From mboxrd@z Thu Jan 1 00:00:00 1970 From: Thomas Gleixner Subject: Re: [PATCH 1/2]cpuidle: makes AMD C1E works in acpi_idle Date: Tue, 19 May 2009 17:24:44 +0200 (CEST) Message-ID: References: <20090519080942.GA12117@sli10-desk.sh.intel.com> <20090519150908.GA27513@sli10-desk.sh.intel.com> Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Return-path: Received: from www.tglx.de ([62.245.132.106]:48644 "EHLO www.tglx.de" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1750721AbZESP0s (ORCPT ); Tue, 19 May 2009 11:26:48 -0400 In-Reply-To: <20090519150908.GA27513@sli10-desk.sh.intel.com> Sender: linux-acpi-owner@vger.kernel.org List-Id: linux-acpi@vger.kernel.org To: Shaohua Li Cc: "linux-kernel@vger.kernel.org" , "linux-acpi@vger.kernel.org" , "lenb@kernel.org" , "thisistempbox@yahoo.com" , "mingo@redhat.com" On Tue, 19 May 2009, Shaohua Li wrote: > On Tue, May 19, 2009 at 08:55:10PM +0800, Thomas Gleixner wrote: > > On Tue, 19 May 2009, Shaohua Li wrote: > > > > > When AMD C1E is enabled, local APIC timer will stop even in C1. This patch uses > > > broadcast ipi to replace local APIC timer in C1. > > > > > > http://bugzilla.kernel.org/show_bug.cgi?id=13233 > > > > Hmm. This was addressed before in > > > > commit a8d6829044901a67732904be5f1eacdf8539604f (x86: prevent C-states > > hang on AMD C1E enabled machines) > > > > We limit C-States to 1 via acpi_processor_cstate_check(). > > > > And we have the broadcast logic in c1e_idle() already. We also select > > c1e_idle as the idle routine, so why is the ACPI code not using this. ? > > We used to call pm_idle_save() for C1, which will call c1e_idle. But this is > removed, guess the reason is we want to do C1 enter per the information gotten > from BIOS. So this bug is a regression actually. The patch follows the logic > of c1e_idle(). Hmm, ok. As long as the code does not undo something which was already set up in c1e_idle() I have no objections. Thanks, tglx