BIG_ENDIAN > >>> +#define LZ4_NBCOMMONBYTES(val) (__builtin_clz(val) >> 3) > >>> +#else > >>> +#define LZ4_NBCOMMONBYTES(val) (__builtin_ctz(val) >> 3) > >>> +#endif > >> > >>It seems at least m68k and sparc don't have the __builtin_clz() functions: > >> > >>m68k-allmodconfig (http://kisskb.ellerman.id.au/kisskb/buildresult/8572593/): > >> > >>ERROR: "__clzsi2" [lib/lz4/lz4hc_compress.ko] undefined! > >>ERROR: "__clzsi2" [lib/lz4/lz4_compress.ko] undefined! > > > > gcc seems to define __builtin_clz as __clzsi2 in some architecture. > > But, kernel doesn't link libgcc.a. > > If kernel should use gcc's built-in function without libgcc.a, > > do we need to port __clzsi2 to 'arch/*/lib/*'? > > That's another option. Without having seen the patch .... Yes, if we fix it this way then we also fix it for future callers of __builtin_clz().