From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id ; Thu, 9 May 2002 23:42:53 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id ; Thu, 9 May 2002 23:42:52 -0400 Received: from 12-224-36-73.client.attbi.com ([12.224.36.73]:9482 "HELO kroah.com") by vger.kernel.org with SMTP id ; Thu, 9 May 2002 23:42:51 -0400 Date: Thu, 9 May 2002 19:42:42 -0700 From: Greg KH To: Keith Owens Cc: linux-kernel@vger.kernel.org Subject: Re: 2.5.15 laziness in export-objs Message-ID: <20020510024242.GA20468@kroah.com> In-Reply-To: <25744.1020998571@kao2.melbourne.sgi.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline User-Agent: Mutt/1.3.26i X-Operating-System: Linux 2.2.20 (i586) Reply-By: Fri, 12 Apr 2002 00:55:19 -0700 Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org On Fri, May 10, 2002 at 12:42:51PM +1000, Keith Owens wrote: > 2.5.15 has four Makefiles where all objects are marked as exporting > symbols. This is lazy coding and causes spurious rebuilds. Please > specify only those objects that really export symbols. > > Also the export list is independent of whether an object is selected or > not. That is, export-objs is unconditional. Does this patch fix up arch/i386/pci/Makefile and drivers/pci/Makefile properly? thanks, greg k-h diff -Nru a/arch/i386/pci/Makefile b/arch/i386/pci/Makefile --- a/arch/i386/pci/Makefile Thu May 9 20:50:36 2002 +++ b/arch/i386/pci/Makefile Thu May 9 20:50:36 2002 @@ -24,6 +24,4 @@ obj-y += irq.o common.o endif # CONFIG_VISWS -export-objs += $(obj-y) - include $(TOPDIR)/Rules.make diff -Nru a/drivers/pci/Makefile b/drivers/pci/Makefile --- a/drivers/pci/Makefile Thu May 9 20:50:36 2002 +++ b/drivers/pci/Makefile Thu May 9 20:50:36 2002 @@ -37,7 +37,8 @@ obj-y += syscall.o endif -export-objs := $(obj-y) +export-objs := access.o hotplug.o pci.o pci-driver.o \ + pool.o probe.o proc.o search.o include $(TOPDIR)/Rules.make