From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S269133AbUJKTu2 (ORCPT ); Mon, 11 Oct 2004 15:50:28 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S269205AbUJKTu2 (ORCPT ); Mon, 11 Oct 2004 15:50:28 -0400 Received: from fw.osdl.org ([65.172.181.6]:18836 "EHLO mail.osdl.org") by vger.kernel.org with ESMTP id S269133AbUJKTu0 (ORCPT ); Mon, 11 Oct 2004 15:50:26 -0400 Date: Mon, 11 Oct 2004 12:54:21 -0700 From: Andrew Morton To: Badari Pulavarty Cc: ak@suse.de, linux-kernel@vger.kernel.org Subject: Re: 2.6.9-rc4-mm1 HPET compile problems on AMD64 Message-Id: <20041011125421.106eff07.akpm@osdl.org> In-Reply-To: <1097509362.12861.334.camel@dyn318077bld.beaverton.ibm.com> References: <1097509362.12861.334.camel@dyn318077bld.beaverton.ibm.com> X-Mailer: Sylpheed version 0.9.7 (GTK+ 1.2.10; i586-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 X-Mailing-List: linux-kernel@vger.kernel.org Badari Pulavarty wrote: > > I get following error while linking kernel on 2.6.9-rc4-mm1. > x86_64/kernel/time.c seems to have a dependency on char/hpet.c > driver. Its forcing me to use CONFIG_HPET. > > LD .tmp_vmlinux1 > arch/x86_64/kernel/built-in.o(.init.text+0x2071): In function > `late_hpet_init': > arch/x86_64/kernel/entry.S:259: undefined reference to `hpet_alloc' > make: *** [.tmp_vmlinux1] Error 1 > > ... > > [hpet_alloc_fix.patch text/plain (638 bytes)] > --- linux.org/arch/x86_64/kernel/time.c 2004-10-11 09:17:15.613107488 -0700 > +++ linux/arch/x86_64/kernel/time.c 2004-10-11 09:14:05.983935504 -0700 > @@ -727,6 +727,7 @@ static unsigned int __init pit_calibrate > return (end - start) / 50; > } > > +#ifdef CONFIG_HPET > static __init int late_hpet_init(void) > { > struct hpet_data hd; > @@ -773,6 +774,7 @@ static __init int late_hpet_init(void) > return 0; > } > fs_initcall(late_hpet_init); > +#endif I assume you have CONFIG_HPET=n and CONFIG_HPET_TIMER=n? Andi, what's going on here? Should the hpet functions in arch/x86_64/kernel/time.c be inside CONFIG_HPET_TIMER?