From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1755084AbaCLPmZ (ORCPT ); Wed, 12 Mar 2014 11:42:25 -0400 Received: from smtp.nue.novell.com ([195.135.221.5]:47208 "EHLO smtp.nue.novell.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1754263AbaCLPmX (ORCPT ); Wed, 12 Mar 2014 11:42:23 -0400 Subject: Re: [RESEND] Fast TSC calibration fails with v3.14-rc1 and later From: joeyli To: "Rafael J. Wysocki" Cc: Julian Wollrath , Thomas Gleixner , x86@kernel.org, LKML , "Rafael J. Wysocki" , "H. Peter Anvin" In-Reply-To: <53642014.xoKWERQIvL@vostro.rjw.lan> References: <20140310110410.5b2218f6@ilfaris> <1394596807.26565.135.camel@linux-s257.site> <1902317.cSAIh6fs84@vostro.rjw.lan> <53642014.xoKWERQIvL@vostro.rjw.lan> Content-Type: text/plain; charset="UTF-8" Date: Wed, 12 Mar 2014 23:41:39 +0800 Message-ID: <1394638899.26565.197.camel@linux-s257.site> Mime-Version: 1.0 X-Mailer: Evolution 2.28.2 Content-Transfer-Encoding: 8bit Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Hi Rafael, 於 三，2014-03-12 於 14:30 +0100，Rafael J. Wysocki 提到： > > But I wonder: Can we simply enable SCI later? In other words, can > we > > split acpi_early_init() so that the part before > acpi_enable_subsystem() > > is done before timekeeping_init() and the part including and after > > is done right after anon_vma_init()? Would the TAD initialization > work > > then? > > Below is a patch implementing that idea. Julian, can you please test > this one too? > > Rafael > > --- > drivers/acpi/bus.c | 20 +++++++++++++++----- > include/linux/acpi.h | 1 + > init/main.c | 1 + > 3 files changed, 17 insertions(+), 5 deletions(-) > > Index: linux-pm/drivers/acpi/bus.c > =================================================================== > --- linux-pm.orig/drivers/acpi/bus.c > +++ linux-pm/drivers/acpi/bus.c > @@ -494,11 +494,21 @@ void __init acpi_early_init(void) > } > > status = acpi_load_tables(); > - if (ACPI_FAILURE(status)) { > - printk(KERN_ERR PREFIX > - "Unable to load the System Description Tables > \n"); > - goto error0; > - } > + if (ACPI_SUCCESS(status)) > + return; > + > + printk(KERN_ERR PREFIX "Unable to load the System Description > Tables\n"); > + > + error0: > + disable_acpi(); > +} > + > +void __init acpi_subsystem_init(void) > +{ > + acpi_status status; > + > + if (acpi_disabled) > + return; > After check the DSDT of target machine, I afraid the above patch is not enough to use _GRT and _SRT because they need opregion support of SystemMemory and SystemIO for trigger SMM to access RTC. In my current semifinished patches, I add code to run the acpi_ev_install_region_handlers() in acpi_enable_subsystem() when acpi_early_init(), for install region handler to support opregion. Before install region_handlers, it call acpi_enable() to enable acpi mode and I think need setup SCI interrupt before run acpi_enable(): acpi_early_init(void) acpi_pic_sci_set_trigger(acpi_gbl_FADT.sci_interrupt, acpi_enable_subsystem(u32 flags) acpi_enable(); /* Enable ACPI mode */ acpi_tb_initialize_facs(); acpi_ev_install_region_handlers(); /* Need it! */ For developing ACPI TAD support, I used SystemIO to access CMOS port in DSDT of Intel DQ57 for simulate ACPI TAD. It also need SystemIO opregion support as the real machine. I just run your patch on this machine but it failed. I think maybe still using ACPI_FADT_NO_CMOS_RTC to check does acpi_early_init() need run before timekeeping_init(). If there have any future machine that applied ACPI TAD but "Fast TSC calibration" fail, at least the alternate TSC calibration can work around issue. Thanks a lot! Joey Lee