From mboxrd@z Thu Jan 1 00:00:00 1970 From: "Rafael J. Wysocki" Subject: [RFC][PATCH 4/7] ACPI Suspend: Call _PTS before suspending devices Date: Sat, 5 Jan 2008 23:48:34 +0100 Message-ID: <200801052348.35899.rjw@sisk.pl> References: <200801052332.44500.rjw@sisk.pl> Mime-Version: 1.0 Content-Type: text/plain; charset="iso-8859-2" Content-Transfer-Encoding: 7bit Return-path: Received: from ogre.sisk.pl ([217.79.144.158]:58215 "EHLO ogre.sisk.pl" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752731AbYAEXQN (ORCPT ); Sat, 5 Jan 2008 18:16:13 -0500 In-Reply-To: <200801052332.44500.rjw@sisk.pl> Content-Disposition: inline Sender: linux-acpi-owner@vger.kernel.org List-Id: linux-acpi@vger.kernel.org To: ACPI Devel Maling List Cc: Arjan van de Ven , Carlos Corbacho , Linus Torvalds , Pavel Machek , pm list , Andrew Morton , Len Brown , Alexey Starikovskiy , "Moore, Robert" , Matthew Garrett From: Rafael J. Wysocki The ACPI 1.0 specification wants us to put devices into low power states after executing the _PTS global control methods, while ACPI 2.0 and later want us to do that in the reverse order. The current suspend code follows ACPI 2.0 in that respect which causes some ACPI 1.0x systems to hang during suspend (ref. http://bugzilla.kernel.org/show_bug.cgi?id=9528). Make the suspend code execute _PTS before putting devices into low power (ie. in accordance with ACPI 1.0x) and provide a command line option to override the default if need be. Signed-off-by: Rafael J. Wysocki --- Documentation/kernel-parameters.txt | 5 +++ drivers/acpi/sleep/main.c | 51 ++++++++++++++++++++++++++---------- 2 files changed, 43 insertions(+), 13 deletions(-) Index: linux-2.6/drivers/acpi/sleep/main.c =================================================================== --- linux-2.6.orig/drivers/acpi/sleep/main.c +++ linux-2.6/drivers/acpi/sleep/main.c @@ -26,6 +26,21 @@ u8 sleep_states[ACPI_S_STATE_COUNT]; #ifdef CONFIG_PM_SLEEP static u32 acpi_target_sleep_state = ACPI_STATE_S0; +static bool acpi_sleep_finish_wake_up; + +/* + * ACPI 2.0 and later want us to execute _PTS after suspending devices, so we + * allow the user to request that behavior by using the 'acpi_new_pts_ordering' + * kernel command line option that causes the following variable to be set. + */ +static bool new_pts_ordering; + +static int __init acpi_new_pts_ordering(char *str) +{ + new_pts_ordering = true; + return 1; +} +__setup("acpi_new_pts_ordering", acpi_new_pts_ordering); #endif int acpi_sleep_prepare(u32 acpi_state) @@ -74,6 +89,14 @@ static int acpi_pm_open(suspend_state_t if (sleep_states[acpi_state]) { acpi_target_sleep_state = acpi_state; + if (new_pts_ordering) + return 0; + + error = acpi_sleep_prepare(acpi_state); + if (error) + acpi_target_sleep_state = ACPI_STATE_S0; + else + acpi_sleep_finish_wake_up = true; } else { printk(KERN_ERR "ACPI does not support this state: %d\n", pm_state); @@ -91,15 +114,17 @@ static int acpi_pm_open(suspend_state_t static int acpi_pm_prepare(void) { - int error; + if (new_pts_ordering) { + int error = acpi_sleep_prepare(acpi_target_sleep_state); - error = acpi_sleep_prepare(acpi_target_sleep_state); - if (error) - acpi_target_sleep_state = ACPI_STATE_S0; - else if (!ACPI_SUCCESS(acpi_hw_disable_all_gpes())) - error = -EFAULT; + if (error) { + acpi_target_sleep_state = ACPI_STATE_S0; + return error; + } + acpi_sleep_finish_wake_up = true; + } - return error; + return ACPI_SUCCESS(acpi_hw_disable_all_gpes()) ? 0 : -EFAULT; } /** @@ -123,10 +148,8 @@ static int acpi_pm_enter(suspend_state_t if (acpi_state == ACPI_STATE_S3) { int error = acpi_save_state_mem(); - if (error) { - acpi_target_sleep_state = ACPI_STATE_S0; + if (error) return error; - } } local_irq_save(flags); @@ -187,6 +210,7 @@ static void acpi_pm_finish(void) acpi_set_firmware_waking_vector((acpi_physical_address) 0); acpi_target_sleep_state = ACPI_STATE_S0; + acpi_sleep_finish_wake_up = false; #ifdef CONFIG_X86 if (init_8259A_after_S1) { @@ -203,10 +227,11 @@ static void acpi_pm_finish(void) static void acpi_pm_close(void) { /* - * This is necessary in case acpi_pm_finish() is not called during a - * failing transition to a sleep state. + * This is necessary in case acpi_pm_finish() is not called directly + * during a failing transition to a sleep state. */ - acpi_target_sleep_state = ACPI_STATE_S0; + if (acpi_sleep_finish_wake_up) + acpi_pm_finish(); } static int acpi_pm_state_valid(suspend_state_t pm_state) Index: linux-2.6/Documentation/kernel-parameters.txt =================================================================== --- linux-2.6.orig/Documentation/kernel-parameters.txt +++ linux-2.6/Documentation/kernel-parameters.txt @@ -167,6 +167,11 @@ and is between 256 and 4096 characters. acpi_irq_isa= [HW,ACPI] If irq_balance, mark listed IRQs used by ISA Format: ,... + acpi_new_pts_ordering [HW,ACPI] + Enforce the ACPI 2.0 ordering of the _PTS control + method wrt putting devices into low power states + default: pre ACPI 2.0 ordering of _PTS + acpi_no_auto_ssdt [HW,ACPI] Disable automatic loading of SSDT acpi_os_name= [HW,ACPI] Tell ACPI BIOS the name of the OS