Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1756198Ab1DBNHO (ORCPT ); Sat, 2 Apr 2011 09:07:14 -0400 Received: from zene.cmpxchg.org ([85.214.230.12]:40096 "EHLO zene.cmpxchg.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1754930Ab1DBNHN (ORCPT ); Sat, 2 Apr 2011 09:07:13 -0400 Date: Sat, 2 Apr 2011 06:06:52 -0700 From: Johannes Weiner To: "H. Peter Anvin" Cc: Maksym Planeta , mingo@redhat.com, kernel-janitors@vger.kernel.org, linux-kernel@vger.kernel.org Subject: Re: [PATCH] page: get_order() optimization Message-ID: <20110402130652.GA2344@cmpxchg.org> References: <1301685493-2567-1-git-send-email-mcsim.planeta@gmail.com> <4D9628C8.3040309@zytor.com> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <4D9628C8.3040309@zytor.com> Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 1844 Lines: 50 On Fri, Apr 01, 2011 at 12:34:32PM -0700, H. Peter Anvin wrote: > On 04/01/2011 12:18 PM, Maksym Planeta wrote: > > Loop was repalaced with __builtin_clz(). This still allows to precompute > > constants, but on some architectures it uses special instruction to > > calculate order. > > > > Signed-off-by: Maksym Planeta > > --- > > include/asm-generic/getorder.h | 8 +++----- > > 1 files changed, 3 insertions(+), 5 deletions(-) > > > > diff --git a/include/asm-generic/getorder.h b/include/asm-generic/getorder.h > > index 67e7245..fe8020c 100644 > > --- a/include/asm-generic/getorder.h > > +++ b/include/asm-generic/getorder.h > > @@ -11,11 +11,9 @@ static inline __attribute_const__ int get_order(unsigned long size) > > int order; > > > > size = (size - 1) >> (PAGE_SHIFT - 1); > > - order = -1; > > - do { > > - size >>= 1; > > - order++; > > - } while (size); > > + order = (__builtin_clzl(size) ^ (BITS_PER_LONG - 1)); > > + if (size == 0) > > + order = 0; > > return order; > > } > > > > You need to guard this with __GNUC__ >= 4; there are still laggards > using gcc 3. Furthermore, on some platforms __builtin_clz*() does a > libgcc call which may be undesirable. > > For the generic case, one can do something like this instead of a loop: It looks odd to me to count from the left and then subtract to get the offset of the msb from the right. Can't we just use fls() here, which is already made available in both generic and arch-optimized versions? pages = size - 1 >> PAGE_SHIFT; if (pages) return __fls(pages) + 1; return 0; -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/