From mboxrd@z Thu Jan 1 00:00:00 1970 From: "Michael S. Tsirkin" Subject: Re: [PATCH 1/3] Enable non page boundary BAR device assignment Date: Wed, 16 Dec 2009 12:41:45 +0200 Message-ID: <20091216104144.GC28541@redhat.com> References: <1260901828-5145-1-git-send-email-agraf@suse.de> <1260901828-5145-2-git-send-email-agraf@suse.de> <20091215183840.GA25724@redhat.com> <4B27D9C7.7020402@suse.de> <20091215185018.GA25764@redhat.com> <4B27DDAA.4040802@suse.de> <20091215210444.GA26712@redhat.com> <4B27FB89.1070306@suse.de> <20091216054709.GM15084@x200.localdomain> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Cc: Alexander Graf , kvm list , Gleb Natapov , Muli Ben-Yehuda To: Chris Wright Return-path: Received: from mx1.redhat.com ([209.132.183.28]:61548 "EHLO mx1.redhat.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1750987AbZLPKod (ORCPT ); Wed, 16 Dec 2009 05:44:33 -0500 Content-Disposition: inline In-Reply-To: <20091216054709.GM15084@x200.localdomain> Sender: kvm-owner@vger.kernel.org List-ID: On Tue, Dec 15, 2009 at 09:47:09PM -0800, Chris Wright wrote: > * Alexander Graf (agraf@suse.de) wrote: > > Michael S. Tsirkin wrote: > > > Existing code first page aligned address, passes this to mmap, > > Did you mean page aligned size? Yes, of course. > > > then adds page offset. I do not know why. Could be e.g. to handle > > > different libc/kernel versions. If we are not sure it is > > > not needed, let's keep it this way. If we are sure, let's > > > cleanup in a separate patch. Makes sense? > > > > Chris, want to make a statement here? :-) > > I have no issue w/ splitting that out. I don't think it is necessary. > The "adds page offset" bit is, however. > > thanks, > -chris I looked at kernel source history, as far as I can tell all mmap implementations round size up to full page size. So yes, this seems safe. -- MST