From mboxrd@z Thu Jan 1 00:00:00 1970 From: Yinghai Lu Subject: Re: [PATCH 05/16] x86, memblock: Add get_free_all_memory_range() Date: Tue, 27 Jul 2010 22:41:38 -0700 Message-ID: <4C4FC312.4010503@kernel.org> References: <1279824241-17582-1-git-send-email-yinghai@kernel.org> <1279824241-17582-6-git-send-email-yinghai@kernel.org> <1280293338.1970.235.camel@pasglop> Mime-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 7bit Return-path: Received: from rcsinet10.oracle.com ([148.87.113.121]:37156 "EHLO rcsinet10.oracle.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752099Ab0G1Fmb (ORCPT ); Wed, 28 Jul 2010 01:42:31 -0400 In-Reply-To: <1280293338.1970.235.camel@pasglop> Sender: linux-arch-owner@vger.kernel.org List-ID: To: Benjamin Herrenschmidt Cc: Ingo Molnar , Thomas Gleixner , "H. Peter Anvin" , Andrew Morton , David Miller , Linus Torvalds , Johannes Weiner , linux-kernel@vger.kernel.org, linux-arch@vger.kernel.org, Jan Beulich On 07/27/2010 10:02 PM, Benjamin Herrenschmidt wrote: > On Thu, 2010-07-22 at 11:43 -0700, Yinghai Lu wrote: >> >> + /* Take out region array itself at first*/ >> + if (memblock.reserved.regions != memblock_reserved_init_regions) >> + memblock_free(__pa(memblock.reserved.regions), sizeof(struct memblock_region) * memblock.reserved.max); >> + > > More of that horror. Don't. > > Another option from what i proposed earlier is to actually have a function > inside mm/memblock to free it (memblock_free_arrays()) that does nothing > if memblock is to survive init. > > I hate exporting variables or data structures like that. Functions are > semantically a lot cleaner. ok, will check if can change that to function instead. but it will be with add-on patch. Thanks Yinghai